Mechanistic Studies of Tetramethylpyrazine Formation under Weak Acidic Conditions and High Hydrostatic Pressure

Article abstract of DOI:10.1021/jf9503671

A significant enhancement of the tetramethylpyrazine (TMP) formation at high pressure was observed in the 3-hydroxy-2-butanone/ammonium acetate model system. In a water system, the activation volume of TMP formation under high pressure was found to be -6.82 mL/mol. A mechanism was proposed to elucidate the formation of TMP under a weak acidic condition and high hydrostatic pressure. Solvents such as propylene glycol (PG), glycerol, methanol, ethanol, propanol, and butanol were found to enhance TMP formation. Kinetic analyses indicated that TMP formation in aqueous, 80% PG, and ethanol systems followed pseudo-zero-order reaction kinetics. The activation energies were found to be 18.84 ± 1.3, 14.19 ± 7.1, and 13.09 ± 4.7 kcal/mol, respectively. The intermediate of TMP formation was characterized as tetramethyldihdyropyrazine using gas chromatography-mass spectrometry. A ¹⁵N-labeled ammonium acetate/3-hydroxy-2-butanone model system was used to confirm the incorporation of a nitrogen atom in the molecule of tetramethyldihydropyrazine. Hydrogen acceptors such as nicotinamide adenine dinucleotide and flavin adenine dinucleotide were found to increase TMP formation, and the formation of TMP from tetramethyldihydropyrazine through dehydrogenation was shown.

Full text of DOI:10.1021/jf9503671

240
J. Agric. Food Chem. 1996, 44, 240−246
Mech a n istic Stu d ies of Tetr a m eth ylp yr a zin e F or m a tion u n d er Wea k
Acid ic Con d ition s a n d High Hyd r osta tic P r essu r e
†
‡
,‡
Tzou-Chi Huang, Hui-Yin Fu, and Chi-Tang Ho*
Department of Food Science and Technology, National Pingtung Polytechnic Institute, Pingtung, Taiwan,
Republic of China, and Department of Food Science, Cook College, New J ersey Agricultural Experiment
Station, Rutgers, The State University of New J ersey, New Brunswick, New J ersey 08903
A significant enhancement of the tetramethylpyrazine (TMP) formation at high pressure was
observed in the 3-hydroxy-2-butanone/ammonium acetate model system. In a water system, the
activation volume of TMP formation under high pressure was found to be -6.82 mL/mol. A
mechanism was proposed to elucidate the formation of TMP under a weak acidic condition and
high hydrostatic pressure. Solvents such as propylene glycol (PG), glycerol, methanol, ethanol,
propanol, and butanol were found to enhance TMP formation. Kinetic analyses indicated that TMP
formation in aqueous, 80% PG, and ethanol systems followed pseudo-zero-order reaction kinetics.
The activation energies were found to be 18.84 ( 1.3, 14.19 ( 7.1, and 13.09 ( 4.7 kcal/mol,
respectively. The intermediate of TMP formation was characterized as tetramethyldihdyropyrazine
1
5
using gas chromatography-mass spectrometry. A N-labeled ammonium acetate/3-hydroxy-2-
butanone model system was used to confirm the incorporation of a nitrogen atom in the molecule
of tetramethyldihydropyrazine. Hydrogen acceptors such as nicotinamide adenine dinucleotide and
flavin adenine dinucleotide were found to increase TMP formation, and the formation of TMP from
tetramethyldihydropyrazine through dehydrogenation was shown.
Keyw or d s: Pyrazine formation; high hydrostatic pressure; kinetics
INTRODUCTION
Pyrazines are nitrogen-containing heterocyclics that
are potent and characteristic flavorants found in a wide
range of raw and processed foods. The odor of tetra-
methylpyrazine (TMP) has been described as pungent,
walnut, and green.
Most of the alkylpyrazines are formed in foods that
are processed at high temperatures. The activation
energies for pyrazine formation are relatively high. Rizzi
F igu r e 1. Proposed overall Schiff base formation mechanism
for TMP synthesis.
tion dramatically. In reactions, though solvents have
been used to replace water as a medium for many years,
the reason for using organic solvents has not been
discussed.
