Molecular products from the thermal degradation of glutamic acid

Article abstract of DOI:10.1021/jf401846t

The thermal behavior of glutamic acid was investigated in N₂ and 4% O₂ in N₂ under flow reactor conditions at a constant residence time of 0.2 s, within a total pyrolysis time of 3 min at 1 atm. The identification of the main pyrolysis products has been reported. Accordingly, the principal products for pyrolysis in order of decreasing abundance were succinimide, pyrrole, acetonitrile, and 2-pyrrolidone. For oxidative pyrolysis, the main products were succinimide, propiolactone, ethanol, and hydrogen cyanide. Whereas benzene, toluene, and a few low molecular weight hydrocarbons (propene, propane, 1-butene, and 2-butene) were detected during pyrolysis, no polycyclic aromatic hydrocarbons (PAHs) were detected. Oxidative pyrolysis yielded low molecular weight hydrocarbon products in trace amounts. The mechanistic channels describing the formation of the major product succinimide have been explored. The detection of succinimide (major product) and maleimide (minor product) from the thermal decomposition of glutamic acid has been reported for the first time in this study. Toxicological implications of some reaction products (HCN, acetonitrile, and acyrolnitrile), which are believed to form during heat treatment of food, tobacco burning, and drug processing, have been discussed in relation to the thermal degradation of glutamic acid.

Full text of DOI:10.1021/jf401846t

Article
pubs.acs.org/JAFC
Molecular Products from the Thermal Degradation of Glutamic Acid
Joshua K. Kibet, Lavrent Khachatryan, and Barry Dellinger*
Department of Chemistry, Louisiana State University, Baton Rouge, Louisiana 70803, United States
*
S Supporting Information
ABSTRACT: The thermal behavior of glutamic acid was investigated in N and 4% O in N under flow reactor conditions at a
2
2
2
constant residence time of 0.2 s, within a total pyrolysis time of 3 min at 1 atm. The identification of the main pyrolysis products
has been reported. Accordingly, the principal products for pyrolysis in order of decreasing abundance were succinimide, pyrrole,
acetonitrile, and 2-pyrrolidone. For oxidative pyrolysis, the main products were succinimide, propiolactone, ethanol, and
hydrogen cyanide. Whereas benzene, toluene, and a few low molecular weight hydrocarbons (propene, propane, 1-butene, and 2-
butene) were detected during pyrolysis, no polycyclic aromatic hydrocarbons (PAHs) were detected. Oxidative pyrolysis yielded
low molecular weight hydrocarbon products in trace amounts. The mechanistic channels describing the formation of the major
product succinimide have been explored. The detection of succinimide (major product) and maleimide (minor product) from
the thermal decomposition of glutamic acid has been reported for the first time in this study. Toxicological implications of some
reaction products (HCN, acetonitrile, and acyrolnitrile), which are believed to form during heat treatment of food, tobacco
burning, and drug processing, have been discussed in relation to the thermal degradation of glutamic acid.
KEYWORDS: flow reactor, decarboxylation, pyroglutamic acid, succinimide, maleimide
INTRODUCTION
Amino acids, from a nutritional standpoint, are perhaps the
most important constituents of our daily diet. They are the
central building blocks of proteins and have applications in the
food, agricultural, cosmetics, and pharmaceutical industries,
where pyrolysis is commonly used.
mutagenics and carcinogenics in pyrolysates of proteinaceous
foods is a health concern in the fields of food processing,
the treatment of ulcers, epilepsy, and Parkinson’s dis-
2
■
,3,23−26
ease.
Hence, it is possible degradation may affect
2,27
1
product quality and safety.
The decomposition products of amino acids are mainly
simple inorganic compounds (CO , H O, NH , and CO), with
2
2
3
2
−4
a variety of volatile organic compounds (amines, nitriles,
The formation of
17,28,29
amides, and hydrocarbons).
Simple amino acids undergo
30
decarboxylation to produce CO and alkylamine. In addition
2
5
to thermally induced loss of small, stable molecules, amino
acids have been reported to undergo dehydration to form
preservation, and safety, implying the existence of a correlation
between mutagenic activity and proteins in food that have
undergone thermal treatment during processing.
3
1,32
6
−8
dipeptides as reactive intermediates.
Degradation of
Glutamic
glutamic acid can proceed via intramolecular dehydration to
acid is present in free form in many different foods such as
32
1
,9
10
10
11
12
form lactam. This observation is consistent with this study,
where it was noted glutamic acid sublimed completely at about
300 °C to form a waxy substance, suggesting possible formation
of a lactam polymer or polyglutamic acid as the major
wheat, soybean, almond, cocoa beans, and coffee. It
has been reported that pyrolysates of glutamic acid show more
1
3,14
potent mutagenicities in the Ames test.
presence of glutamic acid in a variety of heat-treated materials
food, wood, drugs, and tobacco) makes the investigation of the
molecular products from the thermal degradation of glutamic
acid necessary. Amino acid pyrolysis is also important in
understanding hydrothermal systems and the origin of life.
Most biomass materials contain nitrogen, which can be
converted to environmentally harmful products such as HCN,
furans, and aromatic compounds.
resulting from consumption of tobacco products, for instance,
has been blamed on the production of toxic molecular products
such as acetonitrile, acrylonitrile, and HCN
radicals during tobacco burning. Radicals produced during
combustion of tobacco are very reactive and capable of causing
Accordingly, the
3
intermediate product. In the pyrolysis of amino acids, the
(
thermal conditions provide sufficient energy to form inter-
31
mediate peptides in the solid/liquid phase. Generally, the
mechanisms for amino acid pyrolysis are complicated and may
include, in addition to decarboxylation, deamination, dehy-
dration, and condensation, random bond cleavage, thermal
cracking of cylic amides, addition, substitution, recombination,
15
1
6,17
The health consequences
2
8,33,34
rearrangement, and hydrogenation reactions.
