Isothermal Fourier transform infrared microspectrosopic studies on the stability kinetics of solid-state intramolecular cyclization of aspartame sweetener

Article abstract of DOI:10.1021/jf990595l

A novel Fourier transform infrared (FT-IR) microspectrophotometer equipped with differential scanning calorimetry (DSC) was used to investigate the kinetics of intramolecular cyclization of aspartame (APM) sweetener in the solid state under isothermal conditions. The thermal-dependent changes in the peak intensity of IR spectra at 1543, 1283, and 1259 cm^-1 were examined to explore the reaction. The results support that the intramolecular cyclization process in APM proceeded in three steps: the methoxyl group of ester was first thermolyzed to release methanol, then an acyl cation was attacked by the lone pair of electrons available on nitrogen by an S(N)1 pathway, and finally ring-closure occurred. The intramolecular cyclization of APM determined by this microscopic FT-IR/DSC system was found to follow zero- order kinetics after a brief induction period. The bond cleavage energy (259.38 kJ/mol) of thermolysis for the leaving group of -OCH₃, the bond conversion energy (328.88 kJ/mol) for the amide II NH band to DKP NH band, and the CN bond formation energy (326.93 kJ/mol) of cyclization for the DKP in the APM molecule were also calculated from the Arrhenius equation. The total activation energy of the DKP formation via intramolecular cyclization was 261.33 kJ/mol, calculated by the above summation of the bond energy of cleavage, conversion, and formation, which was near to the value determined by the DSC or TGA method. This indicates that the microscopic FT-IR/DSC system is useful as a potential tool not only to investigate the degradation mechanism of drugs in the solid state but also to directly predict the bond energy of the reaction.

Full text of DOI:10.1021/jf990595l

J. Agric. Food Chem. 2000, 48, 631−635
631
Isoth er m a l F ou r ier Tr a n sfor m In fr a r ed Micr osp ectr osop ic Stu d ies
on th e Sta bility Kin etics of Solid -Sta te In tr a m olecu la r Cycliza tion of
Asp a r ta m e Sw eeten er
Yih-Dih Cheng and Shan-Yang Lin*
Biopharmaceutics Laboratory, Department of Medical Research & Education,
Veterans General Hospital-Taipei, Shih-Pai, Taipei, Taiwan, Republic of China
A novel Fourier transform infrared (FT-IR) microspectrophotometer equipped with differential
scanning calorimetry (DSC) was used to investigate the kinetics of intramolecular cyclization of
aspartame (APM) sweetener in the solid state under isothermal conditions. The thermal-dependent
changes in the peak intensity of IR spectra at 1543, 1283, and 1259 cm-1 were examined to explore
the reaction. The results support that the intramolecular cyclization process in APM proceeded in
three steps: the methoxyl group of ester was first thermolyzed to release methanol, then an acyl
cation was attacked by the lone pair of electrons available on nitrogen by an SN1 pathway, and
finally ring-closure occurred. The intramolecular cyclization of APM determined by this microscopic
FT-IR/DSC system was found to follow zero-order kinetics after a brief induction period. The bond
cleavage energy (259.38 kJ /mol) of thermolysis for the leaving group of -OCH3, the bond conversion
energy (328.88 kJ /mol) for the amide II NH band to DKP NH band, and the CN bond formation
energy (326.93 kJ /mol) of cyclization for the DKP in the APM molecule were also calculated from
the Arrhenius equation. The total activation energy of the DKP formation via intramolecular
cyclization was 261.33 kJ /mol, calculated by the above summation of the bond energy of cleavage,
conversion, and formation, which was near to the value determined by the DSC or TGA method.
This indicates that the microscopic FT-IR/DSC system is useful as a potential tool not only to
investigate the degradation mechanism of drugs in the solid state but also to directly predict the
bond energy of the reaction.
