Session T30: Focus Session: Hydrogen Storage III

Role of point defects and additives in kinetics of hydrogen storage materials

First-principles computational studies of hydrogen interactions with storage materials can provide direct insight into the processes of H uptake and release, and may help in developing guidelines for designing storage media with improved storage capacity and kinetics. One important conclusion is that the defects involved in kinetics of semiconducting or insulating H-storage materials are charged, and hence their formation energy is Fermi-level dependent and can be affected by the presence of impurities that change the Fermi level [1,2]. This provides an explanation for the role played by transition-metal impurities in the kinetics of NaAlH$_{4}$ and related materials. Desorption of H and decomposition of NaAlH$_{4}$ requires not only mass transport of H but also of Al and/or Na. This process is mediated by native defects. We have investigated the structure, stability, and migration enthalpy of native defects based on density functional theory. The results allow us to estimate diffusion activation energies for the defects that may be involved in mass transport. Most of the relevant defects exist in charge states other than neutral, and consideration of these charge states is essential for a proper description of kinetics. We propose specific new mechanisms to explain the observed activation energies and their dependence on the presence of impurities. We have also expanded our studies to materials other than NaAlH$_{4}$. In the case of LiBH$_{4 }$and Li$_{4}$BN$_{3}$H$_{10 }$we have found that the calculations have predictive power in terms of identifying which impurities will actually enhance kinetics. Other complex hydrides that we are currently investigating include Li$_{2}$NH and LiNH$_{2}$. \\[4pt] [1] A. Peles and C. G. Van de Walle, Phys. Rev. B 76, 214101 (2007). \\[0pt] [2] C. G. Van de Walle, A. Peles, A. Janotti, and G. B. Wilson-Short, Physica B 404, 793 (2009).   

Super-Activated Microporous Carbon for Cryogenic Hydrogen Storage

A new high surface area microporous carbon material was synthesized by processing the carbon precursor together with the chemical activation reagent prior to carbonization and activation and removal of the non-carbon residues. The specific surface area and pore size distribution of the resulting porous carbon material mainly depended on the weight ratio of the activation reagent to carbon precursor, the processing time, and the carbonization and activation time and temperature. Optimized results were obtained for an activation reagent to carbon precursor weight ratio of 4 and processing time of 60 minutes followed by carbonizing and activating at 700$^{\circ}$C for 4 hours. The resulting super-activated microporous carbon material exhibited an extremely high surface area of 3390~m$^{2}$/g and an excess hydrogen adsorption of $\sim $6.0 wt{\%} at 77 K and 30 bars.   

Measurements of Hydrogen Spillover in Platinum-Doped Superactivated Carbon

Hydrogen uptake was measured of platinum doped superactivated carbon at 296 K where hydrogen spillover was expected to occur. High pressure adsorption measurements using a Sieverts apparatus did not show an increase in gravimetric storage capacity over the unmodified superactivated carbon. Measurements of small samples (0.2 g) over long equilibration times, consistent with reported procedure, showed significant scatter and were not well above instrument background. In larger samples (3.2 g) the hydrogen uptake was significantly above background but did not show enhancement due to spillover; total uptake scaled with the available surface area of the superactivated carbon. Any hydrogen spillover sorption was thus below the detection limit of standard volumetric gas adsorption measurements. Due to the additional mass of the catalyst nanoparticles and decreased surface area in the platinum doped system, the net effect of spillover sorption is detrimental for gravimetric density of hydrogen.   

Chemi- and physisorption of hydrogen on graphitic substrates

We evaluate the possibilities of hydrogen storage on graphitic substrates by chemical and physical sorption using multiscale modeling. Detailed thermodynamic analysis based on ab initio calculations of chemisorptions via catalytic spillover shows that the catalyst saturation and improved C-H binding energies are the key to enhanced storage [1]. The estimation of amount of physisorbed hydrogen under the ambient condition in 3D-foams using grand canonical Monte Carlo simulations demands judicious choice of interaction potential and incorporation of quantum corrections due to large thermal de Brogile wavelength. We will show that the storage capacities (chemi- or physisorption) in graphitic materials can meet the DOE 2015 targets. \\[4pt] [1] A. K. Singh, M. A. Ribas, and Boris I. Yakobson, ACS Nano, 3, 1657 (2009).   