Despite the efforts made on the pyrolysis of amino acids,
1
8,19
there is considerable variation in the amount and type of
as well as free
35
2
0
products reported by various investigators. This discrepancy
has been attributed to adsorption of extremely polar fragments
either in the chromatographic column or on other parts of the
7
,8,21
oxidative stress, cancers, and reproductive health diseases.
35
analytical system. In this work, the System for Thermal
For example, the principal nitrogenous precursors present in
burley tobacco are amino acids such as glutamic acid and
aspartic acid, whereas glutamine, asparagine, and proline are
found in flue-cured tobacco.
glutamic acid is considered interesting due to its wide spectrum
of commercial applications including incorporation in drugs for
Received: April 29, 2013
Revised: July 18, 2013
Accepted: July 22, 2013
Published: July 22, 2013
1
8,22
Also, the thermal behavior of
©
2013 American Chemical Society
7696
dx.doi.org/10.1021/jf401846t | J. Agric. Food Chem. 2013, 61, 7696−7704

Journal of Agricultural and Food Chemistry
Article
Figure 1. Weight percent yields of the major products (A−D) and yields of HCN (based on GC area counts) (B) from the pyrolysis of glutamic acid
in N at 1 atm.
2
Diagnostic Studies (STDS) was used to analyze the gas-phase
reaction products from the thermal degradation of glutamic
acid. This technique has proven credible over the years for
injection port, the cryogenic trap, and the in-line detection system.
37,38
The experimental details for STDS are reported elsewhere.
3
6
Reactor and Sample Loading. A straight quartz tubular reactor
of dimensions 0.3 cm i.d. × 17.7 cm was used for the pyrolysis of
glutamic acid. Glutamic acid (30 ± 0.2 mg) was placed inside the
quartz tube and held in place by quartz wool. A residence time of 0.2 s
was chosen for all temperature runs. Tubular reactors have been in use
for many years and are generally acceptable because in addition to
withstanding high temperatures (about 1400 K), they have very small
analysis of gas-phase products that would be difficult to analyze
37,38
using routine methods.
For this reason, reaction products
(
succinimide and maleimide) that have not been detected
before are reported during the thermal degradation of glutamic
acid. Accordingly, this investigation will focus in detail on the
formation mechanism of succinimide while describing briefly
the mechanistic pathways for other major products, pyrrole,
acetonitrile, 2-pyridone, maleimide, and HCN. The toxicity
implications of some of the reaction products from the
pyrolysis of glutamic acid such as hydrogen cyanide,
acetonitrile, acrylonitrile, and pyrrole will be discussed briefly.
3
8,39
coefficients of thermal expansion.
All of the connections to the
quartz tubular flow reactor were made of silica to maintain an inert
3
9
atmosphere. Nitrogen was the carrier gas for the pyrolytic condition,
whereas 4% O in N was the carrier gas for the reactive (oxidative)
2
2
condition. The flow of the carrier gas was controlled by a digital mass
flow controller (Siera, model 810-DR-2), which has the capacity to
deliver up to 700 mL/min of gas into the reactor system.
The flow reactor effluent is transported through a transfer line
heated at 275 °C to prevent condensation along the transfer line. The
transfer line is coated with deactivated silica lined with a steel tube. In
addition, there is a splitter in the transfer line to deliver only a small
amount of sample to the GC-MS system without damage to the
detector. The splitter also helps to maintain a constant pressure of 1
atm in the reactor. This splitter is controlled by a pressure gauge,
where the excess effluent flows through a charcoal trap and out to a
fume hood.
EXPERIMENTAL PROCEDURES
■
Materials. The L-glutamic acid (≥99.5%, purity) used in this study
was purchased from Sigma-Aldrich (St. Louis, MO, USA) and used
without further treatment.
System for Thermal Diagnostic Studies. To avoid many
experimental pitfalls associated with the analysis of amino acid
17
36
pyrolysis, the STDS was used in this work. The STDS is a
continuous flow reactor system, with collection of the reactor effluent
with an in-line GC-MS at the head of the GC column at −60 °C. It is a
well-established technique for the analysis of a wide range of organic
Fractional Pyrolysis and Fractional Oxidative Pyrolysis. The
fractional pyrolysis technique applied in this study is an experimental
procedure in which the same sample is continuously pyrolyzed at each
36
4
materials including biomass components. STDS contains various
units: the reactor compartment, which houses the electrical heater and
the reactor; the temperature control console, which controls the
pyrolysis temperature within a temperature gradient of ±5 °C; the
pyrolysis temperature. It is defined as a selective in situ conversion of
4,40,41
biomass materials to desired products.
The fractional pyrolysis
technique offers some advantages in comparison with conventional
pyrolysis because only one loading of sample is used and it can be
7
697
dx.doi.org/10.1021/jf401846t | J. Agric. Food Chem. 2013, 61, 7696−7704

Journal of Agricultural and Food Chemistry
Article
Figure 2. Weight percent yields of major products from the pyrolysis of glutamic (A, C, and D) and yields of other major products (based on GC
area counts) (B) from the pyrolysis of glutamic acid in 4% O in N at 1 atm.
2
2
heated multiple times, therefore allowing the intermediate neutral, but
unstable, products to be collected before they disappear in the
secondary processes. The details of this procedure are described in our
pyrolysis products for which standards were not available, the peak
area count obtained from integration of respective total ion
chromatograms (TIC) was plotted as a function of temperature to
determine their yield distribution over the entire pyrolysis temperature
range. The percent yield of each calibrated product was evaluated
using eq 1. After the compounds were calibrated, product distribution
curves displaying the yield of various products with pyrolysis
temperature were generated.
4
previous paper.
Gas Chromatography−Mass Spectrometry (GC-MS) Charac-
terization of Molecular Products. The GC-MS analysis of the
pyrolysate was conducted with an Agilent 6890N gas chromatograph
equipped with a 5973N mass selective detector (MSD) with an ion
source of electron impact (EI) at 70 eV. Two GC columns, a Gas-pro
column (60 m × 0.32 mm i.d. × 0.25 μm) for the analysis of low
molecular weight products and a DB5-MS column (30 m × 0.25 mm
⎛
Y = ⎜
⎝
⎞
wt of product, w
wt of sample, W ⎠
⎟ × 100
×
0.25 μm) for the determination of high molecular weight products,
(1)
41
were used. To analyze small hydrocarbons, a gas sampling valve was
used in place of a cold trap. A detailed description of the procedure for
characterization of molecular products has been presented elsewhere.
