Introduction to Electrodeionization

Introduction to ElectrodeionizationWelcome to Introduction to EDI. This section is dedicated to those who are just learning about Continuous Electrodeionization also known in generic terms as Electrodeionization (EDI). We will provide you with a basic understanding of EDI and why it is important. It will also better prepare you for the content in the Intermediate EDI section

Introduction to EDI gives you the basics. This section helps you obtain information on EDI fundamentals, definitions and how EDI works. It also enables you to see the benefits of EDI, understand how to design an EDI system and watchouts during design and operation to maintain your system.

This section is broken down into the following topics:

Advantages of EDI

Old vs. New Technologies

Development History

EDI or Ion Exchange?

Continuous Operation

Where EDI works

Comparing Old and New Technology

Benefits of EDI

Salt Movement in an Electric Field

Ion Exchange Membranes

Mechanisms of Ion Removal

Advantages of EDI

EDI utilizes chemical-free regeneration. This means a safer workplace because there is no need to store or handle hazardous acid and caustic. There are fewer regulation concerns due to the absence of these corrosive chemicals and there are no waste neutralization or disposal issues.

EDI is a continuous process. The ion exchange resins are being continuously being regenerated by the DC electric field. There is no breakthrough" of ions as happens in conventional ion exchange operations, therefore the quality of the water remains at a constant high level of purity. The electric field also provides a bacteriostatic environment inside of the EDI cell, inhibiting the growth of bacteria and other organisms.

EDI has significantly lower operating costs than conventional ion exchange processes. Only a relatively small amount of electric power is needed to provide high purity water. The lack of acid and caustic regeneration means less operator attention and lower labor costs. Capital costs can also expected to be lower, especially because no chemical storage, pumping and neutralization equipment is required.

EDI has a significantly smaller footprint than conventional ion exchange processes. This means less plant space will be required to provide the same quantity of water.When to Consider Using EDI

EDI may be considered to be a competitive alternate process to:

Regenerable Mixed Bed Deionization

No acid or caustic bulk storage, pumping, waste neutralization or disposal issues. Lower operating cost due lower manpower requirements as well as the lack of chemical regeneration Smaller footprint.

Service Mixed Bed Deionization (off-site regenerated rental vessels)

No ionic breakthrough resulting in a constant high quality of water. No rental vessels, associated freight or monthly demurrage charges.

Second pass of RO

Eliminates the need for a second bank of RO membranes and associated plumbing, pumping and control equipment.Typical applications for EDI

Providing USP (United States Pharmacopeia) grade water

Purified Water

Feed to WFI (Water For Injection) stills

Steam Generation / Boiler Feed

Microelectronics / Semiconductor makeup and rinsewaters

For high quality makeup demineralized water

General Industry

Surface finishing

Chemical manufacturing

Hospital & University Central Systems Advantages of EDI - Pharmaceutical Applications

EDI runs in abacteriostatic service mode. The electric field minimizes bacteria growth in the resin beds. Some EDI devices hot water sanitizable, resulting in more effective sanitization, faster rinseup and more easily validated systems. RO/EDI easily meets Stage 1 conductivity specification, while conventional 2-Pass RO (RO/RO) does not, due to CO2and/or NH3.

Advantages of EDI- Steam Generation Applications

EDIs continuous electrochemical regeneration provides a constant water quality with no breakthrough" of bands of ions. EDI is a low maintenance process requiring little operator attention. EDI is also ideal for remote locations where you might be dependent on delivery of chemicals or DI tanks. EDI can provide low product water conductivity, of much better quality then would be possible with 2-pass RO alone

Advantages of EDI - Microelectronics Applications

EDI provides high quality water, low in particles, partiall due to the fact that there is no resin attrition from backwashing or osmotic shock, as would be the case with conventional ion exchange processes. EDI also provide water with a lower TOC (Total Organic Carbon) content because the electric field will remove charged organic molecules. Boron removal is better on average than mixed bed deionization, unless the mixed bed columns are regenerated very frequently.

Perceived Limitations of EDI

LeaksLeaks have been completely eliminated in some module designs.

Sensitivity to chlorine

EDI is just as sensitive to the chlorine as thin-film reverse osmosis (RO) processes. This is easily addressed with proper pretreatment system design.

Sensitivity to hardness

Most EDI devices have a 1 ppm hardness limit which is easily addressed with proper pretreatment system design.

