Study of the thermal degradation of benzene-containing glycerol carbonate derivatives by a combined TG–FTIR and theoretical calculation

Article abstract of DOI:10.1016/j.tca.2017.05.021

We recently found that the introduction of benzene ring as connection block improved the thermal stability of glycerol carbonate derivatives. To further explore how polysubstitution on benzene influences their thermal stability, another two derivatives, benzene-1,3,5-tricarboxylic acid tris-(2-oxo-[1,3]dioxolan-4-ylmethyl) ester (BATE1) and benzene-1,2,4,5-tetracarboxylic acid tetrakis-(2-oxo-[1,3]dioxolan-4-ylmethyl) ester (BATE2), were synthesized and characterized by a combined TG–FTIR and theoretical calculation. TG results showed that both BATE1 and BATE2 had three weight loss stages. FTIR spectra of evolved gases and the bond dissociation energy for the two compounds indicated that during the pyrolysis process, the C[sbnd]C, C[sbnd]O bonds of five-membered cyclic carbonate fractured preferentially to produce volatile carbonates and CO₂, and then the nearby C[sbnd]O bond of esters continued to crack to produce more fragments. Finally, benzene ring still remains in the solid pyrolysis products. Therefore, improving the thermal stability of compound cannot be achieved simply by increasing the number of substituents on the benzene ring.

Full text of DOI:10.1016/j.tca.2017.05.021

Thermochimica Acta 654 (2017) 179–185
Contents lists available at ScienceDirect
Thermochimica Acta
journal homepage: www.elsevier.com/locate/tca
Study of the thermal degradation of benzene-containing glycerol carbonate
derivatives by a combined TG–FTIR and theoretical calculation
a,⁎
a
a
b,⁎
c,⁎
a
Jing Liu , Rujuan Li , Mengya Guo , Hairong Tao , Donglan Sun , Chengxing Zong ,
a
a
Chunjing Liu , Fengzhi Fu
College of Chemical Engineering and Material Science, Tianjin University of Science & Technology, Tianjin 300457, China
College of Chemistry, Beijing Normal University, Beijing 100875, China
a
b
c
College of Science, Tianjin University of Science & Technology, Tianjin 300457, China
A R T I C L E I N F O
A B S T R A C T
Keywords:
Glycerol carbonate derivatives
TG–FTIR
Bond dissociation energy
Thermal stability
Polysubstitution
We recently found that the introduction of benzene ring as connection block improved the thermal stability of
glycerol carbonate derivatives. To further explore how polysubstitution on benzene influences their thermal
stability, another two derivatives, benzene-1,3,5-tricarboxylic acid tris-(2-oxo-[1,3]dioxolan-4-ylmethyl) ester
(BATE1) and benzene-1,2,4,5-tetracarboxylic acid tetrakis-(2-oxo-[1,3]dioxolan-4-ylmethyl) ester (BATE2),
were synthesized and characterized by a combined TG–FTIR and theoretical calculation. TG results showed that
both BATE1 and BATE2 had three weight loss stages. FTIR spectra of evolved gases and the bond dissociation
energy for the two compounds indicated that during the pyrolysis process, the CeC, CeO bonds of five-mem-
2
bered cyclic carbonate fractured preferentially to produce volatile carbonates and CO , and then the nearby CeO
bond of esters continued to crack to produce more fragments. Finally, benzene ring still remains in the solid
pyrolysis products. Therefore, improving the thermal stability of compound cannot be achieved simply by in-
creasing the number of substituents on the benzene ring.
1. Introduction
[1,3]dioxolan-4-ylmethyl) ester (TABE) by TG–FTIR showed us an in-
teresting result that the 5% weight loss temperature was increased in
As a representative of high-energy green batteries, lithium-ion
the order of TABE > SABE > CABE, which means that the thermal
stability of TABE is the best. Furthermore, it revealed that benzene ring
was introduced as connection block to improve the thermal stability of
the GC derivatives (e.g., greater than saturated hydrocarbon block). To
further extend our previous work, it will be interesting to study how the
number of substituents on the benzene ring affects the thermal stability.
For example, whether improving the thermal stability can be simply
obtained by introducing more substituents on benzene ring.
