Citation
Recent studies of the chemistry of orange aroma and flavor

Material Information

Title:
Recent studies of the chemistry of orange aroma and flavor
Series Title:
Citrus Station mimeo report
Creator:
Attaway, John A., 1930-
Oberbacher, M. F
Dougherty, Marshall H
Wolford, R. W
Buslig, Bâela S
Florida Citrus Experiment Station
Place of Publication:
Lake Alfred FL
Publisher:
Citrus Experiment Station :
Florida Citrus Commission
Publication Date:
Language:
English
Physical Description:
5 leaves : ill. ; 28 cm.

Subjects

Subjects / Keywords:
Orange juice -- Flavor and odor -- Florida ( lcsh )
Oranges -- Quality -- Florida ( lcsh )
Orange oil -- Quality -- Florida ( lcsh )
City of Lake Alfred ( flego )
Station Lake ( flego )
Orange fruits ( jstor )
Terpenes ( jstor )
Juices ( jstor )
Genre:
government publication (state, provincial, terriorial, dependent) ( marcgt )

Notes

General Note:
Caption title.
General Note:
"450-10/12/67-JAA."
Statement of Responsibility:
J.A. Attaway ... et al..

Record Information

Source Institution:
University of Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
76804410 ( OCLC )

Downloads

This item has the following downloads:


Full Text


Florida Citrus Commission and
Citrus Experiment Station CES 68-8F
Lake Alfred, Florida- 450-10/12/67 JAA


RECENT STUDIES OF THE CHEMISTRY OF ORANGE AROMA AND FLAVOR

A.) The Aroma of Intact Oranges

J. A. Attaway and M. F. Oberbacher
Florida Citrus Commission,
Lake Alfred, Florida


During the past decade many of the components responsible for the aroma and
flavor of citrus juices and oils have been identified. However, no work has been
reported on-the identification of compounds responsible for the aroma of intact
oranges, which have been carefully handled to prevent the release of peel oil.
The fact that citrus fruits are not as highly aromatic as are such deciduous
fruits as apples and pears may explain the lack of information in this area.

The initial experiments of our study were based on the observation that
when boxes of 'Hamlin' oranges picked near the end of the season were stored and
later opened, a very pleasant fruity aroma, totally different from that of peel
oil emanated from these fruit. Apparently during storage, they were giving off
volatile compounds which accumulated in the boxes. Consequently, experiments
designed to trap these compounds in sufficient quantity for identification were
carried out as a first approach.

It was also observed that the handling of freshly picked oranges imparted a
characteristic odor to the hands. This odor differed from that of peel oil, but
was not the same as the aroma which accumulated in the boxes on storage as des-
cribed above. It seemed a reasonable assumption that this odor was produced by
relatively non-volatile compounds present on the cuticle of the fruit. Accord-
ingly, experiments were designed to accumulate the compounds responsible for
this aroma as a second approach.

To study the aroma emanating on storage, 140 'Hamlin' oranges were clipped
carefully to prevent rupture of the flavedo, and the stem ends painted with 5%
aqueous thiourea to retard decay. They were divided equally between 2, 5-gallon
glass jars fitted with inlet and outlet tubes. The jars were placed in a con-
stant temperature room at 700F., fitted with activated charcoal traps, and air
was circulated through the system at a rate of 15-20 ml/min. for 12 days. The
charcoal was then extracted for 3 days with diethyl ether in a Soxlet extractor,
after which most of the ether was carefully removed using a Rinco rotary evap-
orator. The yield was about 1/2 ml of ether solution with a fruity odor remin-
iscent of ethyl butyrate.

The collected aroma volatiles were then subjected to gas liquid chroma-
tographic analysis using two different column phases of opposite polarity,
namely Carbowax 20M and Apiezon L, and to mass spectrometric analysis1. Using


Appreciation is expressed to Dr. G. L. K. Hunter and Mr. Wes Busik of the
Minute Maid Company, Plymouth, Florida, who were most helpful in the determi-
nation and evaluation of mass spectra, and Mr. Bela S. Buslig of this laboratory
for technical assistance.










these combined techniques, it was possible to identify the major components of
this mixture as ethyl acetate, ethanol, an unknown compound of molecular weight
114, ethyl butyrate, limonene, ethyl caproate, and ethyl caprylate. In addition,
a number of the minor peaks were tentatively identified as ethyl format, ethyl
propionate, ethyl isovalerate, ethyl valerate, and ethyl heptoate. These ethyl
esters, particularly ethyl butyrate, are undoubtably responsible for the pleasant
fruity aroma which emanated from the stored fruit.

It was further noted that on prolonged gas chromatographic analysis buty-
lated hydroxy toluene (BHT), eluted from the column. It is not yet known
whether this compound is a true constituent of oranges or is an artifact.