(
1988) postulated that 3-amino-2-butanone is a reaction
intermediate in the formation of TMP. Aminocarbonyls,
such as 3-amino-2-butanone, are known to dimerize to
dihydropyrazine and to oxidize to TMP as shown in
Figure 1. Schiff base formation is believed to be the
critical step in pyrazine synthesis. However, the de-
tailed mechanistic steps of Schiff base formation remain
unclear.
Pressure can influence most chemical reactions, since
they often involve a change in volume. Volume-increas-
ing reactions will tend to be inhibited by pressure, while
reactions leading to a decrease in volume will tend to
be promoted. High hydrostatic pressure is of great
value in organic synthesis and is especially useful for
those compounds that cannot be synthesized at ambient
pressure due to steric hindrances or other reasons.
Numerous papers have been published concerning the
pressure effect on organic synthesis, many of which have
been reviewed by Matsumoto et al. (1985). It has been
reported that high pressure could increase organic
synthesis reactions, such as aldol condensation, Michael
reaction, alkylation of aldehydes, Diels-Alder reaction,
and Mannich reaction.
Most of the published pyrazine formation mechanisms
are studied at high temperatures using a model system.
However, Rizzi (1988) reported that alkylpyrazines were
formed in reactions of acyloins and ammonia at acidic
pH and low temperature. Pyrazine formation at ambi-
ent temperature under high hydrostatic pressure has
not been previously studied. Therefore, the objectives
of this research were to study the effects of temperature
and pressure on TMP formation and to identify the
intermediate of the reaction. The mechanism of TMP
Dihydropyrazine has been reported to be an impor-
tant intermediate in pyrazine formation. According to
the generally accepted mechanism for pyrazine forma-
tion in the Maillard reaction, pyrazines are formed in
an oxidation step from their dihydropyrazine precursors.
Shibamoto and Bernhard (1977) summarized the for-
mation of pyrazines via the oxidation of dihydropyra-
zine. At least two possible mechanisms to convert
dihydropyrazine to pyrazine without oxygen were pro-
posed. One is the disproportionation of dihydropyra-
zine, and the second mechanism involves the dehydro-
genation of dihydropyrazine.
Solvent effect on Schiff base formation is another
controlling factor that could promote Schiff base forma-
*
Author to whom correspondence should be ad-
dressed [fax (908) 932-9672; e-mail Chi-tang@
a1.caft1.vax.Rutgers.edu].
†
National Pingtung Polytechnic Institute.
Rutgers.
‡
0021-8561/96/1444-0240$12.00/0
© 1996 American Chemical Society

Pyrazine Formation under High Hydrostatic Pressure
J. Agric. Food Chem., Vol. 44, No. 1, 1996 241
formation in a weak acid condition and under high
hydrostatic pressure is discussed.
Ta ble 1. Effect of Differ en t Am m on iu m Sa lts on TMP
F or m a tion
ammonium
salt
TMP concn
(mg/mL)
ammonium
salt
TMP concn
(mg/mL)
EXPERIMENTAL PROCEDURES
acetate
5.40
0.56
0.21
0.14
0.06
glutamate
chloride
oxalate
0.06
undetectable
a
a
Ma ter ia ls. 3-Hydoxy-2-butanone and tetramethylpyrazine
TMP) were purchased from Aldrich Chemical Co. (Milwaukee,
WI). Ammonium salts (acetate, formate, oxalate, hydroxide,
carbonate, bicarbonate, sulfate, and chloride), propylene glycol
PG), and the solvents for HPLC were of chemical grade and
bicarbonate
carbonate
formate
(
sulfate
hydroxide
(
a
obtained from Fisher Chemical Co. Ammonium glutamate
was obtained from Ajinomoto Co. Nicotinamide adenine
No product was detected due to the poor solubility of the
ammonium salts.
+
dinucleotide (NAD ) and flavin adenine dinucleotide (FAD)
were purchursed from Sigma Chemical Co. (St. Louis, MO).
volume per mole when the activated complex is formed from
1
5
q
[
(
N]Ammonium acetate was purchased from Isotec, Inc.
