Synthesis of sodium amidoborane (NaNH₂BH₃) for hydrogen production

Abstract

The prospect of building a future energy system on hydrogen has stimulated much research effort in developing hydrogen storage technologies. One of the potential materials newly developed is sodium amidoborane (NaNH₂BH₃) which evolves ∼7.5 wt% hydrogen at temperatures as low as 91 °C. In this paper, two methods of synthesizing pure NaNH₂BH₃ were reported. One method is by reacting NaH and ammonia borane in THF at low temperatures, and the other is by reacting NaNH₂ and ammonia borane in THF at ambient temperature. Non-isothermal testing on the thermolysis of solid NaNH₂BH₃ showed that hydrogen evolution was composed of two exothermic steps. More than 1 equiv. H₂ was evolved rapidly at temperatures below 87 °C. After evolving 2 equiv. H₂, NaH was identified in solid products and coexisted with amorphous BN.

---

Introduction

One of the pending issues in the implementation of hydrogen fuel cell technology in the automobile industry is the lack of safe and highly efficient hydrogen storage materials.^1,2 More research efforts in material development are now applied to chemicals comprising light elements and possessing high hydrogen content.^3–8 Because of its high hydrogen capacity (19.6 wt%) as well as its stability in air and non-toxicity, ammonia borane (NH₃BH₃) was considered as one of the most promising candidates for future transportation applications.^9,10 The thermal decomposition of NH₃BH₃ is composed of three steps evolving one equivalent H₂ per step. In the first two steps, ∼2 equiv. H₂ are evolved at temperatures of ca. 110 and ca. 150 °C,^11,12 which are higher than the operation temperature of PEM (Proton Exchange Membrane) fuel cells. Another issue with NH₃BH₃ is that the hydrogen evolution is accompanied by the production of borazine, which is a volatile compound and will poison fuel cells. In an attempt to enhance the performance of thermolysis, NH₃BH₃ was deposited into mesoporous silica scaffolds and the onset temperature for hydrogen production was found to be reduced by approximately 15 °C.¹³ Furthermore, the formation of borazine was largely suppressed. Similar improvement was achieved by loading NH₃BH₃ on carbon cryogels.¹⁴ We have recently reported a derivative of ammonia borane, i.e., sodium amidoborane, NaNH₂BH₃, obtained from the solid-state interaction between NH₃BH₃ and sodium hydride (NaH). Sodium amidoborane evolves ∼7.4 wt% hydrogen at ca. 91 °C, considerably lower than that of neat NH₃BH₃.¹⁵ In the meantime, borazine is undetectable. Since NH₃BH₃ is plastic at room temperature, the mechano-synthetic route for NaNH₂BH₃ needs the help of inert additives, which enhance the efficiency of mechanical milling.¹⁵ However, the existence of these additives in the sample lowered the system gravimetric density of hydrogen. In this study, a wet-chemical route was developed to synthesize pure NaNH₂BH₃. NH₃BH₃ was first dissolved in THF and then reacted with NaH or sodium amide (NaNH₂) at given temperatures. H₂ or NH₃ was generated during the reaction. Upon completion of the reaction, highly pure NaNH₂BH₃ can be obtained by removing THF. Subsequent hydrogen evolution from the NaNH₂BH₃ synthesized was investigated by Temperature Programmed Desorption (TPD), volumetric release and XRD; 7.5 wt% of hydrogen was found to be released from the sample. NaH was identified in the fully dehydrogenated solid residue.