In this work, we increased the number of substituents on benzene in
the molecular structure of terephthalic acid bis-(2-oxo-[1,3]dioxolan-4-
ylmethyl) ester (TABE), designed and synthesized another two new GC
derivatives: benzene-1,3,5-tricarboxylic acid tris-(2-oxo-[1,3]dioxolan-
4-ylmethyl) ester (BATE1) and benzene-1,2,4,5-tetracarboxylic acid
tetrakis-(2-oxo-[1,3]dioxolan-4-ylmethyl) ester (BATE2). The chemical
structures of the two compounds are shown in Fig. 1. In order to un-
derstand how the substituents affect the thermal stability on this type of
ester, TG–FTIR was used to explore the dynamic decomposition and
analyze the gaseous products during the thermal pyrolysis of the two
compounds. Furthermore, the theoretical calculations of the bond
batteries have many outstanding advantages like high voltage, high
energy density, long cycle-life, low self-discharge and environment-
friendly, but they still have some issues in durability, operating tem-
perature and safety to improve [1–4]. Among these, the safety problems
become a serious obstacle to wide application like plug-in electric ve-
hicles and stand-by power sources for communications and modern
airplanes [5,6]. The use of flammable organic electrolytes has been
considered one of the primary reasons causing safety problems [7–9].
Therefore, more and more researchers have investigated the flamm-
ability and thermal stability of the electrolyte at higher temperature.
TG–FTIR technique, having the advantages of minimal material re-
quirement, simple temperature program control, and online recording,
has been widely used to analyze thermal pyrolysis of compounds
[
10–13].
In our recent work [14], the investigation of thermal pyrolysis of
three glycerol carbonate (GC) derivatives: carbonic acid bis-(2-oxo-
1,3]dioxolan-4-ylmethyl) ester (CABE), succinic acid bis-(2-oxo-[1,3]
[
dioxolan-4-ylmethyl) ester (SABE) and terephthalic acid bis-(2-oxo-
⁎
Corresponding authors.
E-mail addresses: jingliu@tust.edu.cn, jingliu2000cn@yahoo.com (J. Liu), hairongtao@bnu.edu.cn (H. Tao), sundonglan@tust.edu.cn (D. Sun).
http://dx.doi.org/10.1016/j.tca.2017.05.021
Received 6 February 2017; Received in revised form 24 May 2017; Accepted 29 May 2017
Available online 02 June 2017
0040-6031/ © 2017 Elsevier B.V. All rights reserved.

J. Liu et al.
Thermochimica Acta 654 (2017) 179–185
Fig. 1. Chemical structure of glycerol carbonate
(GC) derivatives.
dissociation energy for both compounds were carried out for facilitating
to clarify their thermal pyrolysis behaviors. This study may provide
useful information for future molecular design and synthesis of new
carbonates.
a magnetic stirrer, a funnel with pressure-equalising tube and a ther-
mometer was placed 3.69 g (31.25 mmol) of glycerol carbonate,
3.47 mL (25 mmol) of triethylamine and 50 mL of anhydrous acetone as
solvent. A solution of pyromellitic acid chloride (2.05 g, 6.25 mmol) in
anhydrous acetone (30 mL) was added dropwise to the above solution
cooled in ice bath under nitrogen atmosphere. The mixture was stirred
at room temperature for 12 h. After reaction, triethylamine hydro-
chloride was removed by filtration. Then, the filtrate was evaporated
under reduced pressure; the residue was purified by recrystallization
from dichloromethane to give the product (1.04 g, yield 25.43%) (re-
2. Experimental
2.1. Materials
Dimethyl carbonate (DMC, AR grade), calcium oxide (CaO, AR
grade) and triethylamine (TEA, AR grade) were purchased from Tianjin
Damao Chemical Reagent Factory (China). Glycerol (AR grade), thionyl
action 3, Scheme 1).