Miamisburg, OH).
the reactants. ∆V can be obtained from the plot of pressure
verus rate constant. Various methods have been proposed and
q
Rea ction P r oced u r e. Mixtures were composed of 0.01 mol
used to calculate ∆V . The most realistic one is fitting by the
of 3-hydroxy-2-butanone and 0.03 mol of ammonium salts
dissolved in 4 mL of deionized water or solvent. The vials were
shaken regularly to ensure that all of the reactants were
dissolved. The reactions were run in a water bath at the
required constant temperature. The amount of TMP was
analyzed by HPLC.
HP LC An a lysis. Reaction mixtures were routinely ana-
lyzed by HPLC using a Waters Associate liquid chromatograph
Model 6000A and a Model 440 absorbance detector (280 nm)
with a 25 × 0.46 cm RP Partisphere C18 column (Whatman)
under isocratic conditions at ambient temperature. Solvent
was 50/50 (v/v) methanol-water (1.0 mL/min) for all separa-
tions. Tetramethylpyrazine standards were prepared at 0.02,
least-squares method. Among all of the equations proposed
and used, the most popular is the parabolic one, ln K ) a +
2
q
bp + cp , so that then, at p ) 0, ∆V ) -bRT, where R is the
gas constant, 82 cm3 atm K mol (Asano and Le Noble,
-1
-1
1978). The pressure units used are 1 atm ) 1.01325 bar ) 1
2
kg/cm ) 0.101325 MPa. These equations are similar to those
classically calculated for temperature as a variable at constant
pressure.
Ga s Ch r om a togr a p h y. A Varian gas chromatograph
equipped with a fused silica column (60 m × 0.32 mm i.d.,
film thickness, 0.25 nm; DB-1; J &W Scientific) and a flame
ionization detector was used to analyze the volatile compounds.
The operating conditions were as follows: injector and detector
temperatures, 270 and 300 °C, respectively; helium carrier flow
rate, 1.0 mL/min; temperature program, 40-260 °C at 2 °C/
min followed by an isothermal hold at 260 °C for 10 min.
Ga s Ch r om a t ogr a p h y-Ma ss Sp ect r om et r y. GC-MS
analysis was accomplished by using a Varian 3400 gas
chromatograph coupled to a Finnigan MAT 8230 high-resolu-
tion mass spectrometer equipped with an open split interface.
Mass spectra were obtained by electron ionization at 70 eV
and a source temperature of 250 °C. The filament emission
current was 1 mA, and spectra were recorded on a Finnigan
MAT SS 300 data. The operating conditions were the same
as those used in the GC analysis described above.
0
.04, 0.06, 0.08, and 0.10 mg/mL in methanol. Quantitation
was done on base-line-resolved peaks vs external standards
on a Varian 4270 integrator.
High Hyd r osta tic P r essu r e Exp er im en ts. The reactions
were done by putting the reactants into small plastic vials
(polyethylene bottles with a screwed lid with approximately
4
.5 mL capacity) which were filled with water or propylene
glycol. High pressure was applied to the plastic vials with a
hand-type pressure generator (Type K-P5-B, Hikari, Koatsu
Co., Hiroshima, J apan) using water as the pressure medium.
The temperature was maintained by immersing the pressure
vessel in the water bath.
Effect of Differ en t Nitr ogen Sou r ces on TMP F or m a -
tion . Different ammonium salts (acetate, bicarbonate, car-
bonate, hydroxide, oxalate, glutamate, chloride, and sulfate)
were set up to study the influence of nitrogen sources on TMP
formation.
Effect of Hyd r ogen Accep tor s on TMP F or m a tion . The
hydrogen acceptors NAD and FAD were added to the 3-hy-
droxy-2-butanone/ammonium acetate model system in the
amount of 15, 75, and 150 µmol. The reactions were run at
25 °C for 24 h. GC was used to analyze the amount of TMP.
+
Effect of Wa ter Con ten t on TMP F or m a tion . Different
percentages of PG (0, 20, 40, 60, 80, and 100% in water) were
used to investigate the water content on TMP formation at 1
RESULTS AND DISCUSSION
2
and 5000 kg/cm , respectively.