Experimental

Synthesis of NaNH₂BH₃

NaH (95%) and NaNH₂ (95%) were Sigma Aldrich products and used as received. NH₃BH₃ was synthesized on-site according to the literature.¹⁶ The resulting solid was dissolved in tetrahydrofuran (THF) and characterized by ¹¹B NMR. A single –BH₃ resonance was observed at δ −22.1, agreeing well with reference data.¹⁷ The crystal structure of the solid NH₃BH₃ was further confirmed by XRD analysis. The purity of synthesized NH₃BH₃ was estimated at 95%. The chemical reaction of NH₃BH₃ with NaH or NaNH₂ in THF was conducted in a commercial Stirring Tank Reactor (STR). 0.008 mol NaH or NaNH₂ and 0.008 mol NH₃BH₃ were added to 30 ml THF and stirred at constant temperature. As gas was evolved, the pressure variation with time inside the reactor was recorded automatically. At the end of the reaction, gaseous products were analyzed by a Mass Spectrometer (MS). The reaction solution was filtered and evaporated under reduced pressure at ambient temperature for crystallization.

Characterization

Hydrogen evolution from the synthesized NaNH₂BH₃ was conducted on a custom-made Temperature Programmed Desorption (TPD) system. A Gas Chromatograph (GC)–Mass Spectrometer (MS) combined system was employed to trace the gaseous products. Detailed operation was described elsewhere.¹⁸ A 20 mg sample was ramp-heated in argon at 2 K min⁻¹. Volumetric release for the quantitative measurement of hydrogen evolution was performed on a commercial gas reaction controller provided by Advanced Materials Corporation. Approximately 300 mg of sample was employed for each experiment and the heating rate was 1 K min⁻¹. A Netzsch DSC 204 HP unit was applied to detect heat effects in the evolution process.

¹¹B {¹H} magic-angle-spinning Nuclear Magnetic Resonance (NMR) experiments were carried out at room temperature on a Bruker Avance 400 NMR spectrometer operating at 9.7 T on 128.3 MHz ¹¹B frequency. All chemical shifts are reported in ppm downfield.

Structural identifications were performed on a Bruker X-ray Diffractometer (XRD) equipped with an in situ cell. X-ray diffraction data were collected from 10 to 60° in 2θ with a scan step-width of 0.01° using Cu Kα radiation.

Results and discussion

Gas was evolved vigorously after dropping NaH into the NH₃BH₃–THF solution. As can be seen from Fig. 1, gas evolution was completed within 10 min at −3 °C. Mass spectrometer analysis showed that hydrogen was the only gaseous product. By applying the ideal equation of state, the amount of hydrogen evolved was ca. 1 equiv. H₂ per [NaH–NH₃BH₃] mixture. NaH solid was completely consumed and a transparent solution was obtained at the end of the reaction. NMR characterization of the B species in the solution revealed that –BH₃ with a chemical shift at −21.6 ppm was the dominant species. The resonance of –BH₃ in NH₃BH₃ at −22.1 ppm weakened significantly (see Fig. 2). After removing THF at ambient temperature, a white powder with an orthorhombic structure (group Pbca) with lattice constants a = 7.4735 Å, b = 14.6452 Å, c = 5.6739 Å, was obtained, which is identical to the unit cell of NaNH₂BH₃ synthesized by mechano-chemical methods.¹⁵ Obviously, NaH and NH₃BH₃ converted to NaNH₂BH₃ and H₂ in THF through reaction (1).

NaH (s) + NH₃BH₃–THF (l) → NaNH₂BH₃-THF (l) + H₂ (g) (1)

Fig. 1 Time dependence of gas evolution from NaH–NH₃BH₃–THF and NaNH₂–NH₃BH₃–THF solutions at temperatures of −3 °C and 25 °C, respectively.

Fig. 2 ¹¹B NMR spectra obtained at the end of gas evolution from NaH–NH₃BH₃–THF and NaNH₂–NH₃BH₃–THF solutions.

Comparatively, the chemical reaction between NaNH₂ and NH₃BH₃ in THF was slow. Only 6 psi pressure inside the reactor was detected after the solution was stirred at 25 °C for 20 h (see Fig. 1). It was detected by MS that ammonia (NH₃) was the dominant product. Observation of the transparent solution revealed the consumption of NaNH₂. NH₃BH₃ was fully consumed as only the –BH₃ resonance at −21.6 ppm was detected by NMR (see Fig. 2). Removal of the solvent through distillation led to the formation of NaNH₂BH₃ powder, which suggested the occurrence of reaction (2).