H NMR (400 MHz, DMSO-d ): δ 8.17 (s, 2H), 5.21–5.10 (m, 4H),
4.68–4.49 (m, 12H), 4.42–4.32 (m, 4H) (Fig. S2). FTIR (νmax cm ):
1
6
−1
chloride (SOCl
2
, AR grade) and N,N-dimethylformamide (DMF, AR
grade) were obtained from Tianjin Jiangtian Chemical Technology Co.,
Ltd. (China). Acetone (AR grade) was purchased from Tianjin No.1
Chemical Reagent Factory (China). Dichloromethane (AR grade) was
obtained from Tianjin Fuqi Chemical Co., Ltd. (China). Trimesoyl
chloride (98%) and pyromellitic dianhydride (99%) were purchased
from Aladdin Reagent Co., Ltd. (China). CaO was calcined at 800 °C for
2924, 1794, 1730, 1171, 1049, 768.
2.3. TG–FTIR measurement
The TG–FTIR system was composed of a TA SDT-600 Instrument
and a Bruker Tensor 27 FTIR spectrometer. For each TG–FTIR mea-
surement, 10–20 mg sample was weighted into an open alumina cru-
cible. The temperature was set from ambient temperature to 600 °C, the
6
h before using; other chemicals were used as received without further
purification.
−1
heating rate of the TG furnace was 20 °C min , and high purity
2.2. Synthesis of glycerol carbonate (GC) derivatives
−1
(
> 99.999%) nitrogen gas with a flow rate of 140 mL min was used
as carrier gas, this flow had to be kept high enough to avoid long re-
sidence time in the furnace and thus to prevent secondary reactions of
the volatiles [15]. The transfer line used to connect TG and FTIR was a
GC was synthesized by the transesterification of glycerol with DMC
14]. And pyromellitic acid chloride was synthesized from thionyl
chloride and pyromellitic dianhydride. Then, GC was used as organic
synthesis intermediate and further reacted with acid chloride to prepare
GC derivatives.
[
1
m long stainless steel tube with an internal diameter of 2 mm. In order
to reduce the possibility of gases condensing along the transfer line, the
temperature in the gas cell and transfer line were set to 200 °C. The
−1
FTIR spectral region was set as 4000–600 cm , with the resolution of
2
.2.1. Synthesis of benzene-1,3,5-tricarboxylic acid tris-(2-oxo-[1,3]
dioxolan-4-ylmethyl) ester (BATE1)
In a dry 250 mL 3-necked round-bottomed glass flask equipped with
a magnetic stirrer, a funnel with pressure-equalising tube and a ther-
mometer was placed 4.13 g (0.035 mol) of glycerol carbonate, 4.16 mL
−1
4
cm , co-adding 16 scans per spectrum. The experiment started only
when the whole system was stable. The experimental results of TG and
FTIR were recorded automatically by a computer. Data was processed
by the software OPUS 6.0 (Bruker Company, Germany).
(
0.03 mol) of triethylamine and 60 mL of anhydrous acetone as solvent.
A solution of trimesoyl chloride (2.65 g, 0.01 mol) in anhydrous
acetone (40 mL) was added dropwise to the above solution cooled in ice
bath under nitrogen atmosphere. The mixture was stirred at room
temperature for 12 h. After reaction, triethylamine hydrochloride was
removed by filtration. Then, the filtrate was evaporated under reduced
pressure; the obtained residue was washed by deionized water for
several times to give the product (3.83 g, yield 75.10%) (reaction 2,
Scheme 1).
2.4. Bond dissociation energy (BDE) calculation
The quantum chemistry calculations were performed using the
Gaussian 09 program. The molecular structures were optimized at the
M062X/cc-pVTZ level of theory and frequencies and energies of the
molecules were calculated at the same level of theory. All molecular
structures without imaginary frequencies were local minima on the
potential energy surfaces. Formula (1) described by Blanksby and
Ellison [16], was used to calculate the bond dissociation energy (BDE)
for a dissociation process: R − Q → R + Q.
1H NMR (400 MHz, DMSO-d
): δ 8.68 (s, 3H), 5.29–5.18 (m, 3H),
−1
6
4
.82–4.41 (m, 12H) (Fig. S1). FTIR (νmax cm ): 2923, 1794, 1734,
1245, 1050, 738.
