Graphene oxide/metal nanocrystal multilaminates as the atomic limit for safe and selective hydrogen storage

Interest in hydrogen fuel is growing for automotive applications; however, safe, dense, solid-state hydrogen storage remains a formidable scientific challenge. Metal hydrides offer ample storage capacity and do not require cryogens or exceedingly high pressures for operation. However, hydrides have largely been abandoned because of oxidative instability and sluggish kinetics. We report a new, environmentally stable hydrogen storage material constructed of Mg nanocrystals encapsulated by atomically thin and gas-selective reduced graphene oxide (rGO) sheets. This material, protected from oxygen and moisture by the rGO layers, exhibits exceptionally dense hydrogen storage (6.5 wt% and 0.105 kg H2 per litre in the total composite). As rGO is atomically thin, this approach minimizes inactive mass in the composite, while also providing a kinetic enhancement to hydrogen sorption performance. These multilaminates of rGO-Mg are able to deliver exceptionally dense hydrogen storage and provide a material platform for harnessing the attributes of sensitive nanomaterials in demanding environments.


Supplementary Discussion
Corrosion Test for rGO-Mg: The rGO-Mg composite was placed in the environmental chamber with 60 ºC-the upper limit in the range of ambient operating temperature of FCEV by DOE-and 90% of relative humidity for 3 days to verify its safety under environmental exposure, followed by XRD measurement (Supplementary Fig. 4). Remarkably, the Mg crystalline structure was well-maintained without oxidation due to the rGO encapsulation layers.  (1)),

Investigation of Oxidation
where x is the fraction of Mg or MgH 2 hydrogenated or dehydrogenated, k is the reaction rate constant, t is time, and n is the reaction exponent. For the absorption measurement, the best linear behavior was acquired with n=0.79-0.89, though n=0.98-1.00 was obtained for the initial 60 % of absorption fraction ( Supplementary Fig. 7 wt% of H 2 desorption for 300 ºC; hence, the data at 300 ºC was separated into two regions, before and after 1 wt% desorption (labeled as region i and ii, respectively, in Supplementary Fig.  7(b) inset), for an accurate analysis. The best linear behavior was obtained with n=0.95 and 0.89 for 325 ºC and 350 ºC, respectively, while n=3.5 and n=0.99 for 300 ºC, before (i) and after (ii) 1 wt% desorption, respectively, indicating the change of mechanism. Using the two different regimes, different activation energies were obtained: 165.1 kJ/mol (R 2 = 0.912) and 92.9 kJ/mol (R 2 = 0.983) respectively. The curve fitting had a higher R 2 value when the data region with n=0.99 was used. It can be inferred that, at high temperatures, hydrogen is desorbed via rapid nucleation followed by one-dimensional growth, whereas at 300ºC, slow nucleation occurs until 1 wt% of hydrogen is desorbed, followed by one-dimensional growth.

Comparison of Hydrogen Sorption between rGO-Mg and Mg-PMMA:
The hydrogen absorption/desorption properties of the nanolaminate were compared with Mg-PMMA 1 which has similarly sized Mg nanocrystals encapsulated by poly(methyl methacrylate) (PMMA) ( Supplementary Fig. 9). Enhancements of both hydrogen capacity and sorption kinetics were observed for the rGO-Mg multilaminates; clearly, the presence of the rGO-layers has a beneficial effect on sorption and desorption kinetics.

The Role of rGO in Hydrogen Sorption of the Composite:
The amount of GO in the composite was varied in order to examine the effect of mass fraction of rGO on sorption behavior ( Supplementary Fig. 10). Interestingly, relative to the reported abundance of rGO in the manuscript, both additional and less GO in the synthesis resulted in reduced hydrogen capacity and poorer kinetics. Based upon these results, we observe that the catalytic effect of rGO on sorption was diminished when less GO was used, while a larger amount of GO could hinder hydrogen diffusion into and out of the Mg nanocrystals by increasing the diffusion path length.
Consequently, there exists an optimum weight percent range of GO for optimized performance of the nanolaminates, where rGO crucially prevents Mg nanocrystals from oxidization, while also enhancing the kinetics and maximizing hydrogen capacity.

Raman Analysis for rGO Sheet/Sheet Coupling:
To study the evolution of the Raman spectra as sheet coupling is increased, Mg crystals with a large amount of rGO (m-rGO-Mg)-16 times more GO (relative to the reported synthesis) were prepared. With vastly more rGO in this m-rGO-Mg sample, rGO/rGO (e.g. sheet/sheet) interactions are prominent and Mg nanocrystals are embedded in multiple rGO layers as shown in Supplementary Fig. 12c, while they are encapsulated by a single or few layers of rGO in the structure of rGO-Mg used as shown in Supplementary Fig. 12d. The Raman spectra of both composites exhibited distinct differences with respect to that of bare graphite, as expected. As shown in Supplementary Fig. 12e, for bare graphite, the 2D peak appears at ~2710 cm -1 . The 2D peak region of both composites present a new band exhibiting the same characteristics as reported in few layer wrinkled graphene, and the 2D peak itself resembles few layer of graphene sheet which prepared via a scotch-tape method.
As anticipated based upon theory ( Supplementary Fig. 12), the m-rGO-Mg materials have more sheet-sheet interactions which soften and broaden the vibrational states, which is evident by comparing the m-rGO-Mg Raman spectra to that of rGO-Mg, which shows a narrower and more blue-shifted 2D peak. This evolution of vibrational spectra matches theoretical predictions in the graphene literature, and further confirms that peak position and breadth can be used to characterize the extent of sheet/sheet coupling in these composites.