TMP F or m a tion in Differ en t Am m on iu m Sa lts.
At ambient pressure, the amount of TMP was influenced
by the type of ammonium salt. Among the nitrogen
sources, as shown in Table 1, ammonium acetate
produced the most TMP. The pKa values for acetic acid
and ammonium are 4.76 and 9.25, respectively. An
equimolar solution of ammonia and acetic acid had the
maximum buffering capacity at pH 7.0.
According to Scudder (1992), Schiff base formation
could be separated into four sequential steps. These
steps include trimolecular addition (AdE3), proton
transfer, and general acid catalyzed â-elimination (EgA),
followed by another proton transfer. The first proton
transfer increased the polarization of the carbonyl
group, resulting in an increase of nucleophilic activity
of ammonia which led to the formation of a positively
charged transition state. For Schiff base formation to
occur, both the hydroxyl group and the hydrogen on the
nitrogen must be released. The second proton transfer
was necessary to remove the hydroxyl group, which is
converted to a molecule of water. The proton transfer
not only made the hydroxyl group a better leaving group
but also made the lone nitrogen pair push the hydroxyl
group out. The third proton transfer led to the elimina-
Effect of Solven ts on TMP F or m a tion . Solvents such
as methanol, ethanol, propanol, butanol, PG, and glycerol were
used to investigate the solvent effect on TMP formation at 25
°
C for 1 h.
Kin etic An a lysis. The rate of TMP formation in 3-hy-
droxy-2-butanone/ammonium acetate in water, 80% PG, and
ethanol systems at the four reaction temperatures (25, 35, 45,
and 55 °C) was analyzed by HPLC in the reaction mixture
every hour. All kinetic studies were carried out in duplicate.
The kinetics of pyrazine formation was determined using the
basic equation for the rate of change of A with time: dA/dt )
n
KA , where A is concentration, t is time, K is a rate constant,
and n is the reaction order. Slopes and intercepts were
calculated by the linear least-squares method. The activation
energies for formation of the TMP were calculated by the
Arrhenius equation from the slope of the line generated by
plotting the log value of the rate constant verus the reciprocal
of the absolute temperature (Labuza, 1983).
Effect of P r essu r e on P yr a zin e F or m a tion . The rate
of TMP formation in 3-hydroxy-2-butanone/ammonium acetate
in the water system was obtained at 1, 3000, 4000, and 5000
2
kg/cm and 25 °C for 1, 3, or 5 h. The activation volume was
calculated according to the basic relationship between pressure
and rate. The equation originally proposed by van’t Hoff is
q
q
as follows: ∆V ) -RT ln K/P, where ∆V is the change in

242 J. Agric. Food Chem., Vol. 44, No. 1, 1996
Huang et al.
Ta ble 2. Effect of Differ en t Am m on iu m Sa lts on TMP
F or m a tion a t 22 °C u n d er Differ en t P r essu r es
concn (mg/mL) at pressure of
ammonium
salt
1 kg/cm2 500 kg/cm2 1000 kg/cm2 2000 kg/cm2
acetate
12.44
0.27
0.01
a
14.35
0.34
0.04
a
14.29
0.37
0.04
a
22.05
0.45
0.03
a
carbonate
hjydroxide
oxalate
a
Undetectable.
F igu r e 4. Arrhenius plot of TMP formation in water (O), 80%
PG (4), and ethanol (0) systems.
Ta ble 3. Zer o-Or d er Con sta n ts (K) a n d Cor r esp on d in g
2
Regr ession Coefficien ts (R ) of TMP F or m a tion fr om
3
8
-Hyd r oxy-2-bu ta n on e/Am m on iu m Aceta te in Wa ter (1),
0% P G (2), a n d Eth a n ol (3)
conditions
temp (°C)
25
K (mg/mL h)
R2
(1) water
0.0028
0.0089
0.0216
0.0550
0.948
0.989
0.953
0.989
3
4
5
5
5
5
F igu r e 2. Effect of propylene glycol on TMP formation at
2
pressures of 1 (slashed bars) and 5000 (open bars) kg/cm .