D
0
(R − Q) = E
0
(R) + E
0
(Q) − E
0
(R – Q)
(1)
2.2.2. Synthesis of benzene-1,2,4,5-tetracarboxylic acid tetrakis-(2-oxo-
where D (R − Q), E
0
0
(R), E
0
(Q) and E
0
(R – Q) denote to be the calcu-
[
1,3]dioxolan-4-ylmethyl) ester (BATE2)
lated BDE, and the electronic energies at 0 K with zero-point energy
In a dry 250 mL 3-necked round-bottomed glass flask equipped with
(ZPE) corrections for R, Q and R – Q, respectively.
180

J. Liu et al.
Thermochimica Acta 654 (2017) 179–185
Scheme 1. The synthetic route to glycerol carbonate
(GC) derivatives.
3
. Results and discussion
the BATE1’s TG curve was very smooth before 210 °C and the material
weight was almost unchanged. When the temperature rose to 210 °C,
the weight began to decrease gradually, BATE1 went into the thermal
cracking stage at this time, the whole pyrolysis process continued until
540 °C. TG curve flattened again after 540 °C, which indicated that the
cracking was almost completed. The overall weight loss of BATE1
measured at 600 °C was ∼83%. The DTG curve indicated that weight
loss of BATE1 went through three main stages and the three stages were
all partially overlapped. The first decomposition stage was from 210 to
3.1. The thermal pyrolysis of BATE1 and BATE2
Fig. 2 presented the thermogravimetric (TG) and differential ther-
mogravimetric (DTG) curves, the Gram–Schmidt (G–S) signal and the
FTIR absorbance curves of CO
a heating rate of 20 °C min
2
1
during pyrolysis of BATE1 and BATE2 at
under nitrogen gas. As shown in Fig. 2,
−
343 °C, with a weight loss of 32%; the second stage from 343 to 393 °C,
with a weight loss of 13%; the third stage from 393 to 540 °C, with a
weight loss of 35%. The maximum decomposition rate of each stage
occurred at 292 °C, 377 °C and 424 °C, respectively.
Compared to BATE1, BATE2 was observed to have different thermal
pyrolysis. The TG curve showed that the material weight was almost
unchanged under 230 °C. After that, BATE2 went into the pyrolysis
stage and its weight decreased gradually, the whole pyrolysis process
continued until 600 °C. The DTG curve indicated that BATE2 also had
three stages of weight loss. But the weight loss at the first stage
(
230–290 °C) is dramatically different from that of BATE1 (∼4% for
BATE2 versus ∼32% for BATE1). The second and third stages for
BATE2 were more overlapped with each other when compared with the
counterpart of BATE1. Their weight loss reached to 76%, and the
overall weight loss of BATE2 was 85%. There was only negligible mass
loss around 600 °C. The maximum decomposition rate for each stage
occurred at 247 °C, 369 °C and 405 °C, respectively.
5% weight loss temperature was chosen as the initial thermal pyr-
olysis temperature of compounds to evaluate the thermal stability of
BATE1 and BATE2. The thermal decomposition temperatures of BATE1
and BATE2 were 251 °C and 288 °C, respectively. It seemed that in-
crease of substituents on benzene ring could improve the thermal sta-
bility of this type of compounds. However, in our previous work that
TABE had two glycerol carbonate substituents on benzene ring, 5%
weight loss temperature of TABE was as high as 284 °C [14]. Therefore,
for the same type of ester containing the benzene ring, the number of
substituents on the benzene ring has random effects on the compound’s
thermal stability, and improving the thermal stability of compound
cannot be obtained simply by increasing the number of substituents on
the benzene ring.
To explore the thermal pyrolysis process of the two compounds,
coupling of FTIR to thermogravimetry instrument was used to analyze
the volatile components released at every pyrolysis stage. The 3D dia-
grams for the FTIR spectra of the evolved gases produced by pyrolysis of
2
Fig. 2. TG, DTG, Gram–Schmidt and the FTIR absorbance curves of CO during the
thermal pyrolysis of BATE1 (A) and BATE2 (B).
181

J. Liu et al.
Thermochimica Acta 654 (2017) 179–185
Fig. 3. The 3D diagrams for the FTIR spectra of the evolved gases produced by pyrolysis
of BATE1 (A) and BATE2 (B).