Keyw or d s: Aspartame; isothermal stability; solid state; intramolecular cyclization; FT-IR/ DSC
system; bond energy; activation energy
INTRODUCTION
1996c, 1999a,b), and thermal stability and reversibility
of R-crystallin (Lin et al., 1998).
Considerable interest has recently been directed to
Fourier transform infrared (FT-IR) microspectroscopy
for characterization of microscopic samples (Meesser-
schmidt and Harthcock, 1988; Wegmann 1998). Ad-
ditionally, the thermal-dependent properties of samples
have also been studied by combining the FT-IR mi-
crospectroscopy with a differential scanning calorimetry
(DSC). This novel FT-IR/DSC microscopic system is a
simple, rapid, and powerful tool for determining the
thermal-dependent behavior of microsamples (Mira-
bella, 1988). It can measure the effect of temperature
on the conformation of samples and also act as an
accelerated stability testing method to predict product
stability. This unique system has been extensively
approached in our laboratory to investigate simulta-
neously the correlation between the structural change
of the compound and the temperature applied, such as
phase transition and desolvation of polymorphs or
solvation of drugs (Lin, 1992; Tsai et al., 1993), ther-
motropic transition of skin lipid and protein with or
without skin penetrating enhancers (Lin et al., 1994a,
1996a,b), kinetics of curing of silicon elastomers (Lin
et al., 1994b), molecular interaction in Eudragits and
poly(N-isopropylacrylamide) polymers (Lin et al., 1995,
The stability of drugs in pharmaceutical products is
a very critical issue in clinical drug therapy since it may
influence the efficacy and safety of a drug. Before new
methods are developed to stabilize the drugs, the solid-
state stability of drugs has to be first characterized. The
kinetics of the solid-state decomposition reactions of
drugs should be studied to ensure and predict their shelf
life (Byrn, 1982). Aspartame (APM) is a dipeptide
consisting of the methyl ester of N-L-R-aspartyl-L-
phenylalamine and always acts as a sweetener. With
an indirect analytical method, it was found to undergo
thermal-responsive intramolecular cyclization to form
the cyclic compound 3-(carboxymethyl)-6-benzyl-2,5-
diketopiperazine (DKP) (Leung et al., 1997, 1998). The
kinetics of this solid-state reaction has been examined
by the DSC method and thermogravimetric analysis
(TGA) and best fitted the Prout-Tompkins equation
(Leung et al., 1997; Prout and Tompkins, 1944). Our
previous study has successfully used this FT-IR/DSC
microscopic system to quickly and directly investigate
the consecutive pathway of the dehydration process and
intramolecular cyclization of APM dipeptides in the solid
state (Cheng and Lin, 1999).
In the present study, this unique system was used to
isothermally determine the kinetics of intramolecular
cyclization in an APM molecule from thermal-responsive
IR spectra. Moreover, the bond dissociation energy of
* Corresponding author. Fax: 886-2-2875-1562. E-mail:
sylin@vghtpe.gov.tw.
10.1021/jf990595l CCC: $19.00 © 2000 American Chemical Society
Published on Web 03/02/2000

632 J. Agric. Food Chem., Vol. 48, No. 3, 2000
Cheng and Lin
With the increase of heating rates, all of the peak
temperatures shifted to the higher temperature range.
Different heating rates might change the peak shape
and peak temperature of a sample, thus the choice of a
correct heating rate is of vital importance in ascribing
the correct temperature to a transition. The broad
endothermic peaks within 109-125 °C were due to the
dehydration of APM hemihydrate, but the exothermic
peaks at 126-138 °C might be attributed to the recrys-
tallization of APM to form anhydrous APM. Moreover,
the endothermic peaks ranging from 181 to 189 °C were
due to the intramolecular cyclization of APM molecules
to form DKP (Leung et al., 1998). Since APM proceeded
in the solid-state reaction before the melting point of
anhydrous APM, the melting point of anhydrous APM
could not be found on the thermogram. The endothermic
peaks within 240-249 °C should be assigned to the
melting of solid products of DKP.