Y is the yield of the pyrolysis product in wt %.
4
The pyrolysis products were identified using the NIST library.
Additionally, standards for pure compounds were used to compare the
retention times and mass spectral fragmentation pattern with those of
the analyte, and the mass spectral hits were excellent. Each of the
compounds identified was thoroughly checked against literature data
to ensure the correct reaction product was reported. Accordingly,
critical emphasis has been given to those products that can easily be
correlated with the structure of the starting material (glutamic acid).
RESULTS AND DISCUSSION
■
Fractional Pyrolysis. A series of nitrogen-containing
products as well as hydrocarbon products was formed during
pyrolysis of glutamic acid in an inert atmosphere. The reaction
products from the pyrolysis of glutamic acid can be grouped on
the basis of their maximum release temperatures (cf. Figure 2).
Group 1: The maximum release temperature for these
products was between 300 and 400 °C with succinimide as the
major product, peaking at about 350 °C (cf. Figure 1A). The
concentration of compounds in this class peak at low pyrolysis
38
The experimental results were averaged of replicates.
Calibration of Molecular Products. Standards for most reaction
products were purchased from Sigma-Aldrich. Standards of purity
≥99% were used for the calibration of pyrolysis products. For those
7
698
dx.doi.org/10.1021/jf401846t | J. Agric. Food Chem. 2013, 61, 7696−7704

Journal of Agricultural and Food Chemistry
Article
Figure 3. GC-MS chromatogram (DB5-MS column) for pyrolysis (red line) and oxidative pyrolysis (blue line) of glutamic acid in N and 4% O in
2
2
N at 400 °C.
2
Figure 4. Major product distribution from the thermal degradation of glutamic acid in N and 4% O in N at 1 atm in the temperature region 300−
2
2
2
600 °C.
temperature and decrease sharply as the pyrolysis temperature
is increased, implying a short release temperature range.
Group 2: This category of compounds included pyrrole,
hydrogen cyanide (HCN), propionitrile, and acrylonitrile and
generally peaked at about 450 °C (cf. Figure 1B) (the yield of
HCN is depicted in GC area counts, the scale on the right). It is
notable that amino acids are capable of forming a relatively
stable nitrogen−aromatic ring in the early stages of thermolysis,
1C). Whereas the maximum release temperature for com-
pounds in group 2 was about 450 °C, the maximum release
temperature for those in group 3 was slightly higher than 450
°C, with the major compounds being 2-methylpyrrole and
allylcyanide.
Group 4: These products include acetonitrile and 2-
pyrrolidone as the major products (cf. Figure 1D). Acetonitrile
peaked at high pyrolysis temperatures and may be formed from
thermal decomposition of succinimide, pyrrole, and other
heterocyclic products such as indole. This observation can be
noted from Figure 1A,C,D, which shows the concentration of
acetonitrile increased sharply at high temperatures as the
concentrations of succinimide, 2-pyridone, and pyrrole
decreased. The concentration of compounds in this category
reached a maximum at about 500 °C.
42
yielding large amounts of hydrogen cyanide. Nitrogen-
containing rings are known to break down at high temperatures
42−44
to give high levels of hydrogen cyanide.
Previous pyrolysis
of intermediates such as pyrrolidone and 2-pyrodone led to
44
observation of high yields of HCN. This suggests the
concentration of HCN increases sharply at high temperatures
as the concentrations of succinimide, 2-pyrrolidone, 2-pyridone,
and pyrrole decrease.
Generally, the principal products in order of decreasing
abundance in fractional pyrolysis of glutamic acid were as
follows: succinimide, pyrrole, acetonitrile, 2-pyrrolidone, 3H-
Group 3: The reaction products in this group have
similarities to the products discussed in group 2 (cf. Figure
7
699
dx.doi.org/10.1021/jf401846t | J. Agric. Food Chem. 2013, 61, 7696−7704

Journal of Agricultural and Food Chemistry
Article
pyrrolo[2,3-d]pyrimidin-4(7H)-one, 2-methyl-1H-pyrrole, 2-
pyridone, and maleimide.
Fractional Oxidative Pyrolysis. The principal reaction
products from oxidative pyrolysis of glutamic acid can be
classified into three groups.
acetonitrile) were formed at higher temperatures (cf. Figures
1B−D and 2B).
Decomposition Profile for Glutamic Acid. The decom-
position profiles for glutamic acid for both pyrolysis and
oxidative experiments were similar (cf. Figure 5). Conse-
Group 1: The major product in this class of compounds was
exclusively succinimide, which peaked at about 350 °C. Other
products in this group included pyrrole and acetaldehyde (cf.
Figure 2A). Pyrrole, a major product in pyrolysis, was formed in
low amounts under oxidative pyrolysis because an oxidizing
atmosphere may retard the formation of pyrrole while
4
2,45
enhancing the release of CO according to the literature.
2
Previously, intramolecular reactions involving α-lactone fol-
lowed by decarbonylation were proposed to account for the
4
6,47
observed aldehydes,
although aldehydes (acetaldehyde)
were observed as minor products in our experiments (detected
only under oxidative pyrolysis).
Group 2: Many of the reaction products detected in this
group were mainly oxygenated products, with the major
product being propiolactone (reached maximum concentration
at about 400 °C) (cf. Figure 2B). An analogous reaction in the
presence of water suggested direct deamination occurs via an
Figure 5. Weight percent yield of glutamic acid char as a function of
temperature.
quently, glutamic acid appears to exhibit a single decomposition
regime, starting at 300 °C and ending at 600 °C. The highest
rate of decomposition for oxidative pyrolysis was realized
between 300 and 350 °C with a mass loss of 22.4%, whereas the
highest rate of decomposition for pyrolysis was achieved
between 400 and 450 °C with a mass loss of 21.4%. At the end
of the experiment (600 °C), the mass losses for pyrolysis and
oxidative pyrolysis were 75.4 and 81.3%, respectively. This
suggests that glutamic acid has a higher residue content than
internal S 2 mechanism, yielding ammonia and propiolac-
N
4
8
tone. Ethanol and acetic acid were the other oxygenated
products observed in this group.
