Terpenes in the
Aroma
of
Grapes and Wines: A Review
J. MARAIS
Viticultural
and
Oenological Research Institute, Private Bag X5026,
7600
Stellenbosch, Republic
of
South
Africa.
The
author
would like
to
express his
thanks
and
appreciation to Dr. P.
1.
Williams
of
the Australian Wine Research Institute, Glen
Osmond,
South
Australia for critically reading the
manuscript.
Submitted
for
publication: August 1983
Accepted for publication: October 1983
This
paper
reviews some aspects
of
terpenes
and
terpene derivatives in grapes
and
wines, such as the identification
of
these
compounds,
their use in the identification
of
and
differentiation between cultivars, the effect
of
grape maturity
and
wine-making techniques
on
terpene concentrations
and
wine quality, as well as
transformations
of
terpenes in
grapes
and
wine.
Wine quality
is
determined by a complex balance
of
all the wine
aroma
components as perceived by sensory
evaluation. Although several sensory parameters may
play a role in the acceptance
and
enjoyment
of
wine, the
flavour
of
wine
is
possibly the most important factor.
Flavour
is
the result
of
the interaction between certain
chemical compounds in wine
and
the senses
of
smell and
taste
of
the consumer. Approximately 550 volatile com-
ponents have been identified in grapes and wine, and
many
of
these components contribute to the
aroma
(Schreier, 1979). A large number
of
components appear
in micro concentrations in wine,
and
in many cases these
have the
most· significant effect
on
wine quality. With
respect
to
their origin, wine
aroma
components may be
classified into four groups
(Schreier, 1979).
- Originating from the grapes.
-
Produced during the crushing
of
the grapes by the
action
of
certain enzymes.
-
Produced during fermentation.
-
Produced during maturation
of
wine.
A typical cultivar bouquet in wine can be attributed
to
the bouquet
of
the corresponding grape cultivar and
is
caused by:
- Compounds typical
of
the grape cultivar which are
transferred from the grape
to
the wine without being
affected by the fermentation process.
- Compounds equally typical
of
the grape cultivar
which are formed from a precursor(s) during
fermen-
tation (Condonnier & Bayonove, 1978).
GENERAL
The wine
aroma
components responsible for the
characteristic
aroma
of
muscat and
aroma
related
wines, are mainly derived from the grape and are known
as terpene compounds. True muscat cultivars include
Muscat
d'
Alexandrie, Morio Muscat, and Muscat
blanc, whereas aroma related cultivars include Weisser
Riesling, Bukettraube, Gewiirztraminer,
FeriJ.ao Pires
and Scheurebe. Terpene compounds may also contri-
bute
to
the aromas
of
other non-muscat cultivars. Near-
ly
50
terpene compounds in grapes and wine are known
at
this stage,
of
which
46
were identified in grapes and
30
in wines (Table
1).
They include only monoterpenes
and sesquiterpenes.
A number
of
prominent
aroma
components
of
grapes
and wine, which are
of
terpenic origin, are also
of
great
importance, Some are listed in Table 2.
The most prominent terpene compounds occurring
generally
and
in high concentrations in muscat and
aroma
related grapes and wines are linalool, geraniol,
nerol, oc.-terpineol
and
hotrienol.
Of
equal importance
is
the great number
of
terpene compounds which occur
in micro concentrations but which have significant
effects on aroma. Different complex combinations
of
terpene compounds are responsible for the characteris-
tic aromas
of
different muscat and
aroma
related
cultivars. These subtle differences in terpene profiles
may also occur within the same cultivar originating
from different areas. Typical aroma descriptions
of
some important terpenes are floral, rose-like (geraniol,
nerol, rose oxides), coriander (linalool), camphoraceous
(linalool oxides), green (nerol oxide) and herbaceous
(Meilgaard, 1975;
Simpson, 1979 b).
Several studies have been undertaken regarding the
chemistry
of
terpenes, their synthesis, systematic classi-
fications, etc. (Pinder, 1960; Klouwen & Ter Heide,
1962; Ter Heide, 1968; Ter Heide, 1976). Devon
& Scott
(1972) presented a classification
of
4 000 components
of
terpenic nature.
Of
these components, 400 are mono-
terpenes and 1 000 sesquiterpenes. Monoterpenes and
sesquiterpenes are characteristic components
of
the
essential oils produced by plants. Monoterpenes
(C
10
)
and sesquiterpenes (C
15
) are the lower molecular weight
representatives
of
the terpenoid components and com-
prise two and three isoprene units, respectively. A
general representation
of
the biosynthesis
of
mono-
terpenes and sesquiterpenes,
as
well
as
some enzymatic
transformations
such as oxidation, reduction
and
dehydration, has been summarised by Croteau (1975).