BATE1 and BATE2 were shown in Fig. 3. Two significant absorbance
Fig. 4. The FTIR spectra of the evolved gases from the thermal pyrolysis of BATE1 (A) and
BATE2 (B) at the temperature where the Gram–Schmidt signal is maximum.
−1
−1
bands at 2359 cm
gases at an early stage of the pyrolysis for both compounds. The former
band was assigned to CO asymmetric stretching vibration, and the
and 1732 cm
were detected from the evolved
2
Table 1
latter was assigned to C]O of carbonate (R–O–(R’–O–)C]O stretching
vibration [17,18] or its secondary decomposition product: acetaldehyde
O
The infrared intensity ratio of νCO2 and νC] for BATE1/BATE2’s gaseous pyrolysis
products.
[
19–21]. As temperature increased, pyrolysis processed continually.
1
st stage
2nd stage
3rd stage
More organic gaseous fragment was detected. For BATE1, a broad band
centered around 2900 cm was assigned to CH
−
1
a
b
a
b
a
b
2
and CH
3
asymmetric
T
298 °C
T
264 °C
T
401 °C
T
376 °C
T
424 °C
T
407 °C
stretching vibration, and at high temperature, CeH stretching vibration
c
I
I
I
2359/cm−1
0.17
0.047
3.62
0.011
0.0043
2.56
0.034
0.19
0.18
0.25
0.10
2.50
0.11
0.21
0.52
0.10
0.14
0.71
did appear in the spectra. For BATE2, at high temperature (> 400 °C), a
d
1732/cm−1
−1
broad band centered around 2900 cm
did not become stronger as
2359/I1732
BATE1, and CeH stretching vibration did not appear between 3100 and
−
1
a
3000 cm . The different 3D diagrams suggest that BATE1 and BATE2
BATE1.
BATE2.
CO .
2
b
have taken different thermal pyrolysis process.
c
For the purpose of better analyzing the volatile gases released
during pyrolysis, the characteristic FTIR spectra of the gaseous products
obtained at the maxima of the Gram–Schmidt curves had been obtained
d
C]O.
intensity ratio of νCO2 and νC]
products (see Table 1). For both BATE1 and BATE2, at the first stage of
pyrolysis, CO dominated in evolved gases, which I2359/I1732 was 3.62
and 2.56, respectively. However, at the second stage, BATE1 and BA-
TE2 had quite different reaction processes: I2359/I1732 decreased greatly
from 3.62 to 0.18 for BATE1, but it remained small change from 2.56 to
O
for BATE1/BATE2’s gaseous pyrolysis
(
see Fig. 4). According to previous studies [11,13,14], some typical
small molecular gaseous species can be easily identified by their char-
2
−1
−1
acteristic absorbance: CO
2
at 2359 cm
000–3400 cm . Moreover, some characteristic frequencies except
carbonyl group of carbonate could be assigned. For example, the 1400,
2
and 669 cm ; H O at
−1
4
−1
7
40 cm
2
bands were assigned to CH bending and rocking vibration.
2.50 for BATE2. This suggested that organic carbonate became the main
The gaseous carbonate also exhibited asymmetric OeCeO stretching
vibration at 1274 cm , and symmetric OeCeO stretching vibration at
9
evidence that benzene-containing volatile compounds existed.
Additional pyrolysis information can be deduced from the infrared
products instead of CO for BATE1 at the second stage. At the third
2
−
1
stage, organic carbonate became dominant in evolved gas for both
BATE1 and BATE2.
−1
18 cm . For both compounds’ evolved gases, there was no clear
182

J. Liu et al.
Thermochimica Acta 654 (2017) 179–185
Fig. 5. The optimized structures for BATE1 (A) and BATE2 (B).
Distances are in angstroms, and angles are degrees. Red: oxygen atom;
Gray: carbon atom.
3.2. Theoretical calculation of BATE1 and BATE2
Table 2
Data of the bond dissociation energy (BDE) of substituent for BATE1 and BATE2.