Group 3 comprised the products that are formed above 450
°C and includes 5-methylpyridine, acetonitrile, and maleimide
as the principal products (cf. Figure 2C). These products
appear to be formed from the thermal decomposition of major
products such as succinimide. Early studies postulated pyrolysis
49
of succinimide yielded mainly CO, H O, and acetonitrile.
2
16,50
most biomass materials such as tyrosine and cellulose
This may suggest a secondary route for the formation of
acetonitrile. Choud et al. proposed an activation energy of 52
An observation of glutamic acid after heat treatment revealed
a waxy substance (probably polyglutamic acid or diketopiper-
azine) that stuck to the walls of the reactor. Accordingly, this
would imply the gas−solid interface changes during heat
treatment and any pores present in the sample disappear so that
49
kcal/mol for the ring-opening of succinimide. Subsequently,
the ring-opening of succinimide facilitates its decomposition to
other poducts including acetonitrile and HCN.
Succinimide was the most abundant product, contributing
oxygen acts only on the surface but does not penetrate into the
>40% of the total products quantified from oxidative pyrolysis.
50,51
matrix of the (polymer) sample.
Thus, the degradation of
The order of abundance for the major reaction products from
the thermal degradation of glutamic acid can be seen clearly in
the bar graph of Figure 4.
As can be observed from Figure 3 (overlay spectra for
pyrolysis and oxidative pyrolysis at 400 °C), pyrolysis and
oxidative pyrolysis yielded similar reaction products of different
intensities. It is clear from the chromatograms (cf. Figure 3)
that although some products were favored by an inert regime
50
glutamic acid is independent of oxidative reactions.
Consequently, we believe the mass loss due to an oxidizing
and an inert environment will certainly not change significantly
in the entire temperature range. For comparison, it was
observed that the mass loss of cellulose below 300 °C was due
to oxidative reactions, but at temperatures above 300 °C the
rate of pyrolysis was essentially the same in both air and
nitrogen, indicating thermal degradation is independent of
(
N ), some were favored by a reactive regime (oxidative).
2
50,51
oxidative reactions.
This observation is remarkable and
Therefore, a comparison between three major compounds,
pyrrole, succinimide, and 3H-pyrrolo[2,3-d]pyrimidin-4(7H)-
one, revealed interesting results. Under pyrolysis at 400 °C, for
instance, 3H-pyrrolo[2,3-d]pyrimidin-4(7H)-one and pyrrole
were exclusively the major products, whereas for oxidative
pyrolysis at the same temperature, succinimide was the
principal component.
agrees well with observations made during the thermal
degradation of glutamic acid.
Mechanistic Description. The major reactions for
glutamic acid are dehydration, decarboxylation, and deami-
1
8
nation. Glutamic acid contains two acidic (−CO H) groups
2
and one basic (−NH ) group, which can react to form large
molecules at high temperature.
2
3
,52
Therefore, different func-
Succinimide was the dominant reaction product from the
oxidative pyrolysis of glutamic acid according to Figure 4. The
concentration of succinimide was 4 times higher in oxidative
pyrolysis compared to pyrolysis. On the other hand, pyrrole
was 3 times higher in pyrolysis than in oxidative pyrolysis.
Whereas most of the reaction products (succinimide, pyrrole,
and 2-pyrrolidone) appear to be formed at low temperature (cf.
Figures 1A and 2A), other reaction products (maleimide and
tional groups are expected to have different pyrolysis
characteristics as well as to give different pyrolysis products.
53
For example, at low temperatures (about 300 °C) low
molecular weight heterocyclic compounds are formed, whereas
at high temperatures (≥500 °C), polycyclic aromatic hydro-
31
carbons are expected to form.
Decarboxylation Reactions for Amino Acids. The
mechanistic considerations for pyrolysis experiments of amino
7
700
dx.doi.org/10.1021/jf401846t | J. Agric. Food Chem. 2013, 61, 7696−7704

Journal of Agricultural and Food Chemistry
Article
3
,16,34,46,47,54,55
acids have been extensively studied.
More
experimentally by Sharma et al. that direct degradation of
31
recently, a statistical mechanical investigation (QM/MM) has
shown that the most likely pathway for decomposition of amino
acids occurs via direct decarboxylation, where CO departure is
the first as well as the rate-determining step. The use of C-
labeled amino acids provides excellent evidence that decarbox-
ylation is the predominant decomposition pathway for amino
acids. Decarboxylation reactions of amino acids yield an
amine as the major product and clearly constitute the major
decomposition pathway as evidenced from the yield of CO2.
Mechanistic Channel for Formation of Succinimide
and Maleimide. The major product during fractional pyrolysis
as well fractional oxidative pyrolysis was succinimide,
surprisingly nondetected in previous studies. Our studies,
however, have since shown that pyrolysis of glutamic acid in N2
and 4% O2 in N2 would in fact yield succinimide and
maleimide. Maleimide was formed in much lower amounts than
succinimide. The yields of maleimide did not change
significantly when the pyrolysis environment was changed
from N to 4% O in N (Figures 1D and 2). Succinimide was
exclusively the major product under oxidative pyrolysis,
contributing about 40% of the products analyzed, whereas
under pyrolysis it contributed >20% of the products analyzed.
Whereas previously succinimide has been detected from the
thermal degradation of aspartic acid, asparagine, and glutamine,
no succinimide has been detected from the thermal degradation
of glutamic acid. The mechanistic channel for succinimide
and maleimide formation from aspartic and asparagine is
known, however; succinimide and maleimide formation from
glutamine and glutamic acid is not yet understood.
