Graphene oxide (GO) is a promising membrane system for chemical separation applications due to its 2-D nanofluidics properties and an ability to control interplanar spacing for selectivity. The permeance of water, methanol (MeOH) and isopropyl alcohol (IPA) through 5 µm thick membranes was found to be 0.38 ± 0.15, 0.33 ± 0.16 and 0.42 ± 0.31 LMH/bar (liter/m2·h·bar), respectively. Interestingly, the permeance of a water–alcohol mixture was found to be dramatically lower (~0.01 LMH/bar) than any of its components. Upon removing the solvent mixture, the transmembrane flux of the pure solvent was recovered to near the original permeance. The interlayer space of a dried GO membrane was found to be 8.52 Å, which increased to 12.19 Å. 13.26 Å and 16.20 Å upon addition of water, MeOH and IPA. A decrease in d-space, about 2 Å, was consistently observed when adding alcohol to water wetted GO membrane and an optical color change and reduction in permeance. A newly proposed mechanism of a partial reduction of GO through a catalytic reaction with the water–alcohol mixture is consistent with experimental observations.

A multiscale thermomechanical model for a simplified Type-3 cryogenic, compressed-hydrogen (H2) storage vessel is described in this paper. The model accounts for the temperature-dependent elastic-plastic behavior of the vessel's carbon/epoxy composite overwrap and aluminum alloy liner. The homogenized thermo-elastic-plastic behavior for the individual laminae of the vessel layup is obtained using an incremental Eshelby-Mori-Tanaka approach associated with a micromechanical failure criterion to predict laminar failure while a standard elastic-plastic constitutive model is used to describe the behavior of the typical aluminum alloy assumed for the liner. The vessel's response to external loadings is modeled using a finite element method. Four loading scenarios, representing four thermomechanical cycles applied to the vessel, are analyzed to evaluate constituent and laminar stresses as well as the associated failure criterion during the cycle according to these scenarios. The model can provide helpful guidance to mitigate thermal stresses by selecting a suitable loading scenario, optimizing the layup, and tailoring the thermomechanical properties of the resin matrix.

A polyethersulfone (PES)-supported graphene oxide (GO) membrane has been developed by a simple casting approach. This stable membrane is applied for ethanol/water separation at different temperatures. The 5.0 µm thick GO film coated on PES support membrane showed a long-term stability over a testing period of one month and excellent water/ethanol selectivity at elevated temperatures. The water/ethanol selectivity is dependent on ethanol weight percentage in water/ethanol feed mixtures and on operating temperature. The water/ethanol selectivity was enhanced with an increase of ethanol weight percentage in water/ethanol mixtures, from below 100 at RT to close to 874 at a 90 °C for 90% ethanol/10% water mixture. Molecular dynamics simulation of water-ethanol mixtures in graphene bilayers, that are considered to play a key role in transport, revealed that molecular transport is negligible for layer spacing below 1 nm. The differences in the diffusion of ethanol and water in the bilayer are not consistent with the large selectivity value experimentally observed. The entry of water and ethanol into the interlayer space may be the crucial step controlling the selectivity.

Fuel cell vehicles are entering the automotive market with significant potential benefits to reduce harmful greenhouse emissions, facilitate energy security, and increase vehicle efficiency while providing customer expected driving range and fill times when compared to conventional vehicles. One of the challenges for successful commercialization of fuel cell vehicles is transitioning the on-board fuel system from liquid gasoline to compressed hydrogen gas. Storing high pressurized hydrogen requires a specialized structural pressure vessel, significantly different in function, size, and construction from a gasoline container. In comparison to a gasoline tank at near ambient pressures, OEMs have aligned to a nominal working pressure of 700 bar for hydrogen tanks in order to achieve the customer expected driving range of 300 miles. Beyond the need to contain pressure, the hydrogen tanks also differ from gasoline fuel tanks because of the additional vehicle space needed due to the lower hydrogen energy volumetric density even with the highly efficient fuel cell (four times the external volume of a gasoline tank including the fuel cell efficiency benefit). The main difference and challenge of hydrogen tanks is the construction and design that depends on a high utilization of carbon fiber in order to reduce the weight of the pressure vessel although substantially increasing the cost. In 2012, the U.S. Department of Energy (DOE), Office of Fuel Cell Technologies recognized these challenges and initiated a project to research enhance materials and design parameters to reduce the cost of hydrogen storage tanks. The project was led by Pacific Northwest National Lab (PNNL) and involved several other organizations in the value chain of hydrogen tank development: AOC, Ford Motor Company, Hexagon Lincoln, and Toray CFA. The project took a holistic approach to improving performance by investigating: (1) composite matrix resin alternatives including adding nano-reinforcing particles and fiber-matrix sizing for improved adhesion, (2) carbon fiber alternatives, (3) tank design alternatives using hybrid fiber layups, and (4) opportunities with cold gas operating conditions to maintain the hydrogen density while reducing the tank composite utilization. In each of these areas, the project successfully identified the potential benefits: (1) demonstrated new resin with 50% cost reduction at equivalent or better performance than traditional epoxy, (2) identified 4% to 12% improvement with fiber alternatives, (3) developed validated tank design models with improved failure prediction capability, and (4) confirmed value and system level viability of cold gas storage with a combined 22% cost reduction opportunity. This paper examines these modifications and considers the outlook for on-board 700 bar compressed hydrogen tank systems to achieve the commercialization goals for fuel cell vehicles.

