ISSN 1070-3632, Russian Journal of General Chemistry, 2015, Vol. 85, No. 11, pp. 2505–2508. © Pleiades Publishing, Ltd., 2015. Original Russian Text © A.V. Butlak, Yu.V. Kondrat’ev, A.S. Mazur, A.Yu. Timoshkin, 2015, published in Zhurnal Obshchei Khimii, 2015, Vol. 85, No. 11, pp. 1761–1764.
Thermal Decomposition of Ammonia Borane at 357 K A. V. Butlak, Yu. V. Kondrat’ev, A. S. Mazur, and A. Yu. Timoshkin St. Petersburg State University, Universitetskaya nab. 7–9, St. Petersburg, 199034 Russia e-mail:
[email protected] Received July 18, 2015
Abstract—The heat effect of ammonia borane decomposition at 357 K under atmospheric pressure in glass and copper cells has been determined by calorimetry. The enthalpy of BH3NH3 heat-induced decomposition in glass ampoules is of –24.8±2.3 kJ/mol; this value includes the structural reorganization into diammoniate of diborane of –4.0±2.8 kJ/mol and the thermal effect of decomposition into polyamidoborane of –20.8±1.6 kJ/mol. In the case of experiments in copper cells a much higher (–47.6±8.3 kJ/mol) heat evolution is observed. The solidstate 15N and 11B NMR analysis of samples after the calorimetric experiment has shown that partial oxidation of the sample into boric acid occurs in copper cells. Keywords: ammonia borane, thermal decomposition, calorimetry, solid-state NMR
DOI: 10.1134/S1070363215110018 Thermochemical properties of ammonia borane are of interest for the hydrogen energetics in connection with high hydrogen content in this compound (19.6 wt %) [1]. It has been found that ammonia borane thermally decomposes evolving hydrogen and forming polymeric amidoborane (BH2NH2)n [2–5]. BH3NH3(s) = 1/n(BH2NH2)n(s) + H2(gas).
(1)
Enthalpy of reaction (1) at 343–363 K has been determined as of ∆1H0353 = –21.7±1.2 kJ/mol via differential scanning calorimetry (DSC) [4]. Certain salt additives (CuCl2, NiCl2, and CoCl2) affect the kinetics of hydrogen evolution from ammonia borane [6, 7]. In view of that, it is important to elucidate the influence of a cell material on the ammonia borane thermal decomposition. In this work we determined the heat effects of ammonia borane decomposition in glass and copper cells at 357 K and atmospheric pressure via the direct calorimetry. To do so, an ampoule with a weighed amount of ammonia borane was dropped in a preheated calorimetric cell. The resulting thermograms are presented in Fig. 1. In the experiments with glass ampoules (Fig. 1a), the endothermic effect due to the ampoule and the ammonia borane sample heating to the calorimetric cell temperature was observed, followed by an induction period (1–2 h), and the exothermic process assigned to the ammonia borane decomposition.
In the experiments with copper ampoules, a weak exothermic effect was observed immediately after the endothermic effect due to the ampoule and the sample heating, and then the induction period and a more pronounced exothermic effect were recorded, similarly to the case of the glass ampoule (Fig. 1b). The heat of the decomposition process Q was calculated as a difference between the measured gross effect Qmeas and the heat amounts spent on the heating of the ampoule (Qamp) and ammonia borane sample (QAB, estimated using thermal capacity of solid ammonia borane of 75.37 J mol–1 K–1 at 298.15 K [8]). The measurements results are presented in Tables 1 and 2 (with m, the sample mass; t, the experiment duration; ΔH0357, the standard enthalpy of ammonia borane decomposition). As the effects of the ampoule and the sample heating were separated from that of ammonia borane decomposition by the induction period, we directly integrated the second exothermic effect as well (Q'). The average value of the gross effect was of –24.8±2.3 kJ/mol in the experiments in glass ampoules, within experimental errors in agreement with the DSC data (–21.7±1.2 kJ/mol) [4]. When the second exothermic peak was directly integrated, the decomposition enthalpy was of –20.8±1.6 kJ/mol. The difference of –4.0±2.8 kJ/mol was assigned to the heat effect of ammonia borane structural reorganization into diammoniate of diborane [BH2(NH3)2]+[BH4]– during
2505
2506
BUTLAK et al. (b)
Heat flow, mW
Heat flow, mW
(a)
t, h
t, h
Fig. 1. Typical calorimetric curves for the experiments in (a) glass and (b) copper ampoules.