Not suitable for high CO2loads

No ion exchange process is cost-effective for removal of large amounts of CO2.Removal Mechanisms

While both ion exchange and EDI use ion exchange resins, the removal mechanisms are quite different. Conventional ion exchange utilizes chemically regenerated ion exchange resins which function in a capture (exhaustion cycle) and discharge (regeneration cycle) mode. This results in a breakthrough of ions at the end of the service cycle and a rinseout of regenerant at the beginning of the next service cycle. Capacity & selectivity are most important resin properties in this mode of operation.

EDI (continuous electrodeionization) utilizes a reaction/transport mechanism to remove ions (through resin under influence of DC field). This requires continuous path of like-charge resin beads. The transport is largely across the surface of the resin beads. Transport through resin bead (particle diffusion) can be limiting.

Water Treatment - Old vs. New Technologies

Conventional water treatment systems rely on chemically-regenerated ion exchange resins to remove dissolved solids. Regeneration chemicals are costly, hazardous and, even though they are neutralized prior to releasing to streams and rivers, add a significant amount of dissolved solids to the waterways. These systems typically use cation exchange vessels followed by anion exchange to handle the bulk of the demineralization. Mixed bed exchangers are used in many cases to polish the treated water to a resistivity of 18 megohms. The waste regenerants from these systems are usually combined, neutralized and released to the environment. The ion exchange systems are usually supplied in duplicate, to allow one system to provide water while the other one is being regenerated. Regenerations of ion exchangers typically takes several hours, require bulk storage and pumping facilities for regenerant chemicals, and usually require a waste neutralization tank.

State-of-the-art water treatment systems utilize reverse osmosis (RO) membranes to do the bulk of the demineralization. RO systems do not require chemical regeneration and also remove many types of total organic carbon (TOC) which will pass through ion exchange resins. (In some cases, the ion exchange resins actually contribute to the TOC content in the water). The polishing of the RO product water is carried out by continuous electrodeionization (EDI) which is capable of producing water in excess of 18 megohm resistivity. Duplexing of large EDI systems is not required, because a single EDI module may be taken off line for maintenance or repair while the remaining modules operate at a slight increase in flow rate to maintain the required flow through the system. EDI modules can be sized to operate from a fraction of a gpm up to about 50 gpm. The EDI process is a continuous process, utilizes no chemicals for regeneration, does not pollute the environment and requires a fraction of the operator attention necessary for conventional ion exchange systems. EDI systems typically run between 90 and 95% water recovery. The reject stream is usually of better quality than the feed to the RO system, enabling the reject stream to be completely re-used by pumping it back to the pretreatment section of the RO system. RO/EDI systems may achieve overall water recoveries of greater than 90% by recycling a significant portion of the RO reject stream and all of the EDI reject stream back to the front end of the treatment system. Development HistoryEDI theory and practice have been advanced by a large number of researchers throughout the world. It is believed that EDI was first described in a publication by scientists at Argonne Labs in January 1955 as a method for removal of trace radioactive materials from water (Walters, et. al.). One of the earliest known patents describing a EDI device and process was awarded in 1957 (Kollsman). It is thought that the first pilot device incorporating mixed resins was developed by Permutit Company in the United Kingdom in the late 1950's for the Harwell Atomic Energy Authority, as described in a paper (Gittens and Watts) and in more than one patent (Kressman; Tye). One of the first detailed theoretical discussions of EDI was written in December 1959 (Glueckauf). In April 1971, a Czechoslovakian researcher reported results of his experimental and theoretical work that advanced the theory of ionic transport within a EDI device (Mateka). Layered bed devices were described in the patent literature in the early 1980's (Kunz).

EDI devices and systems were first fully commercialized in early 1987 by a division of Millipore that is now part of United States Filter Corporation (Ganzi et. al., 1987). Since then, the theory and practice of EDI has advanced worldwide, and commercial EDI devices are now manufactured by a number of companies (Towe et. al.; Parsi et. al.; Rychen et. al., Stewart and Darbouret). There are now several thousand EDI systems in commercial operation for the production of high purity water at capacities ranging from less than 0.1 to more than 250 m3/h. This includes EDI systems that have been in continuous operation for nearly ten years, producing makeup water for high pressure boilers.

Continuous Electrodeionization or Ion Exchange?

Companies are constantly working to reduce operating costs, improve efficiency and eliminate the use of hazardous chemicals in the workplace. Such goals have caused an increase in the use of continuous electrodeionization technology to produce high-purity water.

Continuous electrodeionization (EDI) uses a combination of ion exchange resins and membranes, and direct current to continuously deionize the water without regeneration chemicals.