Membranes made of stacked layers of graphene oxide (GO) hold the tantalizing promise of revolutionizing desalination and water filtration if selective transport of molecules can be controlled. We present the findings of an integrated study that combines experiment and molecular dynamics simulation of water intercalated between GO layers. We simulated a range of hydration levels from 1 wt.% to 23.3 wt.% water. The interlayer spacing increased upon hydration from 0.8 nm to 1.1 nm. We also synthesized GO membranes that showed an increase in layer spacing from about 0.7 nm to 0.8 nm and an increase in mass of about 15% on hydration. Water diffusion through GO layers is an order of magnitude slower than that in bulk water, because of strong hydrogen bonded interactions. Most of the water molecules are bound to OH groups even at the highest hydration level. We observed large water clusters that could span graphitic regions, oxidized regions and holes that have been experimentally observed in GO. Slow interlayer diffusion can be consistent with experimentally observed water transport in GO if holes lead to a shorter path length than previously assumed and sorption serves as a key rate-limiting step.

Hierarchically stacked graphene oxide (GO) membranes have been observed to undergo physical and chemical changes under ambient conditions. In this study, the slow degradation of GO membranes is mimicked and accelerated by UV light (254 nm) exposure. Spectroscopic and X-ray diffraction analyses confirm that the observed changes, including macroscopic as well as atomic-scale alterations, cannot be attributed to loss of intercalated water molecules between the GO sheets, but UV light exposure triggers a partial reduction of the GO membrane to a more graphene-like material.

We design, built, and demonstrated a photocathode based pulsed electron system to produce optically stimulated luminescence (OSL) for next generation data storage. We successfully demonstrated 10ps pulses of up to 106 electrons and used those pulses to form OSL patterns in LiF without causing mechanical damage. By adjusting the accelerating voltage, we were able to vary the ratio of red to green optical emission, demonstrating a potential path for increasing storage density using orthogonal encoding. We also modeled a new approach to reading data that exploits the unique nature of OSL based data storage where the writing and reading processes have different size scales. Based on the technical demonstrations, we developed a roadmap to increase the storage density beyond 50TB using a standard 5.25” optical storage disc.

Three water adsorption-desorption mechanisms are common in inorganic materials: chemisorption, which can lead to the modification of the first coordination sphere; simple adsorption, which is reversible; and condensation, which is irreversible. Regardless of the sorption mechanism, all known materials exhibit an isotherm in which the quantity of water adsorbed increases with an increase in relative humidity. Here, we show that carbon-based rods can adsorb water at low humidity and spontaneously expel about half of the adsorbed water when the relative humidity exceeds a 50-80% threshold. The water expulsion is reversible, and is attributed to the interfacial forces between the confined rod surfaces. At wide rod spacings, a monolayer of water can form on the surface of the carbon-based rods, which subsequently leads to condensation in the confined space between adjacent rods. As the relative humidity increases, adjacent rods (confining surfaces) in the bundles are drawn closer together via capillary forces. At high relative humidity, and once the size of the confining surfaces has decreased to a critical length, a surface-induced evaporation phenomenon known as solvent cavitation occurs and water that had condensed inside the confined area is released as a vapour.

Membranes for selective water vapor separation were assembled from graphene oxide suspension using techniques compatible with high volume industrial production. The large-diameter graphene oxide flake suspensions were synthesized from graphite materials via relatively efficient chemical oxidation steps with attention paid to maintaining flake size and achieving high graphene oxide concentrations. Graphene oxide membranes produced using scalable casting methods exhibited water vapor flux and water/nitrogen selectivity performance meeting or exceeding that of membranes produced using vacuum-assisted laboratory techniques. (PNNL-SA-117497)

Hierarchically stacked 2D graphene oxide (GO) membranes are a fascinating and promising new class of materials with the potential for radically improved water vapor/gas separation with excellent selectivity and high permeability. This paper details dehumidification results from flowing gas mixtures through free-standing GO membrane samples prepared by a casting method. The first demonstrated use of freestanding GO membranes for water vapor separation reveals outstanding water vapor permeability and H2O/N-2 selectivity. Free-standing GO membranes exhibit extremely high water vapor permeability of 1.82 x 10(5) Barrer and a water vapor permeance of 1.01 x 10(-5) mol/m(2)sPa, while the nitrogen permeability was below the system's detection limit, yielding a selectivity >10(4) in 80% relative humidity (RH) air at 30.8 degrees C. The results show great potential for a range of energy conversion and environmental applications. (C) 2016 Elsevier Ltd. All rights reserved.