Table 1. Results of drop-calorimetric experiments with BH3NH3 in glass ampoules (357 K, 1 atm) m, mg
t, h
Qmeas, J
Qamp, J
QAB, J
Q, J
∆H0357, kJ/mol
Q', J
∆H '3057, kJ/mol
12.1 16.9 15.6 21.3
8.2 9.6 8.3 9.5
9.90 3.84 7.32 6.30
18.08 16.50 16.39 19.07
1.89 2.64 2.44 3.33
–10.08 –15.30 –11.51 –16.10
–25.72 –27.95 –22.78 –23.33
–8.33 –11.89 –9.28 –15.07
–21.25 –21.72 –18.36 –21.85
Average
–24.8±2.3
–20.8±1.6
Table 2. Results of drop-calorimetric experiments with BH3NH3 in copper ampoules (357 K, 1 atm) m, mg
t, h
Qmeas, J
Qamp, J
QAB, J
Q, J
∆H0357, kJ/mol
Q', J
∆H '3057, kJ/mol
28.3 40.0 25.6 25.6 30.3 35.2 16.5
9.1 8.1 8.3 8.7 8.5 8.5 8.7
–34.00 –28.35 –35.00 –35.12 –41.40 –36.68 –18.83
6.09 6.08 6.03 6.54 5.39 6.00 4.69
4.42 6.25 4.00 4.00 4.73 5.50 2.54
–44.51 –40.68 –45.03 –45.66 –51.53 –48.18 –26.06
–48.57 –31.41 –54.32 –55.07 –52.52 –42.26 –48.77
–25.92 –25.04 –21.95 –22.68 –27.93 –33.18 –10.88
–28.28 –19.33 –26.48 –27.35 –28.46 –29.11 –20.36
Average
–47.6±8.3
the induction period. According to the 11В NMR spectroscopy data, [BH2(NH3)2]+[BH4]– was an intermediate of the ammonia borane decomposition formed during the induction period at 361 K [9]. According to the DSC data, the enthalpy of the first stage of diammoniate of diborane thermal decomposition was of –34 kJ per 1 mole of [BH2(NH3)2]+[BH4]– [10]. Assuming that products of the first stage of the decomposition were identical in the cases of ammonia
–25.6±4.0
borane and [BH2(NH3)2]+[BH4]–, the enthalpy of structural reorganization of ammonia borane calculated from the data given in [4, 10] was of –4.7 kJ/mol, coinciding with the value –4.0±2.8 kJ/mol determined in this work. It should be pointed out that the average grosseffect for the experiments in copper ampoules was much more exothermic (–47.6±8.3 kJ/mol), whereas
RUSSIAN JOURNAL OF GENERAL CHEMISTRY Vol. 85 No. 11 2015
THERMAL DECOMPOSITION OF AMMONIA BORANE AT 357 K
2507
the integration of the second exothermic peak gave –25.6±4.0 kJ/mol (the high error was due to the baseline uncertainty). The observed difference (–22± 9 kJ/mol) indicates that additional exothermic process (or processes) being operative in the case of the experiments with copper ampoules, on top of the structural reorganization of ammonia borane into [BH2(NH3)2]+[BH4]– and its thermal decomposition into polymeric amidoborane. IR spectra of the residual after the experiments both in glass and in copper ampoules coincided with the reference polyamidoborane spectrum [3], confirming that the thermal decomposition of ammonia borane followed equation (1). The X-ray diffraction analysis of the samples after the calorimetric measurements revealed that they were X-ray amorphous, indicating the formation of polymeric amidoborane as well. More detailed information on the decomposition products was gained using the solid-state 11B and 15N NMR spectroscopy at room temperature. 15
N NMR spectrum of the starting ammonia borane contained a narrow peak centered at 16.2 ppm, typical of the boron-containing adducts with ammonia [11, 12]. 15N NMR spectrum of the specimens after calorimetric experiments exhibited a broad peak centered at 23.2 ppm, typical of the polymeric compounds with the 15N atom in various surroundings. 11