NaNH₂ (s) + NH₃BH₃–THF (l) → NaNH₂BH₃–THF (l) + NH₃ (g) (2)

The weight difference between NaNH₂ plus NH₃BH₃ (as starting chemicals) and the resulting solid after the removal of THF was found to be equal to the amount of NH₃ evolution (ca. 1 equiv NH₃), further validating reaction (2). The low NH₃ pressure inside the reactor is due to the solubility of NH₃ in THF. It is rather interesting to investigate the mechanism of reaction (2). It could be a direct substitution of NH₃ in NH₃BH₃ by NaNH₂ or by metathesis of Na in NaNH₂ with H in NH₃ in NH₃BH₃. It is also worthy of emphasis that the interaction between NaNH₂ and NH₃BH₃ provides a facile pathway for the synthesis of high purity metal amidoboranes through the reaction of other amides and NH₃BH₃. As a matter of fact, LiNH₂BH₃ can be synthesized by reacting LiNH₂ and NH₃BH₃ at room temperature.

Illustrated in Fig. 3 are TPD, DSC and volumetric release measurements on hydrogen evolution from solid-state NaNH₂BH₃. It can be seen that the TPD-H₂ spectrum exhibits three main evolution steps. A burst was observed at 87 °C, and then broad peaks were found at ca. 174 °C and 325 °C, respectively. Borazine was undetectable, agreeing well with our previous investigations.¹⁵ DSC measurement revealed that the hydrogen evolution in the first two steps was exothermic. An endothermic signal at 57 °C is due to the melting of NaNH₂BH₃. In our previous study, NaNH₂BH₃ produced by a mechano-chemical process was of smaller particle size so that it exhibited better kinetics during dehydrogenation. Quantitative measurements on hydrogen evolution from NaNH₂BH₃ showed that the first and second steps evolved 4.4 wt% H₂ (2.3 equiv. H) and 3.1 wt% H₂ (1.7 equiv. H), respectively. Therefore, ca. 2 equiv. H₂ were detached from one NaNH₂BH₃ molecule at temperatures below 200 °C. By holding the reaction temperature at ca. 91 °C for longer period of time, almost all H₂ can be evolved. As hydrogen evolution in the first two steps is exothermic, the re-hydrogenation may not be thermodynamically favored. Our attempt to re-charge the dehydrogenated product at 120 bar hydrogen was not successful. The third evolution step evolved ca. 1 equiv. H. Due to its high operation temperature this step is not considered for practical usage.

Fig. 3 TPD, DSC and volumetric release measurements on synthesized NaNH₂BH₃ solid.

To gain insight into the hydrogen evolution from NaNH₂BH₃, we collected the samples upon decomposition at 90 °C and 200 °C, respectively, for XRD characterization. As shown in Fig. 4, no phase can be identified from the solid residue collected after the first desorption step, indicating the amorphous nature of the dehydrogenation product. The sample collected at the end of the second desorption step presented a NaH phase, which reminds us that the third desorption step should be with the decomposition of NaH. In fact, our previous investigations showed that NaH decomposed to H₂ at ca. 320 °C.¹⁹ Given that 1 equiv. H was evolved in the third desorption step, one unit of NaH should be in the residue after the second desorption step. Therefore, together with 2 equiv. H₂ evolution in the first two steps, thermolysis of NaNH₂BH₃ at temperatures below 200 °C can be described by reaction (3).

nNaNH₂BH₃ → nNaH + nBN + 2nH₂ 7.55 wt% H₂ (3)

Fig. 4 XRD patterns of a) synthesized NaNH₂BH₃ and its dehydrogenation product collected at b) 90 °C, c) 200 °C.

BN should be in an amorphous form as its crystallization requires high temperatures.²⁰

By summarizing reactions (1) and (3), one could have an overall reaction (4), which allows more than 10.9 wt% of hydrogen to be desorbed from NaH and NH₃BH₃.