The principle behind ion exchange is that polymer resin beads are chemically structured to provide either positively or negatively charged fixed functional groups that attract and remove certain contaminant ions from the water. Conventional ion exchange technology can remove dissolved inorganics such as minerals and salts and some dissolved organics. It does not remove particles, colloids, bacteria or pyrogens.

Cationic resins remove positively charged ions such as calcium, magnesium and sodium, replacing them with hydrogen (H+) ions. Anionic resins remove such negatively charged ions as chloride, nitrate and silica and replace them with hydroxide (OH-) ions. The hydrogen and hydroxide ions then unite to form water molecules.

When the water passes through a tank containing a mixture of both cation and anion exchange resins, the process is called mixed-bed ion exchange. Mixed-bed systems can produce very high-quality water with resistivities up to 18.2 megohms-cm.

Over time, however, the resin beads become saturated with contaminant ions and become less effective at treating the water. Also, the high-purity water flowing past these saturated resins may actually extract trace amounts of contaminant ions by "chromatographic effects," causing a decline in water quality. The exhausted resins must be chemically regenerated off-line before reuse. During regeneration, the cation resin beads are restored to their hydrogen form by treating them with hydrochloric acid (HCl) or sulfuric acid (H2SO4). The anion resin beads are restored to their hydroxide form by treating them with caustic (NaOH).A Continuous OperationLike conventional ion exchange, continuous electrodeionization removes dissolved, ionizable materials such as salts, acids and bases. It can also remove weakly ionized materials such as dissolved silica, carbon dioxide and some organics. Contaminants too large to pass through the ion exchange membrane -- such as large particles and large organic molecules -- are not removed.

Continuous electrodeionization modules consist of mixed-bed resins sandwiched between alternating anion and cation membranes. These membranes are actually ion exchange resins manufactured in sheet form. Resin compartments in this "sandwich" construction alternate between diluting and concentrating compartments. Compartment sets are called cell pairs and form the basic element in a module.

In the module, direct current is applied to the anode (positive electrode) on one end of the module, and to the cathode (negative electrode) on the other end. This electric potential drives the ions captured by the ion exchange resins through the membrane.

Because the resins in the module are continuously regenerated by the electric current, they do not become exhausted. This continuous electro-regeneration enables continuous electrodeionization systems to produce multi-megohm water without the need for chemical regeneration or downtime.

Commercially available continuous electrodeionization modules are normally plate-and-frame devices with varying numbers of cell pairs to accommodate flow rates from 0.5 gallons per minute (gpm) to 1,000 gpm. Customized systems can produce even higher flow rates.

Where it Works

Continuous electrodeionization can be chosen for new projects that require ultra-high purity water or stringent wastewater discharge requirements. It can also be a cost-effective way to upgrade existing ion exchange systems.

Applications include creating process water for the biotechnology and food and beverage industries, and providing high-quality rinse water for electronics, surface finishing and optical-glass processes. For electronics applications, continuous electrodeionization can reduce total organic carbon (TOC) levels in the water.

Because it is able to meet United States Pharmacopoeia (USP) purified water specifications -- including those for bacteria and pyrogens -- continuous electrodeionization systems are frequently used in the pharmaceutical industry, hospitals, university research facilities and dialysis centers.

Electrodeionization (EDI) is a process that removes ionizable species from liquids using electrically active media and an electrical potential to effect ion transport. The electrically active media in EDI devices may function to alternately collect and discharge ionizable species, or to facilitate the transport of ions continuously by ionic or electronic substitution mechanisms. EDI devices may comprise media of permanent or temporary charge, and may be operated batchwise, intermittently, or continuously.

The continuous electrodeionization (EDI) process, a subset of EDI, is distinguished from the EDI collection/discharge processes such as electrochemical ion exchange (EIX) or capacitive deionization (CapDI), in that EDI performance is determined by the ionic transport properties of the active media, not the ionic capacity of the media. EDI devices typically contain semi-permeable ion-exchange membranes and permanently charged media such as ion-exchange resin. The EDI process is essentially a hybrid of two well-known separation processes - ion exchange deionization and electrodialysis, and is sometimes referred to as filled-cell electrodialysis.

This section will focus on EDI devices, and more specifically on the use of such devices in conjunction with reverse osmosis to produce water of sufficient quality to feed to high pressure boilers.Benefits of CEDIConventional ion exchange is a viable treatment option for applications where high-flow and high-conductivity requirements are not critical. In other situations, however, cost comparisons between ion exchange and continuous deionization may reveal potential economic advantages.

