Search

Hydrogen desorption from the mixture of Mg(NH2)2 and 2LiH presents two pressure-dependent segments at a given temperature – ∼2/3 of the hydrogen is released at relatively higher pressures and forms a pressure plateau and the other 1/3 is desorbed at lower pressures. The overall reaction heat measured in a differential scanning calorimeter is 44.1 kJ/mol H 2, while the heat-of-desorption of H2 in the higher pressure plateau is about 38.9 kJ/mol H2, a favorable thermodynamics for PEM Fuel Cell application. However, the relatively higher activation energy (Ea = 102 kJ/mol) sets a kinetic barrier. © 2005 Elsevier B.V. All rights reserved.

High energy density hydrogen storage is a critical technology requirement in a hydrogen-based energy infrastructure. Although there are no current storage methods that meet desired energy density goals for vehicular hydrogen storage, complex metal hydride based systems are among the most promising. These materials form compounds with hydrogen under appropriate conditions and release hydrogen by thermal decomposition. The complex hydride, sodium alanate, is particularly useful due to its favorable reversibility. Thermal properties characterization of sodium alanate has been performed at Sandia National Laboratories to gain a detailed understanding of how complex hydrides will behave in a storage system. Thermal properties were investigated using the thermal probe method (ASTM D5334). Custom test hardware was designed and built to accommodate the complex decomposition and recombination of sodium alanate. Thermal conductivity and thermal wall resistance were determined by utilizing analytical and numerical data analysis methods. The thermal conductivity of sodium alanate was found to vary by more than 90% with changes in phase composition and hydrogen gas pressures between 1 and 100 atm. The quality of thermal contact between the alanate and the vessel wall was characterized numerically for various pressures and phase compositions. The contact resistance is high for all states, indicating poor contact between the material and the vessel wall.

An overview is presented of recent advances in the development of new and improved alanates for applications and in the fundamental understanding of how Ti-doping enhances hydrogen absorption. Sample materials were produced using approaches based on direct-synthesis and dry Ti-doping methods. It is desirable to introduce Ti through non-reactive processes to avoid the hydrogen capacity loss that occurs through the formation of inactive byproducts (for example Na–halide from the decomposition of Ti–halides and Na–oxides from the decomposition of Ti–alkoxides). We show, for the first time, that alanates can be Ti-doped using TiH2 or through indirect-doping by pre-reacting TiCl2 with LiH. Both methods result in enhanced kinetics. However, improved rates were achieved only after a prolonged activation period of about a 10 cycles, suggesting that cycling leads to Ti diffusion and substitution into the alanate lattice which provides the mechanism through which Ti-doping enhances kinetics. Thus, the reactive decomposition of Ti–halide and alkoxide precursors in the doping process serves an important but not necessarily required function.

The kinetics of absorption and desorption of hydrogen from NaAlH 4 have previously been shown to improve upon the addition of a catalyst-precursor such as TiCl 3 . In this paper we demonstrate that TiCl 2 , TiF 3 , and TiBr 4 all effectively improve sorption kinetics. Arrhenius data indicate that the catalyst precursors behave in essentially the same manner. Evidently the valency of Ti in the catalyst-precursor is inconsequential to the role of Ti in altering the kinetic mechanism. The formation of TiAl 3 on doping with TiCl 3 has been observed. The presence of TiAl 3 appears to contribute in part to the enhanced kinetics in these systems.

The reversible hydrogen storage properties of Ti-catalyzed NaAlH 4 (and associated Na 3 AlH 6 ) were studied as a function of Ti-content using a dry preparation technique consisting of the ball-milling of NaAlH 4 +TiCl 3 mixtures (0–9 mol.% TiCl 3 ). This process is believed to result in the in situ solid-state introduction of metallic Ti via the reduction of the TiCl 3 by the Na-component of NaAlH 4 (to form NaCl). Properties studied were hydriding and dehydriding rates and reversible gravimetric H-capacity as a function of starting TiCl 3 content. Detailed isothermal kinetic studies were done over a wide temperature range (20–225 °C) and treated by Arrhenius analysis. All kinetics were found to follow the Arrhenius equation and the changes of thermal activation energies and rate constants with TiCl 3 -content were observed that may give valuable insights into the as-yet unknown mechanistic factors that control H 2 desorption and absorption kinetics. Ti increases both dehydriding and hydriding kinetics (and associated practical engineering rates), but at the substantial expense of H-capacity. The finite room temperature decomposition of the catalyzed NaAlH 4 phase was reconfirmed and rates better quantified.