B NMR spectrum of ammonia borane contained a single asymmetric peak centered at –28.5 ppm, close to the reference values of –23 [12], –25.3 [13], and –24.0 ppm [9]. 11B spectrum of the thermal decomposition products (Fig. 2) contained peaks with maxima at –37 (BH4), –25 (BH3), –11 (BH2), and 0.7 [B(OH)3] along with a weak peak at 10 ppm (presumably BH). The reference data gave the chemical shifts of 37.6, –23.2, and –13.3 ppm for BH4, BH3, and BH2 groups, respectively [13]. The observed spectral pattern was caused by the presence of [BH4]– anions, BH3 terminal groups, and [NH3(BH2NH2)nBH2NH3]+ polymeric cationic species in polyamidoborane [14]. Noteworthily, the NMR spectrum of starting ammonia borane contained a very weak signal at 0.7 ppm, assigned to the presence of small amounts of B(OH)3. In our opinion, it may originate from slow ammonia borane oxidation with air oxygen during preparation of the sample for NMR experiment. It also explains the presence of a weak signal at 0.7 ppm in the spectra of the samples after the calorimetric experiments in glass ampoules (Fig. 2).
δ, ppm 11
Fig. 2. B NMR spectrum of the samples after the calorimetric experiment in (1) copper and (2) glass ampoules.
The significantly stronger NMR signal at 0.7 was recorded for the sample after the experiment in a copper ampoule, evidencing about the boric acid formation. At the same time, the integral intensity of the peak at –37 ppm (BH4) decreased, whereas the integral intensity of the peaks at –25 (BH3) and –11 ppm (BH2) remained almost constant. Hence, the noticeably increased amount of boric acid was observed in the experiments with copper ampoules, suggesting the oxidation of ammonia borane: 3CuO + BH3NH3 = Cu + B(OH)3 + NH3 (oxide film at the copper ampoule surface acted as the oxidizer). Therefore, the value of the gross effect Q (Table 2) corresponded to the sum of two exothermic processes, the decomposition and the oxidation by a copper ampoule surface. In summary, direct calorimetric determination of the enthalpy of ammonia borane thermal decomposition at 357 K and atmospheric pressure demonstrated that partial oxidation of ammonia borane into boric acid was operative along with the thermal decomposition in the case of the experiments in copper ampoules. The overall value of the thermal decomposition enthalpy of BH3NH3 in glass ampoules was of Δ1H0357(BH3NH3) = –24.8±2.3 kJ/mol; that value included the structural reorganization into diammoniate of diborane (–4.0±2.8 kJ/mol) and the heat effect of the decomposition into polyamidoborane (–20.8±1.6 kJ/mol). EXPERIMENTAL Ammonia borane (97%, Aldrich) was purified via recrystallization from dry diethyl ether. Purity of the compound was confirmed by X-ray diffraction analysis and IR spectroscopy. The drop-calorimetric method (a DAK1-1a calorimeter with a modified system of the
RUSSIAN JOURNAL OF GENERAL CHEMISTRY Vol. 85 No. 11 2015
2508
BUTLAK et al.
sample inlet) was used. The experimental technique has been described in detail in [15]. The experi-ments were carried out under argon atmosphere. Glass ampoules (height of 2–3 cm, diameter of 1 cm, and wall thickness of 1 mm) and copper ampoules (height of 1.5 cm, diameter of 0.5 cm, and wall thickness of approximately 0.2 mm) were used as the cells. The heat capacity of each ampoule was measured in the course of the reference experiments.
4. Wolf, G., Baumann, J., Baitalow, F., and Hoffmann, F.P., Thermochim. Acta, 2000, vol. 343, nos. 1–2, p. 19. DOI: 10.1016/S0040-6031 (99 00365-2.
