Direct formation of 4-alkoxy derivatives from 2,5-dimethyl-4-hydroxy-3(2H)-furanone and aliphatic alcohols

Article abstract of DOI:10.1021/jf950451g

The reaction of 2,5-dimethyl-4-hydroxy-3(2H)-furanone (DHF) with aliphatic alcohols using a small amount of strong acid catalyst proceeded, in an analogous way to classical carboxylic acid ester formation, to give good yields of the 4-alkoxy derivatives. This raises the possibility that this type of ester-like formation might occur in foods that contain both DHF and alcohols. Odor threshold determinations of 4-alkoxy DHF derivatives from a number of common food alcohols including methanol, ethanol, butanol, 3-methylbutanol, and (Z)-3-hexenol showed them, however, to be relatively weak odorants.

Full text of DOI:10.1021/jf950451g

1512
J. Agric. Food Chem. 1996, 44, 1512−1514
Dir ect F or m a tion of 4-Alk oxy Der iva tives fr om
2,5-Dim eth yl-4-h yd r oxy-3(2H)-fu r a n on e a n d Alip h a tic Alcoh ols
Ron G. Buttery* and Louisa C. Ling
Agricultural Research Service, Western Regional Research Center, U.S. Department of Agriculture,
Albany, California 94710
The reaction of 2,5-dimethyl-4-hydroxy-3(2H)-furanone (DHF) with aliphatic alcohols using a small
amount of strong acid catalyst proceeded, in an analogous way to classical carboxylic acid ester
formation, to give good yields of the 4-alkoxy derivatives. This raises the possibility that this type
of ester-like formation might occur in foods that contain both DHF and alcohols. Odor threshold
determinations of 4-alkoxy DHF derivatives from a number of common food alcohols including
methanol, ethanol, butanol, 3-methylbutanol, and (Z)-3-hexenol showed them, however, to be
relatively weak odorants.
Keyw or d s: 2,5-Dimethyl-4-hydroxy-3(2H)-furanone; alkoxy derivatives; synthesis; odor thresholds
INTRODUCTION
CaSO4, was attached to the flask. A drying tube was placed
on top of the reflux condensor. The mixture was refluxed for
16 h and then allowed to cool. The p-toluenesulfonic acid was
then removed with NaHCO3 as above and the filtered solution
concentrated under reduced pressure. The yield of 2,5-
dimethyl-4-butoxy-3(2H)-furanone based on the DHF was 90%.
The product was purified by preparative GC.
Syn t h esis of 2,5-Dim et h yl-4-(3-m et h ylb u t oxy)-3(2H)-
fu r a n on e a n d 2,5-Dim et h yl-4-((Z)-3-h exen oxy)-3(2H )-
fu r a n on e. These compounds were synthesized in a way
similar to that of the butoxy compound using benzene with
the Dean-Stark trap to remove water. The yields based on
the original DHF were 50-60%.
Ma ss Sp ectr a . These were obtained with a HP 5970 (mass
detector) quadrupole mass spectrometer. The samples were
introduced into the mass spectrometer using a HP 5890 gas
chromatograph with a 60 m long by 0.25 mm i.d. fused silica
capillary, wall coated with DB-1.
Nu clea r Ma gn etic Reson a n ce Sp ectr a . These were
measured in CDCl3 with a Bruker ARX 400 spectrometer.
In fr a r ed Absor p tion Sp ectr a . These were measured neat
as films between salt plates using a Perkin-Elmer Model 1600
FT-IR infrared spectrophotometer.
2,5-Dimethyl-4-hydroxy-3(2H)-furanone (DHF) has
been well established as an important aroma component
of a number of fruits including pineapple (Rodin et al.,
1965), strawberry (Re et al., 1973), and raspberry
(Honkanen, 1980). The methoxy derivative, 2,5-di-
methyl-4-methoxy-3(2H)-furanone (DMF), is also a known
component of strawberries (Ito et al., 1990) and arctic
bramble (Kalio, 1976). The method commonly used for
the synthesis of DMF is the action of diazomethane on
DHF (cf. Sen et al., 1991; Wilhalm et al., 1965).