Terpene concentrations in grapes and wine would
obviously depend on various factors, such
as
cultivar,
region and wine-making techniques.
Since the majority
of
the terpene compounds occur in micro concentra-
tions in grapes and wines, their quantification
is
quite
difficult.
Consequently,
their
concentrations
are
normally expressed as relative amounts. Examples
of
absolute terpene concentrations in muscat and aroma
related grapes and wines, as well
as
some aroma
threshold values, are shown
in
Table
3.
S.
Afr.
J.
Enol. Vitic., Vol. 4. No. 2. 1983
49
50
Review: Terpenes in the aroma
of
grapes
and
wines
TABLE I
2
~
~
5:0•
R
t"
r
~·
~·
f"
Monoterpenes and sesquiterpenes
in
grapes and wines.
,.C.ryophyU.i>e
~o
p
p
References to these terpenes are available in Supplements 5 (1980, wine)
and
6 (1981, grape) to the
Terprn•
3,7-0.me1hykt01a-l,?-dir1>-3,6""iol
J,7-D1methyloc1-J ...... J.7-diol
cio·S-Hydroxy-2.6,Mrimetbyl-l..,lnyl-
ie1rahydropynn(cis-l'yranllialooloxidol
traru-5-Hydn>•Y·l,6,C.-lrim<thyl-
2-Yinyl-le!rahydropyru (tn,..._
l'ynnlinlloolo•i<kl
Fourth Edition
of
"Volatile Compounds in
Food"
(Van Straten, De Beauveser & Visscher, 1977), except for:
XI
Rapp & Knipser, 1980.
X indicates presence in either grapes
or
wines.
S. Afr.
J.
Enol. Vitic., Vol. 4. No. 2. 1983
~
2
~
~
~
"···f"i_ ~
)l..ox=
Review: Terpenes in the aroma
of
grapes and wines
TABLE 2
51
Some terpene derivatives in grapes and wines.
Ttrprnold
1.l,6-TrlmethyH.2-dihr'-,n•phth.ol•""
~o~
f"/
t"~
~),
Grapoo
c:t::>,
~
References to these terpene derivatives are available in Supplements 5 (1980, wine) and 6 (1981, grape)
to
the
Fourth Edition
of
"Volatile Compounds in
Food"
(Van Straten, De Beauveser & Visscher, 1977), except for:
XI
Schreier & Drawert,
1974
b.
X2 Schreier, Drawert & Junker, 1976 b.
X3
Rapp & Knipser, 1980.
X indicates presence in either grapes
or
wines.
Identification and role of terpene compounds in grapes,
wine and grape leaves: Considerable research has been
done in respect
of
the identification
"-and
contribution
of
terpene compounds to the muscat aroma
of
muscat
grapes and wines. Austerweil (1946) was the first to
suggest that the aroma
of
muscat grapes can be ascribed
to terpene compounds, especially linalool and its deriva-
tives. Cordonnier
(1956)
investigated the aroma
of
muscat wines and tentatively identified linalool,
geraniol,
<X
-terpineol and limonene by means
of
thin
layer chromatography.
Webb
& Kepner (1957), Stevens et
al.
(1966), Webb,
Kepner
& Maggiora
(1966)
and
Wenzel & De Vries
(1968) undertook further investigations and confirmed
the definite contribution
of
linalool and geraniol to the
overall aroma
of
muscat grapes. Research undertaken in
this regard till approximately the end
of
the sixties,
mainly concentrated on Muscat
d'
Alexandrie grapes
and the effects
of
mainly linalool and geraniol on
muscat aroma. Research workers differed in respect
of
the contribution
of
these terpenes to muscat aroma, and
a clear relationship between the concentrations
of
these
compounds and muscat aroma was not always demon-
strable.
Technological development in gas chromatography
and mass spectrometry accelerated identification
of
terpene compounds and the establishment
of
the effect
of
these compounds on the aroma
of
muscat and aroma
related grapes and wines (Usseglio-Tomasset,
1966;
Usseglio-Tomasset, Astegniano & Matta,
1966;
Usseglio-
Tomasset,
1969;
Prillinger & Madner,
1970;
Terrier &
Boidron,
1972
a,b; Terrier, Boidron & Ribereau-Gayon,
1972
a,b; Schreier & Drawert,
1974
a,b; Ribereau-
Gayon, Boidron
& Terrier,
1975;
Bayonove, Richard &
Cordonnier,
1976;
Schreier, Drawert & Junker,
1977;
Usseglio-Tomasset &
Di
Stefano,
1979).