(
2) 80% PG
3) ethanol
25
0.3495
1.3309
2.2264
3.3000
0.983
0.989
0.972
0.995
3
4
5
5
5
5
(
25
0.6570
1.8760
2.6520
5.5000
0.989
0.995
0.965
0.990
3
4
5
5
5
5
Ta ble 4. Activa tion En er gies of TMP F or m a tion fr om
3
8
-Hyd r oxy-2-bu ta n on e/Am m on iu m Aceta te in Wa ter (1),
0% P G (2), a n d Eth a n ol (3)
conditions
Ea (kcal/mol)
linearity (R2)
(
1) water
18.84 ( 1.3
14.19 ( 7.1
13.09 ( 4.7
0.999
0.929
0.967
(2) 80% PG
3) ethanol
(
In the 3-hydroxy-2-butanone/ammonium carbonate
system, the amount of TMP increased with increasing
pressure. The amounts of TMP were 0.27 and 0.34 mg/
2
mL at 1 and 500 kg/cm , respectively. When the
2
F igu r e 3. Solvent effect on TMP formation.
pressure increased to 1000 kg/cm , the amount of TMP
increased to 0.37 mg/mL. When the pressure of 2000
2
kg/cm was applied, the amount of TMP increased to
0
tion of the hydrogen atom on the nitrogen atom. The
acetic acid/ammonia buffer systems served as either a
proton donor or an acceptor, which led to the completion
of the Schiff base formation. The important role of
buffer systems in promoting Schiff base formation in
the ammonium acetate/3-hydroxy-2-butanone model
system reconfirmed the concept discussed by de Kok and
Rosing (1994). Similiar evidence has been reported by
Miyashita et al. (1991) and Lin et al. (1992). However,
in their studies no detailed mechanistic elucidation of
the role of buffer salts on Schiff base formation was
discussed.
.45 mg/mL. These results showed that high pressure
could enhance TMP formation. No TMP was detected
in the oxalate system due to the low solubility of
ammonium oxalate.
A significant pressure effect on TMP formation was
observed in the 3-hydroxy-2-butanone/ammonium ac-
etate system. The amounts of TMP at pressures of 500
and 1000 kg/cm were 14.35 and 14.29 mg/mL, respec-
tively. When the pressure was increased to 2000 kg/
cm , the amount of TMP increased to 22.05 mg/mL. The
amount of TMP increased 1.2-fold under 1000 kg/cm
2
2
2
2
Effects of High P r essu r e on TMP F or m a tion . As
shown in Table 2, in the 3-hydroxy-2-butanone/am-
monium hydroxide system, the amount of TMP was 0.01
mg/mL at ambient pressure, increasing to 0.04 mg/mL
and 1.8-fold under 2000 kg/cm for the sample pressur-
ized at 22 °C for 24 h.
Hamman (1957) proposed that those reactions in
which the transition state is more highly ionic, and
hence more extensitively solvated, than the initial state
are greatly accelerated by pressure; those in which the
transition state is less ionic and less solvated than the
initial state are retarded by pressure. The positive
2
at a pressure of 500 kg/cm . However, when the
2
pressure was increased to 1000 or 2000 kg/cm , the
amount of TMP did not increase. This may be due to
the high pH of the reaction system.

Pyrazine Formation under High Hydrostatic Pressure
J. Agric. Food Chem., Vol. 44, No. 1, 1996 243
F igu r e 5. Proposed mechanism of TMP formation in a weak acidic condition under high pressure.
2
The trend was the same at 5000 kg/cm . An increase
in PG up to 80% showed an increase in TMP. In the
8
0% PG system, a significant increase was observed,
indicating that at this water activity the proximity of
the two reactants was enhanced at higher pressure. The
amount of TMP was about 1.5-fold higher than that of
2
TMP at 1 kg/cm , but in 100% propylene glycol, even at
2
5
000 kg/cm , a low amount of TMP was also obtained
due to the high viscosity of PG. PG has long been
utilized as a water activity lowering agent. The water
activities of 20, 40, 60, and 80% PG in water are 0.96,
0
.91, 0.88, and 0.84, respectively (Alzamora et al., 1994).
