FIELD MUSEUM OF NATURAL HISTORY
PUBLICATION 238
BOTANICAL SERIES VOL. VI, No. i
CITRUS PRODUCTS
PART I
BY
JAMES B. McNAiR
Associate in Economic Botany
B. E. DAHLGREN
Acting Curator, Department of Botany
EDITOR
CHICAGO, U. S. A.
August 19, 1926
CITRUS PRODUCTS
THMlBfUHX
Of THE
mmsmr IF
FIELD MUSEUM OF NATURAL HISTORY.
FRONTISPIECE.
(Courtesy of R. E. Dahlgren, Winter Haven, Florida)
A FLORIDA VALENCIA ORANGE TREE IN BLOSSOM AND FRUIT.
FIELD MUSEUM OF NATURAL HISTORY
PUBLICATION 238
BOTANICAL SERIES VOL. VI, No. i
CITRUS PRODUCTS
PART I
BY
JAMES B. McNAiR
Associate in Economic Botany
B. E. DAHLGREN
Acting Curator, Department of Botany
EDITOR
CHICAGO, U. S. A.
August 19, 1926
PRINTED IN THE UNITED STATES OF AMERICA
BY FIELD MUSEUM PRESS
PREFACE
The literature dealing with citrus products and by-products is
extensive but scattered over a wide range of journals, technical and
consular reports, and other publications and is often difficult of access.
In providing a general account of citrus products, their technology and
economics, the author has, therefore, thought it desirable to supple-
ment his own observations and the results of his own investigations with
a comprehensive summary of all the principal information available on
the subject with the purpose of providing a work of reference useful to
those concerned with the citrus industry in its various branches.
This book is divided into two parts. In the first, consisting of ten
chapters, descriptions are given of the citrus products from various
parts of the plant, such as the entire fruit, the rind, the pulp, the juice,
the stem, the leaves, and the flowers. Methods of their manufacture,
analysis, and use are also included. The second part deals with methods
of production of citrus products in various countries and the economics
of the industry.
The term "citrus products," as used here, applies to all products
of citrus plants, such as citric acid, essential oil, fruit juice, etc. The
term "citrus by-products" refers to by-products of the citrus fruit-
growing industry. From an economic point of view, many citrus pro-
ducts are considered as by-products, while others, such as oils of Neroli
and bergamot, can in no sense be regarded as such.
In a work of this character, including data from so many sources,
individual acknowledgments can not be made. The writer is indebted
to various publishers and authors for illustrations, as indicated in each
instance. Plate VI and the lower half of Plate V were donated to
the author by the La Salle Engraving Co. For references to the num-
erous sources of information the reader is referred to the bibliography
at the end of each chapter and to the author index.
The writer acknowledges gratefully his indebtedness to Dr. B. E.
Dahlgren for critical reading and revision of the manuscript. He also
wishes to express his appreciation to Miss Mary D. Alexander, who pre-
pared the manuscript for the printer and read the proof.
JAMES B. McNAiR
FIELD MUSEUM OF NATURAL HISTORY
April 30, 1925
CONTENTS
PACK
LIST OF FIGURES . jx
LIST OF PLATES xi
CHAPTER
I. CITRUS PRODUCTION AND THE CITRUS PRODUCTS INDUSTRY i
Necessity for the industry. Citrus waste products. Equipment
of citrus products factories. Preparation and condition of the
fruit.
II. PRODUCTS FROM THE RIND — ESSENTIAL OILS 9
The oil cells — position in leaf; quantity and odor varies with
variety ; position in the rind ; shape in relation to species ; origin.
Methods of obtaining the oil. Distillation. Mechanical produc-
tion— expression by hand and by machine. Extraction by dis-
placement. Extraction by solvents. Fruits yielding oil. Oil of
bergamot. Oil of lemon — properties of terpene-free lemon oil;
properties of other special oils; distillation test; detection of
pinene according to Chace ; Patane's method for determining the
amount of constituents soluble in dilute alcohol ; viscosity deter-
mination ; determination of the hydrocarbon content of concen-
trated lemon oils according to E. Bocker; adulterants. Oil of
limes — West Indian limette oils ; Italian limette oil. Oil of man-
darins— Japanese mandarin oil. Oil of sweet orange — Jamaican
sweet orange oil. Oil of bitter orange — Italian bitter orange oil;
West Indian bitter orange oil; Jamaican bitter orange oil. Oil
of citron. Oil of grapefruit. Esters in essential oils — terpinyl
acetate ; detection of citric acid esters ; glyceryl acetate ; non-
volatile esters. Methods of citral assay — Garnett, Walther, Sol-
daini and Berte, Parry, Sadtler, Romeo, Rother, Bruylants, Chace,
Hiltner, Kleber. Terpeneless essential oils.
III. PRODUCTS FROM THE RIND — PECTIN. — ^3
Summary of experimental progress — microchemical reactions;
protopectin ; pectin : orange pectin, physical behavior of pectin ;
pectase: action of pectase ; pectic acid; formation of fruit jelly ;
summary. Production and uses of pectin — pectin extraction;
methods for pectin estimation; jelly manufacture: yields, clari-
fication, aroma and flavor changes in jelly-making, sugar in jelly-
making, acid in jelly-making
IV. OTHER PRODUCTS FROM THE RIND. no
Dried, candied, and preserved rind
V. PRODUCTS FROM THE PULP 113
Canned peeled fruit. Juices, preserved and condensed — orange
juice: yields and composition, clearing, filtration, pasteurization
in barrels, summary ; grapefruit juice; mixed juices; condensed
viii CONTENTS
fruit juice: concentration of fruit juice by dehydration by means
of cold (the Giirber process, the Monti process, and the Gore
process), concentration of fruit juice by dehydration by means of
heat (the McClendon-Dick process). Fermentation products — fer-
mented juice for vinegar-making: control of the micro-organisms,
analyses of commercial orange wine, analyses of sparkling wines ;
partially fermented juice as a beverage; orange vinegar: nature
and origin of vinegar, after-treatment, diseases ; acetone. Citric
acid from lemons. Other sources of citric acid — citric acid by
fermentation; citric acid by artificial synthesis; analysis of cit-
ric acid and citrates; citrate of lime, lime juice, lemon juice, and
factory citric acid liquors, estimation of citric acid in presence
bf other acids, tests of purity of citric acid. Bitter glucosides —
hesperidin and hesperitin : method of preparation, properties,
microchemistry ; isohesperidin and aurantiamarin ; naringin.
Fodder. Paper
VI. PRODUCTS FROM THE SEEDS 167
VII. PRODUCTS IN WHICH THE ENTIRE FRUIT Is USED 170
Marmalade. Cull citrus fruits as fertilizer
VIII. PRODUCTS FROM THE FLOWERS 174
Oil of Neroli Bigarade. Oil of Neroli Portugal. Orange flower
• perfume — extraction with volatile solvents : selection and purifica-
tion of the solvent, the systematic extraction of the flowers, evapo-
ration of the solvent, recovery of the solvent; extraction with
non- volatile solvents with the aid of heat (infusion, maceration or
enfleurage d chaud) : maceration. Lime flower oil
IX. PRODUCTS FROM THE LEAVES 190
Oil of petitgrain. Soporific substance. Alkaloid (dimethylbetain).
Oil of bergamot leaves. Oil of Papeda leaves. Oil of leaves of
Citrus trifoliata. Oil of sweet orange leaves. Oil of grapefruit
leaves. Oil of lime leaves. Oil of mandarin leaves
X. PRODUCTS FROM THE STEMS 195
Oil of petitgrain. Petitgrain citronnier. Oil from immature
oranges. Orange and lemon wood
INDEXES 203
Author Index. Subject Index
LIST OF FIGURES
FIGURE PACK
1. TOTAL AND CALIFORNIA SUPPLY OF LEMONS I
2. TOTAL CONSUMPTION OF LEMONS IN THE UNITED STATES BY YEARS,
1901-20 2
3. WHOLESALE LEMON PRICES COMPARED WITH PRICES OF OTHER COM-
MODITIES, 1914-19 4
4. TYPICAL FLOW SHEETS ILLUSTRATING VARIOUS CITRUS BY-PRODUCTS
INDUSTRIES 5
5. FORMATION OF OIL CELLS IN THE RIND OF THE SOUR ORANGE 9
6. LlMONENE NlTROSOCHLORIDE CRYSTALS FROM LEMON OlL 30
7. LlMONENE AND PlNENE NlTROSOCHLORIDE FROM A LEMON OlL MlXED
WITH 5 PER CENT OF TURPENTINE 30
8. PlNENE NlTROSOCHLORIDE CRYSTALS FROM TURPENTINE 31
9. APPARATUS FOR GRAPEFRUIT OIL EXTRACTION 47
10. APPARATUS FOR ESTIMATION OF FIXED ESTERS 55
11. PECTIN (BLACK) IN INTERCELLULAR SPACES 77
12. A REDWOOD VISCOSIMETER 91
13. DUMPER, ICE-BREAKER, AND CENTRIFUGAL MACHINE USED IN
CRUSHING AND CENTRIFUGALIZING FROZEN APPLE JUICE 125
14. VINEGAR BARREL 142
15. RAPID PROCESS VINEGAR APPARATUS 144
16. APPARATUS FOR THE DETERMINATION OF CITRIC ACID 155
17. APPARATUS FOR NARINGIN EXTRACTION 161
18. NARINGIN 162
19. NARINGENIN 162
ix
LIST OF PLATES
ORANGE TREE IN BLOSSOM AND FRUIT Frontispiece
PLATE *ACING PAGE
I, MANUFACTURE OF THE OIL OF BERGAMOT IN CALABRIA 20
II. FILTERING- APPARATUS FOR THE OIL OF BERGAMOT 36
III. ESSENTIAL OIL STILLS 100
IV. COLLECTING ORANGE FLOWERS IN FRANCE 1 16
V. DISTILLATION OF NEROLI AND PETITGRAIN OILS 164
VI. PRINCESS OF NEROLI, AFTER WHOM ORANGE FLOWER PERFUME WAS
NAMED . . 180
BOTANICAL CLASSIFICATION OF SPECIES
CITED IN THIS WORK1
CITRUS L.
C. aurantifolia (Christm.) Swingle (Limonia aurantifolia Christm.
Citrus lima Lunan, not Aitch. C. limetta Auth., not Risso).
LIME.
C. Aurantium L. (C. Bigardia Risso). SOUR or SEVILLE ORANGE.
C. Bergamia Risso. BERGAMOT.
C. Limetta Risso. LIMETTE.
C. Limonia Osbeck (C. Limonium Risso). LEMON.
C. maxima Merr. (C. grandis Osbeck. C. decumana L.). GRAPE-
FRUIT. Shaddock. Pomelo.
C. medica L. CITRON.
C. mitis Blanco. CALAMONDIN ORANGE.
C. nobilis Lour., var. deliciosa (Tenore) Swingle (C. deliciosa
Tenore). MANDARIN and TANGARINE ORANGES.
C. nobilis Lour., var. Unshiu Swingle. UNSHIU or SATSUMA
ORANGE.
C. sinensis (L.) Osbeck (C. Aurantium, var,. sinensis L.). ORANGE.
C. taitensis Risso. OTAHEITE ORANGE.
FORTUNELLA Swingle. Kumquat.
F. margarita (Lour.) Swingle (Citrus margarita Lour.). OVAL
KUMQUAT.
F. japonica (Thunb.) Swingle (Citrus japonic a Thunb.). ROUND
KUMQUAT.
CITROPSIS Swingle and Kellerm. CHERRY ORANGE.
C. Schweinfurthii (Engler) Swingle and Kellerm. (Limonia
Schweinfurthii Engler).
PONCIRUS Raf.
P. trifoliata (L.) Raf. (Citrus trifoliata L.). TRIFOLIATE ORANGE.
*L. H. Bailey, Manual of Cultivated Plants. New York : Macmillan Co., 1924.
Xll
CHAPTER I
CITRUS PRODUCTION AND THE CITRUS
PRODUCTS INDUSTRY
Citrus fruit growing is an industry of relatively recent develop-
ment in the United States. Confined by limitations of winter tempera-
ture, soil and moisture conditions to a comparatively small area, there
is still considerable acreage available for further expansion which
appears to be rapidly taking place. Groves planted some years ago
are gradually reaching a stage of greater yield, thousands of acres
planted within recent years are coming into bearing and thousands of
acres are being planted annually. There is thus a tendency to large
production which, especially in view of the generally perishable nature
of citrus fruits, eventually threatens to amount to an overproduction
of fruit 'for shipment.
LEMONS TOTAL 8; CALIF.
POPULATION June 30. -CAUF. SHIPMENTS.
— — TOTAL SMIPMEHTS YEAR ENDED '/ao.
CALIF. •• ^IMPORTS.
TOTAU Stft»PLY- IMPORTS. INCREASE OVER 1918-19
~ ESTIMATED AT 1ST 9HOHTMS RATt
CALIF. - ACTUAL <•/,_ TO 3/3tl BALAMCC
OF SfASOH 18.00 CARS PtR MOUTH.
9000.
8 OOO.
5000
4 OOO.
•3 000
2 OOO.
I OOO
1903 '04- 'OS '06 VJ '08
'10 'll '\Z '13 'H 'IS 'ifc '|7 '16 '19 19ZO
'ZZ 'Z3 '34 'Z5 '!*. 'Z7
(California Citrograph)
FIG. i. — Total and California supply of lemons, 1903-20. While California
has secured a larger share of the American lemon market, the total supply has
increased only in proportion to population. From now on, there must be a very
rapid increase in per capita consumption to absorb an anticipated increase in
California shipments of 60 per cent in five years.
2 CITRUS PRODUCTS
An analysis of, e.g., the consumption of lemons in the United States
shows that it increased one-half in ten year*. While the quantity im-
ported remained nearly constant during this period, the domestic pro-
duction doubled. In 1903 California furnished approximately one-
fourth, in 1912, one-half, in 1918, four-fifths, and today produces an
amount of lemons equal to the consumption in this country. This local
increase in production will continue. In 1918, the lemon shipments con-
sisted of 6,913 cars, in 1923 of 8,430 cars. A similar condition exists
in the orange industry; 28,444 cars were shipped in 1918, and 71,971
in 1923.
Economic conditions in this field will undoubtedly become serious
as the market price of the fruit diminishes and the cost of production
increases. The situation may be ameliorated by converting present
waste fruit into marketable products and by securing a profitable use
for surplus fruit.
Wt.ooo.
U.S. LEMON SUPPLY- YEAR ENDING AUG. 31-
1301 "OZ '03 '04 'W '06 '07 '08 "O9 '10 'II *IZ '13 '14 '15 'l6 'l7 'ifl '19 '20
(California Citrograph)
FIG. 2. — Total consumption of lemons in the United States by years, 1901-20.
Carloads are figured on a 38o-box basis, and imports from Sicily are transposed
into a carload equivalent. It will be noted that in general imports have decreased
in proportion as California shipments have increased with the exception of crop-
failure years for California. The substantial increase in total consumption dur-
ing the last two years is shown by the columns at the right.
NECESSITY FOR THE INDUSTRY 3
In this connection it is of interest to note that in the Mediterranean
region the bergamot is grown especially for the essential oil of its fruit
and some other citrus trees for the oil of their flowers. In Sicily approxi-
mately one-third of the lemon crop is made into citrate of lime, lemon oil,
and other "by-products." It is not surprising, then, that a citrus products
industry is beginning to become established in the United States.
Outside of southern California, the lemon-growing industry is al-
most entirely confined at present to southern Italy and Sicily, where
the fruit-growing industry is conducted at a disadvantage because of
the large amount of hand labor required in orchard care. For frost
protection during the winter months the citrus trees there are covered
with straw mats fastened to trellises which are removed when the
danger of frost is past. The terraced arrangement of the groves also
calls for much more care in irrigation and in cultivation than do the
groves in the United States. The economic position of the proprietors
of the orchards along much of the coast from Naples southward to
Salerno is now unsound, because of the large emigration of laborers
from this coast to America and the consequent rise of wages. Only
with cheap labor could the terraces there have been built at a profit to
the landowner, and only with cheap labor is their cultivation now
practicable.
Though fruit not suitable for exportation from Italy and Sicily is
usually converted into citrus products, various circumstances, such as
price of labor, cost of transportation, and the financial condition of
the operator, affect the decision as to whether a crop otherwise ex-
ported is to be so manufactured. Less capital is required to export the
fruit, but it is cheaper to transport citrate and oil to market. The
citrus products industry is profitable, but is still in a rather crude
condition, as machinery, where used at all, is mostly primitive. Since
after the war with Germany economic conditions have been worse and
labor costs higher, good machinery is, however, being sought to take
the place of hand labor.
The citrus products industry has the same relation to lemon and
orange production as the drying and canning industries have to other
fruit-growing. To the citrus fruit-grower it affords a means of utiliza-
tion of waste fruit, a better price for low-grade fruit, and a market for
the crop during periods of low prices.
CITRUS WASTE PRODUCTS
The largest item of waste in the citrus fruit industry is waste or
cull fruit. Estimates of the amount of cull fruit in California vary
CITRUS PRODUCTS
WHOLESALE LEMON PRICES COMPARED.
from i to 5 per cent of the fruit shipped or about 25,000 tons per year.
Culls may be classified as follows : ( I ) fruit which shows physical
injury on the rind such that it is susceptible to rapid mold growth and
decay; (2) decayed and partially decayed fruit; (3) fruit which is
defective in shape, or has blemishes (Euthrip scars, red spider marks,
etc.) which are not a source of attack for molds and fungi, but which
injure its appearance and therefore give the fruit a low market
value; (4) frozen and sunburned fruit. Lemons which have partially
decayed are still valuable as a source of citric acid. Undecayed fruit
may be used for marmalade, citric acid (except oranges and grape-
fruit), juice, oil, peel,
or other products.
As regards the util-
ity of frozen lemons
Chace finds that those
which have been very
badly frozen are still
available for making
citrate of lime or cit-
ric acid. If the fruit
has been thawed grad-
ually enough, little or
no permanent injury
occurs ; but if thawing
is rapid, death of tis-
sue takes place. The
effect of frost on the
fruit is quite similar
to sunburn. The killed
tissues allow the mois-
ture to evaporate and
as a consequence the
fruit shrivels.
Where drying has
taken place, the yield
may be somewhat low-
~*~ Wholesale Price of 96 Commodities. (enA«»T
__ ._ Index of Articles firmer buys. (Dirt or AUBICULT
__ Lemon FViee9.MK KMT ««••«
(California Citrograph)
FIG. 3. — Wholesale lemon prices compared with
prices of other commodities, 1914-19.
ered, but in most cases a fair recovery can be made. Thomas, Young,
and Smith, in their paper upon the composition of frozen oranges, and
lemons recently published in the University of California Bulletin No.
304, show that even two months after the freeze of 1913, frozen lemons
contained as high as 4.79 per cent of anhydrous citric acid in the juice.
NECESSITY FOR THE INDUSTRY
?. T. Jfi//: /owr. 7»rf. awd £»ff. Chem.)
FIG. 4. — Typical flow sheets illustrating various citrus by-products industries
6 CITRUS PRODUCTS
With many frozen lemons, however, decomposition is rapid and the
sooner they reach the factory the better.
So far as oil is concerned, experience shows that, where there is
no serious surface injury, the oil content is not diminished, and the
oil itself is somewhat more easily recovered than in the case of unfrozen
fruit. Where surface injury has not been too great, the peel can be
used for candying, drying, or packing in brine. Where the pulp is not
injured in flavor, it is available for use in marmalade. Past experience
in Florida on grapefruit should teach great caution, however, in bottling
juice from frosted fruit, even where it apparently is of excellent quality
at the time of bottling.
Oranges, frozen or unfrozen, are not available for the manufacture
of citric acid or citrate of lime as the acid content of the juice is not
sufficient to pay for its recovery. If the surface injury is not great, the
oil recovery will not be curtailed, and surface injury, with oranges, is
not usually serious. Candied, dried, and brined peel can be produced
from the rinds. The pulp also is available for use in marmalade. As
shown by Thomas, Young, and Smith, there is a loss of sugar in the
juice of frozen oranges, and our experience is that the loss of flavor is
considerable, so that the material has been lessened in value as marma-
lade stock, although the pectin in the peel, which is valuable on account
of its jellying properties, does not seem to decrease greatly.
Where the fruit has not dried out to too great an extent, it can be
used for vinegar-making; loss of sugar in the juice, however, lessens
its value as vinegar material. It is not possible to produce even with
the roller process a standard vinegar where the juice contains less than
S^2 per cent of sugar. With rapid-process generators, the content must
be even higher. The bitter flavor, sometimes present in the juice from
frozen fruit, does not seem to affect the vinegar made from it.
Aside from satsumas, grapefruit is probably more frost resistant
than any of the other citrus fruits commonly grown. Its uses in by-
products, however, are limited. Citric acid is not present in its juice
in sufficient quantity to make its recovery profitable. It contains an
insufficient amount of sugar to make a standard vinegar, and while the
juice can be satisfactorily bottled, it is best not to use frozen material
in its preparation. The peel from frozen grapefruit is satisfactory for
the preparation of candied peel, and there is no reason why the pulp
cannot be used as usual in the preparation of marmalade.
Cull fruit is the most important source for manufactured citrus
products and is the one most generally considered. Orange flowers
have been used in the manufacture of the essence of neroli since the
PREPARATION AND CONDITION OF THE FRUIT 7
sixteenth century. Orange and lemon flowers are also used in the
manufacture of pomades and perfumes. Stems and leaves are used
in the manufacture of essential oils, and orange leaves contain a
soporific substance.
EQUIPMENT OF CITRUS PRODUCTS FACTORIES
Citrus product factories in the United States, where the cost of labor
is high, must rely upon machinery for the carrying on of successful com-
petition with low-priced foreign products.
For the manufacture of citric acid, a crusher, fermentation tanks,
filter-press, neutralizing and decomposing tanks, evaporating tanks, vacuum
pan, and crystallizing tanks are necessary. The initial outlay for equip-
ment in this process is quite heavy.
Oil production likewise requires a heavy expense for machinery as is
exemplified in the National City (California) plant.
For manufacture of dried peel comparatively cheap machinery is needed.
The peel after its separation from the fruit either by hand or machine is
dried on trays in the sun. Where this is impractical a current of heated
air is resorted to as a means of dehydration.
Marmalade-making machinery consists essentially of fruit slicers and
cooking kettles which, from the standpoint of their capacity, are of low
cost.
Equipment for the manufacture of preserved juice consists, for the
most part, of ordinary bottling machinery.
In all countries the outlay for machinery is essentially the same except
where oil is obtained by hand labor.
Details of factory equipment appear in the later descriptions of the
different processes.
PREPARATION AND CONDITION OF THE FRUIT
Fruit for by-product manufacture must first be examined and
cleaned unless it be obtained from a packing-house where cleaning has
already been performed. If, on the other hand, the fruit be dirty it is
best cleaned by means of brushes in the ordinary fruit-cleaning machine,
together with such necessary water as the fruit may require, since
adhering dirt and scale insects, while not always detrimental to the
manufacture of the product, are not desirable.
The quality and quantity of the final product depends in a measure
on the condition of the ripene'ss of the fruit especially in the manufac-
ture of essential oils, marmalade, and juice. Thus, in the manufacture
of lemon oil a green fruit gives an oil of the highest character, and
fruit of whatever nature that has been subjected to a sweating process
(injured by oxidation), or to slight decay or ageing, is almost unfit for
the manufacture of the finer products. The quantity of oil also varies
with the age of the fruit. Juice and marmalade made from oranges
8 CITRUS PRODUCTS
early in the season have a bitter taste. Oranges from the middle of
the season until its end do not give a bitter juice nor make a bitter
marmalade. Not only the bitterness but the sugar content of the fruits
varies. In ripe fruit the saccharine content is greatest. In lemons and
no doubt in other citrus fruits the citric acid content is greatest in the
unripe fruit particularly in those of summer. In the manufacture of
citric acid, however, the physical condition of the fruit is not as
important as in the manufacture of the essential oils.
REFERENCES FOR CHAPTER I
1. CHACE, E. M.
"The Use of Frozen Citrus Fruits in the Manufacture of By-Products,"
California Cultivator, LII (June 7, 1919), 773, 779.
2. FAWCETT, H. S.
"A Spotting of Citrus Fruits Due to the Action of Oil Liberated from
the Rind," Agricultural Experiment Station Bulletin No. 266 (1916).
Berkeley, California.
3. WEBBER, H. J., et al.
"A Study of the Effects of Freezes on Citrus in California," Agricul-
tural Experiment Station Bulletin No. 304 (1919). Berkeley, Cali-
fornia.
CHAPTER II
PRODUCTS FROM THE RIND— ESSENTIAL OILS
THE OIL CELLS
Even in the seedling stage, the stem of the citrus plant and its first
leaves are studded with essential oil glands. If a leaf of any citrus
plant is held up against the light and examined with a lens, three sizes
of cells are quite distinct. These are the essential oil cells or glands.
It should be noted that the larger ones are at certain distances from each
other, and that the intermediate
spaces are filled with smaller and
smaller oil cells. Every citrus
leaf has this peculiarity repeated
in the edge of the leaves, the large
oil cells occupying the angles be-
tween the large crenations, the
smaller ones, those between the
small crenations. It is curious to
note that in the rind of many cit-
rus fruits a similar disposition of
large and small oil cells is main-
tained.
The outer, or rind, surface of
a citrus fruit often resembles that
of a lump of dough, studded with
large and small pinhole-like de-
pressions, the botanical term for
which is "foveoli." It is interest-
ing to note that in cases of large
fruit, where the oil cells have
plenty of room to develop, the
three sizes are distinct, the larger
depressions corresponding to the
larger oil cells, the smaller de-
pressions corresponding to the
(Engler and Prantl) D
FIG. 5. — Formation of oil cells in the
rind of the sour orange {Citrus Auran-
tium L.) A, iB, first stage; C, before
breaking down of cell wall; D, after
breaking down of cell wall.
represented by still smaller depressions. In short, a longitudinal section
of any part of the rind would very closely resemble, in many cases, the
edge of the leaf of the malta lemon, with the difference that in the rind
cells of the second magnitude;
those of the third magnitude are
io CITRUS PRODUCTS
section the oil cells are much more developed. In small specimens, the
third-size cells are mere points.
The oil cells of the rind are usually balloon-shaped ; some are
pointed like the spindle-shaped cells of the pulp.
In opposition to Risso's theory, the external appearance of the oil
cells of the rind has no definite relation to the quality of the juice of
the pulp, as may be seen from the following instances : ( I ) two vari-
eties of lemon, one sweet, one sour, may be superficially indistinguish-
able, e.g., the nimboo (sweet lemon) of Furruckabod and the Kalair
kaghzi of Lucknow; (2) in one stage of the sweet orange, the simtare
of India, the juice is sour, later the juice is a mixture of acid and
sweet, while, when completely ripe, it has little acid, yet at all stages
the larger oil cells have their openings concave, and the smaller ones
are represented by minute convexities; (3) the sadaphal, which has
scarcely a trace of acid from beginning to end, has also the two
sets of oil cells on its surface; (4) finally, the Kuthairee nimboo
has its large oil cells, which are the only ones visible on its lumpy pro-
jections, convex, while its juice is distinctly sour.
It would, therefore, appear that Risso's notion about the constant
relationships of the concave or convex cells of the rind to the acid or
sweet juice of the pulp is untenable.
According to J. von Sachs, citrus ethereal oil glands originate from
a single mother-cell, which undergoes many divisions in all directions,
so that a multicellular mass of tissue of roundish form arises, the cells
of which subsequently become remarkable as containing very granular,
apparently dead, protoplasm. Later on, the thin cell walls dissolve, the
process commencing in the middle of the spheroidal group and proceed-
ing outward. There thus arises a roundish cavity filled partly with
watery sap, partly with drops of ethereal oil — the products of solution
of the mass of cells. The layers of tissue surrounding this cavity fit
closely on all sides, without intercellular spaces, and thus virtually form
a wall or receptacle for the secretion (Bonavia).
METHODS OF OBTAINING THE OIL
The methods of securing1 the oil from the plant may be divided into
four principal classes :
1United States patents involving extraction of oils are: W. A. Allport and
T. J. W. C. Davenport, No. 1002020, August 29, 1911 (machine for extracting
oil) ; S. L. Ames, No. 1097607, May 26, 1914 (flavoring emulsion) ; A. W.
Giampietro, No. 1116880, November io, 1914 (machine) ; E. J. Sheehan and
W. S. McKay, January 9, 1917 (process for extraction by solvents); F. A.
McDermott, Nq. 1353169, September 21, 1920 (process) ; S. C. Hood, No. 1186317
(machine for peeling).
MECHANICAL PRODUCTION u
1. Distillation under normal or reduced pressure
o) Normal
&) Steam
2. Mechanical processes
a) Expression by hand
6) Expression by machine
3. Displacement
4. Extraction by solvents
a) Volatile solvents
fc) Non-volatile oils or fats
DISTILLATION
All the essential oils are readily volatile in water vapor. Distilled
citrus oils are a commercial commodity but they are of far less value
than expressed oil. Citrus oils even when distilled under diminished
pressure and at a low temperature (which tend to reduce decompo-
sition) decompose more or less rapidly. Ordinary distillation of the
peel leads to a decomposition of the peel with no yield of oil. Finely
ground peel gives no better results. Dry steam distillation results in no
yield of oil, presumably also from the decomposition of the oil. Wet
steam distillation gives the best results. The pulp, when finely ground
in water, is considered as in the best condition for this variety of dis-
tillation. A good yield of clear, colorless oil is obtained.
In connection with steam distillation, it is interesting to note pos-
sible catalytic action. If the distilling vessel is made of galvanized iron,
i.e., lined with zinc, no yield whatever of oil is obtained. The stills
employed are therefore lined with tin, silver, or glass which have no
detrimental effects.
One firm produces lemon oil commercially in southern California. Dis-
tilled oil made by this firm sells at from 40 to 60 cents per pound, and is
used by large soap manufacturers for scenting toilet soaps ; ten pounds
are obtained per ton of fruit. This oil, however, does not have the full rich
odor of pressed lemon oil and is not suitable for flavoring purposes. A
better grade of oil is made in the same factory by a secret process, the
"cold method."
MECHANICAL PRODUCTION
a) Expression by hand. — The sponge process, which is usually em-
ployed in Sicily and Calabria is essentially as follows: The fruit is
first soaked in water for a short time and the peel is then removed in
three parts, each third being pressed flat by the fingers against a sponge
held in the other hand. The oil glands are burst by the pressure and
the oil, together with a considerable quantity of water and some juice,
is absorbed by the sponge The sponge is squeezed from time to time
12 CITRUS PRODUCTS
into a bowl and the oil thus obtained is finally separated from the water
and filtered in closed filters. The oil secured in this manner is the best
commercial oil of today ; it has the best keeping qualities as well as the
best flavor and aroma; and it forms the pharmacopoeia standard. The
sponge process which involves hand labor is impracticable in the United
States, because of the high cost of labor. Pitino Brothers, of Los
Angeles, have put this method to a thorough test with laborers formerly
employed in Sicily.
The ecuelle process is employed in Northern Italy and the south of
France. An ecuelle is a saucer-shaped vessel, 8-10 inches in diameter,
made of tinned copper, the inside of which is covered by short spikes
about a quarter of an inch long. The bottom is connected by a hollow
tube through which the oil passes to a collecting vessel. The whole
fruit is placed on the ecuelle and by rapid rotatory motion the oil glands
are burst and the oil is ejected.
In the Scorsetta process the fruits are cut into halves, the pulp is
removed by a spoon and the peel is twisted and pressed against a sponge.
The pulp is used for the production of lemon juice and the peel is
salted.
Other devices involving hand labor are also used for the production
of these oils. In one of them the fruits are brought into contact with
small knives which puncture or rupture the oil glands. The oil may be
drawn out of the vessel by means of vacuum pumps, and steam may
even be admitted to facilitate the process.
&) Expression by machine. — Fresh unpressed peel will absorb 10
per cent of its weight of oil. By pressure applied to the macerated peel
nearly all the original oil can be recovered as an emulsion. This emul-
sion can be decomposed by heating to a temperature of not more than
90° C. One-tenth of its volume of hot 2 per cent gelatin is then
added, thoroughly mixed, and one-fourth its volume of a hot 10 per cent
tannin solution is stirred in. The emulsifying agent is thus coagulated
and the oil released.
In 1916 Hood invented a machine for peeling citrus fruits. This
machine simply grates the rind from the fruit. It is not very success-
ful in peeling oranges that depart from a spherical shape, as the machine
then tends to peel the central portion only and not the ends of the
fruits.
Experiments conducted on a laboratory scale indicate that finely
ground peel subjected to centrifugal action will yield about 5 Ibs. of
oil per ton of fruit.
EXTRACTION BY SOLVENTS 13
EXTRACTION BY DISPLACEMENT
A novel and mystifying method for the manufacture of lemon oil has
been announced by Liotta. The lemons are pounded in a large vessel and
mixed with a liquid, the composition of which is kept secret. This liquid,
said to be an acid, causes the separation of the oil, which rises to the sur-
face and is removed. A "substance" is then added which liberates the
"acid," so that it can be used again while the residue, consisting of lemon
juice, pulp, peels, etc., may be sold to manufacturers of citrate of lime.
It is claimed for this new method that the yield of oil is 2j^ per cent
higher than with the old method of pressing by hand. As regards cost, this
is stated to be but 2.3 cents per 1,000 lemons, the operation requiring only
twenty-two minutes. It is expected that by perfecting the method and
working on a large scale, the operation can be carried out much more
rapidly.
As long as no further particulars are known about the "acid" and the
"substance" or the manner of recovering it, the possible usefulness of this
method cannot be determined.
EXTRACTION BY SOLVENTS
So far, no essential oil of the citrus fruit is produced commercially
by the volatile solvent processes.
Experiments indicate that when absolute alcohol is used as a solvent,
the proteins, etc., of the rind are coagulated in such a manner as to
prevent the solution of the oil. Fifty per cent alcohol on the other
hand dissolves the oil readily, but also dissolves coloring matter.
Petroleum ether of 86° Baume gives a good yield of colorless oil.
However, it has been found impossible subsequently to separate all the
petroleum ether from the oil ; there always remains enough of the sol-
vent to impair the odor of the citrus oil.
Ethyl ether gives similar results to those of petroleum ether. In-
flammability and expense also tend to prohibit the use of this solvent.
The lemon oil produced by the volatile solvent process has much the
same characteristics as that of the orange, with the exception that the
oil produced from green fruit has a higher citral content than that pro-
duced from fruit which has been subjected to the sweating process, or
is fully ripe.
Sheehan in 1917 patented a process for extracting essential oils by a
volatile solvent. The solvent is said to be obtained from the fruit itself;
its nature he does not reveal although he says it may be obtained also from
other sources.
The extraction by non-volatile solvents is typified by the enfleurage
process, used to obtain the perfumed oil from the flowers. A pure
neutral fat, such as purified lard, is melted and poured onto trays,
technically known as "chassis" and the flowers are laid on the fat for
14 CITRUS PRODUCTS
a short time, then removed and replaced by fresh flowers until the fat
has become saturated. The treated fat is sold as "pomades," or is em-
ployed in the manufacture of perfumes, being extracted by means of
warm alcohol in closed vessels. Such extractions are not at present
commercial in the United States (see chapter viii).
FRUITS YIELDING OIL
Oil may be obtained from the rind of all citrus fruits. Commer-
cially the lemon yields the most important oil manufactured in largest
quantities. The bergamot, the citron, the grape fruit, the limes, the
mandarin, the bitter and the sweet orange, all furnish valuable essential
oils.
OIL OF BERGAMOT
Origin and production. — The bergamot tree (C. Bergamia Risso)
is cultivated exclusively in Calabria, the pear-shaped, pale yellow, thin-
skinned fruit being used for oil production only. In Sicily the tree
does not thrive.
Properties. — Oil of bergamot is a brownish-yellow or honey-colored
liquid often colored green by the presence of copper or chlorophyll
(Liotta, 1899). It has a bitter taste and a very pleasant odor. Its
specific gravity at 15° is 0.881-0.886. The angle of rotation, which
on account of the dark color of the oil can mostly be determined only
in a 50 or 20 mm. tube, varies from +8° to +22°. However, optical
rotations as low as -j-5°24' and as high as +24° have been observed;1
HD20o 1.464 to 1.468; acid value i to 3.5."
The oil yields a clear solution with about one-quarter to one-half
volume of 90 per cent alcohol, and the solution does not become turbid
on the addition of more alcohol. All oils do not dissolve clearly in 80
per cent alcohol. Many, and especially those of a high ester content,
often give turbid mixtures, from which fatty globules separate on the
bottom on standing. The reason for this phenomenon has not yet been
determined, but can probably be sought in the waxlike constituents
which get into the oil during expression. This non-volatile substance,
which partly separates as a deposit when the oil is kept for some time,
consists principally of bergaptene. It remains in the residue when the
oil is evaporated on a water bath or by rectification; it amounts to
5-6 per cent.
Rectified oil of bergamot is colorless and has a lower specific gravity
(0.65-0.875) as well as a slightly higher rotatory power than the orig-
1Report of Schimmel & Co. (April, 1910), p. 59.
Vfctd. (April, 1908), p. 40.
OIL OF BERGAMOT 15
inal oil. The rectified oil is as a rule less valuable because during the
steam distillation part of the ester is always decomposed.
The amount of ester present is a measure of the value of bergamot
oil, i.e., the oil is the better the more linalyl acetate it contains. Al-
though the average content of ester varies somewhat in different years,
it usually amounts to between 34 and 40 per cent, but sometimes rises
as high as 45 per cent. Even within the limits of the same harvesting
period great variations occur. The oils obtained at the beginning of
the harvest from less ripe fruit contain less (down to 30 per cent
linalyl acetate) ; with increasing ripeness the ester content increases,
for which reason the oil expressed from ripe fruit is the best.
Composition. — As early as 1840 Soubeiran and Capitaine called
attention to the presence of different terpenes in bergamot oil. Wallach
showed in 1884 that eMimonene was contained in the fraction boiling
from i75°-i8o°. When he heated the fraction of the oil boiling from
i8o°-i9O°, which, indeed, absorbed bromine but yielded no solid
bromide, to a higher temperature, high-boiling condensation products
were formed, and on again fractionating, the portion going over up to
190° gave dipentene tetrabromide melting at I24°-I25°. It does not
follow from this, whether the dipentene detected in this manner i's to
be considered as an original constituent of the oil, or whether it has
been produced by heating the fraction i8o°-i9O° which no doubt con-
tained linalool. Likewise, the observation made by Semmler and Tie-
mann in 1892, according to which the oil boiling 17° higher than the
limonene fraction yielded dipentene tetrabromide, cannot be considered
as a proof of the presence of dipentene in bergamot oil.
Our knowledge of the most important constituent as far as the
odor of the bergamot oil is concerned is due to two investigations, pub-
lished at nearly the same time, by Semmler and Tiemann, and by
Bertram and Walbaum. By these investigations it was shown that
the principal carrier of the bergamot odor is the acetic ester of /-linalool.
In addition to this ester, free /-linalool and possibly substances not
yet isolated take part in the formation of the aroma.
According to F. Elze, three other alcohols are contained in berga-
mot oil. He examined a fraction which, on a factory scale, was ob-
tained to the extent of 25 per cent in the production of terpene-free
bergamot oil. After 'saponification, distillation with water vapor, and
f ractionation, it yielded an oil with the following properties : dls<>
0.890; OD— 10° (in a loo-mm. tube). Treated with phthalic acid
anhydride, a small amount of an oil was obtained which proved to be
dihydrocuminic alcohol (melting-point of naphthyl urethane 146° to
1 6 CITRUS PRODUCTS
147°) and which upon oxidation with Beckmann's chromic acid mix-
ture yielded an aldehyde boiling at 235° and had a cumin-like odor.
Regenerated from its semicarbazone (melting-point 198° to 199°) di5o
was 0.970 and OD— 39°. The bulk of the alcohol mixture, however,
consisted of nerol (dlaoo.88o) which was further characterized by its
tetrabromide, melting at 118°, and its diphenyl urethane, melting at 50°.
In that portion which did not react with phthalic acid anhydride,
Elze proved the presence of the terpineol melting at 30° (melting-
point of the phenylurethane 110°). This observation is very impor-
tant since it has been found recently that bergamot oil is occasionally
adulterated with terpinyl acetate.1 Hence it is a mooted question at
present whether terpineol is a normal constituent or not.
Charabot has made a comparative study of the oils of bergamot
from the green and the ripe fruit. The oil from the green fruit had a
specific gravity of 0.882 at 14°, OD is -j-i4°38'; it contained 0.289 per
cent of free acid, 33.8 per cent of linalyl acetate, 13.9 per cent of
linalool, and 5.9 per cent of bergaptene. The oil from the ripe fruit
had a specific gravity of 0.883 > ao is -}-2Q°yft 0.283 P61" wnk of acid,
37.3 per cent of ester, 5.9 per cent of linalool, and 5.5 per cent of
bergaptene. He draws the conclusion that in ripening the original
linalool is changed to the ester and that during this process some of
the linalool is dehydrated with the formation of terpenes.
The bergaptene contained in the oil to the extent of about 5 per
cent is completely odorless. A whole series of investigations has been
carried out on this compound.2 Pomeranz in 1891 succeeded in clear-
ing up its constitution. Bergaptene, C12H8O4, forms soft, white, satin-
like, tasteless needles, which are odorless at ordinary temperatures, but
on heating give off aromatic vapors, and melt at 188°. Bergaptene is
the monomethyl ether of a dioxycumarin which is traceable to phloro-
glucin. By treating with methyl iodide and alcoholic potassium hy-
droxide methyl bergaptenic acid and its methyl ester result. According
to this, bergaptene is the inner anhydride of bergaptenic acid. H.
Thorns and E. Baetcke assign to it the following formula:
OCH,
co
H
It is isomeric with xanthotoxin.
^Retort of Schimmel & Co. (April, 1910), p. 60.
'See Mulder, Ohme, Franke, Godeffroy, Tilden and Beck, Crismer, in Refer-
ences at end of this chapter, p. 67.
OIL OF BERGAMOT 17
Several other constituents have been found in a bergamot oil by Bur-
gess and Page. Although these investigators contend that the oil examined
by them was pure, this does not follow from the constants recorded (viz.,
d150o.885; [a]D-f-8°; fgfrpfr'i0080!! ester content not mentioned). Hence
the examination must be repeated with material of unquestioned purity,
before octylene, pinene, camphene, and bisabolene can be regarded as con-
stituents of bergamot oil. In the lowest fractions acetic acid (analysis of
barium salt) was found. It was scarcely necessary to prove the presence
of free acetic acid, since it is a well-known fact that the acids, which in
the volatile oils are combined with alcohols, also occur in the free state to
a slight extent. Moreover, Burgess and Page did not prove the presence
of free acetic acid in bergamot oil, for they accumulated it by means of
fractionation of the oil, a process by which the acid is formed from the
linalyl acetate, as they themselves surmise.
Fraction 150° to 155° revealed a very low specific gravity and a low
refractive index. Hence they infer the presence of an olefinic hydrocarbon,
octylene. This inference Burgess and Page endeavor to substantiate by
pointing out that the odor of this fraction reminds of the octylene found
by the same authors in lemon oil, also by demonstrating that upon oxida-
tion with potassium permanganate butyric acid (identified only by its odor)
results.
Laevogyrate ([o]D — 8.3°) a-pinene and laevogyrate ([a]D — 22.8°)
camphene were identified in fractions 157° to 158° and 164° to 165°
respectively. The former was characterized by its hydrochloride (melting-
point 125°), the latter by its conversion into woborneol (melting-point
203°). The higher boiling fraction of the bergamot oil contained bisabo-
lene (limene) C15H24 (melting-point of hydrochloride 79°).
Examinations. — Formerly the determination of the purity of bergamot
oil was not difficult because adulterations then used changed the physical
constants to a considerable extent. As the specific gravity of pure oils
varies within the comparatively narrow limits of 0.881-0.886, the addition
of turpentine oil,1 lemon oil, orange oil, as well as distilled bergamot oil pro-
duce a decrease, fatty oil, cedar wood oil, or gurjun balsam oil an increase,
in the density. A part of these adulterants would also change the angle of
rotation which lies between -f-8° and -J-2O0 with pure oils.
Nevertheless, in spite of the normal rotation, an oil might be adulter-
ated provided the normal limits were not interfered with by the proper
selection of adulterants such as turpentine oil with lemon oil or orange
oil. In order to detect such adulteration Romeo and Moricca fractionate
30 c.cm. of bergamot oil into fractions of 5 c.cm. and determine the
angle of rotation of the first two fractions. In the case of pure oils, the
rotation of the first fraction is greater than that of the second; in the case
of adulteration, the reverse holds true.
The solubility determination with 90 per cent alcohol gives with ber-
gamot oil results of only slight value, as by it only very extensive adultera-
*As bergamot oil contains no pinene, the presence of this hydrocarbon can be
considered as proof of the adulteration with turpentine oil.
1 8 CITRUS PRODUCTS
tions can be recognized. Only a part of pure bergamot oils, as already
mentioned, is soluble in 80 per cent alcohol. If a bergamot oil dissolves
to a clear solution in this solvent it is free from fatty oil, turpentine oil,
and orange oil. If, however, it does not dissolve, this may be due either
to an adulterant, for instance, fatty oil, or also to the presence of large
quantities of bergaptene or waxlike constituents.
The detection of fatty oil is effected by weighing the residue left by
evaporating the oil at 100°, which with normal oil amounts to 4.5-6 per
cent.
About 5 gm. of oil (weighed accurately to .01 gm.) are weighed off
in a glass or porcelain dish and heated on a water bath until that which
remains has lost all odor of bergamot oil. After cooling, the dish, pre-
viously tared, is weighed with the residue. If this amounts to more than
6 per cent of the oil used, fatty oil is present. Each additional per cent
represents I per cent of adulterant. Thus, e.g., a bergamot oil adulterated
with 5 per cent of olive oil will have a residue of from 10 to n per cent.
In the oils adulterated with turpentine oil, orange oil, or distilled ber-
gamot oil, the residue will in certain cases amount to considerable less than
5 or 6 per cent.
The determination of the residue is of special importance as fatty oil
gives a high saponification number and may, therefore, easily give rise
to mistakes.
Ester content. — The determination of the ester content1 not only al-
lows the detection of adulterations, but also furnishes a criterion of the
quality of the oil. This is the better, the greater the amount of linalyl
acetate.
By the saponification, the addition of the essence obtained by the dis-
tillation from the press residues or the small rejected fruit,2 which shows
a much smaller saponification number than the expressed oil, is also
detected.2
How far, under certain circumstances the adulteration may be carried,
is shown by two samples of "artificial bergamot oil" which were submitted
to Schimmel and Co.1 for analysis. The oils had the following constitu-
ents: (i) d180 1.0432; <ZD 7°3o'; acid value 0.3; ester value 403.3. (2)
rf160 1.0858; aD +7°3o'; acid value 0.9; ester value 577.1.
Even for artificial oils, density and ester values are abnormal to such
an extent that from them alone the peculiar composition of the oils may be
judged. Based on the ester value, the ester, calculated for linalyl acetate,
would be 141 per cent and 202 per cent respectively — figures which are quite
impossible and from which it can be deduced at once that considerable
*As according to A. Borntrager (Zeitschrift fur analytische Chemic, XXXV
[1806], 35) the evaporation residue on saponification gives numbers which cor-
respond to an amount of 2 per cent of linalyl acetate, this number ought by right
to be deducted from the result found. This is, however, not done, as the method
would thereby be only made unnecessarily complex.
2 An oil obtained in Messina by distillation from the expressed bergamot peel
possessed an ester content of only 12 per cent (specific gravity 0.865). Two oils
distilled from the peel of fallen unripe bergamot fruit contained 6.3 and 23.5
per cent of ester (specific gravity 0.868 and 0.889) (Report of Schimmel & Co.
[October, 1894], p. 15).
^Report of Schimmel & Co. (October, 1914 — April, 1916), pp. 18-20.
OIL OF BERGAMOT 19
quantities of esters are present which should not be found even in artificial
oils. Further examination revealed the fact that in both samples these
esters were not linalyl acetate, as should be the case in artificial bergamot
oil, but mostly bodies of no value for the odor, namely, phthalic acid esters
in one sample (probably diethyl phthalate) and glyceryl acetate in the
other. As both oils contained about 70 per cent of these bodies respectively,
the proof of their presence was easy, in spite of the smallness of the
samples. Glyceryl acetate was isolated by snaking out with a 5 per cent
solution of alcohol and was recognized, i.e., by the high saponification value
(676) whereas in the other oil the acid, which did not pass over with
steam in the determination of the acid number II (Report [October, 1910],
p. 61) was identified as phthalic acid (melting-point 203°; melting-point of
the anhydride obtained by sublimation 130° ; fluorescein reaction).
As mentioned above, both samples contained about 70 per cent of these
esters, that is to say, such an extent that one cannot speak of artificial oils
any longer, but more correctly of cheap — and from the point of view of
their strength — valueless esters which had been perfumed more or less
cleverly so as to resemble bergamot oil. Such adulterated substitutes
have, needless to say, nothing in common with a scientifically prepared
artificial bergamot oil.
It is easily explicable why during the Great War many volatile oils
which only occurred rarely in trade or were almost wholly exhausted
should have been subject to adulteration. Bergamot oil belongs to this
class ; the few samples which were submitted for inspection1 were nearly
all adulterated.
A sample from Messina which had been sent them from Switzerland
showed the following characteristics: rf150 0.8761; ao +28°; acid value
2.7; ester value 70.0 equals 24.5 per cent ester, calculated as linalyl acetate.
The constants prove quite alone that the oil was an inferior, adulterated
product. The specific gravity is too low, the degree of rotation much too
high and the ester content, which in a good trade sample ought not to be
under 34 per cent, quite insufficient. The high rotation indicates an admix-
ture of lemon oil.
Two other specimens were of such quality that it was difficult to
decide whether they were adulterated oils or artificial products. One of
them showed the following characteristics: dl50 0.8866; aD -j-I5°» residue
on evaporation 6 per cent (liquid) ; acid value 0.9; ester value 59.7 equals
20.9 per cent ester calculated as linalyl acetate; soluble in one volume and
more of 80 per cent alcohol. In this case also the much too low ester
content immediately characterized the oil as quite inferior. Strange to say,
the oil, in spite of its very low ester content, dissolved surprisingly well in
alcohol, as the majority of bergamot oils do not give a clear solution in 80
per cent alcohol. Even in the case of the oils that are soluble in it the
diluted solution is nearly always turbid; that is all the more to be ex-
pected from an oil with a low ester content. Another anomaly finds its
expression in the quality of the residue on evaporation which in this case
was liquid, instead of butter-like in consistency. All these facts indicate
not a natural oil at all, but an unsuccessful artificial product.
Report of Schimmcl & Co. (April-October, 1917), pp. 16-17.
2O CITRUS PRODUCTS
The same holds good of a second sample which gave the following
constants: d150 0.8864; aD +I5°46'; acid value 37; ester value iio.i; sapon-
ification number 113.8; acid value n 63.5; difference between saponification
value and acid value II 50.3; residue on evaporation 19.3 per cent. The
quality of this product was betrayed by the too great difference between
the saponification value and acid value (normally not above 10), as well
as by the much too great a residue on evaporation (normally 4.5 to 6 per
cent). The sample apparently consisted only of a fatty oil and probably
contained none of those waxlike components which are so characteristic
of bergamot oil. As fatty oils have a high saponification value, the object
was apparently to suggest the presence of a bergamot oil rich in linalyl
acetate, as the above-mentioned ester value would lead to the assumption
of about 38.5 per cent linalyl acetate. In reality, however, this content is
considerably smaller and can approximately be calculated from the acid
value n, which would correspond to a linalyl acetate quantity of 22.2
per cent. As indicated above, this oil also gave the decided impression
of an artificial product the manufacture of which had been a complete
failure.
A sample with the proud label "Oleum Bergamotte la Reggio rect.
pur.," handed by a Leipzig firm to the same chemists for investigation,
was simply artificial, about 80 per cent linalyl acetate, as its constants
prove: d160 0.8997; aD — io°n'; nD20, 1.45082; acid value 1.9; ester value
233.3, corresponding to 81.7 per cent linalyl acetate soluble in 2.7 vol. and
more of 70 per cent alcohol.
Other constituents. — While washed bergamot oil shows a small saponi-
fication number, no attempt has been made to determine the esters present
in it, but it might be expected from the analogy of this oil with the other
citrus fruit oils that both linalyl- and geranyl-acetate are present.
From the analysis and the distillation figures it would seem that the
constituents present in the oil are in about the following proportion:
Per Cent
d-Limonene 90-92
Citral 3-5
o-Pinene 0.5—1.5
Geraniol 1.0-2.0
Linalool 1.0-2.0
Citronellal Some
Linalyl and geranyl esters Some
The oil has a very pleasing odor and flavor and when 30 per cent by
volume of pure ethyl alcohol is added it appears to keep as well as either
lemon, orange, or lime oil. Its properties seem to place it between the
lemon and the orange in value.
Methods of production. — The photographs on Plates I and II illustrate
the details of the operation of the extraction of oil of bergamot on the
estate of M. le Commandatore Spinella at Lazarro.
Plate I shows a view of the workshop where the pressing is carried out
by means of the machine shown there. This machine is composed
of a wheel with pegs turning a drum with bars ending in the bowl. This
bowl is a kind of plate, 20 cm. in diameter, provided with copper points
OIL OF LEMON 21
i cm. long and fitting exactly in a cylinder the bottom of which constitutes
a second similar plate on which the fruits are placed. By its weight, the
drum presses the fruits, the points of the two bowls tear up the rinds and
the liquid which escapes flows into a receiver placed at the lower part of
the machine.
Plate II shows the particular means employed for the filtration of the
essential oil. The crude oil coming from the machine is placed in filters
of coarse flannel which are tied at the top in such a way as to form a
sort of pocket. These are then suspended from iron rods, after having
been cased with a metal arrangement as shown in the figure. The conical
portion of this receiver, loaded with weights, presses the bulging sides of
the filter, which assists and hastens the flow of the liquid into the lower
portion of the receiver, from which it is withdrawn by the tap. There is
thus obtained a mixture of essential oil and water which is subjected to
decantation.
Each apparatus is tended by three men, one to turn the crank, one to
sort the fruits, selecting those of an equal size in order that the pressure
may be exerted uniformly, and one to clean out the bowl when the fruits
have been pressed. 4 •« .,* ,
The fruits from which the essential oil has been pressfed are next sub-
jected to the action of large presses; the juice obtained is utilized for the
manufacture of calcium citrate.
The method for controlling the exhaustion of the rinds of the fruits
is a most simple one and is worth describing. The foreman of the factory
peels a fruit which has come out of the bowl. He then presses the peel
between his fingers holding it a short distance from the flame of a lamp
which is held by an apprentice. If the pressing has been badly done, fine
droplets of essential oil are projected out, making so many small brilliant
flames, which indicate bad workmanship.
This work is done at night time, in order to avoid too great evapora-
tion. The essential oil produced by the machine is far superior to that
yielded by the method called "by the sponge." The latter, in fact, is no
longer used for the bergamot, except in the case of fruits which are too
small to be treated in the machine.
In Calabria lemons are also expressed with the aid of this machine.
However, the oil thus obtained is invariably green (see Properties on p.
24) and is not salable as such, but is said to be used for the adulteration
of bergamot oil.
OIL OF LEMON
This oil is obtained by expression of the fresh peel of lemons (Citrus
limonia Osbeck) and is produced chiefly in Sicily and on the Riviera.
Chemically, it is probably one of the most complex of the essential oils.
Properties. — Lemon oil is a light-yellow liquid possessing the agree-
able odor of fresh lemons and an aromatic, mild, and, somewhat later,
bitter taste. Inasmuch as oils produced in geographically contiguous
districts reveal decided differences in their properties, it is difficult to
22 CITRUS PRODUCTS
establish limit values of general applicability. Moreover, the season
and degree of maturity of the fruit exert a marked influence on the
properties of the oils. Thus, e.g., the oils expressed from the fruits
collected in November reveal the highest optical rotation, but this
diminishes with the increasing maturity of the fruit as the winter ad-
vances. At the same time specific gravity and citral content increase.
A shortage of oil sometimes occurs when a crop of fruit is small, and
lemons are then expressed which in normal years would be used for
other purposes. This happened in 1913, when the price of lemon oil
rose to the height of $10.23 per i kg. During the months of April and
May of that year much oil was expressed from the so-called bianchetti
lemons, which revealed a low citral content and an angle of rotation as
low as -f-54°- It is further noteworthy that climatic conditions, such
as continued drought or cold weather, influence the properties of the
oil.
Generally the specific gravity lies between 0.856 and 0.861, but
0.854 has been observed as a lower limit in connection with pure oils.
The optical rotation,1 as a rule, fluctuates between +57° and -\-6i°.
However, certain districts, more particularly in the western part of
the island, produce oils with an angle of rotation of but +56°. In
one instance Chace observed this to be as low as 54.16°. In other
districts the angle of rotation rises to +67°.
The evaporation residue, the determination of which is discussed
under bergamot oil, varies between 2.1 and 4 per cent. However, in
oils obtained by the machine process it is higher and rises to 5 (Berte
and Romeo) and even 6.6 per cent. Acid value of the evaporation
residue 19 to 39; ester value 100 to 214. However, the ester value of
small amounts of evaporated oil (5 to 10 gm.) does not drop below
150, whereas that of the residue of larger amounts of oil has been ob-
served as low as 100. This apparently is due to the resinification be-
cause of the longer period of evaporation.
The index of refraction nD20o lies between 1.474 and 1.476.
Solubility: because of the mucilaginous and waxlike substances
which accompany the citrapene (see under Composition, p. 28) the
oil is mostly not perfectly soluble in (6 to 8 volumes of) 90 per cent
alcohol. However, it forms a clear solution with 0.5 to I volume of
95 per cent alcohol and with every proportion of absolute alcohol, ether,
chloroform, benzene, and amyl alcohol. Because of the slight water
*It is necessary to determine the angle of rotation at 20° or to compute it with
reference to this temperature. If the determination is made at a temperature
below 20°, 9 minutes should be deducted for every degree below 20° ; if made
above 20°, 8.2 minutes should be added for every degree in temperature.
PROPERTIES OF TERPENE-FREE LEMON OIL
content of the oil, its solutions in carbon disulphide and benzin are
mostly turbid.
The citral content of lemon oil varies from 3.5 to 4 and even 5 per
cent when determined according to Kleber's method. (For details see
p. 64 under Examination.)
As to the optical rotation of the first 10 per cent of fractionated
distillate, see p. 29.
Effects of storage. — Like all volatile oils obtained by expression,
lemon oil deposits a more or less crystalline sediment upon standing.
Both air and light cause rapid changes: the oil loses its color and
a viscous, brown mass is deposited. At the same time the specific
gravity as well as the solubility in 90 per cent alcohol is increased.
These changes correspond to those observed in connection with old
turpentine oil that has been stored improperly. Hence, lemon oil
should be kept in well-filled, closely stoppered containers, stored away
from light and in a cool place.
PROPERTIES OF TERPENE-FREE LEMON OIL
As pointed out on page 66 the methods by which the terpene-free
lemon oils (Boecker) are prepared by the several manufacturers
differ greatly, hence the products differ accordingly. It is, therefore,
impossible to establish limit values. Moreover, a distinction is made
between those deprived merely of their terpenes and those deprived of
both terpenes and sesquiterpenes.
In connection with a number of such concentrated oils, Parry (1913)
has observed the limit values given in Table I.
TABLE I
In connection with two terpene- and sesquiterpene-free oils made by
himself, Boecker observed the following constants: d150 0.8951 and 0.8971;
— 7°io/ and — 5°4o'; the one oil required 1.4 volume of 80 per cent
alcohol to form a clear solution, the other 0.9 volume.
For the determination of the hydrocarbon and citral contents of con-
centrated oils see page 33.
24 CITRUS PRODUCTS
PROPERTIES OF OTHER SPECIAL OILS
An authentic sample of oil prepared by the machine process had the
following properties: color green; d150 0.8630; oD20o -|-5805/; of the first
10 per cent aD2o<> -f"56°34'; nD20o 1.47695; evaporation residue 6.6 per cent
with an acid value of 25.5 and an ester value of 136.7; citral content 6
per cent.1
The constants of a Spanish lemon oil were as follows: d16, 0.862;
oD2o« +63°32'; oD2o<. of the first 10 per cent -}-67°io/.
In connection with a California oil2 the following constants were
found: d15« 0.8598; 00200 +53°56/; aD of the first 10 per cent -f-48°42';
nD200 1.47490; evaporation residue 3.6 per cent.
The color of a concrete oleoresin of lemon was dark olive; d1BO 0.8730
to 0.8836; aD2oo +57°3<>/ to 4-6o°3o'; aD of the first 10 per cent 4-54° n'
to -f-69°4°'; evaporation residue 14.2 to 15.4 per cent; acid value of the
evaporation residue 25.4; ester value 102.1 ; citral content 6 per cent.
Properties of lemon oil terpenes : d150 0.852 to 0.854 ; ao2o<> -f-62° to
-(-72°; nD20o 1-473 to M755 soluble in 6 to 7 volumes or more of 90 per
cent alcohol and in 1.7 volumes or more of 95 per cent alcohol.
Composition. — Although lemon oil has long been known, it required
a relatively long time to acquire an even superficial insight into this
complex mixture. The study of its numerous components was not
rendered more easy by the presence of non-volatile constituents since
their removal, even by steam distillation, could not be effected without
practical decomposition of the volatile constituents.
A critical study of the extensive literature,3 which cannot be taken
up in detail, reveals the fact that some of the oils examined formerly
were adulterated with turpentine oil. This is not surprising for it has
only recently been demonstrated that lemon oil should contain, at most,
traces of pinene.
The early analyses of the oil revealed its low oxygen content. Indi-
vidual chemists even went so far to declare it to be free from oxygen.
This error resulted from the practice of using rectified oils for investi-
gation, in the preparation of which the oxygenated constituents re-
mained in the residue.
Arranged according to their boiling-points, the following constitu-
ents have thus far been observed: (i) octylene, (2) a-pinene, (3)
camphene, (4) yS-pinene, (5) yS-phellandrene, (6) methylheptenone,
(7) y-terpinene, (8) rf-limonene, (9) octylic and nonylic aldehydes,
(10) citronellal, (n) a-terpineol, (12) citral, (13) linalyl acetate,
'Unless otherwise stated the citral content is determined by the Kleber method.
"Report of Schimmel & Co. (October, 1905), p. 28.
"See under Saussure, Dumas, Blanchet and Sell, Soubeiran and Capitaine,
Gerhardt, Berthelot, Oppenheim, Lafont, in the References at end of the chapter,
page 67.
PROPERTIES OF OTHER SPECIAL OILS 25
(14) geranyl acetate, (15) bisabolene, (16) cadinene, (17) acids, (18)
citroptene, citraptene, or lemon camphor.
1. Octylene (?). In connection with the study of the more volatile
constituents of large quantities of lemon oil, Burgess and Page (1904)
isolated a hydrocarbon possessing the following properties : d 0.7275 ;
aD±o°; nDl50 1.4066; boiling-point 123° to 124° (768 mm.); molecular
refraction 38.54. Molecular weight determination and elementary analysis
yielded values corresponding with the formula C8H18 or C8H16. Inasmuch
as the hydrocarbon yielded butyric acid upon oxidation with potassium
permanganate, Burgess and Page concluded that it is octylene and regard
it as a normal constituent of the oil.
2. a-Pinene. At times this terpene seems to be wanting entirely, at
times it is present in such small amounts that its detection is coupled with
difficulty. On account of the importance as to whether pinene is a natural
constituent of lemon oil, Schimmel1 in 1897 endeavored to isolate the hydro-
carbon by careful fractionation of 50 kg. of lemon oil in vacuum. However,
they obtained but 0.016 per cent distilling below 170°. Inasmuch as this
fraction did not possess the properties of pinene and since a second experi-
ment yielded like results, the chemists of this firm denied its presence.
