Group Title: Citrus Station mimeo report - Florida Citrus Experiment Station ; CES 64-8
Title: Volatile flavor and aroma of orange juices
Full Citation
Permanent Link:
 Material Information
Title: Volatile flavor and aroma of orange juices some basic and applied aspects
Series Title: Citrus Station mimeo report
Physical Description: 6 leaves : ; 28 cm.
Language: English
Creator: Wolford, R. W
Attaway, John A., 1930-
Alberding, Gilbert E
Citrus Experiment Station (Lake Alfred, Fla.)
Florida Citrus Commission
Publisher: Florida Citrus Experiment Station :
Florida Citrus Commission
Place of Publication: Lake Alfred FL
Publication Date: 1963
Subject: Orange juice -- Flavor and odor -- Florida   ( lcsh )
Orange juice -- Quality -- Florida   ( lcsh )
Genre: government publication (state, provincial, terriorial, dependent)   ( marcgt )
bibliography   ( marcgt )
non-fiction   ( marcgt )
Bibliography: Includes bibliographical references (leaves 5-6).
Statement of Responsibility: R.W. Wolford, J.A. Attaway and G.E. Alberding.
General Note: Caption title.
General Note: "October 8, 1963."
 Record Information
Bibliographic ID: UF00072421
Volume ID: VID00001
Source Institution: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
Resource Identifier: oclc - 75957087

Full Text

Citrus Station Mimeo Report CES 64-8
October 8, 1963

Volatile Flavor and Aroma of Orange Juices Some Basic and Applied Aspects

R. W. Wolford, J. A. Attaway, and G. E. Alberding
Florida Citrus Commission
Lake Alfred, Florida

Research concerning the volatile organic compounds responsible for the flavor
and aroma of orange juice was met with some unavoidable difficulties the past sea-
son because of the freeze in December, 1962. Curtailment of certain phases of the
program definitely restricted progress. Therefore, it appears timely to present a
review of past work in these investigations, thus providing a summarization of re-
sults and findings to better define the present status of the research.

The question might be asked as to why such a basic approach to flavor in-
vestigation of citrus juices was initiated. While citrus processors are placing
continued emphasis on maintaining good flavor in Florida processed citrus products,
wider ranges of fruit varieties are being employed, the yield of juice per box of
fruit has been increasing, and the employment of methods for improving the physical
stability of these products, to name a few, have created problems. Processing
techniques favorable to one of these objectives are sometimes unfavorable to
another and as a consequence flavor maintenance is a significant problem.

The evaluation of maturity of citrus fruits and the determination of the
general quality of processed products are based to a large extent on standard
chemical and physical methods. However, the flavor evaluation of such products
is still of an empirical nature. Not-withstanding the excellent work of others
in the citrus field on juice and oil constituents, mostly in California, there
remains a continuing need for a better knowledge of the interrelation between
various volatile flavor components and the natural flavor of the juice. From the
standpoint of an overall characterization of fresh orange juice and its products,
canned juice or frozen concentrate, we must not only have qualitative and quanti-
tative information about chemical constituents, but also the knowledge that cer-
tain volatile flavor components are outstanding in influencing the particular
flavor which we recognize as most acceptable. Moreover, if such knowledge can be
obtained, we will be much closer to the development of physical and chemical
methods for the actual determination of flavor. Subsequently, a correlation be-
tween flavor constituents and the accepted maturity standards, which is considered
to be a basic need, may be attainable.

The beginnings of the citrus flavor research project coincided with the
development and application of gas-liquid chromatography for analysis of highly
complex mixtures. Gas chromatography provided a quick and efficient method for
separating the flavor producing compounds under relatively mild conditions. Also
it helped in part to overcome one of the principal obstacles to the successful
completion of the work, namely, the detection and separation of materials pre-
sent in only minute quantities in juices and in recovered juice essences. Whereas,

Florida Citrus Commission and
Florida Citrus Experiment Station,
Lake Alfred, Florida.
400-10/8/63 RWW

investigators working with orange oils could start with several pounds of material
and ultimately analyze for the most minor components, using classical methods of
fractionation, only a few drops of extract from juice or recovered essence were
available for entire analyses.

The initial assault on the problem involved a two-sided approach. A complete
gas chromatographic analysis by programmed temperature gas chromatography (PTGC)
of extracts of recovered orange juice essences was made employing two separate
column liquid phases in order to secure as much information as possible via physi-
cal means, while simultaneously developing highly sensitive chemical tests to use
in confirming the identities of compounds producing gas chromatographic peaks.