We have been studying the importance of DHF to
tomato flavor (Krammer et al., 1994; Buttery et al.,
1995). Some tomato products contain appreciable
amounts of (Z)-3-hexenol and 3-methylbutanol. If DHF
could react directly with these alcohols under the mild
acid conditions of the tomato then there is a possibility
that alkoxy derivatives of these alcohols with DHF
would be present. This study was undertaken to test
whether such direct reaction was possible under normal
(more vigorous) laboratory conditions. It was also
undertaken to obtain spectral reference data and to
evaluate the sensory significance of the most likely
alkoxy compounds.
Ma ter ia ls. 2,5-Dimethyl-4-hydroxy-3(2H)-furanone and
common alcohols were obtained from Aldrich Chemical Co.,
Milwaukee, WI. (Z)-3-Hexenol was obtained from Bedoukian
Research, Inc., Danbury, CT. 2,5-Dimethyl-4-methoxy-3(2H)-
furanone was obtained from Firmenich SA., Geneva, Switzer-
land.
EXPERIMENTAL PROCEDURES
Od or Th r esh old Deter m in a tion . These were determined
on samples purified by preparative GC in odor-free distilled
water (pH 7) following procedures already described (e.g.,
Buttery et al., 1995). The number of experienced judges was
18-21. The solutions were presented to the judges in flexible
Teflon bottles each fitted with a 6 cm long by 0.64 cm diameter
Teflon tube through the bottle cap to facilitate transfer of the
vapor to the judge’s nose when the bottle was gently squeezed.
Solutions were diluted by factors of 2 for each step in the
judgment. For each concentration the judges were given two
coded bottles, one containing the solution and the other odor-
free water. The judge’s task was to record which coded bottle
possessed an odor. Probit analysis (a statistics program for
drug dose tests) was used to determine the threshold point
(taken where 75% of panelists give the correct judgment) and
the 95% confidence intervals.
Syn t h esis of 2,5-Dim et h yl-4-et h oxy-3(2H )-fu r a n on e.
2,5-Dimethyl-4-hydroxy-3-(2H)-furanone (2.0 g) was added to
anhydrous ethanol (120 mL) in a 250 mL round bottom flask.
p-Toluenesulfonic acid (100 mg) was then added, a drying tube
attached to the top of the reflux condenser, and the mixture
refluxed for 6 h. The mixture was then cooled, shaken with
solid NaHCO3 (5 g), and filtered through a ca. 2 mm layer of
NaHCO3 to remove the acid catalyst. The ethanol was then
removed under reduced pressure. The 2,5-dimethyl-4-ethoxy-
3(2H)-furanone was purified by preparative gas chromatog-
raphy (GC). The yield (determined by GC) based on the DHF
was 80%.
Syn th esis of 2,5-Dim eth yl-4-m eth oxy-3(2H)-fu r a n on e.
This was synthesized using essentially the same procedure as
for the ethoxy compound except methanol was used instead
of ethanol. The yield also was better than 80%.
Syn t h esis of 2,5-Dim et h yl-4-b u t oxy-3(2H )-fu r a n on e.
2,5-Dimethyl-4-hydroxy-3(2H)-furanone (2.0 g) was added to
100 mL of butanol, 50 mL of benzene, and 100 mg of
RESULTS AND DISCUSSION
The methods for the synthesis of esters from aliphatic
alcohols and carboxylic acids have long been well
p-toluenesulfonic acid in a 250 mL round bottom flask.
A
Dean-Stark trap, containing ca. 10 g of 8 mesh anhydrous
S0021-8561(95)00451-1
This article not subject to U.S. Copyright. Published 1996 by the American Chemical Society

Direct Formation of 4-Alkoxy Derivatives
J. Agric. Food Chem., Vol. 44, No. 6, 1996 1513
Ta ble 1. Od or Th r esh old s of Alk oxy Der iva tives of 2,5-Dim eth yl-4-h yd r oxy-3(2H)-fu r a n on e in Wa ter (p H 7)
compd
threshold in nL/La (ppb)
95% confidence limits
2,5-dimethyl-4-methoxy-3(2H)-furanone
2,5-dimethyl-4-ethoxy-3(2H)-furanone
2,5-dimethyl-4-butoxy-3(2H)-furanone
2,5-dimethyl-4-(3-methylbutoxy)-3(2H)-furanone
2,5-dimethyl-4-((Z)-3-hexenoxy)-3(2H)-furanone
3400
2900
250
900
1900
2000-6100
1600-16000
170-350
600-1500
1400-3700
a
10-9 L of compound per L of water.
established. It was found that essentially the same
methods were also successful if 2,5-dimethyl-4-hydroxy-
3(2H)-furanone (DHF) was treated as a carboxylic acid.