From a corresponding laevogyrate fraction Burgess and Child (1901) later
obtained a hydrochloride addition product that melted at 124°, hence con-
cluded the presence of /-pinene. A further investigation of the subject2
led to the identification of a very insignificant amount of pinene by means
of its nitrol benzylamide. Still later Schimmel3 examined thirty-six samples
of lemon oil obtained from reliable sources and found traces of pinene in
each.
From these investigations the conclusion may be drawn that pinene
frequently occurs in minimal amounts in normal lemon oil, but that occa-
sionally it is completely wanting. Under what conditions the lemon tree
produces pinene and under what conditions it produces oil free from pinene,
has not been ascertained.
A method worked out by Chace for the detection of minimal amounts
of pinene in lemon oil is described in detail on page 30.
3. Camphene. Schimmel4 having expressed the opinion that camphene
might be contained in lemon oil, its presence was demonstrated by Burgess
and Child (1903) by the preparation of t-yo-borneol from the corresponding
fraction.
4. /J-Pinene. From fraction 165° to 168° Gildemeister and Muller5
obtained, upon oxidation, nopinic acid (melting-point 126°) and nopinone
(melting-point of semicarbazone 188°) thus proving the presence of
y8-pinene in the oil.
1Report of Schimmel & Co. (April, 1897), p. 19; (October, 1897), p. 22.
*Ibid. (April, 1902), p. 32; (October, 1902), p. 38.
3 Ibid. (October, 1908), p. 50.
'Ibid. (October, 1902), p. 39.
*Wallach-Festschrift, p. 441. Gottingen, 1909; also Report of Schimmel & Co.
(October, 1909)^ p. 63.
26 CITRUS PRODUCTS
5. yg-Phellandrene. This was first detected by Schimmel1 and identi-
fied by its nitrite melting at 102°. This was verified by Gildemeister and
Miiller2 who, upon oxidation with dilute permanganate solution, obtained a
liquid glycol which, when boiled with dilute hydrochloric acid, yielded a
hydrocuminic aldehyde the semicarbazone of which melted at 203° to 204°.
6. Methylheptenone. If the mixture of aldehydes and ketones, sep-
arated from the oil by means of bisulphite, be fractionated, methyl hep-
tenone can be identified in fraction 174° to 179° by means of elementary
analysis and the semicarbazone melting at 136° to I37°.3
7. y-Terpinene. Upon oxidation of the hydrocarbon boiling above
173°, Gildemeister and Miiller4 obtained the erythritol of y-terpinene
(melting-point 237°), thereby proving for the first time the presence of
this terpene in a volatile oil.
8. d-Limonene. This is quantitatively the principal constituent of the
oil (melting-point of tetrabromide 104° to IO5°).5
Tilden pointed out that the limonene from lemon oil is much less pure
than that obtained from oil of orange. Upon oxidation of the limonene
fraction, designated by him as citrene, he obtained paratoluic acid and
terephthalic acid, two acids which do not result upon the oxidation of the
corresponding fraction of orange oil. Citrene also behaves differently from
limonene when treated with concentrated sulphuric acid. As with limonene,
resinification, due to the formation of polymeric products, results. In
addition, however, cymene, C10H14, can be identified among the products
readily volatile with water vapor. Naturally the presence of cymene after
the application of so energetic a reagent as sulphuric acid cannot be re-
garded as proof of its presence in the original oil. It is more likely that
the cymene resulted from the phellandrene or from the y-terpinene dis-
covered much later. As a matter of fact neither cymine nor pseudocumene6
have thus far been found in lemon oil not previously treated with sul-
phuric acid.T
9. Octylic and nonylic aldehydes. According to Burgess (1901) lemon
oil contains an aldehyde with a cocoanut-like odor that boils between 80°
and 85° (15 mm.). Von Soden and Rojahn have shown that there are, in
reality, two aldehydes. Judging from the semi-carbazones obtained, they
are of the opinion that the mixture consists of much nonylic aldehyde with
little octylic aldehyde.
10. Citronellal was found by Doebner in oil of lemon. When con-
densed with pyruvic acid and ^-naphthylamine, citronellal-yff-naphthocin-
* Re port of Schimmel & Co. (October, 1897), p. 26.
'Op. cit., p. 441.
* Report of Schimmel &• Co. (October, 1902), p. 39.
4O/>. cit., p. 443.
"Wallach, Liebigs Annalen der Chemie, CCXXVII (1885), 290.
"After treating citrene with sulphuric acid, G. Bouchardat and J. Lafont
(Journal de pharmacie et de chemie, Vol. XXVII, Series 5 [1893], p. 49) found
both cymene and pseudocymene in the reaction product. They seemed to regard
both as pre-existing in the oil of lemon.
'Report of Schimmel & Co. (October, 1897), p. 26. Gildemeister and Miiller,
op. cit., p. 451.
PROPERTIES OF OTHER SPECIAL OILS 27
choninic acid, melting-point 225°, results. Its presence in lemon oil being
thus established, von Soden and Rojahn later verified this by the prepara-
tion of the semicarbazone (melting-point 83° to 84°) from the aldehyde
fraction boiling in the neighborhood of 200°.
Apparently the earlier statement of Ladell, who examined a terpene-
free lemon oil, pertains to citronellal. By means of fractionation he isolated
a dextrogyrate substance boiling at 206° and having the composition
CaoH18O.
As a constituent of lemon oil, Tilden describes a substance that boils
slightly above 200° and has the formula CieH18O. With the exception of
the optical rotation its principal properties corresponded with those of
terpineol. Like Ladell, Tilden probably worked with a mixture of terpineol
and citronellal. Barbier and Bouveault, also Burgess and Child (1902)
deny the occurrence of citronellal in oil of lemon.
11. a-terpineol has been found in the higher boiling fractions which
had been freed from aldehydes. It was identified by means of its phenyl-
urethane melting at no0.1
12. Citral. From the aroma point of view, citral, C10H16O, is the
principal constituent of the oil. This aldehyde was discovered in 1888 by
Bertram in oil of lemon.2 The amount of this aldehyde varies between
3.5 and 5 per cent (Kleber's method). For its assay see pages 57-65.
13 and 14. Linalyl and geranyl acetates. According to Umney and
Swinton the lemon oils of Messina and Palermo contain geranyl acetate.
From so-called concentrated oils, i.e., from the high-boiling fractions rich
in oxygen, they removed the aldehydes with hot bisulphite solution and
saponified the non-aldehyde constituents. From the alkaline solution they
separated acetic acid, and from the oil a fraction which formed a solid
derivative with calcium chloride and which upon oxidation yielded citral.
Hence they regard it as geraniol, which occurs in the original oil as
acetate. From Palermo lemon oil they isolated, in addition to geraniol,
a fraction having the properties of /-linalool. Umney and Swinton are of
the opinion that the differences in the odor of the lemon oils from Palermo
and Messina are due not only to the differences in the ratio between citral
and citronellal of the two oils, but also to the presence of linalyl acetate
in the Palermo oil.
15. Bisabolene. The occurrence of a sesquiterpene in lemon oil was
first pointed out by Oliveri. Schimmel3 determined the constants of this
hydrocarbon and pointed out its remarkably low specific gravity. Burgess
and Page obtained a hydrochloride melting at 79° to 80° and established
its identity with the limene previously obtained by them from oil of limes.
The identity of both sesquiterpenes with the known bisabolene, previously
isolated from Bisabol myrrh (Tucholka) was established by Gildemeister
and Miiller.4
16. Cadinene. Bisabolene is accompanied by a higher boiling sesqui-
terpene which likewise yields a hydrochloride. On account of the small
^Report of Schimmel & Co. (October, 1902), p. 40.
"Ibid. (October, 1888), p. 17.
* Report of Schimmel & Co. (October, 1903), p. 26.
*Op. cit., p. 448, and Report of Schimmel & Co. (October, 1909), p. 64.
28 CITRUS PRODUCTS
amount available, its melting-point could by recrystallization not be brought
higher than 110° to 113°, but all of its properties suggested cadinene.
17. Acids. According to von Soden and Rojahn acids can be re-
moved from lemon oil by shaking it with sodium bicarbonate solution.
When the solution of sodium salts is acidulated with sulphuric acid and
shaken with ether a small amount of acids is obtained, which are in part
volatile, in part non-volatile. The latter shows a light-blue fluorescence
and dissolves in alkali carbonates with a beautiful light-blue color.
Parry (1900) is of the opinion that traces of the methyl ester of
anthranilic acid are present in the oil.
18. Citroptene, citraptene, or lemon camphor. Upon prolonged stand-
ing the expressed lemon oil separates a waxy, soft, more or less slimy
mass, which likewise remains as a yellowish-brown residue when the oil
is rectified or evaporated. Concerning its composition there exists an ex-
tensive literature.1 Schmidt, however, first succeeded in revealing the con-
stitution of citroptene which is identical with limettin. Upon treating the
distillation residues of lemon oil with ether, citroptene is obtained as
granular, crystalline mass. It is insoluble in ether and after repeated
crystallization from acetone and methyl alcohol, and finally from dilute
alcohol to which animal charcoal has been added, it is obtained as shiny,
colorless needles that melt at 146° to 147°. The solutions reveal a violet-
blue fluorescence. Analysis reveals the composition C11H10O4 and meth-
oxyl determinations, the presence of two methoxyl groups. Fusion with
potassium hydroxide yields phloroglucinol and acetic acid. In chloro-
form solution citroptene combines with bromine to form a dibromide
C11H10Br2O4, that melts between 250° and 260°. The supposition that
citroptene is a methylated dihydroxy cumarin was confirmed by synthesis.
Starting with phloroglucinol, Schmidt converted this into phloroglucin
aldehyde and from this, upon cumarin synthesis, a dihydroxy cumarin
which, upon methylation, yielded a compound corresponding in its proper-
ties with citroptene. It melted at 146° to 147°. Hence citroptene is
represented by the following formula.
OCH.
t>-CO
In addition to citroptene and other compounds, Schmidt found in the
lemon oil residues a phenol melting at 89° which dissolves in sulphuric
acid with a deep-red color but does not react with ferric chloride.
Examination. — The examination of lemon oil as to its purity is one
of the most difficult tasks of the analytical chemist. As has already been
pointed out, the properties of the oils vary considerably with the districts
in which they are produced, also with the periods of the year in which the
fruit is harvested. To these difficulties there should be added the uncer-
*See under Mulder, Tilden and Beck, Crismer, Theulier, Burgess, in Refer-
ences at end of chapter, p. 67.
PROPERTIES OF OTHER SPECIAL OILS 29
tainty to detect the addition of hydrocarbons which result upon the produc-
tion of terpene-free oils, provided this addition is restricted to certain limits.
A further difficulty exists in the presence of non-volatile substances which
interfere with the isolation and assay of such constituents as citral. Indeed,
the citral assay of lemon oil is one of the most difficult of tasks. Because
of the low citral content of the oil, its assay must be extremely exact in
order to utilize the results as a means of detecting the additions of non-
aldehydic substances. It should further be remembered that adulterators
can readily bring the citral content of an adulterated oil up to the required
standard.
Gross adulterations can be detected by the determination of the spe-
cific gravity and angle of rotation. Important conclusions can also fre-
quently be drawn from the amount of evaporation residue. Adulterations
with turpentine oil can be detected in most instances by a comparison of
the optical rotation of the first 10 per cent of distillate with that of the
original oil. The solubility test according to G. Patane (cf. p. 31) has
not yet been corroborated sufficiently by experience to judge its value.
Neither has the viscosity determination (see p. 32) found its way into
practice.
THE DISTILLATION TEST
The distillation test is used principally for the detection of turpentine.
In spite of the addition of turpentine oil, an adulterated oil of lemon may
reveal a normal optical rotation provided the reduced rotation is compen-
sated by the addition of orange oil. Such an adulteration may, however,
be detected by the optical examination of special fractions of the oil. In
the case of pure oils, the angle of rotation of the first 10 per cent of
distillate is but 5° to 6° lower than that of the original oil.1 For oils to
which turpentine oil has been added the difference is greater. For this
distillation test a Ladenburg fractionating flask with three bulbs is used.
From 50 c.c. of the oil to be examined exactly 5 c.c. are distilled. A few
drops of water pass over first and render the distillate turbid. These can
be removed by shaking with anhydrous sodium sulphate. After filtration
the distillate is examined in a 50 mm. polariscope tube, the temperature
having been observed accurately. The result is computed for 20° in
accordance with the directions given in footnote i on page 22 and de-
ducted from the result obtained for the original oil likewise computed
for 20 °.2
In certain instances, more particularly when the amount of available
oil is but small, the modification of Soldaini and Berte is preferable. Thiif
consists in distilling off one-half from 25 c.c. of oil. In the case of
pure oils, the optical rotation of the 50 per cent of distillate is higher
than that of the original oil, likewise than that of the residue. In the
case of adulteration with turpentine oil, that of the first half is invariably
lower.
This modification, however, is no better than the original Schimmel
test which, in an emergency, can be conducted with 25 c.c. of oil. For
1In one instance, Chace observed as high as 6.17°.
"Report of Schimmel & Co. (October, 1896), p. 39.
CITRUS PRODUCTS
the detection of the addition of lemon-oil terpenes both methods fail.1
The suggestion made by Burgess to distil under diminished pressure does
not afford any particular advantages.
DETECTION OF PINENE ACCORDING TO CHACE
Assuming that pure turpentine oil is at most to contain but traces of
pinene, Chace2 published a method for the detection of mere traces of
turpentine oil. It is based on the ob-
servation that the nitrosochloride of
pinene and limonene crystallize in dif-
ferent forms readily recognizable
under the microscope. Whereas the
pinene derivative separates in laminae,
the limonene derivative crystallizes in
needles.
Chace proceeds in the following
manner : From 50 c.c. of oil contained
in a Ladenburg flask, he fractionates
5 c.c. From this, with the aid of ethyl
nitrite, he prepares the nitrosochloride
according to Wallach's method. The
reaction mixture is allowed to remain
fifteen minutes in the freezing mix-
ture when the crystals are removed by
means of a suction filter, and washed
(E. M. Chace)
FIG. 6. — Limonene nitrosochloride
crystals from lemon oil.
with 50 c.c. of 95 per cent alcohol.
The mother-liquid is exposed another
fifteen to twenty minutes to the tem-
perature of the freezing mixture. The
second crop of crystals is united with
the first. The united crystals are
washed well with alcohol, dried, and
dissolved in the smallest possible
amount of chloroform. Enough hot
methyl alcohol is then added to result
in crystallization upon cooling. Finally
somewhat more methyl alcohol is added
and the crystals removed by filtration.
For testing under the microscope, the
crystals are best mounted in olive oil.
In this manner Chace could detect an
addition of 2 per cent of turpentine
oil whereas with the aid of a three-
(E. M. Chace)
FIG. 7. — Limonene and pinene ni-
trosochloride crystals from a lemon
oil mixed with 5 per cent of turpen-
tine.
*Cf. also Berte, Boll. Chim. Farn. (May, 1914), No. 10; Report of Schimmel
6- Co. (October, 1904), p. 29. Chemist and Druggist, LXVI (1905), 713.
'Journal of the American Chemical Society, XXX (1908), 1475; Report of
Schimmel & Co. (October, 1908), p. 64. "The Occurrence of Pinene in Lemon
Oil," Bureau of Chemistry, United States Department of Agriculture Circular 46
(October, 1909).
THE DISTILLATION TEST
bulb Glinsky fractionating column
even 0.5 per cent of added turpentine
oil could be detected. Additions of 10
to 15 per cent could be detected by pre-
paring the nitrosochloride from the
original oil.
Both Umney and Parry have pro-
tested against the usefulness and ap-
plicability of this method. Wiley,
however, justifies it by the claim that
since its application by the Bureau
of Chemistry of the United States
Department of Agriculture not a
single imported oil has had to be
questioned.
It is necessary, however, to fol-
low the directions. In the case of the
isolation of but small amounts of
pinene nitrosochloride, the oil should be allowed to pass and questioned only
when larger amounts are obtained.
(E. M. Chace)
FIG. 8. — P i n e n e nitrosochloride
crystals from turpentine.
PATANE S METHOD FOR DETERMINING THE AMOUNT OF THE
CONSTITUENTS SOLUBLE IN DILUTE ALCOHOL
As is well known, lemon oil consists largely of terpenes and sesqui-
terpenes which are of but little value so far as the odor is concerned.
Whereas these hydrocarbons are all but insoluble in 80 per cent alcohol,
the valuable odoriferous constituents are soluble therein. Hence, Patane
regards the determination of the solubility as a valuable criterion in judging
an oil and has based thereon a method of examination.
Indeed, he ernploys two methods. The one consists in shaking thor-
oughly equal volumes of oil and alcohol of definite strength measured at
exactly 20° in a 10 c.c. cylinder graduated into o.i c.c. After the two
layers have separated, the increase in the volume of the alcoholic solution
is noted. According to the other method, equal volumes of the oil and
alcohol of definite strength are mixed in a test tube, and gently heated
until solution has taken place. The solution is then allowed to cool, being
constantly stirred with a thermometer, the scale of which is divided into
tenths of degrees. As soon as the solution becomes turbid, the tempera-
ture is read off. Differences in temperature of one-tenth degree produce
turbidity. All oils that reveal the same turbidity temperature show the
same solubility according to the first method, hence a comparative scale
between turbidity temperature and solubility can be prepared. In most
instances it suffices to carry out the second test which is said to be rapidly
executed and yet exact.
The addition of 10 per cent of terpenes increases the turbidity tem-
perature by more than one degree, that of 20 per cent about two degrees,
etc., with this difference, however, that with the increasing terpene con-
tent, the temperature intervals become smaller. In addition, the relation
32 CITRUS PRODUCTS
between citral content and solubility is said to admit of further conclusion
as to the nature of the oil.
The alcohol used for this purpose must be standardized very carefully
for the turbidity temperature varies with a difference in the strength of
the alcohol of one-tenth of a degree. According to Patane it is more
expedient to use methyl alcohol in place of ethyl alcohol, since the oil is
less soluble in the former. Inasmuch as the turbidity temperature in this
instance lies above room temperature it is not necessary to cool the solu-
tion artificially as has to be done when ethyl alcohol of 93 to 94 per cent
strength is used.
Both tests can also be applied advantageously to the oils of orange
and mandarin which are much less soluble than lemon oil.
In the determination of the solubility it is necessary to consider the
geographic source of the oil since oils from different localities vary in
composition.
In order to judge the usefulness of this method more detailed experi-
ments are necessary.
VISCOSITY DETERMINATION
According to Dowzard the viscosity of the volatile oils is said to be
an important aid in the determination of the purity or quality of volatile
oils. Inasmuch as it seemed highly desirable to increase the means of
ascertaining the value of lemon oil, a number of viscosity determinations
were carried out in the laboratory of Schimmel.1 These are recorded in
Table II. The apparatus used was the same employed by Dowzard, viz.,
the viscosimeter of Reischauer. As unit of viscosity the time of flow of
25 c.c. of water at 20° is indicated as 100. The viscosity number of the
lemon oils tested is computed with the aid of the following formula:
O
V = — X ioo
W
where 0 = the number of seconds required by the flow of 25 c.c. of oil
at 20 °, and W = the number of seconds required by the flow of 25 c.c.
of water at 20°.
So far as lemon oil is concerned, the experiments yielded no encour-
aging results. This is more particularly true of the addition of "citrene,"
for the values for "citrene" and lemon oil are too close to admit of the
detection of this adulterant. Moreover, the difficulty of establishing
standards is increased by the circumstance that different viscosimeters of
the same make do not yield the same results, hence are not comparable.
Whereas Dowzard found the viscosity number for pure lemon oil to be
139.6, the values found by Schimmel varied between 109.8 and 122.9. For
"citrene" Dowzard found 105.8, Schimmel found 103.2.
The cause for these differences is to be looked for in the fact that the
orifices of viscosimeter of this type are different. It may be possible to
secure better results with more exact instruments that guard against this
error. Quercigh and Moreschini regard the viscosimeter of Scarpa as
^Report of Schimmel &• Co. (April, 1901), p. 32.
DETERMINATION OF THE HYDROCARBON CONTENT
33
suited for the determination of the viscosity of volatile oils. Another
advantage of this instrument lies in the fact that smaller amounts of oil
(T to 5 c.c.) suffice for the determination of exact results.
The results obtained by Schimmel & Co. are recorded in Table II.
DETERMINATION OF THE HYDROCARBON CONTENT OF CONCENTRATED
LEMON OILS ACCORDING TO E. BOCKER
This method consists in the removal, first of the citral by means of
sodium sulphite, and then of the other aromatic substances by means of
alcohol. The strength of the latter is chosen so that, with the observa-
tion of certain precautions, the hydrocarbons remain undissolved almost
entirely. Bocker proceeds in the following manner:
The citral content of 10 c.c. of oil is first examined according to the
sulphite method. If the volume of oil that does not enter into the reaction
amounts to less than 6 c.c. the assay is repeated with either 5 or 10 c.c.
TABLE II
The non-aldehyde constituents of both assays are united and 5 c.c. thereof
transferred to a 600 to 700 c.c. separating funnel into which 500 c.c. of 51
per cent alcohol (by volume), cooled to from o to — 2°, have previously
been placed. The separating funnel is stoppered with a cork and the con-
tents are shaken repeatedly thus causing the aromatic constituents to go
into solution, whereas the hydrocarbons remain undissolved almost quanti-
tatively. The inverted separating funnel (stopper downward) is then
transferred to a bath of o° in which it is allowed to remain for eight to
34 CITRUS PRODUCTS
ten hours. After this time it is removed, carefully brought back to its
normal position and suspended from a ring stand. After the alcoholic
solution has become so clear that but a very faint turbidity remains, which
may require up to two days, all but about 10 c.c. of the liquid are drawn
off. Any oil drops adhering to the separating funnel are washed down with
ice-cold 50 per cent (no doubt should read 51 per cent) alcohol so that all
dissolved oil is brought together. The separating funnel is again set aside
until the alcohol has become well-nigh perfectly clear. After the last
traces of alcohol have been separated so far as this is possible, the oil is
transferred to a measuring cylinder, graduated into one-tenth cubic centi-
meters, the separating funnel being again rinsed with some ice-cold 51
per cent alcohol. It is more practicable to use a separating funnel, the
lower portion of which is drawn out to a narrow, calibrated tube. As soon
as the oil has become clear (in case foam persists a few drops of dilute
acetic acid are added), its volume is read off and the percentage with refer-
ence to the original oil computed.
If larger amounts of material are available, the quantitative determina-
tion can be supplemented by a qualitative examination of the separated
hydrocarbons. For this purpose Bocker provides the following directions :
100 to 200 c.c. of oil are fractionated in vacuum. The distillation is con-
tinued only so far as the fractions of 10 c.c. each reveal dextrorotation.
These are united as the terpene-containing portion of the oil whereas the
remainder constitutes the sesquiterpene-containing portion of the oil. Both
of these portions are now deprived of citral by means of sodium sulphite.
The residual oils are treated separately as described above, with 100 times
their volume of ice-cold 51 per cent alcohol. If desired, glass bottles can
be used in the place of the separating funnels, the alcohol being removed
by means of a siphon. The separated oils are measured and the terpene
and sesquiterpene contents of the original oil computed. For further iden-
tification the optical rotation is ascertained and the characteristic deriva-
tives are prepared. The terpene fraction on the one hand, consisting
principally of d-limonene, is characterized by strong dextrorotation, and
affords a good yield of limonene tetrabromide. The sesquiterpene fraction,
consisting principally of bisabolene, is laevogyrate and can be further
characterized by the bisabolene trihydrochloride.
Bocker has tested his method in connection with a number of experi-
mental mixtures and obtained good results. In connection with a terpene-
free lemon oil produced by himself, he obtained 51.5 per cent of citral and
21 per cent of hydrocarbons. In connection with a terpene- and sesqui-
terpene-free oil, also produced by himself, he found 62 per cent of citral
and traces of hydrocarbons too small in amount to be measured. In
terpene-free lemon oils he regards a 25 to 30 per cent hydrocarbon content
as normal.
A comparison of the citral and hydrocarbon contents is said to enable
the detection of the addition of lemon-grass citral to a concentrated oil.
Tests made with two oils, to which 20 per cent of citral had been added,
revealed that the exact amount of the addition could not be proved, but
that only one-half thereof was indicated. Nevertheless, the minimum of
adulteration can be recognized.
DETERMINATION OF THE HYDROCARBON CONTENT
35
For the purpose of valuation, Bocker proceeds from the consideration
that a lemon oil, freed entirely of its hydrocarbons, contains a maximum
of 66 per cent citral. This leads to the results in Table III.
TABLE III
This table can be supplemented as desired, since for every i per cent of
hydrocarbon 0.66 per cent citral should be deducted. Hence, in passing
judgment on a concentrated oil, Table III should be consulted to ascertain
whether the hydrocarbon content agrees with the citral content. If the oil
in question contains more citral than is usual according to this table, this
may be regarded as adulteration with added citral. In order to ascertain
the minimum amount of added citral, the maximum citral content cor-
responding to the hydrocarbon content, as shown in the table, is noted.
This amount of natural citral is deducted from the citral found by the
sulphite assay. The difference stands for the minimum amount of citral
added.
The details of the computation can be explained better with the aid of
data supplied by Bocker. Having ascertained the presence of 15 per cent
of hydrocarbon and 61 per cent of citral, Table III reveals that these values
do not correspond, since according to Bocker the citral content of an oil
containing 15 per cent of hydrocarbon should not exceed 56 per cent.
Hence citral from another source must have been added. Inasmuch as a
normal with 56 per cent citral should contain 100 — 56 = 44 per cent of
non-citral, and since the examined oil contains only 100 — 61 =39 per cent
of non-citral, the normal citral content of the latter is revealed by the
following equation : 44 • 56 = 39 • x ; when x = 49.63. This number cor-
responds to the highest lemon oil citral content of an oil containing 39
per cent of non-citral. Inasmuch, however, as 61 per cent have been found,
an adulteration with at least 61 — 49.63 = H-37 per cent of foreign citral
may be assumed.
Whether this method is of general application to all concentrated oils
produced by different methods can be ascertained only by experiment. It
is noteworthy that, according to the method of production, other constitu-
ents than the hydrocarbons may be removed from the original oil. This
may result in an increased citral content of the concentrated oil. Hence, a
high citral content may not invariably indicate adulteration.
36 CITRUS PRODUCTS
ADULTERANTS
Twenty-five years ago, when the constants of pure lemon oil were not
known and when the polariscope was not utilized in its examination,
French and American turpentine oil1 were the common adulterants. After
the introduction of the polariscope method, the adulterators took recourse
to mixtures of turpentine oil with carvene, orange oil terpenes, or distilled
lemon oil. Later, when the production of terpene-free lemon oil came into
vogue, the citrene, resulting as waste product, became the most favored
and dangerous adulterant, for its detection, like that of citral from lemon-
grass oil, is exceedingly difficult. Of adulterants that can be detected
readily, the following have been observed : cedarwood oil, stearin,2 mineral
oils, alcohol,3 castor oil4 and finally mixtures of alcohol and glyceryl
acetate.6
OIL OF LIMES
Two oils coming from different plants (Citrus aurantifolia
(Christm.) Swingle and C. limetta Risso) and of entirely different
properties are designated by the common name of limette oil,6 but
according to their source may be called West Indian and Italian limette
oils.
They are obtained either by an expression process (generally by
the ecuelle) or by distillation. The pressed oil is the superior of the
two and commands a much higher price.
WEST INDIAN LIMETTE OIL
The West Indian oil, which is the usual oil of commerce, is ob-
tained from the fruits of Citrus aurantifolia,7 the juice of which contains
a large quantity of citric acid. The plant is plentiful in Jamaica, Do-
minica, and Tahite; but the most important plantations are on the
island of Montserrat, one of the Antilles. The lime harvest there lasts
from September to January, and the chief product is, of course, the
lime juice.
The oil obtained from the peel of the fruit by expression (oil of
limette) is of a golden-yellow color and can hardly be distinguished
from a good lemon oil by its odor, if the8 greater intensity of that of
'More recently the strongly dextrogyrate Greek oil is being used for this pur-
pose. E. J. Parry, Perfumery Record, II (1911), 209.
*G. Boswigi, Chemist and Druggist, LV (1899), 710.
^Report of Schimmel & Co. (April, 1008), p. 45.
*E. J. Parry, Chemist and Druggist, LXXIV (1909), 121.
'Report of Schimmel & Co. (April, 1913), p. 61.
•Archiv der Pharmasie, CXXXIII, 174.
''Bulletin of Miscellaneous Information, Royal Gardens Kew (1894), p. 113.
* Pharmaceutical Journal, III, No. 15, 322.
WEST INDIAN LIMETTE OIL 37
the limette oil is not considered. Specific gravity 0.873 at 29° to 0.882
at 15° ; aD +35° to + 38°. The most important constituent of the oil
is citral.
Entirely different from the expressed oil is the distilled oil which is
obtained as a by-product in the evaporation of the juice and is known
in commerce under the name of oil of limes. Its odor is unpleasant,
terebinthinate, and no longer reminds of citral. Probably this alde-
hyde is completely destroyed by the boiling of the acid liquid.
Tempany and Greenhalgh, who have investigated the matter, are of
opinion that the difference is caused by the loss, during the process of
distillation, of part of the lowest and highest boiling fractions. They dis-
tilled hand-expressed oils with steam, and obtained an oil with the char-
acteristic turpentine-like odor of the distilled oil of commerce. The highest
boiling fractions of the hand-expressed oil contain a blue fluorescent,
crystalline body (perhaps methyl anthranilate), which is absent from the
distilled oil. Moreover, limettin, which ordinarily separates out from the
hand-expressed oil when it is left standing, is absent from the distilled oil.
As a rule the citral content of the distilled oil is lower than that of the
hand-expressed oil. The authors mentioned found authentic samples to
show the following properties :
I. Hand-expressed oils: specific gravity, at 30° 0.8712 to 0.8859,
aD3i° -f-3l038' to -|-33043/J specific gravity at 32° 1.4789 to 1.4851, acid
value 1.35 to .8, citral content 2.2 to 6.6 per cent.
II. Distilled oils: d3O° 0.8540 to 0.8858, aD3i° -j-33°O9' to 34°89',
nD320 1.4702 to 1.4713; acid value 0.76 to 1.3, citral content 1.2 per cent
to 2.0 per cent.
ITALIAN LIMETTE OIL1
The fruit of the South European limette, Citrus limetta Risso
(Lima di Spagna dolce, Limettier ordinaire) is distinguished from that
of the West Indian by its sweet juice.
Italian lime oil obtained by expression from the peel, is of a
brownish-yellow color and has a characteristic fragrant odor of the
fruit with a secondary odor of bergamot. It forms a yellow deposit
of limettin in considerable amounts on standing. Its specific gravity
varies from 0.870 to 0.875, its saponification number is 75 and its rota-
tion from +34° to +40°. It contains linalyl acetate and citral. It
also contains a little free linalool, but not more than 3 or 4 per cent.
The bulk of the oil consists of the terpene limonene.
The composition of the Italian limette oil is very similar to that of
bergamot oil, only the limette oil contains more limonene and less
linalyl acetate.
id., Ill, No. 14, 1005 ; Gildemeister.
38 CITRUS PRODUCTS
The limonene is the dextrogyrate modification (an -f-8i°45'; specific
gravity 0.848) and yields a dihydrochloride melting at 50°, as well as a
tetrabromide melting at 105°.
In the oil investigated by Gildemeister1 26.3 per cent of linalyl acetate
were present (boiling-point ioi°— 103° at 13 mm.; specific gravity at 15°
0.898; OD— 9°52')-
After saponification with alkali, acetic acid was found in the alkaline
solution, while from the oil /-linalool (boiling-point 88.3°-89.5° at 13 mm.;
specific gravity at 15° 0.870, (aD — 2o°7') was separated by fractional
distillation. From it citral was formed on oxidation.2 Linalool is present
in the oil partly in the free state, partly as acetic acid ester.
The limettin which separated from the oil on standing melts according
to Tilden at I2i°-i22°. It has the composition C6H3 (OCH3)2 • C8HO2
and yields on melting with potassium hydroxide, besides acetic acid
phloroglucin.
OIL OF MANDARINS
Origin. — The peel of the fruit of Citrus nobilis Loureiro,3 known
as mandarins, contains a very pleasant-smelling oil which is obtained,
like the oils of the other agrumen fruits, by expression.
Properties. — Mandarin oil is a golden-yellow liquid with a slight
bluish fluorescence (probably due to methyl anthranilate), which be-
comes more prominent when the oil is diluted with alcohol. The odor,
although resembling that of lemon oil, is more pleasant and distinctly
different from it. d16» 0.854-0.858 ; <ZD +65° to +75°.
aD of the first 10 per cent of distillate is (cf. under Oil of Lemon, p.
29) slightly lower or but 2 per cent higher than that of the original oil.
According to E. Berte and G. Romeo the ao of the first 50 per cent should
average 3° higher than the original oil; nD20o 1-475 to l-47%> acid valua-
tion up to 1.7; ester valuation 5 to n; ester valuation after acetylation
12.5 (i determination); evaporation residue 2.4 to 3.5 per cent; soluble
in 7 to 10 volumes of 90 per cent alcohol with more or less turbidity.
A Spanish oil from the province of Valencia, which had been expressed
from the immature fruits, had an olive-green color and an agreeable odor,
but less delicate than that of the oil from ripe fruits. Its physical con-
stants also revealed deviations: d150 0.8665; nD2o<> I479°°; acid valuation
0.2; ester valuation 17.3; evaporation residue 8 per cent. On account of
the dark color, the angle of rotation of the original oil could not be ob-
served. The first 10 per cent of the distillate showed OD +55 °12'- In
'The same dihydrochloride was obtained by de Luca in 1860 (Comptes Rendus,
LI, 258) from the terpene boiling at 180° of an oil, which is designated as coming
from Citrus lumia, but which in all probability was Italian limette oil. Cf.
Gildemeister, loc. cit.
*To which compound the "limettsaure" CnH4O« obtained by Vohl in 1853 upon
the oxidation of the oil (Archiv der Pharmasie, CXXIV, 16) owes its origin, is
uncertain.
'According to de Lucca the mandarin is obtained from Citrus bigaradia sinensis
and C. b. myrtifolia.
OIL OF MANDARINS 39
90 per cent alcohol the oil was not completely soluble, but it formed a
clear solution with 0.5 volume and more of 95 per cent alcohol.1
Two mandarin oils2 distilled in Porto Alegra (Brazil) had the follow-
ing properties: d160 0.8515 and 0.8510; OD -j-74°i6' at 17° and -j-75°2o'
at 16°. Both oils were distinguished by a beautiful blue fluorescence.
Composition. — The greater part of distilled mandarin oil consists
of rf-limonene. Citral and citronellal are probably also present. The
constituent to which the characteristic odor as well as the fluorescence
is due is methyl anthranilic ester, present in a quantity of scarcely
I per cent.
The expressed mandarin oil, according to Flatau and Labbe, con-
tains the sarne ester as does orange oil.
(f-Limonene. The oil began to boil at 175° and all except a small
residue went over up to 179°. The fraction boiling at 175°— 177° (OD
-f-76°45') gave on bromination in glacial acetic acid solution a tetra-
bromide melting at 104°— 105° (Gildemeister and Stephan, 1897). By
conducting hydrochloric acid into the same fraction dipentene dihydro-
chloride, melting-point 49°, resulted (de Luca, 1857). According to this
d-limonene was present in large quantity.
Citral and citronellal. If the portion which did not distil over up to
177° be treated with bisulphite solution, an addition product is obtained
from which an oil is separated by alkali. This behaves like a mixture of
citral and citronellal when condensed with pyrotartaric acid and /J-napthy-
lamine. The melting-point of the naphtho-cinchoninic acid formed is not
constant. At 197° (the melting-point of the pure citral compound) the
body begins to run together, but does not melt completely until 222°
(melting-point of the citronellal compound 225°). The positive identifica-
tion of these two aldehydes in the oil has, therefore, not yet been made.
Methyl anthranilic acid methyl ester. This substance was found by
Walbaum in 1900. From 5 kg. of mandarin oil 36 gm. of a base were
obtained by shaking with the sulphuric acid, the bulk of which distilled
between 130° and 131° (13 mm.). Its specific gravity was 1.120 at 15°.
In the cold it congealed and melted at 18.5° to 19.5°. This compound
forms salts and double salts and its odor is similar to that of anthranilic
acid methyl ester. When heated with hydroiodic acid it decomposes with
the formation of methyl iodide. Alcoholic potassium hydroxide saponifies
the ester and from the salt solution acetic acid precipitates the methyl
anthranilic acid.
This acid melts at 179°. When heated with dilute hydrochloric acid
to from 160° to 170° it decomposes into carbonic acid and methyl aniline.
It yields the same derivatives obtained by G. Fortmann from the synthetic
methyl anthranilic acid, e.g., nitrosomethylanthranilic acid (melting-point
128°), acetylmethylanthranilic acid (melting-point 186°) and benzoyl-
methylanthranilic acid (melting-point 161°).
^Report of Schimtnel & Co. (October, 1911), p. 46.
^Report of Schimtnel & Co. (April, 1896), p. 62.
4o
Cixpus PRODUCTS
Synthetically the methyl ester of methylanthranilic acid
CH NH-CH3[i]
4<COOCH3[2]
can be obtained by boiling the methyl alcoholic solution of methylanthra-
nilic acid with sulphuric acid and subsequent decomposition of the ester
salt with soda solution.
Quantitatively (Hesse and Zeitschel) the amount of ester can be deter-
mined as its anthranilic acid methyl ester.
JAPANESE MANDARIN OIL
Three samples of mandarin oil of Japanese origin, are reported upon
in 1914. As they were colorless, it was evident that these oils had not been
obtained by pressure, as is the custom in Italy, but by distillation. Con-
sequently they were devoid of the fine aroma of the Italian oils, but smelled
chiefly of limonene and, therefore, did not compete with genuine mandarin
oils. The method of production is illustrated by the chemical constants,
as specified in Table IV.
TABLE IV
Three oils sent to Schimmel & Co.1 under the name of Mikan oil were
evidently also mandarin oils, for the mandarin fruit is called "Mikan"
(Rein) in Japan. These oils showed somewhat similar constants to the
above-mentioned distillates :
d160 0.8483; aD-f-92°2o'; soluble in 6 volumes and more of 90 per
cent alcohol.
d160 0.8478; aD+90°5^/» soluble in 6.2 volumes and more of 90 per
cent alcohol.
d1B« 0.849 5 aD+9°°35'; soluble in about 10 volumes and more of 90
per cent alcohol.
The values of Italian mandarin oil are within the following figures:
d160 0.854 to 0.859; o,D +65° to -f-75°; aD of the first 10 per cent of the
distillate very little lower, or even up to 2° higher than the original oil
acid value up to 1.7; ester value 5 to n; more or less turbidity when dis-
solved in seven to ten volumes of 90 per cent alcohol.
An English journal (Perfumery Record, V [1914], 721) also reports
on a mandarin oil of Japanese origin. This oil had been obtained by dis-
tillation and was of a very fine odor; it showed a specific gravity of 0.848
and a rotation of +94°. A slight fluorescence noticed in this oil is put
^Report of Schimmel & Co. (October, 1914 — April, 1915), pp. 18-23.
OIL OF SWEET ORANGE 41
down by the author to the presence of methyl ester of anthranilic acid.
We may be permitted to point out that, according to our observations, this
latter body is not present in mandarin oil, but the methyl ester of methyl-
anthranilic acid, which may well be the cause of the fluorescence observed
in the oil in question.
OIL OF SWEET ORANGE
Botanically the sweet orange is Citrus sinensis (L.) Osbeck- Like
lemon oil, this oil is obtained by expression, both in Sicily and Calabria.
In recent years occasional samples have reached Europe from the West
Indies. Likewise, as with lemon, the attempt has been made in Cali-
fornia to obtain a so-called "oleoresin" by means of extraction with
volatile solvents.
Properties. — Oil of orange is a yellow to yellowish-brown liquid of
a characteristic orange odor and a mild, aromatic, not bitter taste.
Specific gravity at 15° 0.848-0.852; OD +96° to 98° at 20°. l
On account of the presence of waxlike, non-volatile substances of
unknown composition, which partly separate on standing for some
time, the oil as a rule does not form a clear solution with 90 per cent
alcohol. It begins to boil at 175° ; up to 180° nine-tenths distil over.
The rectified oil is colorless ; its specific gravity is somewhat lower,
the rotatory power slightly higher than that of the original oil. Recti-
fied orange oil is kept with difficulty, it deteriorates rapidly, and acquires
thereby a stale, unpleasant odor.
The oil yields a clear solution with about one-fourth to one-half
volume of 90 per cent alcohol and the solution does not become turbid
on the addition of more alcohol. All oils do not dissolve clearly in 80
per cent alcohol. Many, and especially those of a high ester content,
often give turbid mixtures, from which fatty globules separate on the
bottom.
The following constants were observed in connection with several oils
from Jamaica: d150 0.8481 to 0.8491; aD20o +97°43' to -(-98° 2'; aD of the
first 10 per cent -|~96°32' to -j-97°3o'; nD2o<> 1.46984 (i determination);
decylic aldehyde 2.3 to 3.8 per cent; evaporation residue 1.4 to 2.0 per cent.
An oil from Dominica2 had the following constants : d150 0.
1Since the angle of rotation of orange oil like that of lemon oil varies greatly
with changes in temperature, decreasing with an increase in temperature, it is
necessary in order to obtain comparable numbers, to ascertain accurately the tem-
perature and to reduce the result to 20° by calculation. As the difference in the
angle of rotation between -f-io° and +20° is 14.5 minutes and between +20° and
-(-30° it is 13.2 minutes for one degree of change in temperature, the reduction to
20° is made by deducting 14.5 minutes for each degree of temperature, when the
polarization was effected at a temperature below 20°. If the determination was
made at a temperature above 20°, 13.2 minutes are to be added to the number
found in order to find the angle of rotation of +20°.
'Examined in the laboratory of Schimmel & Co.
42 CITRUS PRODUCTS
98 °2i/; an of the first 10 per cent 97°4O'; evaporation residue 1.6 per cent,
with saponification value 171.5.
In connection with three samples of oleoresin of orange the following
constants were observed : d150 0.8535 to 0-8723 ; aD2o0 about -(-75° to -\-Sg°
(an approximate determination only was possible because of the dark-
brown color of the preparation), OD of the first 10 per cent -{-90° 54' to
-(-96° 26', evaporation residue n to 14.7 per cent, acid value of the residue
13.5, ester value 107.0 to 112.0.
Properties of the terpenes from orange oil : d150 0.847 to °-854 ;
oD +95° 50' to -{-i 00° 1 8'; nD200 about 1.473; soluble in 8 to 9.5 volumes of
90 per cent alcohol, with partial turbidity.
Composition. — Orange oil contains d-limonene, aldehydes — citral,
citronellal and possibly others of unknown composition — methyl ester of
anthranilic acid and stearoptene.
Wallach (1884) has shown the presence of at least 90 per cent of
rf-limonene (dihydrochloride, melting-point 50° (Soubeiran and Capitaine1)
(1840), tetrabromide, melting-point 104°— 105°). On this account, es-
pecially as other hydrocarbons are completely absent, it is well suited for
the preparation of this terpene in a pure state. The absence of pinene is
of importance for the detection of adulteration with turpentine oil.
Aldehydes. Of oxygenated compounds, aldehydes are present in orange
oil. By shaking with sodium bisulphite solution, crystals of a double com-
pound are formed, which can be isolated by filtration and pressing; by
decomposing with soda an oil is obtained which is purified by steam dis-
tillation. A part of this boils at 224°— 228° and consists of citral (Semm-
ler, 1891). The lower boiling fraction likewise contains an aldehyde, the
composition of which has not yet been determined.
The assertion made by Wright in 1873 that oil of orange peel contains
0.3 per cent of a body boiling at 212°— 218° identical with myristicol
(C10H16O) of the oil of nutmeg, is too little supported by facts.
The most recent investigation of orange oil is that of Flatau and
Labbe. They obtained by shaking orange oil with bisulphite solution a
double compound which yielded besides traces of citronellal, a small amount
of a new aldehyde that had a very characteristic orange color. In addition
an acid of possible twenty-one carbon atoms was isolated. Like the acid,
its ethyl ester is difficultly soluble in alcohol and can be precipitated by this
solvent from the residue of the oil after 95 per cent have been distilled off.
When purified it melts at 64 "-65° and has a pleasant and characteristic
orange odor.
Methyl ester of anthranilic acid. Parry suspects the presence of the
methyl ester of anthranilic acid in sweet orange oil. This observation is
confirmed by Schimmel & Co.2 who have definitely shown its presence.
Stearoptene. Of the nature of the orange oil stearoptene, which finds
its way into the oil by expressing the peel of the fruit and which remains
in the residue when the oil is rectified, nothing is known.
*Cf. also Vochel, Liebigs Annalen der Chcntie, XXXIX (1841), 120; and
Wright and Piesse, Chemical News, XXIV, 147; Chem. Zentralblat (1871), p. 740.
^Report of Schimmel & Co. (April, 1900), p. 18.
OIL OF SWEET ORANGE 43
Examination. — On account of the low specific gravity and the extra-
ordinarily large rotatory power of orange oil, all kinds of foreign additions
can be readily and accurately detected, as there is no adulterant by which
these two properties would not be changed.
Formerly, when the polariscope was not so generally used as it is today
the oil was greatly adulterated with turpentine and sometimes even with
lemon oil.
The terpenes remaining from the manufacture of terpene-free lemon oil
were used to an enormous extent in Messina for the adulteration of
orange oil.1
For the detection of turpentine oil the lowest boiling portions of the
oil are repeatedly fractionated by employing a dephlegmator, and the
pinene may then be recognized by its boiling-point, as well as by its rota-
tory power (strongly laevogyrate with French and slightly dextrogyrate
with American turpentine oil). Should such a test be considered as not
conclusive, the pinene must be converted into pinene nitrosochloride and
into the characteristic pinene nitrolbenzylamine or nitrolpiperidine base.
The decylic aldehyde content can readily be determined by means of
phenylhydrazine. The method to be employed is the same as that used
for the citral assay in lemon oil (cf. p. 64) with this difference that the
mixture of orange oil and phenylhydrazine solution should be allowed to
stand for two hours. Decylic aldehyde reacts slower than does citral so
that complete reaction is not assumed within less time. With mixtures of
decylic aldehyde of known content, there have been obtained very satis-
factory results in this manner. Poorer results are, in turn, obtained by
prolonging the reaction time beyond the two hours. According to observa-
tions made thus far, the decylic aldehyde content of oil of orange varies
between 1.3 and 2.7 per cent.
A sweet orange oil examined by Schimmel in 1917, is worthy of men-
tion and may serve as a warning. It was quite sufficient to determine its
specific gravity and optical qualities, in order to form an opinion on its
quality. The constants indicated were as follows : d1BO 0.8662 ; 00250
4-58° 18'; aD of the first 10 per cent of the distillate -\-$6°2o'. Compare
with this the characteristics of genuine sweet orange oil, namely, at 15°
a specific gravity of 0.848 to 0.853 and an optical rotation of -j-95°3o'
to -{-98° ; the first 10 per cent of the oil has a rotation not measurably
less than the original oil.
The extreme divergence from the normal figures makes it quite obvious
that the sample was at least greatly adulterated. Probably, however, this
was an artificial product in which lemon oil terpenes had been employed,
whereas orange oil is entirely absent, which can be inferred by the exceed-
ingly low degree of rotation of the first 10 per cent of the distillate (boiling
over 172°). However, there is no doubt that this was a case of fraud.
Hood publishes an account of the oil content of Florida oranges.
At various periods he collected oranges during harvest time and made
a determination of the oil value of the peel. Thereby the fact came to
light that conditions of climate and culture greatly influence the oil
*Ibid. (October, 1899), p. 25.
44 CITRUS PRODUCTS
content of the fruit which besides varies considerably with the species.
Only when fully ripe does the peel contain its highest percentage of
oil, which is, however, sufficiently abundant to permit of collecting even
in unripe fruit which is seldom gathered. Rain during harvest reduces
the oil yield considerably. The occurrence of a rust mite has no effect
on the oil yield of the ripe fruit. In tabulated form Hood reports on
the weekly oil yield for eight orange varieties from seventeen different
localities in Florida.1
JAMAICA SWEET ORANGE OIL
Specific gravity at 15° 0.8481 to 0.8491; aD 20° -f-92°43' to +a8°3';
OD of the first 10 per cent of the distillate -{-96° J4' ta -h97°3°'> *n every
instance a little lower than the original oil; specific gravity at 20° 1.46984
(one test only; evaporation residue 1.4 to 2.0 per cent in three aldehyde
determination carried out with phenylhydrazine 3), 2.3, 2.7, and 3.8 per
cent respectively were found, calculated for decylic aldehyde.
OIL OF BITTER ORANGE
The oil of bitter orange (Citrus Aurantium L.), which plays only
a subordinate role in commerce in comparison with oil of sweet orange,
differs from this mainly in its bitter taste. The rotatory power2 is
sometimes slightly lower and varies from +92° to +98°.
All other properties are the same as those of the sweet oil, and it
is impossible to distinguish between the two oils in any other manner
than by their odor and taste.
ITALIAN BITTER ORANGE OIL
According to P. Fenaroli the bitter orange has for some time been made
use of for the production of oil in Italy ; he is of opinion that it may be-
come an important article of commerce in future. The odor resembles
ordinary lemon oil, but has a scarcely pleasant by-odor the elimination of
which may .possibly be attained in time. Fenaroli3 gives the following
figures for this oil: d150 0.847 to 0.848; aD2o° -f-94°-10' to +95°-4°';
nDl70 1.47388 to 1.47408; not readily soluble in 80 per cent alcohol. Soluble
in four to four and one-half volumes of 90 per cent alcohol. It boils at
80 mm. from 102° to 120° and contains 97 to 98 per cent of rf-limonene,
besides small quantities of rf-pinene (?), as well as some bodies boiling
above 176°. Citral could not be found.
^Report of Schimmel & Co. (April-October, 1917), pp. 16-17.
*lbid. (April, 1896), p. 29.
'Compare Report of Schimmel & Co. (April, 1912), pp. 76, 77. Unlike experi-
ments with the citral determination the oil is here allowed to be in contact with
phenylhydrazin for two hours, experiments with decylic aldehyde solutions having
shown what with a contact of one hour only the results obtained are too low.
OIL OF CITRON 45
WJEST INDIAN BITTER ORANGE OIL
The Journal of the Jamaica Agricultural Society reports that before the
great Messina earthquake West Indian orange oil was rarely to be found
in the London market. After this catastrophe, however, more care was
bestowed in Jamaica on the orange oil production which labors under some
difficulties, inasmuch as the trees occur but singly, and not in groves, mak-
ing the gathering of the fruit difficult. Armed with hand presses and
bottles the farmers must proceed from tree to tree, collect the fruit, and
peel it on the spot. The bottles are filled with a mixture of oil, vegetable
slime, and juice. When settled, the oil is poured off into copper vessels.
As the demand is but limited, steps are contemplated to avoid overproduc-
tion if possible.
JAMAICA BITTER ORANGE OIL
d150 0.8517 to 0.8537; 00200 +92°57' to -f96°58'; OD of the first 10
per cent of the distillate -(-92° 20' to -(-96° 40', only in one instance higher
than the original rotation; nD20o 1.47171 (one test only) ; evaporation resi-
due 2.6 to 3.2 per cent. Aldehyde content 0.75 to 1.5 per cent, calculated
for decylic aldehyde.
Two further orange oils, also originating from the West Indies, were
submitted to Schimmel by the Imperial Institute of London. Both had
been obtained by pressure, one from green, the other from yellow oranges.
Unfortunately, the samples were so small that they were only sufficient
for the determination of specific gravity and rotation. These values were
within the recognized limits of sweet orange oil:
Oil from green oranges :d150 0.8515; 0020° +98°2'
Oil from yellow oranges : d150 0.8492 ; aD20o -j-97°22/
OIL OF CITRON
Origin. — According to Gulli the following citron varieties are culti-
vated in Calabria and Sicily: (i) Citrus medica, var. vulgaris, Risso,
known in Calabria and in Sicily as "Cedro"; (2) Citrus medica, var.
gibocarpa or citrea, Risso, designated "Cedrino"; (3) Citrus medica,
var. rhegina, Pasquale, known by the name "Cedrone."
However, in commerce these three varieties of C. medica L. are not
commonly differentiated. As a rule these fruits are not used for the
production of volatile oil but, pickled in brine, they are exported, the
one that is most sought being that designated "cedrone." Citron oil is
produced by expression from the first two varieties, and then usually
on request only at the time of harvesting the fruit in Sicily and in
Reggio-Calabria. It is rarely found in a pure condition since the manu-
facturers add to it varying amounts of lemon or orange oils or the oil
of the sweet lemon.1
1For an oil prepared from the "siisse Limpne" Schimmel & Co. {Report of
Schimmel & Co. [April, 1903], p. 39) ascertained the following constants: di»o
0.8579; OD2oo +64° 15'; HOMO 1.47568. It smells of citral. According to E. Bonavia
{The Cultivated Oranges and Lemons of India and Ceylon [London, 1888], p. 67),
the parent plant of the "sweet lemon" is Citrus Lumia, Risso.
46 CITRUS PRODUCTS
The oil of the "cedrone," does not appear to be known.
Properties. — Although the oils from the fruits known as cedro and
cedrino are not kept separate in commerce, nevertheless they possess dis-
tinctive properties.
1. Oil of cedro or cedrat. d150 0.8706 (Gulli) and 0.871 j1 OD +67°
(Gulli) and -j-67°8'.2
2. Oil of cedrino. d150 0.854; aD +77° to 81°; nD200 (one determina-
tion) 1.47519 (Gulli, 1903).
Composition.— An oil of cedro or cedrat has been examined by Schim-
mel.3 Upon distillation the bulk of the oil passed over between 177° and
220°. The presence of citral was demonstrated by boiling with an alco-
hofic solution of ^-naphthylamine and pyruvic acid, yellow crystalline
laminae of citryl-/J-naphthocinchoninic acid melting at 197° to 200° re-
sulting.
Burgess (1901) has examined an oil which he regards as cedro oil, but
which Gulli (1903) proved to be a cedrino oil. The bulk of the oil con-
sisted of terpenes : according to Burgess mostly limonene with some dipen-
tene (derivatives not mentioned). That portion of the oil which passes
over between 64° and 85° under a pressure of 10 mm. contains citral.
With cyanacetic acid it yielded citralidene cyanacetic acid. Determined
according to the bisulphite method the citral content averaged 6 per cent,
according to the hydroxylamine method 5.7 per cent.
In the bottle of the examined oil a deposit had been formed. By solu-
tion in chloroform a crystalline compound C18H18O6 was isolated that
melted at 145° and probably is identical with citroptene (cf., however, for-
mula on p. 28).
OIL OF GRAPEFRUIT
Grapefruit, C. maxima Merrill, in common with all other citrus
fruits may be made to yield an essential oil, but up to the present time
grapefruit oil has not appeared in the trade. Experimentally it has been
produced in small quantities, but in the presence of an abundant supply
of lemon and orange oil no attempt appears to have been made to manu-
facture it on a large scale. Its production would doubtless be somewhat
complicated by the presence of the bitter principle characteristic of this
fruit.
Properties. — Odor, strongly that of citral ; color, clear greenish
yellow; refractive index at 20 °C. 1.4750 and 1.4785; optical rotation in
loo mm. tube at 2O°C. — 72.5 and — 78.5; specific gravity at 2O°C,
0.845 and 0.860.
Composition. — In the oil analyzed tests were obtained for the pres-
ence of o-pinene, d-limonene, linalool, citral, and geraniol.
^Report of Schimmel & Co. (October, 1895), p. 13.
'Ibid.
'/WA
OIL OF GRAPEFRUIT
47
Large quantities of fresh peel representing the Indian River, DeSoto,
and Excelsior brands of fruit were used by Zoller for oil analysis. They
were thoroughly washed and all pulp removed. It was noticed that the
distribution of the oil sacs in the pericarp varied considerably in the dif-
ferent varieties, some being especially rich in oil, others comparatively
poor.
The cleaned peels were cut by a revolving food chopper into pieces
averaging a centimeter in cross-section. Trial methods of isolating the oil
were tried. A small hand-press cider extractor was employed, yielding a
liquid emulsion of solids, oil and water. When this emulsion was allowed
to stand for several hours there was a slight separation of the oil on the
upper surface but it was always turbid and intensely bitter. Centrifuging
and freezing were tried to free the oil from the accompanying material and
while it was in a measure successful, this would be inadvisable on a large
scale. Precipitation ei the astringent material, pectose, and resins by
means of gelatin and tannic acid solution gave a somewhat clearer oil, but
its flavor was repugnant.
Extraction of the oil by means of volatile solvents gave a good yield
of oil but was laborious, and in case of the solvents employed (acetone and
ethyl alcohol) the bitter glucoside and resins were extracted at the same
time.
The method finally adopted and which resulted in a clear, slightly yellow
oil was as follows : The finely cut peel was introduced into a roomy con-
(Jour. Ind. and Eng. Chem.)
FIG. 9. — Apparatus for grapefruit oil extraction
48
CITRUS PRODUCTS
tainer with an equal weight of water. Slight suction was applied by means
of a water pump furnishing a steady reduction of pressure. Live steam
was, then drawn through the suspended peel and condensed in a suitable
condensing apparatus connected with the suction (see F in Fig. 9). Steam
distillation was continued until the condensate was free from turbidity,
which point indicated that the oil was entirely removed. A second receiver
was connected with the suction in series with F, and this second receiver
was submerged in brine-ice mixture in order to entrap any of the oils which
might tend to be drawn past F. Buffers of glass wool were placed in the
second receiver as an additional precaution. A slight amount of oil was
recovered this way. The oil separated on the surface after a few minutes'
standing, and was drawn off. The remaining traces were removed from the
distillate by centrifuging.
From 0.4 to i.i per cent of oil was obtained. A decrease in the alde-
hyde content of the oil during storage was manifested.
One hundred gm. of the steam-distilled oil, collected under diminished
pressure as indicated above, was washed several times with 5O-c.c. portions
of normal Na2CO3 solution, then with water, and finally dried with anhy-
drous sodium sulphate. The physical properties of the oil in this condition
were determined as stated above under properties.
One hundred gm. of the washed and dried oil was distilled from a
flask fitted with a Hempel column under a constant pressure of 12 mm.
The distilling flask was immersed in a bath of cottonseed oil and the tem-
perature very gradually increased. The distillate was collected in the fol-
lowing fractions, the boiling-point of each fraction being determined and
the approximate results given in Table V.
TABLE V
*Rapidly rising to 225°, then dropping rapidly and browning. Fractions 2 and 3 were mixed
together for qualitative analysis, also Nos. 5 and 6, since the range in either case is similar.
a-Pinene. — Fraction i and the mixtures of Fractions 2 and 3 were
used. Pinene was tested for in these separate fractions by the nitrosyl
chloride method described by Wallach and Ehestadt, and its presence in
both mixtures established by the melting-point of the obtained pinene-
nitrosochloride, 102° and iO3°C. The largest yield was, of course, from
Fraction i. Crystals of pinene-hydrochloride were also prepared from
Fraction i which melted at I32°C.
d-Limonene. — A portion of the mixture of Fractions 2 and 3 distilled
under atmospheric pressure gave a distillate which was collected at the
ESTERS IN ESSENTIAL OILS 49
approximate temperature of pure limonene, 175 °C. The solution was
strongly dextro-rotatory, -}"3^ *n a IO°-mm. tube at 20 °C. As further
proof, 5 gm. of crystalline limonene-tetrabromide were prepared which
melted at I04°C. Limonene was also detected in Fraction 4, though in
small amount.
Aldehydes in limonene fraction. — 10 gm. of the limonene fraction in-
dicated only a possible trace of alcohols. Fraction 4 indicated at least 10
per cent of its weight as of alcohols, calculated as linalool /(CH3)2C:
CH-CH2CH2C(OH)(CH3)CH:CH2. The mixture of Fractions 5 and
6 indicated approximately 4 per cent of their combined weight, calculated
as geraniol /(CH3)2C: CH-CH2C(CH3) : CH-CH2-OH).
Linalool. — Fraction 4 consisting of a little over one gram was shaken
in a small glass-stoppered flask with an excess of 5 per cent sulphuric
acid. After a short time crystals of terpine hydrate (C^HooOaHsjO) sep-
arated, which melted at 116°. The oxidation of linalool to citral was
impossible to detect in this connection, since citral was also present, judg-
ing by the odor and by the test under aldehydes.
Citral. — While citral was known to be present both from the odor and
from the boiling-point of the last two fractions, in the latter case suffering
decomposition, it was further identified by its semicarbazide preparation.
The semicarbazone, prepared according to Zelinsky, melted sharply at 165°.
Geraniol. — One gm. of Fraction 6, before mixing with 5, was shaken
vigorously with 5 per cent aqueous H2SO4 and after standing for one hour
the acid was neutralized with 10 per cent NaOH. Upon further standing
crystals of terpine hydrate separated which melted at n6°C. It was as-
sumed that none of the linalool appeared in Fraction 6, since its boiling-
point is slightly below 200 °C. at atmospheric pressure, while Fraction 6
boils under the same conditions at approximately 230 °C.
The oil from grapefruit is apparently less perfectly known than any
of the citrus oils, which may seem somewhat remarkable in view of the
large and rapidly growing quantity of new material in the form of
unsalable fruit available for its production. The explanation may well
be that grapefruit as a commercial commodity obtained its popularity
within a comparatively short time. Now that its production is growing
at a pace which promises to exceed the present demand a consideration
of possible by-products is becoming a matter of increasing importance,
and this oil should have fuller investigation. There seems to be no
reason why for certain purposes it should not be found equal to lemon
and orange oil.
ESTERS IN ESSENTIAL OILS
Although the estimation of esters in es'sential oils is of extreme
value, considerable difficulty in correctly interpretating results obtained
has recently been introduced on account of the abuse of scientific
knowledge which has led to the preparation of a series of artificial
CITRUS PRODUCTS
esters, most of which have a saponification value which would indicate
the presence of considerably more natural ester than the actual amount
of artificial ester used as an adulterant. The principal esters used in
connection with these frauds are as follows: Terpinyl acetate, glyceryl
acetate, ethyl citrate, ethyl oxalate, ethyl succinate, ethyl tartrate, and
ethyl phthalate. Apart from the actual identification of the acid con-
stituents of these esters, the following notes in regard to one or two of
them will be of value.
TERPINYL ACETATE
This ester is principally found as an adulterant in oil of bergamot, oil
of petitgrain, and similar essential oils. The natural ester present in this
class of oil consists principally of linalyl acetate. Linalyl acetate is hydro-
lyzed at a considerably more rapid rate than terpinyl acetate. It, therefore,
becomes possible to state with certainty that an artificial ester, probably
terpinyl acetate, is present in either of these oils when a marked difference
is found between the saponification value as determined at thirty minutes,
and that determined in sixty minutes. From Table VI1 it will be seen that
hydrolysis of linalyl acetate or of bergamot oil is practically complete in
thirty minutes, whereas the saponification of terpinyl acetate or of bergamot
oil adulterated with this ester is much slower.
TABLE VI
Still greater are the differences when one operates with dilute solu-
tions for the shorter saponification period and with a large excess of
alkali for the longer period. After numerous experiments Schimmel &
Co.2 have prepared the following directions for this fractional saponifica-
tion method:
Four separate experiments are made. About 2 c.cm. of oil are ac-
curately weighed in each case, dissolved in about 5 c.cm. of alcohol and a
few drops of alcoholic phenolphthalein solution (i : 100) added. The acid
value is first determined with N/2 potassium hydroxide. To each of two
of the flasks 10 c.cm. of N/2 potassium hydroxide are added and the con-
tents heated on the water bath for an hour in order to effect saponification
^Report of Schimmel & Co. (October, 1911), p. 116.
