Christian Jallut

Modelling of frost growth and densification

A one-dimensional transient formulation is derived to predict frost growth and densification on a cold wall submitted to a moist air flow. The model is based on a local volume averaging technique that allows the computation of temperature and density distributions throughout the entire frost layer according to time. It is particularly shown that the effective vapour mass diffusivity throughout frost should reach values several times larger than the molecular diffusivity, so that the model is in agreement with experimental data. In order to represent this phenomenon, a new expression for the so-called diffusion resistance factor is proposed. Values of an adjustable parameter appearing in this expression are correlated to heat and mass transfer boundary conditions, and to the global rate of densification. Two possible interpretations of the phenomenon are proposed.

Bond graph modelling for chemical reactors

In this paper we present a bond graph model of a continuous stirred tank reactor which represents the reaction kinetics as well as the heat and mass transport phenomena in the reactor. The consequences of reticulation of the phenomena and of the systematic use of the power conjugated variables on the formulation of the thermodynamic properties, the reaction kinetics and the energy and mass transport are shown. A classical example of chemical reaction is chosen to illustrate this approach: the equilibrated reaction of hydrogen and iodine in hydrogen iodide.

A Dynamic Mechanistic Model of an Electrochemical Interface

We propose a dynamic mechanistic model, based on nonequilibrium thermodynamics and electrodynamics, describing the transient response to current perturbations of an electrochemical double layer at the metal/electrolyte interface in the presence of electrochemical reactions. As an example of application, we have simulated the hydrogen oxidation reaction taking place in a polymer electrolyte fuel cell anode. The model is composed of (i) a 0-D inner layer submodel describing the composition of the metallic phase surface where water and reactant can be adsorbed, and the generated electric potential difference between the metal and the electrolyte phases; and (ii) a 1-D diffuse layer submodel in the electrolyte constituted by spatially moving ions and counterions, describing the ionic transport by migration-diffusion, based on the coupling of a Nernst-Plancks equation with a Poissons equation. At the interface, the reaction kinetics depending on the potential difference is coupled with the inner-layer model through the charge conservation law. The numerical model allows dynamic simulation of the evolution of the local electric potentials (ionic and electronic) and concentrations inside the interface, and the influence of the working conditions on the impedance spectra characteristics.

Structured modeling for processes: A thermodynamical network theory

We review the use of bond graphs for modeling of physico-chemical processes. We recall that bond graphs define a circuit-type language which root on a thermodynamical consistent definition of its network elements. We present the bond graph basic elements in the light of lumped models arising from chemical engineering. We first illustrate it on the historical example of the diffusion process through a membrane. The examples of a Continuous Stirred Tank Reactor and an adsorption process illustrate how the network structure and the choice of variables ease the reusability of submodels and localize the changes in models to some network elements.

Enthalpy based modelling and design of asymptotic observers for chemical reactors

This article proposes to consider the basic thermodynamics based formulation of the energy balance equation for chemical systems with a limited number of simplifying assumptions. The objective is to show, via the design of one typical mass and energy balance state observer, how such design can be modified by considering the proposed thermodynamically based model formulation. The objective is also to emphasise the difference and the links between the energy balance-based temperature equation largely used in process control. The design of the asymptotic observer is illustrated with two examples: one CSTR in liquid phase and another one in gaseous phase.

A multiscale physical model for the transient analysis of PEM water electrolyzer anodes

Polymer electrolyte membrane water electrolyzers (PEMWEs) are electrochemical devices that can be used for the production of hydrogen. In a PEMWE the anode is the most complex electrode to study due to the high overpotential of the oxygen evolution reaction (OER), not widely understood. A physical bottom-up multi-scale transient model describing the operation of a PEMWE anode is proposed here. This model includes a detailed description of the elementary OER kinetics in the anode, a description of the non-equilibrium behavior of the nanoscale catalyst-electrolyte interface, and a microstructural-resolved description of the transport of charges and O(2) at the micro and mesoscales along the whole anode. The impact of different catalyst materials on the performance of the PEMWE anode, and a study of sensitivity to the operation conditions are evaluated from numerical simulations and the results are discussed in comparison with experimental data.

Lyapunov based control for non isothermal continuous stirred tank reactor

Abstract In this contribution we apply the approach of passivity proposed by Ydstie [M. Ruszkowski, V. Garcia-Osorio, and B.E. Ydstie. AIChE Journal, 2005] for physico-chemical processes. The originality of this work lies in the fact we consider a thermodynamically nonlinear consistent model for a continuous stirred tank reactor to built the appropriate Lyapunov function for stabilization purpose. Indeed the kinetics of reaction modelled by Arhenius law leads to non linear model with multiple steady state. We propose to stabilize the reactor around the unstable point. In order to apply the Ydstie approach, we assume that the fluid remains homogeneous. This assumption permits to use the concavity property of the entropy function to build the Lyapunov function. We propose feedback laws in order to ensure the closed loop properties of the Lyapunov function. Finally we propose some simulation results.

Port-based modelling of mass transport phenomena

The goal of this article is to present an extension of the port-based modelling approach (bond graphs) which applies to systems subject to heat and mass transfer. The methodology is based on the first principle, conservation laws and constitutive closure relations. The latter are the phenomenological laws relating fluxes and thermodynamic forces. Then instantaneous power conservation appears naturally as a geometric interconnection structure called Dirac structure. The multi-level case (several macroscopic spatial scales) is investigated with the assumption that the spatial scales are separated and may be considered as two distinct phases. In this case, it is shown that both the interconnection coupling within a phase and the multi-level interconnection coupling are Dirac structures.

Non-isothermal gas-liquid absorption with chemical reaction studies. Temperature measurements of a spherical laminar film surface and comparison with a model for the CO2/NaOH system

The present work deals with the temperature measurements of a thin spherical liquid film surface during non-isothermal absorption with chemical reaction. The absorption of CO2 into aqueous NaOH solutions has been used to test the proposed technique. The measurement is based on the use of an infrared pyrometer through a ZnSe window. The experimental results are compared to calculated values obtained from a previously described non-isothermal model. The measured and computed values of the flux of absorbed gas and the temperature profiles are very close without performing any parameter adjustment. The highest measured temperature increase is about 5 K, and despite this low thermal effect, the system has proved to be highly sensitive.

Modeling adsorption properties on the basis of microscopic, molecular, and structural descriptors for nonpolar adsorbents.

We propose a method for analytically predicting single-component adsorption isotherms from molecular, microscopic and structural descriptors of the adsorbate-adsorbent system and concepts of statistical thermodynamics. Expressions for Henrys constant and the heat of adsorption at zero coverage are derived. These functions depend on the pore size, pore shape, chemical composition, and density of the adsorbent material. They quantify the strength of the solid-fluid interaction, which governs the low-pressure part of the adsorption isotherm. For intermediate and high pressures, the fluid-fluid interactions must also be taken into account. Both solid-fluid and fluid-fluid interactions are combined within the framework of the Ruthven statistical model (RSM). The RSM thus constructs theoretical adsorption isotherms that are entirely based on microscopic molecular and structural descriptors. The theoretical results that we obtained are compared with experimental data for the adsorption of pure CO2 and CH4 on all-silica zeolites. The developed methodology allows for the estimation of the optimum properties of a nonpolar adsorbent for the adsorption of CO2 in cyclic adsorption processes.