A.J. Marquis

Predictions of a turbulent reacting jet in a cross-flow

Abstract The article presents an application of a finite-difference scheme for solving the fluid dynamic equations of three-dimensional elliptic flow to the problem of a turbulent reacting jet discharging perpendicularly into an unconfined cross-flow. The mathermatical model employs a standard two-equation, k -ϵ model to calculate the distribution of Reynolds stresses, with the turbulent nonpremixed combustion process being modeled via the conserved scalar/prescribed probability density function approach. The laminar flamelet concept is used to specify the instantaneous thermochemical state of the combusting mixture. In order to assess the ultimate usefulness of the model for predicting the consequences associated with atmospheric venting and flaring of flammable gases, solutions of the model are compared with experimental data for natural gas flames obtained from wind tunnel studies by Birch et al. Over a range of ratios of cross-flow to jet velocity, predictions of flame trajectory and length are in reasonable qualitative agreement with experimental data. At one particular velocity ratio, for which detailed measurements of the mean temperature field of the flame are available, close agreement between theory and experiment is obtained provided the effects of flame liftoff and radiative heat loss are incorporated into the turbulent flow calculation.

A particle's eye view of crystallizer fluid mechanics

Abstract It has been recognized for some time that flow and mixing in industrial crystallizers has an effect on the kinetics of growth, nucleation and agglomeration and consequently on the crystal size distribution. Yet, a common assumption in population balance modelling is that the fluid mechanical environment experienced by growing crystals is uniform. In practice, however, industrial crystallizers provide extremely varied flow conditions, with local velocities, shear rates and energy dissipation rates varying by orders of magnitude throughout the vessel. A computational fluid dynamics simulation of the flow in a stirred tank was used to illustrate that, in a Lagrangian sense, the particles experienced regions with very different local micromixing characteristics, mean velocities, slip velocities, shear rates and turbulence levels; the sampling of these regions depended only slightly on the particle size and, for the flow considered here, the Eulerian and particle Lagrangian statistics were similar. However, the distribution of slip velocities experienced by the crystals was strongly dependent on the particle microscale and macroscale Stokes numbers. The consequent effects for the estimation of the average growth, nucleation and agglomeration kinetics used in population balances were also considered.

An experimentally validated heat transfer model for thermal management design in polymer electrolyte membrane fuel cells

Abstract The temperature distribution in polymer electrolyte membrane fuel cells (PEMFCs) plays a vital role in defining the overall efficiency and in ensuring the delivery of optimum performance, and understanding the heat transfer taking place is essential for the design of effective thermal and water management systems. This article describes a simple model, validated against experiment, which can be used to investigate the factors such as bipolar plate design, materials of construction, and the external effects of forced convection such as cooling fans and natural convection. The model employs computational fluid dynamics to account for the reactant flows in composite graphite plates and heat distribution within the stack, while convective heat transfer from the external surface of the fuel cell is treated using well-known heat transfer correlations. The computational model was validated using a novel fuel cell analogue composed of an electrically controlled heating plate to simulate the heat generated by the membrane electrode assembly and instrumented with 14 calibrated thermocouples. The model showed good agreement with the experiment over a wide range of gas flowrates, both in terms of local temperature distribution and overall energy balance. This suggests that the novel experimental methodology reported here could be used to support the design of bipolar plates for optimum heat transfer.

Scalar mixing in the vicinity of two disk turbines and two pitched blade impellers

Abstract A LIF line scan system was used to obtain unobtrusive, angle-resolved concentration measurements in the vicinity of four forms of impeller agitating a continuously operated stirred tank. The concentration measurements for the two disc turbines, the Rushton and the `bucket’ impeller revealed strong periodic variations in the radical discharge jet. The measurements for the `bucket’ impeller showed that trailing vortex structures are also generated by curved blades and the trailing vortices of both impellers caused undesired regions of high concentration gradients and fluctuations between blade passages. In the case of the 45- and 60°-pitched blade impeller, periodic concentration variations occurred only in the axial discharge of the impellers and concentration gradients and fluctuations were high close to the blades; this was linked to mean velocity variations between a blade-to-blade passage.

Producing Hydrogen and Power Using Chemical Looping Combustion and Water-Gas Shift

A cycle capable of generating both hydrogen and power with ‘inherent’ carbon capture is proposed and evaluated. The cycle uses chemical looping combustion (CLC) to perform the primary energy release from a hydrocarbon, producing an exhaust of CO. This CO is mixed with steam and converted to H2 and CO2 using the water-gas shift reaction (WGSR). Chemical looping uses two reactions with a re-circulating oxygen carrier to oxidise hydrocarbons. The resulting oxidation and reduction stages are preformed in separate reactors — the oxidiser and reducer respectively, and this partitioning facilitates CO2 capture. In addition, by careful selection of the oxygen carrier, the equilibrium temperature of both redox reactions can be reduced to values below the current industry standard metallurgical limit for gas turbines. This means that the irreversibility associated with the combustion process can be reduced significantly, leading to a system of enhanced overall efficiency. The choice of oxygen carrier also affects the ratio of CO vs. CO2 in the reducer’s flue gas, with some metal oxide reduction reactions generating almost pure CO. This last feature is desirable if the maximum H2 production is to be achieved using the WGSR reaction. Process flow diagrams of one possible embodiment using a zinc based oxygen carrier are presented. To generate power, the chemical looping system is operated as part of a gas turbine cycle, combined with a bottoming steam cycle to maximise efficiency. The WGSR supplies heat to the bottoming steam cycle, as well as helping to raise the steam necessary to complete the reaction. A mass and energy balance of the chemical looping system, the WGSR reactor, steam bottoming cycle and balance of plant, is presented and discussed. The results of this analysis show that the overall efficiency of the complete cycle is dependant on the operating pressure in the oxidiser, and under optimum conditions, exceeds 75%.Copyright

