What Can Nanochemistry do for Hydrogen Storage?

Welcome to one of our guest columns, where active researchers can share their views on topics relevant to materials science. Professor Geoffrey Ozin from the University of Toronto shares his thoughts along with those of Professor David Antonelli from the University of Glamorgan, regarding the role that nanochemistry can play in the current hydrogen storage challenge.

I was intrigued by a paper that David Antonelli and coworkers published in JACS earlier this year on a new generation of enhanced performance hydrogen storage materials. They reported a fascinating new class of metal hydrazide porous gels that instead of storing hydrogen through the usual metal-dihydrogen physisorption or metal-hydride bonding mechanisms of trapping hydrogen in solids, instead works through the well known Kubas metal-dihydrogen chemisorption bond.

I spoke at length to David about the innovation, significance, timeliness and potential technological relevance of his breakthrough and because of my perceived view of the importance of his work in the highly competitive field of hydrogen storage I decided to integrate his thoughts and mine on the matter in this NanoChannel perspective with an eye on the role that nanochemistry can play in the hydrogen storage challenge.

The story of hydrogen storage in solid begins with hydrides that hold more hydrogen per unit volume than liquid hydrogen – some close to twice as much. This sounds promising because hydrogen liquefies at 20 K and it is both expensive and inconvenient to cool to these temperatures. But there is a catch and a big one at that!

While the slow release kinetics of these compounds can sometimes be overcome with a little trickery such as ball-milling and additives, the fact that the thermodynamics of the uptake and release are problematic is something that still remains a challenge, perhaps too large to overcome. The reason is that for those hydrides and hydrogen-containing compounds that reversibly absorb and release hydrogen, an enormous amounts of heat is released on uptake and an equally enormous amount of heat is required to liberate the hydrogen for use as fuel. This necessitates complex engineering solutions that cut into the efficiency of the system in ways that make them unusable.

Because of these difficulties many materials researchers have explored nanochemistry ways of absorbing hydrogen, such as on the surface of metal organic frameworks called MOFs and various forms of nanoporous carbon. These nanomaterials can physisorb up to 7 wt% hydrogen at 77 K and 65 bar which is a feasible pressure considering that everything up to 200 bar is considered safe and convenient from the standpoint of building conformable tanks that can be bent around the design of an automobile chassis.

The aesthetic structures of MOFs and their gigantic surface areas, some even approaching the extraordinary gas adsorption limit of 7000 m2/g, seemed to offer the potential of high weight percentages of hydrogen with no kinetic barrier on release and this opportunity was enough to ignite the imagination and efforts of some of the world’s best academic scientists and chemical industry giants. However these materials have the innate problem that hydrogen binds so weakly to the surface that temperatures of 77 K were necessary to keep the hydrogen intact. Moreover, and this is possibly an even bigger blow, the volumetric densities are far too low for practical use. Even when you include the compressed gas in the pores in addition to that adsorbed to the surface the best materials hold about 50 kg/m3, which is 20 kg/m3 less than liquid hydrogen. So you lose whatever advantage that you had over liquid hydrogen.

But the troublesome question still remains, how do we better store the hydrogen?

As it turns out, nanochemistry may provide an answer to this dilemma. The key to storing hydrogen effectively is to tailor the highest density of spatially accessible binding sites of the correct energy into a solid state material. The irksome point is that hydrides have enthalpies that are too high (ca. 70 kJ/mol) and MOFs have enthalpies that are too low (ca. 8-10 kJ/mol). Calculations show that the optimal for room temperature storage without any kinetic barrier or heat management problem is 20-30 kJ/mol. This is exactly where the Kubas interaction mentioned above falls – a molecular hydrogen chemical binding mode discovered in organometallic compounds in the 1980s by Gregory Kubas.

A challenge for nanochemistry is then to find ways to tailor Kubas binding sites at a high density into nanoporous solids with a very low molecular weight.

While nanochemistry has made great strides to control size, shape, porosity and surface functionality of solid state materials, in this application it is necessary to discover ways and means to also control the coordination sphere about a metal center to allow the maximum amount of Kubas binding in an extended solid while maintaining hydrogen diffusion properties. This is quite a difficult task because transition metal solids and compounds almost always have filled coordination spheres with an average of six ligands per metal, which would not allow room for molecular hydrogen to bind. The only exceptions to this are compounds with very sterically bulky ligands that allow such reactive and unstable coordination numbers lower than six to be stable. The key then is to find a way to jump from the trapped low coordination number in these low-coordinate complexes to a similar coordination number in an extended solid full of accessible space for dihydrogen like nanopores.

I was intrigued to discover that Antonelli and coworkers seem to have found a solution to this problem. In using metal alkyl complexes with low coordination numbers and bulky alkyl ligands as precursors they have found a way to preserve these low coordination numbers through a polymerization reaction involving hydrazine to lead to nanoporous metal hydrazide gels with elimination of the alkyl sheaths by reaction with the hydrazine protons the concept of which is illustrated in the figure above (adapted with permission from J. Am. Chem. Soc. 2010, 132, 11792. Copyright 2010 American Chemical Society). Hydrazine was chosen because it is light weight, and just big enough to bridge the transition metals together without causing clustering. It also has four available protons that can react with the alkyls, which stay on the metal just long enough to protect the low coordination sphere during the reaction. The final compounds have formulas of the type MNxHy and molecular weights of around 70-80 g/mol with extended open frameworks comprised of metal sites connected by M-NH-NH-M and M-NH-NH2 -> M bridging groups, creating thereby a random network of nanopores. This means that binding 2 H2 per metal, which is more than possible, would result in weight % numbers over 5 and volumetric densities close to 100 kg/m3.

In discussions with David my thoughts on the matter were first and foremost that the materials are pretty sensitive and can ignite in air. As hydrogen and fire do not sit well together it will be necessary to develop synthetic pathways to less oxygen sensitive metal hydrazide gels. I thought it might also be advantageous to find ways to make nanoscale metal hydrazide gel particles to reduce diffusion lengths for adsorption-desorption cycling to improve the kinetics and energetics of the system while at the same time minimizing sensitivity to air. Efforts to enhance the porosity of the metal hydrazide gels, which is currently around 200 m2/g, might also prove beneficial as easy access of dihydrogen to Kubas-type metal binding sites may enhance the storage performance. This possibly could be achieved by integrating porogens into the synthesis of the metal hydrazide gels.

While the work described is just a first step for Antonelli and coworkers they have already made materials with as high as 3.2 wt% and 41 kg/m3 volumetric density at pressures under 200 bar with no kinetic barrier and what promises at this early stage to be little heat management issues. These materials already hold about four times as much hydrogen as compressed hydrogen per unit volume at any given pressure and temperature and are totally compatible with compressed gas technologies, as pressure is the toggle switch. This is advantageous given the fact that the industry is currently moving towards compressed gas anyway as a short-term solution.

To Top