F igu r e 1. DSC thermograms of aspartame hemihydrate with
four scanning heating rates. Key: a, 3 °C/min; b, 5 °C/min; c,
8 °C/min; d, 10 °C/min.
According to the Kissinger equation in which the
reaction rate varied with temperature and then the peak
temperature also varied with heating rate (Kissinger,
1957), the activation energy of intramolecular cycliza-
tion of APM can be calculated.
thermolysis and the bond formation energy of cyclization
for DKP formation in the APM molecule were explored.
The total activation energy of the DKP formation by this
FT-IR/DSC microscopic technique was also calculated.
d(ln â/T2)
-Ea
)
R
d(1/T)
MATERIALS AND METHODS
Ma ter ia ls. Aspartame hemihydrate (APM) was obtained
from Tokyo Kasei Indus. Co. Ltd (Tokyo, J apan) without
further treatment. KBr crystals for pellets were purchased
from J ASCO Spectroscopic Co. Ltd (Tokyo, J apan).
Th er m a l An a lysis of Asp a r ta m e. The DSC curves of APM
crystals were determined by differential scanning calorimetry
(DSC-910, TA Instruments Inc., New Castle, DE) with four
heating rates of 3, 5, 8, and 10 °C/min in an open pan system
under a stream of N2 gas.
Tr a n sm ission F T-IR/DSC Tim e-Sca n Mea su r em en ts
(Lin et a l., 1995, 1996). A small amount of APM crystals was
sealed within two pieces of KBr pellets using a hydraulic press
(200 kg/cm2, 15 s). This compressed KBr disk was directly put
in the DSC microscopy cell (FP 84, Mettler, Greifensee,
Switzerland). The DSC microscopy cell was then placed on the
stage of the microscope in the FT-IR microscopic spectrometer
(Micro FTIR-200, J ASCO, Tokyo, J apan) with an MCT detec-
tor. The system was operated in the transmission mode. The
position and focus of the sample were adjusted through the
microscope. The temperature of the DSC microscopy cell was
monitored with a central processor (FT80HT, Mettler, Greif-
ensee, Switzerland). The heating rate of DSC assembly was
controlled at 3 °C/min in the isothermal condition. The
isothermal procedure used a time-scan measurement program
to control the DSC microscopy cell at 150, 155, 160, 165, and
170 °C, maintaining the sample at each temperature for 1 h.
During the experiment, the sample disk was first equilibrated
to the above prescribed temperature for about 3 min and then
time-scanned. The thermal-responsive IR spectra were re-
corded while the sample disk was heated on a DSC microscope
stage.
where â is the heating rate (°C/min), R is the gas
constant (8.314 J /(mol K)), Ea is the activation energy
(J /mol), and T is the peak temperature (K). From the
slope of a plot of ln â/T2) and 1/T, the Ea was calculated
as 250.3 kJ /mol for the intramolecular cyclization of
APM. This value of Ea was close to 265 ( 6 kJ /mol by a
DSC method and 268 ( 8 kJ /mol determined by a TGA
method (Leung et al., 1997, 1998). It should be noted
that the solid-state degradation is usually slower than
solution degradation, producing a greater activation
energy.
Evid en ce of In tr a m olecu la r Cycliza tion in AP M
Molecu les. Parts A-D of Figure 2 represent the three-
dimensional plots of FT-IR spectra of APM between
1800 and 1160 cm-1 with four heating temperatures
(155, 160, 165, and 170 °C) as a function of heating time.