In order to understand the decomposition process of both com-
pounds, we optimized the BATE1 and BATE2’s molecular structures at
the M062X/cc-pVTZ level of theory and calculated their frequencies
and energies at the same level of theory. The optimized geometries for
BATE1 and BATE2 were shown in Fig. 5, and the bond dissociation
energies (BDEs) were listed in Table 2. Since both BATE1 and BATE2
have highly symmetric structures, respectively, each substitution group
has the same parameters for the chemical bonds and the angles, and
thus only the parameters from one representative substitution group
were shown in Fig. 5. The BDE value can reflect the degree of difficulty
of a bond fracture. The smaller the value is, the easier the bond cracks.
As shown in Table 2, the BDE values for most chemical bonds (6 out of
the total 8) of BATE1 and BATE2 are similar with the small difference
less than 6 kJ/mol. However, for the rest two chemical bonds the BDE
values of BATE1 are significantly higher than those of BATE2 by at least
BATE1(kJ/mol)
BATE2(kJ/mol)
O(1)–C(2)
C(2)–O(3)
O(3)–C(4)
C(4)–C(5)
C(5)–O(1)
C(5)–C(6)
C(6)–O(7)
C(8)–C(9)
395.9
400.9
354.6
331.2
345.5
357.3
394.8
448.7
390.1
399.5
349.4
328.3
330.5
360.2
396.8
431.2
15 kJ/mol (i.e., 15.0 kJ/mol for C(5)–O(1), and 17.5 kJ/mol for
C(8)–C(9)). This result is consistent with our experimental observations
that for BATE2, the maximum decomposition rate for each stage oc-
curred at 247 °C, 369 °C and 405 °C, respectively, while for BATE1, the
183

J. Liu et al.
Thermochimica Acta 654 (2017) 179–185
with water’s frequency, and its C]O stretching frequency was much
closer to other carbonates’ C]O group. Therefore, there was no clear
evidence for the existence of methyl carbonic acid in gaseous products.
In addition, the BDE values of ph–C (C(8)–C(9)) for BATE1 and
BATE2 were 448.7 kJ/mol and 431.2 kJ/mol, respectively, the highest
values of the substituent for both compounds. This can explain why
there is no benzene-related infrared absorbance in evolved gases. In
other words, the benzene ring should mainly remain in the final solid
pyrolysis products which were also confirmed by TG results. TG results
showed that the overall weight loss of BATE1 measured at 600 °C was
∼
83%, and the overall weight loss of BATE2 was 85%. That means that
the residual weights after pyrolysis of BATE1 and BATE2 were 17% and
5%, respectively. This is close to the mass fraction of benzene ring in
both of compounds which were 15% and 12%, respectively.
1
4. Conclusions
BATE1 and BATE2, two GC derivatives with different poly-
Scheme 2. The possible pyrolysis mechanisms for BATE1 and BATE2.
substitutions on the benzene ring, were synthesized for further ex-
ploring how polysubstitution on the benzene ring affects their thermal
stability. The thermal pyrolysis analysis was investigated by TG–FTIR in
details. The experimental results showed that both compounds had a
good thermal stability before 210 °C, and three weight loss stages. But
their weight loss curves differed greatly. 5% weight loss temperature
was chosen to evaluate the thermal stability of the two compounds. We
found that the thermal stability cannot be achieved simply by in-
creasing the number of substituents on the benzene ring. According to
the FTIR spectra and the theoretically calculated BDE data of the two
compounds, it implies that the CeC, CeO bonds in five-membered
maximum decomposition rate occurred at 292 °C, 377 °C and 424 °C,
respectively.
3.3. The possible pyrolysis mechanisms
Based on the experimental observation from the FTIR spectra and
the theoretical calculations on the BDE values of BATE1 and BATE2, the
possible pyrolysis mechanisms were proposed in Scheme 2. The BDE
values of C(2)–O(3) were 400.9 kJ/mol for BATE1 and 399.5 kJ/mol
for BATE2, much higher than other BDEs, and thus CO
produced by cleavage of A and B at a low temperature and the reaction
of CO releasing may not occur at the first stage. Detailed theoretical
calculations of the potential energy surfaces are beyond the scope of
this study. As C(5)–O(1) and O(3)–C(4) were 345.5 kJ/mol and
2
was hardly
2
cyclic carbonate fracture preferentially to release CO , water, and di-
methyl carbonate, and the CeO bond of esters nearby the cyclic car-
bonate cracks subsequently to produce some volatile vinylene carbo-
nate which could further decompose to acetaldehyde. Finally, the
benzene ring remains in the solid pyrolysis products.