Recently several new terpene compounds have been
identified in grape and wine aromas. In general, there
are indications that more terpenes other than linalool
and geraniol contribute to muscat and related aromas.
Available literature since
1974
indicate that the follow-
ing new terpenes had been identified.
Schreier, Drawert
& Junker (1974) identified 3,7 - dimethylocta - 1,5,7 -
trien - 3 - ol (hotrienol) in grape and wine aromas
of
the
cultivars Weisser Riesling, Gewiirztraminer, Ruliinder,
Muller Thurgau,
Scheurebe, Optima and Rieslaner.
Rapp
& Knipser (1979) identified
3,
7 - dimethyl - octa -
1,5 - dien -
3,
7 - diol in the grapes and wines
of
the
cultivars
Scheurebe, Weisser Riesling and Forta, while
Rapp, Knipser
& Engel (1980) identified 3,7 -
dimethylocta -
1,
7 - dien - 3 ,6 - diol in the grapes and
wines
of
the cultivars Blauer Muskateller, Schonburger,
Morio Muscat and Sieger. Williams, Strauss & Wilson
(1980a) identified 3, 7 - dimethyloct - 1 - en - 3,6, 7 - trio!,
3,7 - dimethyloct - 1 - en - 3,7 - diol, 3,7 - dimethylocta -
1,7 - dien -
3,6-
diol and 3,7 - dimethylocta -
1,5
- dien -
3,7 - diol in the grapes and wines
of
Muscat
d'
Alexan-
drie. The same group
(1980b) identified 2,6,6 - trime-
thyl - 2 - vinyltetrahydropyran,
cis
- and trans - 5 -
isopropenyl - 2 - methyl - 2 - vinyltetrahydrofuran, 2,2 -
dimethyl - 5 -
(1
- methylpropenyl) tetrahydrofuran,
myrcenol and cis - and trans - ocimenol in heated grape
juice
of
muscat cultivars.
Terrier
et
al.
(1972 b) and Ribereau-Gayon et
al.
(1975) determined the aroma threshold values
of
some
individual terpene compounds as
well
as terpene mix-
tures. The authors found linalool and geraniol to be the
most important compounds because
of
the high concen-
trations in which they normally occur in muscat grapes
and wines, as well as their low threshold values.
It
was
found that terpenes interact to such an extent that one
component can increase the aroma
of
another and that a
mixture could become more aromatic than the most
aromatic single component which belongs to that mix-
ture. For example, the aroma threshold value
of
linalool
was found to be
100
µgit,
while a mixture which
S.
Afr.
J.
Enol. Vitic., Vol. 4. No. 2. 1983
52
Review: Terpenes
in
the aroma
of
grapes
and
wines
TABLE 3
Aroma
threshold values
and
concentrations*
of
some terpenes in muscat
and
aroma
related grapes (G) and wines (W).
Terpene
Aroma
Reference
threshold
( µ..g/
/)
Citronellol
Geraniol
130
Hotrienol
110
Linalool
100
Nero!
400-500
ex
-Terpineol
400-500
3,7 - Dimethylocta - 1,5 - dien -
3,7 - diol
<._
3,7 - Dimethylocta - 1,7 - dien - 3,6 - diol
Nero! oxide
100
Rose oxides
~is
-
Furan
linalool oxide
> 6000
trans -
Furan
linalool oxide > 6000
cis -
Pyran
linalool oxide
3000-5000
trans -
Pyran
linalool oxide 3000-5000
2,6,6 - Trimethyl - 2 - vinyl - tetrahydropyran
p - Cymene
Limonene
Mvrcene
~
Terpinene
*
The
concentration ranges represent the sum
of
different cultivars analysed.
(
+)
= Traces.
4
4
Occur- Concentration (
µ..gl I)
rence
min.
max.
w
5
G
89
1059
G 2
240
G 34 237
w
(+)
69
w
14
187
G
(+)
120
G
7
77
w
3 237
w
3
I
17
G
62
1506
G
II
77
G
167
823
w 6 375
w
(+)
29
w 8
93
G
II
447
G
(+)
43
G
15
57
w
(+)
36
w
3
43
G
19
145
G
(+)
14
G 3
23
w 5 399
w 7 52
G
125
1497
w
24
I
174
G
18
228
w
(+)
86
w I
17
w
(+)
19
G
IO
247
G
(+)
20
w 2 258
w
5
21
G
21
400
G
(+)
15
w
(+)
156
w 2
9
G
10
232
G 3
73
G 20 66
w
(+)
74
w I
7
G
20
349
G 5
89
G
67
271
w
16
58
w
3 64
w
5
35
w
3
30
G
(+)
9
w 6
35
I.