Succinimide and maleimide, nevertheless, are structures that
are characteristic of amino acids with additional carboxylic or
amino functional groups. Accordingly, we propose the
formation of succinimide proceeds via an intermediate, tricyclic
diketopiperazine (DKP), which is a highly unstable inter-
mediate evidenced in the literature. We hypothesize that DKP
would eventually decompose to succinimide and 2-pyridone in
the presence of hydrogen atoms pool through the reactions of
the addition of H to DKP and further disproportionation
reactions (cf. Scheme 1). From this pathway, maleimide
appears to be formed from the dehydrogenation of succinimide
and may explain why succinimide was formed in higher
concentrations than maleimide. It was previously proven
succinmide yielded maleimide.
The formation of DKP requires two molecules of
31
pyroglutamic acid. This product forms in trace and significant
amounts in pyrolysis and oxidative pyrolysis, respectively
(Figure 3). Peptide-forming reactions occur readily because
2
48
14
31
they involve simple dehydration reactions. DKPs also form
3
5
15
promptly upon heat treatment of proteinaceous foods. In
35
principle, peptides will be nondetectable intermediates in
amino acid pyrolysis because of their high thermal reactivity
and low volatility, which keeps them in the thermal zone until
47
31
they react further. Consequently, we postulate DKP thermally
degrades to succinimide and 2-pyridone (cf. Scheme 1). 2-
Pyridone was one of the major products from the thermal
degradation of glutamic acid under pyrolysis but a minor
product under oxidative pyrolysis. This is because, although an
oxidizing environment accelerates the formation of pyrolysis
products, it may also oxidize certain reaction products into
31
water, CO, or CO and subsequently decrease their yields.
2
Scheme 1 predicts equal amounts of succinimide and 2-
pyridone. Experimentally it was found the yields of succinimide
prevailed over the yields of 2-pyridone by a factor of 2. This
may be possible because 2-pyridone is susceptible to hydro-
genation in the presence of an abundant pool of hydrogen
atoms in pyrolysis and char. Also, the high polarizability of 2-
2
2
2
5
6
pyridone in comparison with succinimide (symmetric
31
57
structure) enhances its absorptivity on the GC column and
consequently minimizes its detection. The marked difference in
yields between succinimide and 2-pyridone is evident in
oxidative pyrolysis. 2-Pyridone may be easily oxidized by
oxygen or most importantly by abundant hydroxyl radicals
31
(characteristic of oxidative processes) into water, CO, or CO
2
3
and subsequently decrease its yield, making it a minor product.
The key finding during oxidation is that the yield of
succinimide increases rapidly, predominating over all other
product yields significantly (Figure 4). Pyroglutamic acid, an
important precursor for the formation of succinimide, was
observed experimentally in detectable amounts in oxidative
conditions (Figure 3, blue chromatogram at RT ∼ 17.3 min)
but in trace amounts under pyrolytic conditions. As an
important intermediate product, pyroglutamic acid may form
from internal cyclization of glutamic acid via dehydration
31
2
,58,59
processes.
Dehydration is a very common reaction for
amino acids and may occur in the gas phase through a four-
centered concerted mechanism as presented in the Supporting
Information provided with this paper. It is expected that
hydroxyl radicals during oxidative pyrolysis will facilitate this
concerted mechanism toward the formation of pyroglutamic
acid (by increasing the polarizability of the hydroxyl O−H
bond in the carboxylic group).
Scheme 1. Formation of Diketopiperazine (DKP) from
Pyroglutamic Acid and of Succinimide and Maleimde from
the Unstable DKP Intermediate
Mechanistic Pathways for the Formation of Pyrrole
and Methylated Pyrroles. The thermal degradation of
glutamic acid appears to be a major process in the
pyrosynthesis of not only succinimide but also pyrroles
(
pyrrole, 2-methylpyrrole, 2,4-dimethylpyrrole, and 2,5-di-
methylpyrrole). The production of pyrrole from glutamic acid
clearly indicates one carboxyl group is lost as carbon dioxide
(
cf. Scheme 2), whereas the second carboxyl group is
incorporated into the 2-pyrrrolidone ring before converting
4
2,43
to pyrrole.
In this study, we propose the precursor for the
formation of pyrrole is pyroglutamic acid through the formation
of 2-pyrrolidone. Pyrrole ultimately undergoes methylation to
7
701
dx.doi.org/10.1021/jf401846t | J. Agric. Food Chem. 2013, 61, 7696−7704

Journal of Agricultural and Food Chemistry
Article
Scheme 2. Proposed Mechanism for the Formation of
Pyroglutamic Acid, 2-Pyrrolidone, Pyrrole, and Methylated
Pyrroles
minor product resulting from the dehydrogenation of
succinimide as predicted in the literature.
4
2,43
ASSOCIATED CONTENT
■
*
S
Supporting Information
Hydrocarbon products from pyrolysis. This material is available
free of charge via the Internet at http://pubs.acs.org.
AUTHOR INFORMATION
Corresponding Author
(B.D.) E-mail: barryd@lsu.edu. Phone: +1 (225) 578-6759.
■
*
Funding
We thank R. J. Reynolds for funding this study.
Notes
The authors declare no competing financial interest.
REFERENCES
■
(
1) Elahi, M.; Khan, N. Free amino-acids of Pakistani wheat varieties.
J. Agric. Food Chem. 1973, 21, 743−744.
2) Wu, H.; Reeves-McLaren, N.; Jones, S.; Ristic, R. I.; Fairclough, J.
(
P. A.; West, A. R. Phase transformations of glutamic acid and its
decomposition products. Crystal Growth Des. 2010, 10, 988−994.
(3) Chiavari, G.; Galletti, G. C. Pyrolysis-gas chromatography mass-
spectrometry of amino-acids. J. Anal. Appl. Pyrolysis 1992, 24, 123−
form 2-methylpyrrole and 2,5-dimethylpyrrole as indicated in
Scheme 2.
1
37.
