Electrolyser membrane

To manufacture the brine, a vacuum salt is used to which the producer needs to add a small amount of anti-caking agent which forms a ferrohexacyanide complex in the brine. Because of the acidic process conditions, Fe ions tend to migrate into the electrolyser membranes until encountering a sufficiently high pH and then precipitate [1]. This is an undesirable effect as it can cause void spaces within the membrane and thereby increase the voltage needed for the electrolysis. For this reason the ferrohexacyanide is depleted into Fe(OH)3 under well-defined conditions of temperature, residence time, free chlorine and pH in a process step prior to filtration [2]. 

Another brine species of high interest with respect to the cell voltage and membrane life is aluminium. In the electrolysis cells aluminium forms an aluminosilica complex [1] that can damage the electrolyser membrane. This has a negative effect similar to that of iron migration in terms of power consumption. The necessity then of iron and aluminium removal (to mention only the most important elements) from the brine to their lowest possible levels is obvious. 

Electrolysers can be classified into three basic kinds, based on the type of electrolyte used alkaline water electrolysers, membrane electrolysers (PEM electrolysers) and high-temperature electrolysers. Alkaline water electrolysis, the oldest and, therefore, most widely used technology, is described in more detail below. Figure 10.6 shows a diagram of an alkaline water electrolyser. 

The introduction of NMonNM deposition method has initiated applications as possible route to produce catalysts for fuel cells with improved peifoimance [37, 38, 42]. Different experimental methods were used to characterize electrosorptiOTi characteristics and activity of these modified bimetallic noble metal surfaces for different reactions [38,43,44]. These efforts have led to the design and characterization of one of the most efficient catalysts known today for polymer electrolyse membrane fuel cell anodes [4,45 7]. 

A first application using ferroceneboronic acid as mediator [45] was described for the transformation of p-hydroxy toluene to p-hydroxy benzaldehyde which is catalyzed by the enzyme p-cresolmethyl hydroxylase (PCMH) from Pseudomonas putida. This enzyme is a flavocytochrome containing two FAD and two cytochrome c prosthetic groups. To develop a continuous process using ultrafiltration membranes to retain the enzyme and the mediator, water soluble polymer-bound ferrocenes [50] such as compounds 3-7 have been applied as redox catalysts for the application in batch electrolyses (Fig. 12) or in combination with an electrochemical enzyme membrane reactor (Fig. 13) [46, 50] with excellent results. 

DeNora Permelec SpA (1992) Hydrina membrane electrolysers through (129)... 

Mr K A Stanley ICI Chlor Chemicals, The Heath, Runcorn, Cheshire, WA7 4QF, UK. Practical Operating Differences in Converting a Diaphragm Cell Chlor-Alkali Plant to a Membrane Electrolyser Plant. E-mail keith stanley ici.com... 

For the NaCI electrolysis, finite-gap operation proved initially to be the most promising way to realise good and reproducible cell voltages. Predominantly the reasons were a proper control of the membrane water content, the relatively simple retrofit of existing membrane electrolysers and the ongoing utilisation of the peripheral caustic equipment with the ODC technique. 

Following successful testing of the bubble jet system [3] at pilot scale, the plant was scaled to full technical size (2.5 m2 elements) and successfully tested. The anolyte flow-out of the elements showed a completely pulsation-free operation with all benefits for the membrane lifetime. Despite the rather good results of this first run a design review was started to improve the electrolyser element design. 

DuPont research into high current density and the associated effect on membrane and electrolyser performance has been underway for a decade. It has been the area of greatest concentration for the company during the last 5 years. Studies at the DuPont Experimental Station and Fayetteville Nafion Customer Service Laboratories resulted in polymer innovation and new membrane designs. This work has also identified interactions between membranes and electrolysers... 

This chapter discusses the work at DuPont and provides a simple set of tools to explore the interactions between membranes and electrolysers at high current densities. In fact, these tools help to provide the answer to the following question ... 

If all responses to these tests are linear and typical, and all other independent variables remain within normal operating specifications, it can be assumed that the membrane and electrolyser interactions are optimised for operation within the current density range tested in Section 6.3.1. This procedure has been used successfully to diagnose and optimise operating conditions for both standard and high current density operations where unexpected performance issues have arisen. Furthermore, operators... 

The final phase was completed in May 1999, with the installation of a full-size, commercial pilot cell to enhance the membrane testing capabilities. It is now possible to achieve steady state with the computer-controlled 0.21 m2 electrolyser in just a few hours and operators can obtain meaningful performance test data in 2-3 days. 

The use of membranes for this separation provides the EDC producer with an additional degree of freedom. Higher oxygen contents can be accepted in the cell gas. The cost of a membrane installation can be offset by the cost of upgrading or replacing electrolysers or by the capital and operating cost of providing many connections for additions of acid to the brine. 

Iacopetti, L. (1998) Membrane electrolyser operating at high current density. In Modern Chlor-Alkali Technology, Vol. 7 (ed. S. Sealey), pp. 85-94. Society of Chemical Industry, London and Royal Society of Chemistry, Cambridge. 