Ribereau-Gayon, Boidron & Terrier,
1975
(Muscat d'Alexandrie, Muscat blanc, Saint Vallier, Italia
and
Muscat Hambourg).
2.
Schreier, Drawert &
Junker,
1976 a (Rulander, Gewiirztraminer, Scheurebe
and
Weisser Riesling).
Reference
7
I
2
6
3
7
2
6
3
7
1
2
6
3
5
7
I
2
6
3
7
I
2
6
3
7
6
7
6
3
7
3
1
2
3
7
I
2
3
7
1
2
6
3
7
I
2
6
7
5
5
5
2
5
3.
Schreier, Drawer! &
Junker,
1977
(MUiler
Thurgau,
Weisser Riesling, Morio Muscat, Gewiirztraminer, Rulander
and
Scheurebe).
4.
Simpson, 1979.
5.
Noble, Flath & Forrey, 1980 (Weisser Riesling).
6.
Di
Stefano,
1981
(A series
of
white Muscat grapes from Piemonte).
7. Versini,
lnama
& Sartori,
1981
(Weisser Riesling).
amongst others included linalool presented a threshold
value
of
91
µ..git (Ribereau-Gayon et al., 1975).
Terpene compounds were also found in minor quanti-
ties in some non-muscat cultivars
(Stevens, Bomben &
McFadden, 1967; Prillinger & Madner, 1970). Bayonove
& Cordonnier (1971) found linalool, cx-terpineol,
citral, citronellol, nerol
and
geraniol to occur in I 0 to
I
00 times larger concentrations in muscat
than
in the
non-muscat grape cultivars, Auxerrois, Weisser Riesling
and
Chasselas blanc but in approximately similar relative
S.
Afr.
J.
Enol. Vitic., Vol. 4. No. 2. 1983
Review: Terpenes in the aroma
of
grapes and wines
53
ratios. Theauthorsalsofoundthattheadditionoftheseter- and aroma related cultivars (Hardy, 1970; Terrier et
al.
pene compounds to a neutral grape juice medium in the 1972 b; Cordonnier, 1974; Rapp et
al.
1978). Bayonove
same ratios in which they had been found in muscat culti- & Cordonnier
(1970 a,b) established that in the case
of
varspresentedacharacterwhichcorrespondedwithmuscat several muscat cultivars linalool
is
absent in the unripe
aroma, but lacked balance. Consequently, the terpene grapes, appears at the beginning
of
ripening at the same
compounds mentioned above
do
contribute to muscat time as the muscat aroma, increases in concentration
aroma, but do not represent the complete aroma spectrum. until maturity
is
reached and then decreases in concen-
Terpene compounds were also identified in grape tration with over-ripeness. Bayonove & Cordonnier
leaves. Wildenradt et el. (1975) reported that pentane (1971) observed increases in concentrations, similar to
extracts
of
Chenin blanc leaves contained terpenes and
that
of
linalool,
of
other terpenes until maturity
is
their derivatives
as
prominent compounds apart from reached. Ripe grapes contained about six times more
the grassy C
6
-
compounds. Myrcene, linalool,
ex:
-ter- linalool, five times more cx:-terpineol, four times more
pineol, iso-pulegone, citral, geraniol, nerol and
ex:-
and nerol and 1,4 times more geraniol than unripe grapes,
B-ionone were identified. Augustyn & Rapp
(1982) fail- while citral and citronellol occurred in nearly the same
ed to find measurable concentrations
of
terpenes in the concentrations in the unripe and ripe grapes. Versini et
berries
of
Chenin blanc, and the question arises whether
al.
(1981) determined terpenes in Weisser Riesling
and which biochemical processes are involved in the grapes during the final ripening period (about one
possible translocation
of
terpene compounds from the month) and observed increases as
well
as decreases in
leaves to the grapes. different terpene concentrations. These studies clearly
Identification
of
and differentiation between cultivars,
using terpenes: Generally, it
is
difficult to differentiate
analytically between grape cultivars. An exception in this
regard
is
muscat
and
aroma
related
cultivars,
since
cultivar-typical grape aroma components, such
as
terpene
compounds, are quite suitable for the analytical differen-
tiation between some cultivars
(Schreier, 1979). Terpene
composition, as
well
as their concentrations differ between
cultivars and are responsible for differences in aroma.