It was proposed previously that pyrrole decomposes to yield
60
(4) Kibet, J. K.; Khachatryan, L.; Dellinger, B. Molecular products
predominantly HCN in addition to hydrocarbon products.
from the pyrolysis and oxidative pyrolysis of tyrosine. Chemosphere
013, 91, 1026−1034.
̌
́ ́
5) Djilas, S. M.; Milic, B. L.; Canadanovic-Brunet, J. M. ESR spectral
This implies the pyrrole yield peaks at lower temperatures,
whereas that of HCN and hydrocarbons (propane, propene,
etc.) is expected to peak at high temperatures, remarkably
consistent with the results presented in this study.
2
(
study of free radical formation during the pyrolysis of hydroxyamino
acids. J. Anal. Appl. Pyrolysis 1994, 28, 157−161.
Toxicological Consequences of Molecular Products.
(6) Kanada, T.; Mazaki, M.; Taue, S.; Uyeta, M. Mutagenicity of food
Whereas succinimide, 2-pyrrolidone, and maleimide have no
pyrolysates by ames test. Mutat. Res. 1978, 54, 239−240.
(7) Talhout, R.; Schulz, T.; Florek, E.; van Benthem, J.; Wester, P.;
Opperhuizen, A. Hazardous compounds in tobacco smoke. Int. J.
Environ. Res. Public Health 2011, 8, 613−628.
6
1,62
known toxicological impacts,
HCN, acetonitrile, acryloni-
trile, and benzene are known to be precursors for health-related
6
3,64
problems.
The biological activity of HCN has mostly been
(8) Fujioka, K.; Shibamoto, T. Determination of toxic carbonyl
correlated with the inhibition of the terminal cytochrome
oxidase in the mitochondrial respiratory pathway and attacks
compounds in cigarette smoke. Environ. Toxicol. 2006, 21, 47−54.
(9) Del Molino, I. M. M.; Rojo, B.; Martinez-Carrasco, R.; Perez, P.
amino-acid composition of wheat-grain. 1. Changes during develop-
ment. J. Sci. Food Agric. 1988, 42, 29−37.
63
the nervous system and thyroid glands. Acrylonitrile is known
to cause respiratory effects, whereas benzene is an established
6
5,66
human carcinogen.
Acetonitrile (methyl cyanide) upon
(10) Wolf, W. J. Gel-electrophoresis and amino-acid-analysis of the
ingestion can lead to mental abnormalities, lethargy, tightness
nonprotein nitrogen fractions of defatted soybean and almond meals.
Cereal Chem. 1995, 72, 115−121.
6
7,68
of the chest, and vomiting.
In the gas phase, acetonitrile has
a very high ability to convert to the highly toxic compounds
(11) Hurst, W. J.; Martin, R. A. High-performance liquid chromato-
graphic determination of ochratoxin A in artificially contaminated
cocoa beans using automated sample clean-up. J. Chromatogr., A 1998,
810, 89−94.
68
hydrogen cyanide and cyanogen. Additionally, acetonitrile is
an Environmental Protection Agency (EPA)-classified toxic
6
8
compound. There is no clear record in the literature that
pyrrole is toxic; nevertheless, it is an important precursor for
the formation of the toxic compounds hydrogen cyanide and
(12) Arnold, U.; Ludwig, E. Analysis of free amino acids in green
coffee beans. 2. Changes of the amino acid content in arabica coffees in
connection with post-harvest model treatment. Z. Lebensm.-Unters.
6
0
acetonitrile. The toxicological effects of some of the molecular
products analyzed in this study, including those that were not
analyzed but believed to be formed (ammonia, HNCO, NO(x),
and CO), are critical for understanding biomass burning,
cigarette smoking, and high-temperature cooking of proteina-
ceous foods.
-
Forsch. 1996, 203, 379−384.
(
13) Ames, B. N.; McCann, J.; Yamasaki, E. Carcinogens are
mutagens − simple test system. Mutat. Res. 1975, 33, 27−28.
(14) Ames, B. N.; McCann, J.; Yamasaki, E. Methods for detecting
carcinogens and mutagens with salmonella-mammalian-microsome
mutagenicity test. Mutat. Res. 1975, 31, 347−363.
(15) Li, J.; Wang, X. G.; Klein, M. T.; Brill, T. B. Spectroscopy of
This study proposes that the major product succinimide from
pyrolysis and oxidative pyrolysis of glutamic acid may actually
be formed from the thermal degradation of the intermediate
peptide, a tricyclic diketopiperazine (DKP). We conclude
pyroglutamic acid is an important intermediate product from
the thermal degradation of glutamic acid, which polymerizes to
diketopiperazine before ultimately decomposing to high yields
of succinimide, especially during oxidative pyrolysis. Maleimide,
which was also detected for the first time, is most probably a
hydrothermal reactions, 19: pH and salt dependence of decarbox-
ylation of alpha-alanine at 280−330 °C in an FT-IR spectroscopy flow
reactor. Int. J. Chem. Kinetics 2002, 34, 271−277.
(16) Li, J.; Liu, Y. W.; Shi, J. Y.; Wang, Z. Y.; Hu, L.; Yang, X.; Wang,
C. X. The investigation of thermal decomposition pathways of
phenylalanine and tyrosine by TG-FTIR. Thermochim. Acta 2008, 467,
2
0−29.
(17) Chiavari, G.; Fabbri, D.; Prati, S. Gas chromatographic-mass
spectrometric analysis of products arising from pyrolysis of amino
7
702
dx.doi.org/10.1021/jf401846t | J. Agric. Food Chem. 2013, 61, 7696−7704

Journal of Agricultural and Food Chemistry
Article
acids in the presence of hexamethyldisilazane. J. Chromatogr., A 2001,
(39) Hair, M. L. Hydroxyl-groups on silica surface. J. Non−Cryst.
Solids 1975, 19, 299−309.
9
(
22, 235−241.
18) Sharma, R. K.; Chan, W. G.; Hajaligol, M. R. Product
(40) Agblevor, F. A.; Beis, S.; Mante, O.; Abdoulmoumine, N.
