E-Cell technology represents a new approach to electrodeionisation (EDI). The concept was developed by GE Glegg Water Technologies of Guelph, Ontario, Canada, which was formed in 1999 when GE Power Systems acquired Glegg Industries, Inc.

In a Reverse Osmosis/E-Cell system, E-Cell technology purifies water for use in a power plant without the need for regeneration chemicals. Unlike conventional mixed-bed technology, E-Cell is a continuous operation process. While earlier EDI systems could manage only low flow rates and often had reliability problems, E-Cell’s robust, modular stack-and-rack design is reliable and cost-effective at all flow rates – from 5 to 2000 gallons per minute (1 to 450 m3/h) and higher.

A brief history of water technology

In the 1960s and 70s, the levels of purity demanded by industry were met by chemically regenerated ion exchange technology. The applications of the day were less demanding, and little consideration was given to the long term effects of chemical usage.

In these early systems, the mixed-bed ion-exchange stage was preceded by separate units for both cation and anion exchange. As applications became more demanding, it was clear that these chemically regenerated ion exchange systems had limitations. A key issue was the unacceptable levels of TOCs (total organic carbons) they permitted. Compared with later technology, these systems used excessive amounts of regeneration chemicals, required the ongoing treatment and disposal of chemical waste, and were complex and expensive to operate.

Through the 1970s and into the 80s, growing awareness of the need to reduce chemical usage prompted the shift to a new paradigm within the water industry. The result was the new application of an existing technology – reverse osmosis (RO) – which used membrane technology to replace the cation/anion units in the primary deionisation system. However, the new technology met with initial resistance. RO needed better pretreatment, and the water treatment system as a whole needed refining.

Gradually the initial problems with RO were overcome, as pretreatment processes were improved and more advanced RO membranes were developed. Over time, reverse osmosis achieved widespread acceptance, and complementary ion exchange technologies such as counter current designs, packed bed ion exchange and specialised resins were also developed. As these new processes came into greater use, costs fell and the RO/mixed-bed systems became cost effective compared with the competing chemically regenerated ion exchange systems of the day.

RO/mixed-bed based systems met the requirements of a wide range of industries for improved water quality. They removed dissolved solids down to a few parts per billion, and reduced TOCs. However, industry continued to rely on mixed-bed technology for the final stage of ion removal. The chemical usage and related infrastructure required by the mixed-bed stage meant the full benefits of RO could not be realised. The continued demand for reduced chemical usage is now driving a second paradigm shift.

What is EDI?

EDI was first developed more than 40 years ago as a chemical-free process, primarily for laboratory work. Recent developments in EDI have made the total elimination of regeneration chemicals a practical reality, and have brought a host of other benefits.

A typical system involves a simple process train: pretreatment; RO; and EDI. A continuous process, EDI removes ions from water using conventional ion-exchange resin, but with a key benefit. In EDI an electrical current is used to regenerate the resin continuously, eliminating the need for periodic chemical regeneration.

A typical EDI stack is made up of a number of dual-chambered cells sandwiched between two electrodes. Ion-exchange resin is located between two membranes which make up one of the cell pairs: a cation membrane, specifically designed to allow migration of only cations, and an anion membrane, which allows the migration of only anions. This is the D Chamber in the diagram above.

The resin bed is continuously regenerated by an electric DC current applied across the cell. When contaminant ions such as sodium and chloride are present in the feedwater, they undergo the usual ion-exchange reactions in which they attach to their respective ion-exchange resins, displacing hydrogen and hydroxyl ions as in conventional mixed-bed ion-exchange. Once within the ion-exchange resin, the contaminant ions move through the resin from bead to bead, until they permeate the membranes and enter the adjacent concentrate stream, the C Chambers. Sometimes salt (NaCl) is added to the concentrate loop to improve current efficiency and reduce the amperage needed for continuous regeneration. This concentrate stream of contaminant ions is then swept out of the stack.

In a typical EDI system, 90-95 percent of the feed water is directed through the D Chamber while 5-10 per cent is diverted through the C Chambers. Concentrated ions are removed from the stack by bleeding off a percentage of the water from the “concentrate loop”. This water, with a pH typically in the range of 5 to 8, can be recovered and directed back to the inlet of the RO system.

During the electrodeionisation process, the ionic contaminants in the feed water are removed resulting in highly deionised water.

The advantages of EDI

EDI replaces the primary mixed-bed in conventional water treatment systems, predictably and consistently producing water of the highest quality. EDI’s most important advantage is that no chemicals are required for regeneration. Therefore, no bulk storage or neutralisation tanks for regen chemicals are needed. And the receiving, storage and disposal of hazardous chemicals are dramatically decreased. As a result, EDI offers a significant reduction in system infrastructure.

Where RO reduced the need for especially large facilities, the latest EDI technology completely eliminates it. Tall ion exchange vessels are no longer needed as part of the normal building requirement, since an EDI system typically requires a ceiling no higher than 18 feet. The elimination of special height specifications is especially important when tight deadlines demand that a power plant or processing facility get up and running quickly.