^Report of Schimmel &r Co. (October, 1910), p. 60.
ESTERS IN ESSENTIAL OILS
in the well-known manner. The contents of the third flask are boiled with
20 c.cm. of the alkali for two hours and that of the fourth flask diluted
with 25 c.cm. neutral 96 per cent alcohol and saponified for one hour with
10 c.cm. of the standard alkali. For oils with a high percentage ester con-
tent 30 and 15 c.cm. respectively of the standard alkali are used, and in
the case of artificial esters as much, as 40 and 20 c.cm. respectively for the
third and fourth flasks. At the close of the boiling, the flasks are quickly
and uniformly cooled by placing them in cold water and the contents are
titrated immediately.
In the case of bergamot and lavender oils the difference in the ester
values, obtained by boiling for two hours and with the dilute solution for
one hour, amounted to 3 or at most 4, whereas the ester values for the two
normal determinations represented the approximate mean of these extremes.
If terpinyl acetate be added, this difference increases. Moreover, the ester
value obtained for the two-hour saponification increases in the ratio in
which the terpinyl acetate is added, whereas that obtained for the oil
diluted with alcohol increases but little. The ester values of the two
normal saponifications invariably are intermediate. In Table VII the data
have been compiled for a pure bergamot oil and also for the same oil to
which terpinyl acetate had been added in the proportions indicated.
TABLE VII
Hence an addition of 10 per cent terpinyl acetate produces a difference
of 19 in the ester values. (In the case of normal saponification for one-
half and one hour duration respectively, the difference is but 5.5.) Inas-
much as the addition of but 2 per cent of ester produces a difference of
6.4, the presence of even small amounts of adulteration can possibly be
detected in this manner.
CITRIC ACID ESTERS
If the residue of an oil is found too high, adulteration with fatty oil
is no longer the only explanation, for in recent years citric acid triethyl
ester (triethyl citrate [Wiegand and Rubke]) must likewise be considered.
The latter is a particularly serious adulteration since the addition of but
small amounts materially increases the apparent ester content of the oil.
52 CITRUS PRODUCTS
If a given oil is suspicious because of its high evaporation residue, this
is transferred quantitatively to a saponification flask with the aid of alcohol
and saponified in the usual manner with N/2 potassium hydroxide. The
excess of alkali is titrated back with N/2 sulphuric acid, phenolphthalein
being used as indicator. The first decoloration of the indicator is taken
to indicate the end of the reaction, for subsequent reappearances of the red
color may be due to the liberation of alkali by transformation of bergap-
tenic acid into bergaptene. The reappearance of the red color can be
avoided in part by not diluting the saponification liquid with water previous
to the back titration.
The saponification value of the evaporation residue of pure bergamot
oils lies between 136 and 200. The addition of I per cent triethyl citrate
increases the saponification value of the evaporation residue by 47.7. If
one considers that i gm. of triethyl citrate requires as much alkali for
saponification as 2.13 gm. linalyl acetate, it becomes apparent that, e.g.,
2 per cent of added citric acid ester will increase the apparent linalyl acetate
content of the oil by 4.3 per cent, i.e., by an amount which under certain
circumstances may suffice seemingly to bring an oil poor in ester up to
the normal standard.
Noteworthy is the behavior of oils thus adulterated during saponifica-
tion. Whereas pure bergamot oils yield clear solutions with alcoholic
potassium hydroxide, those adulterated with citric acid ester are rendered
turbid more particularly at the beginning of the saponification because of
the formation of potassium citrate which is difficultly soluble in alcohol.
In order to prove the presence of citric acid as such, the calcium test
may be applied. Two gm. of oil or the evaporation residue of 5 gm. of
oil are saponified with alcoholic potassium hydroxide, the solution diluted
with water, neutralized with hydrogen chloride, the alcohol evaporated
from a water bath, the solution shaken out with ester and then filtered. The
filtered solution is rendered slightly alkaline with sodium hydroxide, a few
drops of concentrated calcium chloride solution are added and the mixture
heated. The presence of citric acid is indicated by the formation of a
precipitate, which, however, may require some time to form.
The following method of identifying citric acid is due to Deniges:
10 c.c. of the saponification liquor from which the separated oil had been
removed are shaken with I gm. of lead peroxide, and 2 c.c. of solution of
mercuric sulphate of about 5 per cent strength. After vigorous shaking
the liquid is filtered and 5 c.c. of the filtrate heated to boiling-point, and
a 2 per cent solution of potassium permanganate added drop by drop, con-
stantly stirring until it ceases to become immediately decolorized. If citric
acid is present a flocculent pale-yellow or white precipitate develops after
the first drop.
This reaction is much more delicate than the calcium test. With some
practice it is even possible to approximate the citric acid ester content of
the oil to be examined by comparing the results with those obtained from
oils of known citric ester content. Great care should, however, be exercised
with the addition of the potassium permanganate, for otherwise manganese
peroxide may be precipitated, which may be mistaken for the precipitate
described above (double salt of mercury and acetone dicarboxylic acid).
ESTERS IN ESSENTIAL OILS 53
When six samples of bergamot oil, guaranteed pure, were tested by
Schimmel & Co.,1 according to Deniges' method, two of these revealed traces
of citric acid. This small amount may have been introduced from the pulp
rich in citric acid during the process of expression. As a matter of fact
the traces were so slight that they need not be taken into consideration in
practice. If, however, but I per cent of triethyl citrate be added to such
an oil, Deniges' test yields a precipitate ten times as voluminous. More-
over, the evaporation residue of these authentic oils revealed low saponifica-
tion values; if the latter exceed 200, the oil may be regarded as suspicious.
GLYCERYL ACETATE
This adulterant is usually a mixture of glyceryl acetates in which tri-
acetin predominates. Its detection is moderately easy on account of the
fact that it is readily soluble in very dilute alcohol, and even fairly soluble
in water. Schimmel & Co.2 have proposed the following methods for its
detection.
Ten c.c. of bergamot oil are mixed in a separating funnel with 10 c.c.
of light petroleum and 2.5 c.c. of alcohol, and vigorously shaken up with
20 c.c. of water. The addition of light petroleum and alcohol causes a
very rapid separation of the oil and the aqueous liquid, so that the latter
can be filtered off when the mixture has been allowed to settle for about
ten minutes. Of the filtrate, ten c.c. are neutralized with potassium hydrox-
ide and saponified on the water bath for one hour with 5 c.c. N/2 potas-
sium hydroxide solution. In the case of pure bergamot oil, the 10 c.c. of
filtrate required for saponification : 0.08 c.c. N/2 potassium hydroxide
solution, i.e., 2.2 mg. KOH. After adding i per cent of glyceryl triacetate
0.58 c.c. 16.2 mg. KOH were used. After adding 2^2 per cent of glyceryl
triacetate 1.43 c.c. 40.0 mg. KOH were used. After adding 5 per cent of
glyceryl triacetate 2.79 c.c. 78.0 mg. KOH were used.
Hence the addition of I per cent triacetate requires about 15 mg. KOH
more for saponification.
Later, Schimmel & Co. proposed omitting the petroleum ether and
shaking up 10 c.c. of the oil with 20 c.c. of 5 per cent alcohol. After
allowing the two layers to separate completely 10 c.c. of the aqueous layer,
which may be filtered, is neutralized, using phenolphthalein as indicator,
and saponified for one hour with 5 c.c. of N/2 potassium hydroxide solu-
tion. With pure oils not more than o.i c.c. of N/2 alkali should be required,
a higher figure indicating the presence of glyceryl esters. The time taken
for the separation of the oil and the aqueous liquids is so long, however,
that the use of petroleum ether is preferable. For the positive identification
of glyceryl acetate, the same chemists describe the following process.
Forty gm. were hydrolyzed with the calculated quantity of sodium
hydroxide in a concentrated aqueous solution; the solution was then dis-
tilled, and, as only water passed over, it was next evaporated to dryness
in a dish. The attempt to separate the alcohol (glycerol ?), which had been
split off from the ester, from the sodium salt of the acid by means of an
extracting agent failed, because the known solvents took up not only the
^Report (October, 1910), p. 63.
2Hdf -Yearly Report (October, 1910), p. 61) ; (April, 1911), p. 150.
54 CITRUS PRODUCTS
alcohol but also considerable quantities of the sodium salt. To obviate this
difficulty the sodium salt was decomposed with the calculated quantity of
dilute sulphuric acid and the organic acid which was liberated removed
by distillation, and subsequently by evaporation. This acid was identified
as acetic acid. It was now easy to separate the alcohol of the ester from
the residual sodium sulphate by means of ethyl alcohol. When the ethyl
alcohol had been evaporated by distilling in vacua, the ester alcohol was
identified as glycerol by its boiling-point 147° (4 mm.).
As Salamon and Seaber have pointed out, glyceryl acetate is so easily
washed out with 5 per cent alcohol, that an adulterated oil, when washed
several times with alcohol of this strength, will show a distinctly lower
ester value than the original unwashed oil. Pure oils of lavender, ber-
gamot, and similar oils show practically no reduction in ester value by
such treatment.
NON-VOLATILE ESTERS
When non-volatile or practically non-volatile esters such as ethyl
citrate are used for the purpose of adulteration, almost the whole of the
esters remain in the residue left on evaporating the oil for from three to
four hours on a water bath. If the weight of this residue is above the
normal for a pure oil, non-volatile esters are strongly indicated and may
be approximately determined by estimating the saponification value of the
residue itself. The process is carried out as follows: Five gm. of the oil
are evaporated in a platinum dish on a water bath until the weight is prac-
tically constant. The residue is washed into a flask and saponified in the
usual manner. Titration should be effected quickly with phenolphthalein
as indicator. After the neutral point has been reached, the liquid will
acquire a red tint in a short time. No notice need be taken of this as it
is due to decomposition of bergaptene or similar bodies. The saponification
value of the residue from pure bergamot oil varies from about 135 to 180,
usually about 170. The addition of I per cent of ethyl citrate will raise
it about 50, while the addition of 2 per cent will raise it by nearly 100.
Schimmel & Co. have devised the following method of estimating the
amount of fixed esters present as an adulterant in this type of oil. The
examination of pure bergamot oil has shown that almost the whole of the
acid which is combined with the potassium hydroxide on saponification can
be distilled off with steam after acidifying the aqueous solution with sul-
phuric acid. With pure bergamot oil, therefore, only a little more potas-
sium hydroxide solution is used in ascertaining the saponification number
than is required for neutralizing the acids which have been distilled off.
The estimation is carried out with 1.5 to 2.0 gm. of the oil as follows:
the acid and ester numbers are ascertained in the usual manner, and the
contents of the saponification flask evaporated to dryness after adding a few
drops of N/2 potassium hydroxide solution. The residue is dissolved in
about 5 c.c. of water and acidified with 2 c.c. of dilute sulphuric acid.
From the flask A (Fig. 10), fitted with an ascending tube, a powerful cur-
rent of steam is generated, by which, in about half an hour, 250 c.c. is
distilled over into the measuring flask C in such a manner that the con-
tents of the saponification flask B are kept down to about 10 c.c. with a
ESTERS IN ESSENTIAL OILS
55
small flame. Afterward 100 c.c. more are carried over in the same manner.
The distillate, after a few drops of phenolphthalein solution have been
added, is titrated with N/2 potassium hydroxide solution (or for the sake
of accuracy, better still with N/io solution) until it assumes a, red color.
The first 250 c.c. contain very nearly the entire volatile acids, as the next
100 c.c. invariably use up only o.i to 0.2 c.c. of N/2 solution. From the
quantity of potassium hydroxide solution required to neutralize the entire
distillate the acid number (II) is calculated for the weight of the bergamot
(Schimmel & Co.)
FIG. 10. — Apparatus for estimations of fixed esters
oil used. The difference between the saponification number of several pure
oils which have been examined and the acid number (II) determined in
the manner described above, varied (as shown in Table VIII, from Allen,
Nos. 1-7) from 5.2 to 6.9.
Hence, unless further examination of pure oils should show a higher
figure, oils with greater difference must be regarded as suspected or
adulterated.
In the case of the pure esters (Nos. 12 to 14) here examined the
difference between the two values was from 596.4 to 745.8. The addition
of such esters to oil of bergamot must, therefore, correspondingly increase
the difference of both numbers of the adulterated oil (Nos. 15 to 23 of
the table).
CITRUS PRODUCTS
TABLE VIII
Nos. 24 and 25 show that adulteration with the esters of the volatile
group cannot be detected by distilling the acids, because in these cases the
difference lies of course within the limits for pure oils. Table IX shows
average figures for a number of artificial esters.
TABLE IX
METHODS OF CITRAL ASSAY 57
METHODS OF CITRAL ASSAY
1. The first attempt to determine quantitatively the amount of citral
in lemon oil was made by H. Garnett, who reduced the aldehydes to alco-
hols by means of sodium and attempted to assay the latter by acetylation.
As detailed experiments have revealed, the reaction is unfortunately not
quantitative, hence the results are useless.1
2. Method of J. Walther. — Walther has utilized citraloxime for the
quantitative determination of this aldehyde. An alcoholic solution of hy-
droxylamine hydrochloride of known strength is boiled with a definite
amount of lemon oil (about 10 gm.) and some sodium bicarbonate for
three-fourths of an hour on a water bath in a long-necked flask connected
with a reflux condenser. After cooling, the amount of unconsumed hy-
droxylamine hydrochloride in the reaction mixture is determined titrimet-
rically with N/io sodium hydroxide solution. The difference between the
amount of hydroxylamine hydrochloride originally employed and that thus
found represents the amount that has entered into reaction with the citral
and admits of the computation of the latter. On account of the inaccurate
results the method has not been followed.2 For this reason it has been
modified by A. H. Bennett.3 In order to avoid the evolution of carbon
dioxide, and with it the occasional loss of hydroxylamine, he has employed
caustic potash in place of sodium bicarbonate. The amount of alkali is
chosen so that an excess of hydroxylamine with reference to the citral is
liberated from its hydrochloride, but that some of the hydrochloride remains
unchanged. To a mixture of 20 c.c. of lemon oil and 20 c.c. N/2 alcoholic
(80 per cent) hydroxylamine hydrochloride solution, 8 c.c. alcoholic normal
potassium hydroxide solution and 20 c.c. of aldehyde-free strong alcohol are
added and the mixture boiled for one-half hour in a flask connected with
a reflux condenser. After the reaction mixture has cooled, 250 c.c. of water
are added, part of which is used to rinse the reflux condenser, and the
hydrogen chloride still combined with the hydroxylamine is neutralized,
phenolphthalein being used as indicator. The hydroxylamine not combined
with the citral is then titrated with N/2 sulphuric acid, the end reaction
being ascertained by removing drops to be tested with methyl orange as
indicator. A blank is conducted in like manner but without lemon oil in
order to ascertain the factor of the hydroxylamine solution. From the
difference in the half-normal sulphuric acid consumed in the two experi-
ments the amount of hydroxylamine that has entered into reaction can be
ascertained and the amount of citral can be computed by multiplication
with 0.076. The serviceability of the method was determined by Bennett
by using pure citral. When lemon oil was tested the citral content fluc-
tuated between 4.3 and 5.2 per cent.
Schimmel & Co.* found that by adding pure citral to the oil the results
were too low by 10 per cent.
^Report of Schimmel & Co. (October, 1896), p. 41.
2Report of Schimmel & Co. (April, 1900), p. 22; ibid. (October, 1901), p. 26.
"From a lengthy experience in these processes, E. J. Parry has no hesitation
in saying that the most accurate method for the estimation of citral in lemon oil
is this modification suggested by Bennett.
'Report of Schimmel &• Co. (October, 1909), p. 153.
eg CITRUS PRODUCTS
3. Method of A. Soldaini and E. Berte. — Five c.c. of lemon oil are
measured at a definite temperature in a cylindrical pipette graduated into
1/40 c.c. and allowed to flow into a 100 c.c. fractionating flask, the neck of
which is so contracted that it can be connected with the inverted pipette by
means of a rubber tube. The lateral tube of the fractionating flask is bent
upward and can be closed by means of a piece of rubber tubing and pinch
cock. To the oil in the fractionating flask 25 c.c. of a saturated solution
of potassium acid sulphite solution, containing some free acid, are added
and the mixture shaken thoroughly. The mixture is then digested for ten
minutes in a boiling water bath, care being taken to shake the flask at
short intervals and to see to it that it does not become hotter than is con-
venient to the touch if the bulb is held in the hand. The contents are then
allowed to cool with constant shaking, again heated for five minutes and
finally allowed to cool completely without interruption in the shaking. The
lateral tube is now connected with a funnel, through which water can be
added as desired to the flask, and the inverted cleansed and dried pipette
is connected by means of a rubber tube with the upper, contracted end of
the fractionating flask. Water is now added through the funnel and,
assisted by rotation and tapping, the floating oil is induced to rise into the
pipette. For the purpose of removing any oily particles that adhere to the
neck, air is forced through the lateral tube. The air will carry with it any
oil. After all of the oil has thus been collected in the pipette, it is allowed
to become clear by standing and its volume read off and the difference
computed as volume of citral. For the determination of the approximate
percentage by weight an average specific gravity of 0.859 is taken.
This method has proved inexpedient because at the border of the oily
and aqueous zones a mucilaginous and salty layer is formed which renders
an exact reading impossible.
4. Method of E. J. Parry? — Parry has suggested to determine the
citral content of lemon oil with the aid of citrylidene cyanacetic acid dis-
covered by Tiemann. From 200 c.c. of lemon oil, about 175 c.c. are re-
moved by fractional distillation under diminished pressure. The residue is
measured and the temperature noted. To 10 c.c. thereof contained in a
cassia flask a solution of 5 gm. of cyanacetic acid in 30 gm. of 15 per cent
sodium hydroxide solution are added and the mixture shaken at a low tem-
perature. The aldehyde of the oil dissolves in the aqueous-alkaline liquid,
whereas the non-aldehyde constituents float on the solution. The volume
of the latter is read off in the graduated neck of the flask at the same
temperature at which the volume of the residue of the oil was determined.
The number of c.c. read off is deducted from 10. This difference with
reference to the original volume of oil used (200 c.c.) computed for 100
parts of lemon oil reveals the percentage content of aldehyde.
Although the results obtained with citral mixtures of known content
agreed fairly well (as a rule they were somewhat too high), the method is
not applicable to lemon oil, since an accurate reading is rendered impossible
'This process according to Parry does not give satisfactory results, and is
only quoted here as Parry definitely withdraws it as a process associated with his
name.
METHODS OF CITRAL ASSAY 59
because of the separation of a mucilaginous layer at the zone of contact
as is the case with the bisulphite method.1
5. Method of S. S. Sadtler. — This is based on Tiemann's observation
that when citral is shaken with neutral sodium sulphite solution citraj
dihydrodisulphonic acid results, sodium hydroxide being formed at the
same time :
C9H15CHO+2Na2SO3-f.2H2O = C9H17(SO3Na)2CHO+2NaOH
By titration with semi-normal hydrogen chloride the amount of liberated
sodium hydroxide is determined and from this the amount of citral com-
puted. The method is carried out in the following manner:
Having neutralized the free acid in 5 to 10 gm. of lemon oil, contained
in an Erlenmeyer flask, with semi-normal potassium hydroxide solution, five
times the amount of a 20 per cent sodium sulphite solution are added, the
sulphite solution having previously been neutralized with semi-normal hy-
drogen chloride at water bath temperature, rosalic acid being used as indi-
cator. The mixture being shaken repeatedly is heated in a boiling water
bath and the resulting sodium hydroxide neutralized from time to time.
The reaction is completed as soon as the solution remains neutral, which
occurs after about one-half hour. The amount of hydrogen chloride con-
sumed is equivalent to the amount of liberated sodium hydroxide and this
to the citral content of the oil. The following formula may be used in
which a represents the number of c.c. of N/2 acid and .s the amount of oil
used.
3-8-0
The end reaction, however, is so indistinct that concordant results are
not obtained. In the process of neutralizing the liberated sodium hydroxide
with the acid, the original red color produced by the rosalic acid gradually
passes over into pink, hence it is difficult to decide which shade of color
is to be regarded as indicating the end of the reaction. Neither will it
do to titrate the solution until an absolute change in color takes place, for
then the results obtained are several per cent too high.2
6. Method of G. Romeo. — This method is very similar to Sadtler's and
suffers from the same defect, viz., that the end reaction cannot be deter-
mined accurately. In trying out the method it was not even possible to
standardize the solution accurately according to directions.3
This method is based on the reaction of citral with a solution of neutral
and acid sulphite with the formation of a trihydrotrisulphonic acid accord-
ing to the following equations:
r TT rH . 2Na2SO3-f 2H2O
9 15 0 + NaHS03
= C9H17- (SO3Na)2-CH(OH)SO3Na-fNaOH.
2. 2NaOH-)-2NaHSO3 = 2H2O-f 2Na2SO3.
^Report of Schimmel & Co. (October, 1900), p. 24.
^Report of Schimmel & Co. (April, 1914), p. 48.
^Report of Schimmel & Co. (October, 1905), p. 30.
60 CITRUS PRODUCTS
According to the foregoing equations three acid equivalents correspond
to one molecule of citral. For the assay a solution of 400 gm. Na2SO8
-|-7H2O in i liter of water -{-160 c.c. of a concentrated potassium acid
sulphite solution are standardized against N/2 potassium hydroxide.
Twenty-five c.c. of this are gently heated with 5 c.c. of lemon oil, titrated
with acid, rosalic acid being used as indicator, and the citral content com-
puted. Experiments were made with pure citral, solutions of citral in
xylene of known aldehyde content, and citral-containing oils (ordinary as
well as terpene-free). The results obtained by the author agree very well.
Four lemon oils revealed a citral content varying between 3.64 and 4.48
per cent.
R. O. Brooks employs a neutralized potassium bisulphite solution in
place of that of sodium sulphite. However, according to Schimmel & Co.,1
this modification does not remove the defects of the method.
7. Method of P. B. Rather. — This method for the assay of aldehydes
and ketones is based on their conversion into phenylhydrazones, the reaction
of the excess of phenylhydrazine with iodine, and finally the tttration of
the excess of iodine with thiosulphate. According to E. Fischer the re-
action between iodine and phenylhydrazine is expressed by the following
equation :
2C6H5 NHNH2+2l2 = C6H6N3+C6H5NH2 - HI+3HI.
However, according to E. von Meyer, the reaction is modified by an
excess of iodine as expressed by the following equation:
C6H5NH - NH2+2l2 = C6H6I+N2+3HI.
Fully 5 gm. of phenylhydrazine are dissolved in about 250 c.c. of warm
water, the solution filtered to remove products into a half-liter flask. After
cooling the flask is filled to the neck with boiled distilled water. The
rather unstable solution is to be kept well stoppered and away from light.
The strength of the phenylhydrazine solution is determined in the fol-
lowing manner: 300 c.c. of water and exactly 40 c.c. of N/io iodine solu-
tion are transferred to a liter flask. With the aid of a pipette, 10 c.c. of
the phenylhydrazine solution are transferred to a small flask containing
50 c.c. of water and the diluted solution transferred to the liter flask.
After about i minute the excess of iodine is titrated with N/io. thiosul-
phate. o.i gm. of pure phenylhydrazine corresponds to 37 c.c. of N/io
iodine solution.
Of the lemon oil to be examined, about 10 gm. are weighed in a
weighing flask and in order to avoid oxidation several c.c. of alcohol are
immediately added. With the aid of about 30 c.c. of alcohol the oil is then
transferred to a flask of 250 c.c. Enough standardized phenylhydrazine
solution is then added so as to supply one molecule of phenylhydrazine for
every molecule of aldehyde or ketone. The mixture is then thoroughly
shaken and set aside for about twenty hours protected against light and
repeatedly shaken during this time. After this the unchanged oil is removed
by means of a separating funnel and repeatedly washed with water, the
washings being added to the liquid to be titrated. This liquid is still further
^Report of Schimmel & Co. (October, 1908), p. 156.
METHODS OF CITRAL ASSAY 61
diluted with water and filtered through a pleated filter (if necessary with
some gypsum), into a liter flask containing about 500 c.c. of water and 10
to 20 c.c. of N/io iodine solution, the latter amount varying according to
the amount of phenylhydrazine used. During the process of filtration the
flask is shaken repeatedly. The filter having been carefully washed wtih
water, the liquid is titrated with N/io thiosulphate solution, starch solution
being used as indicator.
The percentage of aldehyde being indicated by x, this can be computed
with the aid of the following formula:
d-M
x = ,
100-4- s
in which d is the difference between the amount of c.c. of N/io iodine
solution corresponding to the phenylhydrazine used and the actual number
of N/io iodine solution consumed, M the molecular weight, and s the
amount of material used expressed in grams.
Upon trying out this method1 it was found that, in general, useful re-
sults are obtainable but that these do not exceed in accuracy those obtained
by the volumetric bisulphite method and the sulphite method. In daily
practice it offers no advantage over the other two methods, both because
of the instability of the required solutions and because it takes too long
to ascertain the results. Neither is the method sufficiently reliable so that
conclusions as to the quality of the oil can be drawn directly from the
determined citral content. Indeed, the variations in the results are suffi-
ciently large to admit of the passage of adulterated oils as good ones.
8. Method of P. Bruylants. — For a long time the blood spectrum has
been employed for the detection of such substances as oxygen, carbon
monoxide, and hydrogen cyanide by observing the change in the spectra
produced by them. Recently Bruylants proceeding from the following con-
siderations has utilized the spectroscopic behavior of the blood for the
quantitative determination of aldehydes. If yellow ammonium sulphide and
a small amount of aldehyde are added to defibrinated blood, the two absorp-
tion bands of the oxyhaemaglobin lose their intensity. Gradually a third
band appears between them which increases in distinctness and ultimately
acquires the same degree of intensity as the two bands of the oxyhaemaglo-
bin. These are then modified still more. Bruylants points out that, accord-
ing to his method, not only the citral but all of the aldehydes are determined
in the lemon oil. So far as practical results are concerned this is of little
consequence since the other aldehydes are present in small amount only as
compared with citral.
For carrying out the determination the following instruments and re-
agents are needed: (i) A spectrophotometer after d Arson val the two
fields of vision of which are arranged one above the other so as to admit
of direct comparisons; also the containers for solutions belonging thereto,
burettes and test tubes. (2) A 3 per cent blood solution, made from fresh,
defibrinated pig's blood, which in the spectroscope shows the two bands
characteristic of oxyhaemaglobin. (3) A solution of ammonium sulphide
^Report of Schimmel & Co. (October, 1907), pr. 121.
62 CITRUS PRODUCTS
which must be prepared with special care in order to obtain the desired
results. (A definite volume of ammonia is saturated with hydrogen sul-
phide, a somewhat larger volume of ammonia is added, and the ammoniacal
sulphide solution thus obtained exposed to the light for several days.)
(4) Citral solutions of 3, 4, and 5 per cent citral in citral-free oil for
comparison. (5) Aldehyde- free, 94 per cent alcohol prepared by boiling
over potassium hydroxide and m-phenylenediamine hydrochloride.
Preliminary tests are made, preferably by two persons, to ascertain the
approximate percentage of aldehyde. To 9 c.c. each of the blood solution,
i c.c. of the ammonium sulphide solution is added, and then the solution of
the oil to be examined to the one and the 5 per cent citral control solution
to the other. The solutions should be prepared, under exactly the same con-
ditions. Both of the oil solutions are so prepared that for every c.c. of
aldehyde-containing oil 5 c.c. of alcohol are used. At first 0.15 c.c. of each
solution is added and the containers shaken for about one-half minute.
After separation of the insoluble portions, the lower layer is removed by
means of a pipette and transferred to the spectroscope tubes. When work-
ing fairly rapidly the oxyhaemaglobin bands can be located in both fields
of vision. Then the band in the middle appears as first phase of the re-
action. If the aldehyde content is the same in both solutions this middle
band appears in both fields at the same time. If this is not the case the
amount of the alcoholic solution of the oil is increased or diminished by
one-hundredth part of a c.c., whereas the amount of the control solution
remains the same. Thus, after five or six tests the aldehyde content can
be ascertained approximately. In the subsequent exact determination, a
control solution is used, the aldehyde content of which approximates that
of the oil to be examined. If the oil contains less than 5 per cent aldehyde,
a 3 or 4 per cent citral solution is used; if more than 5 per cent, the 5
per cent solution is employed and the solutions are tested until the middle
absorption band appears at the same time. After some practice, a com-
plete determination can really be made within a half-hour. The results
are fairly exact. In an artificial oil with 5.06 per cent of citral, 4.9 per
cent were found.
Several lemon oils were examined by Bruylants and the following vol-
ume percentage of citral ascertained :
Messina oil, 1906 harvest: 4.58 and 4.65 per cent; harvest of January,
1905: 5.42 and 5.52 per cent; harvest of November, 1904: 6.47 per cent.
Reggio oil, 1906 harvest: 5.10 and 5.14 per cent.
Two commercial oils contained 3.5 and 3.7 per cent, a third contained
but 2.86 per cent citral.
Whatever the scientific interest may be that is attached to this method,
from the point of view of the analytical laboratory it is much too compli-
cated to be used in the assay of volatile oils. For this reason it apparently
has not been used in practice.
9. Method of E. M. Chace. — This method is based on the property of
fuchsinsulphurous acid to become red in the presence of aldehydes, the
intensity of the red color depending on the amount of aldehyde present. A
comparison with the color of solutions of known aldehyde content enables
the determination of the aldehyde in the oil to be examined.
METHODS OF CITRAL ASSAY 63
The following reagents are required: (i) Fuchsinsulphurous acid: 0.5
gm. fuchsin are dissolved in 100 c.c. of water and a solution containing 16
gm. of sulphurous acid is added. When decoloration has taken place, the
solution is diluted to i liter. This solution has to be prepared fresh every
two or three days since it spoils readily. (2) Alcohol of 95 per cent by
volume that is absolutely free from aldehydes. In order to remove all but
traces of the aldehyde, the alcohol is allowed to stand over alkali. The
alcohol is then distilled and the distillate boiled for several hours with
w-phenylenediamine hydrochloride, using 25 gm. per liter. After a second
distillation the alcohol is ready for use. (3) A standard citral solution, i.e.,
a solution of o.i per cent of citral in 50 per cent aldehyde-free alcohol.
These, as well as the following solutions, are prepared at a temperature of
15°. At this temperature all of the determinations are made, the several
containers being kept at this temperature in a water bath for ten minutes
before being used. Above all, a higher temperature is to be avoided.
The assay is carried out in the following manner: 2 gm. each of the
oils to be examined are diluted to 100 c.c. with aldehyde-free alcohol. Four
c.c. each of these solutions are transferred to flasks of the same size, 20
c.c. aldehyde-free alcohol added and then 20 c.c. of fuchsinsulphurous acid
and the solution diluted to 50 c.c. with alcohol and well mixed. With the
standard citral solution, like solutions are made for comparison. All solu-
tions are allowed to remain in the water bath for 10 minutes when the
intensity of the colors are compared either directly or with the aid of a
colorimeter.
As tried out by Chace, this method yields fairly satisfactory results with
mixtures of citral and limonene, the values found differing not more than
0.2 per cent on the average from the real citral content. With lemon oils
the results were not as favorable, though here also the errors remained
within 0.5 per cent.
While trying out this method, Schimmel & Co.1 did not use the colori-
meter, but compared the intensity of color directly. They prepared eight
different mixtures of the standard citral solution with the fuchsinsulphurous
acid and alcohol in such a manner that each subsequent mixture contained
i c.c. solution = i mg. citral more than the preceding one : the first mix-
ture containing i c.c. of standard solution = i mg. citral, the eighth one
8 c.c. of solution = 8 mg. citral. Thus it became apparent that up to
5 mg. citral content the mixtures revealed differences in color, whereas
beyond 5 mg. the color became so intense that differences were no longer
observable. Smaller differences than i mg. in the citral content of the
solutions were directly observable, i.e., without colorimeter, only in very
dilute solutions, i.e., up to about 2 mg. citral.
According to Schimmel & Co. the citral content of lemon oils cannot
be determined with certainty in this manner, since the waxy constituents
of the oil prevent the recognition of the tone of color with any distinctness.
This results in appreciable errors. The amount of citral found by the
comparison of colors is to be multiplied by 1250 to yield the citral content
of the oil. If there is any doubt as to whether the color indicated a citral
content of 3 mg. or 4 mg. (lesser differences cannot be determined for
^Report of Schimmel & Co. (April, 1907), p. 167.
64 CITRUS PRODUCTS
lemon oil) the following percentages result: o.oo3XI25°=3-75 of O.OO4X
1250 = 5.0, hence results that differ by 1.25 per cent. For the low citral
content of lemon oils this difference is too great for the oils themselves,
though, no doubt, the method is applicable to lemon extracts. Indeed, Chace
himself has expressed this view in his publication.
10. Method of R. S. Hiltner. — For the determination of citral in lemon
extracts and oils Hiltner likewise recommends a colorimetric method. It
is based on the observation that at room temperature the addition of citral
to a dilute alcoholic solution of m-phenylenediamine hydrochloride causes
the formation of a yellow color, the intensity of which increases with the
citral content. For this purpose a I per cent freshly prepared solution of
w-phenylenediamine hydrochloride in 50 per cent alcohol is employed. It
must be perfectly colorless and clear, conditions that are attained, if neces-
sary, by treatment and filtration with animal charcoal. By comparison with
a solution of known citral content (0.25 gm. citral to 250 c.c. of 50 per
cent alcohol), the citral content of an extract or oil can be determined.
For a determination 1.5 to 2.0 gm. of lemon oil are diluted with 90
to 95 per cent alcohol to make a 50 per cent solution. Of this solution 2
c.c. are transferred to the colorimeter tube, 10 c.c. of the phenylenediamine
solution are added, and the mixture diluted to a definite volume. With the
standard citral solution comparative mixtures are prepared until the same
intensity of color has been produced as has resulted in the solution of the
oil to be tested. From this the citral content of the oil can be computed.
If, according to directions, the tests are made at room temperature the
reaction is influenced neither by the acetaldehyde of the alcohol, nor by
the other constituents of the oil. However, for oils that have been altered
by oxidation, the method cannot be used, for such oils, according to the
degree of oxidation, produce a yellowish-green to greenish-blue color.
11. Method of C. Kleber. — Even before its publication this method had
been used for a longer period by Schimmel & Co.1 who tested it as to its
utility. It was thus demonstrated that it is superior to all methods pub-
lished hitherto, not only because it yields more reliable results, but also
because it can be carried out more readily. It is based on the observation
that phenylhydrazine can be titrated accurately with mineral acids when
ethylorange is used as indicator and that the hydrazones which the phenyl-
hydrazine forms with aldehydes and ketones react neutral toward ethyl-
orange. Inasmuch as the phenylhydrazine solution decomposes very rapidly,
it is best prepared fresh for each assay. At least no solution should be
employed that is more than one day old. If necessary the phenylhydrazine
is previously distilled in vacuum.
According to the directions, slightly modified by Schimmel & Co., about
2 gm. of oil are mixed with a freshly prepared 2 per cent alcoholic phenyl-
hydrazine solution in a 50 c.c. glass-stoppered flask and set aside for one
hour. Then 20 c.c. of N/io hydrogen chloride are added and the liquids
mixed by gentle shaking. Ten c.c. of benzene are then added and after
the mixture has been shaken thoroughly it is transferred to a separating
funnel. After having been allowed to stand for a short time, the 30 c.c.
of acid layer are drawn off and filtered through a small filter.
^Report of Schimmel & Co. (April, 1912), p. 75.
METHODS OF CITRAL ASSAY 65
Twenty c.c. of this filtrate to which ten drops of ethylorange solution
(1:2000) have been added, are titrated with N/io potassium hydroxide
until a distinct yellow color appears. From this the amount of N/io potas-
sium hydroxide required for 30 c.c. of filtrate is computed. For the
standardization of the phenylhydrazine solution a blank experiment, i.e.,
without oil is made. If the 30 c.c. of filtrate of the first experiment demands
a c.c. of N/io potassium hydroxide and b c.c. in the latter experiment, the
citral equivalent of the amount of oil (s grams) use is a — b c.c. N/io
potassium hydroxide. Inasmuch as i c.c. of N/io potassium hydroxide cor-
responds to 0.0152 gm. of citral, the percentage content of the oil can be
computed with the aid of the following formula
(a— &)-i.52
The object of shaking out with benzene is to render clear the solution
which has become turbid upon addition of the hydrogen chloride. This
facilitates the recognition of the change in color.
In Table X are recorded the results obtained by this method with mix-
tures of known citral content. These clearly reveal the usefulness of the
method.
TABLE X
CITRAL CONTENT OF THE SOLUTION (IN PER CENT)
A comparison of the results obtained according to the several methods
for the assay of citral reveals appreciable differences. Hence it is always
necessary to report the method with the results. The citral content of pure
or supposedly pure oils is recorded by the following investigators: Soldaini
and Berte: 6.5 c.c. in 100 c.c. oil; Parry (1900) : 5 to 6 per cent; Parry
(1903): 3 per cent; Sadder: 5.26 per cent; Berte: 6.85 to 7.4 per cent;
Romeo: 3.64 to 4.48 per cent; Rother: 5.25 to 5.9 per cent; Bruylants: 2.86
to 6.47 per cent; A. H. Bennett: 4.3 to 5.2 per cent; Chace: 4.1 to 7.0
per cent.
Determined according to Kleber's method, which is pronounced the
most reliable by several American chemists,1 the citral content of pure
lemon oils varies between 3.5 and 5 per cent and is mostly about 4 per
cent.2 Hence the citral content in different oils fluctuates by more than 70
per cent. For this reason the citral content admits of the detection of
adulteration only then when this is considerable. Thus to an oil which
originally contained 5 per cent of citral, 20 per cent of non-aldehyde con-
stituents might be added without detection by this means. Nevertheless,
the citral assay is indispensable in the valuation of a lemon oil.
JE. M. Chace, Bureau of Chemistry, United States Department of Agriculture
Bulletin No. 137 (July 20, 1911), p. 64.
^Report of Schimmel & Co. (April, 1914), p. 58.
56 CITRUS PRODUCTS
TERPENELESS ESSENTIAL OILS
By subjecting the essential oils to careful distillation under reduced
pressure, there are obtained as distinct fractions, the hydrocarbons or
terpenes of the oils, and an odorous oxygenated portion, which can be
kept behind in the distillation apparatus. The terpene fraction has
comparatively little or no odor in most cases; while, as a rule the
oxygenated or high-boiling fraction retains all the good qualities of
the original oil in a heightened degree. In other words, the oil has been
concentrated, and the residual valuable fraction is found to have a far
greater solubility coefficient (in alcohol) than the original oil, a fact
which is of considerable commercial importance. In certain cases, dis-
tillation, at however low a pressure, would cause decomposition of the
valuable oxygenated constituents of the oil. When this is the case, the
terpeneless oil is of very little value since it does not represent the
unaltered concentrated oil. Some few oils contain so little terpene that
the terpeneless oil possesses little advantage over the neutral product.
The chief constituents of essential oils, prepared in a pure state
(e.g., citral, eugenol, safrol, etc.) have occasionally been termed terpene-
less oils, but this is manifestly a misnomer. In many cases these prod-
ucts are inferior to the real terpeneless oils. The two classes should,
however, be carefully differentiated and the term "terpeneless oil" re-
stricted to the product properly coming under that description.
Volckel and Schweitzer in 1840 and Haensel in 1876 first pointed out
the value of this procedure. Many other observers followed up their re-
searches and the increased usefulness of the terpeneless products has been
widely recognized. The terpenes as a class readily oxidize and resinify on
exposure to air, and at the same time develop an unpleasant odor and taste.
Essential oils, which are thus liable to lose their delicate odor and flavor
on keeping, may therefore frequently be stored in a terpeneless state for a
long time without deterioration. Thus the terpeneless oils are specially
valuable in the manufacture of perfumes, liquors, and essences, and for
medicinal purposes.
Table XI shows the physical characters of many of the best-known
terpeneless oils. The data are chiefly due to Parry, though some are taken
from Haensel's reports (H). With the exception of the first sample of
bergamot oil all the samples were genuine terpeneless products.
The "concentrated oil of lemon" on the market contains about 10 per
cent of terpenes and is therefore to be distinguished from the terpeneless
oil. True terpeneless oil of lemon is freed from the greater portion of the
natural stearopotone by allowing the latter to separate after the distillation
of the terpenes ; or the oil is prepared by distilling 90 per cent of natural
oil of lemons under reduced pressure, and steam-distilling the residue, when
the terpeneless oil comes over practically free from citraptene. The yield
REFERENCES 67
of terpeneless oil is from 5 to 6 per cent, containing approximately half
its weight of citral.
TABLE XI
SPECIAL CHARACTERS OF ESSENTIAL OILS
*Probably contained artificial linalyl acetate (Parry).
tBurgess and Child have recorded a number of figures showing the composition and behavior on
fractionation of terpeneless oils of lemon (Journal of the Society of Chemical Industrial, XX [1901], 1176).
REFERENCES FOR CHAPTER II
BARBIER, P., and BOUVEAULT, L.
Comptes Rendus, CXXII (1896), 85.
BEHAL.
Bulletin de la Societe de Chimie (1914), 306.
BENNETT, A. H.
Annalist, XXXIV (1909), 14.
Chemisches Zentralblatt, I (1909), 593.
BERTE, E.
Chemiker-Zeitung, XXIX (1905), 805.
Chemist and Druggist, LXVI (1905), 682.
BERTE, E., and ROMEO, G.
Annali del laboratoris chimico della camera di commercio ed. arti della
provincia di Messina (1909).
Report of Schimmel & Co. (1909), p. 50.
BERT HELOT.
Annales de Chimie et de Physique, Vol. XXXVII, Series 3 (1853),
p. 233.
Ibid., XXXVIII (1853), 44,
Ibid., XL (1854), 628.
BERTRAM, J., and WALBAUM, H.
Journal fur praktische Chemie, Vol. XLV, Series 2 (1892), p. 602.
BLANCHET and SELL.
Liebigs Annalen der Chemie, VI (1833), 280.
BOCKER, E.
"Uber terpen-sesquiterpen freie 6'le," Wallach-Festschrift, p. 201.
Gottingen, 1909.
Journal fur praktische Chemie, Vol. LXXXI, Series 2 (1910), p. 266.
Ibid., Vol. LXXXIX, Series 2 (1914), p. 199.
68 CITRUS PRODUCTS
Ibid., Vol. XC, Series 2 (1914), p. 393.
Report of Schitnmel & Co. (April, 1914-October, 1914), p. 59.
Ibid. (October, 1914-April, 1915), pp. 18-23.
BONAVIA, E.
The Cultivated Oranges and Lemons, etc., of India and Ceylon, p. 188.
London.
BORNTRAGER, ARTHUR.
Zeitschrift fur analytische Chemie, XXXV, 35.
BROOK, B. O.
American Perfumer, III (1908), 24.
BRUYLANTS, P.
Bulletin de la Academic royale de Belgique (Classe des Sciences)
(1907), 217, 955.
Annales de Pharmacie, XIII (1907), 321.
BURGESS, H. E.
Proceedings of the Chemical Society (London), XVII (1901), 171.
Chemist and Druggist, LXXV (1909), 946.
BURGESS, H. E., and CHILD, J. F.
Journal of the Society of Chemical Industry, XX (1901), 1176.
Chemist and Druggist, LXII (1903), 476.
BURGESS, H. E., and PAGE, T. H.
Journal of the Chemical Society (London), LXXXV (1904), 416, 1327,
1328.
CHACE, E. M.
"A Method for the Determination of Citral in Lemon Oils and Ex-
tracts," Journal of the American Chemical Society, XXVIII (1906),
1472.
"The Occurrence of Pinene in Lemon Oil," Bureau of Chemistry,
United States Department of Agriculture Circular No. 46 (October
30, 1909).
CHARABOT.
Bulletin de la Societe de Chimie, Vol. XXI, Series 3, p. 1083.
CRISMER, L.
Berl. Berichte, XXIV (1891), 661, Referate.
Bulletin de la Societe de Chimie, Vol. VI, Series 3 (1891), 30.
Chemisches Zentralblatt, II (1891), 379.
DENIGES.
Bull Soc. Ph. Bordeaux (1898), p. 33.
DOEBNER.
Archiv der Pharmazie, CCXXXII (1894), 688.
Berichte der deutschen chemischen Gesellschaft, XXVII (1894), 352.
DOWZARD, E.
Chemist and Druggist, LVII (1900), 168.
DUMAS.
Annales de Chimie et de Physique, Vol. LIT, Series 2 (1833), p. 45.
Liebigs Annalen der Chemve, VI (1833), 255.
Ibid., IX (1834), 61.
ELZE, F.
Chemiker-Zeitung, XXXIV (1910), 538.
REFERENCES 69
FENAROLI, P.
Annali Chim. Appt., I (1914), 408.
Chemiker-Zeitung Repert, XXXVIII (1914), 541.
Journal of the Society of Chemical Industry, XXXIII (1914), 710.
Report of Schimmel & Co. (October, 1914 — April, 1915), pp. 18-23.
FLATAU and LABBE.
Bulletin de la Societe de Chimie, III, No. 19, 361, 364.
FRANKE.
Dissertation. Erlangen, 1880.
FORTMAN, G.
Journal fur praktische Chemie, Vol. LV, Series 2 (1897), p. 123.
GARNETT, H.
Chemist and Druggist, XLVIII (1896), 599.
GlLDEMEISTER, E.
The Volatile Oils (2d ed.), Vol. III. Translated into English by E.
Kremers. New York: John Wiley & Sons, 1922.
GILDEMEISTER, E., and STEPHAN.
Archiv der Pharmazie, CCXXXV, 583.
GERHARDT.
Comptes Rendus, XVII (1840), 314.
GODEFFROY, R.
Chemisches Zentralblatt (1881), p. 372.
Zeitschrift der Allgemeinem Ostereichischen Apotheker-Vereines, XIX
(1881), i.
GULL!, S.
Chemist and Druggist, LX (1902), 19.
Ibid., LXII (1903), 22.
Report of Schimmel & Co. (April, 1903), pp. 33—35.
HANUS.
Pharmaceutische Zentralhalle fur Deutschland (1904), p. 37.
HESSE, A., and ZEITSCHEL, C.
Berichte der deutschen chemischen Gesellschaft, XXXIV (1901), 296.
Ibid., XXXV (1902), 2355.
HEYER.
Crells Chemischen Annalen, II (1787), 250.
Ibid., Part I (1789), p. 320.
HILTNER, R. S.
Journal of Industrial and Engineering Chemistry, I (1909), 798.
Chemisches Zentralblatt, I (1910), 1899.
HOOD, G. C.
Journal of Industrial and Engineering Chemistry, VIII (1916), 709.
Perfumery Record, VII (1916), 310.
HOOD, S. C., and RUSSELL, G. A.
United States Department of Agriculture Bulletin No. 399.
KLEBER, C.
American Perfumer, VI (1912), 284.
LADELL, R. S.
Pharmaceutical Journal, Vol. XXIV, Series 3 (1894), p. 586.
70 CITRUS PRODUCTS
LAFONTE, J.
Bulletin de la Societe de Chimie, Vol. XLVIII, Series 2 (1887), 777.
Ibid., XLIX (1888), 17.
Chemisches Zentralblatt, I (1888), 107.
LlOTTA.
Report of Schimmel & Co. (April, 1889), p. 16.
Ibid. (October, 1915).
LIPHARD.
Crells Chemischen Annalen, II (1787), 250-51.
LITTLE.
Journal of the American Pharmaceutical Association, III (1914), 553.
American Perfumer (1914), 74.
LUCA DE.
Comptes Rendus, XLV (1857), 904.
MULDER, G. J.
Liebigs Annalen der Chemie, XXXI (1839), 69, 70.
OH ME, C.
Liebigs Annalen der Chemie, XXXI (1839), 320.
OLIVERI, V.
Gazz. chim. ital., Vol. XXI, Series i (1891), 318.
Berichte der deutschen chemischen Gesellschaft, XXIV (1891), 624.
OPPENHEIM, A.
Berl. Berichte, V (1872), 628.
PATANE, G.
Per un pul rasionale apprezzamento delle essenze di limone, Acireade,
1912.
PARRY, E. J.
Chemist and Druggist, LVI (1900), 376, 462, 993.
Ibid., LXIII (1903), 820.
Ibid., LXXV (1909), 875.
Ibid., LXXXIII (1913), 378.
American Perfumer, IV (1910), 214.
POMERANZ, C.
Monatschrift fur Chemie, XII (1891), 379.
Ibid., XIV (1893), 28.
POWER, FREDERICK B.
"The Detection of Methyl Anthranilate in Fruit Juices," Journal of the
American Chemical Society, XLIII (February, 1921), 377—81.
QUERCIGH, E., and MORESCHINI, D.
Rendiconti della Societa Chimica ital., Fasc. XIII (1913).
Report of Schimmel & Co. (1913), p. 123.
REIN; J. J.
Japan nach Reisen und Studien, II, 105. Leipzig, 1886.
ROMEO, G.
Un nuovo metodo di determinazione quantitativa del citral. Messina,
1905.
ROMEO, G., and MORICCA, G.
Sull 'annalisi della essenza di bergamotto. Messina, 1905.
Report of Schimmel & Co. (October, 1905), p. 26.
REFERENCES 71
ROTHER, P. B.
Die Bestimmung der Aldehyde und Ketone zur Bewertung atherischer
Ole. Inaugural Dissertation. Dresden, 1907.
ROURE-BERTRAND FILS.
Scientific and Industrial Bulletin, Series 3, No. 3 (April, 1911), pp.
62-63.
SADTLER, S. S.
American Journal of Pharmacy, LXXVI (1904), 84.
SALAMON and SEABER.
Perfumery Record, III (1912), 275.
SAUSSURE, THEODORE DE.
Annales de Chimie et de Physique, Vol. XIII, Series 2 (1820), pp.
259-84.
Liebigs Annalen der Chemie, III (1832), 165.
SCHMIDT, E.
Apotheker Zeitung, XVI (1901), 619.
Archiv der Pharmazie, CCXLII (1904), 288.
SEMMLER, F. W.
Berichte der deutschen chemischen Gesellschaft, XXIV, 262.
SEMMLER, F. W., and TIEMANN, F.
Berichte der deutschen chemischen Gesellschaft, XXV, 1182.
SODEN, H. VON, and ROJAHN, W.
Berichte der deutschen chemischen Gesellschaft, XXXIV (1901), 2809.
SOLDAINI, A., and BERTE, E.
Gazz. chim. ital, Vol. XXVII, Series 2 (1897), p. 25.
"Metodi generali por 1'analisi della essenze ed in particolare per li
essenze esperidee," Boll. chim. farm., XXXVIII (1899), 537.
Report of Schimmel & Co. (April, 1900), p. 25.
SOUBEIRAN and CAPITAINE.
Liebigs Annalen der Chemie, XXXIV (1840), 319.
Ibid., XXXV, 313.
Journal de Pharmazie, Vol. XXVI, Series 2 (1840), p. I.
TEMPANY, H. A., and GREENHALGH, N.
West Indian Bulletin, XII (1912), 498-501.
THEULIER, E.
Rev. gen. de Chim., Ill (1900), 421.
THOMS, H., and BAETCKE, E.
Berichte der deutschen chemischen Gesellschaft, XLV (1912), 3705.
TILDEN, W. A.
Pharmaceutical Journal, Vol. VIII, Series 3 (1877), p. 190.
Ibid., IX (1879), 654.
Journal of the Chemical Society (London), XLIX, 316.
Ibid., LXI, 344.
TILDEN, W. A., and BECK, C. R.
Berl. Berichte, XXIII (1890), 500, abstracts.
Chemisches Z entralblatt , I (1890), 719.
Journal of the Chemical Society (London), LVII (1890), 323, 328.
Chemical News, LXI, 129.
72 CITRUS PRODUCTS
TUCHOLKA, W.
Archiv der Pharmazie, CCXXXV (1897), 292.
UMNEY, J. C.
Brit, and Colon. Druggist (1909), p. 447.
Report of Schimmel & Co. (April, 1910), p. 66.
UMNEY, J. C, and SWINTON, R. S.
Pharmaceutical Journal, LXI (1898), 196, 370.
WALBAUM, H.
Journal fiir praktische Chemie, Vol LXII, Series 2 (1900), p. 135.
WALLACH, OTTO.
Terpene and Camphor (2d ed.). Leipzig, 1914.
Liebigs Annalen der Chemie, CCXXVII, 289, 290.
WALTHER, J.
Pharmaceutische Zentralhalle, XL (1899), 621.
Ibid., XLI (1900), 585.
WIEGAND, O., and RUBKE, K.
"Verfalschung von Bergamottol mit Citronensaurerester," Zeitschrift
fur angew. Chemie, XXIII (1910), 1081.
Report of Schimmel & Co. (October, 1910), p. 59.
WILEY, HARVEY W.
Chemist and Druggist, LXXV (1909), 913.
American Perfumer, IV (1910), 226.
WRIGHT.
Chemical News, XXVII, 260.
Berichte der deutschen chemischen Gesellschaft, VI, 148.
ZOLLER, H. F.
Journal of Industrial and Engineering Chemistry, X (1918), 364.
CHAPTER III
PRODUCTS FROM THE RIND— PECTIN1
Pectin, also called parapectin, is produced naturally in the ripening
of fruit. It forms a colloidal solution in water and is the gelatinizing
substance in fruit jellies. In the unripe fruit it is represented by its
mother-substance protopectin, while in over-ripe fruit it becomes pectic
acid. It may be produced by various chemical means from protopectin.
It is present in large quantities in oranges, lemons, and grapefruit.
When pure it is white, odorless, and tasteless with practically the
same gelatinizing properties as gelatin. Its usefulness depends on the
fact that it may be employed in combination with various fruits that
contain little or no pectin, such as strawberries, cherries, and peaches,
to form jellies with the respective character and flavor of the latter.
For various purposes pectin is superior to gelatin. It lacks the odor
of the animal product, is not a vehicle for tetanic infection or of im-
purities such as zinc and arsenic occasionally present in small quantities
in gelatin.
SUMMARY OF EXPERIMENTAL PROGRESS
This "plant gelatin" was discovered by Braconnot in 1833. Nearly
a hundred years have passed since then, yet its chemical constitution is
still the subject of investigations and arguments.
Perhaps the most important of the old experiments on pectin are
those of Fremy. According to him the pectin in unripe fruit occurs
chiefly in an insoluble form as pectose. This body has never been
isolated in a pure state as it has never been 'separated by any solvent
from its accompanying substances, e.g., cellulose and similar bodies.
By treatment with organic or inorganic acids, e.g., acetic, and heat,
pectose changes to soluble pectin, and an abundance of calcium goes
into solution. Therefore, Fremy asserts pectose to be a calcium com-
pound of pectin. Mangin believes, however, that pectin is combined in
the cell membrane with cellulose.
Fremy crushed unripe currants and extracted them several hours
with distilled water, until the acid reaction disappeared. By acidifying
1The formation of pectin in fruits and vegetables, its extraction, purification,
chemical and physical behavior, and use are dwelt upon at length because of the
lack of any single work covering the field.
73
74 CITRUS PRODUCTS
the currant residue with tartaric, malic, or sulphuric acid the solution
became thick. Pectin went into solution.
Pectin was also formed when the fruit was cooked in its own juice
because of the interaction of the fruit acid and the mother substance
of pectin — pectose.
According to Bourquelot and Herissey pectose may be converted
into pectin by the action of an enzyme from Aspergillus niger.
Fremy's further researches showed that separated pectin consisted
of loose wool-like white flakes, a tough, horny, somewhat elastic mass,
or occasionally a hornlike powder. Its solution directs the ray of
polarized light to the right. With water it forms first a sticky mass
and finally with more water an opalescent colloidol solution. It is
insoluble in alcohol and ether. The water solution upon the addition
of alcohol becomes gelatinous, or pectin forms a flocculent precipitate.
Acids do not coagulate the solution, but hydroxides of the alkaline-earth
metals do. Ammonium sulphate or magnesium sulphate when added
until saturation is near completion cause precipitation. Neutral lead
acetate will not precipitate freshly prepared pectin but basic lead acetate
will precipitate it.
If a pectin solution is cooked for a certain time, the addition of
neutral lead acetate will cause precipitation, as pectin has been changed
into its isomer parapectin. If pectin or parapectin is cooked with dilute
acid for a certain time, it is changed to metapectin. This last isomer
gives an acid reaction and is precipitated by barium chloride.
These three isomers of pectin may be distinguished through the per-
centage of lead in their respective precipitates. The pectin-lead precipi-
tate contains 10 per cent of lead oxide, parapectin 19 per cent, and
metapectin 33 per cent lead oxide.
Pectin is unusually sensitive to the alkalies and alkaline-earth
metals. If a pectin solution has Sodium hydroxide added to it and is
acidified after a few minutes, a gelatinous precipitate occurs ; pectin has
been changed over to pectic acid. Ammonia will not cause this change
(Braconnot).
Through the action of an enzyme, pectase, obtained in solution from
carrot and beet roots, pectin may be changed into pectic acid. In ob-
taining this pectase the fresh sap of carrots was precipitated with alco-
hol, and the pectase-holding precipitate was dissolved in water. This
solution when added to a solution of pectin will cause coagulation within
a few minutes or half an hour. If the solution is sufficiently concen-
trated gelatinization occurs, while if dilute, flakes of pectic acid separate
out.
EXPERIMENTAL PROGRESS WITH PECTIN 75
Pectin and pectic acid were hydrolized by Bourquelot and Herissey
by diastase (but not by ptyalin nor emulsin) into reducing sugars. The
foregoing authors considered diastase as specific in bringing about this
reaction.
According to Fremy, pectic acid is an isomer of pectin. They both
form white powders after similar treatment, which swell in water and
are easily soluble in hot water. Their water solutions react weakly
acid, and form gelatinous precipitates with hard-metal salts. Alkalies
and ammonia readily dissolve pectic acid. In acidified water pectic
acid is less soluble than in pure water.
If pectic acid is heated a long time in water, it is changed into a
soluble substance called parapectic acid. If a solution of pectic acid
is heated an hour with a very slight excess of calcium hydroxide, it is
converted into metapectic acid. A similar change occurs when pectic
acid is heated for a long time with strong acids.
Metapectic acid has a syrupy consistency, is not precipitated by
alcohol, its alkali and alkaline-earth salts are easily soluble, and it forms
a lead salt with lead acetate.
Mulder considered the formula of pectic acid as C6H8O5. Regnauld
ascribed to the same body C^H^C^o. Fromberg's result is similar to
Mulder's except the size of the molecule is doubled. Berzelius, who refers
to the work of Fromberg, gives the following formulas : for pectin
C24H32O20, for pectic acid C12H16O10, and for the metapectic acid C6H8O5.
Chodnew gives C28H42O24 as the formula for pectin. He describes a
rational method for obtaining pectic acid from turnips, in which pectin is
not isolated first, but pectose is changed directly to pectic acid. Finely
ground turnips are heated with very dilute potassium hydroxide. The
potassium salt of pectic acid is found in the filtrate. From this salt gelat-
inous pectic acid may be obtained through the action of mineral acids.
An acid from pectin is also obtained by Chodnew from turnips. The
ground and washed turnips are cooked in a very dilute solution of hydro-
chloric acid, and alcohol is added to form a precipitate. It is evident that
this precipitate must be pectin itself.
After extracting with dilute hydrochloric acid, Chodnew heats the
turnip residue with potassium hydroxide, and again obtains a gelatinous
substance which he calls hyperpectic acid. Unripe fruit does not contain
pectin but pectic acid, bound with calcium. Through reduction of this
during ripening pectin is formed.
Chodnew gives the following formulas:
Pectin-like cell tissue C28H.^O22
Pectin C28H42O24
Pectin acid (hydrous) C28H42O25
Pectic acid C28H40O26
Hyperpectic acid C28H38O27
76 CITRUS PRODUCTS
The authors mentioned relied for their formulas mostly upon the analy-
sis of lead, silver, and barium salts and also upon the free substances.
The metapectic acid from turnips was investigated by Scheibler. He
holds that by heating it with strong acids it is split into a sugar and an
acid. The sugar from metapectic acid, pectin sugar or pectinose, is iden-
tical with arabinose, which Scheibler extracted from gum arabic. Meta-
pectic acid, therefore, is nothing else than arabin ; Fremy's formula C8H14O9
must be incorrect. Later Scheibler finds that his metapectic acid from
turnips is not a substance of the pectin group, but arabin, which is in the
insoluble form of meta-arabic acid in the turnip.
Reichardt considers pectin bodies as gelatinous carbohydrates, and holds
them quite similar to gums.
Herzfeldt discovered that upon hydrolysis pectin bodies left arabinose
and by oxidation with nitric acid gave mucic acid. The last result is due
to the presence of a galactose group.
Parapectic acid from turnips gave 29.6 per cent mucic acid and 14
per cent furfurol. It is evidently not one substance but a mixture of
arabinose and galactose. The percentage composition of these two sugars
is not constant.
In regard to the constitution of pectin bodies, Cross gives the follow-
ing: Pectin contains 40.8-43.5 per cent carbon and 56.6-57 per cent hydro-
gen, except currant pectin which has a richer content of both.
The substances of the oxycellulose group lie in the same field. The
same proportions are also given by lignocellulose. Pectin reacts similarly
to lignocellulose with chlorine and forms chinochloride which through
reduction changes to a derivative of pyrogallol. Cross considers pectin as
a soluble lignocellulose.
Tromp de Hass and Tollens consider the question of the position of
pectin from a new angle. Pectin compounds are frequently acidic in their
reaction, while plant slimes are neutral. The last are carbohydrates whose
constitution contains H : O = 1 : 8. For pectin the proportion was about
I : 7.4-8.4.
Tollens' results were slightly high in oxygen but this may have been
due to impurities. The excess of oxygen may probably have been caused
by acidic or basic substances joined to the pectin bodies. They may perhaps
take on in that way one or more anhydrous or esterized carboxyl groups.
Presumably, the ester-forming carboxyl group may be attached to the
CHO — or CH2OH — group of the carbohydrate, or possibly it may be part
of a five carbon atom acid, the formula for which would be (C5H8O4),
CBH8O5 ; or C5 — and C6 — groups could similarly be attached.
As O'Sullivan considers the occurrence of a combination of carbo-
hydrates and acids in the gums, so Tollens considers pectin. Arabic acid,
has according to O'Sullivan the formula C91H142OT4 and leaves besides
glucose geddin acid C2sH38O22- In arabic acid the relation of H : O =
i : 8.33 and in geddin acid H : O = i : 9.26. The original pectin is neutral
because the acid group appears in it in a lactone or ester form. By treating
with alkalies the anhydride binding will first break and the pectin will be
changed to pectic acid and form a salt with the alkali. By hydrolysis
hexoses, pentoses, and acids result.
MlCROCHEMICAL REACTIONS
77
MICROCHEMICAL REACTIONS
The fact that pectin substances are akin to cellulose, and occur in con-
junction with it, renders its identification by microchemical means some-
what difficult. Mangin more particularly has investigated these matters,
and gives the following methods :
1. Methylene blue, Bismarck brown, and fuchsine stain pectic sub-
stances, lignified and suberized walls, but not pure cellulose. If sections
thus stained are treated with alcohol, glycerine, or dilute acids, the lig-
nified or suberized walls retain their coloration, while the pectic substances
are decolorized with rapidity.
2. Croceine and bigrosine stain lignified and suberized walls, but do
not stain pectic compounds.
3. Croceine, naphthol black, and orseille red stain pure cellulose, but
do not stain pectic substances; similarly, pectic compounds are unstained
by Congo red and azo blue, while cellulose and callose are.
4. The middle lamella, which apparently consists of compounds of
pectic acid, may be differentiated from the other pectic substances which
are mixed with the cellulose of the cell walls by the following method:
A thin section is placed in a 20-25 Per cent solution of hydrochloric acid
in alcohol for twenty-four hours; the section is then washed with water
and treated with methylene
blue or phenosafranin. The
middle lamella stains much
more deeply than the rest of
the wall.