Moreover, the critical changes in IR spectra for each
heating temperature are also superimposed in Figure
2A-1-D-1. It is evident that within the critical heating
intervals the peak intensity of IR spectra of APM at
certain wavenumbers changed markedly with temper-
ature. The critical heating intervals for conversion were
46-60 min for the 150 °C heated sample (figure not
shown), 26-40 min for the 155 °C heated sample, 16-
30 min for the 160 °C heated sample, 2-9 min for the
165 °C heated sample, and 0-2 min for the 170 °C
heated sample, respectively. The changes in IR spectral
wavenumber occurred at 1736-1718, 1543, 1377-1362,
1283-1259, and 1225 cm-1. The peak at 1736 cm-1 due
to the carbonyl stretching vibration of ester disappeared
gradually, but a new peak at 1718 cm-1 assigned to the
carbonyl CdO of carboxylic acid increased gradually via
a critical heating time. The amide II-related NH peak
at 1543 cm-1 disappeared stepwise. The peaks at 1377
and 1225 cm-1 assigned to the bending of methyl group
and CsO stretching of ester disappeared also. Moreover,
the peak at 1283 cm-1 assigned to the CN bond of DKP
gradually appeared but the peak at 1259 cm-1 corre-
sponding to the methoxyl group disappeared stepwise.
RESULTS AND DISCUSSION
Th er m a l Sta bility of AP M Deter m in ed by DSC
Meth od s. Thermal analysis techniques such as DSC
and TGA have proven useful in evaluating the kinetic
parameters of various reactions and materials. Figure
1 shows the DSC curves of APM determined by a series
of heating rates between 3 and 10 °C/min. Generally,
one broad and two sharp endothermic peaks and one
exothermic peak were observed in DSC curves, although
peak temperature varied with different heating rates.

Isothermal Stability of Aspartame
J. Agric. Food Chem., Vol. 48, No. 3, 2000 633
F igu r e 2. Three-dimensional plots of FT-IR spectra of aspartame hemihydrate with four heating temperatures as a function of
heating time (A-D) and the critical changes in IR spectra for each heating temperature (A-1-D-1) Isothermal heating
temperature: (A, A-1) 155 °C; (B, B-1) 160 °C; (C, C-1) 165 °C; (D, D-1) 170 °C.
This indicates that the DKP was formed in APM via
intramolecular cyclization. Although the critical heating
time was different from each prescribed heating tem-
perature, the main alteration in the above IR spectral
wavenumbers was similar, suggesting the same in-
tramolecular cyclization process in the APM molecule.
The step of intramolecular cyclization process based on
the changes in IR peak intensity is diagramed in Figure
3. This process was supposed to proceed by three steps:
The methoxyl group of ester was first thermolyzed
during the heating process to release methanol, then
an acyl cation was attacked by a lone pair of electrons
available on nitrogen by an SN1 pathway, and finally
ring-closure occurred.
Kin etics a n d Bon d En er gy of Cycliza tion of
AP M by Micr o F T-IR/DSC System . The changes of

634 J. Agric. Food Chem., Vol. 48, No. 3, 2000
Cheng and Lin
F igu r e 3. Intramolecular cyclization process of aspartame.
F igu r e 5. Kinetics of the temperature-dependent intramo-
lecular cyclization in aspartame. Key: b, 170 °C; O, 165 °C;
[, 160 °C; ], 155 °C; 2, 150 °C.
F igu r e 4. Comparison of the original FT-IR and its second-
derivative spectra determined at isothermal 160 °C heating
temperature as a function of heating time. Heating time: a,
before heating; b, 22 min; c, 60 min.
F igu r e 6. Arrhenius plot for intramolecular cyclization of
aspartame for each specific spectrum. Key: b, 1543 cm-1; [,
1283 cm-1; 2, 1259 cm-1
.
