1. Hydrogen Storage in LiBH 4 -MgH 2 -Al Bjarne R. S. Hansen a, Dorthe B. Ravnsbæk a, Jørgen Skibsted b, Carsten Gundlach c & Torben R. Jensen a a Center for Materials Crystallography and Department of Chemistry, University of Aarhus, Langelandsgade 140, DK-8000 Aarhus C, Denmark b Instrument Centre for Solid-State NMR Spectroscopy, Department of Chemistry, and Interdisciplinary Nanoscience Center (iNANO), Aarhus University, Langelandsgade 140, DK-8000 Aarhus C, Denmark c MAX-II laboratory, Lund University, Ole Römers väg 1, 223 63, S-22100 Lund, Sweden Abstract LiBH 4 is an interesting hydrogen storage material, as it has a high gravimetric hydrogen content of 18.5 wt% [1]. However, the utilization is hampered by lack of reversibility and high decomposition temperatures. One way of improving metalborohydrides, is the utilization of reactive hydride composites, which has be achieved by adding for instance Al [2] or MgH 2 [3]. By adding both MgH 2 and Al the LiBH 4 -MgH 2 -Al composite further improves the hydrogen release and uptake properties of LiBH 4 [4]. In this study the decomposition reactions of LiBH 4 -MgH 2 -Al in molar ratios (4:1:1) and (4:1:5) are investigated using in situ Synchrotron Radiation Powder X-ray Diffraction (SR-PXD) and thermal analysis (TGA/DSC) coupled with mass spectroscopy (MS), Sieverts Measurements (PCT), Fourier transform infrared spectroscopy (ATR-FTIR) and solid state 11 B Magic Angle Spinning Nuclear Magnetic Resonance ( 11 B MAS NMR) Acknowledgements and references Sincere acknowledgements are directed to iNANO, Danscatt, MAX-Lab Bor4Store (EU) and Center for Materials Crystallography (CMC) Decomposition reactions (LiBH 4 -MgH 2 -Al (4:1:5, S2)) MgH 2 decompose at T = 290 °C, but unlike the LiBH 4 -MgH 2 -Al (4:1:1) sample Mg 17 Al 12 is not observed. Mg/Mg 0.9 Al 0.1 is however observed after MgH 2 decomposition. The formation of MgAlB 4 occur at a lower temperature compared to (4:1:1), i.e. T = 300 °C, and seems to form in two different rates. Diffracted intensity from LiAl is observed at T = 410 °C, which is ~50 °C earlier than in (4:1:1). An unknown compound (1) is observed at T = 490 °C and is possibly a stable impurity, as it continues unaltered throughout the experiment. Mechanochemical Synthesis: Fritsch Pulverisette no. 4 with main disk speed: 400 rpm., Relative ratio: -2.25, Mill time: 5 min, Pause time: 2 min, Ball-to- Powder mass ratio: 35:1, Repetitions: 24 (Total mill time: 120 min). [1] L. Schlapbach, Nature 2009, 460, 809-811; [2] D. B. Ravnsbæk, T. R. Jensen, J. Appl. Phys. 2012, 111, 112621 ; [3] U. Bösenberg, S. Doppiu, L. Mosegaard, G. Barkhordarian, N. Eigen, A. Borgschulte, T. R. Jensen, Y. Cerenius, O. Gutfleisch, T. Klassen, M. Dornheim, Acta Materialia 2007, 55, 3951–3958; [4] Y. Zhang, Q. Tian, H. Chu, J. Zhang, L. Sun, J. Sun, Z. Wen, J. Phys. Chem. C 2009, 113, 21964–21969 Thermal analysis (comparison) TGA/DSC-MS analysis complement the decomposition pathway observed from in situ data well. No diborane was detected with MS during decomposition of either sample (not shown). At T = 300 °C the decomposition of MgH 2 is observed in