Real-time monitoring of proton exchange membrane fuel cell stack failure

Uneven pressure drops in a 75-cell 9.5-kWe proton exchange membrane fuel cell stack with a U-shaped flow configuration have been shown to cause localised flooding. Condensed water then leads to localised cell heating, resulting in reduced membrane durability. Upon purging of the anode manifold, the resulting mechanical strain on the membrane can lead to the formation of a pin-hole/membrane crack and a rapid decrease in open circuit voltage due to gas crossover. This failure has the potential to cascade to neighbouring cells due to the bipolar plate coupling and the current density heterogeneities arising from the pin-hole/membrane crack. Reintroduction of hydrogen after failure results in cell voltage loss propagating from the pin-hole/membrane crack location due to reactant crossover from the anode to the cathode, given that the anode pressure is higher than the cathode pressure. Through these observations, it is recommended that purging is avoided when the onset of flooding is observed to prevent irreparable damage to the stack.Graphical Abstract

Application of Infrared Thermal Imaging to Map Stress Distributions in a Solid Oxide Fuel Cell

The application of infrared thermal imaging to the study of solid oxide fuel cell components with applied thermal gradients is demonstrated. Thermal imaging of a technologically relevant SOFC sample was carried out while a temperature gradient was induced by heating the sample in a furnace and blowing cold nitrogen at its centre. Subsequently the stress field in the sample was determined by a novel method of converting the thermal image into nodal temperatures for a finite element analysis. Asymmetry in the temperature distribution across the samples was marked, whereas previous versions of the thermal gradient experiment had assumed axisymmetry.

Assessment of Conventional Droplet Evaporation Models for Spray Flames

The present work investigates droplet evaporation rates in inert and reactive environments using fully resolved Direct Numerical Simulation (DNS). The droplets are arranged in regular droplet layers and the evaporation of two different fuels, n-heptane and kerosene, is investigated under engine like conditions. It is found that the performance of standard models fort he evaporation rate strongly depends on the modelling of the representative properties. The conventional 1/3-rule for their computation does not necessarily lead to good agreement between model and DNS. This holds for droplet evaporation in non-reacting and reacting environments. Conditions at the droplet surface would need to be more heavily weighted for better model performance. The droplet loading has a minor effect on the validity of the standard single droplet evaporation models.

Gas-Phase Mixing in Droplet Arrays

Droplet evaporation is usually modelled as a subgrid process and induces local inhomogeneities in the mixture fraction probability density function (PDF) and its scalar dissipation. These inhomogeneities are usually neglected, however, they can be significant and determine the combustion regime. In the present work, Direct Numerical Simulations (DNS) of fully resolved evaporating methanol droplets are analysed, assessing fuel vapour mixing in laminar and turbulent flows. The results show that scalar probability distributions and scalar dissipation vary greatly depending on the position relative to the droplet position, on droplet loading and on flow conditions. The β-PDF seems to capture the global behaviour for laminar flows around droplet arrays with low droplet density, however, mixing characteristics for higher droplet densities in stagnant and turbulent flows cannot be approximated by a β-PDF, and modelling approaches based on cell mean values will lead to erroneous results.

Carbon Capture and Reduced Irreversibility Combustion Using Chemical Looping

Power generation traditionally depends on combustion to ‘release’ the energy contained in fuels. Combustion is, however, an irreversible process and typically accounts for a quarter to a third of the lost work generation in power producing systems. The source of this irreversibility is the large departure from chemical equilibrium that occurs during the combustion of hydrocarbons. Chemical looping combustion (CLC) is a technology initially proposed as a means to reduce the lost work generation in combustion equipment. However, renewed interest has been shown in the technology since it also facilitates carbon capture. CLC works by replacing conventional “oxy-fuel” combustion with a two-step process. In the first, a suitable oxygen carrier (typically a metal) is oxidised using air. This results in an oxygen depleted air stream and a stream of metal oxide. The latter is then reduced in the second reaction step using a hydrocarbon fuel. The products of this second step are a stream of reduced metal, which is returned to the oxidation reaction, and a stream of CO2 and H2 O that can be separated easily. The thermodynamic benefits of CLC stem from the fact that the oxygen carrier is recirculated and can thus be chosen with a reasonable degree of freedom. This enables the chemistry to be optimised to reduce the lost work generation in the two reactors – the reactions can then be operated much closer to chemical equilibrium. It is widely accepted in the literature that a key issue in CLC is identifying the most effective oxygen carrier. However, most previous work appears to consider systems in which a solid phase metallic oxygen carrier is recirculated between two fluidised bed reactors. In the current paper, we explore the possibility of using liquid or gas phase reactions in the two reaction steps since it is hypothesised that these might be compatible with a wider range of fuels including coal. The paper, however, starts by reviewing the existing literature on CLC and the basic thermodynamics of a conceptual CLC power plant. The thermodynamic analysis is extended to include a general method for calculating the lost work generation in a given chemical reactor. Finally, this method is applied to the oxidation reaction of a proposed CLC reaction scheme.Copyright