5. If, after the foregoing
treatment with acid alcohol,
the section be washed in a
10 per cent solution of am-
monia, it is found that the
cells separate with ease one
from the other. According
to Mangin, the combined pec-
tic acid is freed from its
bases by the treatment with
acid alcohol, and is then dis-
solved by the ammonia. A
recombination of the pectic
acid may be brought about by
treatment with baryta water,
and after this process the
cells will not separate one
from the other.
6. The cellulose may be
separated in the following
manner: A thin section is
treated with cuprammonia
for twenty-four hours ; it is
then washed, first with water,
( Rosenberg-tfein)
Fig. ii. — Pectin (black) in intercellular
spaces.
78 CITRUS PRODUCTS
and, finally, with 2 per cent solution of acetic acid. The cellulose is thus
dissolved and fills the cells and intercellular spaces. On treatment with
chlorzinc-iodide the middle lamella gives either no color reaction or turns
a pale yellow, while the cellulose gives the familiar blue reaction; the
membrane stains very deeply with safranin or methylene blue, and is easily
soluble in a solution of ammonia.
Rosenberg found that some of the basic dyestuffs as safranin, methylene
blue, neutral violet, and ruthenium red (ruthenium sesquichloride) were
able to color pectin in unripe fruit without coloring the remainder of the
fruit.
But for ripe fruit another test was necessary : concentrated sugar solu-
tion was placed on the fruit section and the entire heated. If pectin had
been present in the intercellular substances it was lost, its place filled with
air, the cell walls distorted and separated from one another. The pectin
accumulation filled the intercellular space as a gelatinous mass which upon
exposure to air darkened to black.
Pectin is never found in the inside of a cell or on the inside of the
cell wall but always on the outside. Most of the intercellular spaces are
lined with pectin. This lining is continuous with the middle lamella and
appears in the first development.
The gelifying pectin is not the first of the pectin bodies to appear in
the intercellular substance but is seen only during ripening. It is soon
replaced by calcium pectate which marks the end of the pectin series.
There is much confusion in the literature of pectin substances. The
older authors described a large number of different bodies using various
terms.
The following terminology will be used for the remainder of con-
sideration of the pectins: (i) pectose, the equivalent of protopectin;
(2) pectin, under which designation parapectin falls ; (3) pectic acid.
PROTOPECTIN
Under protopectin may be designated pectose, which exists in unripe
or barely ripe fruit (but not in over-ripe or decayed fruit) as an
insoluble substance, which through ripening is transformed into pectin.
Through heating with water, or still better with weak acids, this con-
version is accelerated. Bourquelot and Herissey obtained pectin by
heating protopectin in alcohol. Protopectin according to Tschirch and
Rosenberg is colored with basic dyes and appears as the outer coat of
the intercellular covering and the middle lamella of the fruit. Rosen-
berg considered it as pectin in his dissertation.
Fremy considered tenable the possibility that protopectin is a union
of pectin and calcium, a calcium pectinate. Other investigators as
Payen and Mangin speak of calcium pectate, i.e., a compound of pectic
PECTIN 79
acid and calcium. The latter conception is not correct as pectic acid
is a decomposition product of pectin and not a precedent.
In order to decide whether or not protopectin is a combination of cal-
cium and pectin the following experiments were conducted by von Fellen-
berg.
Two kg. of apples were cooked with water and pressed. The residue
was then cooked one-half hour with i kg. of 50 per cent sugar solution
to separate water-insoluble but sugar-soluble pectin bodies. The residue
was washed with water; 42 gm. of gray, protopectin powder were obtained.
This gave by Zeisel's method 2.12 per cent methyl alcohol in comparison
to apple pectin which contains 10.5 per cent methyl alcohol.
Ten gm. of this protopectin was cooked one-half hour with 150 c.c. water
and pressed. The nitrate was centrifuged until clear. HC1 was added,
alcohol added until precipitation was complete, precipitate pressed and
dried. The resultant dried pectin contained: 0.440 gm. pectin and o.oio
calcium or 100 parts pectin to 2.3 parts calcium by titration.
Another portion of protopectin was cooked in a 2 per cent acetic acid
solution, filtered, pressed, washed with water, and filtrate and wash water
concentrated in vacuum. The analysis of calcium and pectin in this sample
gave results as follows: 0.004 gm. pectin and 0.080 gm. calcium; 100 parts
pectin to 2,000 parts calcium.
A third portion of protopectin was treated with 150 c.c. of I per cent
hydrochloride acid in the cold and pressed. The solution gave 0.069 Sm'
pectin and 0.012 gm. calcium; 100 parts pectin to 1,733 parts calcium.
It is evident, therefore, that the combined calcium is soluble in cold
acetic acid. Acetic acid has no effect on protopectin. The principal part
of the calcium is therefore not united with the pectin, but is a water-
insoluble calcium salt. On account of the colloidal nature of protopectin it
is unlikely that a part of the calcium lost through acetic acid was held back
by absorption. This is especially doubtful since the protopectin had just
been treated with the stronger hydrochloric acid. It has been shown further
that protopectin is not combined with any metal. Concentrated acids do
not split protopectin into pectin, although this may possibly be accomplished
by hydrolysis with dilute acids. In the formation of pectin from protopectin,
no reducing sugar is formed, therefore it is not a glucoside.
PECTIN
Many methods for obtaining pectin are found in literature. It was
first obtained by Braconnot and Fremy as a precipitate from clear,
filtered fruit juice upon the addition of a double volume of alcohol.
The pectin is precipitated in a gelatinous mass. By filtering through
linen or canvas, pressing out, and drying, it can be obtained as a horny
or flaky, more or less colored, mass. It can be freed from minerals by
dissolving in water, filtering several times, adding hydrochloric acid,
and again precipitating with alcohol.
This original method is unsatisfactory for three reasons. First, it
is difficult to get a clear filtrate of many fruit juices, e.g., currants.
8o CITRUS PRODUCTS
Filtration requires a long time. Secondly, the pectin obtained in this
way contains impurities, among them coloring matter. Thirdly, in this
way only the free pectin in the juice is secured, since as a rule the larger
part of pectin is in the form of protopectin.
Another method resorted to in pectin purification is that of dialysis.
The fruit jelly is first dialyzed. The colloidal residue is then precipi-
tated with alcohol. Dialysis, however, does not remove much of the
coloring matter or tannin. Dialysis is much more effective when com-
bined with other methods of purification such as acids and salts. Pectin
itself diffuses but very slightly through the membrane.
The best-known method for pectin purification is that of Bourquelot
and Herissey. By this method all the alcoholic soluble impurities are
first removed by extraction with alcohol. During this extraction a
large part of protopectin is converted into pectin. The material is now
heated under pressure with water and the filtrate precipitated with
alcohol.
Until recently the known constituents of pectin have been confined to
arabinose and galactose. These were hydrolysis products of pectin. Mucic
acid was also obtained from pectin by oxidation with HNO3.
Von Fellenberg discovered methyl alcohol as another constituent. The
alcohol is so loosely bound that large amounts are easily split off in a
minute by the action of alkalies. Pectin by this reaction is changed to
pectic acid. Acids do not split off methyl alcohol as easily as alkalies.
Von Fellenberg certifies to the existence of another pectin constituent,
methyl pectose. By the Tollens method of pentose examination phloro-
glucin is obtained, which is soluble in alcohol giving a brown color. The
soluble portion has many similarities to methylfurfurolphloroglucin. The
exact nature of the methylpentose is at present unknown.
According to the investigations of Tollens the pectin molecule consists
of an ester or anhydride containing a carboxyl group, in which the carboxyl
group has the position of an esterized CHO — or CH2OH — group of the
carbohydrate. The splitting-off of methyl alcohol through the action of
sodium hydroxide and the formation of an acid may also be due to the
ester formation. Von Fellenberg considers pectin, therefore, as a methyl
ester of pectic acid. The statement of Fremy that pectin, pectic acid, and
metapectic acid are isomers or polymers is not in accord with the foregoing
experiments.
To ascertain whether methyl alcohol and no other alcohol was present
the methoxyl was estimated by two methods: first by Zeisel's and secondly
by the colorimetric method of Deniges. Zeisel's method consists in estimat-
ing the methoxyl groups by decomposing the substance with strong hydri-
odic acid and eliminating the alkyl group as alkyl iodide. The alkyl iodide
is passed through an alcoholic solution of silver nitrate, which decomposes
the alkyl iodide and the silver iodide is weighed:
R-OCH3-fHI=R-OH+CH3I.
PECTIN 81
This method, however, does not eliminate error from the possible pres-
ence of other alcohols besides methyl.
By Deniges' method the alcohol is oxidized to aldehyde, and estimated
with acid fuchsin. The formaldehyde frees fuchsin, while of the aldehydes
from other alcohols only amyl and isobutyl give a minimum color.
Both Deniges' and Zeisel's method were tried on currant pectin. The
pectin was obtained by precipitation from the filtered juice with alcohol, and
while not very pure it was pure enough for the experiment. It contained
2.9 per cent ash and contained 9.3 per cent methyl alcohol by Zeisel's
method (determined as ash-free substance) and 9.3 per cent by Deniges'.
This would indicate that methyl alcohol was the only alcohol in pectin.
Orange pectin. — 2.3 kg. oranges were boiled in a reflux condenser six
to eight times (twenty-minute periods) with 4 to 6 liters of 95 per cent
alcohol. This process tended to free them from coloring matter, acids, etc.
Between each boiling the material was pressed. Three liters of water were
added to the final press cake and the mixture heated in an autoclave for
an hour at no°C. At the end of this period the contents were pressed
and the liquid filtered until clear. The pectin was obtained from the filtrate
by adding a double volume of alcohol to which 7 c.c. of concentrated hydro-
chloric acid had been added per liter. The coagulum was pressed in a
cloth and kneaded in several changes of alcohol until the acid reaction dis-
appeared. Finally the pectin was washed with ether and dried in a vacuum
desiccator over sulphuric acid. In this way 17 gm. of a white, feathery
pectin were obtained. When analyzed it gave the following:
Ash 0.2 per cent.
Methyoxyl estimation gave : 0.2495 Sm- substance gave 0.2095 gm. Ag I
= 11.44 Per cent methyl alcohol; 0.2134 gm. substance gave 0.1851 gm. Agl
= 11.76 per cent methyl alcohol.
The pentose estimation gave: 0.2749 gm. substance gave 0.1048 gm.
phloroglucid ; alcohol extracted therefrom 0.0060 gm. methylfurfurolphloro-
glucid. Hence there was present 0.1126 gm. arabinose or 41.0 per cent
arabinose and 0.0185 gm. methyl pentose or 6.7 per cent methyl pentose.
Estimation of galactose : 0.5 gm. lactose and pectin when treated simi-
larly yielded mucic acid. Both samples of mucic acid melted at 2i5°C.
(uncorr.) and solidified at 2i8°C. 0.5 gm. lactose = 0.25 gm. galactose
gave 0.1484 gm. mucic acid; 0.5 gm. pectin gave 0.1626 gm. mucic acid.
Assuming that the galactose of pectin had formed mucic acid similarly
to the galatose of lactose, the galactose content of pectin is 54.8 per cent.
It is possible that as plants have specific proteins, so each plant may
have a pectin of different and specific chemical composition. The pectin
is such a complex molecule that the existence of very many pectins is
easily possible. The pectin molecule apparently contains pentoses,
methyl pentoses, mucic acid-forming groups, carboxyl groups, and
methoxyl groups. These various constituents might readily occur in
various numbers and in various different arrangements.
82
CITRUS PRODUCTS
Q
7,
<
•Z.
s
PH
£
o
M
PH
5
o
tn
.2 o
•g*
jelly when
with sugar
in-free fruit
fljj
OT-— P«
B5
g .'
E * S u
; ^ W rl
,i ilJ <-.5
>'5 G
j o a
•2-2.a«
c
•o .S;
°§«
_73.a-
p -(-> O
•-> o
E"o
j-tjS-O C-O
CJ3
if
t> o
. -
oca .jM^
s-a -go »
c _:j
O C3
r2 G
^§•3 d-S fe
°
S3"3'" a 0-3
|ie.-ssi-§s
;i.i^ji
h«0 O o »<
PECTIN 83
If the results of the analysis of orange pectin be added in the form of
their anhydrides we have the following:
41.0 per cent arabinose=36.i per cent araban
6.7 per cent methyl pentose= 6.0 per cent methyl pentosan
54.8 per cent galactose=49-3 per cent galactan
11.5 per cent methyl alcohol=ii.5 per cent methyl alcohol
102.7 per cent
The methyl alcohol is considered as such and not as methylene. This
calculation does not take into account the effect of the introduction of a
carboxyl in a sugar molecule.
Physical behavior of pectin. — Pectin is a reversible colloid, soluble in
water, forming an opalescent solution. It shows Brownian movement.
It is precipitated from solution by concentrated alcohol as a gelatinous
mass, and by dilute alcohol as flakes.
Besides alcohol some of the metal salts coagulate pectin. Turnip,
apple, and quince pectin are precipitated by copper sulphate, lead nitrate,
and basic and neutral lead acetate. Turnip and quince pectin are precipi-
tated by ferric chloride while a fresh solution of apple pectin remains clear
when treated with this reagent. After standing one hour, however, coagu-
lation occurs in this case also. Zinc chloride when added to a fresh solu-
tion of turnip pectin causes precipitation. Quince pectin after standing
for some time with zinc chloride forms a precipitate.
Entirely analogous behavior was noticed by Fremy in lead acetate pre-
cipitation. He considered pectins from different fruits isomeric. Today
it is considered that the similarity is physical not chemical. Possibly the
particles of pectin become larger upon standing in water solution. When
the particles attain a sufficiently large size they are subject to coagulation
by metal salts which had no action at first.
Neutral lead acetate may also be placed under the same category as
the metal salts in relation to coagulation. When the solution is of pectin
a month old, precipitation occurs, but not in all cases where fresh pectin
is used. The freshly separated pectin will often not be precipitated by
neutral lead acetate but will be precipitated by basic lead acetate. Investi-
gations have been carried on to ascertain the difference between pectin
which was not precipitated by lead acetate and its isomer parapectin which
was coagulated. Difference in physical behavior is not always accompanied
by difference in chemical constitution.
Pectin is not coagulated by the following metal salts: AgNO3, HgCl2,
CO(NO3)2, SrCl2, BaCl2; nor by salts of the alkalies. Coagulation with
the above-named salts is reversible; by acidifying, precipitation occurs, but
the addition of ammonia causes solution. When precipitation occurs by
the addition of CuSO4, further addition of ammonia causes solution and
the formation of a complex blue copper salt. If precipitation be caused
by ferric chloride, added ammonia allows the precipitate to remain, and
iron precipitates also as hydroxide. However, this is only the case when
a great excess of ferric chloride is present. If in only a very slight excess,
a clear solution forms, and the ferric hydroxide remains in colloidal solu-
tion in which the pectin is held. When the lead nitrate coagulum is made
84 CITRUS PRODUCTS
alkaline with ammonia, easily soluble basic lead nitrate is formed. This
in itself forms a gelatinous substance and prevents the solution of the
pectin.
The old investigators considered the combination of pectin and metal
salts precipitants as chemical. The chemical compound is very weak, if
any, as the reaction of pectin and the ready reversibility of the coagulum
attest. This coagulation is more electrolytic, analogous to the coagulation
of some other colloids with metal salts.1
Pectin is not precipitated by tannin nor albumen, nor by acid or basic
dyes.
A gelatinous precipitate forms when calcium or barium hydroxide is
added to a pectin solution. The hydroxyl ion has changed pectin into pectic
acid, and calcium or barium pectate forms. Acids free the pectic acid by
combining with the metals.
Pectin as has already been stated is a methyl ester of pectic acid.
Consequently if pectin be treated with sodium hydroxide, methyl alcohol
is split off. This separation approaches more nearly the theoretical amount
when the solution has been heated. The following experiments show
this:
When a 5 per cent solution of quince pectin was used, the methyl alco-
hol content by Zeisel's method was found to be 10.05 Per cent. Therefore,
100.^2
a normal pectin solution corresponds to — = 318.4 gm. of pectin per
liter or i gm. pectin = — ^-= 3.14 c.c. N/i NaOH or 5 c.c. of the .5 per
3*°
cent solution of pectin = .0785 c.c. N/i NaOH.
Five c.c. of pectin solution were used in each case — different amounts
of N/i NaOH used for different definite times. They were brought back
to acid reaction with N/i HC1 and diluted with 5 c.c. water, then distilled
and the methyl alcohol determined in the distillate. The first 4 c.c. car-
ried all of the methyl alcohol liberated. The methyl alcohol was determined
by Deniges' method.
The results showed that in a 5 per cent solution of quince pectin all
the pectin was hydrolyzed to pectic acid and methyl alcohol by heating
(for 5 c.c. this was .0785; .08 c.c. N/i NaOH was used). Double the
theoretical amount of N/i NaOH hydrolyzed it completely in five minutes
in the cold, while four times the theoretical amount hydrolyzed it completely
in two minutes in the cold.
1Tartar and Gailey ("Role of H-ion Concentration in the Precipitation of
Colloids," Journal of the American Chemical Society, XIV [1922], 2212) report
that the addition of neutral salts to solutions of colloids, results in jelly formation
at much lower hydrogen ion concentration than colloidal isoelectric points. Hal-
liday and Bailey ("Effect of Calcium Chloride on Acid- Sugar-Pectin Gels,"
(Journal of Industrial and Engineering Chemistry, XVI [1924], 595) have shown
that the addition of CaCU in this connection to pectin jellies acts in this way and
if increased sufficiently may cause syneresis. Tarr ("Fruit Jellies No. I, The
Role of Acids," Delaware Agricultural Experiment Station Bulletin, No. 134,
Tech. No. 2, 1923) showed that syneresis and precipitation of pectin occurs in
pectin jellies at pH's above 3.1.
PECTASE 85
PECTASE
Action of pectase. — The pectase of turnip juice was precipitated by
alcohol and dried. This dried pectase (with adherent pectin) was added to
boiled pear juice. On the following day coagulation had taken place. On
distillation it gave a strong Deniges' methyl alcohol reaction. Heated
juice which had not had pectase added to it gave no methyl alcohol reaction.
The distillate of the natural juice on standing without heating gave a
methyl alcohol reaction.
An apple coated with paraffin and allowed to stand turned brown in-
side. When the juice was distilled the distillate gave the reaction for
methyl alcohol.
It was also shown that juice which had started to ferment contained
methyl alcohol from pectin.
Decayed apples contain no protopectin, but reduced amounts of pectin
and large amounts of pectic acid and methyl alcohol.
The enzyme which causes the coagulation of vegetable saps containing
pectin was discovered by Fremy in 1840. At present there is a certain
amount of confusion in the terminology of this and related enzymes.
Euler applies the name pectase to the enzyme which converts pectose into
pectin, and calls the enzyme which coagulates pectin, pectinase. Other
writers retain the original name pectase for the enzyme which coagulates
pectin, and in the following account this enzyme will be denoted by this
name. With regard to the two other related enzymes, the most generally
accepted terminology seems to be that in which the enzyme which hydro-
lyzes pectin to d-galactose and /-arabinose is called pectinase, and the
enzyme which converts pectose into pectin, pectosinase.
Previous workers on this subject — Fremy, Bertrand, Mallevre, Bour-
quelot, and Bourquelot and Herissey simply allowed coagulation of the
pectin by the enzyme to take place under various conditions, and noted the
time taken before this was complete.
In the research by N. G. Ball an attempt was made to study the action
of pectase by observing the electrical conductivity of a solution of pectin
when acted upon by the enzyme, and also by determining the change in
viscosity.
The pectin used in this research was obtained exclusively from the
roots of the carrot, Daucus carota. The first extraction was based on a
method suggested in C. A. Browne's Handbook of Sugar Analysis for
obtaining pectin from the juice of ripe pears. The chopped-up carrots
were put into a can and steamed in a Koch's sterilizer for about forty-five
minutes, and the juice was then squeezed out and filtered under reduced
pressure. A little oxalic acid was added to precipitate calcium, and a little
tannic acid to precipitate albumins. The juice was filtered, and centrifuged
until almost clear. The pectin was precipitated by addition of an equal
volume of alcohol, and was filtered off and redissolved by pouring a little
hot water onto the filter. The pectin was subsequently reprecipitated by
addition of alcohol, and the gelatinous precipitate obtained was collected
by means of a centrifuge and dried on a watch glass placed on a water
bath. A horny residue was obtained in this way, and about 0.3 gm. of
pectin was extracted from about noo gm. of carrots.
86 CITRUS PRODUCTS
In subsequent extractions the carrot roots were finely minced, and the
pulp covered with water and heated on a water bath for one or two hours.
The pulp was then squeezed, and the liquid obtained was treated as before.
This method was based on one described by Bourquelot and Herissey, who
obtained pectin by heating chopped-up gentian roots with water in an
autoclave at no°C.
Browne recommends the addition of tannic acid to juice which has
been extracted in the cold. In cases where the carrot pulp had been heated
this was found to be unnecessary, as all albuminous substances had been
coagulated.
The pectin obtained from these various extractions was dissolved in
sufficient water to form a 2 per cent solution, and a few drops of toluene
were added to prevent growth of micro-organisms. The pectin solution
thus obtained was in a fairly pure state, but was faintly acid to litmus
paper. In all the experiments this solution was diluted with an equal
quantity of water.
Sap released from the leaves of Syringa vulgaris was used as a source
of pectase. This plant was chosen both for the sake of convenience, as
the leaves could be obtained easily, and also owing to the fact that Bertrand
and Mallevre state that sap pressed from these leaves is fairly active in
coagulating pectin.
The method of extraction was as follows : Leaves were stripped from
fresh shoots of Syringa, the petiole of each leaf being removed. They
were then placed in a small steel cylinder fitted with a piston, and provided
with a hole at its lower end. The piston was squeezed in by means of a
vice, and the expressed sap was centrifuged until clear, and was then ready
for use.
o) Measurement of the electrical conductivity during the coagulation
of pectin. — If the product of the action of pectase on pectin is a true gel,
there would be little change in electrical conductivity during its formation,
as the resistance of a gel to the passage of ions is practically the same as
that of the sol from which it has been formed. One c.c. of distilled water
was added to i c.c. of 2 per cent pectin solution in a test tube, and i c.c.
of freshly extracted sap from the leaves of Syringa was put into a similar
tube. The two tubes were partly immersed in a large glass tank of water
for about ten minutes, in order that their contents might come to the same
temperature. At a noted time the contents of the tubes were mixed and
the mixture poured into a Hamburger conductivity tube, which was also
immersed in the same tank of water at 13 °C. Measurements of the resis-
tance of the mixture were made every ten minutes by comparison with a
standard resistance, using a meter bridge with an alternating current and
telephone.
The resistance of the mixture remained practically constant for over
two hours, and at the end of this time a solid jelly had been formed in
the conductivity tube. During the experiment the temperature of the
water in the tank was constant within 0.5 °C.
The product of the action of pectase, therefore, consists of a spongy
network composed of a more or less solid phase in the meshes of which
a more liquid phase is distributed.
PECTASE 87
b} Measurement of the change in viscosity. — In order to measure the
change in viscosity, a viscosimeter of the Ostwald type was constructed,
this consisted of a U-tube made of glass tubing. Part of one limb was
composed of a piece of thermometer tubing, and above this capillary the
tube was dilated into a small bulb. Three c.c. was the amount of liquid
which was always used, and this was introduced by means of a pipette.
In use, the liquid was sucked up to a definite height above the bulb in the
limb of the U-tube whieh contained the capillary, and was then allowed
to run back. The time taken by the meniscus in passing between marks
on two constrictions above and below the bulb was determined by means
of a stopwatch. As this piece of apparatus was not sufficiently accurate
to permit absolute determinations of the viscosity being made, the times
taken for the meniscus to pass between the two marks were compared with
one another.
In the experiments at a temperature above o°C. the viscosimeter was
partly immersed in a tank of water containing about 20 liters. This was
kept at a constant temperature by the introduction of either hot or cold
water from a tap connected with another tank. The water in the tank was
kept well stirred, and a sensitive thermometer attached to the viscosimeter
was easily maintained within o.i°C. of any desired temperature.
The viscosimeter was standardized with 3 c.c. of distilled water at
o°C., and the time for emptying the bulb was 3.0 seconds.
These experiments show that the product of action of pectase on pectin
has the structure of a gel composed of a semi-solid reticulum, in the
meshes of which a liquid is distributed, and also that the gel, when formed,
is broken up in the presence of electrolytes by the clumping of the more
solid portion into separate aggregates.
It appears that what has hitherto been described as the coagulum formed
by the action of pectase is composed of calcium pectate, and not of pectic
acid, as previously stated by Fremy. They base this conclusion on the fact
that the coagulum is completely insoluble in feeble alkaline liquids, but
dissolves almost instantaneously after having been macerated in dilute
hydrochloric acid, and the resulting solution is found to contain calcium.
They also state that if juice, extracted from carrots, which has been decal-
cified by addition of oxalic acid, be added to a solution of pectin from which
calcium has been carefully removed, the mixture remains indefinitely liquid,
but the least addition of a soluble calcium salt causes rapid coagulation.
They point out that, owing to the very high molecular weight of the pectic
compounds, and to the enormous volume which they occupy in the gelatin-
ous state, the amount of calcium required for the formation of a coagulum
consisting of calcium pectate would be very small.
In view of the results obtained in the experiments, it seems possible
that the calcium or other metallic ions act in a purely physical manner in
causing coagulation. In this experiment sap was extracted from the leaves,
and the enzymes precipitated by the addition of five times its volume of
alcohol. The precipitated enzymes were redissolved in water, and added to
pectin solution, from which dissolved calcium salts were absent. In this
way any electrolytes, including calcium salts, would have been almost com-
88 CITRUS PRODUCTS
pletely removed. The experiment, therefore, affords no evidence tnat the
gel which is formed is composed of calcium pectate.
If the statement of Bertrand and Mallevre is correct, namely, that in
the absence of calcium salts a mixture of pectase and pectin remains
indefinitely liquid, then it would seem that gelification, as well as coagu-
lation, is due to a minute trace of electrolytes. As far as gelification is
concerned, the action of the electrolytes is possibly indirect, and it may
be that their presence is necessary in order to allow the enzyme to exert
its activity.
If this is the case, then possibly the events which take place during
the coagulation of pectin are somewhat as follows: Under the action of
pectase some kind of pectic acid is produced from the pectin. If electro-
lytes are completely absent, the mixture remains liquid ; but if any ions are
present, and especially if they belong to one of the divalent metals such
as calcium, partial precipitation of the colloid takes place. At first these
colloidal particles will be free from one another, and while this is the
case the viscosity will remain almost constant, thus explaining the first
portion of the curves. As the action of the pectase proceeds, and more
material is formed, which can be precipitated by the electrolytes, a semi-
solid reticulum is gradually built up with a consequent rise in viscosity.
As the reticulum becomes denser, the viscosity will increase rapidly; but
if electrolytes are present in larger proportion, or if they are allowed to act
for a sufficiently long time, clumping together of the particles of colloid
forming the reticulum of the gel takes place, and a suspension is formed
with a consequent dimunition in viscosity. A maximum viscosity will be
reached when the rate of gel-formation becomes insufficient to counter-
balance the clumping effect.
PECTIC ACID
Pectic acid may be obtained by taking about 100 c.c. of .5-1 per cent
pectin solution, adding 5 c.c. of 10 per cent NaOH and allowing to stand
two minutes. Then acidify with HC1, filter off the gelatinous precipitate
through a cloth, press out, wash several times with alcohol until all the
HC1 is removed, wash with ether, and dry in a desiccator over sulphuric
acid. It may also be obtained by heating the fruit under pressure in an
autoclave in order to change protopectin and pectin into pectic acid.
Weighed pectin was hydrolyzed with NaOH, neutralized with HC1,
dried to constant weight in a glycerine drying oven at 103 °C., weighed,
and the NaCl determined and subtracted. Pectic acid was found in the
proportion of 95.16 gm. acid to the total weight of pectin. .2484 gm. pectin
gave .2364 gm. pectic acid. By the direct distillation of hydrolyzed pectin
10.54 per cent of methyl alcohol was found (the equivalent of 4.62 per cent
of CH2) which proves up very well with 4.83 per cent by the indirect
estimation noted above. It is quite evident, therefore, that pectin contains
no other alcohol but methyl or otherwise these two results would not agree
so closely.
One hundred gm. of pectin acid is theoretically equivalent to .35 gm.
H according to the amount of methyl alcohol contained in apple pectin —
THE FORMATION OF FRUIT JELLY 89
10.5 per cent. By titration its equivalent is .43 gm. H per 100 gm. pectic
acid.
Orange pectin contains 11.7 methyl alcohol. Therefore 100 gm. of
pectic acid from oranges should equal .42 gm. NaOH. By actual titration
with phenolphthalein as an indicator the pectic acid equivalent is .46 gm.
H per 100 gm. pectic acid. This excess may be due to the breaking down
of a lactone group by the NaOH. The freed COOH group may combine
loosely with NaOH. This latter combination is easily broken upon acidi-
fication.
The pectic acid molecule takes up one-half molecule of water. Barium
pectate contains 22.5-23 per cent Ba. Copper pectate contains 9.78 per
cent Cu.
The solubility of pectic acid is greatly influenced by small amounts of
electrolytes. The white pectic acid powder is soluble in water and forms
a colloidal solution of smaller particles than those of pectin solution.
Pectic acid is precipitated by most metal salts, e.g.: NaCl, CaCl2,
SrCl2, BaCl2, MgCl^, A1C12, FeSO4, FeCl3, CuSO4, CO(NO3)2, NiSO4,
CdCl^ ZnSO4, SnCl2, MnG3, AgNO3, PbNO3, etc. It is not precipitated
by HgCl2.
In accordance with its character as a negative hydrosel, pectic acid
is precipitated by albumin but not by tannin. The basic dyes, e.g., fuchsine,
crystal violet, methylene blue, bismarck brown, safranin, precipitate pectic
acid in intensely colored flakes. Pectin is unaffected by these dyes. Acid
dyes as eosin, picric acid, azolithmin, and curcumin form no precipitate;
neither do Congo or indigo carmin.
THE FORMATION OF FRUIT JELLY
Just what chemical substances are necessary and how they combine
to form a fruit jelly of the proper consistency has been a subject of
much investigation and discussion.
Fremy claimed that jelly formed from a change of pectin into pectic
acid. He considered pectic acid the essential part of jelly, as fruit juice
spontaneously gelatinized when pectin was transformed to pectic acid by
the action of an enzyme pectase. This jelly, however, has no connection
with fruit jelly for three reasons, viz.: (i) When a pectic acid solution
and sugar are mixed no jelly forms. (2) Pectic acid together with sugar
and another acid or salt gives a syrupy liquid. A jelly of the proper con-
sistency does not form because pectic acid is not soluble in this media.
(3) An analysis of fruit jelly does not disclose any pectic acid.
Tschirch experimented with currants. He considered jelly as due to
a pectin and sugar solution or a combination between pectin and sugar.
Without sugar he could obtain no jelly. As a disproof of this theory
we have the experience of von Fellenberg. Upon dialyzing currant jelly,
he observed that it attracted water quickly. The liquid increased in the
apparatus and in fifteen hours the jelly dissolved. No clear solution re-
sulted; flocks of a body similar to cellulose remained which could be
removed by filter or centrifuge. When the clear filtrate was dialyzed,
flocks were formed during dialysis which were not present beforehand in
9O CITRUS PRODUCTS
the jelly. These flocks dissolved in a sugar solution, and on successive
heating and cooling a jelly formed. These flocks were considered by
Tschirch as pectin. But this substance is not taken as pectin by von
Fellenberg, first because the filtrate free from these flocks when heated
with sugar gives a jelly. Secondly, this flocking body in currants is not
present in other fruits. Thirdly, it differs from the pectin of von Fellen-
berg in not containing methoxyl groups. Lastly, when Tromp de Hass
and Tollens analyzed this flocculating body, they found it evidently a
carbohydrate containing 54.4 per cent carbon and 5.05 per cent hydrogen.
It contains, therefore, a higher percentage of carbon than either cellulose
or pectin.
To prove that pectin plays an important role in jelly formation von
Fellenberg dissolved quince jelly in boiling water. From this mixture he
precipitated pectin by adding alcohol. The filtrate would form only a
syrupy solution, not jelly. If the precipitated pectin were added jelly
formed. A similar result was obtained when purified pectin was added to
quince syrup.
A purified pectin and sugar solution alone gave no jelly. Pectin from
various fruits and varying quantities when added to a sugar solution re-
sulted in syrups only. He concluded, therefore, that the formation of jelly
is due to other substances than pectin and sugar.
If jelly formation be considered as a coagulation it might be supposed
that a jelly would result by the addition of mineral salts which coagulate
pectin to a solution of pectin and sugar. This was not the case, for when
ferric chloride was added to a solution of pectin and sugar, flocks of a
pectin-iron coagulum separated ; no jelly formed.
Von Fellenberg obtained better results by using organic salts, i.e.,
malates of calcium, magnesium, and aluminum. These gave jellies when
added to a solution of pectin and sugar. Such jellies were not as stiff as
those formed by the addition of pectin to pectin-free fruit syrups. In
summarizing his work with jellies, von Fellenberg concluded that besides
sugar and pectin certain salts are necessary to jelly formation. He con-
sidered the possibility of other parts of fruit juice playing active roles in
jelly-making.
A previous investigator on jelly-making, Goldthwaite, noticed that jelly
formation would occur when 0.5 per cent tartaric acid solution, i per cent
pectin, and three-fourths volume of sugar were heated for not more than
fifteen minutes. She considered the presence of free acid necessary for
jelly. In fact she claimed that neutral juice would not gelatinize.
In considering earlier researches, I believe that the malates of calcium,
magnesium, and aluminum may owe their virtue in gelatinization to the
presence of the hydrogen ion. Acid malates may have been present as
impurities in the salts used by von Fellenberg. It is doubtful if neutral
malates would aid appreciably in gelatinization.1
In commencing work with pectin I considered the principal factors
in securing jelly of proper consistency to be pectin, sugar, and acid.
JTarr ("Fruit Jellies, No. I, The Role of Acids," Delaware Agricultural Ex-
periment Station Bulletin, No. 134, Tech. No. 2, 1923) has found that the mini-
mum pH for jelly formation is 3.40, and that the pH for optimum jelly is 3.1.
THE FORMATION OF FRUIT JELLY
The pectin used was obtained from Villa Franca and Eureka lemon rinds
by the von Fellenberg process. Four kg. of lemon rinds were boiled in a
reflux condenser six to eight times (twenty-minute periods) with 4—6
liters of 95 per cent alcohol. This process tended to free them from color-
ing matter, acids, etc. Between each boiling the material was pressed in
a screw press. Three liters of water were added to the final press cake
and the mixture heated in an autoclave for an hour at uo°C. At the end
of this period the contents were pressed and the liquid was filtered through
a Seitz asbestos filter until clear. The pectin was obtained from the filtrate
by adding a double volume of alcohol to which 7 c.c. of concentrated hydro-
chloric acid had been added per liter. The coagulum was pressed in canton
flannel and kneaded in several changes of alcohol until the acid reaction
disappeared. Finally the pectin was washed with ether and dried in a
vacuum desiccator over sulphuric acid.
The results show: (i) All three of the substances tested, when in
aqueous solution, independently increase the viscosity of distilled water.
(2) Mixtures of any two or all three of the substances result in an increase
of viscosity greater than the viscosities independently. (3) Jelly will form
when acid, pectin, and sugar are in solution of
certain concentration. (4) Jelly will form from
a solution of 3 gm. per cent pectin and 65 gm.
per cent sugar. (5) The viscosities of the sepa-
rate substances are not additive.
According to Hardy and Robertson jelly forma-
tion is analogous in some instances to emulsifica-
tion and may therefore be dependent on surface
tension phenomena. From the Tolman theory of
colloids in the case of a lyophylic colloid in equi-
librium with a dispersing medium, an increase in
the concentration of the hydrogen ion in the dis-
persing medium lessens surface tension and hence
increases dispersion. Thus an increase in hydro-
gen ion concentration would result in an increase
in the total surface of the pectin and consequently
the viscosity of the solution could be increased.
Still further additions of acid might lead to such
high degrees of dispersion that the condition of
affairs in so-called true solutions would be ap-
proached and thus the viscosity again decreased.1
The function of sugar may be that of a dehy-
drating agent. It also imparts a certain amount
of viscosity to the jelly presumably by combining
the wattf which would otherwise lead to fluidity.
Glycerol acts similarly. The comparative anhy-
drous state of jelly may be shown by the addition
of a small amount of either cupric chloride or
cobalt chloride to the jelly. Green and red colors
respectively result, which are the anhydrous col-
(A. H. Thomas Co.)
Fig. 12. — A Redwood
viscosimeter.
JSee footnote i, page 84.
92
CITRUS PRODUCTS
ors of these salts. Upon dilution these colors change to blue and pink, the
respective hydrous colors of the salts.
If the function of sugar is principally that of dehydration, then a higher
concentration of pectin together with the same concentration of acid and
no sugar should give a jelly of equal firmness to the one in which sugar
is present. Indeed experimental results indicate such to be the case.1
In the viscosity experiments a Redwood viscosimeter was used. The
results are tabulated in Table XIII.
TABLE XIII
VISCOSITY RESULTS ALL OBTAINED AT 45° c.
SUMMARY
1. Pectin, acid, and sugar are the principal factors in fruit
jelly-making.
2. Pectin was obtained from lemon rinds by boiling in several
changes of alcohol, heating in an autoclave for one hour at no°C. in
'Confirmatory evidence toward the functioning of sugar as a dehydrating
agent has been obtained by Holmes and by Tarr and Baker ("Fruit Jellies, No. 2,
The Role of Sugar," Delaware Agricultural Experiment Station Bulletin, No.
136, Tech. No. 3, 1924. H. N. Holmes and H. A. Howe, "The Role of Pectin
in Gel Formation," Science, New Ser. LVIII [1923], 314).
THE PRODUCTION AND USES OF PECTIN 93
the presence of water, filtering precipitating by a double volume of
acidified alcohol, removal of acidity with alcohol and ether, and finally
drying in vacuum desiccator over sulphuric acid. 1-
3. The results of the viscosity experiments indicate that: (a)
Pectin, acid, and sugar independently or combined increase the viscosity
of water, (b) Jelly will form from a mixture of acid, pectin, and sugar
when in certain concentration, (c) Jelly will form when pectin and
sugar are dissolved in certain concentration, (d) The viscosities of the
various substances are not additive, (e) The increase of viscosity
may be due to the "presence of the hydrogen ion. (/) That sugar may
have two functions, viz., that of a dehydrating agent and of increasing
the viscosity.
THE PRODUCTION AND USES OF PECTIN
PECTIN EXTRACTION2
Boyles of the Washington State University has perfected a process
and patented the methods of extracting pectin from apples and oranges.
The pectin is extracted with hot water and then precipitated from solu-
tion by the addition of alcohol. The alcohol is then evaporated and
impure pectin remains.
It has been shown by Bigelow, Gore, and Howard that six hours'
boiling with water under a reflux condenser, with changes of water
at the end of every hour, does not completely remove all pectin from
apple pulp previously freed of sugars and other materials soluble in
cold water, alcohol, or ether. (Pectin solution can be greatly acceler-
ated by heating under pressure in an autoclave.) The pectin is thus
dissolved in a very large volume of water, and the problem to be
solved is the reduction of this dilute solution to a small volume. Pectin
cannot be prepared by evaporating a dilute solution in water unless the
work is carried on in a partial vacuum at a low temperature, since long-
continued boiling brings about chemical changes which involve the loss
of gelatinizing properties. It could be precipitated from such solutions
by the addition of alcohol, but precipitation does not occur until the
1Singh ("Practical Experiments in Jelly Making," Journal of Industrial and
Engineering Chemistry, XIV [1922], 710) has carried on confirmatory experi-
ments.
Sucharipa ("Experimental Data on Pectin- Sugar- Acid Gels," Journal of
American Official Agricultural Chemists, VII [1923], 57-68) consider jelly forma-
tion to be due to the coagulation of pectin in the liquid sugar-acid medium, there
being no chemical reaction involved. Pectin is described as being very slightly
soluble in sugar-acid solutions of certain concentrations.
'United States patents involving pectin manufacture are: R. Douglas, No.
1082682, December 30, 1913; R. Douglas, No. 1235666, August 7, 1917 (pectin
solution) ; R. D. O. McDill, No. 1365000, January n, 1921.
94 CITRUS PRODUCTS
percentage of alcohol in the mixture has reached 51 per cent. This
would necessitate the use of large quantities of alcohol, impossible to
recover except by repeated distillation, and would be entirely too
expensive for general use.
The method employed by Caldwell for securing concentrated solu-
tions of pectin was suggested by Gore's method for concentrating fruit
juices. This method consists essentially in separation of a portion of
the water of the juice by freezing, removal of the concentrated liquid
from the ice by centrifuging with repetition of this process until the
desired concentration has been attained. The following description of
a typical experiment will make the method clear :
Twenty kilos (44 Ib.) of mature Winesap culls were ground to a fine
pulp in a small meat chopper, placed in a graniteware vessel, enough cold
water (3,000 c.c.) added to cover them, and slowly heated to boiling. After
two hours of gentle simmering the whole was placed in a double muslin
bag and suspended over a vessel to drain. When dripping had entirely
ceased the pulp was returned to the cooker, 2,500 c.c. cold water added,
and again slowly cooked. Four successive cookings of approximately two
hours each extracted the pectin so completely that the extract no longer
gave an appreciable precipitate of pectin when allowed to fall drop by
drop into strong alcohol. The pulp was then gently pressed to extract as
much liquid as could be forced out without forcing fragments of pulp
through the muslin, and the various extracts were collected and measured.
During the cooking 12,000 c.c. of water had been added and the total quan-
tity of extract measured 23,456 c.c. (24.8 qt.).
This liquid was now placed in tall, enamel-lined tin cans and subjected
to freezing by exposing it on a window-sill over night, at a temperature
of 15°— 2O°F. The cans were filled next morning by a mass of knifeblade-
like platelets and spicules of pure water ice, holding between them a quan-
tity of liquid containing all the solids of the extract. The cans were
emptied into a large graniteware vessel, the ice was crushed by the use
of a wooden mallet, and the liquid separated from the ice in an ordinary
cream separator turned slightly faster than for separating cream. Five
minutes sufficed to separate the liquid completely from a charge of ice.
This treatment reduced the original 23,456 c.c. to 9,308 c.c., which was
returned to the cans, again frozen and separated, yielding 5,626 c.c. A
third and a fourth freezing and separation reduced the volume to 2,260
c.c. The liquid was now of a very dark-brown color and of the consistency
of a thick syrup, as it contained not only the pectin, but also the sugar
and the coloring matter of the entire extract. (The nitrogenous constitu-
ents had been coagulated by the long heating to which the fruit had been
subjected, and hence did not pass into the extract.)
The extract was now further concentrated by placing it upon a shelf at
sucb a distance above a radiator that it would be kept at a temperature of
i58°F. Here it remained four days, becoming reduced to 1,509 c.c. or
less than one-fifteenth its original volume, when solidification at the sur-
face practically stopped further loss of water. The extract was now
THE PRODUCTION AND USES OF PECTIN 95
treated by the Goldthwaite method for precipitating pectin. The mass
was slowly poured from a beaker into a vessel containing 1,600 c.c. of 95
per cent alcohol, which caused precipitation of the pectin as a gelatinous,
rubber-like mass, the sugars and coloring matter for the most part re-
maining in the alcohol. The mass of pectin was collected on a cheese-
cloth filter, the alcohol worked out of it with a spatula, and it was then
washed with previously filtered alcohol which had been used in the pre-
cipitation, followed by small quantities of fresh alcohol. To further purify
it, it was now dissolved in 1,000 c.c. of lukewarm water, which required
occasional stirring for several hours to secure complete solution, and
reprecipitated by pouring into 1,000 c.c. of 95 per cent alcohol, after which
the alcohol was pressed out with a spatula. The mass of pectin was then
dried slowly at a temperature of I58°F. The dry pectin is a whitish
gray mass easily ground into a grayish powder which dissolves readily
in warm water, and has not undergone change upon keeping in a corked
bottle on the laboratory shelf for more than five months. A good jelly
was made by adding i per cent by weight of this powder to water in which
YZ per cent tartaric acid had previously been dissolved, adding 65 per cent
by weight of sugar, and boiling for fifteen minutes.
It occurred to Caldwell that if the activity of the concentrated extract
were reduced to such a point that the pectin would not be affected thereby,
it should be possible to keep such a preparation for a long period without
deterioration, as the very high content of sugar and the absence of nitrog-
enous material makes such an extract a very unfavorable medium for the
growth of fungi. This hypothesis was tested out in the following way:
12,000 c.c. of water extract made from 10 kilos of Ben Davis culls was
reduced by two successive freezings to 4,763 c.c. The liquid was trans-
ferred to a tall vessel and 12 gm. powdered lime carbonate (precipitated
chalk) was slowly added with constant stirring. Lime carbonate reacts
with the malic acid of the juice to form an insoluble salt, calcium malate.
As soon as the addition of the carbonate had been made, the liquid was
thoroughly stirred and divided into two equal parts in order that two
methods of removing the calcium malate might be tested.
One lot was immediately transferred to a graniteware vessel, brought
to a boil, and filtered through filter paper while boiling. Filtration was
effective in removing the calcium salt but was rather slow.
The second lot of extract was allowed to stand undisturbed for twelve
hours after the addition of the lime carbonate. At the end of that time
the calcium malate formed had collected at the bottom and upon the sides
of the vessel as a thin, granular precipitate. The clear liquid could be
siphoned off without disturbing this precipitate, and it was also possible
to decant without the loss of more than a few c.c.
The two portions of extract, after precipitation and separation of their
acid as the calcium salt, consisted of 2,365 c.c. each. They were kept sep-
arate, reduced to 725 and 743 c.c. respectively by two successive freezings,
then transferred to beakers and reduced by slow evaporation over a radi-
ator at 70 °C. to a volume of 400 c.c. each. They now had the consistency
and the color of tomato catsup. Two grapejuice bottles were sterilized
by thorough boiling and the contents of the two beakers were transferred
96 CITRUS PRODUCTS
to them and sealed. "These bottles have been kept upon an open laboratory
shelf for four and one-half months, have been twice opened and resealed
after removal of a part of the contents, and are at the present time free
from evidence of growth of bacteria or fungi, while the pectin retains its
power of causing gelatinization without decrease." Preparations made
without removal of the acid slowly lose their power to cause gelatinization,
losing somewhat more than half their efficiency in the first three weeks
after preparation. Preparations of dry pectin made by precipitating with
alcohol and drying at a low temperature undergo no discoverable change
in their efficiency when kept in sealed bottles for months.
The concentrated pectin solution made by freezing and evaporating is,
of course, very far from pure, since it contains not only the sugars present
in the fruit used but also the coloring matters and such flavoring sub-
stances as were not volatilized by heat. However, its addition to any fruit
juice will not produce greater change in color and flavor than would be
produced by the addition of an equivalent volume of apples. The powdered
pectin, on the other hand, is practically free from coloring matter, has no
discoverable apple odor or taste, and will not cause deterioration in flavor
of even the most delicate jellies, to which its use may well be restricted
because of the greater labor and cost of preparation.
Caldwell later developed a method for preparing pectin by the use
of ammonium sulphate as a precipitant. The principle of this method is
based upon the fact that pectin as extracted from the pulp or pomace is in
a colloidal state and can be readily changed by electrolytes. Since pectin,
after precipitation, must be dispersed again in order to be of any value
as a gelatininzing agent, an electrolyte that will produce a reversible pre-
cipitation must be chosen. Also the electrolyte chosen must be non-
poisonous. Lead acetate will precipitate pectin, but the precipitation is an
irreversible one, and the amount of lead absorbed or combined may be
poisonous. For these reasons ammonium sulphate was selected. Bourque-
lot and Herissey used ammonium sulphate as a precipitant for pectin ob-
tained from gentian root.
Sixty gm. of dried apple pomace were boiled three successive times
with 200 c.c. of water, filtering after each boiling. To each of the 100
c.c. of filtrate 25 gm. of ammonium sulphate were added1 and then heated
to 70° C., whereupon the pectin was precipitated as a grayish-white floccu-
lent precipitate. The precipitate was separated from the mother-liquor by
filtering. (The mother-liquor can be evaporated and the residue used again
or the residue can be used as a fertilizer.) The precipitate was dissolved
in hot water and again precipitated with ammonium sulphate. Again it
was filtered and the precipitate was removed from filter paper and dried
at 6o°— 70 °C. and when dry was washed several times with cold water to
remove adhering ammonium sulphate. The precipitate was dried again
and its gelatinizing power was tested by adding to a i per cent solution of
the pectin, 0.5 per cent solution of citric acid, and 65 gm. of sugar. This
*If wet pomace is used it will require a somewhat larger amount of ammonium
sulphate. First add 25 gm. per 100 c.c. and if precipitation does not occur, add
successive portions of 5 gm. until precipitation occurs. The pectin may also be
precipitated by saturating the solution in the cold with ammonium sulphate.
THE PRODUCTION AND USES OF PECTIN 97
solution was boiled for ten to twenty minutes and upon cooling a jelly was
produced. The taste did not indicate the presence of ammonium sulphate
and upon dissolving the jelly in hot water only a slight milkiness was
produced when tested for sulphates.
In order to determine whether the yield of pectin by the foregoing
method was equal to the yield produced by the alcohol precipitation method,
two samples of apple pomace from the same lot were treated exactly alike,
except that ammonium sulphate was used in one case and alcohol in the
other as the precipitating medium. The pectin was dissolved and repre-
cipitated in each case, then filtered, and the precipitate was removed from
filter paper and dried. The ammonium sulphate was removed from the
one by washing with cold water, again dried and weighed. The amount
of pectin recovered by each method is recorded in Table XIV below.
TABLE XIV
Pectin
Precipitant Per Cent
Ammonium sulphate 6.33
Alcohol 6.91
The amount of ammonium sulphate used can be reduced by concen-
trating the extract, either by evaporating on a steam bath, in a partial
vacuum, or by freezing. The quality of the pectin is not impaired in
either case.
It has been found quite difficult, however, to free the final pectin ob-
tained by this method from ammonium sulphate.
METHODS FOR PECTIN ESTIMATION
Perhaps the method most generally used at present is simply that of
precipitation from aqueous solution by alcohol, filtering, washing with
alcohol and ether, and drying in a desiccator over sulphuric acid. This,
however, requires a long time and is not absolutely accurate owing to
impurities and to the deliquescence of pectin.
If the pectin solution is free from impurities the amount of pectin
may be obtained by determining the viscosity of the solution, as shown by
McNair (p. 92).
T. B. Robertson suggested to the writer that a method might be evolved
for the determination of pectin through use of its refractive index. The
refractive index of a 2 per cent water solution of lemon pectin was found
to be 0.00130 per gm. per cent when observed in a Pulfrich refractometer
which indicates to within i' of the angle of total reflection. A sodium
flame was employed as a source of light. The solution was adjusted to
the temperature of the room so that the temperature of the refractometer
prism and that of the solution used would be the same. It has been shown
by Robertson that in alkaline or acid aqueous solutions of proteins the
change in the refractive index of the solvent is directly proportional to
the concentration of the dissolved protein. In respect to its refractive
index pectin acts similarly to protein. This relation can be expressed by
the equation :
n — HJ. = a X c
98 CITRUS PRODUCTS
when n is the observed refractive index of the pectin solution, n1 that of
the solvent in which the pectin is dissolved, c the percentage of pectin in
solution, and a a constant, expressing the change in the refractive index
of the solvent by the addition of i gm. of pectin per 100 c.c. In using this
method the refractive indices of impurities would also have to be
determined.
For most purposes the estimation of pectin through its methyl alcohol
content may be most satisfactory. This can be done as outlined on pages
80— 81 by making use of either Zeisel's or Deniges method for methoxyl
estimation.
JELLY MANUFACTURE1
To produce a jelly of the proper consistency, the fruit used must con-
tain a good supply both of pectin and acid, or if not a large supply of
pectin, a large amount of compounds that break down into pectin at the
temperature of boiling water. Some fruits were found by Cruess and
McNair to contain a sufficiency of acid and pectin while others lacked one
or the other of these necessary constituents. Where the fruits were low
in acid, attempts were made to produce jelly without addition of acid and
also with addition of citric acid or lemon juice. In most cases the pectin
was extracted in the usual way by cutting the fruit into small pieces, adding
water to cover, boiling slowly until tender and expressing the hot juice by
pressing in a small fruit press or by straining through a coarse cloth with
gentle pressure.
The solution so obtained was analyzed for Balling or Brix degree (per-
centage of dissolved solids) and acidity, and, if much more dilute than
the juice of the fresh fruit itself, was concentrated before being made into
jelly. In making jelly, sugar was added in most cases at the rate of one
and one-quarter volumes of sugar to one of solution. The mixture was
then heated to boiling and boiled down to a boiling point of 104°— IO5°C.,
or to a concentration of 65-70 per cent dissolved solids. At these concen-
trations, a jelly will form if the fruit is suited to the purpose.
Yields. — The amount of jelly obtainable from any fruit without addition
of acid will depend on the pectin and acid content of that fruit. Maximum
yields from several fruits were obtained by making from four to six suc-
cessive extractions of pectin by boiling the fruit with water and pressing.
The several extracts were kept separate and tested individually for their
jelly-making properties. They were also combined in various amounts to
ascertain the maximum amounts of pectin solutions from the last extrac-
tions that could be blended with the first two extracts and still give a
jelly. Oranges and lemons mixed in the ratio of two oranges to one lemon
gave a maximum yield of 1,585 c.c. jelly per 1,000 gm. of fruit, or approxi-
mately 392 gal. of jelly per ton of fruit, or 8,363 6-oz. glasses of jelly
per ton. Yields of 300 gal. of jelly from the mixed fruits have been ob-
tained often in the laboratory. Red loganberries gave a maximum yield of
'United States patents involving jelly manufacture are: F. L. Jefferies, No.
1045849, December 3, 1912; P. R. Boyles, No. 1067714, July 15, 1913 (jelly base,
pectin and acid mixed) ; R. Douglas, No. 1304166, May 20, 1919; M. O. Johnson,
No. 1362869, December 21, 1920 (concentrated jelly) ; R. D. O. McDill, No.
1365001, January n, 1921 (dried jelly).
THE PRODUCTION AND USES OF PECTIN 99
1,890 c.c. jelly per 1,000- gm. fruit or approximately 467 gal., or 9,962
6-oz. glasses per ton. Similarly, Mammoth blackberries gave 290 gal.
jelly per ton. With addition of acid, these yields were considerably in-
creased so that the yields were limited rather by lack of acid than lack of
pectin.
Clarification. — A jelly to be most attractive should be clear. Two
methods of clarification are in general commercial use. The most common
practice is to filter the hot pectin solution. Often the boiled fruit and
liquid are thrown together into some form of cloth or felt bag filter. This
process is slow and troublesome and does not ordinarily give a brilliantly
clear filtrate.
The second method consists in sterilizing the hot juice from the press
in 5-gal. cans. These are stored until the sediment deposits. This will
ordinarily be a period of several months. The settled juice is then decanted
or siphoned off from the sediment directly into the cooking kettles. The
process is often modified to the extent of giving the hot juice a bag filtra-
tion before canning.
Laboratory tests were made to ascertain the effect of Spanish clay and
infusorial earth on the rate of filtration. These substances were added in
powdered form to the hot juices and the rates of filtration and clearness
of filtrates compared with those of the untreated juice. The Spanish clay,
when added dry, gave an "earthy" taste and did not appreciably hasten
filtration. Infusorial earth, when added at the rate of 5 gm. per 100 c.c.,
greatly increased the rate of filtration and gave a clearer filtrate with one
filtration than could be obtained without the addition of this substance. It
was also found that the filtration through short fiber asbestos pulp in the
Seitz type of filter was very much more rapid and resulted in a clearer
filtrate than was obtained with the bag filter. A mixture of Seitz asbestos
No. 5 and Seitz "Brilliant" asbestos added to and mixed with the juice to
give a filtering layer about ^ in. thick gave good results. The flavor of
the juice is not impaired. Filtration in all cases is made before any sugar
is added to the juice.
Jelly stocks from loganberries, currants, and a mixture of oranges and
lemons were prepared by boiling the fruits with a small amount of water
until soft and pressing through a coarse cloth. The oranges and lemons
were mixed in the ratio of two oranges to one lemon before boiling. The
juices were divided into small portions. To these portions were added
casein from a 2 per cent solution in dilute NH4OH at the rate of 20, 40,
60, and 100 gm. per hectoliter, respectively. To other portions were added
egg albumen in the same amounts as noted for casein. To others were
added 250, 500, 1,000, 1,500, and 2,000 gm. of Spanish clay per hectoliter
from a 10 per cent suspension of this substance in water. Untreated checks
were also prepared. The various lots were bottled and sterilized one-half
an hour at ioo°C. It was found that the lemon-orange jelly stock settled
very satisfactorily in twenty-four hours after sterilization without the
addition of any clarifying material. The untreated loganberry and currant
checks did not settle very satisfactorily in twenty-four hours, but after sev-
eral weeks' storage did settle fairly well. Currant juice prepared by heating
the crushed fruit to 85 °C, pressing and sterilizing at 85 °C, settled better
ioo CITRUS PRODUCTS
than the juice from the same fruit heated to ioo°C. The casein and egg-
albumen findings gave poor results in all cases and resulted in making the
liquids more cloudy than the untreated checks. Spanish clay at 250 gm.
per hectoliter did not aid in clearing; 500 gm. per hectoliter seemed to aid
considerably in clearing. In some cases, 1,000 gm. per hectoliter gave a
perfect clarification in twenty-four hours; in other cases, 1,500 to 2,000 gm.
of Spanish clay per hectoliter were needed to effect clearing of the juice.
This was especially true of loganberry juice. In applying this method in
practice, preliminary tests on a few hundred c.c. of the material should
be made before clarifying any large amount.
The Spanish clay was prepared by soaking it in a small amount of
water into a finely divided, thin "mud" or suspension. This can be made
to a definite concentration, e.g., 20 per cent or 10 per cent mixtures were
found satisfactory. The flavor of the juice is not impaired by the use of
the clay in this form.
Preliminary tests with fire clay indicate that this substance can prob-
ably be used in a way similar to that employed with Spanish clay.
Aroma and flavor changes in jelly -making. — Fruit jellies when made in
the ordinary way usually differ in aroma and flavor from the fresh fruits
from which they are made. Ordinarily, jellies are made by extracting the
pectin by boiling, followed by addition of sugar and boiling until the mix-
ture boils at about 220 °F., or until the mixture jells, or until a Balling or
Brix degree of 65 (corrected for temperature) is obtained. The high
temperatures result in marked changes in flavor, and give a so-called
"cooked" taste to the product. Experiments have proved that the usual
high temperature of the second boiling is not required in all cases and that
the fruit flavor may then be retained.
It was thought that this change in flavor and aroma might be due in
part to hydrolysis and in part to loss by volatilization. To throw light on
loss by evaporation, attempts were made to make jelly -from currants,
loganberries, blackberries, and a mixture of orange and lemons, at tem-
peratures of room temperature, 60°, 70°, 80°, 90°, and ioo°C. The fruits
were crushed and heated in water- jacketed aluminum pots to the tempera-
tures indicated and pressed. Cane sugar was added to increase the Brix
degree to 65 and to dissolve the sugar the juices were heated to the respec-
tive temperatures indicated.
To note whether loss of flavor was also due to hydrolysis, a second
series at the same temperatures indicated above was carried out in a
5oo-c.c. flask fitted with a long water-cooled reflux condenser.
The jellies made by the two methods were compared shortly after they
were made. The jellies made in the open kettle at the lower temperatures
were superior to those made at the higher temperatures as regards amount
of fresh fruit flavor and aroma retained. The same applied to the jellies
made under the reflux condenser. After two to three months' storage, the
differences in flavcr and aroma were not so pronounced. The orange
jelly, after long storage, developed a "turpentine"-like taste, probably due
to oxidation of the orange oil.
The jellies made at room temperature by the addition of sufficient
sugar (to 65° Brix) and citric acid (to a total of l/2 per cent) to jelly
THE PRODUCTION AND USES OF PECTIN 101
stocks were especially close to the flavor of the fresh fruits from which
they were made. Loganberry and in one or two cases blackberry and
orange jelly were thus made without application of heat, but strawberries
did not yield a jelly in any case. Loganberries very readily give a highly
flavored aromatic and firm jelly in this way. So far as known, this method
has not been described before.
The facts that jellies made at high "temperatures (85°— 105° C.) were
poorer in fresh fruit flavor than those made at room temperature and 60 °—
75 °C, and that jellies made under the reflux condenser were stronger in
fresh fruit aroma and flavor than jellies made at the same temperatures
in an open kettle, indicate that loss of flavor and aroma is due both to
volatilization and to decomposition of flavoring and aromatic bodies.
Sugar in jelly-making. — The addition of sugar in jelly-making is neces-
sary to raise the concentration of dissolved solids sufficiently to cause the
formation of a gel with the pectin. The amount of sugar necessary will
depend upon the pectin and acid concentrations, but the amount used nor-
mally is controlled by the concentration of sugar necessary to prevent fer-
mentation or molding. Tests with loganberry and orange jellies showed
that jellies could be obtained at 60° Brix or lower, but that in such cases
molding or fermentation took place unless the jellies were sterilized in
sealed containers. Jellies with varying amounts of sugar were inoculated
with mold and yeast and caps were placed on the glasses. It was found
that 65° Brix prevented spoilage,, but that much below this concentration
mold growth took place. Where the "Sugar Concentration was increased
from 72° to 75° Brix crystallization of sugar often took place. This
crystallization depends on the amount of cane sugar present. The tendency
to crystallize was most noticeable in jellies of low acid content and in
which little hydrolysis of the cane sugar had probably taken place. Where
glucose is used for jelly-making, the concentration may be considerably
above 70° Brix without crystallization.
A simple method for calculating the amount of sugar to add to a
juice of a given degree Brix to bring the liquid to 65° Brix is as follows:1
Let a = Brix of juice.
V = Volume of juice.
5" = Grams sugar necessary to bring to 65° Brix.
Then (65 — a)F/35 = S.
Example :
V — 250 c.c. a = 5 Brix.
(65 — 5) 250/35 — 5" = 428.5 gm. sugar.
For practical purposes, a little more than one and one-half of sugar to
one of juice by weight will be found satisfactory.
If this method is used, the fruit juice cannot be so dilute as where a
mixture of one of sugar to one of juice is used and where the pectin is
concentrated by boiling off excess water. The juice must contain enough
aA more exact formula is y — 1. 4^ + 0.0025 12jr3 where x = degrees Brix
of syrup and 3; = the ounces of sugar added per gallon of water at 17 °C. (J. B.
McNair, "Syrups for Canning and Preserving," Journal Industrial and Engineer-
ing Chemistry, IX [1917], 151).
IO2 CITRUS PRODUCTS
pectin to give a jelly without concentration of the juice after addition of
sugar.
Sugar added to the crushed fruit before pectin extraction seemed to
help retain the flavor and the jelly so made appeared to be superior in this
respect to jelly made by the usual methods.
A solution of 65° Brix will boil at IO3.9°C. or 2i9.o2°F. at sea-level.
This point can be used to determine the proper density of the boiling jelly
if the proper allowance is made for elevation of locality in which the jelly
is made. The boiling-point of water for any given locality may be deter-
mined and then the boiling-point of jelly at the proper concentration will
be 3.9° C. or 7.02° F. above the boiling-point of water.
