68
E. Zinigrad et al. / Thermochimica Acta 457 (2007) 64–69
All the B2O3 produced by the first step (Eq. (1)) is con-
consumes only 1/3 of the Li oxalate formed in the first. This
correlates well with the fact that the final products of LiBOB
decomposition contain Li oxalate, as evidenced by both XRD
and FTIR (Figs. 6 and 7). The calculated weight loss for decom-
position equation (3) is 61.6%. This matches the experimental
result (DTA–TGA, Fig. 2) of ca. 62%. The Li triborate (LiB3O5)
is formed as a non crystalline glass phase and cannot therefore be
identified by X-ray diffraction. However, Fig. 7 presents FTIR
spectra of pristine LiBOB, of a sample after heating to 350 ◦C,
and of Li oxalate for comparison. The spectrum of the decom-
posed sample clearly shows oxalate bands at 449, 516, 775,
1330, and 1665 cm−1, and three new bands at 1087, 1382 and
866 cm−1 (not related to oxalate). These peaks can be assigned
to the strong bands of LiB3O5 at 1372, 1087 and 860 cm−1 [13].
The ICP analysis of the solid product after heating LiBOB to
350 ◦C showed the presence of 9.3% Li and 13.9% B. This fits
well the composition expected for solid product suggested by
Eq. (3), namely a mixture of Li2C2O4 + LiB3O5 (9.4% Li and
14.5% B).
Fig. 8. FTIR spectrum of a gas phase that is formed in the ARC calorimetric
bomb due to a thermal decomposition of LiBOB (a). Spectra of CO2 (b) and CO
(c) are presented for comparison.
that of CO2 at room temperature [10]. The solid product of
LiBOB thermal decomposition was dissolved in distilled water
and analyzed by ICP and found to contain 9.3% Li and 13.9%
B.
The following experiment was carried out to confirm the
occurrence of the reaction postulated in Eq. (2). Mixtures con-
taining Li2C2O4:B2O3 1:3 mol/mol were thoroughly ground
in an agate mortar and pestle and were then subjected to
DSC in closed crucibles up to 500 ◦C. A pronounced broad
endothermic peak appears in the temperature range that over-
laps the range in which LiBOB decomposes (Fig. 3). The
onset of this peak is near 210 ◦C that is ca. 80 ◦C lower
than the onset of the LiBOB decomposition peak (Eq. (1)).
This temperature is far for the oxalate decomposition (above
400 ◦C) and the endotherm is attributed only to reaction (2).
Hence, reaction (2) begins immediately upon the formation
of Li2C2O4 and B2O3 by LiBOB decomposition (Eq. (1)).
The heat of the endothermic reaction between Li oxalate and
B2O3 was found to be about 67.9 kJ mol−1 of the product
LiB3O5. According to Eq. (3), only 1/3 mol of LiB3O5 is formed
theformationof1/3 molofLiB3O5. Thisvalueispartoftheover-
all heat of decomposition LiBOB, of namely about 38 kJ mol−1
(see Table 1). Consequently, the heat of the first decomposition
step of LiBOB, Eq. (1), is near 38 − 23 = 15 kJ mol−1. XRD
measurements of the product of heating the Li2C2O4–B2O3
mixture (up to 350 ◦C) showed no evidence of a crystalline
phase.
A mixture Li2O–B2O3 containing 75 mol% B2O3 should
undergo slow crystallization upon storage at elevated temper-
atures, above 500 ◦C [14]. The crystals of LiB3O5 were grown
on a seed from the solution-melt using the method of tempera-
ture lowering [15]. The end product of LiBOB decomposition
was stored at 300 ◦C during 10 days by further cooling at a rate
highest temperature at which a mixture containing Li oxalate
can be heated and remains stable). No evidence for the forma-
tion of any new crystalline phase, except the existing Li oxalate,
was observed by X-ray diffraction (Fig. 6c).
4. Discussion
Based on the above results and as detailed below, it is sug-
gested that the decomposition of LiBOB at ∼300 ◦C occurs as
follows:
6LiBC4O8(cryst) → 3Li2C2O4(cryst) + 3B2O3(glass)
+ 9CO(gas) + 9CO2(gas);
(1)
(2)
(3)
Li2C2O4(cryst) + 3B2O3(glass)
→ 2LiB3O5(glass) + CO(gas) + CO2(gas).
The overall thermal decomposition process of LiBOB is:
6LiBC4O8(cryst) → 2Li2C2O4(cryst) + 2LiB3O5(glass)
+ 10CO(gas) + 10CO2(gas).
The calculated weight loss for decomposition of LiBOB
according to Eq. (1) is 55.7%. However, the TGA and ARC
experiments show a greater weight loss upon thermal decompo-
sition. The TGA data (Fig. 2) show a weight loss of about 62%.
Clearly, an additional reaction involving gas evolution is tak-
ing place. This cannot be the decomposition of lithium oxalate
well as literature data [12] this occurs only above 400 ◦C. Fur-
thermore, FTIR spectra of the LiBOB samples after heating to
350 ◦C indicated the absence of B2O3 as a final decomposition
product (Fig. 7). The IR spectrum of the vitreous B2O3 contains
a stronger absorption band at ∼1265 cm−1, a weak absorption
band at ∼720 cm−1, and also a shoulder situated at ∼1400 cm−1
[11].
Hence, we suggest reaction (2) above as a follow up process
that well explains both the weight loss observed and the absence
of B2O3 in the end product.