By
employing
mathematical
techniques,
considerable
research has been done in order to identify cultivars and to
distinguish between them on the basis
of
their terpene com-
position and concentrations. These analyses were useful in
establishing which compounds were cultivar-typical, such
as
terpenes,
as
well
as
in
determining their effect on the
quality
of
grapes and wine (Wagner, Dirninger & Fuchs,
1974; Rapp&Hastrich, 1976; Schreier, Drawert&Junker,
1976 a; Rapp & Hastrich, 1978 b).
Terrier
et
al.
(1972 b) demonstrated cultivar-typical
distribution patterns
of
geraniol, nerol, oc-terpineol and
the linalool oxides
in
different white
V.
vinifera cultivars.
Schreier et
al.
(1976
c)
and Drawert & Schreier (1978) suc-
ceeded in distinguishing mathematically between the
cultivars Rulander, Morio Muscat, Gewi.irztraminer,
Scheurebe, Weisser Riesling and
Mi.iller
Thurgau using
the concentrations
of
the terpene compounds linalool,
hotrienol, geraniol, nerol, cx:-terpineol, nerol oxide, the
rose oxides and the linalool oxides
of
grapes as
well
as
wines. In this case, the rose oxides and nerol oxide emerg-
ed as the aroma components with the most significant
discriminating abilities. Rapp, Hastrich & Engel
(1977),
Rapp & Hastrich (1978
a)
and Rapp et
al.
(1978) under-
took detailed investigations into the aroma
("finger-
print")
patterns
of
different cultivars.
By
employing the
concentrations
of
certain terpene compounds the authors
could differentiate between the cultivars
Sylvaner,
Weisser Riesling and Morio Muscat and the cross B 6 -
18
irrespective
of
grape maturity, vintage year and origin.
More subtle differentiations
("fine pattern") were also
possible between Weisser Riesling grapes cultivated
in
different wine areas.
The effect of grape maturity on terpene concentrations:
Several studies were undertaken to determine the effect
of
grape maturity on terpene concentrations in muscat
indicated that the maximum aroma can be attained
before all the sugar has been accumulated.
In contrast to the above-mentioned findings, prelimi-
nary investigations
(A.
Rapp,
0.
P. H. Augustyn &
L.
Engel, 1980; personal communication) indicated
decreases in the concentrations
of
certain terpenes
of
some muscat and aroma related cultivars in South
Africa with an increase in grape maturity. Reasons for
these contradicting tendencies must still be found, but
may possibly be ascribed to, amongst others, the warm
South African climatic conditions.
Wagner
et
al.
(1977), Cordonnier & Bayonove (1978)
and Simpson (1979 b) stressed the importance
of
harves-
ting grapes at the correct maturity. In order to exploit
the aromatic potential
of
muscat and aroma related
grapes, they should be harvested before maximum sugar
accumulation, at least before over-ripeness, but not too
early in order to prevent the detrimental effect
of
com-
ponents derived from unripe grapes on wine quality.
The optimum maturity
of
grapes in terms
of
the opti-
mum terpene composition must still be determined.
The effect
of
certain wine-making techniques on terpene
concentrations and wine quality: In general, there
is
proportionately higher concentrations
of
nerol
in
muscat grapes with small berries compared to those with
larger berries, such as Muscat
d'
Alexandrie. Geraniol
and linalool represent approximately
7 5
OJo
to
80%
of
the
total terpene content
of
grapes and wine, and the largest
amounts were found in the case
of
Muscat
d'
Alexandrie
(Cordonnier & Bayonove,
1978). The volatile compo-
nents
of
the berry are mainly found in the skin and the
solid parts
of
the cells. In the case
of
terpene com-
pounds, Bayonove, Cordonnier & Ratier
(1974),
Cordonnier & Bayonove (1978) and Cordonnier &
Bayonove
(1981) indicated that in Muscat
d'
Alexandrie
grapes,
94%
of
geraniol and
96%
of
nerol are located
in
the skin, whereas linalool
is
almost equally distributed
between the juice
(50%)
on the one hand and the skin
(26%)
and cellular debris
(24%)
on the other. Versini
et
al.
(1981) found similar trends for geraniol, nerol and
citronellol in the case
of
Weisser Riesling grapes, name-
ly
the occurrence
of
high concentrations in the skins.
Linalool also occurred in greater concentrations
in
the skins, namely
60%.
The linalool oxides appeared
in evenly distributed concentrations between the skins
S. Afr.
J.
Enol. Vitic., Vol. 4. No. 2. 1983