MEMBRANES set to clean up

Last month's Dubai Desert Classic golf tournament gave a colourful view of what the future could hold for membrane technologies. The 18-hole verdant oasis that hosted the tournament, in the middle of what must be the world's biggest bunker, only exists thanks to desalination of seawater. And desalination on an economically viable world scale is now possible thanks to advances in membrane technologies.

According to US industry analysts, the McIlvaine Company, the ability of reverse osmosis (RO) systems to convert seawater to drinking water at a cost of less than $4/1000gal (about 70p/m3) is justifying projects as far afield as Florida, California, Singapore and many coastal cities around the world. McIlvaine even predicts that such developments will begin to affect population growth, with previously arid areas attracting more immigrants (although not necessarily all wearing Pringle sweaters and spiked shoes).

Desalination apart, membrane-based technologies such as ultrafiltration (UF) and microfiltration (MF) can now also remove potentially harmful organisms such as cryptosporidium from drinking water. And in industry the power, semiconductor and pharmaceutical industries are accelerating their interest in membrane technologies to produce ultrapure water for their processes.

In economic terms, the result predicted by McIlvaine is a worldwide growth of RO, UF and crossflow MF system sales from $3.7billion in 1997 to $6.1billion by 2003. Sales of membranes alone will increase from $900million to $1.7billion over the same period, with the total annual cost of owning and operating these systems rising to $14billion in 2003. In all, sales are expected to grow at an average 9 per cent a year over the next five years.

This burgeoning interest in membrane technologies was underscored at a recent joint Filtration Society/Society of Chemical Industry meeting in London. Presenting an overview of membranes in drinking water treatment, Peter Hillis of NorthWest Water referred to a case study carried out by NWW in 1996 to build a 76Ml/d treatment plant. Membrane plant costs, representing around a half of the total capital expenditure on the treatment plant, would have come to £5.5million - but today the costs would be half that.

As Hillis explained, no single technology will necessarily be the right one for any specific application. As shown in the panel on the previous page, combining technologies can often deliver better results. Another hybrid process described by Hillis uses coagulation together with MF or UF for removing most organic material, reducing disinfectant by-products and controlling fouling. NorthWest Water has recently built a 45Ml/d plant in Preston based on this process.

Fouling can be a problem with any filtration system, but as Linda Dudley of Houseman Desalination Products explained to the meeting, particularly so with membranes. Their high packing density, a requirement for good production rates, results in low voidage for feedwater flow and a consequently high potential for fouling. Although not a new problem, it does need addressing because membrane replacement can be a significant cost in RO plant operation.