It was known that DHF has acidic properties, the -OH
and CdO arrangement having some similarities to that
of some other noncarboxylic acidic compounds such as
ascorbic acid and tropolone. We had noted (Buttery et
al., 1995) that a 2% water solution of DHF had a pH of
2.8, not very different from that of a 2% water solution
of acetic acid (pH 2.7). Using a published procedure
(Wilcox, 1984), titration with 0.5 N NaOH showed DHF
to have a pKa of ca. 8.2. This value shows DHF to be a
weaker acid than aliphatic acids (pKa ) 4-5) but a
stronger acid than phenol (pKa ) 9.9).
The yields of the DHF alkoxy derivatives were of the
same order as generally found with carboxylic acid ester
formation. The DHF seemed reasonably stable under
the conditions of ester formation, although precautions
were taken to use conditions so that the temperature
remained below 100 °C by using benzene for the higher
boiling alcohols. The benzene also functioned to remove
water to a Dean-Stark trap.
GC retention index, 1305; 1H NMR spectrum δ 0.93, d,
J ) 7 Hz, 6H (CH3CH(CH3)CH2CH2-), 1.43, d, J ) 7
Hz, 3H (CH3- at 2 position of the furanone ring), 1.54,
q, J ) 7 Hz, 2H (-CH2CH2O-), 1.76, hep, J ) 7 Hz,
1H (CH3CH(CH3)CH2CH2-); 2.18, s, 3H (CH3- at 5
position of the furanone ring), 4.04, m, 2H (-CH2CH2O-
), 4.40,q, J ) 7 Hz, 1 H (H at 2 position of the furanone
ring); 13C NMR spectrum δ 198.5, 178.2, 135.7, 80.3,
70.5, 38.6, 24.8, 22.55, 22.52, 16.5, 13.5; IR spectrum
major absorption bands at 2958, 1703, 1631, 1426, 1311,
1199, 1005 cm-1
.
2,5-D im e t h y l-4-((Z)-3-h e x e n o x y )-3(2H )-fu r a -
n on e. Mass spectrum showed ions at 41 (47), 43 (84),
55 (83), 57 (21), 67 (23), 72 (33), 83 (10), 85 (77), 128
(100), 129 (17), 167 (4), 210 (8) M+; Kovats’ GC retention
index on the DB-1 capillary, 1480; 1H NMR spectrum δ
0.97, t, J ) 7 Hz, 3H (CH3CH2CH)CH-), 1.43, d, J )
7 Hz, 3H (CH3- at 2 position of the furanone ring), 2.06,
p, J ) 7 Hz, 2H (CH3CH2CHdCH-), 2.18, d, J ) 1 Hz,
3H (CH3- at 5 position of the furanone ring), 2.40, q, J
) 7 Hz, 2H (-CHdCHCH2CH2-), 4.03, m, 2H
(-CH2CH2O-), 4.40, q , J ) 7 Hz, 1H (H at 2 position
of the furanone ring), 5.36, m, 1H (-CHdCH-), 5.49,
m, 1H (-CH)CH-); 13C NMR spectra δ 198.3, 178.3,
135.5, 134.2, 124.2, 80.4, 71.4, 28.0, 20.7, 16.5, 14.2,
13.5; IR spectra, major absorption bands at 2965, 1702,
2,5-Dim et h yl-4-m et h oxy-3(2H )-fu r a n on e.
The
sample obtained from the reaction of methanol with
DHF had mass spectra, GC retention index, and IR
spectrum identical to those of an authentic commercial
sample.