The water holding capacity of PG shifts the Schiff base
reaction to the final product.
Effect of P r otic Solven ts on TMP F or m a tion .
Significant solvent effect on TMP formation in the
ammonium acetate/3-hydroxy-2-butanone model system
was observed. Protic solvents such as methanol, etha-
nol, propanol, butanol, and glycerol were used to replace
water in our model system. The amount of TMP
increased 27-, 57-, 53-, 36-, and 2-fold, respectively, as
shown in Figure 3.
Binding the solvent molecule to the more highly
charged activated complex resulted in a decrease in the
volume of the system. For the AdE3 addition mecha-
nism, the three reactants of nucleophile, acid, and
carbonyl group must be roughly coplanar. High hydro-
static pressure can facilitate mutual collision, thus
promoting the formation of the transition state inter-
mediate.
F igu r e 6. Mass spectra of tetramethyldihydropyrazine from
1
4
N-labeled (A) and 15N-labeled (B) ammonium acetate.
Kin etic Stu d ies of TMP F or m a tion . For water,
0% PG, and ethanol systems, the rates of TMP forma-
pressure effect of TMP formation was caused by an ionic
intermediate. The apparent activation volume obtained
from the plot of pressure versus the rate constant was
found to be -6.82 mL/mol.
8
tion at each temperature followed pseudo-zero-order
reaction kinetics as shown by the linear plot of TMP
concentration verus time. This indicates that pyrazine
formation in all systems tested is relatively complex.
Rate constants were calculated from the slope using the
least-squares fit method.
The natural log of the rate constant as a function of
the reciprocal of temperature is shown in the Arrhenius
plot (Figure 4). Linear regression data summarizing the
effects of temperature on TMP formation are shown in
Table 3. The activation energies were found to be 18.84
( 1.3, 14.19 ( 7.1, and 13.09 ( 4.7 kcal/mol in water,
80% PG, and ethanol, respectively (Table 4).
Effects of P r op ylen e Glycol on TMP F or m a tion
in 3-Hyd r oxy-2-bu ta n on e/Am m on iu m Aceta te Sys-
tem . Figure 2 shows the different percentages of PG
used for TMP formation in the 3-hydroxy-2-butanone/
ammonium acetate system. At ambient pressure, the
amount of TMP increased with increasing percentage
of PG up to 80%. The concentration of free ammonia
increased the nucleophilic attack of the carbonyl group
of 3-hydroxy-2-butanone and accelerated the formation
of TMP. However, a percentage of PG higher than 80%
did not increase TMP formation due to high viscosity.

244 J. Agric. Food Chem., Vol. 44, No. 1, 1996
Huang et al.
F igu r e 7. Fragmentation pathways of tetramethyldihydropyrazine from 14N-labeled (A) and 15N-labeled (B) ammonium acetate.
1
4
15
Mech a n ism of TMP F or m a tion a t a Wea k Acid ic
Con d ition . As shown in Figure 5, a mechanism for the
formation of TMP under a weak acidic condition was
proposed. The proton from acetic acid protonated the
carbonyl carbon of 3-hydroxy-2-butanone, which facili-
tated the formation of an ionic intermediate. This was
confirmed by the high-pressure experiments. Subse-
quent rearrangement of the intermediate was catalyzed
by the acetate ion and led to the formation of an amino
ketone. Condensation of two molecules of the amino
ketone produced one molecule of TMP.
from [ N]- and [ N]ammonium acetate, respectively.
The mass ranges of these two spectra were, unfortu-
nately, different. The mass range of the spectrum
shown in Figure 6A started at m/ z 40 and that for
Figure 6B started at m/ z 50. In Figure 6A, a molecular
+
ion [M] at m/ z 138 (20%), which corresponded to a
possible elemental composition of C8H14N2, was similiar
to the fragmentation pattern of 2,3-dimethyl-5,6-dihy-
dropyrazine (Porter and Baldas, 1972), which is less
stable to electron impact than the fully aromatic pyra-
zines. The proposed fragmentation pathway is sum-
marized in Figure 7A. The peak observed at m/ z 123
(45%) was the loss of the CH3 group from the parent
molecule. Peaks at m/ z 96 (10%) and m/ z 82 (15%)
represented a loss of HCN and CH3CN fragments,
respectively, from the m/ z 123 ion. The other pathway,
yielding the ion m/ z 56 (48%), is due to expulsion of
two molecules of CH3CN. The compound was, therefore,
tentatively characterized as tetramethyldihydropyra-
zine.
