Sandia National Laboratories has developed a vehicle-scale demonstration hydrogen storage system as part of a Work for Others project funded by General Motors. This Demonstration System was developed based on the properties and characteristics of sodium alanates which are complex metal hydrides. The technology resulting from this program was developed to enable heat and mass management during refueling and hydrogen delivery to an automotive system. During this program the Demonstration System was subjected to repeated hydriding and dehydriding cycles to enable comparison of the vehicle-scale system performance to small-scale sample data. This paper describes the experimental results of life-cycle studies of the Demonstration System. Two of the four hydrogen storage modules of the Demonstration System were used for this study. A well-controlled and repeatable sorption cycle was defined for the repeated cycling, which began after the system had already been cycled forty-one times. After the first nine repeated cycles, a significant hydrogen storage capacity loss was observed. It was suspected that the sodium alanates had been affected either morphologically or by contamination. The mechanisms leading to this initial degradation were investigated and results indicated that water and/or air contamination of the hydrogen supply may have lead to oxidation of the hydride and possibly kinetic deactivation. Subsequent cycles showed continued capacity loss indicating that the mechanism of degradation was gradual and transport or kinetically limited. A materials analysis was then conducted using established methods including treatment with carbon dioxide to react with sodium oxides that may have formed. The module tubes were sectioned to examine chemical composition and morphology as a function of axial position. The results will be discussed.
This paper describes mitigation technologies that are intended to enable the deployment of advanced hydrogen storage technologies for early market and automotive fuel cell applications. Solid State hydrogen storage materials provide an opportunity for a dramatic increase in gravimetric and volumetric energy storage density. Systems and technologies based on the advanced materials have been developed and demonstrated within the laboratory [1,2], and in some cases, integrated with fuel cell systems. The R&D community will continue to develop these technologies for an ever increasing market of fuel cell technologies, including, forklift, light-cart, APU, and automotive systems. Solid state hydrogen storage materials are designed and developed to readily release, and in some cases, react with diatomic hydrogen. This favorable behavior is often accomplished with morphology design (high surface area), catalytic additives (titanium for example), and high purity metals (such as aluminum, Lanthanum, or alkali metals). These favorable hydrogen reaction characteristics often have a related, yet less-desirable effect: sensitivity and reactivity during exposure to ambient contamination and out-of-design environmental conditions. Accident scenarios resulting in this less-favorable reaction behavior must also be managed by the system developer to enable technology deployment and market acceptance. Two important accident scenarios are identified through hazards and risk analysis methods. The first involves a breach in plumbing or tank resulting from a collision. The possible consequence of this scenario is analyzed though experimentally based chemical kinetic and transport modeling of metal hydride beds. An advancing reaction front between the metal hydride and ambient air is observed to proceed throughout the bed. This exothermic reaction front can result in loss of structural integrity of the containing vessel and lead to un-favorable overheating events. The second important accident scenario considered is a pool fire or impinging fire resulting from a collision between a hydrocarbon or hydrogen fueled vehicle. The possible consequence of this scenario is analyzed with experimentally-based numerical simulation of a metal hydride system. During a fire scenario, the hydrogen storage material will rapidly decompose and release hydrogen at high pressure. Accident scenarios initiated by a vehicular collision leading a pipe break or catastrophic failure of the hydride vessel and by external pool fire with flame engulfing the storage vessel are developed using probabilistic modeling. The chronology of events occurring subsequent to each accident initiator is detailed in the probabilistic models. Technology developed to manage these scenarios includes: (1) the use of polymer supports to reduce the extent and rate of reaction with air and water, (2) thermal radiation shielding. The polymer supported materials are demonstrated to provide mitigation of unwanted reaction while not impacting the hydrogen storage performance of the material. To mitigate the consequence of fire engulfment or impingement, thermal radiation shielding is considered to slow the rate of decomposition and delay the potential for loss-of-containment. In this paper we explore the use of these important mitigation technologies for a variety of accident scenarios.
Mitigating and overcoming environmental problems brought about by the current worldwide fossil fuel-based energy infrastructure requires the creation of innovative alternatives. In particular, such alternatives must actively contribute to the reduction of carbon emissions via carbon recycling and a shift to the use of renewable sources of energy. Carbon neutral transformation of biomass to liquid fuels is one of such alternatives, but it is limited by the inherently low energy efficiency of photosynthesis with regard to the net production of biomass. Researchers have thus been looking for alternative, energy-efficient chemical routes inspired in the biological transformation of solar power, CO2 and H2O into useful chemicals; specifically, liquid fuels. Methanol has been the focus of a fair number of publications for its versatility as a fuel, and its use as an intermediate chemical in the synthesis of many compounds. In some of these studies, (e.g. Joo et al., (2004), Mignard and Pritchard (2006), Galindo and Badr (2007)) CO2 and renewable H2 (e.g. electrolytic H2) are considered as the raw materials for the production of methanol and other liquid fuels. Several basic PFD diagrams have been proposed. One of the most promising is the so called CAMERE process (Joo et al., 1999 ). In this process, carbon dioxide and renewable hydrogen are fed to a first reactor and transformed according to: H2 + CO2 <=> H2O + CO Reverse Water Gas Shift (RWGS) After eliminating the produced water the resulting H2/CO2/CO mixture is then feed to a second reactor where it is converted to methanol according to: CO2 + 3.H2 <=> CH3OH + H2O Methanol Synthesis (MS) CO + H2O <=> CO2 + H2 Water Gas Shift (WGS) The approach here is to produce enough CO to eliminate, via WGS, the water produced by MS. This is beneficial since water has been proven to block active sites in the MS catalyst. In this work a different process alternative is presented: One that combines the CO2 recycling of the CAMERE process and the use of solar energy implicit in some of the biomass-based process, but in this case with the potential high energy efficiency of thermo-chemical transformations.