A further advantage is that the EDI reject stream contains only the impurities from the feed water and is usually of higher quality than the feed water from the pretreatment system. As a result, reject waste can be fed continuously into the system directly ahead of the RO, effectively eliminating a waste stream. In contrast, mixed-bed regeneration is a batch process, and because chemicals are used to regenerate the resin bed, the waste stream contains 3 to 4 times the ionic waste of the typical EDI waste stream. This batch waste stream is usually not fed back into the pretreatment system but is discarded in a waste neutralisation tank.

RO-EDI is a continuous process. It produces consistent water quality, unaffected by the ionic leakage that occurs at the beginning and end of every regeneration cycle in the mixed bed. An RO-EDI system only requires a quick flush and is typically operational within a few minutes of starting the system. This continuous process also simplifies operation. The operators and operating procedures associated with recurring regeneration cycles are no longer required.

The latest EDI technology

Until ten years ago, EDI had not been developed for use in industrial applications. As a result the technology had serious shortcomings. High costs limited its use to applications that required only low flow rates. What’s more, early EDI technology had some reliability issues, and system designs were complex.

Today, EDI can handle even the highest flow-rate applications. GE-Glegg’s E-Cell system, the latest EDI technology, is cost effective at flow rates as high as 2000 gallons/minute and beyond. Capacity can be easily increased or reduced simply by adding or removing stacks.

The latest E-Cell equipment is comparable with primary mixed-bed systems in cost. Innovative approaches that simplify membrane composition and decrease membrane utilisation have resulted in substantial savings over earlier technology. EDI operating costs typically are below those for mixed bed, with lower labour and maintenance costs, and the cost of electricity roughly matches the cost of chemical regeneration for mixed beds.

The demands being met by current E-Cell EDI technology are well beyond those seen in pharmaceutical and electronics applications in the past. For the first time, EDI systems are being truly industrially designed, with the capacity to operate continuously at pressures of 100 psi without leaks.

Using a modular stack-and-rack design, E-Cell systems are expandable, and easy to install and maintain. Similar to an RO system, they incorporate standard building blocks that are simple to replace and can be mass produced, further lowering costs. While spare capacity can be designed into the system, only the optimum number of stacks required to meet existing flow requirements need be purchased initially.

Unlike mixed-bed equipment, where the entire system must be taken off line if maintenance or repair is required, with EDI – again, similar to RO – overall system performance is unaffected if any one stack requires servicing. The load is simply distributed among the remaining stacks. As mentioned earlier, such a stack-and-rack approach also allows the water treatment system to be designed to accommodate the specific space requirements of a given building.

E-Cell stacks currently in use are guaranteed to produce resistivities in excess of 16 Mohm-cm. In fact, RO-EDI based systems incorporating final treatment are capable of meeting the most demanding water quality requirements in the industry today. As a result of recent improvements such as E-Cell, current EDI technology is meeting the requirements of today’s most demanding applications and is achieving mass acceptance across a broad range of industries.

A powerful case history

Water treatment applications for the power generation industry currently account for more than 30 per cent of global industrial water treatment sales. More than 200 E-Cell based water treatment systems have been sold; one recent example is Transalta’s Sundance Generating Station located near Duffield, Alberta, Canada, on the south shore of Lake Wabamun.

This 2100 MWe coal-fired plant meets approximately one-third of Alberta’s electrical demand. Over the past few years, two ageing and high-maintenance, conventional ion-exchange water treatment plants and a change in EPRI guidelines for boiler feedwater prompted Transalta to look for a new feedwater treatment technology.

The nominal feedwater usage is around 350 US gpm (gallons/minute) for each of the two water treatment plants. To meet this capacity, a dual-train, 700 US gpm micro-filtration/reverse osmosis/EDI feedwater treatment plant was designed, built and commissioned in 1999, and began producing water in February 2000. The new plant has consistently been available to produce high quality, ultra-pure boiler feedwater at full capacity since it was commissioned.

The raw water source is either lake water or recycled cooling water. A GE-Glegg two-pass RO system produces treated water below 1 µS/cm at 75 per cent water recovery and feeds the E-Cells. The ultra-pure water from the E-Cells typically is above 17 Mohm-cm, and is used as the boiler feedwater.

Transalta reports that the new water treatment plant has reduced the cost of producing boiler feedwater by one-third. In addition, chemical usage has been reduced by more than 90 per cent. Added benefits from the use of ultra-pure water in steam generation and power production include a large reduction in cycle chemistry production derates, fewer high pressure boiler tube failures, and increased turbine blade life. Transalta selected the GE Glegg E-Cell EDI system because of its modular design, cost and Glegg’s history as a technology leader in the field of industrial water treatment.

The future of water treatment

Current RO-EDI technology is substantially reducing chemical usage, simplifying operations and reducing overall inherent costs. Although there are a number of challenges to the continued pursuit of using fewer and fewer chemicals in industrial water treatment, new advances in filtration and RO membrane technology will further improve the effectiveness of RO-EDI based systems.

What does the future hold? In all industry segments including power generation, there is growing emphasis on reduced chemical usage. In three to five years it is expected that 85 per cent of all industrial water treatment systems will be RO-EDI based systems. The remaining 15 per cent will be special applications where the incremental costs of pretreatment and other considerations do not offset the inherent benefits of a chemical-free RO-EDI system.

Electrodeionisation is now well established in the power generation marketplace and is achieving wide acceptance across a broad range of industries. The direction of the technology will be determined by the continued needs of industry for pure water – produced without chemicals.