Fractional catalytic pyrolysis of hybrid poplar wood. Ind. Eng, Chem.
Res. 2010, 49, 3533−3538.
compositions from pyrolysis of some aliphatic α-amino acids. J. Anal.
Appl. Pyrolysis 2006, 75, 69−81.
(
19) Matsubara, K.; Nakato, T.; Tomida, M. H-1 and C-13 NMR
(41) Kibet, J.; Khachatryan, L.; Dellinger, B. Molecular products and
radicals from pyrolysis of lignin. Environ. Sci. Technol. 2012, 46,
12994−13001.
(42) Haidar, N. F.; Patterson, J. M.; Moors, M.; Smith, W. T. Effects
of structure on pyrolysis gases from amino-acids. J. Agric. Food Chem.
characterization of poly(succinimide) prepared by thermal poly-
condensation of L-aspartic acid. Macromolecules 1997, 30, 2305−2312.
(
20) Maskos, Z.; Khachatryan, L.; Dellinger, B. Formation of the
persistent primary radicals from the pyrolysis of tobacco. Energy Fuels
008, 22, 1027−1033.
21) Ghosh, A.; Choudhury, A.; Das, A.; Chatterjee, N. S.; Das, T.;
2
(
1
(
981, 29, 163−165.
43) Patterson, J. M.; Haidar, N. F.; Papadopo, Ep.; Smith, W. T.
Chowdhury, R.; Panda, K.; Banerjee, R.; Chatterjee, I. B. Cigarette
smoke induces p-benzoquinone-albumin adduct in blood serum:
implications on structure and ligand binding properties. Toxicology
Pyrolysis of phenylalanine, 3,6-dibenzyl-2,5-piperazinedione, and
phenethylamine. J. Org. Chem. 1973, 38, 663−666.
(44) Johnson, W. R.; Kang, J. C. Mechanisms of hydrogen cyanide
2
(
012, 292, 78−89.
22) Stedman, R. L. Chemical composition of tobacco and tobacco
smoke. Chem. Rev. 1968, 68, 153−207.
23) Chen, S. H.; Lei, H. L.; Huang, L. R.; Tsai, L. H. Protective
formation from pyrolysis of amino acids and related compounds. J.
Org. Chem. 1971, 36, 189−192.
(45) Patterson, J. M.; Haidar, N. F.; Shiue, C. Y.; Smith, W. T.
(
Thermally induced side-chain to ring migrations in aromatic systems. J.
Org. Chem. 1973, 38, 3052−3054.
effect of excitatory amino acids on cold-restraint stress-induced gastric
ulcers in mice − role of cyclic nucleotides. Dig. Dis. Sci. 2001, 46,
(46) Ratcliff, M. A.; Medley, E. E.; Simmonds, P. G. Pyrolysis of
2
(
285−2291.
amino-acids − mechanistic considerations. J. Org. Chem. 1974, 39,
481−1490.
47) Simmonds, P. G.; Shulman, G. P.; Ratcliff, M. A.; Medley, E. E.
24) Lloyd, K. G.; Hornykie, O. L-Glutamic acid decarboxylase in
Parkinsons disease − effect of L-dopa therapy. Nature 1973, 243, 521−
23.
25) Lloyd, K. G.; Hornykiewicz, O.; Davidson, L.; Shannak, K.;
1
(
5
(
Thermal-decomposition of aliphatic monoamino-monocarboxylic
acids. Anal. Chem. 1972, 44, 2060−2066.
Farley, I.; Goldstein, M.; Shibuya, M.; Kelley, W. N.; Fox, I. H.
Biochemical-evidence of dysfunction of brain neurotransmitters in the
Lesch-Nyhan syndrome. N. Engl. J. Med. 1981, 305, 1106−1111.
(48) Alexandrova, A. N.; Jorgensen, W. L. On the mechanism and
rate of spontaneous decomposition of amino acids. J. Phys. Chem. B
011, 115, 13624−13632.
49) Choudhar, G.; Cameron, A. M.; Back, R. A. Photolysis and
pyrolysis of succinimide vapor. J. Phys. Chem. 1968, 72, 2289−2292.
50) Mamleev, V.; Bourbigot, S.; Yvon, J. Kinetic analysis of the
2
(
(
26) Lloyd, K. G.; Shibuya, M.; Hidaka, H.; Hornykiewicz, O.
Neurochemical alterations in the basal ganglia in the lesch-nyhan
syndrome. Can. J. Neurol. Sci. 1982, 9, 278−279.
(
(
27) Bernstein, J.; Henck, J. O. Disappearing and reappearing
thermal decomposition of cellulose: the main step of mass loss. J. Anal.
Appl. Pyrolysis 2007, 80, 151−165.
polymorphs − an anathema to crystal engineering? Mater. Res. Bull.
998, 119−128.
28) Li, J.; Wang, Z. Y.; Yang, X.; Hu, L.; Liu, Y. W.; Wang, C. X.
1
(
(51) Mamleev, V.; Bourbigot, S.; Le Bras, M.; Yvon, J.; Lefebvre, J.
Model-free method for evaluation of activation energies in modulated
thermogravimetry and analysis of cellulose decomposition. Chem.. Eng.
Sci. 2006, 61, 1276−1292.
Decomposing or subliming? An investigation of thermal behavior of L-
leucine. Thermochim. Acta 2006, 447, 147−153.
(
29) Ren, Q. Q.; Zhao, C. S.; Chen, X. P.; Duan, L. B.; Li, Y. J.; Ma,
(
52) Choi, S. S.; Ko, J. E. Analysis of cyclic pyrolysis products formed
from amino acid monomer. J. Chromatogr., A 2011, 1218, 8443−8455.
53) Hansson, K. M.; Samuelsson, J.; Amand, L. E.; Tullin, C. The
C. Y. NO(x) and N(2)O precursors (NH(3) and HCN) from biomass
pyrolysis: co-pyrolysis of amino acids and cellulose, hemicellulose and
lignin. Proc. Combustion Inst. 2011, 33, 1715−1722.