The decomposition products were identified by spectral methods. IR spectra (KBr) were measured using a Fourier IR Prestige-21 Shimadzu spectrophotometer. Solid-state 11B and 15N NMR spectra were registered at room temperature with a Bruker AVANCE III 400 [128.41 (11B) and 40.55 MHz (15N)] instrument. The solid samples were centrifuged at 12 kHz using a 4 mm rotor. Aqueous 1 М H3BO3 solution (11B) and solid NH4NO3 (15N) were used as reference.
7. He, T., Xiong, Z., Wu, G., Chu, H., Wu, C., Zhang, T., and Chen, P., Chem. Mater., 2009, vol. 21, no. 11, p. 2315. DOI: 10.1021/cm900672h. 8. Wolf, G., van Miltenburgb, J.C., and Wolf, U., Thermochim. Acta, 1998, vol. 317, no. 2, p. 111. DOI: 10.1016/ S0040-6031 (98 00381-5. 9. Stowe, A. C., Shaw, W. J., Linehan, J. C., Schmid, B., and Autrey, T., Phys. Chem. Chem. Phys., 2007, vol. 9, p. 1831. DOI: 10.1039/b617781f.
ACKNOWLEDGMENTS This work was performed with financial support from the Russian Scientific Foundation (project no. 1413-00151) using the equipment of the Research Center of St. Petersburg State University “Research Resources Center for Magnetic Resonance.” REFERENCES 1. Jena, P., J. Phys. Chem. Lett., 2011, vol. 2, p. 206. DOI: 10.1021/jz1015372. 2. Baitalow, F., Baumann, J., Wolf, G., Jaenicke-Roβlerb, K., and Leitner, G., Thermochim. Acta, 2002, vol. 391, nos. 1–2, p. 159. DOI: 10.1016/S0040-6031(02)00173-9. 3. Baumann, J., Baitalow, F., and Wolf G., Thermochim. Acta, 2005, vol. 430, nos. 1–2, p. 9. DOI: 10.1016/ j.tca.2004.12.002.
5. Storozhenko, P.A., Svitsyn, R.A., Ketsko, V.A., Buryak, A.K., and Ul’yanov, A.V., Russ. J. Inorg. Khim., 2005, vol. 50, no. 7, p. 980. 6. Kalidindi, S.B., Joseph, J., and Jagirdar, B.R., Energy Environ. Sci., 2009, vol. 2, p. 1274. DOI: 10.1039/ B909448M.
10. Bowden, M., Heldebrant, D. J., Karkamkar, A., Proffen, T., Schenter, G.K., and Autrey, T., Chem. Commun., 2010, vol. 46, p. 8564. DOI:10.1039/c0cc03249b. 11. Gunaydin-Sen, O., Achey, R., Dalal, N. S., Stowe, A., and Autrey, T., J. Phys. Chem. B, 2007, vol. 111, p. 677. DOI: 10.1021/jp0649347. 12. Gervais, C., Babonneau, F., Maquet, J., Bonhomme, C., Massiot, D., Framery, E., and Vaultier, M., Magn. Res. Chem., 1998, vol. 36, p. 407. DOI: 10.1002/(SICI)1097458X (199806)36:6 3.0.CO;2-Y. 13. Shimoda, K., Doi, K., Nakagawa, T., Zhang, Y., Miyaoka, H., Ichikawa, T., Tansho, M., Shimizu, T., Burrell, A.K., and Kojima, Y., J. Phys. Chem. C, 2012, vol. 116, p. 5957. DOI:10.1021/jp212351f. 14. Staubitz, A., Robertson, A.P., and Manners, I., Chem. Rev., 2010, vol. 110, p. 4079. DOI: 10.1021/cr100088b. 15. Butlak, A V., Kondrat’ev, Yu V., and Timoshkin, A.Y., Russ. J. Gen. Chem., 2014, vol. 84, no. 12, p. 2455. DOI: 10.1134/S1070363214120184.
RUSSIAN JOURNAL OF GENERAL CHEMISTRY Vol. 85 No. 11 2015