In conventional hydrochloric acid electrolysis [1], aqueous hydrochloric acid (HClaq) is electrolysed in a cell, constructed basically from graphite, which is divided by a porous diaphragm or a membrane. The overall reaction is... 

Kvaerner Chemetics have developed a novel, patented process [1] for the removal of multivalent anions from concentrated brine solutions. The prime market for this process is the removal of sodium sulphate from chlor-alkali and sodium chlorate brine systems. The sulphate ion in a brine solution can have a detrimental effect on ion-exchange membranes used in the production of chlorine and sodium hydroxide consequently tight limits are imposed on the concentration of sulphate ions in brine. As brine is continuously recycled from the electrolysers back to the saturation area, progressively more and more sulphate ions are dissolved and build up quickly in concentration to exceed the allowable process limits. A number of processes have been designed to remove sulphate ions from brine. Most of these methods are either high in capital or operating cost [2] or have large effluent flows. 

Iodide is oxidised to iodate or periodate in the membrane cell during the electrolysis process. Iodide, iodate and periodate are therefore present in the brine of a membrane electrolyser. Figure 12.5 shows comparative plots of laboratory adsorption test data for the removal of iodide and other relevant species. 

Practical Operating Differences in Converting a Diaphragm Cell Chlor-Alkali Plant to a Membrane Electrolyser Plant... 

Another aspect that may be taken into account is that of membrane electrolysers having a lower power consumption (Fig. 15.4). Not only does the new technology save power but it also requires less steam to evaporate the cell caustic product to 50%. Additionally, salt removal equipment required in diaphragm plants uses power. This benefit can also be turned around so that for the same power consumed by a diaphragm cell room extra volumes of rayon-grade caustic soda can be produced from the membrane electrolysers. 

Typically, a single-effect membrane caustic evaporator takes around 1.2 tonnes of steam per tonne of caustic soda while a double-effect uses around 0.7 tonne of steam per tonne of caustic soda. An equivalent single-effect diaphragm evaporator uses 4.2 tonnes per tonne. However, a well-run multiple four-effect diaphragm evaporator consumes about 2.1 tonnes of steam to produce one tonne of caustic soda. At a price of US 20 per tonne of steam, a saving of US 4 million per year for a 200 000 tonnes per year plant can be achieved (Fig. 15.6). So what extra equipment is required besides electrolysers ... 

Fig. 15.7 The additional equipment required in a membrane electrolyser plant. 

The chlorine-free brine, still as a weak solution, can then be recirculated to the resaturator. Care should be taken in partial or staged conversions not to feed any diaphragm cell evaporated salt to membrane electrolysers as it may contain chromium and nickel from the evaporators, which are harmful to the membrane. 

For the membrane cellroom of the same capacity there are two choices of technology type either monopolar or bipolar electrolysers. In the case of monopolar membrane electrolysers (Fig. 15.9), such as the ICI FM1500, one membrane electrolyser can replace one diaphragm cell. Since the membrane electrolyser has smaller dimensions there is an overall space saving. The monopolar membrane electrolysers may use the same pipework galleries and overhead crane from the... 

If bipolar membrane electrolysers are installed (Fig. 15.10), such as the ICI BiChlor, then even less floor area is required for the same production capacity. In both membrane cases the space available also depends on the chosen operating current density. Utilising the 150 kA available, nine bipolar electrolysers can operate up to 16.7 kA each. This would require 112 anodes per electrolyser to manufacture the 200 000 tonnes per year of caustic soda, utilising about 360 V of the 450 V available. With bipolar electrolyser centres of 6.5 m, including operator walkways, an area of around 60 m by 14 m or 840 m2 will be required. However, more extensive pipework modifications are required with bipolar arrangements. 

If the choice is to utilise the full capacity of the existing rectifiers and install more membrane electrolysers then adequate space is available. In the 200 000 tonnes per year example, utilising the voltage saved and adding 16 extra monopolar electrolysers would take less space than the original diaphragm cells. In the case of bipolar electrolysers, the length of the electrolyser could be increased as more anodes and cathodes are added to each electrolyser. The number of electrolysers, however, would stay the same. 

Fig. 15.12 Utilisation of rectifier capacity in monopolar membrane electrolysers.

Fig. 15.13 Utilisation of rectifier capacity in bipolar membrane electrolysers.

Staged conversions may also be carried out whereby a number of diaphragm cells is replaced with the equivalent number of membrane electrolysers. For example, four diaphragm cells are removed, to be replaced by five monopolar membrane electrolysers. A special switch would be required of, say, 20 V capacity to enable the work to be carried out with minimal loss of production. A special manifold header enables the pipe connections of the five membrane electrolysers to be fed into the original four diaphragm cell flanges on the headers. 

Replacement of diaphragm cells with bipolar membrane electrolysers requires a different electrical layout (Fig. 15.17) since each bipolar membrane electrolyser can only take about 17 kA of the 150 kA available (for a selected current density). This means that all nine electrolysers need to be installed together. The number of anodes in each bipolar electrolyser can be set depending on the number of diaphragm cells left on load, up to the maximum voltage of the rectifiers. 

---