Hydrogen storage of calcium atoms adsorbed on graphene: First-principles plane wave calculations

Based on the first-principles plane wave calculations, we showed that Ca adsorbed on graphene can serve as a high-capacity hydrogen storage medium, which can be recycled by operations at room temperature. Ca is chemisorbed by donating part of its $4s$-charge to the empty $\pi^*$-band of graphene. At the end adsorbed Ca atom becomes positively charged and the semi-metallic graphene change into a metallic state. While each of adsorbed Ca atoms forming the (4$\times$4) pattern on the graphene can absorb up to five H$_2$ molecules, hydrogen storage capacity can be increased to 8.4 wt \% by adsorbing Ca to both sides of graphene and by increasing the coverage to form the (2$\times$2) pattern. Clustering of Ca atoms is hindered by the repulsive Coulomb interaction between charged Ca atoms.   

Hydroxyl group-Calcium complex for hydrogen storage media

Using first - principles calculations based on the density functional theory, we study the hydroxyl group-Ca complex for hydrogen storage application. The Ca atom is bound to the hydroxyl group with a binding energy comparable to the cohesive energy of bulk Ca and the Ca atom can bind up to 7H$_2$ molecules in molecular form, which can give a very high weight percent storage. However, the average binding energy is about 0.1 eV per hydrogen molecule, which is somewhat smaller than the requirement for the room temperature application. This result shows a possibility of organic materials functionalized with hydroxyl group for hydrogen storage media at near ambient conditions. We also show that, in addition to the metal-H$_2$ hybridization, the polarization of H$_2$ molecules induced by the ionized Ca atom plays an important role in the binding of H$_2$ molecules to the Ca atom.   

Record Hydrogen Storage Capacities in Advanced Carbon Storage Materials

Carbons can be engineered to achieve exceptional storage capacities: the ALL-CRAFT (\underline{www.all-craft.missouri.edu}) nanoporous carbon achieves gravimetric excess adsorption of 0.073 kg H$_2$/kg C, gravimetric storage capacity of 0.106 kg H$_2$/kg C, and volumetric storage capacity of 0.040 kg H$_2$/l C (80 K, 100 bar). The nanopores generate high storage capacity by having: high surface area (2,600 m$^2$/g); high H$_2$-wall interaction; and multi-layer H$_2$ adsorption (cryogenic). We we show how the experimental characteristics of the ALL-CRAFT carbon correlate to the observed H$_2$ storage, with help from theoretical considerations and GCMC simulations. The ALL-CRAFT carbon is composed of a large variety of pore sizes which generates substantial heterogeneity. We explain most features observed by considering superpositions of low- and high-binding energy sites (9 kJ and 5 kJ/mol), corresponding to wide and narrow ($<$ 1 nm) pores. We further explain: exceptional low-temperature storage (in excess of the usual Chahine's rule); and absence of an excess adsorption peak (for $0 < P < 100$ bar).   

Ultrananopores in Carbons by Boron-neutron Capture and Their Effect on Hydrogen Storage

The Alliance for Collaborative Research in Alternative Fuel Technology (ALL-CRAFT) has been optimizing high surface area activated carbon nanospaces for high capacity hydrogen storage. Boron doped samples have been prepared by vapor deposition of decaborane. Neutron irradiation of Boron doped activated carbon was done at the University of Missouri Research Reactor (MURR). Ultrananopores created by alpha particle fission tracks from Boron-neutron capture alter the surface and the adsorption properties of activated Carbons. A detailed theoretical model of the creation and the structure of defects on graphene sheets was developed. BET surface areas, porosity, and pores size distributions of modified activated carbons were measured using sub-critical nitrogen isotherms. Hydrogen adsorption isotherms of irradiated samples were indicative of record fraction of high binding energies and record fraction of sub-nm pores compared to their unirradiated parent samples.   