(
temperature’s influence on the selectivity between HNCO and HCN
from pyrolysis of 2,5-diketopiperazine and 2-pyridone. Fuel 2003, 82,
(
30) Bada, J. L. Amino-acid cosmogeochemistry. Philos. Trans. R. Soc.
London Ser. B−Biol. Sci. 1991, 333, 349−358.
2
(
163−2172.
(
31) Sharma, R. K.; Chan, W. G.; Wang, J.; Waymack, B. E.; Wooten,
54) Basiuk, V. A.; Douda, J. Pyrolysis of simple amino acids and
J. B.; Seeman, J. I.; Hajaligol, M. R. On the role of peptides in the
nucleobases: survivability limits and implications for extra-terrestrial
pyrolysis of amino acids. J. Anal. Appl. Pyrolysis 2004, 72, 153−163.
delivery. Planetary Space Sci. 1999, 47, 577−584.
(
32) Povoledo, D.; Vallentyne, J. R. Thermal reaction kinetics of the
(55) Basiuk, V. A.; Douda, J. Pyrolysis of poly-glycine and poly-L-
glutamic acid-pyroglutamic acid system in water. Geochim. Cosmochim.
Acta 1964, 28, 731−734.
alanine: analysis of less-volatile products by gas chromatography/
Fourier transform infrared spectroscopy/mass spectrometry. J. Anal.
Appl. Pyrolysis 2000, 55, 235−246.
(
33) Tudorachi, N.; Chiriac, A. R. TGA/FTIR/MS study on thermal
decomposition of poly(succinimide) and sodium poly(aspartate).
(56) Hammes, G. G.; Lillford, P. J. Kinetic and equilibrium study of
Polym. Testing 2011, 30, 397−407.
the hydrogen bond dimerization of 2-pyridone in hydrogen bonding
(
34) Basiuk, V. A. Pyrolysis of valine and leucine at 500 °C:
solvents. J. Am. Chem. Soc. 1970, 92, 7578−7585.
identification of less-volatile products using gas chromatography
Fourier transform infrared spectroscopy mass spectrometry. J. Anal.
Appl. Pyrolysis 1998, 47, 127−143.
(57) Cowley, E. G.; Partington, J. R. Studies in dielectric polarisation.
Part XV. The dipole moments of five-membered nitrogen ring
compounds: indole, skatole, carbazole, isatin, phthalimide, and
succinimide. J. Chem. Soc. 1936, 47−50.
(
35) Vollmin, J.; Kriemler, P.; Omura, I.; Seibl, J.; Simon, W.
Structural elucidation with a thermal fragmentation-gas chromatog-
raphy-mass spectrometry combination. Microchem. J. 1966, 11, 73−86.
(58) Nunes, R. S.; Cavalheiro, E. T. G. Thermal behavior of glutamic
acid and its sodium, lithium and ammonium salts. J. Therm. Anal.
Calorim. 2007, 87, 627−630.
(
36) Rubey, W. A.; Grant, R. A. Design aspects of a modular
instrumentation system for thermal diagnostic studies. Rev. Sci.
Instrum. 1988, 59, 265−269.
(59) Vlase, T.; Vlase, G.; Doca, M.; Doca, N. Specificity of
decomposition of solids in non-isothermal conditions. J. Therm. Anal.
Calorim. 2003, 72, 597−604.
(
37) Dellinger, B.; Rubey, W. A.; Hall, D. L.; Graham, J. L.
Incinerability of hazardous wastes. Hazard. Waste Hazard. Mater.
986, 3, 139−150.
38) Rubey, W. A.; Carnes, R. A. Design of a tubular reactor
1
(
(60) Blank, D. A.; North, S. W.; Lee, Y. T. The ultraviolet
photodissociation dynamics of pyrrole. Chem. Phys. 1994, 187, 35−47.
(61) Stadler, R. H.; Verzegnassi, L.; Varga, N.; Grigorov, M.; Studer,
A.; Riediker, S.; Schilter, B. Formation of vinylogous compounds in
instrumentation assembly for conducting thermal-decomposition
studies. Rev. Sci. Instrum. 1985, 56, 1795−1798.
7
703
dx.doi.org/10.1021/jf401846t | J. Agric. Food Chem. 2013, 61, 7696−7704

Journal of Agricultural and Food Chemistry
Article
model Maillard reaction systems. Chem. Res. Toxicol. 2003, 16, 1242−
1250.
(
62) Lin, D.; Saleh, S.; Liebler, D. C. Reversibility of covalent
electrophile − protein adducts and chemical toxicity. Chem. Res.
Toxicol. 2008, 21, 2361−2369.
(
63) Siegien, I.; Bogatek, R. Cyanide action in plants − from toxic to
regulatory. Acta Physiol. Plant. 2006, 28, 483−497.
64) Khan, H. A.; Alhomida, A. S.; Arif, I. A. On the mechanism of
nitriles toxicity. Toxicol. Sci. 2009, 110, 246−248.
65) Adlkofer, F.; Scherer, G.; Conze, C.; Angerer, J.; Lehnert, G.
(
(
Significance of exposure to benzene and other toxic compounds
through environmental tobacco-smoke. J. Cancer Res. Clin. Oncol.
1
(
990, 116, 591−598.
66) Smith, C. J.; Hansch, C. The relative toxicity of compounds in
mainstream cigarette smoke condensate. Food Chem. Toxicol. 2000, 38,
37−646.
67) Geller, R. J.; Ekins, B. R.; Iknoian, R. C. Cyanide toxicity from
acetonitrile-containing false nail remover. Am. J. Emergency Med. 1991,
, 268−270.
68) Chaban, V. V.; Prezhdo, O. V. How toxic are ionic liquid/
acetonitrile mixtures? J. Phys. Chem. Lett. 2011, 2, 2499−2503.
6
(
9
(
7
704
dx.doi.org/10.1021/jf401846t | J. Agric. Food Chem. 2013, 61, 7696−7704