Energy Systems Division

A fast-start capability is a key requirement for on-board fuel processors for automotive fuel cell systems operating on gasoline fuel. This paper reports on the design and fabrication of a suitable fuel processor having this capability and discusses estimates of the start-up fuel consumption for the laboratory unit. Also discussed are the start-up strategy and the results of a start-up simulation, which showed that the fuel processor can deliver 90% of the rated hydrogen capacity in 60 s, producing a product gas that contains >30% hydrogen and <50 ppm carbon monoxide. The start-up fuel consumption was estimated on the basis of the thermal mass of the fabricated components; the 10-kW e laboratory unit was estimated to require >2.9 MJ of fuel energy. (S. Ahmed).

Seed particles are added to a high temperature reacting flow to achieve the desired electric conductivity for magnetohydrodynamic (MHD) power generation. The evaporation rate of the seed particles and the uniformity of the seed vapor distribution in the reacting flow are major factors affecting the performance of the MHD system. An integral reacting flow (1RF) computer code was used to calculate the multi-phase flow characteristics in an MHD combustor and predict the optimum operating conditions of the combustor. The IRF code is a general hydrodynamics computer code for multi-phase, two-dimensional, steady state, turbulent, and reacting flows, based on mass, momentum, and energy conservation laws. A unique integral reaction submodel in the IRF code enhances numerical convergence while preserving the major physical effects of the complex combustion processes. The combustor is a rectangular channel in which a mixture of fuel gas and seed particles enters from one end and exits at the other, while the oxidizer is injected from the sides. The major findings of a parametric study included: (1) the seed particle evaporation rate varies for different oxidizer injection angles; (2) the optimal evaporation rate was found at an injection angle around 130°; (3) the seed particle size had a major effect on particle dispersion and vapor mixing; (4) seed particles with a diameter of 34 #m or larger were only partially vaporized in the combustor, and the seed vapor distribution at the combustor exit was highly non-uniform; and (5) significant particle deposition was found on the side walls upstream of the oxidizer injection slots. Theoretical predictions were in general agreement with experimental trends.

The main focus of this paper is the modeling, simulation, and analysis of the calcium bromide hydrolysis reactor stage in the calcium-bromine thermochemical water-splitting cycle for nuclear hydrogen production. One reactor concept is to use a spray of calcium bromide into steam, in which the heat of fusion supplies the heat of reaction. Droplet models were built up in a series of steps incorporating various physical phenomena, including droplet flow, heat transfer, phase change, and reaction, separately. Given the large heat reservoir contained in a pool of molten calcium bromide that allows bubbles to rise easily, using a bubble column reactor for the hydrolysis appears to be a feasible and promising alternative to the spray reactor concept. The two limiting cases of bubble geometry, spherical and spherical-cap, are considered in the modeling. Results for both droplet and bubble modeling with COMSOL MULTIPHYSICSÔ are presented, with recommendations for the path forward. ª

This study is focused on the simulation of open channel turbulent flow over flooded laboratory scale bridge decks and formation of scour holes under various flooding conditions. Solutions for turbulent flow field are based on Reynolds Averaged Navier-Stokes (RANS) equations and turbulence closure models using the STAR-CD commercial computational fluid dynamics (CFD) software. An iterative computational methodology is developed for predicting equilibrium scour profiles using the single-phase flow model with a moving boundary formulation. The methodology relies on an empirical correlation for critical bed shear stress that is used to characterize the condition for onset of sediment motion and an effective bed roughness that is a function of sediment particle size. The computational model and iterative methodology were stable and converged to an equilibrium scour hole shape and size that compares reasonably well with experiment using a constant critical shear stress value.