Adsorption of hydrogen in boron-substituted nanoporous carbons

Nanoporous carbons are promising for hydrogen storage. However, the heat of physisorption is low (4.5-8 kJ/mol), which limits the total amount of hydrogen adsorbed at room temperature to \~{}2 wt.\% at 100 bar. To enhance sorption the surface must be modified by substitution or doping/intercalation of some atoms in the carbon skeleton by other elements. Here we present coupled {\em ab initio} calculations and Monte Carlo simulations showing that partial substitution of carbon atoms in nanoporous matrix with boron increases significantly the adsorption energy (up to 10-13.5 kJ/mol) and storage capacity (\~{}5 wt.\% at 298 K, 100 bar), even for relatively low substitution ratios (5-10\%). Although substitution causes both energetic and structural heterogeneity of the adsorbent, at room temperature the delivery of the stored gas during adsorption-desorption cycle is almost complete (\~{}97 \%). We analyze whether the location of substituted atoms (within the graphene plane or between two adjacent planes) and randomness of its distribution modify either the adsorption mechanism or/and storage parameters. In particular the heterogeneity of energy landscape is discussed in a context of optimization of system delivery.   

Isosteric heats of adsorption for activated carbons made from corn cob

Activated carbons made from corn cob show promise as materials for high-capacity hydrogen storage. As part of our characterization of these materials, we are interested in learning how different production methods affect the adsorption energies. In this talk, we will present experimentally measured isosteric heats of adsorption for various activated carbons calculated using the Clausius-Clayperon equation and hydrogen isotherms at temperatures of 80 and 90K and pressures up to 100 bar measured on a volumetric instrument. We discuss differences observed between isosteric heats determined from Gibbs excess adsorption vs. absolute adsorption curves.   

Quantization of Adsorbed Hydrogen for Inhomogeneous Materials Characterization using Inelastic Neutron Scattering

The motion of adsorbed hydrogen is heavily quantized, especially at cryogenic temperatures. Previous theoretical work has taken a mostly classical approach. Using a slit-shaped pore model, we show that quantizing the transverse degree of freedom has a significant effect on adsorption isotherms. The model is also used to create inelastic neutron scattering spectra, which are highly variable for different size slits. When compared with previous experimental work in the literature, the results suggest that while the slit-shaped pore model works well for some materials, it is inadequate for activated carbon. Planned experimental work to probe the structure of inhomogeneous materials is discussed.   

Nanopore structure from USAXS/SAXS in advanced carbon materials for hydrogen storage

Despite their mass-production and industrial use, there is still no generally accepted structural model of non-graphitizing activated carbons. We will show how USAXS/SAXS can be used to estimate the average shape and size of nanopores in amorphous carbon used for methane and hydrogen storage. Simulated scattering curves constructed from explicit experimental N$_{2}$ isotherm pore size distributions reveal that nanoporous activated carbons scatter as correlated networks of pores with scattered intensities that depend largely on sample porosity. Graphical methods will be used to show how porosity can be calculated from SAXS data, using minimal model-dependent assumptions. The results are shown to be in excellent agreement with porosity values measured via N$_{2}$ sorption isotherms.   

Fractal Structure in Hydrogen and Methane Storage Materials

For nearly 30 years, fractal characteristics have been used to describe physical properties of disordered materials. Small Angle X-ray Scattering (SAXS) and adsorption isotherms are two experimental techniques that have been used successfully estimate the fractal dimension ( the $D$-dimensional Hausdorff measure) of porous media systems. We present the fractal structure of amorphous nanoporous carbons used for hydrogen and methane storage. Measurements from USAXS$\backslash $SAXS and N$_{2}$ isotherms are compared and contrasted. The implication of how the sample topology of these materials may influence both the availability and number of potential binding sites for hydrogen or methane storage is discussed.