For the study of the less volatile compounds present on the cuticle, 160
'Hamlin' oranges were carefully clipped, washed gently with laboratory deter-
gent to remove dirt and scale insects, dried with cheesecloth, and allowed to
stand for 24 hours to replenish aroma compounds lost because of the washing.
Each fruit was then dipped in a 2-liter quantity of methylene chloride contained
in a large beaker, taking care to prevent scratching of fruit which would re-
lease peel oil. The methylene chloride was then filtered to remove scale or
other particles which had escaped the detergent wash, after which it was concen-
trated to 1 ml. in the rotary evaporator. The resulting extract had a strong
odor of fresh oranges.

Analyses of this extract on a 12 ft. X 1/2 in. Carbowax 20M column showed
that almost all of the volatile components emerged in the sesquiterpene region
of the chromatogram. Accordingly, several collecting runs were made so that
the sesquiterpene cut could be accumulated and analyzed using a 50 ft. X 1/8 in.
column. In this manner it was shown that the predominant compound was valencene,
with lesser amounts of beta-elemene, beta-caryophyllene, farnesene, humulene,
and delta-cadinene. Apparently these sesquiterpene hydrocarbons are the major
contributors to the odor of fresh oranges.























Florida Citrus Commission and
Florida Citrus Experiment Station,
Lake Alfred, Florida. 450-10/12/67-JAA










B.) Oxygenated Terpene and Aldehyde Concentrations in Orange Juices during
the 1966-67 Season

J. A. Attaway, M. H. Dougherty, and R. W. Wolford
Florida Citrus Commission,
Lake Alfred, Florida


The object of this study was to determine the concentrations of oxygenated
terpenes and saturated aldehydes in orange juices at various stages of maturity.
Analytical methods used were the vanillin procedure for estimating oxygenated
terpenes as CIOHI80, and the p-phenylenediamine-H202 procedure for estimating
saturated aliphatic aldehydes as octanal. It was not feasible to used the
o-dianisidine procedure for measuring unsaturated aldehydes or the hydroxamic
acid procedure for measuring esters because the concentrations of these compounds
in fresh juice were too low to be detected.

The juices were prepared for analysis by taking 50 ml aliquots, diluting
them to 100 ml with distilled water, and then distilling a 50 ml sample. For
C10H180 determination, a 5 ml aliquot of this sample was pipetted into 5 ml
distilled water in a 50 ml Erlenmeyer flask, after which 5 ml of 2% wt/vol
vanillin in H2SO4 solution was added with swirling. In approximately 10 minutes
an absorbance reading was made using the Fisher Electrophotometer II with a 585
mu filter. This reading was compared with a standard curve (1, 3, 5, 10, 15, and
20 ppm) to get a CIOHI80 value.

For determination of saturated aldehydes 10 p.l of distillate were pipetted
into a 50 ml Erlenmeyer flask, after which 2 ml of 3% H202 were added. Following
this, 1 ml of 0.25% aqueous p-phenylenediamine solution was added and allowed to
react for exactly 15 min, after which the absorbance was determined at 425 m[I
using the Fisher Electrophotometer II. The reading was compared to a graph pre-
pared with each set of determinations using known solutions of 10, 20, 30 and 50
ppm octanal in 50% ethyl alcohol.

The first varieties studied were 'Parson Brown' and 'Hamlin' oranges beginn-
ing in late September, 1966. After this 'Pineapple' and the 'Valencia' oranges
were used. Samples were taken from raw juice furnished by the Pounds-Solids
program. Results of the analyses are reported in Table: 1.

These data show a relationship between the concentration of saturated alde-
hydes and maturity of the fruit. 'Parson Brown' juice contained only 16 ppm
saturated aldehydes in mid-September, but by mid-October this had risen to 22 ppm
and to 40 ppm by mid-November. 'Hamlin' juice analyzed 14-18 ppm in September
and October, over 30 ppm in November and early December, and as high as 50 ppm
in late December and January. Juice from 'Pineapple' oranges varied from 24-36
ppm near the first of December to as high as 54 ppm in late December. Only 8 ppm
octanal were found in 'Valencia' juice in late February. However, the values rose
rapidly to 31 ppm in early March, over 50 ppm in mid to late March, and as high as
60 ppm in early May.

Examination of the data shows no clear cut trend for oxygenated terpene con-
centrations relative to maturity. The early varieties, 'Hamlin' and 'Parson
Brown', showed minima for ppm ClOHI80 during the middle portion of their respect-
ive seasons; however, 'Pineapple' and 'Valencia' did not.

Florida Citrus Commission and
Florida Citrus Experiment Station,
Lake Alfred, Florida. 450-10/12/67-JAA














Table 1. Saturated aliphatic aldehydes, as octanal, and oxygenated terpenes, as CloH180,
in orange juices extracted from different varieties of fruit during the 1966-67 citrus season.