1630, 1311, 1198, 1138, 1075, 1009 cm-1
.
Except for the methoxy compound these DHF alkoxy
compounds do not seem to have been previously de-
scribed in the literature. The spectral data are quite
consistent with these structures. One other possible
product from this type of reaction is the formation of
ketals from the reaction of the alcohols with the CdO
group in the 3-position of the furanone ring. Ketals of
this type would have a considerably higher molecular
weight and show quite different IR and NMR spectral
data. No ketals were detected.
From this study the ease with which DHF reacts with
alcohols raises the possibility that this type of alkoxy
formation might occur in foods, where both are found,
especially because of the acidic nature of many fruits
and vegetables. Reactions that occur easily in the
laboratory can also frequently be catalyzed by enzyme
systems and such a pathway has been suggested by
Kalio (1976) for formation of DMF from DHF in arctic
bramble.
Od or Th r esh old Stu d ies. Table 1 lists the odor
thresholds determined for these compounds in odor-free
distilled water (pH 7.0). They are all considerably
weaker odorants than 2,5-dimethyl-4-hydroxy-3(2H)-
furanone itself (threshold 60 mg/L at pH 7; Buttery et
al., 1995). The most potent was the butoxy derivative
with a threshold of 250 nL/L. The least potent was the
well-known methoxy derivative with a threshold of 3400
nL/L. It might be noted that in other studies (Pyysalo
et al., 1977) a threshold of 0.03 nL/L (ppb) had been
reported for the methoxy derivative. Our data on the
GC purified material is ca. 105 times higher than this.
Pyysalo et al. (1977) did not report that their sample
2,5-Dim eth yl-4-eth oxy-3(2H)-fu r a n on e. This com-
pound had a mass spectrum (intensities in parentheses)
showing ions at 41 (5), 43 (100), 55 (9), 57 (54), 69 (5),
72 (15), 83 (2), 85 (58), 99 (2), 112 (2), 113 (3), 127 (1),
128 (15), 141 (2), 156 (48) M+; Kovats’ GC retention
1
index on DB-1, 1090; H NMR spectrum δ 1.27, t, J )
7 Hz, 3H (CH3CH2-), 1.44, d, J ) 7 Hz, 3H (CH3- 2
position of the furanone ring), 2.19, d, J ) 1 Hz, 3H
(CH3- at 5 position of the furanone ring), 4.07, m, 2H
(CH3CH2O-), 4.40, q, J ) 7 Hz, 1 H (H at 2 position of
the furanone ring); 13C NMR δ 198.5, 178.5, 135.3, 80.3,
67.6, 16.48, 15.3, 13.5; IR spectrum, major absorption
bands at 2982, 1704, 1632, 1426, 1311, 1202, 1031, 998
cm-1
.
2,5-Dim eth yl-3-bu toxy-3(2H)-fu r an on e. Mass spec-
trum showed ions at 39 (12), 41 (22), 43 (80), 55 (9), 57
(49), 69 (2), 72 (40), 85 (100), 99 (1), 110 (1), 128 (67),
141 (2), 184 (26) M+; Kovats’ GC retention index on DB-
1, 1260; 1H NMR spectrum δ 0.94, t, J ) 7 Hz, 3H (CH3-
CH2CH2-), 1.42, m, 2H (CH3CH2CH2-), 1.43, d, J ) 7
Hz, 3H (CH3- at 2 position of the furanone ring), 1.62,
m, 2H (CH3CH2CH2CH2-), 2.18, d, J ) 1 Hz, 3H (CH3-
at 5 position of the furanone ring), 4.0, m, 2H
(-CH2CH2O-), 4.4, q, J ) 7 Hz, 1H (H at 2 position of
the furanone ring); 13C NMR spectrum δ 198.5, 178.3,
135.7, 80.3, 71.8, 31.9, 18.9, 16.5, 13.8, 13.5; IR spec-
trum, major absorption bands at 2961, 1703, 1631, 1311,
1200, 1140, 1075, 1003 cm-1
.