The first chemical confirmation tests used involved the formation of 2,4-
dinitrophenylhydrazine derivatives of aldehydes and ketones. Based on molecular
weight determinations, infrared spectra, and paper chromatographic properties of
the carbonyl derivatives that group of compounds were largely identified. Most of
this class of compounds reacted readily with the DNP reagent under the conditions
of the experiment; however, gas chromatographic studies in conjunction with sub-
tractive analytical procedures have further shown that acetone, n-nonanal, n-
decanal, and citronellal are also present but presumably in lesser quantity. The
initial identification of the chief hexenal present was cis-2-hexenal-l, however,
subsequent research has shown it is actually trans-2-hexenal-l. Reference to
Table 1 shows the carbonyl constituents which have been identified.

Classes of compounds other than aldehydes and ketones were also studied in
the initial gas chromatographic work. Among the terpenes indicated were the
following: d-limonene, beta-myrcene, terpinolene, alpha-terpinene, gamma-ter-
pinene, alpha-pinene, p-cymene, and delta-3-carene. Among the alcohols were the
following: methanol, ethanol, butanol, pentanol, hexanol, 3-hexen-l-ol, nonanol,
octanol, linalool, citronellol, decanol, alpha-terpineol, geraniol, nerol, and
carveol, while among the esters it appeared that ethyl-n-caprylate, terpinyl
acetate, and methyl-n-methyl anthranilate might be present.

From this point the research divided quite naturally into two categories;
that devoted to the positive identification of the individual flavor producing
compounds, and that devoted to improved resolution and higher sensitivity analyses
using gas chromatography. A third category involved the application of results
obtained from the previous two categories to the analysis of the flavor and aroma
of different varieties of oranges.

The identification of individual components involved the development of new
chemical techniques accompanied by experiments designed to simplify the component
mixtures so that physical methods of analysis, such as infrared spectroscopy could
be used with accuracy. It was possible to positively identify ethyl alcohol,
ethyl butyrate, d-limonene, and linalool by condensing the effluent from the gas
chromatograph and determining the infrared spectrum. However, these were the
four major components and were easily separated by PTGC and collected as relatively
pure materials, whereas, the condensate from other peaks was shown by infrared to
be mixtures of more than one component.

Florida Citrus Commission and
Florida Citrus Experiment Station,
Lake Alfred, Florida.
o00-10/8/63 RWW

The first chemical technique to be developed was one devoted to the identifi-
cation of alcohols through derivatives. Older methods involving xanthogenates and
3,5-dinitrobenzoates were tried and rejected. A new method based on the formation,
followed by paper chromatographic and infrared spectral analysis, of urethan de-
rivatives was evolved.

A study of the paper chromatographic properties of seven different substituted
urethans of aliphatic alcohols showed that m-nitrophenylurethans were best suited
for analysis of short chain alcohols, and o-nitrophenyl- and p-phenylazophenyl-
urethans were best for the longer chain alcohols. However, improvement in the
procedure was necessary to obtain the alcohols free of other classes of components.
Using columns of either activated alumina or silicic acid, the alcohols were sep-
arated as a class from the other components. The procedure was further refined by
using carbonyl-free extracts prepared by forming bisulfite addition products of
carbonyls or Girard T reagent adducts.

Using the above procedures, alcohol identifications were made as shown in
Table 1. Strong evidence for the presence of methyl heptenol, nerol, geraniol,
and carveol was also presented.

Following the successful identification of alcohols, attention was shifted to
the free organic acids. Early attempts to identify this class of components uti-
lized paper chromatography alone. Unfortunately, this was not sufficient, as only
the acids present in greatest quantity could be located on the chromatogram and
even then their identification was not conclusive. Consequently, a procedure was
sought which would free the acids from the other essence components so that they
could be analyzed by gas chromatography. The finally accepted method employed re-
peated evaporations of essence solutions made slightly basic to fix the organic
acids...Proper adjustment of pH was made to avoid hydrolysis of the esters which
might produce misleading results. The acids identified are shown in Table 1. In
addition, evidence for the identification of isovaleric, valeric, isocaproic, and
caproic acids was also reported.

The most recent identification studies have been devoted to the positive con-
firmation of some tentatively identified terpenes in recovered essences. Column
chromatography was used to separate the nonpolar terpenes from other constituents.
Separation by PTGC permitted collection of individual peaks in cold traps for
infrared analysis. Positive identifications were obtained for alpha-pinene,
myrcene, alpha-terpinene, and gamma-terpinene as constituents of recovered orange
essences. Also, a small peak not sufficient for collection was studied using en-
richment techniques and retention data on two column stationary phases and was
found to be beta-pinene.