Acid in jelly-making. — It was found that pectin solutions made by
boiling and pressing fruit and containing less than 0.5 per cent acid (as
citric or tartaric) usually did not jell, but it was also found that if the
acid were increased to 0.7 or 0.8 in such cases, it almost invariably gave a
jelly. This was noticed especially with citron melon and fig solutions,
both naturally deficient in acid, but containing sufficient pectin for jelly-
making. Tests made with a I per cent solution of pure orange pectin in
distilled water indicated that a concentration of 0.3 per cent acid in the
final jelly was necessary to give a firm jelly. Increase of acid in the final
jelly to 1.9 per cent acid resulted in softening of the jelly; i per cent acid
in the final jelly gave good results and corresponds to about 1.5 acid in a
fruit juice before addition of sugar; 0.3 per cent acid in the final jelly
would similarly correspond to about 0.5 per cent acid in the fruit juice
before addition of sugar. This would indicate a range of 0.5 to 1.5 per
cent acid in the original fruit juice for the proper concentration of acid
calculated as citric acid. This corresponds very well with results obtained
with pectin-containing orange juices made by boiling this fruit and press-
ing. In this case an acidity of 1.5 per cent in the juice gave a finished
jelly that set quickly but which finally softened, while 0.5 per cent gave a
fairly good jelly, and less than 0.5 per cent did not give a jelly or resulted
in a very soft jelly.
Working with purified pectin and different amounts of pure citric acid,
it was found that for the same treatment as regards amounts of sugar
added, time of boiling, etc., the color of the resulting jelly was darkened
in direct proportion to the amount of acid used. The colors were com-
pared by means of a Dubosq colorimeter. This darkening is probably due
to caramelization of the sugar.
REFERENCES FOR CHAPTER III
CHEMISTRY OF PECTIN
AUDRLK.
Zeitschrift Zucker-Ind. Bohm, XIX (1894), 101.
Chemisches Zentralblatt, LXVI (1895), i, 833.
BAILEY, F. S.
Science (June 26, 1914).
REFERENCES
BALL, NIGEL G.
"On the Action of Pectase," Scientific Proceedings of the Royal Dub-
lin Society, XIV, No. 28 (1914), 349-57-
BANDRIMONT.
Journal de Pharmacie, Vol. XII, Series 3, p. 25.
BARFOLD.
Journal fur praktische Chemie, Vol. XI, Series 2 (1875), p. 186.
BATTIET.
Sucrerie indigene, XXXII (1888), 285, 311, 333, 357, 415, 456.
BAUER.
Journal fur praktische Chemie, CXXXVIII, 367.
Landwirtschaftliche Versuchs-Stationen, XXXVIII, 319.
Ibid., XLI (1892), 477.
Ibid., XLIII (1894), 191.
Journal fur praktische Chemie, Vol. XLIII, Series 2 (1891), p. 112.
Chemisches Z entralblatt , Vol. LXXII, Series 2 (1901), p. 196.
BERTRAND and MALLEVRE.
"Sur la pectase et sur la fermentation pectique," Comptes Rendus,
CXIX (1894), 1012.
"Nouvelles recherches sur la pectase et sur la fermentation pectique,"
ibid., CXX (1895), 1 10.
"Sur la diffusion de la pectase dans le regne vegetal sur la preparation
de cette diastase," ibid., CXXI (1895), 726.
BIGELOW, W. D.
Journal of the American Chemical Society, XXIII, 347.
BIGELOW, W. D., GORE, H. C., and HOWARD, B. J.
"Studies on Apples," United States Department of Agriculture Bulletin
No. 94, Bureau of Chemistry (1905), pp. 67-89.
BOURQUELOT.
"Sur les pectines," Journal de pharmacie et de chimie, IX (1899), 563.
Comptes Rendus, CXXVIII (1899), 1241.
BOURQUELOT and HERISSEY.
Journal de pharmacie et de chimie, Vol. VII, Series 6 (1898), p. 473.
"De 1'action des ferments solubles sur les products pectiques de la
racine de gentiane," ibid., Vol. VIII, Series 6 (1898), p. 145.
Ibid., Vol. IX, Series 6 (1899), p. 281.
"Sur la pectine de cynorrhodon," ibid, X (1899), 5.
BRACONNOT.
Annales chim. phys., Vol. XXVIII, Series 2, p. 173.
Annalen de chimie et de physique, Vol. XXVII, Series 2 (1831), p. 266.
Ibid., Vol. L, Series 2 (1832), p. 376.
Liebigs Annalen der Chemie, V (1883), 275.
BROWN, C. A.
Handbook of Sugar Analysis. New York: John Wiley & Sons.
BROWN, C. A., JR.
"A Chemical Study of the Apple and Its Products," Pennsylvania State
Department of Agriculture Bulletin 58 (1899), PP- 1—46.
IO4 CITRUS PRODUCTS
CALDWELL, J. S.
"A New Method for the Preparation of Pectin," Agriculture Experi-
ment Station (Pullman, Washington) Bulletin 147 (1917).
CHEVRON.
Chemiker-Zeitung R., XII (1888), 82.
Neue Zeitung fur Rubens., XX (1888), 169.
Sucre beige, Vol. XIII (1888).
CHODNEW.
Liebigs Annalen der Chemie, LI (1844), 355.
CROSS.
Berichte der chemischen Gesellschaft, XXVIII (1895), 2609.
EULER, H.
General Chemistry of the Enzymes. English translation by T. H. Pope.
New York: John Wiley & Sons, 1912.
FELLENBERG, TH. VON.
Mitteilungen Lebensmittel Hyg., Vol. V, Part I (1914), pp. 225-26.
Chemical Abstracts, IX, 488.
FREMY.
"Recherches sur la pectine et 1'acide pectique," Journal de pharmacie et
de chimie, Vol. XXVI, Series 2 (1840), p. 368.
Ibid., Vol. XII, Series 3 (1847), p. 13.
Comptes Rendus, XXIV (1847), 1046.
Liebigs Annalen der Chemie, LXIV (1847), 383.
Annalen de chimie et de physique, Vol. XXIV, Series 3 (1848), p. 5.
Journal fur praktischc Chemie, XLIV (1848), 385.
Liebigs Annalen der Chemie, LXVII (1848), 257.
Comptes Rendus, XLIX (1859), 561.
Journal de pharmacie et de chimie, Vol. XXXVI, Series 3 (1859), p. 5.
FROMBERG.
Liebigs Annalen der Chemie, XLVIII (1843), 56.
Journal fur praktische Chemie, XXXII (1844), 179.
Berselius Jahresbericht, XXV (1846), 565.
Scheik. onderz. d. Utrechtsche Hoogesch, II Decl., iste stuk.
GIRARD.
Berichte der chemischen Gesellschaft, VIII (1875), 340.
Comptes Rendus, Vol. LXXXa, p. 477.
GREEN, R.
Science Progress, VI (1896), 344.
GUIBERT.
Schweigg. Journal, XLIV (1827), 136.
HAAS, R. W. TROMP DE.
Untersuchungen uber Pectinstoffe, Cocosschalen, Oxycellulose. Dis-
sertation, 1894.
HAAS, R. W. TROMPE DE, and TOLLENS.
Liebigs Annalen der Chemie, CCLXXXVI (1895), 278.
HARDY, W. B.
Journal of Physiology, XXIV (1899), 158.
Journal of Physical Chemistry, IV (1900), 254.
* REFERENCES 105
HARLAY, V.
Journal de pharmacie et de chimie, Vol. V, Series 7 (1912), pp. 344-47.
Journal of the Chemical Society (London), Vol. CII, Part 2 (1912),
p. 479.
HEBERT.
Annales Agron., XXVI (1900), 34-50.
HERZFELDT.
Zeitschrift Ver. d. Zucker-Ind., XLI (1891), 295, 667.
HUNT, CHARLES H.
"A Method for Preparing Pectin," Science, XLVIII (1918), 201.
JAVILLIER.
Journal de pharmacie et de chimie, Vol. IX, Series 6 (1899), pp. 163,
5i3-
KIRCH NER and TOLLENS.
Liebigs Annalen der Chemie, CLXXV (1875), 205.
KOLB.
Jahresbericht der Chem. Technologie (1868), p. 611.
LIPMAN.
Die Chemie der Zuckerarten (1895), PP- 924-36.
LIPPMAN, VON.
Berichte der chemischen Gesellschaft, XX (1887), 1001.
Die Chemie der Zuckerarten (3d ed.), II, 1602-17. Braunschweig,
1904.
McNAiR, J. B.
"Factors Determining the Consistency of Pectin Gels," Journal of
Physical Chemistry, XX (1916), 633-39.
MANGIN.
Comptes Rendus, CVII (1888), 144.
Ibid., CIX (1889), 579.
Ibid., CX (1890), 295, 644,
Ibid., CXVI (1893), 653.
Journal Bot., V (1891), 400, 440.
Ibid., VI (1892), 12.
7Wd./VII (1893), 37, 121, 325.
MULDER.
Liebigs Annalen der Chemie, XXVIII (1838), 280.
MUNTZ.
Annalen de chimie et physique, Vol. X, Series 6 (1887), p. 566.
PAYEN.
Annalen de chimie et physique, Vol. XXVI, Series 2 (1824), p. 329.
Rev. sav. etrang., Vol. IX, Series 2 (1846), p. 148.
Comptes Rendus, XLIII (1856), 769.
PELLET.
Sucrerie indigene, XXXII (1888), 390.
POUMAREDE.
Comptes Rendus, X (1839), 660.
Ibid. (1847), p. 17.
Rev. scient. Quesneville, Vol. XIV, Series 2 (1848), p. 68.
Ibid., Vol. XV, Series 2 (1849), p. 98.
io6 CITRUS PRODUCTS •
POUMAREDE and FlGNER.
Comptes Rendus, XXIII (1846), 918.
Journal de pharmacie, III (1847), 12.
REGNAULD.
Journal de pharmacie et de chimie, XXIV (1838), 201.
Journal fur praktische Chemie, XIV (1838), 270.
Journal de pharmacie, Vol. XII, Series 3, p. 27.
REICHARDT.
Berichte der chemischen Gesellschaft, VIII (1875), 807.
Archiv der Pharmacie, CCIX (1876), 97.
Ibid., CCX (1877), 116.
ROBERTSON, T. BRAILSFORD.
Zeitschrift Kolloidchemie , III (1908), 49.
"The Proteins," University of California Publications in Physiology,
III (1909), 115.
Journal of Physical Chemistry, XIII (1909), 469, 473.
Journal of Biological Chemistry, VII (1909-10), 359.
Ibid., VIII (1910-11), 287, 441, 507.
Ibid., XI (1912), 179, 307.
Ibid., XIII (1912-13), 455, 499.
Die physikalische Chemie der Proteine (1912), pp. 269, 317.
ROBERTSON, T. B., and GREAVES.
Journal of Biological Chemistry, XIX (1911), 181.
SCHEIBLER.
Neue Zeitschrift fur Rubens, III (1879), 341.
Berichte der chemischen Gesellschaft, I (1868), 58, 108.
Ibid., VI (1873), 612.
SOUBEIRAN.
Journal fur praktische Chemie, XLI (1847), 309.
Annalen der Chemie et Pharmacie, Vol. LI, p. 355.
STUDE.
Liebigs Annalen der Chemie, CXXXI (1864), 241.
TOLLENS.
Handbuch der Kohlenhydrate (1888 ed.) pp. 242—46; (1895 ed.), pp.
242-47.
Bulletin de la Societe botanique de France (1889), p. 388.
Liebigs Annalen der Chemie, CCLXXXVI (1895), 292.
TOLMAN, RICHARD C.
Journal of the American Chemical Society, XXXV (1913), 317.
ULLIK. ; ^
Osterr.-Ung. Zeitschrift Zucker-Ind. Landw., XXI, 564.
Ibid., XXIII, 268.
Chemie der Zuckerarten von Lippman (1895), PP- 927» 92&
VANQUELIN.
Annalen de chimie et de physique, Vol. XLI, Series 2 (1829), p. 46.
VOTECEK and SEBOR.
Zeitschrift Zucker-Ind. Bdhm, Vol. XXIV (1899).
Chemisches Zentralblatt, Vol. LXX, Series 2 (1899), p. 1022.
Journal of the Chemical Society, LXXVIII (1900), 208.
REFERENCES 107
WEISBERG.
Neue Zeitschrift fur Rubens., XXI (1888), 325.
WIDTSOE and TOLLENS.
Berichte der chemischen Gescllschaft, XXX (1900), 132.
WOHL and VAN NIESSEN.
Zeitschrift Ver. d. Zucker-Ind., XXXIX (1889), 924.
PECTIN IN REGARD TO PLANT PHYSIOLOGY
AGARDH.
Essai sur developpment interieur des plantes (Lund, 1828).
CZAPEK, FREDERICK.
Biochemie der Pflanzen, p. 546.
DsBARY.
Vgl. Anat., p. 126.
GARDINER.
"The Continuity of the Protoplasm in Plant Tissue," Nature (1885),
P- 391-
HOFMEISTER.
PHanzenzelle (1867), p. 241.
ISCKII.
Imperial University, College of Agriculture Bulletin, II (1894), 97.
Tokyo. • '•';.;'£.'.
KABSCH.
Untersuch. uber d. chem. Beschaffenh. de. Pfianzengewebe mit Bezug
auf d. neuesten Arb. Fremys. Zurich, 1863.
KOENIG.
Chemie der menschl. Nahrungs und Genussmittel, II, 167.
LURSSEN.
Botanische Zeitschrift (1873), pp. 641, 644.
MANGIN.
Bulletin de la Societe botanique de France (1889), pp. 279, 388.
Journal de Botan., VII (1894), 37. Also in volumes for 1891, 1892,
1893-
MEYER.
Pfianzenphysiologie, Vol. I, pp. 160 ff.
Wiegmanus Archiv (1835), p. 151.
Ibid., II (1837), 30.
MOHL.
Beitrage zur Anatomie und Physiologie der Gewachse (1834).
Ober d. Verbing d. Zellen untereinander. Dissertation, 1835.
MULDER.
Poggendorfs Annclen, XLIV (1838), 432.
Berzelius Johresbericht, XVIII (1839), 283.
Journal fur praktische Chemie, XIV, 277.
Liebigs Annalen der Chemie, V (1838), 278.
Physiologische Chemie (1844), p. 244.
ROSENBERG-HEIN, EDGAR.
Pectinmetamorphose. Inaugural Dissertation, 1908.
io8 CITRUS PRODUCTS
Russow.
"Uber die Anskl. der Interzellularen," Zeitzber. Dorp. Naturf. Ges.
(1884), p. 19.
SCHENK.
Berichte dcr botanischen Gesellschaft, III (1885), 217 ff.
t)ber die Stabchen der Marattiaceen," ibid., IX (1886), 86 ff.
SCHLEIDEN.
Grundsiige, etc. (4th ed., 1861), p. 122.
Botanik, II, 328.
SCHRODER, BRUNO.
Botanische Zentralblatt, Ber-Hefte, X (1901), 123.
TSCHIRCH.
Angewandte Pnanzen-anatomie, p. 181.
TSCHIRCH and OESTERLE.
Anatomischer Atlas, p. 45, Table XII.
UNGER.
Vber die Lentizellen-Flora (1836), pp. 577-604.
Botanische Zeitschrift (1847), P- 2%9-
VALENTIN.
Repertor. fiir Anatomic und Physiologie, I (1836), 96.
VOGEL.
Wien Akademie, Vol. XLVIII (II) (1863), p. 672.
WlESSNER.
Wien Akademie, VoL XLVIII (II) (1864).
WlGAND.
Interzeltularsubstanz und Cuticula (1850).
JELLY-MAKING
BlGELOW, W. D.
Bureau of Chemistry, United States Department of Agriculture Bul-
letin 66, p. 50.
CRUESS, W. V., and McNAiR, J. B.
"Jelly Investigations," Journal of Industrial and Engineering Chemis-
try, VIII (1916), 417-21.
GOLDTHWAITE, NELLIE E.
"Contribution to the Chemistry and Physics of Jelly-Making," Journal
of Industrial and Engineering Chemistry, I (1909), 333-40.
"Studies on Jelly and Jelly-Making," ibid., II (1910), 457-62.
"The Principles of Jelly-Making," University of Illinois Bulletin II,
No. 31 (4th ed., March, 1915), pp. 1—23.
McNAiR, J. B.
"Failures in Jelly-Making and Their Remedies," California Cultivator,
XLVI( 1916), 736-37-
"Syrup for Canning and Preserving," Journal of Industrial and En-
gineering Chemistry, IX (1917), 151-53.
PACRETTE.
The Art of Canning and Preserving as an Industry.
REFERENCES 109
PARLOA, MARIA.
"Canned Fruit, Preserves, and Jellies : Household Methods of Prepa-
ration," United States Department of Agriculture Farmers' Bulletin
203, pp. 1—31 (1904; reprinted without change, 1916).
ROLET.
Les Conserves de fruits.
SNOW, JENNY H.
"Effect of Sugar and Temperature on Fruit Juices," Journal of Home
Economics, I (1909), 261-66.
CHAPTER IV
OTHER PRODUCTS FROM THE RIND
The rinds of the orange, lemon, and citron are made into articles of
commerce by candying or drying.
For this purpose the peel is separated from the fruit either by hand
or machine1 and is dried either by laying it in trays in the sun, as in
fruit drying, or by a current of heated air. Market for this product is
found with wholesale druggists, although some is sold for use in dried
mince meat and similar preparations. Candied peel is made somewhat
similar to glaced fruit.
Glaced kumquats are prepared in China and Japan. The fruit is
picked when ripe and several slits are made in the sides, sometimes the
seeds are squeezed out through the slits, sometimes not. The fruit is
next boiled until tender and then placed in a dilute sugar solution. From
this it is soaked in a series of sugar solutions of gradually increas-
ing density. They are next dried and then dipped into a very hot
syrup and again dried. The final product has the sugar glaced on the
surface of the fruit (Coit). Glaced kumquats acquire an unattractive
"turpentine" taste in time, due to the decomposition of the rind oil.
Some of the candied citron of commerce is prepared where the fruit
is grown, chiefly in Mediterranean localities, the principal production
centering in Leghorn, Italy, and in Corsica, but far the greater propor-
tion of candied citron is in recent years manufactured in England or in
the United States, the Corsican or Italian fruit being imported preserved
in brine.
The fruit is simply cut in halves, the seeds scooped out, and without
peeling the fruit is thrown into brine, cooped up, and shipped in casks.
Though either green or ripe fruit may be used, the demand is mostly
for fruit having a green color and therefore the green fruit is preferred.
aOne of the most successful types of machines used consists of an axle to which
is attached in the middle a steel pulley 3 or 4 in. in diameter, and on each end a
conical bur. The axle passes through the centers of the bottoms of two aluminum
tea kettles each of which surrounds a bur. The spouts of these kettles are turned
toward the floor and serve as exits for the juice. In operating the machine the
burs revolve about 1,000 times per minute, and a half of a fruit is held against each
bur. The burs are frequently made of apple wood although often of bronze. The
machine requires about one-half a horse-power to operate. An experienced man
can handle one lug box (60 Ib.) of lemons in twelve minutes or twenty-five boxes
in eight hours. If the capacity of the machine be considered from the standpoint
of the juice secured, it can make 35 gallons of lemon juice in one hour.
1 10
Ill
The candying process involves boiling in water to remove the salt
and soften the fruit which is then immersed in cold water to intensify
its green color. It is next covered with hot syrup and allowed to stand
three or four weeks, during which time the strength of the syrup is
gradually increased or in practice the' fruit is transferred from tank to
tank with a gradually increasing sugar concentration and the fruit boiled
with syrup, cooled, more sugar added, and again boiled until no more
sugar is taken up. The pieces of fruit are then dried, coated with white
sugar crystals, wrapped in tissue paper, and packed in wooden boxes.
The annual consumption in the United States averages about 4,000,-
ooo Ibs.
In Dominica a small business is done in shipping limes pickles in
sea- water. These limes go chiefly to Boston. The trade is only a small
one, and during late years the export of pickled limes has fallen off
somewhat. This is probably not due to a decreasing demand for pickled
limes, but rather to increasing shipments of this product from other
TABLE XIII A
IMPORTS OF CITRON OR CITRON PEEL INTO THE UNITED STATES
1898-1923
112
CITRUS PRODUCTS
TABLE XIII B
IMPORTS OF ORANGE AND LEMON PEEL INTO THE UNITED STATES
1898-1923
West India islands. The average annual export of pickled limes from
Dominica for the five-year period ending 1896, was 1,505 casks, for a
similar period ending 1901, 1,117 casks, and for five years ending 1906,
1,000 casks. A cask holds about 2,000 limes.
For pickling, the finest specimens of sound yellow limes are selected
and placed in vats into which sea-water is pumped in two or three days,
this water is run off and fresh sea-water pumped in. This process is
repeated several times until the limes are cured. The fruit is then placed
in casks which are in turn filled with sea- water to which a small amount
of salt has been added. The casks are then closed, and are ready for
export.
CHAPTER V
PRODUCTS FROM THE PULP1
The pulp of citrus fruit serves for a variety of products. That of
oranges and grapefruit is preserved by canning; the juice of these as
well as of limes is extracted and bottled as such, or mixed with other
fruit juices, or concentrated; citrus juice, particularly that of oranges,
is converted by fermentation into wine, alcohol, vinegar, or acetone;
citric acid is manufactured from the juice of lemons and limes.
CANNED PEELED FRUIT
Owing to higher cost and poorer flavor canned citrus fruit cannot
compete directly with fresh, but it is acceptable both in flavor and
appearance. It serves to supply vessels taking long voyages, mining,
lumber and construction camps, and other inaccessible places. In Cali-
fornia the orange canning season for navel oranges ranges between
February and June, and for Valencia oranges between May and Sep-
tember. The season for canning grapefruit in Florida naturally co-
incides with height of the shipping season there. Fruit canned when
immature becomes bitter and poor in flavor; on the other hand, fully
ripe oranges having a good full flavor are sometimes lacking in acidity.
This deficiency is remedied by the addition of lemon juice.
Oranges for canning are peeled by cutting the rind around the center,
dropping in hot water to separate skin from pulp, and slipping the
peels off by hand. They are next sliced and placed for a certain time
in a definite strength of cane sugar syrup to impregnate the fruit with
syrup. They are then placed in cans, covered with a 32 per cent syrup,
exhausted, capped, and sterilized at a moderate temperature.
It should be noted that oranges when heated above i85°F. have
their flavor greatly spoiled. The orange oil should not enter the
product as it decomposes, producing what is called a "turpentine"
taste. Sterilization is accomplished by heating the fruit for three hours
at I2O°F. or for twenty minutes at I45°-I55°F.
JUnited States patents pertaining to the use of pulp are: D. F. Sherman, No.
744810, November 24, 1903 (dried with binder) ; D. F. Sherman, No. 793615,
June 27, 1905 (dried by heat) ; H. A. Shepard, No. 1257410, February 26, 1918
(concentrated pulp by heat) ; D. F. Sherman, No. 793614, June 27, 1905 (dried by
heat) ; P. R. Welch, No. 1277672, September 3, 1918 (preserved by heat) ; R. D.
O. McDill, No. 1361079, December 7, 1920 (concentrated pulp paste with addition
of sugar).
"3
H4 CITRUS PRODUCTS
JUICES, PRESERVED1 AND CONDENSED
Orange juice apparently satisfies the average taste best when it
contains eight parts of sugar to one of acid and, therefore, must be
secured when the fruit is fully ripe. Bitterness must also be guarded
against; oranges from the middle of the season until its end do not
give bitter juice. Lemon, grapefruit, and lime juices also vary accord-
ing to their degree of ripeness. They, too, are liable to become bitter.
None of these fruit juices will ordinarily retain their natural flavors
longer than twelve hours and many methods for their preservation have
been undertaken. It is a comparatively easy matter to clarify these
juices, bottle, and preserve them against fungus attack, but such products
have not the color, flavor, or aroma of fresh juice. In attempting to
find a suitable method for preservation many procedures have been
tried, e.g., pasteurization, sterilization by heat, ozone, sulphurous acid,
carbon dioxide, fluorides, benzoates, and the ultra-violet rays, together
with the addition of sugar or partial fermentation.
One of the chief defects so far insurmountable in this line has been
a "limey" or "musty" taste to the juice. The cause of this limey taste
has been frequently attributed to oxidation. If such is the case it
would seem that the condition should be prevented by the presence of
reducing agents. Experiments conducted at the University of Cali-
fornia by Cruess and his students indicate that oxidation is not the cause
of the limey taste. Immature oranges will yield bitter juice while
oranges fully mature do not do so. The presence of a reducing agent,
such as sulphur dioxide, when added as soon as the juice is pressed
from the fruit does not prevent the formation of the limey taste, nor
does a similar addition of carbon dioxide prevent its development.
The addition of sugar to orange or lemon juice helps to retain the
flavor, but does not result in a thoroughly saisfactory product. Juices
which contain oil from the rind tend to develop a "turpentine" taste.
Lemon and grapefruit juices tend to darken in the bottle, especially
if the bottles are not well filled. Darkening, in opposition to the devel-
opment of the limey or musty taste, is apparently caused by oxidation
and may be prevented by the addition of a small amount of sulphurous
acid.
'United States patents involving preservation of fruit juices are: Hans Zurck,
No. 1045245, November 26, 1912 (by heat) ; W. Rueff, No. 1066153, July I, 1913
(with COO ; E. Monti, No. 1758266, October 26, 1915 (by heat) ; C. E. Burke,
No. 1097442, September 5, 1916 (with SO2) ; A. Fonyo, No. 1207041, December
5, 1916 (with lactic acid) ; H. C. Gore, No. 1284187, November 5, 1918 (by
absorbent black) ; P. R. Welch, No. 1258627, March 5, 1918 (by heat) ; F. L.
Dunlap and R. Kuever, No. 1338684, May 4, 1920 (by proteolytic enzymes and
heat) ; Max Henius and F. Mendelsohn, No. 1343915, June 22, 1920 (with SOz)-
JUICES, PRESERVED AND CONDENSED
The best and most palatable preserved juices seem to be those of
the grapefruit and lime.
ORANGE JUICE
Yields and composition. — A number of tests of sound and of frozen
oranges were made to determine the volume of juice which should be
expected from a ton of fruit. The yields were obtained by crushing the
pulp from weighed lots of oranges and pressing in a small meat press.
TABLE XV
YIELDS OF JUICE
The unfrozen oranges yielded nearly as much juice as can be extracted
from grapes; the frozen oranges only about 25 to 30 per cent as much.
The loss of juice in the frozen fruit is not due simply to concentration by
evaporation, but also to an actual disappearance of the juice with all its
constituents as shown by analyses in Table XVI.
TABLE XVI
COMPOSITION OF JUICE
These data, showing that the concentration of the solids in the juice
from the frozen and unfrozen oranges is about the same and the fact that
there is a much greater quantity of juice in the unfrozen oranges, indicate
a disappearance of both solids and liquid after freezing.
A comparison of samples i and 2 indicate a slight concentration due
to evaporation, but samples i and 3 show practically the same composition.
On the whole, the composition of the juice seems to be little affected by
the freezing of the oranges.
Orange juice according to these analyses, shows nearly three times the
amount of acidity found in grape juice and five or six times the amount
n6 CITRUS PRODUCTS
found in apple juice. The total solids are about two- thirds those of grape
juice and a little less than those of apple juice.
Clearing. — The juice should be made permanently bright, so that it will
have an attractive appearance when bottled. Fresh juice will not filter
easily and is difficult to make bright by filtration until it has stood a certain
length of time. The length of time necessary varies considerably, but, in
all the tests made in the laboratory, twenty-four to seventy-two hours was
the maximum variation.
The following observations bring out the effect of preliminary defeca-
tion by standing and settling on the clearing of the juice by filtration.
TABLE XVII
EFFECT OF STANDING FOR VARIOUS TIMES BEFORE FILTERING
Sample Observations
1. Fresh juice Filters slowly and filtrate is cloudy
2. Same juice after 18 hours. .Filters slowly and filtrate is cloudy
3. Same juice after 52 hours. . . .Filters slowly but filtrate is clear
4. Same juice after 76 hours. .. .Filters easily and filtrate is clear
Other samples of juice behave similarly, except that in most cases the
time necessary for defecation was less than the seventy-six hours noted in
Table XVII. One sample became jelly-like in a few hours after expressing
it from the oranges, but two days later the jelly-like material coagulated
and settled out, carrying down all suspended matter, leaving a perfectly
bright liquid above. Apparently a coagulating enzyme is active in bringing
about the clearing of the juice. Further evidence for the existence of such
an enzyme in orange juice is given by the fact that, if the juice is pasteur-
ized, it will not clear, but will remain cloudy until filtered or clarified by
other means. Many enzymes are destroyed by heat; therefore the fact
that unpasteurized juice clears of its own accord and the pasteurized un-
defecated juice does not indicates the presence of such an enzyme.
Several samples of the same juice were treated as follows:
1. Defecated fifty hours and then heated in bottles to i85°F.
2. Defecated fifty hours, egg albumen added at the rate of 5 oz. per
100 gal., then heated in bottle to i85°F.
3. Defecated fifty hours, casein added at the rate of 5 oz. per 100 gal.,
then heated in bottle to i85°F.
4. Not defecated; casein added at rate of 5 oz. per 100 gal. immedi-
ately after pressing and then heated in bottle to i85°F.
5. Not defecated; heated in bottle to i85°F. immediately after ex-
traction from the fruit.
Seven months later, samples I, 2, and 3 were bright, but 4 and 5 were
cloudy. These tests demonstrate the utility of defecation in clearing the
juice. (Compare especially tests 3 and 4.) They also indicate that finings
are unnecessary for the clearing of the juice. (Compare test i with tests
2 and 3.)
Three other small lots of juice were treated in the following ways:
6. Same juice as No. 5, defecated seventy-six hours with sulphurous
acid, filtered, pasteurized at i85°F. in bottle. This juice came from un-
frozen oranges.
FIELD MUSEUM OF NATURAL HISTORY. BOTANY, VOL. VI, PART I, PLATE IV.
Parfiimerie ^Inderne)
(Petite Revue)
COLLECTING ORANGE BLOSSOMS IN FRANCE.
JUICES, PRESERVED AND CONDENSED 117
7. Juice from frozen oranges tested in same manner as No. 6.
8. Same juice as No. 7. Not treated with sulphurous acid. Filtered
several times till bright and pasteurized twice at 185 °F.
Seven months after pasteurization, samples 6, 7, and 8 were all bright.
No. 6 was of a slightly lighter shade than No. 7. No. 8 was dark brown in
color. Samples 6 and 7 exhibited very little cooked flavor; No. 8, on the
other hand, had enough of the cooked taste to make it considerably inferior
to samples 6 and 7. Although not treated with sulphurous acid, sample 8
did not have a very pronounced bitter flavor, probably because it was
sterilized so soon after extraction, thus not allowing time for the develop-
ment of the bitter flavor.
In the settling-out process, or defecation of the juice, two serious
difficulties are met with. The first is that the juice will start to ferment if
allowed to stand long enough to defecate, unless treated in some way to
delay fermentation. Secondly, a bitter taste develops in the untreated juice
if it is long exposed to the air. Tests have shown that the addition of
moderate amounts of sulphurous acid will prevent fermentation for the
desired length of time and will also prevent the development of the bitter
taste. Since sulphurous acid acts in the opposite manner to the oxygen of
the air, it may be surmised that the development of the bitter flavor in
orange juice is due to the oxidation of some tasteless constituent flavor in
juice to a bitter form. An experiment indicated that the sulphurous acid
must be added very soon after the oranges are crushed in order to check
the bitter flavor, as, in this particular test, a bitter taste was perceptible in
one-half hour after the juice had been expressed. The amount of sulphur-
ous acid necessary in any case will probably not exceed 2 or 3 oz. per 100
gal. of juice, or if reckoned in terms of the form in which it is most
usually applied, not more than 4 to 6 oz. of potassium metabisulphite. The
latter is most conveniently added as a water solution which is made up so
that each gallon contains the amount of sulphite necessary for 200 gal. of
the juice. For example, if it is desired to add 8 oz. to each 200 gal. a
solution is made containing 8 oz. of the metabisulphite per gallon of water.
After the addition of the potassium metabisulphite, the juice may be
allowed to stand in convenient containers until it has defecated long
enough to permit rapid filtration. The amount of sulphites recommended
are well below the limits allowed by law in various food products. It may
also be stated that a great deal of the sulphurous acid disappears during
subsequent treatment, so that the amount left in the juice is negligible.
Filtration. — After the juice has defecated a sufficient length of time, it
may be filtered without difficulty to give a bright liquid. In this process,
it will be necessary to draw off the clear liquid from the sediment. This
juice can undoubtedly be filtered in a commercial way in any of the good
filters that are on the market. A pulp filter is probably the best. The
sediment from the defecating vessel may be thrown on bag filters and in
this way brightened; it cannot be passed through the pulp filter (because
of clogging) without a preliminary filtering through a bag filter.
Orange juice is much more easily filtered than grape or apple juice and
gives on filtration a very brilliant liquid. This juice also differs from
filtered fresh apple or grape must in that it remains bright after heating;
n8 CITRUS PRODUCTS
fresh grape and apple juices may be filtered bright but will often not remain
bright after heating.
Pasteurisation in barrels. — The juice must be pasteurized after filtration
to prevent fermentation. It may either be bottled before pasteurization or
may be pasteurized and stored in barrels until it is convenient to bottle.
If the juice is stored in barrels the latter must be new or must have
been used only for juice and kept sterile by burning sulphur tapes in them
when empty. Before filling, the barrels should be thoroughly steamed to
sterilize the inner surface. Several forms of pasteurizers may be used in
the heating of the juice to the desired temperature. One of the simplest
types may be made by placing a non-corrosive metal coil in wooden con-
tainer of convenient size. Juice may be placed in this container and steam
passed through the coil. The coil should be kept in motion during the
heating in order to avoid scorching the juice. A tin or aluminum coil will
not be attacked by the orange juice. Double- jacketed aluminum kettles are
used with success in the pasteurization of grape juice and apple cider, and
would no doubt give satisfaction in the sterilization of orange juice. Steam
is passed between the walls of the kettle to heat the juice.
Both of the foregoing heaters are discontinuous in their action. Where
a continuous flow of sterilized juice is desired, a continuous pasteurizer
may be constructed by placing a tin pipe inside of an iron pipe. Steam may
be passed through the outer pipe and juice through the inner tin pipe. By
varying the steam pressure and the flow of juice, the desired temperature
may be attained. Continuous pasteurizers suitable for this purpose are
obtainable.
In any case, the juice should reach a temperature of i85°F. It should
then be run into clean barrels. The barrels should be bunged immediately
with a new bung covered with a clean cloth. The barrel should then be
rolled on its side to sterilize the bung with the hot juice. The flavor of
orange juice seems less easily injured by overheating than that of grape or
apple juice.
During the first few weeks of storage, the barrels must be carefully
watched in order that those which start to ferment may be detected in time
to save them. With careful work few, if any, should ferment.
Pasteurisation in bottles. — Whether the juice is bottled immediately
after filtering or stored first, for a time, in barrels, it must receive a final
pasteurization after bottling. The bottles and corks and caps that are used
must be clean. Capping is preferable to corking, because it gives less
trouble in handling and gives a neater appearance to the bottle. There
are two types of caps, the Crown cap or the ordinary beer bottle or soda
bottle cap, and the Goldy stopper, the aluminum cap seen on grape juice
and pineapple juice bottles. Of these the latter type is preferable. The
bottles must be sterilized immediately after filling and capping. A con-
venient form of pasteurizer may be used by placing a false bottom in a
rectangular wooden tank. Under the false bottom is placed a steam coil.
The bottles are placed on the false bottom, water is admitted so that its
level is about three-fourths the height of the bottles. The pasteurizer must
be covered in order that the caps and tops of the bottles will be heated by
the escaping steam. One bottle may be left uncapped. A thermometer is
JUICES, PRESERVED AND CONDENSED 119
placed in this bottle and the rise in temperature noted. The liquid in the
bottles must reach a temperature of i8o°F.
The bottles may then be removed and allowed to cool. They should
be stored until it can be seen whether the pasteurization has been success-
ful and whether any of the bottles develop cloudiness; three weeks or a
month will probably be sufficient in most cases. In laboratory tests the
juice pasteurized in bottles developed a slight sediment after three months,
but so small in volume as to be scarcely noticeable.
To some, the unclarified juice might be preferable to the clear. If
such a juice is to be produced, no nitration is necessary and the juice can
be placed in bottles immediately, and sterilized. Its appearance would prob-
ably not be pleasing, hence would probably be disguised in dark bottles.
Summary. — It is recommended that the freshly expressed juice be
allowed to defecate until it becomes fairly clear. To prevent fermentation
during this period and to check the development of a bitter flavor, a mod-
erate amount of sulphurous acid should be added to the juice immediately
after crushing. Potassium metabisulphite is a convenient form in which
to add the sulphurous acid. The defecated juice should be filtered. It
may then be bottled immediately and pasteurized, or may be pasteurized in
barrels and kept until it is desired to bottle it. The bottled juice should be
sterilized at 180°— i85°F. to prevent fermentation and mold growth, es-
pecially the latter.
GRAPEFRUIT JUICE
According to the Chace method, the fruit is washed, cut in halves or
smaller pieces, and the juice extracted by means of a hydraulic press. It
is then passed through about 100 in. of sterling silver tubing surrounded
by boiling water, the time of passage being approximately five seconds. It
goes immediately from this pasteurizer into a hot 5~gal. carboy which is
filled as far as possible into the neck, leaving, however, a slight air space
jusr beneath the cork. This is put in cool storage, the temperature not to
exceed 50° F., and there kept until the sedimentation has finished, leaving
a clear liquid in the larger part of the carboy. The clear liquid is decanted
and treated with about I Ib. of Kieselguhr per 100 gal., the Kieselguhr
having been previously washed several times with boiling water until the
extract is tasteless. The residues, after drawing off the supernatant liquid,
are treated with about 5 Ib. of Kieselguhr per 100 gal. of material, and
filtered. Chace uses the ordinary rack-and-plate-filter press for this filtra-
tion, although he is not sure but other forms would probably be more rapid
and give just as clear a product.
After this has been done, the juice is sweetened until the solid contents
are about from 17 per cent to 20 per cent. This usually requires about 5
per cent of sugar, with grapefruit. The product is then again passed
through the flash pasteurizer, using the same method as before. If the
bottling is done hot care should be taken lest the paraffin melt off the Crown
caps and deposit in the beverage. The grapefruit juice thus prepared re-
mains clear for months. Recently a Jensen milk pasteurizer was substi-
tuted in the first process. This is considerably more rapid and permits of
rapidly filling the carboys with the juice.
I2O CITRUS PRODUCTS
MIXED JUICES
Chace bottled several combinations of grapefruit with other fruit juices,
and considered two of these blends as very good. Both pomegranate and
loganberry juices make fine blends. He uses about 20 per cent of the
blending juice to 80 per cent of grapefruit juice, and sometimes sweetens
until the solids are raised to 20 per cent. This gives a product of brilliant
color and fine flavor. He used infusorial earth as a filtering medium and
a flash sterilization above 85°C. Juices have kept for six months without
deposition of sediment or deterioration of flavor.
Other blends, with California grape juice, have been tried out by Chace
but lack characteristic flavor. The muscats probably do better than the
others, but all blends were inferior to the first two mentioned. Concord
grape juice has not been tried, but there seems no reason why it would
not blend well with grapefruit juice.
CONDENSED FRUIT JUICE1
Concentrated juices, if they could be satisfactorily produced, would
have obvious advantages over unconcentrated juice. In the production
of concentrated juices two dehydrating methods are usually employed,
namely, by vacuum distillation or freezing. Neither of these result in
a product which when diluted to its original volume with water has a
satisfactory taste. The chief defect, apart from the loss of flavor, is
in a darkening of the juice presumably from a carbonization of the
sugar by the concentrated acid. The latter may be overcome by neutral-
izing the acid.
One firm in concentrating juice by vacuum distillation maintains a
vacuum of 28 inches and a temperature of io8°-no°F. The final
product contains 61-65 Per cent citric acid.
The methods of concentrating the various juices will be discussed
under the various localities involved, e.g., Sicily, West Indies, and
Florida. Dehydration by freezing has not been used widely. As it
seems to have advantages over any other process used so far, a de-
tailed description will be given here with the hope that its use will be
stimulated.
'United States patents involving the concentration of fruit juices by means of
cold are: A. Giirber, No. 723152, March 17, 1903; E. J. Sheehan and W. S.
McKay, No. 1211361, January 2, 1917 (addition of sugar) ; M. O. Johnson, No.
1362868, December 21, io2o (addition of sugar). United States patents involving
the concentration of fruit juices by means of heat are : C. Graef , No. 640289,
January 2, 1900; R. Oehme, No. 950950, March I, 1910 (with solvents); A.
Fernbach, No. 981405, January 10, 1911 (in COa) ; P. Kestner, No. 1005554,
October 10, 1911; J. N. G. Singleton, No. 1093081, April 14, 1914; H. C. Gore,
No. 1141458, June i, 1915; J. L. Kellogg, No. 1189127, June 27, 1916; A. C.
Braden, No. 1278297, September 10, 1918 (addition of sugar); C. A. Kern, No.
14633, April 15, 1919. United States patents involving the concentration of fruit
juices by desiccation are: L. C. Merrell, I. S. Merrell, and \V. B. Gere, No.
878977, February 19, 1908; W. S. Osburn, No. 1088267, June 4, 1912; C. Ellis.
No. 1068047, July 22, 1913; C. E. Gray, No. 1057935, October 26, 1915.
121
Concentration of fruit juice by dehydration by means of cold: the
Giirber method. — Perhaps the first patent for concentrating solutions in this
manner was applied for by August Giirber in Germany, July 15, 1899. Later
(1903) he received a United States patent for the process. A liquid to be
concentrated is placed in a suitable centrifugal machine, preferably having a
central coil or chamber containing a refrigerating medium, such as ammonia,
. brine, or the like. The centrifugal machine is then started, and as the
freezing action proceeds, the watery constituent of the liquid is congealed
into ice crystals, which crystals are, by reason of their less specific gravity,
displaced by the heavier solids and caused to accumulate about, and adhere
to, the central coil or refrigerating chamber, the more concentrated solu-
tion, deprived of a large portion of its water, finding its way to the outer
periphery of the drum, whence it may be drawn off in any well-known
manner. The product may be withdrawn at an intermediate stage of con-
centration and subjected to further similar treatment or the exit valves
may be so adjusted as to accomplish the desired high degree of concentra-
tion in one and the same continuous treatment by subjecting the content's
of the drum to the action of refrigeration and centrifugal separation during
a sufficiently extended space of time.
Concentration of fruit juice by dehydration by means of cold: the
Monti process. — If we consider that the evaporation of one kilo of water
at 15° C. requires theoretically 600 positive calories, while the separation
of one kilo of ice from a solution cooled to o°C. requires only 40 negative
calories, it is plain in view of the progress in methods of refrigeration that
there would be a theoretical gain on an industrial scale in substituting for
direct evaporation by heat the freezing and separation of the water in the
state of ice as a means of concentrating large volumes of dilute solutions.
In this field, Italy has contributed, largely through the numerous inves-
tigations of Endo Monti, to the progress of the concentration of liquids
by means of cold.
During the winter of 1902, Monti was seeking a practical and econom-
ical method to concentrate the weak wines of 1901. He found that on
freezing the wine, and then inverting the vessel, keeping the bottom part
insulated, and allowing the ice to melt slowly in the upper part, he ob-
tained at first a concentrated wine, then wine progressively more and more
dilute until finally a cake of ice crystals was left containing no soluble
matters, but only tartar, lees, and other insoluble matters.
Monti noticed also that the first liquid to drip from the ice was much
more concentrated than the liquid which remained uncongealed in the center
of the vessel. In later experiments on a larger scale made at the estab-
lishment of the Italian Society for Artificial Ice, he demonstrated that, by
passing the dilute liquids obtained in a previous operation and cooled to
o°C. through the mass of ice crystals, he could free the ice of all soluble
matters without melting the crystals. Studying, later, this fact established
experimentally, he demonstrated by a very simple experiment that when a
solution is frozen rapidly, the dissolved substances are concentrated not in
the uncongealed portion of the liquid, as formerly supposed, but in the
water interposed between the ice crystals, and only later and slowly do
they diffuse into the uncongealed liquid. He took three metallic cylinders
122 CITRUS PRODI/GTS
about 10 cm. in diameter and 20 cm. in height closed at the bottom. These
he filled with equimolecular solutions (one-tenth moleculegram to 100 gm.
of water) of alcohol, saccharose, and citric acid and immersed them in a
freezing mixture of about i8°C.
At the end of an hour a little more than half of the liquid was frozen,
forming a ring adhering to the walls of the vessel. On removing 50 c.c.
of the liquid and determining its density with a Mohr's balance he found-
that there had been no sensible change and therefore no concentration.
After pouring out the uncongealed liquid, he placed the frozen contents of
each cylinder on a funnel and drained off liquids which showed a concen-
tration double that of the original solution. In another experiment, in-
stead of pouring out the uncongealed liquid and draining the ice on a
funnel, he surrounded the vessel with cotton and left it for several hours
and found that the density of the uncongealed liquid gradually increased
and its temperature fell as the cold, concentrated solution between the ice
crystals gradually diffused into the liquid in the center.
In other experiments he froze various solutions (wine, beer, coffee,
meat extract, etc.) in insulated cylinders of convenient height, furnished
with an opening at the bottom. By allowing the dilute, cooled solutions
obtained by allowing the ice in the upper part of one cylinder to melt and
to filter through the ice of another cylinder containing the same solution he
was able to displace completely the concentrated solution between the crys-
tals without melting more than a very small part of these crystals. At
each operation he obtained pure ice at the top of the cylinder while ex-
tracting a concentrated solution from the bottom. The solutions of an
ever increasing degree of dilution which served for the extraction were
reproduced without increase of volume for an indefinite number of suc-
cessive operations.
Monti showed further that if the cooling was pushed too far the
permeability of the mass diminished, the dissolved substances precipitated
or crystallized in the interpolated water and the extraction become more
and more difficult and finally impossible. The limits of cooling varied with
the concentration and with the viscosity of the solutions, and were in the
case of fermented liquids about three times the temperature expressed in
negative degrees centigrade, at which the solution commenced to freeze.
With grape juice and concentrated solutions the limit was about twice the
freezing temperature and in all cases the point of congelation of the
saturated solution.
For this reason, it is advisable to limit the congelation to the separa-
tion of 60 to 65 per cent of the ice in dilute solutions and 40 to 50 per cent
in more concentrated solutions. When a greater concentration is desired,
it is best to obtain it by two or more successive operations. In the par-
ticular case of sugar solutions he found that when the syrup contains over
50 per cent of sugar, the extraction becomes very slow and difficult, owing
to the viscosity of the liquid.
Practically eighty-eight negative calories are needed for every kilo of
ice to extract 60 per cent of the ice from each 10 per cent sugar solution
at 2°C. by freezing at 3°C. A 25 per cent solution frozen at — 6°C.
requires ninety-five negative calories to extract 45 per cent of the ice.
JUICES, PRESERVED AND CONDENSED 123
To this amount of calories must be added about 10 per cent to allow for
the heat of the vessels and tubes and for the ice which melts, owing to
the imperfect insulation of the apparatus and the imperfect exchange of
temperature in the coils.
Monti obtained various patents for the application of his method to the
concentration of diverse liquids and grape must.
By a method patented April 17, 1906, he prepared concentrated must
from freshly crushed grapes from which the juice was extracted and cooled
rapidly before the flavor changed. He prepared four aromatic musts from
various varieties of grapes, Dolcetto, Banarda, Freisa, Nebbiolo, Mosca-
tello, and preserved them at ordinary temperatures after pasteurizing at
50° C. in bottles sealed with paraffin. These concentrated musts which are
fluid and perfectly clear taste very delicate, recalling in the highest degree
the true, natural aroma of the grape.
The characteristic of the Monti patents which distinguishes them from
others dealing with concentration by the application of artificial cold, is
the extraction of the soluble substances interposed between the ice crystals
by means of osmosis and progressive extraction with solutions of increasing
degree of dilution obtained by the fractional liquefaction of a previous
operation. Naturally these solutions must be cooled to the point of freezing
with the ice separated in the preceding operation. This is accomplished
automatically without mechanical work and therefore economically. By a
patent of 1908 Monti has still further simplified the construction of his
apparatus by suppressing coils and cocks, so that the whole can be made
by ordinary country blacksmiths and mechanics and adapted to any form
of freezing machine.
With regard to cost, Monti has demonstrated that his process is more
economical than those of evaporation by heat or in vacuum. This supposes
that the motive power does not cost more than $40 per horse-power per
year and that efficient freezing machines are used which with a suitable
supply of refrigerating water are capable of absorbing at least about 2,000
calories per horse-power per hour. In this regard Monti makes the following
statements :
"It was shown that with a concentration plant of moderate size, not
more than 120 frigories were used for every kilo of water extracted, rep-
resenting 36 kilos of fuel for each hectoliter while the direct evaporation of
one kilo of water required 720 calories representing 1.25 kilos of steam
and 150 kilos of fuel for each hectoliter. This indicates that evaporation
requires about four and one-half times as much fuel as congelation for the
separation of the same amount of water. In each case, proper allowance
is made for all losses based on the average results of a continuous operation
on an industrial scale.
"Moreover, the cost of present installation of apparatus for concentra-
tion and the buildings necessary for refrigeration, conservation, and clarifica-
tion of the musts and wines is much exaggerated, as it is based on imperfect
construction and unfavorable working conditions.
"In fact, the installation of an ice plant of the capacity of 25 hectoliters
per day of four -extraction columns made of reinforced concrete and prop-
erly insulated of 250 hectoliters each, of which three are used to concen-
124 CITRUS PRODUCTS
trate 250 hectoliters of must or wine per day and the fourth as a reservoir
for the frozen water, and the recuperators necessary for the refrigeration
of the wines and the extraction liquids with the cold water of the separated
ice costs less than $2,000. The tinning and varnishing of the 1,000 ice
forms needed cost less than $600. Assuming, therefore, that the factory
works 1 80 days producing ice and concentrating the must or wine, 27,000
hectoliters would be handled in six months which reduces the cost of the
plant to 20 cents for each hectoliter of must or wine concentrated. This
cost corresponds to 2 cents per hectoliter if we calculate on amortization
in ten years.
"The labor cost is also much reduced by the suppression of centrifugal
extraction and the movement and filling of forms. The extraction of the
ice is accomplished automatically by the crane used in all ice factories of
any size.
"It is true that the tinning and varnishing of the forms must be re-
newed at rather frequent intervals and a part of the amortization and
interest on the cost of the plant must be charged against the concentration.
"Where an ice factory is not available the cost of the plant is naturally
greater. However, wooden concentrators with straight aluminum or tinned
copper tubes are of such simple construction that they can be made any-
where by skilful mechanics. The cost of a concentrator of the capacity of
50 hectoliters should not exceed $800 in any case. A battery of four, of
which one would serve as a reservoir for the frozen water, would cost
$3,200. Reckoning $2,400 as the cost of a freezing machine capable of
absorbing 20,000 calories per hour, it follows that the cost of a plant to
concentrate 50 hectoliters a day or 15,000 hectos per annum, varies, exclud-
ing the cost of the buildings, from $6,000 to $8,000 or 50 cents per hectoliter,
according to local conditions. This with a ten-year amortization represents
an expense of 5 cents per hectoliter of wine concentrated. The cost of
cooling and conserving wines and musts in cold storage is much higher.
"In buildings of a capacity of from 20,000 to 40,000 hectoliters, such as
might be established in Italy for the treatment and storage of wines and
musts in refrigerated containers of reinforced concrete, an area of I sq. m.
for each 10 hectoliters may be reckoned, representing an expense of 20
cents for each hectoliter of must treated. Provision also must be made for
the refrigeration of the rooms and containers.
"This refrigeration, assuming a maximum difference of 30 °C. between
the inside and outside temperatures, would necessitate the absorption of
15 calories per hour per cubic meter, that is, 1.5 calories per hectoliter or
per 30,000 hectoliters, 45,000 frigories per hour.
"It will be necessary, therefore, in this case to provide for the absorp-
tion of 100,000 calories per hour which would require a force of about 50
horse-power.
"It should be noted, however, that the unit cost of plant and running
expenses of the establishment and freezing machinery increases rapidly as
the output of the plant diminishes. And also, with equal productive capacity
the unit cost may be doubled or quadrupled, if the warehouses and reser-
voirs are not planned to completely utilize the space, or if the ice machines
and the insulation of the walls is not as perfect as possible. Under normal
JUICES, PRESERVED AND CONDENSED 125
conditions, therefore, the conservation in cold storage costs twice as much
as their concentration.
Concentration of fruit juice by dehydration by means of cold: the Gore
process. — This process of concentration of cider by freezing has become
quite popular. For this reason it is given here as a possible means of con-
centrating citrus fruit juice.
An ice-making plant was equipped with special 3OO-pound tin-lined
freezing cans, an ice crusher, a centrifugal machine, necessary machinery
for handling the frozen cider, and containers for the finished product
(Fig- 13).
Scftfrf XT CavceffTAAr/oj? of
A me C/£>f* *r S*
(Gore: Yearbook, U. S. Dept. of Agric.)
FIG. 13. — Dumper, ice-breaker, and centrifugal machine used in
crushing and centrifugalizing frozen apple juice.
The fresh cider is placed in the tin-lined ice cans and frozen in the
brine tanks of the ice-making plant. For the first freezing the brine tem-
perature should range between 10° and 20° F., which will freeze the cider
into a solid mass in from thirty-six to forty-eight hours. The frozen cider
is then loosened from the cans by removing it from the brine tanks, thawing
it at the sides and bottom, and dumping it, just as ice is removed from the
126 CITRUS PRODUCTS
cans in ice-making. The blocks of frozen cider are then passed through a
power ice crusher which breaks them into pieces no larger than a walnut.
The crushed frozen cider drops from the crusher into a standard sugar
centrifugal machine which provides a mechanism for rapidly whirling it,
and thus separating the syrupy part from the water ice, just as in the case
of sugar, where the molasses is whirled off from the crystallized sugar.
The frozen cider drops into the perforated metal basket of the machine.
This basket, which is whirled rapidly, causes the syrupy part of the cider
to fly off from the mass of ice into the collecting chamber opening into
the collecting tank below. A centrifugal operating at about the same speed
used on sugar will separate most of the syrupy content from a charge of
ice in three minutes.
The partially concentrated cider in the receiving tank is then put again
into the freezing cans and refrozen at a temperature from 0° to 10° F.,
which will refreeze the syrupy cider in from two to three days. The
refrozen cider does not become very solid and can readily be removed from
the cans without thawing. It is passed through the crusher and again
whirled in the centrifugal machine and reaches the receiving chamber as
a fairly thick, somewhat viscous syrup. One gal. of this syrup represents
5 gal. of the original cider, which means that nearly 4 gal. of water have
been removed by freezing and centrifugalizing.
As the ice remaining in the centrifugal basket still contains some of
the sugar and solids of the cider, it may be removed from the basket by
use of the unloader provided, allowed to warm up slightly, and again be
run through the centrifugal machine. This will remove practically all the
valuable material. As there will now be less than I per cent of apple
solids left in the ice, it probably does not merit further treatment. The
concentrate obtained on the second centrifugalizing of the ice is somewhat
richer in solids than ordinary cider and may be added to the fresh cider
or refrozen and treated as fresh cider.
The fully concentrated cider will be a somewhat thick liquid with the
cloudy appearance and color of unfiltered fresh cider. If kept sealed at
household refrigerator or in cold-room temperatures, it will gradually fer-
ment, but will spoil much more slowly than ordinary cider. Concentrated
cider intended for use during the following spring or summer should be
put into cold storage at once and kept at or below 32° F. It can be shipped
anywhere in winter weather in unheated cars without danger of spoiling.
It can be bottled, canned, or shipped in clean kegs. Inasmuch as a i-gal.
can represents 5 gal. of cider, the equivalent of 250 gal. of fresh cider can
be shipped profitably for much longer distances to market than can the
bulky fresh cider. In addition to saving 80 per cent on the freight, the
reduction in volume permits the shipper to use containers one-fifth the
size, which is an important saving in cooperage. The saving in con-
tainers, which have always been relatively expensive items in shipping
cider, should alone nearly offset the cost of freezing and concentration.
In addition there is the saving of drayage and other handling costs. On
its arrival at the market the retailer can at once restore it to cider by
adding four parts of water, or he can sell it in the convenient concen-
JUICES, PRESERVED AND CONDENSED 127
trated form to his customers. Moreover, as has been explained, this cider,
instead of fermenting on the journey, or a day or two after its arrival,
will keep for two or three weeks in closed containers if not allowed to get
warm, and if kept closed in an ice box or refrigerator will remain in good
condition for five or six weeks. The cider in its concentrated form is useful
also as a flavoring syrup for desserts. When stored at low temperatures,
the concentrated cider will keep indefinitely. In this way concentrated
cider made in the fall can be kept over and used in the soft-drink trade
during the hot months, at soda fountains, and in the home.
Those who wish to make a clear, brilliant cider concentrate can filter
it by adding to the concentrated product 3 per cent or more by weight of
infusorial earth, which should be thoroughly stirred in. This mixture can
then be pumped through a plate-and-frame filter press, such as has been
described in connection with the manufacture of apple syrup.
a) Cost of concentrating by freezing. — Until the process is well
worked out on the commercial scale, statements of the probable cost of
preparation are necessarily estimates only. The principal elements of cost
are cost of raw material, cost of freezing, labor, power, interest, deprecia-
tion, and superintendence.
b) Cost of raw material. — Apples should yield at least 150 gal. of
cider per ton, with apples at $6 per ton, allowing 2 cents per gallon as
the cost of pressing, the raw material charge is 6 cents per gallon, or 30
cents per gallon of finished cider concentrated by freezing, on the assump-
tion that the reduction in volume is 5 to i.
c) Cost of freezing. — If the brine tanks of a modern ice plant can be
used, the cost of freezing for the first time, including dumping and delivery
to the ice crusher, should not exceed $3 per ton of cider measuring about
230 gal., provided ice-making is carried on during the balance of the year,
so that the yearly fixed charges of the ice, factory are borne by cider and
ice in proportion to the amounts frozen. In the second operation about one-
third of the original volume of cider is refrozen, bringing the total cost of
freezing up to $4 per ton, approximately, or 8.7 per gallon of concentrated
cider. Where an ice plant is to be erected for the purpose of freezing cider
alone, using it at full capacity for but a few months each year, the cost of
freezing will necessarily be much higher.
d) Labor cost. — Two men, with the assistance of the engineer at the
ice plant, whose labor is included in the cost of freezing, can easily oper-
ate the centrifugal machine at the rate of 300 Ib. of frozen cider per ten
minutes, amounting to a little more than 7 tons per eight-hour day. About
one-third by volume of the original cider is returned for refreezing. On
the following day the ice obtained on the day previous is recentrifugalized,
requiring about a half-day's work, and two days later the centrifugalizing
in the second operation, requiring about a half-day's work, is accomplished,
producing cider concentrated by freezing in finished form, except for filter-
ing, which may or may not be done, at the option of the manufacturer.
Thus, approximately two days' work of two men is required in working up
7 tons, about 1,600 gal., of fresh cider, or 320 gal. of cider concentrated by
freezing. At $2.50 per day the labor charge is thus $10, or 3.12 cents per
gallon. The power required for crusher and centrifugal machine and for
128 CITRUS PRODUCTS
small hoist for elevating the ice for recentrifugalizing totals less than 100
horse-power. The crusher requires 2 horse-power, centrifugal 5 horse-
power, and hoist 2 horse-power. The centrifugal costs $560 complete and
the ice crusher $84. These are the prices paid by the department on com-
petitive bids for a 3O-in. centrifugal and a standard crusher for 3OO-lb.
blocks of ice. Erection of the equipment and necessary pumps and vats
for economically carrying out the process should bring the equipment cost
to between $1,000 and $1,500, not including the cost of the building. A
plate-and-frame filter press and feed pump costs about $300. It is worth
while to call attention to the fact that centrifugal and ice crusher, as well
as hoist, filter press, and feed pump are made to stand heavy, constant
service.
We thus have the following estimate of cost per gallon for the prepa-
ration of cider concentrated by freezing:
Cents
Raw material 30
Freezing 8.7
Labor 3.12
41.82
Filtering, power, interest, depreciation, and
other charges 8.18
Total 50.00
The cost items classed under filtering, power, interest, etc., cannot be
determined at the present time with anything like accuracy. They will
necessarily vary with the volume of cider concentrated. On the whole,
however, a cost figure of 50 cents per gallon is a conservative estimate for
the preparation of cider concentrated by freezing, on the assumption that
the brine tank of a going ice plant is available for the freezing.
Concentration of fruit juice by dehydration by means of heat: the
McClendon-Dick process. — A unit or dehydrating cell of J. P. McClendon
and E. M. Dick consists of an octagonal chamber 20 ft. in diameter and
20 ft. high with twelve hot-air ports in three tiers. The upper tier admits
air at a velocity of 300 ft. per minte, the middle at 600 ft., and the lower
150 ft. The centrifugal spray is in the center and just below the level of
the upper tier of ports. It consists of a hollow shaft (admitting the juice)
and a head made of a series of concave disks serrated at their margins
(12 in. in diameter). Between the disks is a series of grooved collars
distributing the juice from the hollow shaft to the concave sides of the
disks. The shaft and head rotate on ball bearings 5,000 revolutions per
minute and transform the juice into a fine spray or fog. The air currents
cause eddies and retard the falling of the spray until it is dried. The
exhaust ports are at the bottom of the cell and are so large as not to
retard the falling of the dried juice in the dead air space. The incoming
air is heated to 55°— 70°, but the evaporation on the surface of the droplets
prevents their reaching this temperature before they fall into the dead air
space where the temperature is lower. Since the droplet is exposed but
one minute to the hot air, any temperature effect on the vitamine is mini-
FERMENTATION PRODUCTS 129
mized. Orange juice is one-eighth solids. If it is evaporated to one-fifth
of the original weight the pectin forms a clot from which a syrup oozes.
This separation is not hastened by the centrifuge but is by a filter press.
If the juice is evaporated to one-sixth or less, no clot separates. The high
content in monosaccharides makes it hygroscopic when dried. The taste
is practically unaffected and the dried juice, reconstituted with water,
makes a pleasant drink. Cane sugar may be added before drying.
Milk is dried in the same type of cell and the orange preparation may
be added to milk powder to restore the antiscorbutic lost in pasteurization.
(The milk is first pasteurized since wild yeasts and some bacteria are not
killed by drying.) The orange juice imparts a lactic acid flavor to the
reconstituted milk in proportion to the amount added, but the milk does
not curdle unless enough citric acid (juice) is added to bring it to the
isoelectric point of casein.
FERMENTATION PRODUCTS
The citrus fermentation products1 so far known commercially and
experimentally are based primarily upon alcoholic and citric fermenta-
tions. Only citrus fruits of appreciable sugar content, can therefore,
be used, and fermentation products are usually derived from oranges,
though yielded also by grapefruit. The products in question are four :
alcohol, wine, vinegar, and acetone. Alcohol as such is seldom made
from citrus fruits. The preparation of "wine" is the first step in the
manufacture of vinegar and of acetone from citrus fruits.
FERMENTED JUICE FOR VINEGAR-MAKING
Orange wine may be defined shortly as the product of the alcoholic
fermentation of oranges and the usual wine cellar treatment.
The samples of orange juice examined at the Zymological Labora-
tory of the University of California have averaged by chemical test
about 1 1 per cent actual total sugars. On fermentation this would give
about 5.5 per cent alcohol if the fermentation were carefully conducted.
Control of the micro-organisms. — Given oranges of suitable com-
position, the quality of the wine depends on the work of micro-organ-
isms. The art of the wine-maker consists almost entirely in the control
of these micro-organisms. His success in facilitating the work of the
useful form (true wine yeast) and in preventing or hindering the
work of injurious forms determines the quality of his product.
a) Before fermentation. — On the skins of sound ripe oranges as
they hang in the orchard the micro-organisms are comparatively few
'This chapter in so far as it touches upon matters not directly applicable to
the preparation of wine for subsequent vinegar or acetone making is to be re-
garded in the United States and in some of its territories as of purely theoretical
and historical interest.