The calcium-bromine cycle being investigated is a novel continuous hybrid cycle for hydrogen production employing both heat and electricity. Calcium bromide (CaBr 2) hydrolysis generates hydrogen bromide (HBr) which is electrolyzed to produce hydrogen. The CaBr 2 hydrolysis at 1050 K (777 C) is endothermic with the heat of reaction dG T ¼ 181.5 KJ/mol (43.38 kcal/mol) and the Gibbs free energy change is positive at 99.6 kJ/ mol (23.81 kcal/mol). What makes this hydrolysis reaction attractive is both its rate and that well over half the thermodynamic requirements for water-splitting heat of reaction of dG T ¼ 285.8 KJ/mol (68.32 kcal/mol) are supplied at this stage using heat rather than electricity. Molten-phase calcium bromide reactors may overcome the technical barriers associated with earlier hydrolysis approaches using supported solid-phase calcium bromide studied in the Japanese UT-3 cycle. Before constructing the experiment two design concepts were evaluated using COMSOLÔ multi-physics models; 1) the first involved sparging steam into a calcium-bromide melt, while 2) the second considered a ''spraydryer'' contactor spraying molten calcium bromide counter-currently to upward-flowing steam. A recent paper describes this work [6]. These studies indicated that sparging steam into a calcium-bromide melt is more feasible than spraying molten calcium bromide droplets into steam. Hence, an experimental sparging hydrolysis reactor using a mullite tube (ID 70 mm) was constructed capable of holding 0.3-0.5 kg (1.5-2.5 Â 10 À3 kg mol) CaBr 2 forming a melt with a maximum 0.08 m (8 cm) depth. Sparging steam at a steam rate of 0.02 mol/mol of CaBr 2 per minute (1.2-2.3 Â 10 À5 kg/s), into this molten bath promptly yielded HBr in a stable operation that converted up to 25% of the calcium bromide. The kinetic constant derived from the experimental data was 2.17 Â 10 À12 kmol s À1 m À2 MPa À1 for the hydrolysis reaction. The conversion rate is highly dependent on melt depth and the design for steam sparging. This experimental data provides a basis for designing a largerscale sparging hydrolysis reactor for the calcium bromide thermochemical cycle where the endothermic heat of reaction can be effectively supplied by heat transfer coils embedded in the melt.

A three-dimensional stream bed scour modeling methodology was developed using well-benchmarked commercial Computational Fluid Dynamics (CFD) software to compute the bed shear stress distribution used to calculate bed displacements and to re-mesh the computational domain as the bed is displaced. This study extends a previously developed two-dimensional iterative scouring procedure to predict the final shape and size of the scour-hole under pressure-scour flow conditions for flooded bridge decks using commercial CFD software. The current approach uses single phase flow models with an assumed flat water surface using a symmetric slip top boundary to simulate a free-surface flow condition, quasi-steady simulation to obtain the bed shear, and a moving boundary formulation based on an empirical correlation for critical shear stress to iteratively deform the bed under supercritical shear conditions until an equilibrium scour condition is obtained. The model solves the flow field using Reynolds Averaged Navier-Stokes (RANS) equations and the high Reynolds number kepsilon turbulence model using the commercial CFD software STAR-CD. A Bash script was developed to use a Python script to compute bed displacements from the computed shear stress distribution and generate a STAR-CD processor command file to displace the bed followed by a step using the STAR-CCM+ software to remesh the domain as the bed is displaced and bed

In this p ap er, a h e a t conduction equation and a dynam ic th erm oelastic eq uatio n a re briefly dedu ced and estab lish ed based on C ontinuum M echanics. F irst, an q u alitative discussion is em p h atically cen tered around th e co up le term an d th e dynam ic term o f th e equation by m eans o f the dim ensional analysis and by considering th e com bination o f th e ch aracteris tic s o f the m aterials and o f th e therm al load e ffe c te d on the n u clear p ow er station pump u n d er study. S e co n d , form ulations o f th e FEM fo r non-coupled h ea te d equations and q u a s is ta tic th erm o elastic eq uatio n s a re derived in this paper. T h ird , a h a lf space th erm al shock problem is used as a com putational e x a m p le in th e highlighted research on the varying b ehavio r o f th e dynam ic th erm al stress on th e te m p e ra tu re slo p e. T h e conclusion o f th e p ap er provides re lia b le justification fo r applying th e num erical m ethod. F in ally, th e distribution and variety o f th e tem peratu re fie ld , th e therm al stress field an d th e th erm al defo rm atio n field a t various transient m om ents on the pum p a re given.

This study is focused on the simulation of open channel turbulent flow over flooded laboratory scale bridge decks and formation of scour holes under various flooding conditions. Solutions for turbulent flow field are based on Reynolds Averaged Navier-Stokes (RANS) equations and turbulence closure models using the STAR-CD commercial computational fluid dynamics (CFD) software. An iterative computational methodology is developed for predicting equilibrium scour profiles using the single-phase flow model with a moving boundary formulation. The methodology relies on an empirical correlation for critical bed shear stress that is used to characterize the condition for onset of sediment motion and an effective bed roughness that is a function of sediment particle size. The computational model and iterative methodology were stable and converged to an equilibrium scour hole shape and size that compares reasonably well with experiment using a constant critical shear stress value.