'Hamlin' oranges


'Parson Brown' oranges


'Valencia' oranges


Picking
date
Sept. 19
23
Oct. 6
13
27
Nov. 3
10
19


Octanal
ppm
16

27
22
24
38
42
40


Cl0H180
ppm
9.0

8.0
7.5
6.0
8.0
9.0
10.5


'Pineapple' oranges
Picking Octanal C10H180
date ppm ppm


Picking
date
Sept. 19
23
Oct. 6
f" 13
27
Nov. 3
10
19
28
Dec. 1
"1 9
15
Sr ". 27
Jan. 3
6
t" 16
If )1.


C10H180
ppm
7.5
7.0
4.5
5.0
3.0
3.0
5.0
2.5
3.0
5.0
5.0
5.5
5.0
4.5
6.5
8.5
6.0


Octanal
ppm


Nov.
Dec.
It



Jan.
"i


Octanal
ppm


4.5
6.0
7.0
7.5
8.0
8.0
7.5


Picking
date
Feb. 27
Mar. 2
13
21
23
26
Apr. 3
12
20
24
May 2
11 17


Cl0HI80
ppm
4.0
6.0
6.0
6.0
7.0
4.0
6.0
6.5
5.5
6.0
6.5
6.5









C.) New Approaches to Understanding the Formation of Orange Flavor
Compounds by the Living Tree

J. A. Attaway and B. S. Buslig
Florida Citrus Commission,
Lake Alfred, Florida


This paper introduces a completely new line of thinking in flavor research.
All previous studies of volatile citrus flavor components have been carried out
using citrus juices, essences, oils, or other static materials which do not per-
mit the study of interconversions between compounds. This new approach, however,
involves the introduction of radioactive compounds into living systems. The
changes which the tagged compounds undergo under various conditions of the system
may then be monitored by tracing the formation of other radioactive compounds from
the original labeled material. Among the labeled compounds to be used are CO2,
acetate-14C, mevalonate-14C, linalool-14C, and other labeled terpene and terpen-
oid type compounds. Among the systems to be studied are excised shoots, excised
shoots containing fruit, small trees with and without fruit, freshly picked
respiring fruit, and others. In addition, stored orange oil will be studied with
the object of determining the mechanism of deterioration.

The substrates 14CO2 and acetate-14C have been widely used to trace the
formation of many classes of compounds. Mevalonate-14C has been used in tracer
studies of steroids. However, neither linalool-14C nor any other labeled ter-
pene has been used in the study of terpene interconversions. Linalool-14C has
been selected for the initial phase of this study because linalool is the most
prominent oxygenated terpene of citrus, and may very well be the precursor of
all the citrus terpenes. Evidence for this has been presented in earlier papers
from this laboratory dealing with seasonal changes in the peel and leaf oil ter-
penes. In a paper1 entitled "The Origin of Citrus Flavor Components III. A
Study of the Percentage Variations in Peel and Leaf Oil Terpenes During One
Season", it was shown that the (+)-limonene concentration of tangerine peel oil
increased as the tangerines matured, and that this increase was in direct pro-
portion to a decrease in concentration of linalool. This is shown graphically
in Fig. 1 below.



100
90 ..

80
(+)-limonene
70 -
60 ,

8 50
40
30 '

20c -n

10- Linalool


Sample collection date
1 Attaway, John A., Arthur P. Pieringer, and Leonard J. Barabas.
Phytochem. 6, 25-32 (1967).










Only a few preliminary experiments using tracers have been made to date.
All of these involve linalool- C and no conclusive data ,have yet been obtained.
The first problem has been getting the linalool into the plant as it is only
very slightly soluble in water and is toxic in large concentrations. Direct
injections of linalool into succulent stems with a syringe was unsuccessful be-
cause the plant tissue around the point of injection was killed preventing
translocation. Several surface-active agents were tried with the hope of find-
ing a non-phytotoxic material capable of solubilizing linalool. To date,
dimethylsulfoxide (DMSO) has produced the best results, and by using a dilute
solution of linalool-14C in 1% aqueous DMSO it has been possible to get 14C into
both leaves and fruit. This technique involved placing both excised stems with
leaves and fruit, and excised stems with leaves only, into the linalool-14C -
DMSO water solutions, and allowing them to take up the solution for several
days while illuminated by a 100 watt light bulb. Analysis of leaf and fruit
oils obtained by steam distilling the leaves and whole fruit separately from the
first few experiments showed that linalool-14C was seemingly recovered unchanged.
Some level radioactivity was found in other compounds but it was not definite
whether this resulted from conversion of the linalool or from impurities origi-
nally found in the linalool.

Two possible explanations were that the linalool-14C is not actually getting
inside the oil glands where transformations would presumably take place; or that
linalool is an end-product of terpene formation rather than an intermediate.
However, in a more recent experiment almost 25% of the total radioactivity was
found to have moved into two other compounds and this work is being continued.

















1 Appreciation is expressed to the USDA Fruit and Vegetable Products
Laboratory, Winter Haven, Fla., for the use of their Packard Liquid Scin-
tillation Counter.








Florida Citrus Commission and
Florida Citrus Experiment Station,
Lake Alfred, Florida. 450-10/12/67-JAA