Ch a r a cter iza tion of Tetr a m eth yld ih yd r op yr a -
zin e. An unknown compound with a molecular weight
of 138 was formed in the reaction mixture of ammonium
acetate. This compound was suspected to be tetram-
ethyldihydropyrazine, an intermediate compound to-
ward the formation of TMP. The reaction mixture of
1
5
14
3
-hydroxy-2-butanone/[ N]ammonium acetate or /[ N]-
ammonium acetate was then analyzed by GC and GC-
MS to confirm the incorporation of a nitrogen atom in
this unknown molecule. Parts A and B of Figure 6 show
the electron impact mass spectra of this intermediate
For the reaction mixture of 3-hydroxy-2-butanone/
[ N]ammonium acetate, a molecular ion [M] at m/ z
1
5
+

Pyrazine Formation under High Hydrostatic Pressure
J. Agric. Food Chem., Vol. 44, No. 1, 1996 245
Ta ble 5. Com p a r ison of th e F r a gm en ts of
Tetr a m eth yld ih yd r op yr a zin e Der ived fr om [14N]- a n d
1
5
[
N]Am m on iu m Aceta te
m/z
fragment
14N
15N
I
II
III
IV
V
138
123
96
82
56
140
125
97
83
56
Ta ble 6. Effect of Hyd r ogen Accep tor s on TMP
F or m a tion
µmol added
TMP concn (mg/mL)
0.97
control
NAD+
150
2.57
2.02
1.60
7
1
5
5
FAD
150
1.80
1.60
1.20
7
1
5
5
1
6
40 (35%) rather than m/ z 138 was observed in Figure
B. This could be interpreted as the replacement of two
F igu r e 8. TMP formation from tetramethyldihydropyrazine
through dehydrogenation.
1
4
N atoms with two 15N atoms. The fragmentation
scheme is shown in Figure 7B. The base peak observed
at m/ z 125 (100%) corresponded to a loss of CH3
the formation of TMP through the intermediate, tet-
ramethyldihydropyrazine, is presented in Figure 8.
1
5
from the parent molecule. This revealed that two
N
atoms remained in this fragment. Further loss of the
neutral fragments, CH3CN and HCN, led to the frag-
ments m/ z 83 and 97, respectively. Comparison of the
Con clu sion . Application of high hydrostatic pres-
sure was found to enhance the formation of TMP in a
3-hydroxy-2-butanone/ammonium acetate system. Sol-
vents such as PG, glycerol, methanol, ethanol, propanol,
and butanol were found to enhance TMP formation. The
use of high hydrosatic pressure in combination with an
adequate solvent may lead to a new way to prepare
flavor compounds.
1
4
fragmentation pattern of the intermediate with
N
showed that only one nitrogen remained in these
fragments. The sequential loss of two neutral frag-
ments corresponding to CH3CN from the parent mol-
ecule resulted in the formation of a m/ z 56 frag-
ment. Apparently, there was no nitrogen atom in the
m/ z 56 fragment in either fragmentation pattern. A
comparison of the two suggested fragmentation patterns
LITERATURE CITED
1
4
of the intermediates derived from [ N]ammonium
Abeles, R. H.; Frey, P. A.; J encks, W. P. Biochemistry; J ones
and Bartlett Publisher: New York, 1992; pp 446-476.
Alzamora, S. M.; Chirife, J .; Gerschenson, L. N. Determination
and correlation of the water activity of propylene solutions.
Food Res. Int. 1994, 27, 65-67.