The production of citric acid by fermentation of sugar is discussed on p. 151.
130
CITRUS PRODUCTS
TABLE XVIII
VARIATION IN COMPOSITION OF ORANGE JUICE
and in an inactive condition. When ordinary care is exercised there is
little danger that they will injure the wine. On broken or injured
oranges the number is greater and the forms more active. If many
such oranges occur they should not be mixed with the sound fruit if
the best wine is to be made, even though it be designed for subsequent
vinegar making.
Care should be taken to avoid unnecessary bruising of fruit which
cannot be worked immediately. Cleanliness is essential throughout the
process. Oranges, gathered in moldy boxes, hauled in dirty wagons
or cars, and passed through dirty crushers, conveyors, and presses, may
be so completely infected with injurious germs that it is impossible to
obtain a good fermentation. Dust or soil is less injurious and, if
excessive, may often be removed by 'sprinkling. Washing with anti-
septics is not permissible. A weak solution of potassium metabisulphite
might be used with benefit if it were not for the difficulty of regulating
the amount of sulphurous acid entering the fermentation vat.
If the oranges have to be kept for some time before crushing they
should be kept as cool as possible to delay the growth of molds. Gather-
ing in the cool of the morning is desirable, and if oranges are gathered
when warm they should be left in boxes to cool off during the night
whenever possible. If the fruit is cool when it reaches the fermenting
vat its temperature will neutralize a certain proportion of the heat of
fermentation, and accordingly the difficulties connected with an injuri-
ously high temperature are diminished.
However carefully the oranges are handled, a certain amount of
dust, containing germs and other injurious matter, will reach the vats
and presses. It is desirable to get rid of this matter before fermenta-
tion. This is best accomplished by settling and decantation.
FERMENTATION PRODUCTS 131
When whole, unpeeled fruit is used, the presence of the oil from
the peel greatly depresses fermentation. This oil can be eliminated by
boiling before inoculating, however, at the risk of impairing the natural
flavor.
Thorough crushing is necessary as the must should be well satu-
rated with air. As the juice runs from the press it is pumped into a
settling tank or cask. If it is cold, below 15 °C. and of full normal acid-
ity, the impurities may settle in twenty to forty-eight hours. If the
temperature is higher than I5°C. and the acidity low, molds and yeasts
will develop or fermentation will start and interfere with the settling.
To prevent such interference a slight sulphuring with the fumes of
burning sulphur or with a solution of potassium metabisulphite is usu-
ally necessary at this stage. The sulphuring should be as light as pos-
sible with acid musts, as it tends to preserve the fixed acids. For the
same reason it benefits musts of low acidity. In from twelve to twenty-
four hours the must should become purged of all its gross impurities
including micro-organisms, and solid particles derived from the skins,
stems and pulp of the oranges.
This separation of undesirable substances, the so-called "defecation,"
is of great value, ridding the must of extraneous matter that would
affect the flavor of the wine in the heat of fermentation and eliminat-
ing the excess of protein that would serve as food for injurious bac-
teria. Centrifugal machines have been devised to hasten the process,
but their work is not perfect.
Sterilization by heat has been tried for the same purpose but with
indifferent success. A high temperature causes caramelizing of a part
of the sugar and oxidizes the must, thus injuring the flavor. Discon-
tinuous heating at lower temperatures in an atmosphere of carbon
dioxide is efficient, but troublesome and expensive ; all methods involv-
ing heat have the defect of extracting undesirable substances from the
solid matters which are heated with the must.
Chemical sterilization is still less practicable. No substance could
be used for this purpose except sulphur dioxide, but this can be em-
ployed only in minute quantities without seriously injuring the flavor
of the wine.
All the methods discussed have for their object the diminution or
elimination of micro-organisms of all kinds, and to the extent of their
removal the true yeast is also lost. Hence the more efficient the method
used, the more necessary it is to add wine yeast. Without this addition,
in fact, all the precautions described may result in harm, for the wine
yeast, being present in much smaller quantity than many of the injuri-
132
CITRUS PRODUCTS
ous forms, may be completely eliminated, while enough of other forms
of micro-organisms are left to spoil the wine after this is drawn off
into clean casks for fermentation.
In order that the proper kind of fermentation may be initiated a
"starter" of some kind is, therefore, added to the defecated must. An
improvement on a natural starter is a pure culture of tested yeast. The
methods of handling this would require too much space to describe here,
but they are simple and such as could easily be devised by anyone with
some knowledge of microbiological technique. They do not aim at
obtaining an absolutely pure fermentation, which is unnecessary, but
endeavor to have an overwhelming proportion of a thoroughly tested
and suitable yeast which will attenuate the wine rapidly and perfectly
before the injurious micro-organisms present have time to do any harm.
This starter is used only for the first vat or cask. Those following are
started from the first fermentations, care being taken always to use
the must from a tank only at the proper stage of fermentation and to
avoid all tanks that show any defect.
If grown in well-defecated and sulphited juice, the yeast will re-
main sufficiently pure, throughout the season1
The following figures were obtained in laboratory fermentations of
orange juice.
Sample No. I was fermented with pure yeast; No. 2 was allowed to
ferment naturally. Although the Balling was 1.6 per cent higher in the
latter case, the yield of alcohol was only .3 per cent higher, indicating a
greater efficiency in the pure yeast fermentation. It may be stated that
the natural fermentation of No. 2 was carried on largely by undesirable
types of yeast. This also happened in most cases where the juice was
TABLE XIX
YIELDS OF ALCOHOL FROM ORANGE JUICE
allowed to ferment spontaneously. In Sample No. 2, a heavy growth of
film-forming yeast developed, giving a disagreeable flavor as well as
causing the liquid to clear very slowly after fermentation. On the other
hand, the juice fermented with the pure yeast, had a clean flavor, and was
easily cleared.
"Formerly, directions always accompanied the wine-yeast distributed by the
Enology Laboratory of the University of California.
FERMENTATION PRODUCTS 133
6) During fermentation. — However carefully the injurious germs
have been excluded and the proper yeast increased, fermentation will
not be successful unless conditions are maintained as favorable as pos-
sible to the wine yeast and unfavorable to other micro-organisms.
The temperature of the crushed oranges or of the expressed must
is of importance. If it is below 15°, unless the weather is warm, the
oranges should be warmed to 20° or 25°. If this is not done, the
molds and the undesirable Saccharomyces apiculatus, which require less
heat than the wine yeast, 5\ ellipsoideus, will develop more quickly.
This is of especial importance when! starters are not used. Once the
impurities are removed by the preliminary defecation the must is
ordinarily not too warm for the commencement of fermentation. How-
ever, the warmer the must, the more artificial cooling will be necessary
later, and the sooner this will have to be applied.
At the beginning of fermentation the must should be thoroughly
saturated with air to insure the multiplication of the yeast. The
aeration received in the first process of crushing and pressing the fruit
is usually sufficient for this purpose. One objection to the sterilization
of must by heat is the expulsion of the air and the difficulty of replacing
it in the proper amount. Too much aeration on the other hand is
harmful by injuring the flavor and color of the wine through over-
oxidation and by promoting the growth of injurious aerobic organisms.
The proper employment of sulphurous acid in the regulation of fer-
mentation is one of the most important but least understood parts of
the wine-maker's art. Only by its proper use can wholesome wine of
the highest quality be produced, while improper use of sulphurous acid
will injure or completely spoil the wine. Its beneficial effects are due
not only to its action on micro-organisms, but also to its influence on
enzymes and on the color of the wine.
In the small quantities properly used in wine-making, it is antiseptic
in a degree varying with the amount. All micro-organisms are sus-
ceptible to its action in different degrees. Bacteria are particularly
sensitive, molds and pseudo-yeasts less so, while wine yeast is the most
resistant of the usual forms found in must and wine.
Its effect on the color of wines is also of importance. By the action
of oxygen, the color of wine is gradually darkened; this is prevented
or much diminished by the use of minute quantities of sulphurous acid.
The most commonly used source of sulphurous acid is fumes of
burning sulphur. Sulphur is burned in a cask and the must caused to
take up the fumes by being pumped into the cask through the upper
bung hole.
134 CITRUS PRODUCTS
The method is defective in many ways. It is impossible to tell
within very wide limits how much sulphur dioxide has been absorbed
by the wine. Moreover, the sulphur burns incompletely and the vola-
tilized sulphur acted upon by the yeast may produce sulphuretted hydro-
gen. Other sulphur compounds are also produced during the burning
to some of which the so-called sulphur taste of wine is said to be due.
Several devices have been invented to decrease these defects but none
removes them completely and progressive wine-makers are adopting
more reliable sources of sulphur dioxide.
An improvement is the use of potassium metabisulphite (K^Os),
a salt which can be obtained in the requisite purity in commerce, and
contains 50 per cent by weight of sulphur dioxide. The amount of
potash added by this salt in the quantities required is very small, and
well within the limits of variation of the normal potash content of dif-
ferent wines. By the use of this salt, exact amounts of sulphur dioxide
can be applied. Other sulphites are not permissible.
The best source of the acid, however, is the liquefied gas recently
brought into limited use and manufactured comparatively cheaply in
great purity. By the employment of this all the action of sulphurous
acid is obtained and all uncertainty eliminated.
Some oranges, owing to their composition, especially their high
acidity, are very resistant to the attacks of injurious bacteria. Others,
owing to low acidity or highly nitrogenous nature, are more susceptible.
The quality and character of the wine depends greatly on the tem-
perature of fermentation. If too low, the fermentation may be unduly
prolonged, the wine yeast may have difficulty in overcoming its com-
petitors and the wine may remain inferior and cloudy. On the other
hand, if the temperature is too high the results are worse. The growth
of bacteria is promoted, injuring the wine by the production of volatile
organic acids, by displeasing flavors produced and by preventing the
proper action of the yeast. Such wines may remain sweet on account
of the failure of the yeast to do its work and become unpleasantly acid
owing to the volatile acids produced by the bacteria.
Some means of controlling the temperature is, therefore, always
needed. Where heat is deficient it may be supplied by direct heating
of the must or part of it, or by heating the cellar. Small fermenting
vats promote radiation, diminishing the heat, and cooling machines may
be applied directly to the fermenting wine.
The main part of the fermentation should be completed in from
three to five days. If it is desirable to retain the slight sweetness due
to a small amount of unfermented sugar, this may be accomplished by
FERMENTATION PRODUCTS 135
the judicious use of sulphurous acid, prompt clarification by filtration
or by "fining" and, when necessary, by pasteurization. The pasteuriza-
tion tends to remove those proteins which are coagulated by heat and
which are the preferred food of bacteria.
In the case of dry wines, protection from bacteria is best obtained
by prompt and complete attenuation. Fermentation should not be al-
lowed to cease until all the sugar has disappeared. For this purpose,
one or two aerations by pumping over may be required immediately
after the end of the tumultuous fermentation. The temperature of the
wine should not be allowed to fall sufficiently to check the action of
the yeast until all the sugar has disappeared.
c) After fermentation. — As soon as all the sugar has been de-
stroyed, in the case of dry wines, or when the desired degree of attenu-
ation has been obtained in the case of sweet wines, all the useful work
of micro-organisms has been accomplished. The quality and safety of
the wine then depend on freeing it from the organisms present and
preventing the entrance and action of all others.
As soon as bubbles of carbon dioxide cease to be given off, the
yeast and other solid matters will settle to the bottom and the liquid
become clear. This often occurs before the fermentation is complete.
In this case the yeast should be stimulated by aeration as described
above.
If the wine is dry it should be racked (drawn off, decanted) from
the sediment into clean casks. The first racking is usually done to re-
move the more bulky sediment while the wine is still slightly cloudy
during the first month or six weeks. If left too long on the yeast the
autophagy or degeneration of the latter may produce substances which
injure the brightness and flavor.
A second racking is necessary at the end of the winter before the
spring rise of temperature tends to renew the activity of the micro-
organisms which always remain in the wine. A well-made wine at
this time should be perfectly bright and all solid matters consisting of
yeast, bacteria, and coagulated proteins should have accumulated in the
sediment.
Racking should take place when possible only in settled weather,
when the barometric pressure is high. Low atmospheric pressures
diminish the solubility of the carbon dioxide with which the wine is
saturated. Under these conditions, therefore, bubbles of gas are apt
to be given off, bringing up particles of sediment and rendering the
wine cloudy. However long wine is kept in wooden casks, it will con-
tinue to deposit sediment owing to chemical changes due to the action
136 CITRUS PRODUCTS
of oxygen which penetrates slowly through the wood. Repeated rack-
ings are therefore necessary, at least twice a year until the wine is
bottled or consumed.
Abundant aeration is required during active fermentation; a more
moderate supply of oxygen is important later for the proper ageing
of the wine. Experience has shown that exactly the requisite amount of
pure filtered air for the latter purpose will obtain access to the wine
through the pores of the wood of ordinary oak casks of proper size.
If the casks are too small the oxidation will be too rapid, if too large
the maturing of the wine will be unduly prolonged. The temperature
of the storage cellar is a modifying factor. The warmer the cellar the
larger the casks should be.
With sound, completely fermented wines, all aeration, other than
that due to the porosity of the wood, should be avoided as much as
possible. This is accomplished by keeping the casks tightly bunged and
completely filled. Evaporation through the wood continually diminishes
the volume of wine and the lack must be supplied by filling up, at first
two or three times a month and later every month or two. The drier
the air of the cellar, the more frequent the fillings neccessary.
A light sulphuring of the clean casks into which the wine is racked
is usual. This should be practiced with great caution. Very little is
needed with sound wines, especially if sulphuring has been used also
before or during fermentation. A slight excess will injure the flavor,
and the amount should not exceed 2 gm. per hectoliter, while one-half
to one-third of this is sufficient for old wines. The amount can be
accurately measured only when using metabisulphite or the liquefied
gas.
All the manipulation of the wine should be conducted with strict
attention to cleanliness. This applies especially to empty casks, pumps,
and hoses. These should be thoroughly cleaned immediately after use
and, if made of metal or of other non-absorbent material, should be
kept perfectly dry. Utensils of rubber, wood, or other porous material
should be preserved from bacterial or mold growth with sulphurous
acid.
The clarification, or fining, of a perfectly sound wine may be
facilitated and hastened by thorough stirring of the yeast immediately
before racking. The yeast in settling carries down much of the finer
suspended matter, thus effecting a rough fining. Materials such as
kaolin, pure silica sand, charcoal, and filter paper can be used with
the same effect. The fining, however, is never perfect and the flavor
of the wine is often injured.
FERMENTATION PRODUCTS
The best wines are nearly always fined at least once, immediately
before bottling. One or two finings may precede this to hasten ageing,
defecation, and bottle ripeness.
The materials used for fining are soluble gelatinous or albuminous
substances which are capable of being coagulated and precipitated by
some ingredient of the wine. The best of the commonly used sub-
stances are isinglass (ichthyocol) 2 or 3 gm. per hectoliter, the white
of fresh eggs, i or 2 per hectoliter, and gelatin, 10 or 12 gm. per
hectoliter.
The proper quantity of the finings is dissolved in a little water
diluted with wine and stirred into the cask. Acids (in grape wine,
tannin, etc.), of the wine cause a gradual coagulation in minute par-
ticles throughout the liquid. These particles gradually coalesce,., form-
ing larger particles which include all the other floating solid matter
of the wine as in a net. These larger particles contracted by the alcohol
then settle to the bottom, leaving the wine bright.
The amount required varies with the quality of the fining agents
used and the composition and temperature of the wine.
To precipitate commercial gelatin of good quality about an equal
quantity of good tannin is necessary; therefore if the tannin is insuf-
ficient, as in citrus wines, it must be added ; isinglass properly prepared
requires only from one-half to one-third this amount ; egg-white requires
only minute quantities.
Specially prepared casein of milk is sometimes used for fining
wine. Its chief merit is that the acids of the wine suffice to cause its
complete precipitation and no addition of tannin is needed. Many other
albuminous substances such as milk, blood, and various proprietary
preparations are also used, but are all inferior to these mentioned and
many of them introduce foreign matters, such as milk sugar and bac-
teria, which are a source of danger to the wine.
Wines containing many wine disease-producing bacteria may be
injured by the introduction of finings. The evolution of gases due
to the bacterial action may prevent the settling and the protein matter
introduced will favor the multiplication of the wine disease organisms.
By the use of 5 to 10 gm. of sulphurous acid per hectoliter added to
the wine immediately before the addition of the gelatin, the bacteria
with the other floating particles are precipitated.
The bright wine should be racked from the finings very soon after
the sediment has settled, especially when its disease-producing bacteria
are numerous. This will be in from ten to twenty days. If the wine
is not clear in three weeks it should be filtered.
138 CITRUS PRODUCTS
Filtering is inferior to fining in producing a perfectly bright wine.
It is more rapid, however, and is useful in clearing wine refractory to
fining.
Filters of innumerable forms are used. They are of two main
types. For rough clearing of very cloudy wines some form of bag
filter is usually employed in which the wine passes through a cloth
tissue. The passage is rapid at first and the filtration imperfect. As
the solid matter accumulates on the filtering surface, the filtration im-
proves but the passage of the wine is retarded. The first wine is
passed a second time through the filter but as soon as the rate of filtra-
tion becomes too slow the operation must be stopped and the filtering
surface removed.
For wines containing little sediment the filter must be "primed."
This is accomplished by putting some finings in the wine first passed
through the filter. The priming is more effective and the output of the
filter much increased if a little infusorial earth is used with the gelatin.
For the more perfect clearing of old wines from which it may be
desired to make vinegar some form of pulp filter is used. There are
various devices by which the wine is forced through a mass of cellulose
or asbestos pulp and freed from all floating matter. Some of the best
of these, carefully used, remove nearly all of the bacteria present.
Wine that cannot be assured of storage in a cool place should be
pasteurized, unless it contains sufficient alcohol to protect it against
spoiling by bacterial fermentation. A pasteurization temperature of
1 40°- 1 50° F. is usually sufficient.
ANALYSES OF COMMERCIAL ORANGE WINE
Samples of beverages bearing the name of orange wines examined at
the Enology Laboratory of the University of California all proved to be
sweet liquors with medium to high alcoholic content and with a flavor of
orange extract or orange oil. They gave evidence of having been made
from sherry, sweetened by the addition of sugar and flavored by the addi-
tion of orange extract or oil in some cases, and in others of having been
made from poorly fermented orange juice fortified by the addition of
alcohol and sweetened by large additions of sugar. Analyses of such
"wines" are given in Table XX.
Sample 3 was evidently a sherry flavored with orange oil or extract
and sweetened. Nos. I and 2 may have been partially fermented orange
juice that had started to turn to vinegar and had been then fortified by
the addition of alcohol or brandy and sweetened by the addition of sugar
or syrup. These were all "liquors" of bad quality and mislabeled, as their
composition and flavor showed that they had no right to the title "Orange
Wine."
FERMENTATION PRODUCTS
139
TABLE XX
ANALYSES OF "ORANGE WINES"
TABLE XXI
ANALYSES OF ORANGE WINES MADE IN THE LABORATORY
Neither of these wines gave any perceptible taste of sugar. Both were
very low in alcohol as compared with the artificial orange liquors cited in
Table XX. The acid in the true orange wines is very much higher than
that in the artificial product. Some of the wines were made in the labora-
tory by treating the fresh juice with potassium metabisulphite at the rate
of 4-6 oz. per 100 gal. and allowing the juice to settle until clear. Pure
yeast was added to the clear juice after drawing it off the sediment. An-
other lot of the juice was allowed to ferment naturally. Both lots of wine
were filtered after fermentation. Neither gave any trouble and both gave
a brilliantly clear wine.
SPARKLING WINES
In some countries the filtered wine is made into sparkling wine as
follows, after the first fermentation is over. To the filtered wine is added
1.5 per cent cane sugar previously made into a thick syrup and boiled with
a little citric acid. This would be about 1.8 oz. of sugar per gallon of
wine. To this, a little champagne yeast is added and the wine bottled
in champagne bottles. The bottles are corked with champagne corks and
left in a warm place for a few days until fermentation starts in the bottles.
The corks must fit very tightly and must be well tied down. During the
first few weeks, the bottles should be turned often to prevent the yeast
from sticking to the sides of the bottles. They are then placed in a cool
place until fermentation in the bottle is complete after which they are
turned cork downward for several months. The yeast sediment settles
out on the cork. To remove this the bottle is then held in a slanting
position and the cork released by cutting the string that holds it. The cork
is shot out of the bottle by the gas pressure within and carries the yeast
140 CITRUS PRODUCTS
sediment with it. It must be replaced quickly by a new cork, before too
much gas escapes or before too much of the wine is lost.
A sparkling orange wine made in the Zymology Laboratory of the
University of California, though not relieved of its yeast sediment as de-
scribed above was pronounced preferable to the still wine made from the
same juice.1
PARTIALLY FERMENTED JUICE AS A BEVERAGE
One of the most satisfactory processes for securing a pleasant drink
from orange juice is that in which fermentation (with a special yeast)
is allowed to proceed for only a very short time in order to keep the
alcohol percentage within the legal limit after which the juice is pas-
teurized. The result is a sweet, light, sparkling juice even after seven
months' storage.
ORANGE VINEGAR
Nature and origin of vinegar. — Vinegar is made from various
sugary or starchy matters by alcoholic and subsequent acetic fermenta-
tion. It should contain from 4 to 8 per cent of acetic acid and natural
flavoring, coloring, and other matters varying according to its origin.
Theoretically, I per cent of alcohol converted into acetic acid will
yield 1.2 per cent of the latter. Actually, I per cent of alcohol gives
approximately i per cent of acetic acid. Therefore orange juice of II
per cent sugar, yielding alcohol of 5.5 per cent, should give vinegar con-
taining considerably over the minimum legal limit of 4 per cent acetic
acid.
Fermentation. — The transformation of the alcohol of the fermented
orange juice into acetic acid takes place only with an abundant supply
of air because it consists in the addition of the oxygen of the air to the
alcohol, in this way changing it into acetic acid or vinegar.
The quality of the vinegar will depend on the quality of the raw
material from which it is made. Wine spoiled by bacterial fermenta-
tion, moldy casks, etc., will make inferior vinegar. The wine should be
perfectly clear and clean tasting and, if necessary, should be fined,
filtered, or pasteurized immediately before use. It should contain no
antiseptic which would interfere with the development of the acetic
bacteria. Sulphurous acid is particularly troublesome in this respect,
and should be removed or oxidized by thorough aeration.
Before starting the acetic fermentation, it is a usual and good prac-
tice to add about 10 per cent of good vinegar to the liquid, which is
thus rendered acid and therefore less liable to alteration by injurious
bacteria and other micro-oranisms.
FERMENTATION PRODUCTS 141
All the processes of vinegar-making depend on the same principle,
which is to expose the wine prepared as above to the action of acetic
bacteria with full access of atmospheric oxygen at a suitable tempera-
ture. The rapidity of the process depends on the number of active
bacteria present, the nutritive value of the liquid, the temperature, and
especially the free access of oxygen.
o) Starters and pure cultures. — The 10 per cent of vinegar added to
the liquid to be fermented usually contains sufficient bacteria to insure a
prompt start. Where this is not the case, a starter may be prepared by
exposing a suitable liquid in a shallow vessel to the air of a warm room
for several days. Any liquid containing about 4 per cent of alcohol, 2 per
cent of acetic acid, and a moderate amount of nitrogenous matter is suit-
able. A decoction made by boiling 50 gm. of fresh yeast in 1,000 c.c. of
water, filtering, and adding the proper amount of vinegar and wine will
serve. After thorough aeration, such a liquid in a few days becomes cov-
ered with a film of acetic bacteria. This film may be used as a starter by
gently submerging the vessel in which it is formed in the liquid to be
acetified, or by removing with a clean sliver of wood which is afterward
floated in the liquid.
In practice, such a starter gives a sufficiently pure fermentation of
acetic bacteria. The particular species of acetic bacteria, however, is left
to chance. Pure cultures of a special selected form would in all prob-
ability improve the certainty of the production of good vinegar, but the
method has not entered into general practice.
b) Apparatus. — Most metals of all kinds should be avoided as much
as possible. The hoops of barrels and buckets may be protected by a
coating of paraffin. Pumps may be of wood or of special non-corrosive
alloys, or they may be so constructed that they will not come in contact
with the liquids.
c) Domestic method. — A cask of convenient size (40 to 200 liters)
is fitted as illustrated in Figure 14.
The wine to be acetified is poured, after filtering, if necessary, into the
cask until it is about one-half to two-thirds full, the object being to have
as large a surface as possible for the growth of the bacterial film. Free
circulation of air is insured by a 5-cm. hole in each head of the cask, one
near the surface of the liquid and one near the top of the cask. These
holes should be covered with varnished metal netting to prevent the
entrance of flies.
The top bung hole is then closed with a cork, through which a funnel
passes, furnished at its lower end with a glass or rubber tube extending to
within a few inches of the bottom of the cask. By means of this funnel
new liquid can be added without disturbing the surface film. The lower
bung hole is closed with a cork, through which passes an L-shaped glass
tube which serves as an indicator of level and which also can be used to
draw off the vinegar.
When this apparatus is working well, one-fifth to one-quarter of the
contents may be taken off every three or four weeks. This depends on
the temperature, which should be between 10° and i8°C. The vinegar
142
CITRUS PRODUCTS
drawn off is immediately re-
placed with wine which, if
added slowly, will, owing to
its lower specific gravity, re-
main at the surface in con-
tact with the bacterial film.
d) Orleans m e t h o d. —
This is practically the same
as the method just described
with slight modifications to
adapt it to large-scale opera-
tions. It is the oldest com-
mercial method and produces
vinegar of the highest quality.
Barrels of about two hec-
toliters are usually employed,
fitted essentially like that
already described but with
the omission of the funnel
and drawing-off tubes.
(Biolctti in Marshall: Microbiology)
FIG. 14. — Vinegar barrel. L, surface of
liquid; O, O, openings for entrance of air; t,
tube for introducing new supplies of wine with-
out disturbing surface films ; E, glass tube to
show level of liquid and for drawing off vine-
gar.
The wine is first cleared in a vinegar filter. This consists of a wooden
vat filled with beech chips which have been extracted by soaking for sev-
eral days in cold water. The wine remaining in contact with these chips
for three or four days deposits most of its sediment.
The cask is first one-third filled with good vinegar and ten or fifteen
liters of the filtered wine added. The same amount of wine is added every
week for four weeks by which time the cask is half full. At the end of
the fifth week an amount of vinegar equal to the wine added is drawn off
and the operation repeated. The vinegar is filtered as soon as it is drawn
off, placed in full, tightly bunged casks and kept in a cool cellar.
e) Pasteur method. — Pasteur long ago pointed out the defects of the
old Orleans method and suggested improvements. The main defects of
the old method are that it is cumbersome, laborious, slow, and costly. There
is a loss of about 10 per cent of material by evaporation and the repeated
additions of liquid break the bacterial film, which then sinks to the bot-
tom, grows anaerobically, and exhausts the nutrients of the solution without
producing acetic acid. These submerged bacteria finally form a large
gelatinous mass which interferes with the regular progress of the opera-
tions, depreciates the quality, and necessitates frequent expensive cleanings
of the casks. Many attempts, more or less successful, to overcome these
defects in accordance with Pasteur's ideas have been made, but that of
Claudon is one of the best and will serve to exemplify all.
It requires the use of a wide, shallow, covered square vat, furnished
with numerous openings near the top by which the entrance of air can be
facilitated and regulated. This vat is filled to the bottom of the air vents
with a mixture of four parts of good new vinegar and six parts of wine
which has been pasteurized at 55° and, when necessary, filtered. On top
of this liquid is floated a light wooden grating which helps to support the
bacterial film and prevent its breaking and submerging during the various
FERMENTATION PRODUCTS 143
operations. When the vat is filled, the process is started by placing a small
quantity of a good bacterial film on top of the liquid which soon becomes
completely covered when the proper conditions of temperature and aeration
are maintained.
Each acetifying vat is connected with a small measuring vat from
which the proper amount of liquid is added every day after a corresponding
amount of vinegar has been removed. These two vats constitute a unit,
several of which, usually six, are united in a battery. A factory includes
several of these batteries.
The batteries are fed from a large vat or reservoir, where the mixture
of wine and vinegar is prepared and stored. The vinegar drawn from the
batteries runs directly to filters, thence to a pasteurizer, and finally to the
storage casks.
The output of these batteries is from two to five times as great per
square meter of acetifying surface as that of the old method; the cost of
the operation is considerably less, the loss by evaporation much reduced
and the quality equal and much more under control of the manufacturer.
f) German method. — In all the methods described, the surface of the
liquid exposed to air, where alone acetification occurs, is small compared
to the volume of the liquid. In order to hasten and therefore cheapen the
process, various devices for increasing the surface in contact with air have
been designed. The simplest of these is one sometimes employed in wine-
making countries. The present pomace from wine is broken up and placed
loosely but uniformly in a tall, narrow vat. In a few days, acetic fer-
mentation commences in all parts of the mass. Wine is then sprinkled
periodically on top, and, trickling down over the pomace, it is changed to
vinegar by the bacterial film which incases every particle of the mass.
The "quick" or German method of vinegar-making is based on this principle.
The apparatus used in this method consists of a tall cylindrical or
slightly conical wooden vat provided with a perforated false head a few
inches from the bottom and another, similar in structure, at the same dis-
tance from the top. The space between these two false heads is filled with
long thin chips or shavings of beech wood which have been thoroughly
extracted, first with water and then with good, strong vinegar.
In operation, the liquid to be acetified is distributed over the top false
head intermittently in small amounts. This intermittent supply is accom-
plished by various automatic devices. If the supply is continuous, the
liquid tends to run in streams or currents in certain parts of the vat and
much of the acetifying surface is lost; if too rapid, the bacterial film is
removed from the upper part of the mass of beech chips and only the lower
part is effective.
From the false head, the liquid passes through numerous small holes
to the mass of beech chips, over which it trickles slowly and is acetified by
means of the bacterial film which covers them. By the time it reaches the
lower false head, the alcohol is in greater or less amount converted into
acetic acid. Usually the liquid must pass through from two to five times
or through an equal number of vats before it is completely changed into
vinegar. The number of passages depends on the amount of alcohol pres-
144
CITRUS PRODUCTS
DpaaaDDCjaanaqg
»•••*» l!0
(Bioletti in Marshall: Microbiology)
FIG. 15. — Rapid process vinegar apparatus. V ,
mass of beech chips over which the alcoholic
liquids run from H ; H, false head with numerous
small holes and threads for the slow and equal
distribution of the liquid; E, filtering trough for
the intermittent supply of liquid; O, opening for
the entrance and exit of air.
^ path of air. X path of liquid.
ent, the height of the
acetifying column, the ra-
pidity of the flow, the
temperature, and on the
perfection of the appara-
tus.
Oxygen is supplied by
the air which, entering
holes in the vat below the
lower false head, passes
through numerous holes
in the latter, through the
interstices between the
chips and out through
short tubes fixed in the
upper false head and holes
in the top. The passage
of air is insured by the
heating of the interior
due to the fermentation.
It can be regulated by the
number and diameter of
the air holes.
The temperature, which
should be close to 30°,
must be carefully regu-
lated. If the temperature
rises too much, the loss by
evaporation will be greatly
increased ; if it remains
too low the acetification
will be retarded.
Many modifications of
this method exist, having
principally for their ob-
jects the more complete
regulation of the temper-
ature and air supply, re-
covery of the volatile mat-
ters and avoidance of the
need of repassing the
liquid through different
acetifying columns.
g) Rotating barrels. — Several methods are in use which attempt to
combine the rapidity of the German machines with the quality of the
Orleans method. Wine and cider cannot be acetified conveniently by
the German method on account of the large amount of solids and extractive
matter they contain. This coats the beech chips rapidly and interferes with
the perfect working of the process.
FERMENTATION PRODUCTS
145
These methods make use of a barrel filled partially or wholly with
beech chips and half-filled with the liquid to be acetified. By rotating the
barrel at short intervals the liquid trickles repeatedly over the chips and
with proper aeration, the acetification is rapid and complete.
After-treatment. — Wine (and cider) vinegars, for the best results,
require ageing and careful treatment. They should be filtered and
pasteurized as soon as made and stored in clean casks which are well
bunged and kept constantly full in a cool place of even temperature.
If too dark in color they may be decolorized with pure animal charcoal
carefully extracted with acids and water.
Before using or bottling, the vinegar should be fined with isinglass.
The results obtained from the juice made into vinegar by yeast fer-
mentation followed by vinegar fermentation brought about by the addition
of strong vinegar equal in volume to one-fourth the volume of the fer-
mented juice, are given in Table XXII.
TABLE XXII
YIELDS OF ACETIC ACID IN ORANGE VINEGAR FROM FROZEN ORANGES
The juice for sample i was defecated with the help of sulphurous acid
and the clear juice was fermented with pure yeast. The clear wine was
allowed to stand several days after alcoholic fermentation and was then
drawn off the yeast and fermented into vinegar by use of a vinegar starter.
The juice for samples 2 and 3 was made from the same oranges as that of
sample I. This juice was divided into two equal portions. Neither re-
ceived any sulphurous acid or defecation and both were fermented with
pure yeast. The wine of sample 2 was drawn off the yeast and acetic
fermentation carried out as in sample i. Sample 3 was treated in the
same manner as sample 2 except that the wine was not drawn off the
yeast and sediment before acetic fermentation.
Most of the orange flavor was lost during vinegar fermentation and
the flavor of the finished vinegars was not so agreeable as that of apple
or wine vinegar. The inclusion of a little of the orange oil from the
skins during extraction of the juice might improve the flavor, or at least
increase the orange flavor.
Diseases. — The most troublesome pest of vinegar factories is a minute
nematode, the Anguillula aceti or vinegar eel. It often develops around
the edges of the surface of the liquid in vinegar barrels and in the acetify-
ing columns and, if neglected, may cause putrefaction and spoiling of the
vinegar. Frequent and thorough cleaning of all apparatus, pasteurization
146 CITRUS PRODUCTS
of liquids, and light sulphuring of empty casks will prevent its development.
Microscopic mites are sometimes troublesome in neglected factories.
They can be reduced by the methods recommended for vinegar eels and
their entrance into the barrels or acetifying columns prevented by painting
a ring of turpentine or some viscid substance around each air hole.
Vinegar flies (Drosophila cellaris) are often troublesome, but can be
excluded by proper screening of buildings and barrels.
Bacteria other than acetic may develop in vinegar and some of them
may depreciate its quality. These have been little studied but the most
harmful seem to be anaerobic forms which develop in the lower parts of
the liquid protected from oxygen by the screening film of the acetic bac-
teria. They produce butyric acid and putrid odors and, if neglected, may
completely spoil the vinegar. Sulphuring, fining, and pasteurization are
the remedies.
Darkening or persistent cloudiness may. be caused by oxidase as in
wine and cider and is controlled in the same way. A similar defect may
be caused by the tannic extractive matters of new casks or contact with
iron. Aeration followed by fining will remove the cause of the trouble.
ACETONE
Cruess has suggested the possibility of making acetone as a citrus
by-product. Acetone is obtained by the neutralization of acetic acid
with calcium carbonate or oxide followed by dry distillation of the cal-
cium acetate. During distillation the calcium acetate decomposes to
form calcium carbonate and acetone, the acetone distilling over. By
this process I gal. of alcohol when passed through the processes of
acetification, neutralization, and dry distillation will yield not more
than .6 gal. of acetone. One ton oranges <== 6 gal of absolute alcohol
= 3^ gal. acetone at $2.25 per gal. = $7.85 per ton of oranges. The
manufacture of acetone from oranges appears to be a more attractive
possibility than the manufacture of alcohol.
The manufacture of acetone, like many other fermentation processes,
requires a large number of fermentation vats for alcoholic fermenta-
tion, large crushers, pressers, many acetic acid or vinegar generators, and
a still. It is desirable to distil the alcohol from the fermented juice
before acetifying because this facilitates working of the generators,
gives a purer acetic acid, and permits the recovery of the citric acid,
a very important consideration.
The total value of a ton of oranges for acetone, citric acid, and oil
would be about $18.85 to $20.85.
CITRIC ACID FROM LEMONS
Citric acid, as its name implies, occurs principally in the juice of the
citrus fruits and is chiefly responsible for the sour taste of these fruits.
CITRIC ACID FROM LEMONS 147
Commercially it is derived mainly from the juice of the lemon, although
smaller amounts are obtained from the lime, bergamot, and orange.
Citric acid is usually made from the inferior fruit or "cull lemons"
which have been damaged by insects, fungi, or frost, or which are mis-
shapen, undersized or even oversized. It may be regarded, therefore,
as a by-product of the lemon-growing industry.
The principal use of citric acid is in the manufacture of beverages
and effervescent salts. It is also used in the manufacture of many salts
which are used in medicine, including the citrates of ammonium, bis-
muth, caffeine, iron, lithium, magnesium, potassium, quinine, and
sodium. Citric acid and sodium citrate find some application in textile
printing and in the manufacture of a few dyes. Many formulas for
photographic developers and toning baths contain citric acid or sodium
citrate. Ferric ammonium citrate is used in the manufacture of blue-
print paper. Citric acid and ammonium citrate are important labora-
tory reagents. They are essential for the determination of phosphates
in fertilizers, which ranks as one of the most important of analytical
operations.
From 8,000 lemons, pressed in a suitable press, 700 liters of juice,
containing 4.5 to 6 per cent of citric acid are obtained. Fresh lemon
juice contains also 7 to 9 per cent of glucose, 0.2 to 9.8 per cent of
saccharose (according as the lemons are sour or ripe), certain extrac-
tive, gummy, and pectic substances (about 0.2 per cent for ripe and 0.8
per cent for unripe fruit), and about 0.5 to 0.7 per cent of inorganic
salts. The presence of these substances renders it impossible to crystal-
lize the citric acid merely by concentrating the juice, even when all the
glucose is transformed into alcohol (5 to 6 per cent), so that, even at
the present time, the citric acid is separated by Scheele's classical and
rather costly process, according to which it is first converted into cal-
cium citrate.1
The high price of fuel has prevented the establishment of the citric
acid industry in Sicily, and the preparation of the acid has been monopo-
lized for a long time by England and, at the present time, largely by
Germany. Both these countries receive the raw material from Sicily, to a
small extent as lemons packed in barrels containing sea water, partly as
concentrated juice (agro cotto), but mostly as calcium citrate.
In consequence of the development of lemon-growing in Spain, and
especially in California and Australia, and also owing to an agreement
entered into by the manufacturers of citric acid, the condition of the
Sicilian growers became so critical that in 1903 the Italian Minister of
Agriculture offered a prize of £6,000 for improvements in the industry
or new processes of value to the cultivators. This sum was largely wasted
JFor a method covering the manufacture of citric acid from impure calcium
salts see H. Tobler, United States Patent No. 1288293, December 17, 1918.
148 CITRUS PRODUCTS
by commissions who achieved nothing, or by rewarding certain favored
individuals. However, at the end of 1904, Professor Restuccia, of Messina,
announced to the government the discovery of a process for the direct
extraction of citric acid by simple concentration of the juice, to which was
previously added a trace of a substance — the nature of which he did not
reveal (picric acid) — and a little animal charcoal, but this process only
led to further waste of money.
In 1910, Peratoner and Scarlata suggested the following new process
for extracting the essence and citric acid from the lemons directly, without
a conversion of the acid into the calcium salt. The juice obtained by
squeezing the minced lemons in hydraulic presses is partly distilled in a
vacuum on a water bath at 60° to recover the essence and then concentrated
in vacuo at 70° until it acquires a syrup consistency (one-tenth of the
original weight). When the syrup is cold, all the citric acid is extracted
by treatment with a mixture of alcohol and ether, in which many of the
impurities are insoluble. The alcohol and ether are recovered by distilla-
tion, and the residue diluted with a little water, filtered, and concentrated
in vacuo; after standing for twelve to twenty-four hours it sets to a
yellowish-red crystalline mass which, after defecation and decolorization
in the ordinary way (animal charcoal, etc.), gives pure colorless crystals,
the yield being 60 to 70 per cent. The remaining acid may be separated
from the mother-liquor as citrate.
In spite of the favorable opinion expressed by Professors Garelli and
Paterno, this process does not seem to have been applied practically.1
Meanwhile the crisis in Sicilian citrus industry, which had apparently
lessened as a result of the good crops and prices of 1906 and 1907, became
aggravated in 1908 owing to the diminished demand for lemons, the Amer-
ican financial crisis, and the agreement between the producers of citric
acid to limit the amount of raw material required — thus lowering prices
and exhausting the usual stock of treated products — and finally, to the
abundant production, since refuse lemons did not sell for enough in 1908
to pay for gathering.
Various measures have been taken by the Italian government to pro-
tect the citric acid industry in Sicily, but it should be possible in the
present advanced condition of technical chemistry to develop this industry
without such aid. The sulphuric acid required is now made in Sicily it-
self, and by the use of a multiple-effect evaporating plant, the consumption
of coal may be reduced to a minimum. In 1911 a large citric acid factory
was erected in the vicinity of Palermo by the firm of Goldenberg, of
Winckel, near Wiesbaden.
'Poore made (1923) experiments using collodion sacs in dialyzing lemon juice,
followed by dialysis on a large scale in an osmogene containing colodion-impreg-
nated cloth membrane, giving the following results: (i) The small quantity of
colloids present in lemon juice does not affect crystallization, but the ash and
other non-acid constituents prevent satisfactory crystallization of concentrated
fermented juice. (2) In the presence of ash and other impurities, needle and
leaflet modifications of citric acid crystals are obtained; upon the removal of half
of the ash, characteristic citric acid crystals are formed. (3) At the concentra-
tion necessary for crystallization, the mother-liquor is so viscous that the crystals
cannot be separated.
CITRIC ACID FROM LEMONS 149
In the large modern Sicilian factories, the juice is treated in almost
the same manner as in the manufacture of tartaric acid: into loo-hectoliter
masonry vessels provided with stirrers and cold-water coils are placed 20
hectoliters of concentrated juice and 80 hectoliters of water, the liquid
then being well mixed for thirty minutes and allowed to ferment, the
glucose thus converted into alcohol and the juice clarified. By passing
very cold water through the coil, the temperature of the liquid is lowered
to 5°, and a large part of the dissolved and suspended extractive and muci-
laginous matters separated; in presence of a little tannin, these matters
coagulate and do not redissolve (50 liters of sumach extract at 10° Be.
are sufficient, the liquid being stirred for fifteen to twenty minutes immedi-
ately after the addition). The solution is then passed by the filter presses
and thence into 2O-hectoliter wooden vats or into brickwork vessels simi-
lar to the preceding ones, but provided with perforated coils for direct-
steam heating. The boiling liquid is now neutralized exactly with dense
milk of lime or with powdered calcium carbonate. The latter causes froth-
ing and sometimes overflow of the liquid, but precipitates a purer calcium
citrate, while the hydroxide throws down many pectic and coloring mat-
ters. In some cases two-thirds of the acidity is neutralized with calcium
hydroxide and the remainder by the carbonate. For every 100 kilos of
citric acid present (titrated) 45 kilos of quicklime (57 of slaked lime per
80 of the carbonate) are added. After stirring while hot, the insoluble
tricalcic citrate — which forms immediately — is passed at once through the
filter presses and washed for ten minutes with very hot water, for ten
minutes with tepid water, and for five minutes with cold water, which
should remain almost colorless. In other parts of Sicily, calcium citrate
is prepared in a primitive method (with slaked lime often containing mag-
nesia, which yields soluble magnesium citrate, this being lost) and is sold
dry with a content of 64 per cent of citric acid. Three hundred kilos of
calcium citrate of this strength require, on the average, 100,000 lemons,
the peel of which yields 37 kilos of essence, selling at $1.52 per kilo. The
total cost of manufacturing calcium citrate and essence from 100,000
lemons was, before the war, about $48.60. The cakes of calcium citrate from
the filter presses are mixed in 2O-hectoliter lead-lined vessels with 15 hec-
toliters of cold water, the lime of the citrate being then neutralized exactly
with dilute sulphuric acid (1:5) (with 100 kilos of citric acid in the juice
correspond 400 kilos of this dilute acid) ; a slight excess of sulphuric acid
is always added, since the presence of unaltered calcium citrate would
hinder the crystallization of the acid.
The acid is added in portions at the rate of 5 liters per minute, the
liquid being kept well mixed and direct steam applied through a perforated
leaden coil. The mass is boiled for ten to fifteen minutes, the steam being
then suspended and the whole mixed for thirty minutes. The calcium
sulphate is then removed by means of a filter press and is washed with 200
liters of boiling water, which is added to the first filtrate, and then with
cold water, which is afterward used for treating fresh calcium citrate.
The citric acid solutions from the filter presses contain only minimal quan-
tities of sulphuric acid and certain blackish extractive matters. Concentra-
tion of the solution was formerly carried out in lead-lined wooden vessels,
150 CITRUS PRODUCTS
4 m. long, 2 m. wide, and 25 cm. deep, containing closed steam coils.
Evaporation should be rapid and the temperature should never exceed 65°
to 70°. When the liquid reaches 46° (specific gravity 1.3), almost all the
calcium sulphate previously remaining in solution separates ; the clear
liquid is then siphoned into a similar vessel underneath, the concentration
being continued until a crystalline skin forms at the surface of the liquid,
which is next transferred to wooden crystallizing vessels, 2 m. X 7° cm-
X 20 cm. (deep) ; the inner surface is polished with plumbago. After two
days, the dark-brown mother-liquors are removed and the yellowish-brown
crystals centrifuged. In order to separate traces of dissolved iron from the
mother-liquor, this is treated with a little potassium ferrocyanide and fil-
tered ; two or three further crops of dark-colored crystals are obtained, the
very dark mother-liquor finally obtained being added to fresh lemon juice.
In modern factories the citric acid solution, freed from calcium sul-
phate by filter-pressing, is concentrated in. vacuum apparatus, just as in the
sugar and tartaric acid industries, the density 45° to 50° Be. in the hot
being attained. In this way the temperature does not exceed 60° to 65°,
and with a triple-effect apparatus not only rapidity, but also economy of
fuel, is attained.
In order to remove the calcium sulphate remaining in solution, the
concentration is effected in two phases : in the first to 26° to 28° Be., the
liquid being then cooled in suitable vessels in which the gypsum deposits;
the residual liquid is then concentrated further to 48° to 50° Be. This
liquid is discharged into the crystallizing vessels, which are of lead-lined
wood and of large surface ; the mother-liquors are reconcentrated and re-
crystallized two or three times, and are finally worked up to crude calcium
citrate. The blocks of crystals left in the crystallizing vessels are broken
up with wooden mallets and centrifuged.
The brown crystals first obtained are refined and decolorized by dis-
solving them in rather more than double their weight of water (to a solu-
tion of 20° Be.) and boiling the solution with animal charcoal previously
treated with hydrochloric acid, as in the refining of tartaric acid.
The hot liquor is filter-pressed under low pressure until it becomes
clear and free from particles of charcoal. The filtrate is concentrated in a
vacuum at about 60° to 65° until crystals of citric acid form, and is then
heated to 90° and discharged into lead-lined crystallizing vessels, in which
it is stirred at intervals so as to obtain small crystals; after forty-eight
hours these are centrifuged and washed in the centrifuge with pure citric
acid solution, just as is done with sugaa.
If chemically pure citric acid free from metals is required, the concen-
tration is carried out in thickly tinned vessels and the crystallization in
wooden vessels; the traces of iron present are eliminated by addition of a
little potassium ferrocyanide and sodium sulphide.
In all the washing and refining operations, pure water with little hard-
ness is always employed.
OTHER SOURCES OF CITRIC ACID 151
OTHER SOURCES OF CITRIC ACID
CITRIC ACID BY FERMENTATION
Five different species of fungi, namely, Citromyces Pfefferianus, Citro-
myces glauber, Mucor piriformis, Citromyces citricus, and Sterigmatocystis
nigra, have been shown to produce citric acid by fermentation but compara-
tively little work has been done by the experimenters to find the value of
this formation, while up to the present time only one firm in America has
found it of commercial value.
From a review of the literature on the subject the general conditions
for this fermentation are : A neutral slightly acid solution containing proper
nutrient salts and dextrose, not exceeding 10 per cent. (However, the
fungi are not dependent on this form of sugar alone, for sucrose, lactose,
mannose, xylose, arabinose, glycerol, and alcohol all produce citric acid.)
The temperature should be carefully regulated.
The citric acid fermentation induced by certain fungi has been elabo-
rately studied, especially by Wehmer. His work on this subject and also
on the oxalic acid fermentation is familiar to all students of fungi. Wehmer
believed that the production of citric acid in more than mere traces was
characteristic of the group of fungi to which he gave the generic name
Citromyces and that oxalic acid fermentation was characteristic of Asper-
gillus niger. This seems to have been accepted by all the other workers
who have investigated either the citric or oxalic acid fermentation. Martin,
in a study of the citric acid fermentation, discarded all cultures of Asper-
gilli with the assumption that their fermentative action was well known
and that they did not produce citric acid.
It has been noted1 that many cultures of Aspcrgillus niger produced
citric acid. Although the literature on the chemical activity of Aspergillus
niger is voluminous, only one reference has been found relative to citric
acid production by this mold. In 1913 Zahorski was granted a patent in the
United States on a method for producing citric acid by fermenting sugar
solutions with Sterigmatocystis nigra. This is one of the many names that
have been used to designate fungi of the black Aspergillus group. Zahorski,
however, states that Sterigmatocystis differs distinctly from Aspergillus.
Currie at first supposed that Zahorski had worked with some very un-
usual culture of Aspergillus niger. This impression was probably wrong,
for any one of about twenty cultures studied under certain conditions pro-
duced citric acid in abundance. In fact almost any culture of Aspergillus
niger upon concentrated sugar solutions will produce much more citric
acid than oxalic acid. For conducting the citric acid fermentation a well-
selected culture of Aspergillus niger is far superior to any culture re-
sembling Wehmer's Citromyces with which Currie worked.
No cultures produced citric acid only under all conditions or oxalic acid
only under all conditions.
Many of the workers who have studied the citric acid fermentation
performed only a few experiments without being guided by a fundamental
knowledge of the metabolism of fungi or of the conditions favorable to
the reaction with which they were concerned. Experiments conducted in
^•Journal of Agriculture Research, VII (1916), I.
152 CITRUS PRODUCTS
this way are not likely to make a very definite contribution to any prob-
lem. In Currie's work three fundamental factors with regard to Asper-
gillus niger have been considered: (i) the inorganic salt requirements; (2)
the general equation of metabolism; and (3) the reaction of the medium.
Few concise statements can be made concerning the metabolism of an
organism capable of producing such a variety of chemical transformations
as Aspergillus niger. What is true for one set of conditions may not be
true for another set of conditions differing ever so little. The general
equation for the metabolism of Aspergillus niger may be written in the
following form:
carbohydrate ->• citric acid-> oxalic acid ->• carbon dioxide ->> mycelium.
These four products are nearly always present. Although their proportions
may vary widely with the culture employed and the- conditions of growth,
their sum will account for approximately 95 per cent of the consumed
carbohydrate. By a judicious selection of cultures and conditions citric
acid can be varied from none at all to over 50 per cent of the cane sugar
consumed.
The conditions especially favorable to the citric acid fermentation are
low nitrogen supply, high concentration of sugar, and nitrogen supplied as
ammonium salts rather than as nitrates. Wehmer and also others who
have studied the citric acid fermentation worked on the supposition that
the acid should be neutralized as formed or the rise in acidity would inter-
fere with the growth of mold. Oftentimes, therefore, calcium carbonate
was introduced. However, in the limitation of the growth of fungi the
hydrogen ion concentration of the media is of much more importance than
their titrable acidity. The limiting pH for Aspergillus niger is about 1.6
and 1.4. Ten per cent of citric acid is equivalent to pH 1.6 and 20 per
cent of citric acid to pH 1.4. After a concentration of 20 per cent citric
acid is reached the increase in hydrogen ion concentration is very slight in
comparison to the added acid. Consequently Aspergillus niger was found
to make considerable growth on a medium containing 40 per cent citric
acid. However, it is hardly practicable to consider producing a liquid con-
taining more than 10 per cent citric acid, for the concentration of sugar
required would exceed the point where fermentation proceeds most rapidly.
It has been found that the addition of calcium carbonate caused in many
cases contamination with foreign organisms which interfered with citric
acid production. The growth of foreign organisms could be restricted by
the addition of hydrochloric acid to a pH of about 3.5. Also the fermenta-
tion proceeded more rapidly in an acid medium than in one to which cal-
cium carbonate had been added. There is also a possibility of recovering
the citric acid directly from the fermented liquor, without going through
the expensive process of separating and decomposing the calcium citrate.
This cannot be done with lemon or lime juice because of the high per-
centage of sugar, pectin, and protein.
CITRIC ACID BY ARTIFICIAL SYNTHESIS
Citric acid can be obtained synthetically by a method evolved by
Grimaux and Adam. A saturated solution of dichloracetonic acid is neu-
ANALYSIS OF CITRIC ACID AND CITRATES 153
tralized with sodium carbonate, Na2CO3, and heated with two molecules of
potassium cyanide, KCN. The resulting solution of dicyanoacetates is sat-
urated with hydrochloric acid gas, HC1, and upon heating on a water bath
for fifteen hours, the citric is then separated as calcium citrate by neutraliz-
ing with milk of lime.
CH2C1. CH,Cl CHaCl CH2C1 CH2— CN CH,— CO,H
-fOjxx I I^-OH | OH | OH | /OH
CHOH CO -> CCT -> CC+KCN ->*CC^ -> C^
| CN | ^CO,H | COzH XC02H
CH2CI CH,Cl CH2CI CH2C1 CH,— CN CH2— C02H
dichlor- dichlor- citric acid
hydrin acetone
cyanhydrin hydrolysis hydrolysis
synthesis
More recently Lawrence has synthesized citric acid by decomposing
ethyl citrate which had been formed by heating ethyl bromacetate and ethyl
oxalylacetate in the presence of zinc.
Neither of these methods, however, has commercial importance at
present but it must not be forgotten that artificial synthesis may some-
times result in furnishing dangerous competition for the natural product as
is often the case in industrial chemistry.
ANALYSIS OF CITRIC ACID AND CITRATES
Citrate of lime. — The analysis of commercial calcium citrate (citrate
of lime) is now generally carried out by Warington's method modified as
follows :
Four gm. of citrate are boiled with 30 c.c. of 2N-hydrochloric acid in
a loo-c.c. standard measuring flask for ten minutes, the solution being then
cooled and made up to the mark with water. It is then shaken and filtered
through a dry filter paper, 50 c.c. of the filtrate being measured by a stan-
dard pipette into a beaker of 3OO-c.c. capacity and exactly neutralized with
dilute sodium hydroxide free from carbonate, using phenolphthalein as
indicator. The solution is next made slightly acid by the addition of three
or four drops of N-hydrochloric acid, 2 c.c. of a 45 per cent solution of
calcium chloride added, the liquid raised to the boiling-point and kept
boiling for fifteen minutes; to avoid bumping it is necessary to stir the
liquid well until boiling, after which it can safely be left. The hot liquid
is filtered and the precipitate on the filter paper washed with boiling water
six times. The filtrate and washings are then made just alkaline by adding
a drop or two of dilute ammonia, and boiled down to about 15 c.c. The
precipitate which forms is collected on a small filter paper and washed
with boiling water six times, using a very small quantity of water for each
washing. The filtrate and washings are treated with a drop of .ammonia,
if they have become acid, and are boiled down to about 10 c.c., but as a
rule no further precipitate will be obtained while the liquid is hot; any
precipitate which forms on cooling can be neglected.
The filter papers with their precipitates are dried at 100° and burned
together in a platinum dish with a cover. The flame should be kept low
154 CITRUS PRODUCTS
until the whole is charred, and then gradually raised until the ash is
white. The mass is then carefully treated with 30 c.c. of N-hydrochloric
acid, and boiled until all is dissolved and all carbon dioxide expelled; the
resulting solution is titrated with N/5 or N/2 sodium hydroxide, using
phenolphthalein as indicator.
The sodium hydroxide is standardized by pure potassium hydrogen
tartrate, and the N/i hydrochloric acid by the alkali; phenolphthalein is
used as indicator.
The number of c.c. of N/i HC1 used for the neutralization of the
ash X 0.070 gives the weight of citric acid in the portion tested.
An almost identical method has been described by L. and J. Gadais.
If the citrate contains much sulphate it is advisable to ash at as low
a temperature as possible, preferably with an alcohol flame. Before dis-
solving in hydrochloric acid, the ash should be treated with 10 c.c. of
hydrogen peroxide. (If, as is usual, the hydrogen peroxide contains free
acid, allowance must of course be made for it.)
Lime juice, lemon juice, and factory citric acid liquors. — The analysis
of these materials is conducted as follows: 15—20 c.c. of unconcentrated
juice, or an amount corresponding with 3 c.c. of concentrated juice (40
gm. per 100 c.c.), previously diluted to facilitate exact measurement, are
exactly neutralized with pure potassium hydroxide (N/5). The liquid,
having a volume of about 50 c.c., is heated to boiling, mixed with a slight
excess of concentrated calcium chloride solution, and kept at a gentle boil
for half an hour. The precipitate is filtered off immediately while hot,
washed with boiling water six times, and the mother-liquor and washings
again evaporated and worked up as described above under calcium citrate.
The whole of the calcium citrate collected is then dissolved in 30 c.c. of
N/i hydrochloric acid and the excess of acid estimated as above. In
dealing with the cruder factory liquors three or four evaporatings are gen-
erally necessary to separate all the calcium citrate.
The foregoing methods are not entirely free from error (O. von Spind-
ler) but have not yet been replaced by better. Incorrect results are obtained
when the calcium citrate or juice contains other acids which yield sparingly
soluble calcium salts. The presence of oxalic acid or of tartaric acid may be
detected by the fact that the cold, neutralized solution gives a precipitate
in the cold with calcium chloride.
The above-described methods are those which are generally adopted
in the citric acid industry. The following methods, selected from the
numerous processes which have been suggested may be here briefly sum-
marized.
Ulpiani and Parozzana have described a method of analysis which,
according to Klapproth gives satisfactory results for citric acid even in
presence of other organic acids. This method depends upon the fact that
citric acid, in presence of a sufficient quantity of calcium chloride, is pre-
cipitated by sodium hydroxide in the cold, when the whole of the acid is
saturated, and the hot solution when one-third of the acid is saturated.
Spica has described a method based on the formation of carbon mon-
oxide when citric acid is gently warmed with concentrated sulphuric acid;
one molecule of carbon monoxide is obtained for each molecule of citric
ANALYSIS OF CITRIC ACID AND CITRATES
155
acid present. The citric
acid is first precipitated as
calcium citrate (as in
Warington's method) and
the citrate then decom-
posed by the sulphuric
acid in the apparatus de-
scribed below (Fig. 16) :
The upper part A, is
fitted to the flask, B, (150-
c.c. capacity), by a ground
joint, and the tubes, D
and E and C and E, may
be connected respectively
through the tap, R, as
also may the flask and the
exterior. Two gm. of the
calcium citrate, moistened
with water, are intro-
duced into B, and the air
in the flask is completely
displaced by carbon diox-
ide, the absence of air be-
ing ascertained by means
of an auxiliary nitrometer,
filled with potassium hy-
droxide solution (1:5)
and attached to the T-
piece. Twenty-five c.c. of
concentrated sulphuric acid
are then run into B from
A and a slow current of
carbon dioxide occasion-
ally shaken, the carbon
monoxide evolved being
collected in a nitrometer
of 200-c.c. capacity, of
which the lower part Bf
(loo-c.c. capacity) is
graduated in fifths of a
c.c. When the volume of
gas becomes constant, the
nitrometer is allowed to
stand for half an hour
and then after washing
the gas with potassium
hydroxide solution, intro-
duced through I, the volume is read and the usual corrections are made for
temperature and pressure. One c.c. of carbon monoxide at o°C. and 760
(Allen: Commercial Organic Analysis)
FIG. 16. — Apparatus for the determination of
citric acid.
156 CITRUS PRODUCTS
mm. indicates 0.009407 gm. of citric acid (C6H8O7, H2O). The same ap-
paratus may be used for the determination of carbonate in a citrate, by de-
composing with a known volume of concentrated hydrochloric acid and
measuring the evolved carbon dioxide over water.
The foregoing process is obviously useless when other organic acids are
present which evolve carbon monoxide.
Estimation of citric acid in presence of other acids. — Beau's modifica-
tion of the method of Deniges is shown by Gowing-Scopes to give un-
reliable results owing to the action of the hydrogen peroxide (added to
remove the manganese dioxide) upon the precipitated mercury dicarboxy-
sulphoacetone. Other reagents tried for removing the manganese dioxide
either oxidized or reduced the precipitate, which in hot solutions was also
oxidized by manganese dioxide. A compound containing 72.5 to 74.0 per
cent of mercury and apparently similar to, if not identical with, the com-
pound of Deniges' was obtained by the use of a reagent prepared by adding
68 c.c. of strong nitric acid to 51 gm. of mercuric nitrate and 51 gm. of
manganese nitrate, diluting the mixture with 100 c.c. of water, and finally,
making up the solution to 250 c.c. and filtering. For the determination of
citric acid a quantity of the substance containing not more than 0.04 gm.
nor less than o.ooi gm. of the acid is exactly neutralized with N/io
alkali, using phenolphthalein as indicator, and, after the addition of 10 c.c.
of the reagent, the liquid is diluted to 200 c.c., and boiled for three hours
beneath a reflux condenser. The precipitate is washed by decantation,
collected on a weighed Gooch crucible, and again washed in the crucible,
which is then dried in a water-oven until nearly constant in weight (about
five hours). The residue should then be of a cream color, any yellow
coloration indicating the formation of basic salts, which will cause the
results to be too high. One-sixth of the weight of the precipitate gives the
amount of citric acid. In sixteen test estimations with pure citric acid
within the limits stated above, the maximum error was -(-0.0004 gm. Good
results were also obtained in the presence of tartaric, succinic, oxalic, ben-
zoic, and phosphoric acids, but when malic, lactic, or salicylic acids were
present the results were too high. Salicylic acid gives, with the reagent, a
salmon-colored precipitate, probably a nitro-derivative, while gallotannic
acid gives an orange-brown precipitate. The precipitate given by citric
acid decomposes suddenly when heated. It is very soluble in hydrochloric
acid, in strong sulphuric or nitric acid, and in solutions of halogen salts.
In the absence of sugars it is not necessary to precipitate the citric acid
as barium citrate and redissolve with phosphoric acid. An accuracy of 2
per cent to 4 per cent is claimed.
The foregoing process does not give satisfactory results in the presence
of malic and tartaric acid and is therefore not applicable to the examina-
tion of fruit juices. In such cases, according to D. S. Pratt the following
method is of especial value.
Fifty gm. of fruit juice are treated with no c.c. of 95 per cent alcohol
to remove pectin bodies. After fifteen minutes the solution is filtered and
the residue washed with 95 per cent alcohol. The resulting solution is
diluted with water to an alcoholic strength of about 50 per cent and a 20
per cent aqueous solution of barium acetate is added to precipitate the
ANALYSIS OF CITRIC ACID AND CITRATES 157
citric acid. After stirring, allowing the precipitate to settle, and filtering
the solution, the precipitate on the paper is washed with 50 per cent alcohol
to remove sugars and then the paper and its contents are dried to remove
the alcohol. The residue is then warmed with 50 c.c. of water and 3-5 c.c.
of syrupy phosphoric acid to dissolve the barium citrate. This mixture is
filtered into a graduated flask and the paper washed until the filtrate meas-
ures 100 c.c. An aliquot part of this solution containing 0.05—0.15 gm.
citric acid is measured into a 50o-c.c. distilling flask, 5—10 c.c. phosphoric
acid are added with 400 c.c. hot water, and the flask is heated. When
briskly boiling 0.05 per cent potassium permanganate solution is run in by
means of a dropping funnel at the rate of one to two drops per second
until the pink coloration is permanent.
The acetone formed by the oxidation distils off as fast as it is formed
into 30-40 c.c. of Deniges reagent. The distillation is continued till only
50—100 c.c. of solution remain in the flask.