The use of risk information in establishing code and standard requirements enables: (1) An adequate and appropriate level of safety; and (2) Deployment of hydrogen facilities are as safe as gasoline facilities. This effort provides a template for clear and defensible regulations, codes, and standards that can enable international market transformation.
Objectives are to enable development and implementation of codes and standards for H{sub 2} containment components: (1) Evaluate data on mechanical properties of materials in H{sub 2} gas - Technical Reference on Hydrogen Compatibility of Materials; (2) Generate new benchmark data on high-priority materials - Pressure vessel steels, stainless steels; and (3) Establish procedures for reliable materials testing - Sustained-load cracking, fatigue crack propagation. Summary of this presentation are: (1) Completed measurement of cracking thresholds (K{sub TH}) for Ni-Cr-Mo pressure vessel steels in high-pressure H{sub 2} gas - K{sub TH} measurements required in ASME Article KD-10 (2) Crack arrest test methods appear to yield non-conservative results compared to crack initiation test methods - (a) Proposal to insert crack initiation test methods in Article KD-10 will be presented to ASME Project Team on Hydrogen Tanks, and (b) Crack initiation methods require test apparatus designed for dynamic loading of specimens in H{sub 2} gas; and (3) Demonstrated ability to measure fatigue crack growth of pressure vessel steels in high-pressure H{sub 2} gas - (a) Fatigue crack growth data in H{sub 2} required in ASME Article KD-10, and (b) Test apparatus is one of few in U.S. or abroad for measuring fatigue crack growth in >100 MPa H{sub 2} gas.
A permeability model for hydrogen transport in a porous material is successfully applied to both laboratory-scale and vehicle-scale sodium alanate hydrogen storage systems. The use of a Knudsen number dependent relationship for permeability of the material in conjunction with a constant area fraction channeling model is shown to accurately predict hydrogen flow through the reactors. Generally applicable model parameters were obtained by numerically fitting experimental measurements from reactors of different sizes and aspect ratios. The degree of channeling was experimentally determined from the measurements and found to be 2.08% of total cross-sectional area. Use of this constant area channeling model and the Knudsen dependent Young & Todd permeability model allows for accurate prediction of the hydrogen uptake performance of full-scale sodium alanate and similar metal hydride systems.
Advances are reported in several aspects of clathrate hydrate desalination fundamentals necessary to develop an economical means to produce municipal quantities of potable water from seawater or brackish feedstock. These aspects include the following, (1) advances in defining the most promising systems design based on new types of hydrate guest molecules, (2) selection of optimal multi-phase reactors and separation arrangements, and, (3) applicability of an inert heat exchange fluid to moderate hydrate growth, control the morphology of the solid hydrate material formed, and facilitate separation of hydrate solids from concentrated brine. The rate of R141b hydrate formation was determined and found to depend only on the degree of supercooling. The rate of R141b hydrate formation in the presence of a heat exchange fluid depended on the degree of supercooling according to the same rate equation as pure R141b with secondary dependence on salinity. Experiments demonstrated that a perfluorocarbon heat exchange fluid assisted separation of R141b hydrates from brine. Preliminary experiments using the guest species, difluoromethane, showed that hydrate formation rates were substantial at temperatures up to at least 12 C and demonstrated partial separation of water from brine. We present a detailed molecular picture of the structure and dynamics of R141b guest molecules within water cages, obtained from ab initio calculations, molecular dynamics simulations, and Raman spectroscopy. Density functional theory calculations were used to provide an energetic and molecular orbital description of R141b stability in both large and small cages in a structure II hydrate. Additionally, the hydrate of an isomer, 1,2-dichloro-1-fluoroethane, does not form at ambient conditions because of extensive overlap of electron density between guest and host. Classical molecular dynamics simulations and laboratory trials support the results for the isomer hydrate. Molecular dynamics simulations show that R141b hydrate is stable at temperatures up to 265K, while the isomer hydrate is only stable up to 150K. Despite hydrogen bonding between guest and host, R141b molecules rotated freely within the water cage. The Raman spectrum of R141b in both the pure and hydrate phases was also compared with vibrational analysis from both computational methods. In particular, the frequency of the C-Cl stretch mode (585 cm{sup -1}) undergoes a shift to higher frequency in the hydrate phase. Raman spectra also indicate that this peak undergoes splitting and intensity variation as the temperature is decreased from 4 C to -4 C.