Following development and installation of an essence recovery system at the
Citrus Experiment Station studies were begun during the 1961-62 season on the
comparison of volatile flavor constituents between the juices of three established
varieties of oranges, namely, Hamlin, Pineapple, and Valencia. Those comparisons
were based on the identifications outlined above and summarized in Table 1. The

Florida Citrus Commission and
Florida Citrus Experiment Station,
Lake Alfred, Florida.
o00-10/8/63 RWW

Table 1
Volatile Flavor Components in Recovered Orange Essence
(Confirmed Identifications Unless Noted)

Aldehydes and Ketones

Methyl heptenol*
2-N onanol-*


Terpene Hydrocarbons

Ethyl Butyrate

Strong tentative identification ,acks confirmation.

Florida Citrus Commission and
Florida Citrus Experiment Station,
Lake Alfred, Florida.
00-10/8/63 RWW

experiments were designed and aimed toward extending the knowledge of flavor
composition in known commercial varieties of Florida oranges. Some specific
aims in the study were; (1) to examine compositional differences in control
juices simulating commercially extracted juice, peel oil-free prepared juice, and
peel oil of the three varieties; (2) to determine the contribution to the flavor
by volatile components inherently present in the juice and/or peel oil; (3) to
define any qualitative or relative quantitative changes in flavor components in
the three varieties with advancing maturity; and (4) to compare analyses of the
fresh aroma of. the juice of each variety.

Preliminary investigations revealed no significant qualitative differences in
gross analyses conducted on either the thermal conductivity or ionization detection
system using PTGC. However, it was indicated that relative quantitative differences
were responsible for the flavor differences noted among varieties. Similar results
were obtained on direct injection of vapors from a sealed headspace over orange

It was possible to show the development of volatile flavor components with
advancing maturity in each of the varieties although any correlation was difficult
until the trend could be observed in successive seasons. Unfortunately, the plans
which had been established for rechecking the data obtained, using the same three
varieties, during the 1962-63 season had to be altered because of the freeze.
However, a three stage maturity study using East Coast Valencia oranges was made.
The results obtained with advancing maturity compared very well with the findings
during-the 1961-62 season on Valencia oranges from the Citrus Experiment Station

Throughout the entire course of investigations outlined above continued pro-
gress has been made in improved resolution and sensitivity in the gas chromato-
graphic analyses. Using dual column programmed temperature gas chromatography
with thermal conductivity detection upward of 56 resolved and indicated component
peaks, each representing one or more components, have been observed from sepa-
rations of organic extracts of recovered orange essences. Using the more sensi-
tive flame ionization detection system and similar column stationary phases as
many as 70 component peaks have been observed on those chromatograms. These
figures do not imply that all of the chromatographic peaks have been identified;
however, to date 40 components, shown in Table 1, have received positive identi-
fication while 16 additional components have received strong tentative identifi-
cation based on peak coincidence on at least two column stationary phases. As
most of the major volatile components have now been identified, future emphasis
will be shifted to the quantitative evaluation of the volatile flavor components
of orange juice.

Continued success has been achieved in direct injection of vapor samples taken
from controlled headspace above single-strength or reconstituted orange juice,
This procedure has resulted in the resolution of some hO-50 components from a 4-5
ml. sample of vapor. These analyses were accomplished using dual columns 1/8" X 6'
(Apiezon L, 8% w/w on 80-100 mesh HMDS treated Chromosorb W). The range of compon-
ents detected went beyond linalool which has a boiling range of 198-2000C. The
headspace samples were taken at approximately 40C.

Florida Citrus Commission and
Florida Citrus Experiment Station,
Lake Alfred, Florida.
00-10/8/63 RWW

Aside from direct aroma analyses injections of 0.5 pl. (0.0005 ml.) of aqueous
essences and 0.7 to 1.0 pl-. of orange juice have been made. Satisfactory chromato-
grams were obtained.

To permit a direct comparison between the volatile components recovered from
juice with those in fresh juice required that direct extraction of orange juice be
accomplished. Through manipulation of solvent combinations it was found that the
azeotropic mixture of methylene chloride and isopentane (30:70 v/v) was of reason-
able polarity for the extraction. Its specific gravity, lower than the juice, also
facilitated phase separation. Centrifugation of the solvent mixture was then
accomplished giving a clear top solvent extract layer. Following concentration,
the extracts were directly analyzable by gas chromatography. The addition of this
technique to these studies will be of considerable value.

With the instrumentation we now have at our disposal, including the recently
acquired Bendix Time-of-Flight Mass Spectrometer permitting coupled gas chromato-
graphy and mass spectrometry, and the analytical methods which have been developed
along with an efficient volatile component recovery system, the ultimate character-
ization of the flavor of citrus juices and processed products is much nearer to
becoming a reality. When we can characterize the flavor of orange juice and
correlate the chemical and physical measurements with the organoleptic testing,
objective flavor maintenance will be possible,

A bibliography has been included as part of this paper for your convenience.
All publications listed are directly related to various phases of the research
outlined above.


