SubscribeCan’t make up your mind what process to use- ion exchange or membranes?
Jeremy Wakeham, ELGA Process Water
"OK," said the Finance Director, "We need this 0.1µS/cm high purity water for the new boilers and you two," he glared at the Technical Director and Chief Process Engineer, "can't make up your mind what process to use. It's a choice between ion exchange and membranes, is that right?" "In my last company," said the Technical Director, "we looked at two options.
On the one hand we had ion exchange - that's cation-anion exchange with mixed bed polishing - and, on the other, a membrane system consisting of reverse osmosis followed by polishing with continuous electrodeionisation. We found that ion exchange was lower in capital cost, but the high chemical usage more than outweighed the saving."
Chemicals!" said the Environmental Officer, "From what I've heard RO doesn't use chemicals and doesn't give us a waste disposal problem. As far as I'm concerned that's worth a lot." "It's not quite that simple," interrupted the Chief Process Engineer, "I know lots of companies who've had problems with membrane fouling, and they've had to use quite a lot of chemicals to clean the membranes, not to mention dosing acid and anti-scalants. And polishing by electrodeionisation doesn't always meet the 20ppb silica residual we need." "I know that, Jim, but when you're regenerating an ion exchange plant, over half the chemicals go straight down the drain: it's very inefficient.", responded the Technical Director.
"Sorry, Bill, but you're out of date. One of my guys showed me an article from The Chemical Engineer by a person at ELGA Process Water - they were the people who invented SCION, the short cycle ion exchange system that we have down on the process line. Well, they've developed a new regeneration system called Pulsed Regeneration Sequence. The regenerant chemicals - hydrochloric acid and caustic soda - are passed through the exhausted resin bed in short pulses instead of continuous flow.
The static periods allow a more rapid and much closer approach to equilibrium, reducing the regeneration time and further improving the chemical utilisation. ELGA Process Water's new generation of RAPIDE™ deionisers which use this new technology use between 30 and 50% less regenerant chemical than most conventional plants." "Well, they would say that, wouldn't they," chipped in the Environmental Officer, "they're trying to sell ion exchange plants." "Not really, Dave," said the Technical Director rather unexpectedly, "Actually they're the UK's biggest supplier of both technologies, and their new generation of continuous electrodeionisation systems can easily meet 20ppb silica, so they don't really have an axe to grind. And before you start complaining about having to neutralise the spent regenerants, it's usually possible to set an ion exchange plant up to produce a consistently neutral effluent."
"It sounds to me, Jim, that you're really convinced about this," said the Finance Director, "And maybe you're prepared to reconsider, Bill? Let's have a look at my area! What are the relative costs?" "I've done some sums using Bill's figures for reverse osmosis and mine for ion exchange. Our mains water costs 70p/m3 and sewer discharge is the same again. The water is fairly hard with a TDS about 480mg/l. Oh, and it's got 10mg/l of silica.
Bill's reverse osmosis system operates at 15barg at a recovery of 75%, that is 25% of its feed water is rejected to drain, with acid dosing to reduce scaling problems. My ion exchange figures assume a wastewater volume of about 1.5% per 50mg/l/l of TDS in the feed water with 75% utilisation of hydrochloric acid regenerant and 70% utilisation of caustic soda. Here's the analysis." The Chief Process Engineer handed round copies:
Approximate Comparative Costs of Ion Exchange And Ro
| IEx | Membrane | ||
| Raw water | p/m3 treated water | 80.0 | 93.0 |
| Wastewater to sewer | p/m3 treated water | 8.4 | 17.5 |
| Hydrochloric acid | p/m3 treated water | 15.5 | 5.0 |
| Caustic soda | p/m3 treated water | 19.5 | 0.0 |
| Power | p/m3 treated water | 0.3 | 3.5 |
| Media replacement | p/m3 treated water | 2.0 | 10.0 |
| Cost of polishing | p/m3 treated water | 0.5 | 5.0 |
| Capital amortisation | p/m3 treated water | 6.0 | 9.0 |
| Total treated Water Cost | p/m3 treated water | 132.2 | 143.0 |
"The difference isn't all that much, is it?" said the Finance Director, "But water, sewer discharge and power costs are all likely to rise and ion exchange seems less dependent on them." "We've had quite a few instances of mains water quality changes recently," said the Environmental Officer, "It's likely to get worse with the current water shortages. How do the two processes cope?" "Basically," said the Technical Director, "The operating costs of membrane systems are largely independent of mains water quality but ion exchange costs will vary directly - if the mains water TDS goes up so does the operating cost and if the TDS goes down the operating costs do too."
"Media replacement costs are high for reverse osmosis," said the Environmental Officer, "Does that include disposal?" "No, I have to admit it doesn't," said the Chief Process Engineer. "It looks like both used membranes and ion exchange resins would have to go to landfill as process wastes. I've based the sums on replacement every five years. The ion exchange resin volume will be lower than the membranes and replacement is less frequent, so probably ion exchange will just win."
"So it looks like ion exchange?" asked the Finance Director. "I'd say so," replied the Chief Process Engineer, "But it's not a universal panacea - there're lots of applications where reverse osmosis will win. It depends so much on the water quality. If we were buying this for the East Anglia factory, where the water has a higher TDS and lower alkalinity, then I'm sure the membrane option would be more economical." "I'm convinced," said the Technical Director, "But I'd like to know more." "Have a look at the ELGA Process Water web site, Bill, it's atwww.elgaprocesswater.com.