Electrode shape change

Martino et al. have demonstrated the use of BN felt separators in engineering tests. They have high porosity ( 90%), and hence, low ionic resistance, in addition to excellent compatibility with other cell materials at the operating temperature of 470 °C. However, this separator is too expensive and has poor mechanical properties and so cannot prevent electrode shape change during cell operation. ... 

The separators may be simple absorbent material except in Ni/Zn where zinc solubility creates zinc electrode shape change and zinc dendrites, resulting... 

Temperature variations. Essentially all kinetic phenomena are temperature dependent ion diffusion (in both electrolyte and active materials), electron transfer, desolvation, adsorption, etc. Additionally, thermodynamic equilibrium constants are temperature dependent, so any temperature variations within the cell will produce uneven plating and stripping, electrode shape change effects and uneven utilization again leading to compromised performance. 

R. C. Alkire and D. B. Reiser, Electrode Shape Change During Deposition onto an Array of Parallel Strips, Electrochim. Acta, 28, No. 10, 1309-1313 (1983). 

J. Deconinck, Current Distributions and Electrode Shape Changes in Electrochemical Systems, Lecture Notes in Engineering, Vol. 75, Springer-Verlag, Berlin, 1992, pp. 281. 

Electrode shape change between plane parallel electrodes. 

The slightly modified differential form of Faraday s law makes it possible to describe electrode shape change due to reactions as well as mechanical displacement, By introducing the Wagner number in the classical one-dimensional examples of electrode growth and electrochemical machining, the important properties of electrode shape change were obtained. 

For the simulation of electrode shape change or electrochemical machining the BEM offers the advantage that no internal nodes are to be reordered after each time step. This makes the method flexible. Disadvantageous is the fact that the interconnection matrices [G ] and [H ] are to be recalculated completely after each time step which is not necessary using FEM or FDM. 

Even to simulate electrode shape change in ECM, conducting paper analogue methods are described [ 71 ], but the growing power and availability of digital computers makes these methods less competitive. 

The necessity of a simple manipulation of elements -which is very important when electrode shape change has to be simulated - and the fact that the integration a-long the boundary should be fast and accurate led us to opt for straight elements with two nodes, one at each end. 

In retrospect, this restriction was certainly necessary to develop the electrode shape change algorithms. 

In the latter case, this offers also the advantage that all nodes of one element have either essential or natural boundary conditions which are essential in electrode shape change simulations. On the other hand the system of equations is increased. 

In the previous chapter we discussed the possibility to calculate quickly current distributions. The simulation of electrode shape change can now be tackled. Referring to section 1.8.4 >one has to solve equation... 

Finally, in the case of copper electrorefining, experimentally obtained data of electrode shape changes are compared with calculated values. 

In section 3.5.1 we discussed the current distribution in a Hull-cell. This cell has an obtuse and a sharp angle and is therefore an interesting sample problem to check the electrode shape change algorithm. 

These examples show clearly that the boundary element method is very well adapted to electrode shape change simulations. At most 75 nodal points were used and the use of quadratic shape functions could still reduce that number. Using a domain method, an order of magnitude more nodal points would be required to obtain comparable results. 

Partially to check the electrode shape change algorithm and partially because the edge effect in itself is a fascinating phenomenon, we decided to perform experiments. For that purpose, the cell in which the current distributions were measured (section 3.6) was adapted. Also the electrolyte was slightly changed. 

Fig. 4.30 Electrode shape change in a cell with screen. The overshut near the singularity is totally suppressed.

In this chapter we studied the simulation of electrode shape change governed by the potential model. That potential model was discretized by the boundary element method. 

In this work we applied the boundary element method for the solution of many current distribution problems in electrochemical systems, including electrode shape change simulations. 

Predicting Electrode Shape Change with Use of Finite Element Methods". 

Application of a Finite Element Method to Transient Electrode Shape Change". 

The Boundary Element Method (BEM) for the Si-mulation of Electrode Shape Change. ... 

Calculation of Current Distribution and Electrode Shape Change by the Boundary Element Method. ... 

Current Distributions and Electrode Shape Changes in Electrochemical Systems XV, 281 pages. 1992... 

Five years later, Riggs et al. [ 94] presented the first electrode shape change simulations. They used also the finite difference method. 

In 1978, Bergh [ 12] applied at first the finite element method to predict electrode shape changes. Since then, an increasing number of publications on -computer modelling of electrochemical systems, appeared. Mainly the finite difference or the finite element method were used. 

The goal of this investigation was to contribute to this evolution by applying the recently developed boundary element method to the solution of current distribution and electrode shape change problems governed by the potential model. 

Chapter Four deals with the simulation of electrode shape change in which the potential model was discretized by the boundary element method. Electrodeposition, electrochemical machining and electrochemical levelling are treated. The integration with respect to time is performed by an Euler integration as well as by a simple predictor-corrector method. In the former case accuracy and stability conditions are derived. 

---