NaH + NH₃BH₃ → NaNH₂BH₃ + H₂ → NaH + BN + 3 H₂ 10.9 wt% H₂ (4)

Due to the exceptionally high hydrogen content and low reacting temperatures, the NaNH₂BH₃ and NaH–NH₃BH₃ systems are potential candidates for on-board hydrogen storage and, therefore, are worthy of further investigation.

Conclusion

Sodium amidoborane (NaNH₂BH₃) was synthesized through the reaction of NaH and NH₃BH₃ or NaNH₂ and NH₃BH₃ in THF. Investigations on sodium amidoborane revealed that, at temperatures below 200 °C, its thermolysis was composed of two steps. More than 7.5 wt% of hydrogen can be evolved. NaH and BN are the solid products.

Acknowledgements

The present work is financially supported by the Agency for Science, Technology & Research (A*STAR) in Singapore. Z. Xiong, G. Wu and P. Chen would like to thank the Dalian Institute of Chemical Physics (DICP) in China for startup funding support on laboratory building.

References

1. L. Schlapbach and A. Zuttel, Nature, 2001, 414, 353.
2. W. Grochala and P. P. Edwards, Chem. Rev., 2004, 104, 1283.
3. B. Bogdanovic and M. Schwickardi, J. Alloys Compd., 1997, 1, 253.
4. A. Zuttel, P. Wenger, S. Rentsch, P. Sudan, P. Mauron and C. Emmenegger, J. Power Sources, 2003, 118, 1.
5. P. Chen, Z. T. Xiong, J. Z. Luo, J. Y. Lin and K. L. Tan, Nature, 2002, 420, 302.
6. Z. T. Xiong, G. T. Wu, J. J. Hu and P. Chen, Adv. Mater., 2004, 16, 1522.
7. Z. T. Xiong, G. T. Wu, J. J. Hu, Y. F. Liu, P. Chen, W. F. Luo and J. Wang, Adv. Funct. Mater., 2007, 17, 1137.
8. F. E. Pinkerton, G. P. Meisner, M. S. Meyer, M. P. Balogh and M. D. Kundrat, J. Phys. Chem. B, 2004, 109, 6.
9. T. B. Marder, Angew. Chem., Int. Ed., 2007, 46, 8116.
10. F. H. Stephens, V. Pons and R. T. Baker, Dalton Trans., 2007, 2613.
11. V. Sit, R. A. Geanangel and W. W. Wendlandt, Thermochim. Acta, 1987, 113, 379.
12. M. G. Hu, R. A. Geanangel and W. W. Wendlandt, Thermochim. Acta, 1978, 23, 249.
13. A. C. Stowe, W. J. Shaw, J. C. Linehan, B. Schmid and T. Autrey, Phys. Chem. Chem. Phys., 2007, 1831.
14. A. Feaver, S. Sepehri, P. Shamberger, A. Stowe, T. Autrey and G. Cao, J. Phys. Chem. B, 2007, 111, 7469.
15. Z. T. Xiong, C. K. Yong, G. T. Wu, P. Chen, W. Shaw, A. Karkamkar, T. Autrey, M. O. Jones, S. R. Johnson, P. P. Edwards and W. I. F. David, Nat. Mater., 2008, 7, 138.
16. P. V. Ramachandran and P. D. Gagare, Inorg. Chem., 2007, 46, 7810.
17. J. S. Wang and R. A. Geanangel, Inorg. Chim. Acta, 1988, 148, 185.
18. Z. T. Xiong, P. Chen, G. T. Wu, J. Y. Lin and K. L. Tan, J. Mater. Chem., 2003, 13, 1.
19. Z. T. Xiong, G. T. Wu, J. J. Hu and P. Chen, Catal. Today, 2007, 120, 287.
20. R. T. Paine and C. K. Narula, Chem. Rev., 1990, 90, 73.

---