2
354.6 kJ/mol for BATE1, and 330.5 kJ/mol and 349.4 kJ/mol for BA-
TE2, respectively, cleavage of A and C to generate CO and water most
2
Acknowledgements
likely occurred, and the thermal degradation could dominates in the
early pyrolysis. Furthermore, C(4)–C(5) and C(5)–C(6) were 331.2 kJ/
mol and 357.3 kJ/mol for BATE1 and 328.3 kJ/mol and 360.2 kJ/mol
for BATE2, respectively. Therefore, from the theoretical calculations
cleavage of D and E or cleavage of E may also occur at the early stage to
This work was financially supported by the National Natural Science
Foundation of China (Grant No. 21373150). The authors would like to
thank software application engineer Xuju Jiang at Bruker company for
her help on TG–FTIR measuremen.
release dimethyl carbonate or vinylene carbonate. But νC]
O
of di-
−
1
methyl carbonate would occur at 1818 cm , which was not observed
in IR spectra [18]. The reason could be that dimethyl carbonate is less
than instrument detection limit. Another possibility is that product of
dimethyl carbonate could be a thermodynamically favored reaction, not
kinetically favored one. The thermal degradations could make con-
tributions for gaseous carbonate in gaseous pyrolysis products. How-
Appendix A. Supplementary data
Supplementary data associated with this article can be found, in the
online version, at http://dx.doi.org/10.1016/j.tca.2017.05.021.
References
O
ever, the infrared intensity ratio of νCO2 and νC] for BATE1/BATE2’s
gaseous pyrolysis products (see Table 1) suggested that the organic
carbonate dominated pyrolysis products at high temperature. Thus,
cleavage of D and E or cleavage of E was the main thermal degradation
at high temperature, especially for BATE1 since the infrared intensity
ratio was decreased from 3.62 to 0.18 when temperature increased
from 298 °C to 401 °C, and then the ratio was increased from 0.18 to
[1] M.Q. Xu, W.S. Li, X.X. Zuo, J.S. Liu, X. Xu, Performance improvement of lithium ion
battery using PC as a solvent component and BS as an SEI forming additive, J.
Power Sources 174 (2007) 705–710.
[
2] E.-G. Shim, T.-H. Nam, J.-G. Kim, H.-S. Kim, S.-I. Moon, Effects of functional
electrolyte additives for Li-ion batteries, J. Power Sources 172 (2007) 901–907.
3] C. Shi, X. Li, X. He, J. Zhao, New insight into the interaction between carbonate-
based electrolyte and cuprous sulfide electrode material for lithium ion batteries,
Electrochim. Acta 174 (2015) 1079–1087.
[
0
.52 when temperature rose from 401 °C to 424 °C. This indicated that
part of carbonate (R–O–)(R’–O–)C]O decomposed to produce CO at
high temperature. For BATE2, the ratio was slightly decreased from
.56 to 2.50 when the pyrolysis processed from the first stage to the
second stage, and temperature was changed from 264 °C to 376 °C. This
indicated that the thermal degradation to release CO dominated in the
[4] G. Zhu, K. Wen, W. Lv, X. Zhou, Y. Liang, F. Yang, Z. Chen, M. Zou, J. Li, Y. Zhang,
W. He, Materials insights into low-temperature performances of lithium-ion bat-
teries, J. Power Sources 300 (2015) 29–40.
2
[
5] E.-G. Shim, T.-H. Nam, J.-G. Kim, H.-S. Kim, S.-I. Moon, Effect of vinyl acetate plus
vinylene carbonate and vinyl ethylene carbonate plus biphenyl as electrolyte ad-
ditives on the electrochemical performance of Li-ion batteries, Electrochim. Acta 53
2
(2007) 650–656.
2
[
6] Z. Zeng, B. Wu, L. Xiao, X. Jiang, Y. Chen, X. Ai, H. Yang, Y. Cao, Safer lithium ion
first two stages. As temperature increased, carbonate-related thermal
degradation became the main process. Methyl carbonic acid could be
formed by cleavage of A and D, which seemed favorable to happen if
the BDE was considered. Its eOH stretching frequency exhibited a weak
batteries based on nonflammable electrolyte, J. Power Sources 279 (2015) 6–12.