four IR peaks at 1543, 1495, 1283, and 1259 cm-1 were
studied to investigate the intramolecular cyclization of
APM, as shown in Figure 4. The peaks at 1543 and 1259
cm-1, assigned respectively to the amide II N-H band
and the CsO of ester, disappeared gradually with an
increase of heating time, but the peak at 1283 cm-1
corresponding to the C-N band in the DKP was formed
stepwise. The peak at 1495 cm-1 due to the CdC
stretching of the benzene ring maintained a constant
intensity with heating and could be used as an internal
marker. Thus the kinetics of intramolecular cyclization
of APM to form DKP were studied by plotting the peak
area ratio of 1543, 1259, or 1283 cm-1 to 1495 cm-1
against the heating time, as indicated in Figure 5. This
the DSC method or TGA technique (Leung et al., 1997,
1998), the present study found that the intramolecular
cyclization of APM still obeyed zero-order kinetics after
a brief induction period, since the plot of the fraction
decomposed or formed versus time was nearly linear (R
> 0.99). Thus the rate constants for the ester cleavage
of -OCH3, the disappearance of the amide II NH band,
and the CN bond formation of the DKP in APM molecule
at different temperatures were obtained from the slope
of linear plots.
The rate constants (k) are then plotted versus tem-
perature (T) according to the Arrhenius equation
clearly indicates that the peak area ratios for 1543 cm-1
/
k ) A exp(-Ea/RT)
1495 cm-1 or 1259 cm-1/1495 cm-1 decreased with
heating, but the peak area ratios for 1283 cm-1/1495
cm-1 increased. DKP formation was apparently heat-
dependent, increasing between 150 and 170 °C. The
values of peak area ratio reached a constant plateau (0
or constant), except for the 150 °C treated sample,
suggesting the reaction was almost complete throughout
the heating course.
The solid-state reactions of drugs have been exten-
sively analyzed but are usually found not clearly zero-
order or first-order, although the Prout-Tompkins
equation has also been used (Byrn, 1982; Leung et al.,
1997). Although the kinetics of the solid-state reaction
in the APM molecule via intramolecular cyclization can
best fit the Prout-Tompkins equation determined by
where A is a constant known as the frequency factor
and R is the gas constant. A plot of ln k against 1/T
was obtained as indicated in Figure 6. A good linear
relationship for each of three plots was obtained (R >
0.99). According to our proposed scheme (Figure 3) for
the intramolecular cyclization in APM, three steps have
been delineated. Thus, the energy for the representative
step was calculated. The bond dissociation energy of
thermolysis for the leaving group of -OCH3 was 259.38
kJ /mol, the bond conversion energy for the amide II NH
band was 328.88 kJ /mol to DKP NH band, and the CN
bond formation energy of cyclization for DKP was 326.93
kJ /mol, respectively. The total activation energy of the
DKP formation via intramolecular cyclization was 261.33

Isothermal Stability of Aspartame
J. Agric. Food Chem., Vol. 48, No. 3, 2000 635
Lin, S. Y.; Duan, K. J .; Lin, T. C. Microscopic FT-IR/DSC
combined system used to investigate the thermotropic
behaviors of lipid in porcine stratum corneum after pre-
treatment with skin penetration enhancers. Skin Res.
Technol. 1996a , 2, 186-191.
Lin, S. Y.; Duan, K. J .; Lin, T. C. Microscopic FT-IR/DSC
system used to investigate the conversion process of protein
structure in porcine stratum corneum after pretreatment
with skin penetration enhancers. Methods Find. Exp. Clin.
Pharmacol. 1996b, 18, 175-181.
Lin, S. Y.; Liao, C. M.; Hsiue, G. H. Isothermal kinetics of
anhydride formation in Eudragit L-100 films determined by
reflectance FTIR/DSC microspectroscopy. Polymer 1996c,
37, 269-273.
Lin, S. Y.; Ho, C. J .; Li, M. J . Thermal stability and revers-
ibility of secondary conformation of R-Crystallin membrane
during repeated heating processes. Biophys. Chem. 1998,
74, 1-10.
Lin, S. Y.; Chen, K. S.; Liang, R. C. Thermal micro ATR/FT-
IR spectroscopic system for quantitative study of the mo-
lecular structure of poly(N-isopropylacrylamide) in water.
Polymer 1999a , 40, 2619-2624.