An Argonne two-phase combustion ilo';, _c,,_;puter code rh Mass flow rate (ks/s) is u,_..dto simula_ re_tLng flows to aid in the development of n Particle number density (#/m "_) an advanced combustor for magnetohydrodynamic power Nu Nusselt number generation. The combustion code is a general hydrodynamics P Pressure (atm) computer code for two.phase, two-dimensional, steady state, I nonexclu$ive, _oyalty.iree licenr,,e _0 I:x_bli_;h ] or reproduce the pubh_hed lofm of thls conlri_uDOl'l, or allow olhe¢l& 10 ck) 10o,tor

Quality Assurance Statement The Federal Highway Administration (FHWA) provides high-quality information to serve Government, industry, and the public in a manner that promotes public understanding. Standards and policies are used to ensure and maximize the quality, objectivity, utility, and integrity of its information. FHWA periodically reviews quality issues and adjusts its programs and processes to ensure continuous quality improvement. 1. Report No. FHWA-HRT-12-034 2. Government Accession No. 3. Recipient's Catalog No.

Analysis of drag and lift forces and evaluation of scour around bridge structures are necessary to assess risk of bridge failure during storms and floods. Simulation of scour-hole formation under the bridge deck and around the bridge piers, due to sediment entrainment and transport caused by flooding flow conditions, is of significant interest to computational fluid dynamics (CFD) and hydraulics researchers. This study is focused on simulation of 3-D open channel turbulent flow over an inundated bridge deck to obtain the final shape and size of the scour-hole. This 3-D study extends the previous 2-D simulation scouring methodology. Solutions for flow field and turbulence, using an effective bed roughness value, are based on the Reynolds Averaged Navier-Stokes (RANS) equations, and a k- turbulence closure model using a commercial CFD code. An iterative computational methodology is developed to predict the equilibrium 3-D scour-hole using a single phase model moving boundary formulation, based on an empirical correlation for critical shear stress to determine the condition for sediment removal. The computational model has demonstrated capability to reach a converged solution for the equilibrium scour-hole shape and size that agree reasonably well with experiments. The method provides a basis for implementing enhanced particle entrainment physics and other scour model improvements in well benchmarked commercial CFD software.

The rapid advancement of computational capability including speed and memory size has prompted the wide use of computational fluid dynamics (CFD) codes to simulate complex flow systems. CFD simulations are used to study the operating problems encountered in system, to evaluate the impacts of operation/design parameters on the performance of a system, and to investigate novel design concepts. CFD codes are generally developed based on the conservation laws of mass, momentum, and energy that govern the characteristics of a flow. The governing equations are simplified and discretized for a selected computational grid system. Numerical methods are selected to simplify and calculate approximate flow properties. For turbulent, reacting, and multiphase flow systems the complex processes relating to these aspects of the flow, i.e., turbulent diffusion, combustion kinetics, interfacial drag and heat and mass transfer, etc., are described in mathematical models, based on a combination of fund...

This paper discusses the use of various technologies and advanced computational modeling techniques that were combined for monitoring the performance of pier riprap on the basis of a field case study-Pier 3 of a bridge over the Middle Fork Feather River-in northern California, USA. The first phase involved capturing the field condition of the bridge site using sonar instrumentation technology in order to obtain high resolution bathymetry data. The second phase entailed enhancement and transformation of the scanned bathymetric data into a 3D CAD model to be used as the initial geometry for numerical modeling. A Fluid Structure Interaction (FSI) numerical approach was applied to simulate the rock incipient motion i.e. shear failure by coupling Computational Fluid Dynamics (CFD) software STAR-CCM+ and a Computational Structural Mechanics (CSM) software LS-DYNA. Several coupled simulations have been performed with varying flow conditions to identify shear failure conditions for the riprap apron.

Riprap is one of the most common materials used to protect bridge abutment and pier foundations from scour. A key element of the design of riprap countermeasures is rock sizing, which is based on equations generally derived from simplified laboratory experiments. In this study, an advanced modeling approach is developed and applied to evaluate rock stability. The advantage of this approach is that it can incorporate site-specific conditions that complicate riprap design. This report describes this advanced, numerical modeling procedure for analyzing the stability of riprap at bridge abutments and piers. The report will be useful for designers and engineers responsible for protecting bridge foundations. The study described in this report was conducted at the Federal Highway Administration's (FHWA) Turner