sample LiBH 4 -MgH 2 -Al (4:1:1), but not in the (4:1:5)-sample, although a release of hydrogen is observed in the MS data for both samples. The formation of MgAlB 4 is observed as a broad endotherm and the formation of LiAl is seen at T = 410 °C. In LiBH 4 -MgH 2 -Al (4:1:1) and LiBH 4 -MgH 2 (2:1) the formation of LiMg is observed. This is however not observed in the in situ SR-PXD measurement (it was only heated to 500 °C) Decomposition reactions (LiBH 4 -MgH 2 -Al (4:1:1, S1)) Figures show SR-PXD of LiBH4-MgH2-Al (4:1:1). MgH 2 decompose at T = 300 °C, and react with Al to form Mg 17 Al 12 and Mg 0.9 Al 0.1. MgAlB 4 is observed at T = 380 °C. This formation might be slow, since diffracted intensity from Mg 0.9 Al 0.1 increase as Mg 17 Al 12 decompose. LiAl is observed at T = 460 °C. From the molar composition LiAl should not form. However, since the formation of MgAlB 4 is relatively slow, it is possible that LiAl forms instead since both Al and Li are present. Temperature (°C)T onset T peak T end LiBH 4 -Al (2:1)390425470 LiBH 4 -Al (2:3)325375450 LiBH 4 -MgH 2 (2:1)350420440 (565) LiBH 4 -MgH 2- Al (4:1:1)275415440 (550) LiBH 4 -MgH 2- Al (4:1:5)275350/400485 λ = 1.103671 Åλ = 1.102050 Å Mg/Mg 0.9 Al 0.1 and Al Bragg peak overlap * = spinning sidebands/satellite transitions LiBH 4 -MgH 2 -Al (4:1:5) After 3 PCT cycles λ = 1.54056 Å 11 B MAS NMR S2, LiBH 4 -MgH 2 -Al (4:1:5) after 3 PCT cycles Centerband resonance from LiBH 4 (-41.3 ppm) Centerband resonance consistent with Li 2 B 12 H 12 (-9 ppm) * * * * * * * * * ** * Comparison with the LiBH 4 -Al and LiBH 4 -MgH 2 samples show that the onset temperature for hydrogen release is lowered up to ~75 °C. Reversibility PCT desorptions of the LiBH 4 -MgH 2 -Al samples show a decrease in hydrogen storage capacity. In S2 (4:1:5) three gas release steps are observed. The release related MgAlB 4 is decreasing the most. Hence, the capacity drop is likely related to a boron-containing species. This is also the case in S1 (4:1:1), but LiAl is not observed. The 11 B MAS NMR spectra of S2 after three hydrogen release and uptake cycles show a centerband resonance from LiBH 4, and a chemical shift consistent with Li 2 B 12 H 12 which was not detected by PXD. This is also the case in S1. Li 2 B 12 H 12 is a stable closo-borate and it does not seem to react/rehydrogenate at the physical conditions used in this study. Li 2 B 12 H 12 was also observed by FT-IR in the cycled samples. Diffractograms were obtained after the first desorption and first absorption. These diffractograms confirm the reactions proposed from the PCT desorptions. Unknown 1 is also observed after absorption. Diffracted intensity from LiBH 4 is weak compared to the diffractograms of the ball milled samples. S1: LiBH 4 -MgH 2 -Al (4:1:1) S2: LiBH 4 -MgH 2 -Al (4:1:5) LiBH 4 -MgH 2 -Al (4:1:1) After 3 PCT cycles λ = 1.54056 Å More information: brsh@chem.au.dk λ = 1.54056 Å LiBH 4 -MgH 2 -Al (4:1:1) More Al  lower T onset No B 2 H 6 detected ATR-FTIR Hence, Li 2 B 12 H 12 is likely responsible for the hydrogen storage capacity loss in the samples. λ = 1.54056 Å LiBH 4 -MgH 2 -Al (4:1:5)