Lin, S. Y.; Yu, H. L.; Li, M. J . Formation of six-membered cyclic
anhydrides by thermally induced intramolecular ester con-
densation in Eudragit E film. Polymer 1999b, 40, 3589-
3593.
kJ /mol, calculated by the summation of bond conversion
and bond formation energies. The 261.33 kJ /mol for the
intramolecular cyclization of APM determined by micro
FT-IR/DSC system was near the value of 265 ( 6 kJ /
mol by the DSC method and 268 ( 8 kJ /mol by the TGA
method (Leung et al., 1997, 1998). This strongly dem-
onstrates that the microscopic FT-IR/DSC system not
only is a potential tool to investigate the degradation
mechanism of drug or sweetener in the solid state but
also can directly predict the bond energy of the reaction
site.
LITERATURE CITED
Byrn, S. R. Solid-state Chemistry of Drugs; Academic Press:
New York, 1982.
Cheng, Y. D.; Lin, S. Y. Microscopic FT-IR/DSC system quickly
and directly evidenced the pathways of dehydration and
intramolecular cyclization of solid-state aspartame dipep-
tides. Submitted for publication.
Kissinger, H. Reaction kinetics in differential thermal analysis.
Anal. Chem. 1957, 29, 1702-1706.
Leung, S. S.; Grant, D. J . W. Solid-state stability studies of
model dipeptides: Aspartame and aspartylphenylalanine.
J . Pharm. Sci. 1997, 86, 64-71.
Leung, S. S.; Padden, B. E.; Munson, E. J .: Grant, D. J . W.
Hydration and dehydration behavior of aspartame hemi-
hydrate. J . Pharm. Sci. 1998, 67, 508-513.
Lin, S. Y. Isolation and solid-state characteristics of a new
crystal form of indomethacin. J . Pharm. Sci. 1992, 81, 572-
576.
Lin, S. Y.; Liang, R. C.; Lin, T. C. Lipid and protein thermo-
tropic transition of porcine stratum corneum by microscopic
calorimetry and infrared spectroscopy. J . Chin. Chem. Soc.
Taipei 1994a , 41 425-429.
Lin, S. Y.; Tsay, W. J .; Chen, Y. L.; Lee, C. J . Application of a
new micro FTIR/DSC technique to the study of curing
kinetics of silicone elastomer. J . Controlled Release 1994b,
31, 277-282.
Messerschmidt, R. G., M. A. Harthcock, Eds. Infrared Mi-
crospectroscopy: Theory and Applications; Maecel Dekker:
New York, 1988.
Mirabella, F. M. Simultaneous FT-IR Microspectroscopy and
differential scanning calorimetry: Polymer studies. In
Infrared Microspectroscopy: Theory and Applications; Maecel
Dekker: New York, 1988; pp 85-92.
Prout, E.; Tompkins, F. The thermal decomposition of potas-
sium permanganate. Trans. Faraday Soc. 1944, 40, 488-
498.
Tsai, S. Y.; Kuo, S. C.; Lin, S. Y. Physicochemical characteriza-
tion of anthraquinone 2-carboxylic acid. J . Pharm. Sci. 1993,
82, 1250-1254.
Wegmann, R, D.; Wetzel, L. B.; LeVine, S. M. Fourier
Transform Infrared (FT-IR) Microspectroscopy. Cell. Mol.
Biol. 1998, 44, 1-272.
Lin, S. Y.; Liao, C. M.; Hsiue, G. H. A reflectance FTIR/DSC
microspectroscopic study of nonisothermal kinetics of an-
hydride formation in Eudragit L-100 films. Polym. Degrad.
Stab. 1995a , 47, 299-303.
Lin, S. Y.; Liao, C. M.; Hsiue, G. H. Thermal-dependent
anhydride formation of Eudragit L-100 films determined by
reflectance FTIR/DSC microspectroscopy. Polymer 1995b,
36, 3239-3241.
Received for review J une 2, 1999. Revised manuscript received
December 2, 1999. Accepted December 29, 1999.
J F990595L