The suggestions and assistance of Mr. C. D. Atkins in the recovery of the
orange juice volatiles, using pilot plant equipment, was very much appreciated.

The authors wish to thank Libby, McNeill, and Libby for supplying samples of
orange essences used in the.flavor research investigations.


1. Attaway, John A., Richard W. Wolford, and George J. Edwards. Isolation and
Identification of Some Volatile Carbonyl Components from Orange Essence.
Agr. and Food Chemistry, 10, 102-104, (1962). Fla. Agr. Exp. Sta. Jour.
Ser. No. 1236.

2. Wolford, Richard W., John A. Attaway. The Application of Gas Chromatography
to the Analysis of Flavor Components of Citrus Juices. Gas Chromatography -
Third International Symposium, Brenner, Callen, and Weiss, 1962, Academic
Press, Inc. Fla. Agr. Exp. Sta. Jour, Ser. No. 1282.

Florida Citrus Commission and
Florida Citrus Experiment Station,
Lake Alfred, Florida.
400-10/8/63 RWW


3. Wolford, R. W., G. E. Alberding, and J. A. Attaway. Analysis of Recovered
Natural Orange Essence by Gas Chromatography. Agr. and Food Chemistry,
10, 297-301, (1962). Fla. Agr. Exp. Sta. Jour. Ser. No. 1307.

4. Attaway, John A., Richard W. Wolford, Gilbert E. Alberding, and George J.
Edwards. Paper Chromatographic Analysis of Urethan Derivatives of
Saturated Aliphatic Alcohols. Analytical Chemistry, 34, 671, (1962).
Fla. Agr. Exp. Sta. Jour. Ser. No. 1413.

5. Attaway, John A., Richard W. Wolford, Gilbert E. Alberding, and George J.
Edwards. Paper Chromatographic Analysis of Terpene Alcohols as Their
o-Nitrophenyl- and p-Phenylazophenylurethan Derivatives. Analytical
Chemistry, 35, 234-236, (1963). Fla. Agr. Exp. Sta. Jour. Ser. No. 1502.

6. Attaway, John A., Richard W. Wolford, and Gilbert E. Alberding. Identifi-
cation of Alcohols and Volatile Organic Acids from Natural Orange Essence.
Agr. Exp. Sta. Jour. Ser. No. 1644. Presented to the Division of
Agricultural and Food Chemistry at the 144th ACS Meeting, Los Angeles,
Calif., Mar. 31 -April 5, 1963.

7. Wolford, R. W., J. A. Attaway, G. E. Alberding, and C. D. Atkins. Analysis
of the Flavor and Aroma Constituents of Florida Orange Juices by Gas
Chromatography. J. Food Sci., 28, 320-328, (1963).

8. Wolford, R. W. Preliminary Studies of Volatile Flavor Components in Citrus
Juices Using Gas Chromatography. Citrus Station Mimeo Report 60-1.
September, 1959.

9. Wolford, Richard W. and John A. Attaway. The Recovery and Analytical
Separation of Volatile Flavor Components of Valencia Orange Gas
Chromatography. Citrus Station Mimeo Series 61-6. September, 1960.

10. Attaway, John A. and Richard W. Wolford. Identification of Chromatograph-
ically Separated Flavor Components. Citrus Station Mimeo Series 61-6.
September, 1960.

11. Wolford, R. W., J. A. Attaway, and G. E. Alberding. I. Recent Studies on the
Recovery, Separation, and Analysis of Volatile Flavor Components in Orange
Juice. Citrus Station Mimeo Series 62-4, September, 1961.

12. Alberding, G. E., R. W. Wolford, and J. A. Attaway. II. Some Techniques
Employed for Subtractive Analysis of Complex Mixtures of Flavor Components.
Citrus Station Mimeo Series 62-4, September, 1961.

13. Attaway, J. A., R. W. Wolford, and G. E. Alberding. III. Further Progress in
the Chemical Identification of Volatile Flavor Components of Orange Juice.
Citrus Station Mimeo Series 62-4, September, 1961.

14. Wolford, Richard W., Gilbert E. Alberding, and John A. Attaway. I. Present
Status on Analysis of Flavor and Aroma of Orange Juices by Gas Chromato-
graphy. Citrus Station Mimeo Series 63-1, October, 1962.

15, Attaway, John.A. Attaway, Richard W. Wolford, and Gilbert E. Alberding.
II. The Identification of Some Alcohols and Acids in Orange Essences.
Citrus Station Mimeo Series 63-1, October, 1962.

University of Florida Home Page
© 2004 - 2010 University of Florida George A. Smathers Libraries.
All rights reserved.

Acceptable Use, Copyright, and Disclaimer Statement
Last updated October 10, 2010 - - mvs