Asano, T.; Le Noble, W. J . Activation and reaction volumes in
solution. Chem. Rev. 1978, 78, 407-439.
de Kok, P. M. T.; Rosing, E. A. E. Reactivity of peptides in
Maillard reaction. In Thermally Generated Flavors, Mail-
lard, Microwave and Extrusion Process; Parliment, T. H.,
Morello, M., McGorrin, R. J ., Eds.; ACS Symposium Series
1
5
acetate and [ N]ammonium acetate is summarized in
Table 5.
Effect of Hyd r ogen Accep tor on TMP F or m a -
+
tion . NAD functions as an oxidizing agent in the
+
formation of TMP. The oxidized forms of NAD and
FAD engage in a two-electron transfer by accepting a
-
hydride ion H from the reduced substrate AH (Abeles
et al., 1992). A series of experiments was developed to
examine the possible effects of oxidative factors on the
formation of pyrazine from dihydropyrazine. In our
experiments a model system of 3-hydroxy-2-butanone/
ammonium acetate in a water system was used which
5
43; American Chemical Society: Washington, DC, 1994;
pp 158-179.
+
Hamman, S. D. Physico-chemical Effects of Pressure; Academic
Press: New York, 1957.
Labuza, T. P. Computer-Aided Techniques in Food Technology;
Dekker: New York, 1983; Vol. 1, 71 pp.
Lin, T. F.; Yakushijin, G. H.; Buchi, G. H.; Demain, A. L.
Formation of water soluble Monascus Red pigments by
biological and semi-synthetic processes. J . Ind. Microbiol.
consisted of NAD and FAD selected as hydrogen
acceptors. The results of the quantitative analyses of
+
TMP are presented in Table 6. Both FAD and NAD
showed a significant effect on the transformation of
dihydropyrazine to the corresponding TMP. The amount
+
of TMP increased with increasing NAD concentration
linearly, in the range of 15-150 µmol. FAD showed a
1
992, 9, 173-179.
Maga, J . A. Pyrazines in foods: an update. Food Rev. Int. 1992,
, 479-558.
similiar effect on TMP formation. When hydrogen
+
acceptors such as FAD or NAD were added to the
8
reaction system, no peak for the intermediate dihydro-
pyrazine was observed.
Matsumoto, K.; Sera, A.; Uchida, T. Organic synthesis under
high pressure I. Reviews. Synthesis 1985, 1-26.
These results confirmed the mechanism proposed by
Turksma, which was cited in a paper by Weenen and
Tjan (1991). It was reported that pyrazines could be
formed from dihydropyrazines in a redox reaction. A
pathway similar to the previous reaction which shows
Miyashita, K.; Kanda, K.; Takagi, T. A Simple and quick
determination of aldehydes in autoxidized vegetable and fish
oils. J . Am. Oil Chem. Soc. 1991, 68, 748-751.
Porter, Q. N.; Baldas, J . Mass Spectrometry of Heterocyclic
Compounds; Wiley: New York, 1971; pp 491-499.

246 J. Agric. Food Chem., Vol. 44, No. 1, 1996
Huang et al.
Rizzi, G. P. Formation of pyrazine from acyloin precursors
Received for review J une 19, 1995. Accepted September 19,
1995. This is publication D-10535-1-95 of the New J ersey
X
under mild conditions. J . Agric. Food Chem. 1988, 36, 349-
3
52.
Agricultural Experiment Station supported by State Funds
and the Center for Advanced Food Technology (CAFT). CAFT
is a New J ersey Commission on Science and Technology. This
work was also supported in part by the U.S. Army Research
Office.
Scudder, P. H. Electron Flow in Organic Chemistry; Wiley:
New York, 1992; pp 90-112.
Shibamoto, T.; Bernhard, R. A. Investigation of pyrazine
formation pathways in glucose-ammonia model systems.
Agric. Biol. Chem. 1977, 41, 143.
Weenen, H.; Tjan, S. B. Analysis, structure, and reactivity of
J F9503671
3
-deoxyglucosone. In Flavor Precursors: Thermal and En-
zymatic Conversions; ACS Symposium Series 490; Teranishi,
R., Takeoka, G. R., G u¨ ntert, M., Eds.; American Chemical
Society: Washington, DC, 1992; pp 217-231.
X
Abstract published in Advance ACS Abstracts, No-
vember 1, 1995.