In future nuclear fuel reprocessing plants and possibly for nuclear power plants, the cleanup of tritiated water will be needed for hundreds of thousands of gallons of water with low activities of tritium. This cleanup concept utilizes graphene oxide laminar membranes (GOx) for the separation of low-concentration (10-3-10 µCi/g) tritiated water to create water that can be released to the environment and a much smaller waste stream with higher tritium concentrations. Graphene oxide membranes consist of hierarchically stacked, overlapping molecular layers and represent a new class of materials. A permeation rate test was performed with a 2-µm-thick cast Asbury membrane using mixed gas permeability testing with zero air (highly purified atmosphere) and with air humidified with either H2O or D2O to a nominal 50% relative humidity. The membrane permeability for both H2O and D2O was high with N2 and O2 at the system measurement limit. The membrane water permeation rate was compared to a Nafion® membrane and the GOx permeation was approximately twice as high at room temperature. The H2O vapor permeation rate was 5.9 × 102 cc/m2/min (1.2 × 10-6 g/min-cm2), which is typical for graphene oxide membranes. To demonstrate the feasibility of such isotopic water separation through GOX laminar membranes, an experimental setup was constructed to use pressure-driven separation by heating the isotopic water mixture at one side of the membrane to create steam while cooling the other side. Several membranes were tested and were prepared using different starting materials and by different pretreatment methods. The average separation result was 0.8 for deuterium and 0.6 for tritium. Higher or lower temperatures may also improve separation efficiency but neither has been tested yet. A rough estimate of cost compared to current technology was also included as an indication of potential viability of the process. The relative process costs were based on the rough size of facility to accommodate the large surface area of the membranes and the energy needed to evaporate the water and pass through the membranes as compared to the currently used combined electrolysis and catalytic exchange process.

• By 2020, develop and verify onboard automotive hydrogen storage systems achieving 1.8 kWh/kg system (5.5 wt% H2) and 1.3 kWh/L system (0.040 kg H2/L) at a cost of $10/kWh ($333/kg H2 stored). The progress toward targets is shown in Table 1. The gravimetric and volumetric capacities decrease slightly, due to the required insulation weight and volume. The storage system cost decreases 22% from the 2012 cost, due to the reduced carbon fiber composite in the 500 bar tank vs. the 700 bar tank. IV.D.1 Enhanced Materials and Design Parameters for Reducing the Cost of Hydrogen Storage Tanks

Solid oxide fuel cell (SOFC) power systems can be designed to produce electricity from fossil fuels at extremely high net efficiencies, approaching 70%. However, in order to penetrate commercial markets to an extent that significantly impacts world fuel consumption, their cost will need to be competitive with alternative generating systems, such as gas turbines. This report discusses a cost model developed at PNNL to estimate the manufacturing cost of SOFC power systems sized for ground-based distributed generation. The power system design was developed at PNNL in a study on the feasibility of using SOFC power systems on more electric aircraft to replace the main engine-mounted electrical generators [Whyatt and Chick, 2012]. We chose to study that design because the projected efficiency was high (70%) and the generating capacity was suitable for ground-based distributed generation (270 kW).

Gold, palladium, platinum or nickel ohmic contacts on Mg doped p -type Al x Ga 1-x N with x = 0.4 and x = 0.45 have been examined. The Au contact provided the lowest contact resistivity with p c = 1.8 (± 1.1) x 10 −3 Ωcm 2 , but only following annealing at 850°C. For the Pd, Au, and Pt contacts annealed at greater than 700°C, a rapid degradation in the current-voltage curves was observed upon testing. The degradation was induced by exposure to sub-bandgap light and was reversed with a mild anneal at 500°C. Possible mechanisms for the degradation are discussed.

AlGaN/GaN heterojunction field effect transistors (HFET) on sapphire substrates have demonstrated ability as power devices operating with high current densities and high breakdown voltages. Additionally, AlGaN/GaN HFET devices have a very low on-state resistance. This makes these devices ideal for automotive applications such as switching relays, DC-DC converters, and power inverters. By 2006, switching devices using GaN-based FETs are anticipated to be employed in luxury automobiles and transitioned to the mass market by 2009. In this presentation, data from AlGaN/GaN HFET’s grown in an Emcore D180 MOCVD system will be presented. Typical production-scale material results (on 2” sapphire substrates) for these wafers were: μ ∼ 1000 cm2/Vs, Ns = 1.0×1013 cm−2, and Rs ∼ 450 Ω/square with <3% variation across the wafer. These wafers were then processed into devices using Pt/Au gate contacts with 2 μm gate length, 200 μm gate width, and a source to drain spacing of 13 μm. A total of 1000 FETs were combined in parallel for an effective gate width of 20 cm for high current operation (10A). These devices have a lower on-state resistance (<0.01 Ω-cm2) and higher Schottky breakdown voltages (400 V) than the theoretical limit of Si MOSFET devices. These devices demonstrate suitability for insertion in automotive electrical harnesses.