The mixture in the receiver is then boiled gently under a reflux con-
denser for forty-five minutes after the liquid becomes cloudy. It is then
filtered hot through a Gooch crucible, washed with water, alcohol, and
ether, and dried in a water oven for thirty minutes. The weight of
precipitate multiplied by 0.22 gives the weight of citric acid originally
present.
Fresenius and Grunhut claim that the methods based on the precipita-
tion of acetone dicarboxylic acid as given above are untrustworthy in the
qualitative detection of citric acid in wines. For this purpose they recom-
mend Krug's modification of Moslinger's test which is carried out as
follows :
Fifty c.c. of wine are evaporated to a syrup. The syrup is treated with
95 per cent alcohol and filtered to remove tartrate and then evaporated to
remove alcohol. Ten c.c. of the resulting liquid are treated with acetic
acid and lead acetate. In the presence of citric acid a precipitate will be
obtained which dissolves on heating and reappears on cooling.
Tests of purity of citric acid. — Lead, arsenic, and ash are treated for as
under tartaric acid, the limiting quantities allowed being the same as in
the case of tartaric acid. It is usual, however, in commerce to require a
higher degree of freedom from lead than in the case of tartaric acid; the
amount present seldom exceeds ten parts per million. Arsenic is generally
entirely absent.
The following are the requirements of the British Pharmacopoeia, 1914:
I gm. dissolved in water requires for neutralization 14.2 c.c. of N/i solu-
tion of sodium hydroxide; yields no characteristic reaction for copper or
iron and not more than a very slight reaction for calcium or sulphates.
Lead limit 20 parts per million. Arsenic limit 1.4 parts per million. One
gm. of powdered citric acid mixed with 10 c.c. of sulphuric acid in a
test tube previously rinsed with sulphuric acid acquires not more than a
pale-yellow color when kept at a temperature of 90° for one hour (absence
of tartaric acid) ; ash not more than 0.05 per cent
The following color tests, based on Pusch's method of detecting tar-
taric acid in citric acid are described by Hill. 0.5 gm. of the sample
and of pure citric acid are placed in separate test tubes (6 in. X ^ in.)
158 CITRUS PRODUCTS
and 5 c.c. of sulphuric acid added to each. The tubes are placed simul-
taneously in an ordinary Bunsen flame; at the end of thirty seconds they
are withdrawn and examined. The results obtained with acids of different
degrees of purity are as follows :
Pure Acid Lemon-yellow Solution
5.0 per cent tartaric acid black, sulphur dioxide abundant
i.o per cent tartaric acid deep brown-black, sulphur dioxide evident
0.5 per cent tartaric acid deep red-brown
0.25 per cent tartaric acid red-brown
o.i per cent tartaric acid reddish brown
o.oi per cent tartaric acid brownish yellow
With careful manipulation and sixty seconds heating, o.oooi per cent of
tartaric acid is said to be easily detected; even o.ooooi per cent gave a
distinctly deeper tint, in good light, than the control tube. The test is also
applicable to citrates, i per cent of tartrate being easily detected. The
foregoing method is also applicable to detect sugar in citric acid; the
reactions obtained are:
i per cent of sugar •. . . cherry-red, sulphur dioxide distinct
o.i per cent of sugar sherry color, sulphur dioxide evident
o.oi per cent of sugar yellowish red, sulphur dioxide perceptible
o.ooi per cent of sugar ... . reddish yellow
o.oooi per cent of sugar reddish yellow
It was also found that 0.5 per cent of sugar could be detected, in tar-
taric acid, by the coloration after an hour's contact with cold sulphuric acid ;
by heating for ten seconds in the flame the control tube remained prac-
tically unaffected, with I per cent of sugar a red solution was obtained,
and with 0.5 per cent a red-brown ring.
Haussler describes a characteristic color reaction given by citric acid
with vanillin; the solution is evaporated to dryness after adding an alco-
holic solution of vanillin and the residue treated with three drops of dilute
sulphuric acid, heated on a water bath for fifteen minutes dissolved in
water and ammonia added. A bright-red coloration is obtained with 0.002
gm. of citric acid. The red coloration is not given by tartaric, malic,
oxalic, malonic, benzoic, salicylic, acetic, lactic, or succinic acids.
BITTER GLUCOSIDES
HESPERIDIN AND HESPERITIN
Hesperidin is a glucoside freely distributed in the family Rutaceae.
It has been shown to be present in the ripe and unripe fruit pulp of
C. aurantium, C. limonum, C. limetta, C. sinensis, C. twbilis, in the
leaves and twigs of C. aurantium, different species of Dioswia, Barosma,
and other plants. On the other hand it is absent in C, decumana.
Hesperidin was discovered by Lebreton in 1828, but was more fully
studied by Hoffmann, I876.1
*Berichte der deutschen chemischen Gesellschaft, g, 26, 685; ibid. (1876), pp.
250-52.
BITTER GLUCOSIDES 159
Method of preparation. — Cut and bruised oranges are covered with
dilute alcohol, with potassium hydroxide added in excess to neutrality.
The liquor is filtered after two days, when impure hesperidin is precipitated
by hydrochloric acid. The precipitate is boiled with acetic acid for ten
minutes, after cooling filtered from the resinous mass that is left; on
standing the hesperidin gradually separates from filtrate in white fine
needles. From 4,000 oranges about 6 oz. of hesperidin is obtained.
Properties. — Its formula is C22H26O12 and it is a glucoside, as is shown
by the reaction with dilute sulphuric acid, whereby it is decomposed into
hesperitin, C16H14Oa, glucose, C6H12O6, and rhamnose.
The white, odorless needles, are almost insoluble in cold water, diffi-
cultly soluble in hot water, more soluble in alcohol and hot acetic acid,
insoluble in benzol, chloroform, and ether. Ammonia, dilute alkalies, alka-
line earths, and pyridine dissolve hesperidin easily, the solution being
yellow to orange. If hesperidin is evaporated with dilute KOH and the
residue heated with dilute H2SO4 a red to violet color is obtained. By
heating with dilute H2SO4 hesperidin splits into rhamnose, glucose, and
hesperitin. It does not reduce Fehling's solution. Hesperidin is always
dissolved in the cell sap and separates out on the removal of water. The
addition of glycerine causes crystals to form. The melting-point of hes-
peridin crystals is 270° according to Borodin. It is not appreciably fer-
mented by yeast.
Tutin has given the following formula to hesperitin which has been
confirmed by the work of Oesterle and Kueny:
Me(K >CH:CH-CCK >OH
Microchemistry. — It is present in living cells in dissolved form. If
ripe or unripe oranges are placed in absolute alcohol hesperidin crystallizes
out in the form of sphaerocrystals. The same result may be obtained by
placing the pieces of fruit in glycerine, though in this case the sphaero-
crystals are less beautiful. Also by drying the tissue of the orange, hes-
peridin crystallizes out in the form of needles or incomplete sphaerites.
The crystallization of the glucoside out of alcohol takes place slowly; one
must wait a long time. If it is desired to produce this result quickly the
following method may be used. Cut an orange in two crosswise and
simply lay down the two halves with the cut surface upward. After one-
quarter to one-half hour the hesperidin forms on the cut sections of the
peel in the form of needles and sphaerocrystals. One might think that the
same results could be obtained by pressing out a drop and allowing it
to evaporate. Such is not the case. Hesperidin crystals from the tissue
show the same solubility relations as pure hesperidin ; e.g., their solubility
in aqueous and alcoholic alkalies with yellow color, their difficult solubility
in water by virtue of which they may be distinguished from inulin sphae-
rites to which they are similar in appearance. Hesperidin crystals are
160 CITRUS PRODUCTS
doubly refracting, and the sphaerites behave in polarized light like those
of inulin. According to Pfeffer hesperidin appears not only in the fruit
but also in all pith, bark, leaf, and flowers of the orange. It is especially
plentiful in the fruit nodes and bloom buds and in the unripe fruits. It is
found in the fruit in all the parenchymatic cells of the fruit flesh and pulp.
In order to exclude disturbing substances (alkaloids) treat the preparation
with tartaric acid alcohol and then with hot water. Put a large number of
sections on a slide in a few drops of KOH, remove the sections after some
time, let the solution dry up, and add the H2SO4.
ISOHESPERIDIN AND AURANTIAMARIN
Tanret found in the rind of the bitter orange (C. aurantium L.) (i)
a crystalline acid, C44H28O14; (2) a non-crystalline resinous body; (3)
hesperidin; (4) isohesperidin, a crystalline glucoside isomeric with hes-
peridin; (5) aurantiamarin, another glucoside to which, in part, the
bitterness of the peel is due.
NARINGIN
The bitter principle from the chopped, fresh peel of the grapefruit is
called naringin. A method for its extraction was devised by Zoller. An
extraction apparatus according to Figure 17 was set up. The percolator
A held the finely chopped peel from fifteen large grapefruit. A small tuft
of glass wool was placed in the bottom of the percolator to prevent the
return tube from being stopped up. One and one-half liters of 96 per cent
alcohol were placed in B. C contained water for the bath. The operation
is automatic and simple after the first dumping of A. Ten hours are suf-
ficient for the complete extraction of the glucoside from each charge of
peel. Of course, the oils, resins, and other substances are extracted along
with the glucoside, but no pectose material is present in the extract. To
facilitate the evolution of a continuous current of alcohol vapor from B,
a glass tube of 0.5 to i.o cm. bore is sealed at one end and inverted in the
round bottom flask. The vapor expanding in this tube causes a continuous
succession of bubbles, which tend to prevent both bumping and froth. The
extract from B is then poured into flask D of Figure 9 on page 47. E is
again used for a water bath, the temperature of which must not remain
above 80°. D is connected with a Hopkins condenser with a glass seal,
and the condenser in turn connected with a receiver to which suction to
the pressure of 10 mm. may be applied. A fine capillary tube with stop-
cock extends to the bottom of D to regulate the distillation by means of a
free current of air. The alcohol and oils are collected in F and may be
used for the extraction of another charge of peel without distilling.
The residue in D should be a golden-yellow syrup, which is taken up
in water and treated with a few cubic centimeters of 25 per cent basic lead
acetate solution to precipitate out the material other than naringin, the
glucoside. The mass is filtered without suction and the excess lead re-
moved with hydrogen sulphide, from hot solution. The clear filtrate from
the PbS precipitate is allowed to stand for a few hours, when the white
rosettes begin to form on the sides of the containing vessel, analogous to
those from the fermenting liquor. Mild shaking induces an instantaneous
BITTER GLUCOSIDES
161
(Zoller : Jour. Ind, and Eng. Chem.)
FIG. 17. — Apparatus for naringin extraction
1 62
CITRUS PRODUCTS
FIG. 1 8. — Naringin. Glucose and rham-
nose ester of naringenin.
crystallization filling the con-
tainer. Considerable heat is
evolved. Purification is accom-
plished as previously mentioned.
The white crystals are mon-
oclinic, glistening, and when
compressed are light cream in
color. They are exceedingly
fluffy so that the quantitative
yield appears larger than it
really is — from 0.2 gm. to 1.6
gm. per fruit. The crystals are
soluble in water at 20° to the
extent of one part in 8,000 of
water, though even at this dilu-
tion it is intensely bitter. This
emphasizes the fact that it is of
a greater degree of bitterness
than quinine, for which it has
often been mistaken.
After the determination of its physical constants, elementary analysis,
and chemical reactions together with the physical and chemical properties
of its hydrolytic products, Zoller was certain that he was working with
the same compound that DeVry discovered in 1857, in the flowers of the
grapefruit trees in Java. DeVry states that it occurs in all parts of the
Citrus decumana though to a much greater extent in the freshly opened
flowers. While both he and Hoffmann, and later Will, conducted researches
on this glucoside, they state that they obtained their raw product from the
residue remaining in the distillation pots at Java after removing the
"neroli oil" from the flowers of the grapefruit tree by steam distillation.
Both DeVry and Hoffman were
unable to find this same bitter
substance in the flowers or fruit
of any of a host of other cit-
rus fruits, including the bitter
orange.
Hoffman applied the name
"naringin" to the glucoside
which DeVry and he investi-
gated. Will retained the same
term in his investigations with
Tiemann. The term originated,
according to Hoffman, from the
Sanskrit word "naringi" for
orange.
Solutions of naringin in
ethyl alcohol and water are
laevo-rotatory ; tEfe molecular
rotation in alcohol at i8°C. is FlG I9._Naringenin. Phloroglucinol es-
—65.2. Its empirical formula ter of parahydroxy cinnamic acid.
BITTER GLUCOSIDES
i63
as determined from carbon and hydrogen combustion, as well as from a
study of its cleavage products, appears to be C21H26O11-4H2.O (air-dried).
Over sulphuric acid it loses three molecules of water, and when dried at
I20°C. it loses the remaining molecule of water. In the latter state it is
in the form of an impalpable powder, colored a faint tinge of yellow.
When naringin is hydrolyzed with dilute (5 per cent) HC1 or H2SO4, it
forms a mixture of rhamnose and glucose, though the quantity of glucose
is much smaller than that of rhamnose. At the same time a highly crystal-
line solid separates, insoluble in water, and was found to be the phloro-
glucinol ester of />-hydroxy-cinnamic acid.
Zoller calls attention to the behavior of the glucoside which governed
the choice of the extraction method above for its isolation from the peel.
When the air-dry naringin is heated in a receptacle over a free flame it
melts at about 83 °C., and forms a syrupy mass which turns brown on
gently increasing the temperature to ioo°C., above which violent evolution
of H2O vapor takes place and a hard, glassy, dark-brown mass results.
This mass is still bitter but dissolves with difficulty in water. On the
other hand, when a water solution of the naringin (pure crystals) is boiled,
it rapidly turned yellow to brown, the bitterness gradually disappeared and
when evaporated on a steam bath a resinous mass resulted, possessing the
odoriferous principles of caramelized sugars together with those of cu-
marin and certain phenols. No naringin could be extracted from the
mass. Likewise, when the peel was ground, dried at iio°C, and lixiviated
with water, very little naringin was obtained. Hence in the separation of
the glucoside from the fresh fruit, temperatures above 80° C. should be
avoided. In the steam distillation of the peels some of the naringin is
hydrolyzed by the steam, though the temperature in the1 distilling flask
seldom registered above 85 ° C. ; stronger suction would prevent this
decomposition.
The quantity of glucoside in the grapefruit examined was approxi-
mately as given in Table XXIII.
TABLE XXIII
*"Old" signifies that the fruit was of the same variety as the fresh market, but that it had
been kept for some time between analyses, in order to determine the effect of deterioration on
the bitterness. The "Walters" was secured directly from a fruit exchange and some of the
fruit kept at constant temperature for the time specified.
tit is difficult to give the age to any great degree of exactness.
164 CITRUS PRODUCTS
From Table XXIII it is quite evident that there is a diminution in
the naringin content during storage, more noticeable in some varieties
than in others. What has happened to the glucoside is revealed when the
sugar content of the pulp is examined over a like period. Of course, it is
not argued in Zoller's article that the increase in sugar content of the
pulp during storage is traceable solely to the glucoside, in the face of the
large pectose content. But a portion of it, he considers, may be derived
from the hydrolyzed glucoside by the aid of the enzymes present in the
fruit. Certain flavors and "pink spots" which develop simultaneously with
the decrease in naringin and increase in sugars may be traceable to the
glucoside, when the reactions of this substance are better understood.
FODDER
According to Cheney the lemon pulp from which the juice has been pressed
is used in Sicily for either fertilizer or fodder. In southern California this prac-
tice has not yet come into use. Many refuse nitrogenous substances are now being
utilized in the United States for cattle food, substances which of themselves are
unpalatable. These when mixed with black strap or refuse from sugar refineries
are converted into edible food material.
PAPER
Among the uses which have been suggested for the refuse orange and lemon
pulp is that for paper. So far as known, however, no experiments have been con-
ducted along this line.
REFERENCES FOR CHAPTER V
ALLEN, ALFRED H.
Commercial Organic Analysis, Vol. IX. Philadelphia: P. Blakiston
Sons & Co.
ANONYMOUS.
"Manufacture and Industrial Application of Ozone," Scientific Ameri-
can (February u, 1911).
"Powdered Fruit Juices," Pure Products, XVI (October, 1920), 10,
502.
BUCHNER and WUSTENFELD.
Biochemische Zeitschrift, XVII (1909), 395.
CHACE, EDWARD W.
"Grapefruit Juice," California Citrograph.
"Clarification and Preservation of Fruit Juices," ibid., V (1920), 264.
CHACE, EDWARD W., and POORE, HOMER D.
"Orange Vinegar by the Rapid Process," California Citrograph, V,
No. 9 (July, 1920), 282, 296, 297.
CHENEY, ARTHUR S.
Consular Report on the Lemon Industry in Sicily. Los Angeles : Kings-
ley, Moles, & Collins Co.
CRUESS, W. V.
"Utilization of Waste Oranges," Agricultural Experiment Station
(Berkeley, California) Bulletin No. 244 (March, 1914).
FIELD MUSEUM OF NATURAL HISTORY. BOTANY, VOL. VI, PART I, PLATE V.
(Petite Revue)
(!M Parfumerie Moderne)
DISTILLATION OF NEROLI AND PETITGRAIN OILS.
THE LIBRARY
OF THE
mrmtr *
REFERENCES 165
CRUESS, W. V., ZION, J. R., and SIFREDI, A. V.
"The Utility of Sulphurous Acid and Pure Yeast in Cider Vinegar
Manufacture," Journal of Industrial and Engineering Chemistry,
VII, No. 4 (April, 1915), 324.
CURRIE, JAMES NORMAN.
"The Citric Fermentation of Aspergillus niger," Journal of Biological
Chemistry, XXXI, No. i (July, 1917), 15-37.
DENIGES.
Ann. chim. phys. (1899), p. 18.
DUNLAP, F. L., and KUEVER, R. A.
"Enzymic Removal of Proteins from Fruit Juices," United States
Patent No. 1338684. May 4, 1920.
DUNLAP, W. B.
Bulletin of the Imperial Institute, XIII (1915), 66-87.
FERNBACH, AUGUST.
"Preparing Fresh Lemon Juice to Prevent Decomposition," United
States Patent No. 981405. January 10, 1911.
GADAIS, L., and J.
Bulletin de la Societe de chimie, IV (1909), 5, 287.
GlLDEMEISTER, E.
The Volatile Oils (ad ed.).
GORE, H. C.
"Apple Syrup and Concentrated Cider. New Products for Utilizing
Surplus in Cull Apples," United States Department of Agriculture
Yearbook (1914), pp. 227-44!
GOWING-SCOPES.
Annalist, XXXVIII (1913), 12.
GRIMAUX and ADAM.
Comptes Rendus, XC, 1252.
HAUSSLER.
Chemiker-Zeitung, XXXVIII (1914), 937.
HEMINS, M.
"Preserving Orange Juice," United States Patent No. 1343915. June
22, 1920.
KLAPPROTH.
Lunge-Keane, Vol. III.
McDERMOTT, F. ALEX.
"The Utilization of Cull Citrus Fruits in Florida," Agricultural Experi-
ment (Gainesville, Florida) Bulletin No. 135 (April, 1917).
MARSHALL, CHARLES F.
Microbiology. Philadelphia: P. Blakiston Sons & Co., 1911.
MOLINARI, ETTORE.
General and Industrial Organic Chemistry. Translated by T. H. Pope.
Philadelphia: P. Blakiston Sons & Co.
OTTAIR and MARESCALCHI.
L'Unal le sue unove utilissazioni. An English translation by F. T
Bioletti.
i66 CITRUS PRODUCTS
PRATT, D. S.
Bureau of Chemistry, United States Department of Agriculture Circu-
lar No. 88 (1912).
FERRET, M.
(Citric Acid Preparation with the Aid of Magnesia), Journal de
pharmacie et de chimie, Vol. IV, Series 4, p. 48.
PFEFFER.
"Hesperidin, ein Bestammtheil einiger Hesperideen," Botanische Zeit-
schrift, XXXII (1874), 529.
POORE, H. D.
"Effect of Dialysis on Direct Crystallization of Citric Acid from Lemon
Juice," Journal of Industrial and Engineering Chemistry, XV (1923),
775-78.
SPICA.
Chemiker-Zeitung, XXXIV (1910), 1141.
SPINDLER, O. VON.
Chemiker-Zeitung, XXVII (1903), 1263.
THOMPSON, JAMES.
"Chemical Action of B. Cloacene (Jordan) on Citric and Malic Acids
in Presence and Absence of Oxygen," Proceedings Royal Society of
London (B), LXXXVI, 1-12.
TUTTIN.
Journal of the Chemical Society, XCVII (1910), 2.
ULPIANI, C., and PAROZZANI, A.
Atti. R. Accid. Lincei (V) (1906), 15, it. 517.
WEHMER.
Beitrage zur Kenntnis einheinischer Pilze. Hannover, 1893, No. I.
Handbuch technischen Mykologie, IV (1905-7), 242.
WEISS, JOHN M., and DOWNS, C. R.
(Synthetic Citric Acid), American Chemical Society News Service
Bulletin No. 252.
Pure Products, XVI, No. 3 (March, 1920), 154.
ZAHORSKI.
United States Patent No. 1066358. July I, 1913.
ZOLLER, HARPER F.
"Some Constituents of the American Grapefruit (Citrus decumana),"
Journal of Industrial and Engineering Chemistry, X, No. 5 (May,
1918), 364.
CHAPTER. VI
PRODUCTS FROM THE SEEDS
The "pastazzo" or residue from washing citrate of lime, etc., in
the manufacture of citric acid contains a large quantity of seeds.
Bertolo considers that this waste product, of no commercial value,
could be profitably utilized for oil extracted by pressure or with sol-
vents. Little oil is obtained from the seeds by pressure, but extraction
with carbon disulphide, petroleum ether, or benzene gives 30-35 per
cent oil according to the degree of maturity. The oil from the press
is light yellow and rather fluid, with the odor of lemon and with a
somewhat bitter taste. That extracted with solvents is somewhat turbid
and gradually deposits a considerable quantity of pasty sediment con-
sisting chiefly of solid saponifiable matter. The dark-yellow color of
TABLE XXIV
167
1 68 CITRUS PRODUCTS
the oil and its green fluorescence are not removed by repeated washing
with hot water or dilute sulphuric acid. The oil is semi-drying and
resembles cottonseed oil, containing a very high percentage of liquid
glycerides and less than 28 per cent solids.
Diedrichs in investigating orange and lemon seeds found oils in them
which had the following chemical and physical properties: Lemon seed
oil, yield 50 per cent, iodine No. 107.26, saponification No. 195.98. Orange
seed oil, yield 57 per cent, iodine No. 97.26, and saponification No. 196.37.
S. Kobayashi investigated the following orange seeds and orange seed
oils. The air-dried seeds from Sinensis Engl. (yield 2 per cent), of Fortu-
nclla Japonica Swingle (3.7 per cent), and of Junos Mak. (9.50 per cent)
were crushed into small pieces and the oils extracted with petroleum ether.
The oils were purified by steam distillation, and in the latter two cases the
oils were treated with Kambara-clay (10 per cent) for decoloration.
Hewer analyzed oils from orange seeds which seeds were separated by
centrifuge from orange pulp used in marmalade. The inodorous oil was
extracted by petroleum ether with a yield of 37.5 per cent. This was
easily saponified and had a saponification No. 193.7, iodine No. 100.3, and
specific gravity at 15 °C. of 6.9208.
Fisch and Gattefosse (1920) analyzed the seed oil from "Citrus auran-
tium s. sp. Lima var. fusca." They secured the following values: specific
gravity 0.930, melting-point — 2.05, solidifying-point — 3, viscosity (Engler)
at 25° 5.7, flash point 210°, combustible point 210°, saponification No.
177.3 (easily saponified), iodine No. 83.8, total volatile acidity 94 per cent,
combined volatile acidity 91.1 per cent, free volatile acidity 2.1 per cent,
melting-point volatile acidity 28°, solidifying-point volatile acidity 27.40°.
Fisch and Gattefosse (1921) also worked with citron seed oil ("Citrus
medico var. acida") which they obtained from French equatorial Africa.
The expressed oil is an amber viscous liquid with an agreeable odor and
fluorescence analogous to certain mineral oils. It has the following physical
and chemical constants: melting-point 2.05°, solidifying-point 3.00°, spe-
cific gravity 0.930, optical rotation o.o, refractive index at 15° 1.4757,
viscosity (Engler) at 25° 5.7, point of inflammability 210.00°, point of
combustion 247.00°, total volatile acidity 94 per cent, combined volatile
acidity 91.1 per cent, free volatile acidity 2.9 per cent, melting-point volatile
acidity 28.00°, solidifying-point volatile acidity 27.40°, iodine No. 83.8,
saponification No. 177.3.
It is thought that this oil may be used in soap-making and in the
replacement of raisin-seed oil in rubber-making.
REFERENCES 169
REFERENCES FOR CHAPTER VI
BERTOLO, P.
Giom. Chim. Ind. Applic., I (1920), 54-55.
Chemical News, CXXI, No. 3168 (December 31, 1920), 322.
DIEDRICHS, A.
Z. Nahr. Genussm,, XXVII, 132.
FISCH and GATTEFOSSE.
"L'huile de graine de citron de 1'Afrique equatorial," Le Matiere
Grasses, XII (1920), 5371-72.
In Junelle, Henri, Les Huiles vegetables, pp. 385-86. Paris, 1921.
HEWER.
In Junelle, Henri, Les Huiles vegetates, p. 385. Paris, 1921.
KOBAYASHI, S.
Journal of the Chemical Industry (Japan), XXI (1918), 1235-45.
Chemical Abstracts, XIII, 1125.
CHAPTER VII
PRODUCTS IN WHICH THE ENTIRE FRUIT IS USED1
MARMALADE
Orange marmalade, most familiar to us in the form of the "Dundee
marmalade," consists essentially of a cooked mass of sugar and fruit.
Dundee marmalade is made in Scotland of bitter Seville oranges. The
oranges are imported from Southern Europe, and, after necessary
cleaning, the pulp and the peel are separated and the seeds extracted
by special highly developed machinery, patented in England. The
white pith adhering to the peel and containing bitter substances, is
separated from the peel, and the peel is then placed in large vats and
subjected to steaming for a number of hours in order to soften materi-
ally and possibly drive off a portion of the essential oil so that the
sugar in the after-process of cooking may more readily penetrate and
produce a transparent finished product. The softened peel is then run
through a peel-shredding machine, which automatically cuts the peel
into very fine shreds. The pulp meanwhile has been subjected to a
thorough grinding process to produce as fine a division of particles as
is reasonable and practicable. This pulp is then mixed with a proper
proportion of shredded peel and placed in kettles for cooking. Some
factories add sugar directly to this prepared mass, while others give a
preliminary boiling before the sugar is added with water, in order to
permit the sugar more completely to penetrate the peel and the fibrous
bodies contained in the pulp. When the mass of orange peel has been
subjected to boiling for some time with sugar, and the proper con-
sistency is reached, it is placed in jars and sealed.
The characteristic of the Dundee marmalade is its slight bitterness
and its tart flavor; sugar being the cheapest ingredient in it, formerly
approximately 2 cents per pound compared to American sugar at 4 and
'United States patents involving the use of whole fruit are: H. A. Hughes,
No. 534368, February 19, 1895 (preserved in sugar) ; A. P. Gaines, No. 683112,
September 24, 1901 (dehydration of fruit in hot sugar) ; F. M. Libby, No.
737119, August 25, 1903 (fruit compound with limes, pickled) ; A. T. Jones, No.
827730, August 7, 1906 (dry fruit in hot vacuo) ; O. Paucksch, No. 92421, June
8, 1909 (desiccating apparatus) ; G. D. Harris and J. S. Pollard, No. 1017411,
February 13, 1912 (pieced and dried, hot air) ; E. W. Cooke, No. 1025374, May
7, 1912 (dehydration by air) ; C. N. Tinkelpaugh, No. 1033637, July 23, 1912
(desiccation with sun and sugar) ; H. H. Harrison, No. 1062969, March 27, 1913
(dried fruit).
170
MARMALADE 171
5, the product is loaded with sugar in order to reduce the cost of manu-
facture to as low a point as possible.
In California the bitter Seville orange is not available. Here mar-
malade is manufactured on a large scale not only with different fruit
but by very different methods.
The fruit is cleaned if required, and then sliced with what is
known as a "kraut cutter," which is simply a revolving disk upon
which is screwed semi-circular knives which cut the fruit into slices.
These slices are sometimes subjected to a preliminary soaking in water,
but more often are placed immediately in kettles with sugar and water,
brought to a boil, and boiled until a proper consistency is reached. As
the fruit is the cheapest ingredient entering into this product, the
marmalade is loaded with fruit, although large consumers of this
product in bulk usually specify 50 per cent sugar and 50 per cent fruit
in the finished product.
The orange marmalade so made is a rather sweet mass not having
many characteristics of the Dundee marmalade. It has, however, cer-
tain uses, as in the manufacture of biscuits and the like.1
To produce from the sweet orange a marmalade very similar to
Dundee marmalade the fruit is treated as described under this but with
the addition of some more acid fruit, such as the grapefruit or lemon.
One of the characteristics in which the usual orange marmalade
produced in America differs from the Dundee product is the fact
that it jells very readily. It is a solid mass, rather than a syrupy
mixture. This is due to the fact that the white pith of the rind bearing
the pectin is not removed.
The combination of various fruits have also been made into marma-
lade, which have their own peculiar pleasant characteristics. The grape-
fruit has been made into marmalade quite extensively and has found
favor as a competitor with the Dundee product because of its inherent
tartness and slight bitterness.
One California plant makes orange marmalade according to the follow-
ing process. The fruit is first sorted, then goes to an automatic slicing
machine so that all of the valuable properties of the fruit may be easily
got into suspension in the first cooking process.
From the slicing machine the material is dropped into large steam-
jacketed kettles, where sufficient water is added to keep the fruit from
burning, and the entire mass is boiled into a heavy pulp. After a given
time, this pulp is run through a continuous press, where the acid, fruit
sugar, and other valuable properties of the fruit are pressed out and col-
JOne American concern has used many cars of this orange marmalade in
making a product similar to what is known as "Fig Newtons."
172 CITRUS PRODUCTS
lected in a large tank, in which the temperature is kept uniform and the
juice standardized for all of its valuable properties.
From this tank the juice is pumped through a large filter, which has
a capacity of over 200 gal. an hour, and in this filter all of the flocculent
matter that is carried through the press by the juice is removed, and the
resultant product is the essence of the fruit — clear, sparkling, of uniform
composition, and without any of the fibrous substances of little or no food
value, so abundant in orange products.
From the filter the juice is carried by glass-lined pipes to the cooking
department where it is mixed with a uniform quantity of sugar and the
individual lots cooked to a uniform consistency by expert women cooks. A
small part of the very best fruit is selected, choice strips of peel are taken
off from the center of these and run through a special device known as a
"shredding machine," which chops the strips of peel into very fine shreds.
This raw shred is then put in a steam cooker and cooked until it becomes
very soft. The water is then pressed off and the dry cooked shred is
added to the marmalade before it reaches the finished point. This shred
adds the flavor of the peel as well as gives a delicate body to the finished
marmalade.
After the pans of marmalade have reached the finishing point, they are
immediately removed from the cooking-table. The marmalade is emptied
into a large enameled tank, which supplies the filling machine. On one
side of the filling machine is the glass-washing machine, where the empty
glass is carried through a bath of hot water and live steam. From the
washing machine the dry sterile glasses are delivered to the filling machine.
After leaving the filling machine the glass is passed over an inspection
table, where each glass is candled by a woman inspector. At this point,
any jars containing thick or scorched shred or other discolorations are re-
moved. The inspector places the metal cap on the glasses that pass in-
spection. The glass and cap are taken off the belt by a capping-machine
operator and the caps are automatically sealed down after exhausting in a
25-inch vacuum.
At Anaheim, California, a company was organized in 1921 with a capitaliza-
tion of $100,000. This company concentrated marmalade, making it into cubes,
i cu. in. in size. A cube placed in boiling water for ten minutes forms marmalade.
A small amount of sugar and a bit of orange peel may be added. The entire
process of manufacture requires one day and one ton of fruit will make 4,000
cubes.
CULL CITRUS FRUITS AS A FERTILIZER
Waste citrus fruits are of value as an orchard fertilizer. As such they
add humus to the soil, a constituent in which citrus soils generally have
only a small amount. They also by their decomposition return to the soil
many of their inorganic constituents in very nearly the proportion that is
required. The organic contents are mostly lost during the decay of the
fruit. Several undesirable factors, however, are met with in this practice
such as the putrid smell of the decaying organic matter and the danger of
infecting the fruit on the trees with brown rot and other fungi. These
undesirable factors are sometimes eliminated (at the sacrifice of humus
value) by burning the fruit and using the ashes as fertilizer, or the fruit
may safely be thoroughly plowed under.
REFERENCES 173
REFERENCES FOR CHAPTER VII
ANONYMOUS.
"How Sunkist Marmalade Is Made," California Citrograph, IV, No. 4
(February, 1919), 80-81.
CRUESS, W. V.
"Jellies and Marmalades from Citrus Fruits," Agricultural Experiment
Station (Berkeley, California) Circular No. 146 (January, 1916).
McDERMOTT, F. ALEX.
"The Utilization of Cull Citrus Fruits in Florida," Agricultural Ex-
periment Station (Gainesville, Florida) Bulletin No. 135 (April,
1917).
MUNSON, L. S., TOLMAN, L. M., and HOWARD, B. J.
"Fruits and Fruit Products." United States Department of Agriculture
Bulletin No. 66 (Revised) (1905).
THOMAS, E. L. P.
"Preserving Citrus Fruit without Altering Its Exterior Appearance,"
U. S. Patent 1047592. December 17, 1913.
WILL, R. T.
Journal of Industrial and Engineering Chemistry, VIII (1916), 78.
CHAPTER VIII
PRODUCTS FROM THE FLOWERS1
Oil of orange flowers was known as early as the sixteenth century.
Its distillation was described for the first time by Porta. About a cen-
tury later, in the year 1680, it appears to have become a fashionable per-
fume through the Duchess Flavio Orsini, Princess of Neroli, hence
the name essence of neroli. On account of its delicate pleasant odor,
the oil has been able to hold its reputation as one of the finest of flower
perfumes. This is also true of the distilled orange-flower water, or
Aqua naphae, which is used extensively to impart an aroma to food,
confections, beverages, and toilet articles. The distillation of orange
flower oil was described by Benatius in 1806. The oil was investigated
in 1825 by Bonastre and in 1828 by Boullay.
True neroli oil is produced from the flowers of the bitter orange,
Citrus aurantium. Other trees of citrus also produce delicately scented
flowers, but their fragrance is inferior to that of the bitter orange.
The center for the neroli oil industry is in the southern part of
France where the bitter orange is cultivated for this purpose only. The
following localities are noteworthy because of their large groves :
Cannes, Le Cannet, Vallauris, and Golf e Jouan ; also the following locali-
ties near to the base of the Maritime Alps, viz., Biot, Cagnes, La Colle,
La Gaude, Gattieres, St. Jeannet, Vence, Tourettes, Gorges du Loup,
and Le Bar.
The principal harvest of blossom lasts from the beginning of May
to the middle of June and, in good years, amounts to from 2.5 to 3
million kg. Much smaller is the harvest of fall blossoms.2 In 1909 it
amounted to 270,000 kg. The flowers are collected by women with the
aid of ladders. Each blossom has to be pinched off with the finger
nails and is dropped on cloths spread underneath the tree. Only the
fully expanded flowers are picked and all injury to the buds must be
*Orange flower oil is used in perfume and as a flavoring agent in syrups. Its
odor is due principally to methylanthranilate. This substance is made synthetically
from coal tar, and in 1920 was produced by six United States firms. A synthetic
product sold under the name of neroli is used largely as a perfume by soap
makers. Comparative prices of genuine and synthetic oil of neroli are tabulated
in Part II of this book.
"Details as to yields see Report of Schimmel & Co. (October, 1899), p. 38;
(October, 1902), p. 52; (October, 1903), p. 49. Also P. Jeancard and C. Satie,
Bulletin de la Socicte de Chinne, Vol. XXIII, Series 3 (1900), p. 605; Vol.
XXIX (1909), p. 992.
174
PRODUCTS FROM THE FLOWERS 175
carefully avoided. In order to obtain satisfactory prices for the flowers,
the owners of the groves have not long ago combined in a syndicate
(Societe cooperative de production des proprietaires d'oranges des
Alpes-Maritime), to utilize the flowers not sold to others. The price
of the blossoms fluctuates between 0.5 to 1.35 fr. per kilogram.
The essential oil obtained by distillation is called "neroli" ; that by
maceration or absorption in fats and volatile solvents is called true
"orange flower oil." There is a marked difference in the products ob-
tained in these various ways; this is especially true of the oil obtained
by distillation. The oil thus produced, although suprisingly delightful,
has quite a different odor from the fresh orange flowers. Of the other
methods, the ones by solvents and by absorption (enfleurage) yield the
finest article true in odor to the natural flower.
The method most generally followed is distillation. Among the
reasons for this preference is the higher yield of oil obtained. From
1,000 kg. of flowers the following are the results:
Gm. of Oil
1. By distillation 1,000
2. By maceration 400
3. By absorption (i.e., enfleurage') .... 100
4. By volatile solvents 600
Another advantage of the distillation method consists in the simul-
taneous production of a quantity of orange flower water. In former
times when the perfumery industry was still in its infancy, the orange
flower water was the main item of production, the oil itself being
considered a by-product and sold cheaply. Now conditions are reversed.
The yield of oil is greatly influenced by the temperature and at-
mospheric conditions prevailing at the harvest time. In warm and dry
weather it may run up as high as 1,400 gm. per 1,000 kg. of flowers;
under exceptionally favorable conditions, it may even reach 1,700 gm.
per 1,000 kg. But under adverse conditions such as damp, cool, and
variable weather, a considerable dimunition is experienced. As a rule,
toward the end of the flowering season, on account of the warmer tem-
perature, the larger yields are obtained. The flowers are distilled im-
mediately after gathering. Special distilling apparatus is employed
whereby the distillation is effected very rapidly and completely, and in
a few weeks enormous quantities of flowers are worked over.
In connection with orange flower oil it is interesting to note that
the oil from spring flowers is similar to the oil from the young branches
whereas oil from fall flowers is similar to oil from the older branches.
Chemically the oil from the spring flowers is poorer in esters and total
176 CITRUS PRODUCTS
alcohols. From such a consideration Laloue concludes terpenes to be
formed in the green parts of plants.
The importation of neroli into the United States has greatly in-
creased during the last sixteen years as is shown in the table in the
Appendix in Part II of this book. The value per pound, however, has
decreased. This particular branch of the citrus product industry has no
direct bearing on American citrus production as neroli is not made here.
Nevertheless it has an indirect influence on the other phases of the situa-
tion as can be easily comprehended. Orange flower oil might prove a
worthy field for investigation.
OIL OF NEROLI BIGARADE
Oil of neroli from the flowers of the bitter orange is more valuable
and is known as "Oil of Neroli Bigarade" (ordinary neroli oil ; the oil
from the sweet orange flower is known as "Oil of Neroli Portugal").
The greater part of the oil of commerce is distilled in Southern France,
one of the chief centers being Vallauris. In this region the yield of oil
varies greatly according to the time of season when the flowers are
collected. The season commences about the last week in April and
ends about the last week in May. The early flowers give ^2 gm. of
oil per kilo, the late flowers double this amount.
In preparation for distillation the petals are separated from the
sepals and are covered with water in the still. The contents of the still
are heated by means of coils of superheated steam. The oil which
comes over with the steam is separated and the water is sold as orange-
flower water.
Although this industry is of greatest importance in France, a small
amount of oil of equal perfume value is prepared in Yunis and Algeria.
The chief Yunisian center is the district of Nabeul. This district pro-
duces 60,000 kilos of flowers annually. The oil is distilled by Arabs,
many of whom have but small stills which hold about 3 kilos.
The industry is also being well developed in Malaga.
Properties. — The oil of neroli of commerce is a yellowish, slightly
fluorescent liquid, which becomes brownish red when exposed to light, of
an intensive, highly pleasant odor reminding of orange blossoms, and a
bitter aromatic taste. Its specific gravity is 0.870-0.880. The oil is soluble
in T.l/2—2 volumes of 80 per cent alcohol. On the further addition of alco-
hol the liquid becomes turbid, and on standing crystalline flakes consisting
of paraffin collect on the surface. The alcoholic solution of neroli oil
distinguishes itself by a beautiful violet-blue fluorescence, which becomes
especially prominent when some alcohol is poured in a layer above the oil.
On cooling strongly, the oil becomes turbid on account of the separation
of paraffin. At times it even solidifies to a butter-like mass.
PRODUCTS FROM THE FLOWERS
177
The saponification number of good oils lies between 20 and 52, cor-
responding to an amount of 7—18 per cent of linalyl acetate. Oils with
a saponification number higher than 55 are suspicious.
In order to ascertain the properties of oils which were undoubtedly
genuine, fresh orange blossoms which were partly preserved with salt,
partly with sea water, for transportation, were distilled by the firm of
Schimmel & Co.1 and about o.i per cent of oil obtained which possessed
properties given in Table XXV.
TABLE XXV
The rotatory power of No. 2 could not be determined on account of
its dark color. Nos. 3 and 4 are distillates of the same shipment of blos-
soms. No. 4 consists only of the oil which separated directly in the receiver
on distillation. Its preparation corresponds to the method usually employed
in Southern France, where orange blossom oil is obtained as a by-product
in, the manufacture of orange blossom water. No. 3 is a normal product,
i.e., a mixture of oil separating at once and that obtained by cohobation
from water. The oils obtained by these two methods differ but slightly.
Of still greater importance for the determination of the constants of
pure commercial oils than these experiments with preserved blossoms are
the distillations of fresh material by Charabot and Fillet in Southern
France. The oils distilled in May, 1898, in Cannes and Antibes behave
as follows: The specific gravity was between 0.8720 and 0.8757, tne angle
of rotation, aD, between —1.420 and —4.06°. One-tenth part of the oil
dissolved in 1.3 to 1.6 parts of 80 per cent alcohol at 20. The amount of
ester (C10H17OCOH3) present was 13.4-18.0 per cent.
Some interesting quantitative distillations were made by Jean Gras in
Cannes during the harvest of 1899. Thirty samples were distilled at dif-
ferent periods throughout the harvest. The yield increased as the season
advanced, from 0.80 to 1.23 per cent. Schimmel & Co.2 investigated
twenty-five of the thirty samples. No marked differences were noted.
The specific gravity varied from 0.873-0.877 at 15°, saponification No.
35.3-44.8, aD — 3°22' to — 5°24' at 20°. All the oils were soluble in one
and one-half and more parts of 80 per cent alcohol.
With few exceptions, the following oils have been examined in the
laboratory of Schimmel & Co. The number in parentheses behind the
geographic designation indicates the number of samples examined.
1Report of Schimmel & Co. (October, 1891), p. 26; (October, 1894), p. 40.
"Report of Schimmel & Co. (October, 1899), p. 42.
178 CITRUS PRODUCTS
SPAIN (7)
d150 0.870 to 0.885; aD +9° 30' to +29°; nD200. 1.4705 to I.4720; acid
value 0.7 to 2.0; ester value 18 to 47, methyl anthranilate content 0.45 to
0.5 per cent.1
CALABRIA AND SICILY (ll)
dj 5o 0.860 to 0.924; aD +2°54' to +56°30'; nD200, 1.468 to 1.474! acid
value 0.3 to i.o; ester value 6 to 127, methyl anthranilate content (i
determination) 0.22 per cent.
VENEZUELA (2)
d150 0.884 to 0.887; aD — o°55' to — 1°54'; nD200 1.463 to 1.465; acid
value 1.3 to 1.9; ester value 96 to 102.
PARAGUAY (l)
d150 0.9076; aD +°°25'; acid value 6.0; ester value 72.5.
MAYOTTE (l)
d180 0.8562; QD -j-46°2'; nD200 i.47°5 ; ester value 4.6.
ALGERIA
d15« 0.8723 to 0.8768; oD +5°42/ to -j-6°6'; saponification value 72 to
91 (Chapus).
SYRIA (i)
d150 0.8758; ao i°6'; saponification value 51. 5-2
Composition. — Neroli oil is of special interest to the perfume chem-
ist because it is the first oil in which there has been found a nitrogenous
substance of great fragrance. Inasmuch as this valuable oil is indis-
pensable to the perfumer, it has recently been subjected to repeated
scientific investigation. As a result the presence of a considerable
number of substances has been established.
The following compilation of constituents pertains to the ordinary neroli
oil obtained by steam distillation.
l-y.-Pinene (melting-point of pinene nitrolbenzylamine 122° to 123°. )3
l-Camphene. Upon treating fraction 168° to 170° with glacial acetic
acid— sulphuric acid, H. Walbaum and O. Hiithig3 obtained woborneol
which, however, melted at 195° instead of at 212°. Neither did A. Hesse
and O. Zeitschel (1902) succeed in raising the melting-point beyond 200°
although they purified the woborneol through its phthalic acid ester.
Dipentene was identified in fraction 175° to 179° (aD — 1°6') by
means of its tetrabromide (melting-point 125° ).3 Previously F. Tiemann
and F. W. Semmler had obtained a tetrabromide melting at 105° from
fraction 75° (15 mm.) and had concluded the presence of limonene. Ap-
1Ibid. (October, 1903), p. 77.
^Report of Roure-Bertrand fits (April, 1911), p. 26.
^Report of Schimmel & Co. (October, 1902), pp. 54-57; also Journal fur
praktische Chemie, Vol. LXVII, Series 2 (1903.), pp. 315-25.
PRODUCTS FROM THE FLOWERS 179
parently they did not examine the optical rotation of the fraction, hence
it may be presumed that the tetrabromide melting per chance at 105° was
an impure dipentene tetrabromide. If this explanation is rejected, it must
be assumed that the neroli oil in question was adulterated with orange oil,
for pure oils yield fractions with but a slight rotation from which limonene
tetrabromide cannot be obtained directly.
Decylic Aldehyde. From fraction 70° to 82° (7 to 8 mm.) shaking
with bisulphite separated a solid compound which, when decomposed with
alkali carbonate, developed the odor of decylic aldehyde.1 Inasmuch as
characteristic derivatives of this aldehyde have not been prepared, its
presence has not yet been demonstrated. Its presence may, however, be
regarded as probable since it occurs in the closely related orange oil.
l-Linalool had firs't been found in the oil by Tiemann and Semmler.
Its presence was definitely proved by Walbaum and Hiithig2 who prepared
its phenylure thane melting at 65°. According to Hesse and Zeitschel
(1901) the oil contains about 30 per cent of linalool, partly combined with
acids.
l-Linalyl Acetate. According to Tiemann and Semmler fraction 97°
to 104° (15 mm.; dD200 0.8972) consists of linalyl acetate. When boiled
with potassium hydrochloride, it was broken up into acetic acid and
linalool. The presence of this ester was proved a second time by Hesse
and Zeitschel (1901) who assumed that the oil examined by Tiemann and
Semmler was largely adulterated with petitgrain oil. This assumption was
not correct as demonstrated in detail by Walbaum and Hiithig.
Phenyl Ethyl Alcohol. This alcohol having been found by Hesse and
Zeitschel in the oil extracted from the orange flower water, its presence in
the oil proper was demonstrated by Walbaum and Hiithig (oxidation to
phenyl acetic acid).
a-Terpineol (melting-point above 33° ; melting-point of terpinyl phenyl
urethane iia0)3 (Hesse and Zeitschel, 1902).
Nerol. This alcohol, previously unknown, was discovered by Hesse and
Zeitschel (1902) in neroli oil. Its boiling-point is somewhat lower than
that of geraniol, but it does not combine with calcium chloride, hence can
be separated from geraniol by this means. Its diphenyl urethane, however,
does not melt at 73° to 75° as stated by Hesse and Zeitschel, but at 52°
to 53°. Nerol is contained in the oil both free and as acetate.
Geraniol was first found in the oil by Tiemann and Semmler, but its
presence was first definitely established by Hesse and Zeitschel who
oxidized it to citral (melting-point of citrylidene cyanacetic acid 122°).
Jasmone. The presence in neroli oil of this ketone, first found by A.
Hesse in jasmin oil, is probable, since the corresponding fraction yielded
a semicarbazone melting at 200° to 204°. 4
^Report of Schimmel & Co. (October, 1902), pp. 54-58. Also Journal filr
praktische Chemie, Vol. LXVII, Series 2 (1903), pp. 315-25.
'Ibid.
•Ibid.
'Report of Schimmel & Co. (April, 1903), p. 55.
180 CITRUS PRODUCTS
Nerolidol. As found by Hesse and Zeitschel, the high boiling frac-
tions contain a hitherto unknown sesquiterpene alcohol C15H26O, nerolidol.
It is identical with peruviol and is characterized by a phenyl urethane melt-
ing at 37° to 38°, the formation of which, however, requires several weeks'
standing of the alcohol with the reagent.1
Farnesol, which has a still higher boiling-point than nerolidol, was
isolated by Schimmel & Co.2 by treating fraction 127° (4 to 5 mm.) with
phthalic acid anhydride. It was identified by means of its constants and
by its conversion into farnesal (melting-point of semicarbazone 127° to
132° in place of 133° to 135° ).s
Paraffin. The stearoptene of orange flower oil, also known as neroli
camphor or aurade, was first found by Boullay. It is a paraffin which
occurs in almost all flower oils and, when pure, is completely odorless and
tasteless. It melts at 55°.*
Acids. In addition to the acetic acid already mentioned, neroli oil con-
tains phenyl acetic acid5 (Hesse and Zeitschel, 1902) which is found in
larger amounts in the oil shaken out of the aqueous distillate, benzoic acid,8
also traces of palmitic acid (Hesse and Zeitschel, 1902). All of these acids
presumably exist in the oil as esters.
Phenols are present, but only in traces (Hesse and Zeitschel, 1902).
Anthranilic acid methyl ester. The most important role in the forma-
tion of the orange flower perfume is played by the small amount of anthra-
nilic acid methyl ester, which was found in the oil in 1894 in the laboratory
of Schimmel & Co.7 It is the presence of this substance that causes the
fluorescence of the oil. Anthranilic acid methyl ester NH2-C6H4-COOCH3
boils at 132° under 14 mm. pressure, melts at 25°, and while in the liquid
condition at 15° has a specific gravity of 1.168. In the undiluted condition
its odor is unpleasant. Only when greatly diluted does its odor remind of
the fragrance of orange blossoms.
Several weeks after the Report of Schimmel & Co. and with it the
report of Walbaum had been issued, there was published a paper by E. and
H. Erdmann (1899) who claimed the scientific priority of the discovery
of the ester in neroli oil.
After a rejoinder by Walbaum (1899), the Erdmanns (1900) claimed
to prove their scientific priority by a German application for letters patent,
E. 5958, in which the occurrence of anthranilic acid methyl ester in neroli
oil is mentioned without, however, producing analytical proof. These let-
VWtf. (1914), p. 72.
'Ibid.
*Cf. M. Kerschbaum, Berichte der deutschen chemischen Gesellschaft, XLVI
(1913), 1732.
*Fluckiger and Hanbury, Pharmacographia, p. 127 (ad ed.). London, 1879.
Also E. and H. Erdmann, (Berichte der deutschen chemischen Gesellschaft,
XXXII (1899), 1214 footnote.
'Report of Schimmel & Co. (October, 1902), pp. 54-58. Also Journal fiir
praktische Chemie, Vol. LXVII, Series 2 (1903), pp. 315-25.
"Ibid.
''Report of Schimmel & Co. (April, 1899), p. 32. Also H. Walbaum, Journal
fiir praktische Chemie, Vol. LIX, Series 2 (1899), p. 350.
FIELD MUSEUM OF NATURAL HISTORY.
BOTANY, VOL. VI, PART I, PLATE VI.
(Hill: The Princesse dts Ursins)
MARIE DE LA TREMOUILLE (1635-1722), PRINCESS OF NEROLI, PRINCESS OF CHALAIS,
DUCHESS OF BRACCIANO.
She is reported to have popularized orange flower perfume which has been named,
after her. Oil of Neroli.
PRODUCTS FROM THE FLOWERS
181
ters patent were published July 5, 1900, hence fifteen months after the
.publication of Walbaum's first report. It thus becomes apparent that Wal-
baum's report was made public sooner than any of the communications by
the Messrs. Erdmann.
Indol, likewise an important constituent of the oil, was first found by
Engels in neroli pomade, later by Hesse and Zeitschel in the oil itself.
Pyrrol and pyrrol derivatives, found by Erdmann (1899) in neroli oil,
do not, according to Schimmel & Co.1 and Hesse and Zeitschel, appear to
occur in genuine, unadulterated neroli oil. Probably the appearance of the
pyrrol reaction (cherry-red coloration of a pine shaving moistened with
hydrochloric acid when exposed to the vapors of the first fraction) may
be attributed to an adulteration with petitgrain oil.
At the close of their second contribution on neroli oil, Hesse and
Zeitschel presented a compilation of the constituents of the oil with ap-
proximate amounts. In somewhat modified form, in which the more recent
investigations are considered, it is reproduced in Table XXVI.
Constituents
Hydrocarbons
35 per cent
Terpene al-
cohols and
their
acetates
47 per cent
Sesquiterpene
derivatives
6 per cent
Nitrogenous
compounds
0.7 per cent
Acids and
phenols
o.i per cent
Other constitu-
TABLE XXVI
COMPOSITION OF NEROLI OIL
Approximate
Amount in Per Cent
35
30
7
2
4
1. Pinene
2. Camphe.ne'"
3. Dipentene
4. Paraffin C27
5. /-Linalool
6. /-Linalyl acetate
7. d-Terpineol
8 and 9. Geraniol -f- Nerol
10 and ii. Geranyl acetate -J-
Neryl acetate
12. d-Nerolidol
13. Anthranilic acid methyl ester 0.6
14. Indol less than o.i
15. Acetic acid
16. Palmitic acid
Decylic aldehyde (?) and esters of
ents, resn s J phenylacetic acid and benzoic
acid, jasmone and farnesol
and loss
II.2
^Report of Schimmel & Co. (October, 1902), p. 54.
1 82 CITRUS PRODUCTS
In the orange blossom water oil all of the above-mentioned constitu-
ents are likewise found, but in different ratios. Those substances which
are more soluble in water, such as the alcohols and anthranilic acid methyl
ester, exist in larger percentage in the water oil than in the neroli oil. The
more difficultly soluble constituents, such as the esters of the terpene
alcohols, are present in smaller amounts. The water oil possibly also con-
tains phenyl acetonitrile, likewise a nitrogenous substance melting at 159°
(Hesse and Zeitschel, 1902).
The orange blossom extract oil, according to Schimmel & Co.1 con-
tains, in addition to the substances found in the distilled oil, the following
substances: traces of benzaldehyde (melting-point of semicarbazone 214°) ;
a basic substance, with a decided odor of nicotine, that boils above 110°
(6 mm.) ; and a nitrogenous substance which, upon saponification of the
oil, is hydrolyzed to ammonia and phenyl acetic acid and which, in all
probability, may be regarded as the nitrile of phenyl acetic acid. The
nitrogenous substance, melting at 150° which had been found by Hesse
and Zeitschel in the water oil, has also been obtained from the extract oil.
Finally, it contains a ketone with the odor of jasmine, presumably jasmone
(melting-point of semicarbazone 204° to 205°).
Examination. — The most common and most dangerous adulterants are
the oils of bergamot and petitgrain. As these for the greater part possess
the same constituents as oil of neroli — linalool and linalyl acetate — the
detection of small amounts is impossible. Larger additions cause an in-
crease in the specific gravity and the amount of esters, which in pure neroli
oil is 7—24 per cent (saponification No. 20—69) *n bergamot oil 34—45 per
cent (saponification No. 100—130), in petitgrain oil 38—85 per cent.
Orange flower oils which show a saponification number higher than 70
are therefore rejected as suspicious.
The property of neroli oil to separate paraffin in a freezing mixture
has been employed as a test, which is not wholly irrational, as the addition
of paraffin free oils might decrease the relative amount of paraffin to such
an extent that a separation no longer takes place on cooling. It must,
however, be remembered in employing this test that some unadulterated
oils may in certain cases be poor in paraffin. When, for instance, at the
time of harvest a large amount of blossoms is to be quickly distilled, it
happens that the distillation is not carried on to its complete exhaustion
and that a smaller amount of the difficultly volatile paraffin gets into the
oil. For the rest it is necessary in testing to depend on the comparison of
the physical properties with those of good oils, especially of the odor.
OIL OF NEROLI PORTUGAL
Oil of sweet orange blossom, Neroli Portugal, i.e., the oil distilled from
the blossoms of the sweet orange does not occur at all in commerce in a
pure state. The goods sold under this designation are always a mixture of
different aurantiaceous oils.
An oil distilled in Germany from the fresh blossoms of the sweet
orange had entirely different properties from that procured from Southern
1 Re port of Schimmel & Co. (October, 1903), p. 49.
ORANGE FLOWER PERFUME 183
France.1 The blossoms used for the distillation were transported from
Southern Spain in iron casks from which the air was pumped after filling.
Properties. — The amount of oil obtained was 0.154 per cent, having the
specific gravity 0.893 at 15° and the angle of rotation ao -j-i6°8'.
An oil2 distilled in Spain revealed the following properties : dieo
0.8571; an +42° 47'; "0200 * 47274; acid value 1.6; ester value 6.8; insol-
uble in 80 per cent alcohol, readily soluble in 90 per cent alcohol with
fluorescence and the separation of traces of paraffin.
Another oil from the same source examined by Schimmel & Co. had
the following properties: d150 0.8746; aD +43°22'; nD200 1.47450; acid
value 3.7; ester value 16.7; insoluble in 10 volumes of 80 per cent alcohol,
soluble in 0.3 volume of 90 per cent alcohol, from 2.5 volumes on opales-
cence is produced.
The properties of an Algerian oil are recorded by A. Chapus: d150
0.8731; o.D -(-26° 15'; ester content, computed as linalyl acetate, 34.18 per
cent.
For scientific purposes an oil was prepared by E. Theulier in Southern
France. The oil obtained by simple distillation without cohobation was
dark yellow in color. Its odor did not remind of that of ordinary neroli
oil. Its specific gravity was 0.860 at 23°, its angle of rotation -f^g^o',
the ester content 6.35 per cent. With 90 per cent alcohol the oil produced
a silky turbidity. In the cold it separated paraffin melting at 55°.
An oil sent to Schimmel & Co.3 from Southern France had the follow-
ing properties: d150 0.8686; aD-f-45°i6'; nD200 1.47352; acid value 1.8;
ester value 16.7 ; soluble in 0.5 volume of 90 per cent alcohol, the addition
of more than 5 volumes produce produced opalescence.
Composition. — The oil examined by Theulier boiled between 160° and
233° and appeared to contain appreciable amounts of the higher terpenes.
d-Camphene was identified by means of uoborneol melting at 212°, limonene
by means of its tetrabromide melting at 105°, and rf-linalool by means of
Doebner's compound of citral melting at 198.5° which had been obtained
upon oxidation of the linalool fraction. Those portions of the oil boiling
above linalool were not examined. Anthranilic acid methyl ester was not
contained in the oil. However, this ester was found to the extent of 0.3
per cent in a Spanish oil and was identified by means of its benzoyl
derivative (melting-point 100° to 102°) and by means of its picrate
(melting-point 105° to 106°).
ORANGE FLOWER PERFUME
Orange flower perfume cannot be obtained best by steam distillation.
In order to prepare the substances which impart to them their fragrant
odor, other methods are employed. The volatile oils may be extracted
either with volatile solvents or with fat, or they may be allowed to be
absorbed by fat.
Hence a distinction is made as in the oil from citrus rind between;
^Report of Schimmel & Co. (October, 1889), p. 38.
^Report of Schimmel & Co. (October, 1903), p. 77.
3Ibid. (April, 1910), p. 79.
184 CITRUS PRODUCTS
(i) extraction with volatile solvents; (2) extraction with non-volatile
solvents, i.e., fats: (a) without the aid of heat, enfleurage, (&) with the
aid of heat, maceration. Which of these methods is best adapted to the
extraction of the perfume of a particular flower has been determined in
Southern France by long years of experience.
The extraction of orange blossoms has thus been restricted to the use
of volatile solvents and maceration. They can be treated equally well
either with volatile solvents or with warm fats.
The perfume of the orange flower is ready made and is not manufac-
tured after the flower has been separated from the plant as is the case
with the jasmine and tuberose.
According to Hesse and Zeitschel, 1,000 kilos of orange blossoms yield
only 100 gm. of volatile oil when submitted to the enfleurage process,
whereas maceration yields 400 gm. and distillation with steam 1,200 gm.
of oil, of which, however, about 400 gm. are dissolved in the aqueous
distillate.
EXTRACTION WITH VOLATILE SOLVENTS
A more detailed study of the extraction of flowers with volatile sol-
vents was made by Millon in Algiers. In addition to ether he recommended
chloroform, carbon disulphide, wood alcohol, and the low-boiling fractions
of benzin.
Millon placed the flowers into a percolator (appareil d de placement}
covered them with ether and renewed the menstrum after ten to twenty
minutes. The extract, obtained after evaporation of the ether, he kept in
open containers because he thought that the air acted favorably on the
odoriferous principle. This, however, was a deception. Unless the solvent
is carefully removed with the aid of a vacuum, mere traces of it can be
detected if the extract is kept in closed containers. Millon was not ignor-
ant of the fact that the greater part of the extract consists of vegetable
wax which is well-nigh insoluble in alcohol. Hence he determined the
amount of odoriferous substances by ascertaining the difference in weight
of the extract before and after treatment with alcohol.
The substitution of petroleum ether for ether, a practice now in com-
mon use, was suggested by Hirzel of Leipzig. His apparatus constructed
for this purpose was patented as early as 1864 in France, England, Aus-
tria, and several of the German states. The problem of utilizing on a
technical scale the extraction with ether, carbon disulphide, chloroform,
and petroleum ether was developed about the same time by Piver. The
use of methyl chloride as a solvent was recommended by C. Vincent.
In the early seventies, Roure devised a method for the preparation of
the so-called essences concretes, concentrated alcoholic perfumes, obtained
by extraction. These were exhibited at the Vienna Exposition in 1873. A
very complicated apparatus was patented by Naudin in 1875. This per-
mitted of the vacuum distillation of the solvent charged with the perfume.
However, it is only during the last twenty-five to thirty years that the
method of floral extraction with volatile solvents has found industrial ap-
plication. It was at the time that Massignon in Cannes erected a battery
of extraction apparatus similar to the diffusion apparatus employed in sugar
ORANGE FLOWER PERFUME 185
factories. The solvent saturated with perfume was evaporated in a vacuum
still. As a menstrum Massignon first used ether, carbon disulphide, methyl
chloride, and benzene. Finally he arrived at the conclusion that a petro-
leum ether of the specific gravity 0.650 was best suited. Later Massignon
sold his factory and his patents to Leon Chiris. Gradually other factories
were equipped with extraction batteries.
Practically the process of extraction with volatile solvents resolves
itself into four steps: (a) selection and purification of the solvent; (&)
the systematic extraction of the flowers; (c) evaporation of the solvent and
production of the vegetable wax saturated with perfume (cire parfumee) ;
(rf) recovery of the solvent.
Selection and purification of the solvent. — The solvent most commonly
used is petroleum ether of a specific gravity 0.650 (15°). It is purified
by consecutive treatment with sulphuric acid and caustic soda, washing
with water, and rectification with the aid of a column in a still over solid
paraffin for the purpose of removing the lighter as well as the heavier
fractions. Less frequently benzene is used since it has the disadvantage of
yielding a highly colored extract. Carbon disulphide is not serviceable
since the extract always retains some of the disagreeable odor of the
solvent. Neither has carbon tetrachloride maintained itself. The high
price of ether stands in the way of its general use.