2,5-Dim e t h yl-4-(3-m e t h ylb u t oxy)-3(2H )-fu r a -
n on e. Mass spectrum showed ions at 41 (21), 43 (100),
55 (14), 57 (22), 71 (7), 72 (34), 83 (4), 85 (74), 111 (2),
128 (82), 129 (14), 141 (1), 155 (1), 198 (25) M+; Kovats’

1514 J. Agric. Food Chem., Vol. 44, No. 6, 1996
Buttery and Ling
coside from tomatoes. J . Agric. Food Chem. 1994, 42, 1595-
1597.
Pyysalo, T.; Suihko, M.; Honkanen, E. Odour thresholds of the
major volatiles identified in cloudberry and arctic bramble.
Lebensm. Wiss. Technol. 1977, 10, 36-39.
Re, L.; Maurer, G.; Ohloff, G. A simple synthesis of 4-hydroxy-
2,5-dimethyl-3(2H)-furanone (furaneol) an important aroma
component of pineapple and strawberry. Helv. Chim. Acta
1973, 56, 1882-1894.
Rodin, J . O.; Himmel, R. M.; Silverstein, R. M.; Leeper, R. W.;
Gartner, W. A. Volatile flavor and aroma components of
pineapple. J . Food Sci. 1965, 30, 280-285.
Sanz, C.; Richardson, D. G.; Perez, A. G. 2,5-Dimethyl-4-
hydroxy-3(2H)-furanone and derivatives in strawberries
during ripening. In Fruit Flavors: Biogenesis, Characteriza-
tion and Authentication; Rouseff, R. L., Leahy, M. M., Eds.;
ACS Symposium Series 596; American Chemical Society:
Washington, DC, 1995; Chapter 24.
Sen, A.; Schieberle, P.; Grosch, W. Quantitative determination
of 2,5-dimethyl-4-hydroxy-3(2H)-furanone and its methyl
ether using a stable isotope dilution assay. Lebensm. Wiss.
U. Technol. 1991, 24, 364-369.
Wilcox, C. F., J r. Experimental Organic Chemistry: Theory and
Practice; Macmillan: New York, 1984; pp 305-308.
Wilhalm, B.; Stoll, M.; Thomas, A. F. 2,5-Dimethyl-4-hydroxy-
2,3-dihydrofuran-3-one. Chem. Ind. (London) 1965, 1629-
1630.
was purified by GC. Small concentrations of impurities
can frequently lower the apparent thresholds of weak
odorants significantly.
The methoxy compound has been reported as occur-
ring in strawberries at a concentration as high as 18500
nL/L (Sanz et al., 1995) which is several times over its
threshold and at that level it has a good probability of
contributing to the odor and flavor. As the other alkoxy
compounds have not yet been reported in fruits and
vegetables, if they are present, they probably only occur
at trace levels. With the high odor thresholds found it
would seem that they have a very low probability of
contributing to the odor and flavor of the food.
ACKNOWLEDGMENT
We thank Rosalind Y. Wong for the NMR spectra.
LITERATURE CITED
Buttery, R. G.; Takeoka, G. R.; Ling, L. C. Furaneol: odor
threshold and importance to tomato aroma. J . Agric. Food
Chem. 1995, 43, 1638-1640.
Honkanen, E.; Pyysalo,T.; Hirvi, T. The aroma of finnish wild
raspberries. Z. Lebensm. Unters. Forsch. 1980, 171, 180-
182.
Ito, O.; Sakakibara, H.; Yajima, I.; Hayashi, K. The changes
in the volatile components of strawberries with maturation.
In Flavor Science and Technology; Bressiere, Y., Thomas,
A. F., Eds.; Wiley: Chichester, England, 1990; pp 69-72.
Kalio, H. Identification of vacuum steam-distillable aroma
compounds in the press juice of arctic bramble. J . Food Sci.
1976, 41, 555-562.
Received for review J uly 17, 1995. Revised manuscript received
J anuary 11, 1996. Accepted April 10, 1996.X
J F950451G
X Abstract published in Advance ACS Abstracts, May
15, 1996.
Krammer, G. E.; Takeoka, G. R.; Buttery, R. G. Isolation and
identification of 2,5-dimethyl-4-hydroxy-3(2H)-furanone gly-