The discovery that hydrogen can be reversible absorbed and desorbed from NaAlH{sub 4} by the addition of catalysts has created an entirely new prospect for lightweight hydrogen storage. NaAlH{sub 4} releases hydrogen through the following set of decomposition reactions: NaAlH{sub 4} {r_arrow} 1/3({alpha}-Na{sub 3}AlH{sub 6}) + 2/3Al + H{sub 2} {r_arrow} NaH + Al + 3/2H{sub 2}. These decomposition reactions as well as the reverse recombination reactions were directly observed using time-resolved in-situ x-ray powder diffraction. These measurements were performed under conditions similar to those found in PEM fuel cell operations (hydrogen absorption: 50--70 C, 10--15 bar Hz, hydrogen resorption: 80--110 C, 5--100 mbar H{sub 2}). Catalyst doping was found to dramatically improve kinetics under these conditions. In this study, the alanate was doped with a catalyst by dry ball-milling NaAlH{sub 4} with 2 mol.% solid TiCl{sub 3}. X-ray diffraction clearly showed that TiCl{sub 3} reacts with NaAlH{sub 4} to form NaCl during the doping process. Partial desorption of NaAlH{sub 4} was even observed to occur during the catalyst doping process.

The hydrogen storage properties of catalyzed NaAlH 4 (and associated Na 3 AlH 6 ) were studied in relation to various practical engineering considerations. Properties measured were cyclic capacity, charging and discharging rates, thermal effects, gaseous impurities, volume changes, low temperature plateau pressures and detailed isothermal desorption kinetics over the temperature range 23–180°C. Two materials were evaluated, one mechanically milled with the liquid alkoxides Ti(OBu n ) 4 and Zr(OPr i ) 4 and one milled with dry TiCl 3 as catalyst precursors. The alkoxide-catalyzed materials had low reversible capacities and released significant levels of hydrocarbon impurities during H 2 discharge. These problems were virtually eliminated with the inorganic TiCl 3 catalyst precursor. The NaAlH 4 and Na 3 AlH 6 decomposition kinetics of TiCl 3 -catalyzed Na-alanate conform to Arrhenius behavior with activation energies of 79.5 and 97 kJ/mol H 2 , respectively. Measured absorption and desorption kinetics were surprisingly good and it is shown that 3–4.5 wt.% H 2 can be stored and recovered in reasonable times at 100–125°C. It may even be ultimately possible to use the NaAlH 4 decomposition reaction to provide 3 wt.% H 2 at room temperature for low-rate applications.

A number of laboratories have now demonstrated that catalyst-assisted NaAlH 4 can reversibly absorb and desorb hydrogen in the solid state at moderate temperatures. An understanding of the mechanisms by which bulk decomposition and reformation of the compound can occur in the presence of a surface catalyst is important to improving the kinetic and thermodynamic properties of alanates for use in hydrogen storage applications. Using scanning electron microscopy (SEM) and energy dispersive spectroscopy (EDS), we have examined the microstructure and elemental composition of Na alanate samples, doped using a liquid Ti/Zr catalyst precursor, for a number of conditions. First, microscopy and compositional analyses were performed at different stages of the decomposition process within the first desorption cycle. Second, the material was characterized after multiple absorption/desorption cycles (five cycles). Finally, the effects of the catalyst doping procedure on particle size, surface morphology and surface composition were examined. Significant changes in particle morphology and in elemental distribution were found to be induced by the desorption and cycling processes. Importantly, our measurements indicate that the initial dehydriding reactions were accompanied by significant enhancement of Al concentration toward the surface of particles and that elemental segregation occurred with repeated absorption/desorption cycles.

The dehydriding of sodium aluminum hydride, NaAlH4, is kinetically enhanced and rendered reversible in the solid state upon doping with selected titanium compounds. Following the initial reports of this catalytic effect, further kinetic improvement and stabilization of the cyclable hydrogen capacity have been achieved upon variation in the method of the introduction of titanium and particle-size reduction. Rapid evolution of 4.0-wt % hydrogen at 100 °C has been consistently achieved for several dehydriding/rehydriding cycles. An improved, 4.8-wt % cyclable capacity has been observed in the material doped with a combination of Ti and Zr alkoxide complexes. Doping the hydride with Ti(OBun)4 and Fe(OEt)2 also produces a synergistic effect, resulting in materials that can be rehydrided to 4 wt % at 104 °C and 87 atm of hydrogen within 17 h. The improved kinetics allowed us to carry out constant-temperature, equilibrium-pressure studies of NaAlH4 that extended to temperatures well below the melting point of the hydride. The 37-kJ/mol value determined for enthalpy of the dehydriding of NaAlH4(s) to Na3AlH6 and Al and the hydrogen plateau pressure of 7 atm at 80 °C are in line with the predictions of earlier studies. The nature of the active catalyst and the mechanism of catalytic action are unknown. The catalytically enhanced hydrides appear to be strong candidates for development as hydrogen carriers for onboard proton exchange membran (PEM) fuel cells. However, further research and development in the areas of rehydriding catalysts, large-scale, long-term cycling, safety and adjustment of the plateau hydrogen pressure associated with dehydriding of AlH6- are required before these materials can be utilized in commercial onboard hydrogen-storage systems.