The systematic extraction of the flowers. — The extractors 6f the
mounted extraction batteries are cylindrical vessels, varying in size, but
usually of about 5oo-liter capacity. They contain three or four low cylin-
ders made of wire gauze or of perforated tin, one placed over the other.
The extractors are air-tight and sealed with a removable cover. The
batteries are mounted in one of several ways: either the extractors are all
on a level in a circle or a row or in two tiers one over the other. With
the aid of a pump, the solvent can be passed from one extractor to another.
As a rule, three or four extractions are made within twenty-four hours.
Hence the menstrum remains in contact with the flowers from six to eight
hours. Then the flowers are replaced by fresh ones. However, during the
height of the harvest the flowers in the extractors are replaced every eight
or even every four or five hours. This hastening of the process, however,
results in an imperfect extraction.
Each batch of flowers is usually extracted three times, more rarely
twice or four times. For the third (or last) extraction fresh petroleum
ether is used, for the second, that which served for a former third extrac-
tion, and for the first extraction (i.e., for fresh flowers) the menstrum
which has once served for a second and once for a third extraction. The
percolate of each extraction is collected in a separate receiver.
How many times the same batch of flowers is to be extracted depends
upon the nature of the flowers to be examined and also on the views of the
manufacturer. A difference of opinion exists as to how often the same
petroleum ether can be used before it is to be recovered by distillation.
In order to increase the yield, the extractors have been so constructed
in several instances as to enable their being warmed. This has resulted
in a larger yield of wax but not of odoriferous constituents.
1 86 CITRUS PRODUCTS
Evaporation of the solvent, — The manufacturer operates one or several
stills under ordinary pressure or in vacuum. As a rule the bulk of the
solvent is recovered in a large still under atmospheric pressure until the
temperature has reached a point that would prove detrimental to the fine-
ness of the perfume. The distillation is then continued in a vacuum still
and finally concluded in a glass flask.
The last traces of solvent are removed by passing small amounts of
alcohol into the molten wax thus causing a violent ebullition.
The floral extracts thus obtained are designated "concrete oils" (essences
concretes). The alcoholic extracts prepared from these, according to the
method described below, are known as "floral extracts" (extraits aux
fleurs) ', the pure oils remaining after the removal of the alcohol as "quin-
tescences."
Recovery of the solvent. — On account of the expense of the solvent,
suitable devices are provided to prevent loss. Hence long and well-cooled
spirals are used to condense the vapors from the stills, the extractors, and
the reservoirs. The ejectors of the vacuum pump must also be provided
with an efficient condenser. If the process is conducted without a vacuum,
the petroleum ether losses are less, but evaporation under atmospheric
pressure, as already pointed out, results in an impairment of the fragrance.
The solvent adhering to the flowers after their extraction is recovered
by passing steam through the extractors which, for this purpose, are con-
nected with the condensers ; or by transferring the flowers to a still from
which the solvent is recovered by steam.
a) Yield. — The yield varies greatly according to the solvent and tem-
perature employed, also according to the duration of the extraction.
At ordinary temperature, petroleum ether yields for every kilo of
orange flowers from 2.0 to 4.0 gm. of concrete oils.
b) Removal of the plant wax from the concrete oils. — For this pur-
pose the essences concretes, i.e., the extracts obtained by percolation and
subsequent evaporation of the solvent, are shaken for several days with
strong alcohol, preferably in shaking machines (batteuses). The alcoholic
solution, after being separated from the insoluble wax, is cooled for
some time to 0°. This causes the separation of the dissolved wax, which is
removed by filtration. To the alcoholic filtrate a solution of common salt
is added, and the perfume, which has risen to the surface, is separated.
If necessary, the last traces of alcohol are removed with the aid of a
vacuum.
Inasmuch as the plant wax when shaken with alcohol agglutinates to a
magma from which the perfume is extracted with difficulty, Massignon has
suggested to mix it with a very hard and odorless mineral wax, thus yield-
ing a mass that can be broken up in a mortar. The particles then remain
suspended in the alcohol and are more readily extracted by it.
EXTRACTION WITH NON-VOLATILE SOLVENTS WITH THE AID OF HEAT:
INFUSION, MACERATION OR "ENFLEURAGE A CHAUD*'
The extraction by fixed solvents is, of course, obtained by the means
of fats. It is a rather ancient method of extracting perfumes. The fats
used are those of pork and beef, or a mixture of them.
ORANGE FLOWER PERFUME 187
This process of extraction is used in numerous cases, varied only ac-
cording to the flowers being dealt with. Some of these yield the whole
amount of their perfume at the time of the extraction ; others are able to
produce an extra quantity of perfume under the conditions, and the period
of association with the fatty absorbent is extended. In the former case
the extraction is performed by "hot maceration" (maceration d chaud), in
the latter by "enfleurage." The principle of this extraction is, in both
cases, intimate contact between the flowers and the fat. The odoriferous
products then leave the vegetable cells of the flower to incorporate with
the fat in which they are absorbed.
These fats must, of course, be of A-i quality, and perfectly odorless.
Before using them they are subjected to an extremely careful purification,
the object of which is to free them from all putrescible substances. The
preparation of these fats takes place in February and March, in order to
have the refined material — the corps, as it is called — ready to be used at
the time of the first deliveries of flowers (violets). After olfactory exam-
ination, the fats are crushed, melted, and filtered. The fat is then very
slowly cooled, while being constantly mashed. This operation is called
lissage. Before being employed, these corps, contained in very clean
"piles," are stored in cool cellars, connected with the various floors of the
factory, where the making of pomades takes place.
The preparation of pomades is a very delicate one, requiring much
experience from the foreman in charge who is called contremaitre
pommadier.
Maceration. — this operation is made by the hot process. It is used for
the extraction of the perfume of rose, violet, orange flower, and cassia.
The infusing material is prepared in the laboratory, and consists of
lard, to which has been added beef fat, deprived of all impurities, and
washed with a solution of alum ; this second fat is necessary to impart
the required degree of firmness. This compound grease, called the corp
prepare, is placed in a basin which is warmed on a water bath. The opera-
tion is carefully watched, and the heat regulated according to the ther-
mometer, so as to avoid overheating.
Immediately over this department is the weighing-room, where the
flowers are piled in heaps. By the aid of long sacks the workman pours a
weighed quantity into each vat through an opening in the floor just above
it. Another employee mixes together the flowers and grease with a stirrer.
The heat (50° to 70°) then bursts the cells containing the essential oil,
and the floral odor is absorbed by the warm fat. After half an hour the
contents of the vat are thrown onto a great sieve, which retains the
flowers and allows the grease to run through. The recovered, partly aroma-
tized fat is treated with fresh flowers. This is repeated ten to fifteen times1
until the fat has acquired the desired strength as to odor. There remains
the fat held by the flowers which have been kept back by the sieve. These
are filled without delay into sacks. These are then placed in a hydraulic
press which separates the last trace of perfumed fat from the flowers.
1 According to Labbe, the average is six to eight times (H. Labbe, Essais des
Huiles esscntielles, p. n. Paris).
1 88 CITRUS PRODUCTS
The fineness of the aroma of the pomade is inverse to the time of
exposure of the flowers to the action of the fat. If prolonged unduly,
other odoriferous substances that interfere with the quality of the perfume
are also absorbed.
The floral extracts obtained by maceration are known by the same names
as those obtained by the enfleurage process and are treated in like manner
for the separation of the volatile oil. Here also the recovered fat, the
corps equise, is not used again in the perfume factory but finds its way into
the soap industry.
LIME FLOWER OIL
Parry examined an oil distilled from the flowers of the Italian lime
tree (Citrus limetta), and found it to have the following characteristics:
Specific gravity 0.870; optical rotation -j-2i°3o'; it contained linalool and
methyl anthranilate.
REFERENCES FOR CHAPTER VIII
ANONYMOUS.
"The Distillation of Orange Flowers at Grasse," Journal of the Royal
Society of Arts (June 23, 1911), p. 807.
Perfumery and Essential Oil Record, Vol. XII (July 8, 1921).
BENATINS.
Berl. Jahrbuch der Pharmazie (1806), p. 256.
BONASTRE.
Journal de Pharmacie, Vol. II, Series 2 (1825), p. 529.
BOULLAY.
Journal de Pharmacie, Vol. XIV, Series 2 (1828), p. 496.
Tromsdorffs Neues Journal de Pharmazie, Vol. XIX, Series i (1829),
p. 226.
BURGESS, H. E.
Annalist, XXVI (1901), 260.
Chemisches Zcntralblatt, II (1901), 1226.
CHAPUS, A.
Journal de Pharmacie et de Chimie, Vol. XXX, Series 6 (1909), p. 484.
Report of Schimmel & Co. (October, 1910), p. 88.
CHARABOT and PILLET.
Bulletin de la Societe de Chimie, Vol. XIX, Series 2, p. 853.
ENGELS.
Journal fur praktische Chemie, Vol. LXVI, Series 2 (1902), p. 504.
ERDMANN, E. and H.
Berichte der deutschen chemischcn Gesellschaft, XXXII (1899), 1213.
Ibid., XXXIII (1900), 266.
GlLDEMEISTER, E.
The Volatile Oils, Vol. Ill (sd ed.). Translated by G. E. Kremers.
New York: John Wiley & Sons, 1922.
HIRZEL, H.
Toiletten-Chemie (3d ed.), p. 77. Leipzig, 1874.
REFERENCES 189
HESSE, A.
Berichte der deutschen chemischen Gesellschaft, XXXII (1899), 2619.
HESSE, A., and ZEITSCHEL, O.
Journal fur praktische Chemie, Vol. LXIV, Series 2 (1901, pp. 245-60.
Ibid., Vol. LXVI, Series 2 (1902), pp. 481-576.
JEANCORD and SATIE.
Abrege de la Chimie des Parfum, p. 12. Paris, 1904.
MASSIGNON.
In P. Jeancard, "Volatile Solvents Applied to Flowers," American
Perfumer, I (1907), 10.
MENAGIO.
Origini della lingua Italiana. 1685. Dictionnaire universal franqois et
latin vulgairement appele dictionnaire de trevoux, VI (1771), 178.
Paris.
MILLON.
"Memoire sur la nature des parfums et sur quelques fleurs cultivables
en Algerie," Journal de Pharmacie et de Chimie, Vol. XXX, Series
3 (1856), p. 407-
Comptes Rendus, XLIII (1856), 197.
NAUDIN, LAURENT.
Bulletin de la Societe de Chimie, Vol. XXXVIII, Series 2 (1882),
pp. 586-600.
PARRY, E. J.
The Chemistry of Essential Oils and Artificial Perfumes (2d ed.).
London, 1908.
PIVER.
In Hirzel, Toiletten-Chemie (3d ed.), p. 77. Leipzig, 1874.
PLISSON.
Journal de Pharmacie, Vol. XV, Series 2 (1829), p. 152.
PORTAE, Jo. BATT.
Neapolotanae Magial Naturalis libri viginti, etc. Romae. Jol. 118
(1563).
ROURE, LOUIS.
Report of Roure-Bertrand fils (October, 1900), p. 27.
THEULIER, E.
Bulletin de la Societe de Chimie, Vol. XXVII, Series 3 (1902), p. 278.
TIEMANN, F., and SEMMLER, F. W.
Berichte der deutschen chemischen Gesellschaft, XXVI (1893), 853,
2711 note.
VINCENT, CAMILLE.
In Piesse, Chimie des parfum (1903), p. 69.
WALBAUM, H.
Berichte der deutschen chemischen Gesellschaft, XXXII (1899), 1512.
CHAPTER IX
PRODUCTS FROM THE LEAVES
Petitgrain oil which is obtained by distillation from the leaves of
Citrus Aurantium is dealt with in chapter x, as this oil is also obtained
from the stems of the plant.
Besides volatile oils citrus leaves, especially those of oranges, con-
tain a soporific substance and an alkaloid.
Hoffman— La Roche & Co. have recently succeeded in preparing from
orange leaves a solid substance which has hypnotic properties. In the
manufacture of the hypnotic substance an aqueous extract of the leaves
is evaporated at a low temperature to a syrupy consistency. Alcohol is
added to the residue.
An alkaloid known as stachydrin, methylbetain of hygrin acid, or
dimethylbetain of o-prolin, whose structural formula is as follows:
H.C
CHfl
4
N 0
H2C CH- C : 0
H2C
= CTH18N02
CH,
has been found in the leaves of Citrus aurantium, as well as in the tubers
of Stachys tuberifera (E. Jahns). This alkaloid has a sweet taste and
H2C
H2C-
N-CH
/\
CH-COOH
•CH,
= Hygrin acid or
t-methyltetrahydropyrrole.
separates in optically inactive crystals from a mixture of alcohol and ether.
These crystals are colorless, hygroscopic, and contain one molecule of
water of crystallization. By heating at ioo°C. this water is given off.
When the hydrochloride of the ethyl ester of the substance is decomposed
190
OIL OF BERGAMOT LEAVES 191
by HC1 gas /-methyltetrahydropyrrole (2-carboxylic acid) is formed. De-
composition with potassium hydroxide yields dimethylamine and thereby
evidence is given for the foregoing constitutional formula. This formula
is further established by synthesis from the treatment of the methyl ester
of the substance with methyl iodide and then with a silver oxide.
H2C. CH-C:O > H2C. .CH-C:O
0-CH8
by
Methyl betain distillation Methyl ester
of to isomeric of
Hygrin acid form Hygrin acid
The experimenters considered that according to the constitution of
stachydrin it might be formed as a decomposition product of albumen. One
support of this hypothesis lies in the fact that in the sap of the starchy
tubers examined stachydrin was accompanied by glutamin, tyrosin, and
arginin which are products formed when albumen decomposes.
OIL OF BERGAMOT LEAVES
According to S. Gulli the distillation of bergamot leaves Citrus Ber-
gamia Risso, is conducted on a restricted scale. It is practiced between
February and April when the trees are cleaned and trimmed. The yield
is small (100 kg. of leaves yield but 150 gm. oil) and the entire output for
a year is said not to exceed 20 to 25 kg. The pure oil has a specific gravity
of about 0.870 to 0.873, an optical rotation of +25° to -{-26°. It is soluble
in an equal volume of 90 per cent alcohol. It contains about 32 to 34 per
cent of esters, computed as linalyl acetate, some of which, however, is
methyl anthranilate.
The oil is rarely met with in a pure condition. Frequently the bergamot
leaves are distilled with the addition of turpentine oil. Still more frequently
the young plants of the bitter orange tree are admixed. Rind oils also are
used freely as adulterants. However, the oil is not only adulterated itself,
but it in turn is used for the adulteration of petitgrain oil and orange
flower oil, both bitter and sweet.
192 CITRUS PRODUCTS
OIL OF PAPEDA LEAVES (CITRUS HYSTRIX D. C.)
According to B. T. Brooks, the leaves of this group of Malayan citrus
trees do not yield more than 0.08 per cent of volatile oil with the follow-
ing properties: d^ 0.9150; aD — 10.50°; nD300 1.4650; saponification value
50.2. Its odor resembles that of the oil from grapefruit.
OIL OF LEAVES OF "CITRUS TRIFOLIATA"
A Chinese neroli oil, said to have been obtained from the leaves of
Poncirus trifoliata Raf. (Citrus trifoliata, L.) has been described by Umney
and Bennett.
The yellowish-brown oil had a peculiar sweetish odor, revealed a very
faint blue fluorescence and had the following constants : d150 0.850 ; <ID
~h35° > ester content 4.97 per cent, computed as linalyl acetate; total alco-
hol content 25.17 per cent, computed as linalool. According to the investi-
gators the oil contains limonene, camphene, linalool, linalyl acetate, anthra-
nilic acid methyl ester, and a paraffin. How these substances were
identified is not stated.
OIL OF SWEET ORANGE LEAVES
Two oils (petitgrain Portugal) distilled in Southern France from the
leaves of the sweet orange tree (Citrus sinensis Osbeck) had the follow-
ing properties:1 d160 0.8602 and 0.8584; aD -{-$6° 46' and 53° 52'. An
Algerian oil was examined with the following results (Chapus) : dls<>
0.8705; aD -f-2i°33'; ester content, computed as linalyl acetate 21.6 per
cent. The oil contains (Litterer) d-camphene (melting-point of uoborneol
212°), limonene (melting-point of tetrabromide 104°), linalool (?) citral,
about 4 per cent (melting-point of naphthocinchoninic acid 197°) and
geraniol, about 12 per cent (calcium chloride compound).
OIL OF GRAPEFRUIT LEAVES
According to Brooks (cited by Parry) the leaves of a variety of
grapefruit (Citrus grandis Osbeck) growing in the Philippines, yields 1.7
per cent of an oil having the following characters:
•T/\ e
Specific gravity at 35. 0.870
Optical rotation at 30° -{-22.9°
Refractive index at 30° 1.4644
Ester number 10
It contains dipentane, 15 per cent of linalool, and I per cent of citral.
OIL OF LIME LEAVES
The leaves of the West Indian lime tree (Citrus aurantifolia Swingle)
yield, according to Parry, on distillation, an oil which has the following
characteristics :
Specific gravity 0.877-0.878
Optical rotation _[_37°_38°
Acid value 3—5
Ester value 22-25
Aldehydes 43 per cent
1Report of Roure-Bertrand fils (October, 1904), p. 35; (October, 1910), p. 42.
OIL OF MANDARIN LEAVES 193
OIL OF MANDARIN LEAVES
Production. — Upon distillation with water vapor, the leaves of the
mandarin tree (Citrus nobilis var. deliciosa Swingle) yield from 0.2 to
0.35 per cent of volatile oil which in the receiver usually separates into a
lighter and a heavier oil.
Properties. — A Spanish oil described by Schimmel1 which had been
obtained from Carcagente, was yellowish in color and showed a decided
bluish fluorescence, d150 1.0142; ttD +7°46'; ester value 216; soluble in
6 to 6.5 volumes of 80 per cent alcohol. The properties of an oil from
Southern France2 were d15» 1.005; ao -\-7°l9''> saponification value 159.
Another oil from the same source3 had d1BO 1.0643; "D -j-2°2o/; saponifica-
tion value 26.5.
The oil examined by Charabot (see under Composition) deviated the
polarized light -j-6°4o' ; saponification value 160 ; content of methyl anthra-
nilic acid methyl ester about 50 per cent.
Upon the distillation of 250 kg. of leaves in Grasse, Hesse obtained
618 gm. of light oil and 159 gm. of heavy oil. The former showed d150
0.993; «D -h5°42'; saponification value 191; content of methyl anthranilic
acid methyl ester 55.7 per cent. The heavy oil showed d150 1.033; ao
-|~3°2o'; saponification value 238; content of methyl anthranilic acid
methyl ester 67.65 per cent. Both oils congealed in a freezing mixture.
From the aqueous distillate an additional 80 gm. of oil were obtained that
contained over 80 per cent of ester. The methyl anthranilic acid methyl
ester content of the total oil was computed at 65 per cent. There were
also found in this leaf oil a small amount of other esters, likewise a non-
saponifiable base with an odor of nicotine.
Composition. — As shown by Charabot mandarin leaf oil consists prin-
cipally of methyl anthranilic acid methyl ester/ It was isolated and iden-
tified according to the method employed by Walbaum in his investigation
of mandarin oils. The melting-point of the ester was 19°. It was identified
as methyl anthranilic acid methyl ester by elementary analysis and by the
acid obtained upon saponification which melted at 179° and showed all of
the properties of methyl anthranilic acid.
^Report of Schimmel & Co. (October, 1902), p. 83.
'Ibid. (April, 1902), p. 81.
^Report of Roure-Bertrand fils (October, 1910), p. 42.
*E. Charabot and G. Laloue (Comptes Rendus, CXXXVII [1903], 996. Bul-
letin de la Societe de Chimie, Vol. XXXI, Series 3 [1904], p. 195) have made a
study of the origin of the volatile oil, more particularly of the methyl anthranilic
acid methyl ester, in the leaves and petioles of the mandarin tree. It was found
that the formation of the volatile oils in the leaves is most active while their
organs are young. Moreover, the leaves contain more oil than the petioles. Later
the terpene content of the leaves is diminished whereas that of the petioles is in-
creased. The water-soluble constituents (methyl anthranilic acid methyl ester)
seem to accumulate in the leaves during the process of vegetation, the reverse
being true of the petioles.
194 CITRUS PRODUCTS
REFERENCES FOR CHAPTER IX
BROOKS, B. T.
Philippine Journal of Science, VI, A (1911), 349.
CHAPUS, A.
Journal de Pharmacie et de Chimie, Vol. XXX, Series 6 (1909), p. 484.
CHARABOT, E.
Comptes Rendus, CXXXV (1902), 580.
GULL!, S.
Chemist and Druggist, LX (1902), 995.
HESSE, A.
Cheni. Zeitschrift, II (1903), 497.
HOFFMAN, F.-LA ROCHE & Co.
Pharmaceutical Journal, XCI, 217.
Chemical Abstracts, VII, 3198.
(Solid Soporific from Orange Leaves). German Patent No. 260481,
February I, 1912.
JAHNS, E.
"Verkommen von S. in den Blattern von C. vulgaris," Berichte der
deutschen chemischen Gesellschaft, XXIX (1896), 2065—68.
Ibid., XLII (1909), 2962, 4654.
LITTERER, G.
Bulletin de la Societe de Chimie, Vol. XXXIII, Series 3 (1905), p.
1079.
PARRY, E. J.
The Chemistry of Essential Oils and Artificial Perfumes, 2d ed. Lon-
don, 1908.
UMNEY, J. C., and BENNETT, C. T.
Pharmaceutical Journal, LXIX (1902), 146.
CHAPTER X
PRODUCTS FROM THE STEMS
OIL OF PETITGRAIN
Origin and preparation. — Petitgrain oil is obtained from the leaves,
twigs, and immature fruit of the bitter orange (Citrus Aurantium L.),
by distillation with water. Formerly the oil was principally produced
in Southern France, but toward the end of the seventies French colo-
nists began the distillation in Paraguay. The poor quality of the oils
there produced in the beginning improved in the course of the years
to such an extent that the South American oil is now generally preferred
on account of its greater reliability and uniformity to the often adul-
terated French product. The market for the Paraguay oil is Ascun-
cion1 ; the principal place of distillation is said to be the little town of
Yaguaron.
Properties. — The odor of petitgrain oil is similar to that of neroli,
but far less delicate, the taste is aromatic and somewhat bitter, the color
yellowish. Its specific gravity is 0.886-0.900. It turns the polarized
ray of light either slightly to the right or to the left; aD +5° to
— 2°45'. The oil is soluble to a clear solution in two parts of 80 per
cent alcohol. The saponification number is 110—245=38—85 per cent
of linalyl acetate.
Those oils with a lower specific gravity and a larger dextrorotation
are mostly poorer in linalyl acetate and hence inferior.2 Nevertheless,
oils have been frequently observed which, in spite of the deviations
mentioned, revealed a normal ester content. Thus for oils with from
37 to 46 per cent linalyl acetate content, the following constants have
been observed: di5« as low as 0.883; and «D +5°io' to -f n°3'. In
isolated cases the oils differed from the normal only by a stronger dex-
trorotation (e.g., d15« 0.8907; aD -f-7°48'; 39.7 per cent ester; yielded
a turbid solution with about 5 volumes and more of 70 per cent alco-
hol, soluble in i volume and more of 80 per cent alcohol). Presumably
these deviations are due to the selection of the crude material, though
unfavorable climatic conditions may be accountable therefor.
Eight different petitgrain oils recently distilled from the leaves of
the bitter orange by Charabot and Pillet in Cannes, had the following
'See Part II of this book.
"Report of Schimmel & Co. (October, 1910), p. 107.
195
196
CITRUS PRODUCTS
properties: di5o 0.8910-0.8934; aD — 5°i2' to — 6°I5'. Soluble in
i-i.i parts of 80 per cent alcohol. The amounts of ester varied from
51.5-69.6 per cent.
TABLE XXVII
PROPERTIES OF PETITGRAIN OILS OF VARIOUS SOURCES
*J. C. Umney and C. T. Bennett, Pharmaceutical Journal, LXXII (1904), ai?. .
^Report of Roure-Berlrand fils (April, ign), p. 26.
The influence of climatic conditions on the formation and compo-
sition of petitgrain oils has been studied by Jeancard and Satie for the
years 1901 to 1903.
Composition. — Pyrrol and furfurol. The vapors of the first dis-
tillate of the oil color a pine shaving, moistened with hydrochloric acid,
cherry-red (H. and E. Erdmann), a reaction that indicates the pres-
ence of pyrrol derivatives.1 Besides, the lowest fraction contains fur-
furol (color reaction with aniline hydrochloride).2
Camphene. When fraction 160° to 170° was treated with glacial acetic
acid— sulphuric acid, the odor of woborneol resulted. However, no crystals
of this alcohol could be obtained.
f$-Pinene. Upon oxidation of fraction 167° to 170° (d160 0.8474; on
— 7°io') with potassium permanganate in alkaline solution, a good yield of
sodium nopinate, crystallizing in shining laminae, was obtained. Decom-
posed with sulphuric acid, it yielded nopinic acid melting at 126° to 127°.
For further identification nopinic acid was converted into nopinone which
condensed readily with benzaldehyde forming the benzylidene compound
C9H12O : CHC6H5 melting at io7°.3
Dipentene and Limoncne. Walbaum and Hiithig proved the presence
of dipentene by means of its tetrabromide (melting-point 125°), whereas,
according to Tiemann and Semmler, limonene is contained in the oil.
Charabot and Fillet have found that the petitgrain oil distilled from the
leaves only contains no limonene and that the d-limonene occasionally
found in the oil is due to the small fruits.
^Schimmel (October, 1902), p. 69.
"Ibid. (April, 1914), p. 83.
PRODUCTS FROM THE STEMS
197
l-Linalool, which occurs both free and as ester, was first shown to be
present by Tiemann and Semmler. This observation was later verified by
Walbaum and Hii.th.ig by means of the phenylurethane of the linalool
melting at 65°, a compound previously unknown.
d-a-Terpineol. (Melting-point 34°; melting-point of phenylurethane
112° [Walbaum and Hiithig].)
Nerol. Von Soden and Zeitschel found about 2 per cent of this alcohol
in petitgrain oil.
Geraniol is present both free and as acetate (Parry) ; melting-point of
diphenyl-urethane 82° ; analysis of the silver salt of the acid (Walbaum
and Hiithig).
The highest boiling portions of petitgrain oil contain sesquiterpenes
(Tiemann and Semmler) that have not been further investigated. The
blue fluorescence of the oil is in all probability due to anthranilic acid
methyl ester.1 Upon shaking the fraction which boils in the neighborhood
of 200° with dilute sulphuric acid, Walbaum and Hiithig obtained traces of
an oil which had a strongly basic odor and which probably produces the
characteristic petitgrain odor.
Adulteration and examination. — Petitgrain oil is adulterated with
orange oil, lemon oil, and turpentine oil. These additions are readily
recognized by the lowering of the specific gravity, decrease in the saponi-
fication number and the solubility, and finally by the change in rotatory
power.
Of other adulterants that are more difficult of detection, the following
have been observed: ethyl tartrate (Parry, 1909), oleic acid (Parry, 1909),
also terpinyl acetate.2
A number of specimens of petitgrain oil have been found during the
last few years, having a somewhat low ester value, and a high optical
TABLE XXVIII
(October, 1902), p. 68.
^Report of Schimmel & Co. (October, 1912), p. 92.
198 CITRUS PRODUCTS
rotation. This is probably due to the more or less careless selections of
the material to be distilled, in which some young fruits, more mature than
usual, have been included. The ester value of such oils falls to 35 per
cent, and the optical rotation rises to 19°. A sample of petitgrain oil, of
authentic origin, distilled in Jamaica1 has been found to have a rotation
-6°45/-
Terpinyl acetate has recently been found as an adulterant of this oil.
The results in Table XXVIII are obtained on the analysis of several
samples of petitgrain oil so adulterated. The analyses are by Parry and
by Schimmel.
PETITGRAIN CITRONNIER
By this name an oil is designated which is obtained now and then from
the twigs, leaves, and unripe fruit of the lemon tree. Its odor is similar
to that of the petitgrain oil, but the lemon-like odor accompanying it be-
trays the source of the oil. Its specific gravity is 0.868-0.874; OD +22°5'
to -|~34°i2r; saponification number 14.5-32.2. The odor, also the property
of forming a crystalline compound with bisulphite solution caused the pres-
ence of citral to be suspected. In fact the citryl /3-naphtho cinchoninic acid
was obtained, and the presence of this aldehyde thereby proved.2 This oil
is not made in southern California although the territory looks promising
especially as regards oil from lemon leaves and branches. There is an
abundance of lemon prunings which at present are either burned up and
their ashes used as fertilizer or they are cut up by machines to be left as
humus in the orchards.
OIL FROM IMMATURE ORANGES
From the unripe bitter Sicilian orange, Haensel obtained upon distil-
lation 0.712 per cent of oil (aD about -(-49°), and from Spanish material
0.372 per cent (aD about -(-58°). This oil contains appreciable amounts of
a pyrrol derivative, for the vapors, even of the unfractionated oil, produce
a bluish-red color on a pine shaving moistened with hydrochloric acid
(Erdmann). The principal constituent of the oil is limonene (Haensel).
ORANGE AND LEMON WOOD
Orange and lemon wood is used for manicure sets, dentists' tools,
wagon eveners, single-trees and neck yokes. For this purpose it has been
found that young trees have greater strength than old and trees grown in
windy sections are stronger and tougher than those grown in sheltered
places. In California August and September, which are dry, are the best
months for cutting the branches or felling the trees for this purpose. For
the best results, the limbs are allowed to lie with the leaves on them for
ten days after felling; much of the moisture is evaporated from the leaves.
If treated in this manner, the wood will not check nor be damaged by
insects.
One firm received an order for fifteen carloads of wood and paid the
growers $10 per ton for logs 3 or more inches in diameter.8
\Btt//. Imp. Instit. (1913), pp. II, 437.
^Report of Schimmel & Co. (October, 1896), p. 59.
'California Citrograph, VII, No. 7 (May, 1921), 231, 256.
REFERENCES 199
REFERENCES FOR CHAPTER X
ALLEN, ALFRED H.
Commercial Organic Analysis, Vol. IX. Philadelphia: P. Blakiston's
Sons & Co., 1917.
COIT, J. E., and HODGSON, R. W.
"The June Drop of Washington Naval Oranges," Agricultural Experi-
ment Station (Berkeley, California) Bulletin No. 290 (January,
1918).
CHARABOT, E., and PILLET, L.
Bulletin de la Societe de Chimie, Vol. XXI, Series 3 (1899), pp. 73-74.
ERDMANN, E. and H.
Berichte der deutschen chemischen Gesellschaft, XXXII (1899), 1217.
GlLDEMEISTER, E.
The Volatile Oils (2d ed.), Vol. III. Translated into English by E.
Kremer. New York: John Wiley & Sons, 1922.
HAENSEL, H.
Pharm. Zeitung, LI (1906), 352.
HAWLEY, L. F., and PALMER, R. C.
"Yields from the Destructive Distillation of Certain Hardwood,"
United States Department of Agriculture Bulletin No. 120 (Septem-
ber 10, 1914).
JEANCARD, P., and SATIE, C.
Bulletin de la Societe de Chimie, Vol. XXIX, Series 3 (1903), p. 1089.
PALMER, R. C.
"Yields from the Destructive Distillation of Certain Hardwoods,"
United States Department of Agriculture Bulletin No. 508 (March
6, 1917).
PARRY, E. J.
Chemist and Druggist, LI, no.
Ibid., LXXV (1909), 410.
PASSY, J.
Bulletin de la Societe de Chimie, Vol. XVII, Series 3 (1897), P- 5J9-
SODEN, H. VON, and ZEITSCHEL, O.
Berichte der deutschen chemischen Gesellschaft, XXXVI (1903), 265.
Report of Schimmel & Co. (April, 1903), p. 62.
TIEMANN, F., and SEMMLER, F. W.
Berichte der deutschen chemischen Gesellschaft, XXV (1892), 1186.
WALBAUM, H., and HUTHIG, O.
Journal fur praktische Chemie, Vol. LXVII, Series 2 (1902), p. 321.
Report of Schimmel & Co. (October, 1902), p. 69.
INDEXES
AUTHOR INDEX
Adam, 152
Allport, W. A., 10
Ames, S. L., 10
Baetcke, E., 16
Bailey, F. S., 84
Baker, 92
Ball, N. G., 85
Barbier, P., 27
Beau, 156
Beck, C. R., 17, 28
Beckmann, 16
Benatuis, 174
Bennett, A. H., 57, 65
Bennett, C. T., 192, 196
Berte, E., 22, 29, 30, 38, 58, 65
Berthelot, 25
Bertolo, P., 167
Bertram, J., 15
Bertrand, 85, 86, 88
Berzelius, 75
Bigelow, W. D., 93
Bioletti, Frederic T., 142, 144
Blanchet, 25
Bocker, E., 33, 34, 35.
Boecker, 23
Bonastre, 174
Bonavia, E., 45
Borntrager, A., 18
Borodin, 159
Boswigi, G., 36
Bouchardat, G., 26
Boullay, 174, 180
Bourquelot, 74, 75, 78, 80, 85, 86, 96
Bouveault, L., 27
Boyles, P. R., 93, 98
Braconnot, 73, 74, 79
Braden, A. C., 120
Brooks, B. T., 192
Brooks, R. O., 60
Browne, C. A., 85, 86
Bruylants, P., 61, 62, 65
Burgess, H. E., 17, 25, 26, 27, 28, 30,
46, 67
Burke, C. E., 114
Caldwell, J. S., 94, 95, 96
Capitaine, 15, 25, 42
Chace, Edward McKay, 4, 22, 25, 29,
30, 31, 62, 63, 64, 65, 120
Chapus, A., 178, 183, 192
Charabot, E., 16, 177, 193, 195, 196
Cheney, A. S., 164
Child, J. F., 25, 27, 67
Chiris, Leon, 185
Chodnew, 75
Claudon, 142
Coit, J. E., no
Cooke, E, W., 170
Crismer, L., 17, 28
Cross, 76
Cruess, William V., 98, 114, 146
Currie, James N., 151, 152
Davenport, T. J. W. C., 10
Deidrichs, A., 168
Deniges, 52, 53, 80, 81, 84, 85, 98, 156,
157
De Vry, 162
Dick, E. M., 128
Doebner, 26, 183
Douglas, R., 93, 98
Dowzard, E., 32
Dumas, 25
Dunlap, F. L., 114
Ehestadt, 48
Ellis, C., 120
Elze, F., 15, 16
Erdmann, E., 180, 196, 198
Erdmann, H., 180, 196, 198
Fellenberg Th.f von, 80, 89, 90, 91
Fenaroli, P., 44
Fernbach, A., 120
Fisch, 168
Fischer, E., 60
Flatau, 39
Fluckiger, 180
Fony6, A., 114
Fortmann, G., 39
Franke, 17
Fremy, 73, 74, 75, ?6, 78, 79, 80, 83, 85,
87,89
Fresenius, 157
203
2O4
CITRUS PRODUCTS
Fromberg, 75
Gailey, 84
Gaines, A. P., 170
Garelli, 148
Garnett, H., 57
Gattefosse, 168
Gere, W. B., 120
Gerhardt, 25
Giampetro, A. W., 10
Gildemeister, E., 25, 26, 27, 37, 38, 39
Godeffroy, R., 17
Goldthwaite, N. E., 90, 95
Gowing-Scopes, 156
Gore, Howard C., 93, 94, 114, 120, 125
Graef, C.f 120
Gras, Jean, 177
Gray, C. E., 120
Greenhalgh, N., 37
Grimaux, 152
Gninhut, 157
Gulli, S., 46, 191
Gurber, A., 120, 121
Haensel, 66, 198
Halliday, E. G. 84
Hanbury, 180
Hardy, W. B., 91
Harris, G. D., 170
Harrison, H. H., 170
Hass Tromp R. W., de, 76, 90
Haussler, 158
Henius, Max, 114
Herissey, 74, 75, 78, 80, 85, 86, 96
Herzfeldt, 76
Hesse, A., 40, 178, 179, 180, 181, 182,
184, 193
Hewer, 168
Hill, 157
Hiltner, R. S., 64
Hirzel, H., 184
Hoffman, 158, 162
Holmes, H. N., 92
Hood, S. C., 10, 12, 43, 44
Howard, 93
Howe, H. A., 92
Hughes, H. A., 170
Huthig, O., 178, 179, 196, 197
Jahns, E., 190
Jeancard, P., 174, 196
Jeffries, F. L., 98
Johnson, M. O., 98, 120
Jones, A. T., 170
Kellogg, J. L., 120
Kern, C. A., 120
Kerschbaum, M., 180
Kestner, P., 120
Klapproth, 154
Kleber, C., 23, 24, 27, 64, 65
Kobayashi, S., 168
Krug, 157
Kueny, 159
Kuever, R., 114
Labbe, H., 39, 187
Ladell, R. S., 27
Lafont, J., 25, 26
Laloue, G., 176, 193
Lawrence, 153
Lebreton, 158
Libby, F. M., 170
Litterer, G., 192
Liotta, 13, 14
Luca, de, 38, 39
McClendon, J. P., 128
McDermott, F. A., 10
McDill, R. D. O., 93, 98, 112
McKay, W. S., 10, 120
McNair, James B., 97, 98, 101
Mallevre, 85, 86, 88
Mangin, 73, 77, 78
Martin, 151
Massignon, 184, 185
Mendelsohn, F., 114
Merrell, I. S., 120
Merrell, L. C., 120
Millon, 184
Monti, Endo, 114, 121, 122, 123
Moreschini, D., 32
Moricca, G., 17
Moslinger, 157
Mulder, G. J., 17, 28, 75
Muller, 25, 26, 27
Naudin, L., 184
Oehme, R., 120
Oesterle, 159
Ohme, C., 17
Oliveri, V., 27
Oppenheim, A., 25
AUTHOR INDEX
205
Orsini, Flavio, 174
Osburn, W. S., 120
O'Sullivan, 76
Page, T. H.f 17, 25, 27
Parozzani, A., 154
Parry, E. J., 23, 28, 31, 36, 42, 57, 58,
65, 66, 67, 188, 192, 197, 198
Passy, J., 197
Pasteur, Louis, 142
PatanS, G., 29, 31, 32
Paterno, 148
Paucksch, O., 170
Payen, 78
Peratoner, 148
Piesse, 42
Fillet, L., 177, 195, 196
Pitino, 12
Piver, 184
Pollard, J. S., 170
Pomeranz, C., 16
Poore, Homer D., 148
Pratt, D. S., 156
Pusch, 157
Quercigh, E., 32
Regnauld, 75
Reichardt, 76
Reischauer, 32
Risso, 10
Robertson, T. Brailsford, 91, 97
Rojahn, W., 26, 27, 28
Romeo, G., 17, 22, 38, 59, 65
Rosenberg-Hein, E.r 78
Rother, P. B., 60, 65
Roure, L., 184
Roure-Bertrand fils, 178, 192, 196
Rubke, K., 51
Rueff, W., 114
Sachs, J., von, 10
Sadtler, S. S., 59, 65
Salamon, 54
Satie, C., 174, 196
Scarlata, 148
Scarpa, 32
Scheibler, 76
Schimmel & Co., 14, 16, 18, 19, 24, 25,
26, 27, 29, 30, 32, 33, 36, 39, 40, 41,
42, 43, 44, 45, 46, 50, 53, 54, 57, 59,
60, 61, 63, 64, 65, 174, 177, 178, 179,
180, 181, 182, 183, 193, 195, 196, 197,
198
Schmidt, E., 28
Schweitzer, 66
Seaber, 54
Sell, 25
Semmler, F. W., 15, 179, 196, 197
Sheehan, E. J., 10, 13, 120
Shepard, H. A., 112
Sherman, D. F., 112
Singh, 92
Singleton, J. N. G., 120
Smith, 4, 6
Spica, 154
Spindler O., von, 154
Spinella, Commandatore, 20
Soden H., von, 26, 27, 28, 197
Soldaini, A., 29, 58, 65
Souberian, 15, 25, 42
Stephan, 39
Sucharipa, 92
Swinton, R. S., 27
Tanret, 160
Tarr, 84, 90, 92
Tartar, 84
Taussure, 25
Tempany, H. A., 37
Theulier, E., 28, 183
Thomas, 4, 6
Thorns, H., 16
Tiemann, F., 15, 58, 59, 162, 178, 179,
196, 197
Tilden, W. A., 17, 27, 28, 38
Tinkelpaugh, C. N., 170
Tobler, H., 147
Tollens, 76, 80, 90
Tolman, Richard C., 91
Tschirch, 78, 89, 90
Tucholka, W., 27
Ulpiani, C., 154
Umney, J. C., 27, 31, 192, 196
Vincent, C., 184
Vohl, 38
Volckel, 66
Walbaum, H., 15, 178, 179, 180, 181,
193, 196
Wallach, Otto, 15, 25, 26, 30, 42, 48
Walther, J., 57
206 CITRUS PRODUCTS
Warington, 153 Young, 4, 6
Wehmer, 151, 152
Welch, P. R., 112, 114 Zahorski, 151
Wieeand On ZeiSel> 79( 8°' 8l> 84' 98
an£ ~y 5I Zeitschel, O., 40, 178, 179, 180, 181, 182,
Wiley, H. W., 31 !84r 197
Will, 162 Zelinsky, 49
Will, R. T., 5 Zoller, H. F., 47, 160, 162, 163
Wright, 42 Zurck, Hans, 114
SUBJECT INDEX
Acetone from citrus fruit, 146
Acid value of essential oils: bergamot,
19, 20; Japanese mandarin, 40; lemon,
22; lime leaves, 192; Neroli Bigarade,
178, 181; Neroli Portugal, 183; petit-
grain, 196
Acids in essential oils: lemon, 27; Neroli
Bigarade, 180, 181. See Acid value;
Acetic acid; Citric acid; Palmitic acid
Adulterants of essential oils: bergamot
in Neroli, 182; glyceryl acetate in ber-
gamot, 53; lemon, 36; lemon in petit-
grain, 197; non- volatile esters in berga-
mot, 54; orange in petitgrain, 197;
petitgrain in Neroli, 182; terpinyl
acetate in bergamot, 50; terpinyl ace-
tate in petitgrain, 50; triethyl citrate
in bergamot, 52; turpentine in oil of
lemon, 24, 29, 36
Adulteration of essential oils: bergamot,
17-20; bergamot leaves, 191; Neroli
Bigarade, 182; sweet orange, 43
Alcohols: in oil of bergamot, 15, 16. See
Alcohol content; Citronellol; Farnesol;
Geraniol; Linalool; Myristicol; Nerol;
Nerolidol; Phenyl ethyl alcohol; Ter-
pineol
Alcohol content of oil of leaves of trifo-
liate orange, 192
Aldehydes in essential oils: Jamaican
bitter orange, 45; Jamaican sweet
orange, 41 ; grapefruit, 49; lime leaves,
192; sweet orange, 142. See Citral;
Citronellal; Decylic aldehyde; Furfur-
ol; Nonylic aldehyde; Octylic alde-
hyde
Amido and imino derivatives. See An-
thranilic acid methyl ester; Indol;
Methyl anthranilic acid methyl
ester; Pyrrol
Anthranilic acid methyl ester in essen-
tial oils: Neroli Bigarade, 180, 181;
Neroli Portugal, 183; petitgrain, 197;
trifoliate orange leaves, 192
Aqua naphae, 174, 182
Aurantiamarin, 160
Bergamot. See Oil of bergamot
Bergaptene, in bergamot oil, 14, 16
Bisabolene in oil of lemon, 27
Cadinene in oil of lemon, 27
California lemon supply, i, 2
Camphene in essential oils : bergamot, 1 7 ;
lemon, 25; Neroli Bigarade, 178, 181;
Neroli Portugal, 183; petitgrain, 196;
sweet orange leaves, 192; trifoliate
orange leaves, 192
Citral in essential oils: bergamot, 20;
citron, 46; grapefruit, 49; grapefruit
leaves, 192; Italian bitter orange, 44;
Italian limette oil, 37; lemon, 27;
mandarin, 39; petitgrain citronnier,
198; sweet orange leaves, 192; West
Indian limette, 37
Citral content of oil of lemon, 22, 23; of
terpeneless oil of lemon, 23
Citral determination, 57; Bruylant's
method, 61; Chace's method, 62;
Hiltner's method, 64; Kleber's meth-
od, 64; Parry's method, 58; Romeo's
method, 59; Rother's method, 60;
Sadtler's method, 59; Soldaini and
Berte's method, 58; Walther's method,
57
Citraptene. See Citroptene
Citric acid: analysis in citrate of lime,
153; analysis in juice and factory
liquors, 154; by artificial synthesis,
153; equipment for production of, 7;
by fermentation, 151; identification,
52; manufacture from lemons, 146;
in presence of other acids, 156; tests
for purity of, 157
Citron, candied, no
Citron seed oil, 168
Citronellal in essential oils: bergamot, 20;
in oil of lemon, 26; mandarins, 39;
sweet orange, 42
Citroptene: in oil of citron, 46; in oil of
lemon, 28
Citrus products: condition of fruit for,
7, 8; enumeration of, 6, 7; equipment
for factories, 7; preparation of fruit
for, 7, 8; production in Sicily of, 3;
quality and quantity of, 7, 8
Citrus products industries (flow sheet), 5
Citrus products industry in relation to
fruit industry, 3
Citrus fruits, frost resistance of, 6
Citrus juice production, equipment for, 4
Citrus waste products, 3, 4
207
208
CITRUS PRODUCTS
Climate, effect of, on production of oil of
sweet orange, 43, 44
Color of essential oils: grapefruit, 46;
Italian limette, 37 ; Japanese mandarin,
40; mandarin leaves, 193; Neroli
Portugal, 183; petitgrain, 195; sweet
orange, 41; trifoliate orange leaves,
192; West Indian limette, 36
Cull fruit: definition of, 4; as fertilizer,
172
Cymene in oil of lemons, 26
Decylic aldehyde in oil of Neroli Biga-
rade, 179, 181
Dipentene in essential oils: citron, 46;
grapefruit leaves, 192; Neroli Biga-
rade, 178, 181; petitgrain, 196
Distillation: of oil of citron, 46; oil of
lemon, 29
Distillation test for oil of lemon, 29
Dominica, limes from, in, 112
Essence of Neroli, 174
Esters in essential oils: bergamot, 15, 18;
lemon, 27, 28; non-volatile, as adulter-
ants of oil of bergamot, 54. See Ester
value; Geranyl acetate; Glyceryl ace-
tate; Linalyl acetate; Methyl anthra-
nilic acid methyl ester (methyl anthra-
nilate); Neryl acetate; Phthallic acid
esters; Saponification number; Ter-
pinyl acetate; Trie thy 1 citrate
Ester content of essential oils: bergamot
leaves, 191; sweet orange leaves, 192;
trifoliate orange leaves, 192
Ester value of essential oils: bergamot,
19, 20; grapefruit leaves, 192 ; Japanese
mandarin, 40; lemon, 22; mandarin
leaves, 193; Neroli Bigarade, 177, 178,
181; Neroli Portugal, 183; petitgrain,
196
Farnesol in oil of Neroli Bigarade, 180,
181
Fermentation products, 129
Fermented juice, 129
Fertilizer from cull citrus fruit, 172
Flowers, products from, 174. See Grape-
fruit flowers, Orange flowers
Fruit: canned, 113; cull, denned, 4; as
fertilizer, 172; frozen, see Lemons,
Oranges, etc.
Fruits yielding oil, 14
Fruit juice, condensed, 120; Gore process,
125; Gurber method, 121; Monti
process, 121
Fruit juices: mixed, 120; partially fer-
mented as beverage, 140
Fodder, 164
Furfurol in oil of petitgrain, 196
Geraniol in essential oils: bergamot, 20;
grapefruit, 49 Neroli Bigarade, 179,
181; petitgrain, 197
Geranyl acetate in essential oils: berga-
mot, 20; lemon, 27; Neroli Bigarade,
181
Glucosides, bitter, 158; aurantiamarin,
160; hesperidin, 158; hesperitin, 158;
isohesperidin, 160; naringin, 160
Glyceryl acetate: as adulterant of oil of
bergamot, 53; in oil of bergamot, 19
Grapefruit: bottling juice, 6; citric acid
of, 6 ; frozen juice of, 6 ; peel of, 6 ; pre-
served juice of, 114, 119; marmalade,
of, 6; naringin from, 160; no hesperidin
from, 158, vinegar, 6. See Oil of grape-
fruit
Grapefruit flowers, naringin from, 162
Hesperidin, 158; from lemons, 158; from
mandarins, 158; microchemistry, 159;
preparation of, 159; properties of, 159
Hesperitin, 158
Hydrocarbon content: determination of,
in oil of lemon, 33; relation of, to citral
content, 35. See Terpenes
Indol in oil of Neroli Bigarade, 181
Isohesperidin, 160
Jasmone in oil of Neroli Bigarade, 179,
181
Jelly, formation from pectin, 89
Jelly manufacture, 98; acid in, 102;
aroma in, 100; clarification in, 99;
flavor changes in, 100; sugar in, 101;
yields, 98
Juices, preserved and condensed, 114.
See Lemon juice, Orange juice, etc.
Ketones. See Jasmone, Methylheptenone
Kumquats, glaced, no
Leaves, products from, 190. See Oils
Lemons: hesperidin from, 158; imported
in United States, i, 2
SUBJECT INDEX
209
Lemons, frozen : citric acid yield from, 4 ;
juice of, 6; oil yield from, 6; peel of,
6; pulp of, 6; utility of, 4. See Oil of
lemon
Lemon camphor. See Citroptene
Lemon consumption in the United States,
2
Lemon juice, preserved, 114
Lemon oil, condition of fruit in relation
to, 7
Lemon peel, imports into United States,
112
Lemon production in Sicily, economics
of, 3
Lemon seed oil, 167, 168
Lemon wood, uses for, 198
Limene in oil of bergamot, 17
Limes: hesperidin from, 158; pickled,
imports in United States, 1 1 1 ; pickling
process of, 112. See Oil of limes
Lime juice, preserved, 114
Lime seed oil, 168
Limettin: in Italian limette oil, 37, 38;
in West Indian limette oil, 37
Limonene in essential oils: bergamot, 15,
20; citron, 46 ; grapefruit, 48 ; immature
oranges, 198; Italian bitter orange, 44;
Italian limette, 37, 38; lemon, 26;
petitgrain, 196; sweet orange, 43; tri-
foliate orange leaves, 192
Linalool in essential oils: bergamot, 15,
16; grapefruit, 49; grapefruit leaves,
192; Italian limette, 37, 38; lime flow-
ers, 188; mandarin, 39; Neroli Biga-
rade, 179, 181; Neroli Portugal, 183;
petitgrain, 197; sweet orange leaves,
192; trifoliate orange leaves, 192
Linalyl acetate in essential oils: berga-
mot, 15, 16, 17, 19, 20; Italian limette,
37, 38; lemon, 27; Neroli Bigarade,
179, 181; petitgrain, 195; trifoliate
orange leaves, 192; West Indian
limette, 37
Mandarins, hesperidin from, 158. See
Oil of mandarins
Marmalade: equipment for production
of, 7; "Dundee," 170; manufacture of,
170
Methyl anthranilate in essential oils:
bergamot leaves, 191; lemon, 28;
lime flowers, 188; mandarins, 38, 39;
Neroli Bigarade, 178; sweet orange,
42 ; West Indian limette, 37. See Meth-
yl anthranilic acid methyl ester
Methyl anthranilic acid methyl ester in
essential oils: Japanese mandarin, 40,
41; mandarin, 39, 40; mandarin
leaves, 193
Methyl ester of anthranilic acid in essen-
tial oils. See Methyl anthranilate
Methylheptenone in oil of lemon, 26
Myristicol in oil of sweet orange, 42
Naringin, 160; found only in grapefruit,
162
Nerol in essential oils: bergamot, 16;
Neroli Bigarade, 179; petitgrain, 197
Neroli oil, 174
Nerolidol in oil of Neroli Bigarade,
180, 181
Neryl acetate in oil of Neroli Bigarade,
181
Nonylic aldehydes in oil of lemon, 26
Octylene in essential oils: bergamot, 17;
lemon, 17, 25
Octylic aldehyde in oil of lemon, 26
Odor of essential oils: bergamot due to
linalool acetate, 15; grapefruit, 46;
Italian bitter orange, 44; Italian lim-
ette, 37; Japanese mandarin, 40; man-
darin, 38; Neroli Portugal, 183 ; Papeda
leaves, 192; petitgrain, 195; sweet
orange oil, 41 ; West Indian limette, 37
Oil, methods for obtaining, 10, n
Oil of bergamot, 14; adulteration of, 17-
20; adulterants of, 50, 52, 53, 54; as
adulterant of oil of Neroli, 182; as
adulterant of oil of petitgrain, 191;
change during ripening, 15, 16; chem-
ical composition of, 15; comparison
with Italian limette oil, 37; ester con-
tent of, 18; examination of, 17; evap-
oration residue of, 20; methods of pro-
duction of, 20; properties of, 14; origin
and production of, 14
Oil of bergamot leaves, 191
Oil of bitter orange, 44; Italian, 44;
Jamaican, 44; properties, 44; taste of
compared with oil of sweet orange, 44;
West Indian, 45. See Oil of immature
orange
Oil of bitter orange leaves. See Oil of
petitgrain
Oil of citron: adulteration of, 45; com-
position of, 46; origin of, 45; produc-
210
CITRUS PRODUCTS
tion of, 45; properties of, 46; sediment
in, 46
Oil of Citrus trifoliate, leaves, 192
Oil of grapefruit: composition of, 46;
origin of, 46; properties of, 46
Oil of grapefruit leaves, 192
Oil of Italian limette, 37
Oil of lemon, 21; adulterants, 29, 36;
composition of, 24; change during
ripening, 22; constants of California
and Spanish, 24; constituents soluble
in alcohol, 31; difference in odor be-
tween Palermo and Messina oils, 27;
distillation test for, 29; effects of air,
distillation, light, storage on, 23; ex-
amination of, 28; hydrocarbon deter-
mination in, 33; oxygen content of, 24;
properties of , 2 1 , 24 ; properties of ter-
peneless, 23 ; sediment in, 23 ; terpene-
less, 66; viscosity determination of, 32
Oil of lemon leaves. See Oil of petitgrain
citronnier
Oil of limes, 36 '
Oil of lime leaves, 192
Oil of limette: Italian, 37; West Indian,
36
Oil of mandarins, 38; composition of, 39;
Japanese, 40; Japanese compared with
Italian, 40; origin of, 38; properties
of, 38
Oil of mandarin leaves, 193
Oil of Neroli : history of, 174; oil of petit-
grain as adulterant, 182
Oil of Neroli Bigarade, 176; composition
of, 178; oil of bergamot as adulterant,
182; properties of, 176
Oil of Neroli Portugal, 182; Algerian,
183; composition of, 183; occurrence
of, 1 83; properties of, 182; Spanish, 183
Oil of oranges, immature, 198
Oil of Papeda leaves, 192
Oil of orange flowers: adulterated with
oil of bergamot leaves, 191; from fall
flowers similar to oil from old branches,
175; history of, 174; production of,
174; from spring flowers similar to oil
from young branches, 175; synthetic,
174; use of, 174
Oil of petitgrain, 195; adulteration and
examination of, 197; adulterants, 50,
191; as adulterant of oil of Neroli, 182;
composition of, 196; origin and pro-
duction of, 195; properties of, 195;
source of, 195; taste of, 195
Oil of petitgrain citronnier, 198
Oil of petitgrain Portugal. See Oil of
sweet orange leaves
Oil of sweet orange, 41; change in stor-
age, 41; composition of, 42; examina-
tion of, 43; Jamaican, 44; properties
of, 41; sediment in, 41; source of, 41;
taste of, 41
Oil of sweet orange leaves, 192
Oil of West Indian bitter orange, pro-
duction of, 45
Oil of West Indian limette, 36
Oil from seeds, 167
Oil from the stems, 195, See Oil of petit-
grain
Oil cells: morphology of, 10; position in
plant, 9; shape in relation to quantity
of juice, 10
Oil production: equipment for, 7; ex-
pression by hand, n, 12; expression
by machine, 1 2 ; extraction by displace-
ment, 13; extraction by solvents, 13;
by distillation, 1 1
Oils, esters in: citric acid, 51; glyceryl
acetate, 53; non- volatile, 54; terpinyl
acetate, 50. See Esters in essential
oils
Oil, terpeneless, 66
Oils: formation in leaves, 193; terpene
content of petioles and leaves, 193
Oleoresin of oil of sweet orange, 42
Optical rotation of essential oils: berga-
mot, 14, 16, 17, 19, 20; bergamot
leaves, 191; bitter orange, 44; citron,
46; Dominican sweet orange, 41, 42;
grapefruit, 46; grapefruit leaves, 192;
immature orange, 198; Italian bitter
orange, 44; Italian limette, 37; Jamai-
can bitter orange, 45 ; Jamaican sweet
orange, 41, 44; Japanese mandarin, 40;
lemon, 22; lime flowers, 188; lime
leaves, 192; mandarins, 38, 39; man-
darin leaves, 193; Neroli Bigarade,
177, !78; Neroli Portugal, 182, 183;
Papeda leaves, 192; petitgrain, 195,
196; sweet orange, 41; sweet orange
leaves, 192; trifoliate orange leaves,
192 ; terpeneless bergamot, 67 ; terpene-
less lemon, 23, 67; terpeneless lime, 67;
terpeneless orange, 67; West Indian
limette oil, 37
Oranges: aurantiamarin from, 160; citric
acid from, 6; hesperidin from, 158;
isohesperidin from, 160
Oranges, frozen: flowers of, 6; oil of, 6;
SUBJECT INDEX
211
pectin of, 6; peel of, 6; pulp of, 6;
sugar content of, 6; vinegar from, 6
Orange flowers: oil, 174; perfume, manu-
facture of, 183; perfume, ready made
in plant, 184; perfume, use of, 188;
wax of, 1 86
Orange juice: filtration of, 117; pas-
teurization of, 118; preserved, 114;
yields and composition, 115
Orange marmalade, relation of maturity
of fruit to, 7, 8
Orange peel, imports into United States,
112
Orange seed oil, 167, 168
Orange vinegar, 140; after-treatment,
145; apparatus for, 141; diseases of,
145; domestic method, 141; fermenta-
tion, 140; German method, 143; nature
and origin of, 140; Pasteur method,
142; starters and pure cultures, 141;
rotating barrel method, 144
Orange wine: analysis of, 138; sparkling,
139
Orange wood, uses for, 198
Palmitic acid in oil of Neroli Bigarade,
181
Paper, 164
Paraffin in essential oils: Neroli Biga-
rade, 126, 180, 181, 182; trifoliate
orange, 192
Parapectin, definition of, 78
Pectase, action of, 85
Pectic acid, sources and properties, 82, 88
Pectin, 73; composition of, 80; experi-
mental progress, 73; microchemical
reactions, 77; methods for estimaljon
of, 97 ; physical behavior, 83 ; production
and composition of orange, 8 1 ; pro-
duction and purification, 79, 93 ; uses,
93; sources and properties, 82
Pectin extraction, 93 ; Caldwell's method,
93
Pectose, definition of, 78
Peel, dried, equipment for production of,
7. See Orange peel, Grapefruit peel
Phellandrene in oil of lemon, 26
Phenols in oil of Neroli Bigarade, 180,
181
Phenyl acetonitrile in aqua naphae, 182
Phenyl ethyl alcohol in oil of Neroli
Bigarade, 179
Phthallic acid esters in oil of bergamot,
19
Pinene in essential oils: absent in sweet
orange, 42; bergamot, 17, 20; grape-
fruit, 48; Italian bitter orange, 44;
lemon, 25; Neroli Bigarade, 178, 181;
petitgrain, 196
Pomelo. See Grapefruit
Products from the stems, 195
Protopectin, 78; definition of, 78; sources
and properties, 82
Pseudocumene in oil of lemon, 26
Pulp: as fodder, 164; for paper, 164; uses
for, 113
Pyrrol in essential oils: from immature
oranges, 198; Neroli Bigarade, 181;
petitgrain, 196
Refractive index of essential oils: grape-
fruit, 46; grapefruit leaves, 192; lemon,
22 ; Neroli Bigarade, 178; Neroli Portu-
gal, 183; Papeda leaves, 192; of ter-
peneless lemon oil, 23
Residue, evaporation, of oil of lemon, 22
Rind, use of, no
Rubber manufacture, use of citrus seed
oils in, 1 68
Saponification number of essential oils:
bergamot, 20; mandarin leaves, 193;
Neroli Bigarade, 177; Papeda leaves,
192; petitgrain, 195; petitgrain cit-
ronnier, 198
Seeds, products from, 167
Seed oil, uses for, 168
Sesquiterpenes. See Terpenes
Soap: from citrus seed oil, 168; orange
flower perfume in, 188
Soporific substance from leaves, 190
Solubility of essential oils: bergamot
leaves, 191; Italian bitter orange, 44;
lemon, 22; Japanese mandarin, 40;
mandarin leaves, 193; Neroli Biga-
rade, 176, 178; Neroli Portugal, 183;
petitgrain, 195, 196; terpeneless berga-
mot, 67 ; terpeneless orange, 67
Specific gravity of essential oils: berga-
mot, 14, 16, 17, 18, 19, 20; bergamot
leaves, 191; bitter orange, 44; citron,
46; Dominican sweet orange, 41;
grapefruit, 46; grapefruit leaves, 192;
Italian bitter orange, 44; Italian lim-
ette, 37; Jamaican bitter orange, 45;
212
CITRUS PRODUCTS
Jamaican sweet orange, 41, 44; Japan-
ese mandarin, 40; lemon, 22, 23; lime
flowers, 1 88; lime leaves, 192; mandar-
ins, 38, 39; mandarin leaves, 193;
Neroli Bigarade, 176, 177; Neroli
Portugal, 183; Papeda leaves, 192;
petitgrain, 196, 197; petitgrain citron-
nier, 198; sweet orange, 41; sweet
orange leaves, 192; terpeneless berga-
mot, 67; terpeneless lemon oil, 23, 67;
terpeneless lime oil, 67; terpeneless
orange oil, 67; trifoliate orange leaves,
192; West Indian limette, 37
Stearoptene in oil of sweet orange, 42
Stems, products from, 193. See Petitgrain
Sugar content of fruit in relation to
maturity, 8
Terpene properties of oil of sweet
orange, 42
Terpinene in oil of lemon, 26
Terpineol in essential oils: bergamot, 16;
lemon, 27; Neroli Bigarade, 179, 181;
petitgrain, 197
Terpinyl acetate in essential oils: as
adulterant in bergamot, 50; adulter-
ant in petitgrain, 50; bergamot, 16
Turpentine oil: adulterant of oil of
lemon, 24, 29, 36; detection in oil of
sweet orange, 43
Triethyl citrate as adulterant of oil of
bergamot, 52
Terpenes and sesquiterpenes: formed in
green parts of plants, 176; in oil of
Neroli Portugal, 183. See Bergaptene;
Bisabolene; Cadinene; Camphene; Cit-
roptene; Cymene; Dipentene; Limon-
ene; Phellandrene; Pinene; Pseudo-
cumene; Octylene; Stearoptene; Ter-
pinene
Terpeneless oil of lemon, 66
Vinegar. See Grapefruit, Orange vinegar
West Indies, limes from, 1 1 1
Wine. See Orange wine
Wood: lemon, uses for 198; orange, uses
for, 198
Yield of essential oils: bergamot leaves,
191; grapefruit leaves, 192; immature
oranges, 198; mandarin leaves, 193
FIELD MUSEUM OF NATURAL HISTORY
PUBLICATIONS
BOTANICAL SERIES
August IQ, 1926
VOL. I, Complete in 7 numbers.
VOL. II, Complete in 1 1 numbers.
VOL. Ill, Only 2 numbers so far issued.
VOL. IV, Only 4 numbers so far issued.
VOL. V, Complete in i number.
VOL. VI, Complete in 2 numbers.
(Number 2 in preparation.)