[7] S.-c. Kinoshita, M. Kotato, Y. Sakata, M. Ue, Y. Watanabe, H. Morimoto, S.-
i. Tobishima, Effects of cyclic carbonates as additives to γ-butyrolactone electrolytes
for rechargeable lithium cells, J. Power Sources 183 (2008) 755–760.
[
8] L. Zhao, S. Okada, J.-i. Yamaki, Effect of VC additive on MFA-based electrolyte in
−1
sharp band near 3580 cm in the vapor phase, which was overlapped
Li-ion batteries, J. Power Sources 244 (2013) 369–374.
184

J. Liu et al.
Thermochimica Acta 654 (2017) 179–185
[
9] P.-L. Kuo, C.-H. Tsao, C.-H. Hsu, S.-T. Chen, H.-M. Hsu, A new strategy for preparing
oligomeric ionic liquid gel polymer electrolytes for high-performance and non-
flammable lithium ion batteries, J. Membr. Sci. 499 (2016) 462–469.
[15] J. Giuntoli, W. de Jong, S. Arvelakis, H. Spliethoff, A.H.M. Verkooijen, Quantitative
and kinetic TG?FTIR study of biomass residue pyrolysis: dry distiller's grains with
solubles (DDGS) and chicken manure, J. Anal. Appl. Pyrolysis 85 (2009) 301–312.
[16] S.J. Blanksby, G.B. Ellison, Bond dissociation energies of organic molecules, Acc.
Chem. Res. 36 (2003) 255–263.
[
[
[
[
[
10] D. Chen, D. Liu, H. Zhang, Y. Chen, Q. Li, Bamboo pyrolysis using TG-FTIR and a
lab-scale reactor: analysis of pyrolysis behavior, product properties, and carbon and
energy yields, Fuel 148 (2015) 79–86.
11] L. Jie, L. Yuwen, S. Jingyan, W. Zhiyong, H. Ling, Y. Xi, W. Cunxin, The in-
vestigation of thermal decomposition pathways of phenylalanine and tyrosine by
TG–FTIR, Thermochim. Acta 467 (2008) 20–29.
12] J.M. Cervantes-Uc, J.V. Cauich-Rodríguez, H. Vázquez-Torres, L.F. Garfias-Mesías,
D.R. Paul, Thermal degradation of commercially available organoclays studied by
TGA–FTIR, Thermochim. Acta 457 (2007) 92–102.
13] R.H. Aguirresarobe, L. Irusta, M.J. Fernandez-Berridi, Application of TGA/FTIR to
the study of the thermal degradation mechanism of silanized poly(ether-urethanes),
Polym. Degrad. Stabil. 97 (2012) 1671–1679.
[17] G. Socrates, Infrared and Raman Characteristic Group Frequencies: Tables and
Charts, third ed., Wiley, Chichester, 2001.
[18] R.A. Nyquist, Interpreting Infrared, Raman, and Nuclear Magnetic Resonance
Spectra Vol. 1 Elsevier, New York, 2001.
[19] R. Mogi, M. Inaba, Y. Iriyama, T. Abe, Z. Ogumi, Study on the decomposition
mechanism of alkyl carbonate on lithium metal by pyrolysis-gas chromatography-
mass spectroscopy, J. Power Sources 119–121 (2003) 597–603.
[20] T. Li, L. Xing, W. Li, Y. Wang, M. Xu, F. Gu, S. Hu, How does lithium salt anion affect
oxidation decomposition reaction of ethylene carbonate: a density functional theory
study, J. Power Sources 244 (2013) 668–674.
14] R. Li, J. Liu, H. Tao, D. Sun, D. Xiao, C. Liu, The investigation of thermal pyrolysis of
[21] J.C. Evans, H.J. Bernstein, The vibrational spectra of acetaldehyde and acet-
glycerol carbonate derivatives by TG–FTIR, Thermochim. Acta 624 (2016) 76–81.
1
aldehyde-d , Can. J. Chem 34 (1956) 1083–1092.
185