The biggest difference is that continuous electrodeionization eliminates all chemical regeneration and waste neutralization steps. While capital equipment costs may be higher with continuous deionization systems, operating expenses are lower because there are no regeneration chemicals, and labor or maintenance costs are less.

Electrical requirements are nominal. A typical system uses one kilowatt-hour (KwH) of electricity to deionize 1,000 gallons, based on a feed conductivity of 50 micromhos/cm and a product water resistivity of 10 megohm-cm.

Continuous electrodeionization systems do not require duplexing (two separate treatment units) which can increase the cost, complexity and size of the system.

Both conventional ion exchange and continuous electrodeionization may require pretreated feedwater to prevent scale formation and plugging by colloids and particles. Pretreatment is also required to reduce high levels of free chlorine and organic foulants. The type of pretreatment required is determined by product water quality requirements. For most high-purity water needs, a reverse osmosis (RO) pretreatment step is sufficient.

With RO pretreatment, continuous electrodeionization systems can achieve better than 99.5 percent salt removal, reduce the levels of individual ionic species to parts-per-billion or even parts-per-trillion levels, and produce high-purity water with resistivities of 10 to 18 megohm-cm (or 0.1 to 0.055 micromho/cm conductivity).

Continuously regenerating the ion exchange resins also removes the possibility of exhausted or improperly regenerated resins contaminating the product water.

Continuous electrodeionization systems typically convert 80 to 95 percent of the feedwater into product water. The waste stream can be discharged without treatment or recycled back to the RO pretreatment unit.

It can also reduce facilities costs because waste neutralization equipment and hazardous fumes ventilation equipment are not required. The elimination of regeneration chemicals can help improve workplace health and safety, as well as prevent corrosion from hydrochloric acid fumes. The costs of meeting EPA and OSHA requirements and "right to know" laws are also reduced.

There are significant tangible cost benefits associated with the elimination of regeneration. The costs of regeneration labor and chemicals are replaced with a small amount of electrical consumption. A typical EDI system will use approximately 1 kW-hr of electricity to deionize 1000 gallons from a feed conductivity of 50 microsiemen /cm to 0.1 S/cm product conductivity. Since the EDI concentrate (or reject) stream contains only the feed water contaminants at 5-20 times higher concentration, it can usually be discharged without treatment, or used for another process. Thus facility costs can also be reduced since waste neutralization equipment and ventilation for hazardous fumes are not necessary.

There are also less tangible cost reductions, which are harder to quantify, but usually favor the use of EDI systems. By eliminating hazardous chemicals wherever possible, workplace health and safety conditions can be improved. With today's increasing regulatory influence on the workplace, the storage, use, neutralization, and disposal of hazardous chemicals result in hidden costs associated with monitoring and paperwork to conform to safety and environmental requirements (such as the US EPA and OSHA laws). In addition, the fumes, particularly from acid, often cause corrosive structural damage to facilities and equipment.

For the most part the elimination of regenerant chemicals is considered advantageous, but the chemicals do offer at least one benefit. In conventional demineralizers, acid and caustic is typically applied to the ion exchange resins at concentrations of 2-8% by weight. At these concentrations the chemicals not only regenerate the resins but clean them as well. The electrochemical regeneration that occurs in a EDI device does not provide the same level of resin cleaning. Therefore proper pretreatment is even more important with a EDI device, in order to prevent fouling or scaling. This is one of the reasons that RO pretreatment is normally required upstream of a EDI system. In general, the feed water requirements for EDI systems are more stringent than for chemically regenerated demineralizers.

In summary: EDI does not require chemicals (as does DI resin regeneration)

EDI does not require shutdowns

EDI modules are the smallest and lightest per unit flow on the market, EDI systems are compact as well.

Consistent, continuous ultrapure water

Requires little energy

Economic use of capital, no operating expenses, just the cost of power.

Reduced facility requirements Minimal operator supervision requiredHow does it WorkFirst, lets start with a container of water with salts (cations and anions) in solution.

A typical EDI device contains alternating semipermeable anion and cation ion-exchange membranes. The spaces between the membranes are configured to create liquid flow compartments with inlets and outlets. A transverse DC electrical field is applied by an external power source using electrodes at the ends of the membranes and compartments.

When the compartments are subjected to an electric field, ions in the liquid are attracted to their respective counter-electrodes. The result is that the compartments bounded by the anion membrane facing the anode and the cation membrane facing the cathode become depleted of ions and are thus called diluting compartments. The compartments bounded by the anion membrane facing the cathode and cation membrane facing the anode will then ?trap? ions that have transferred in from the diluting compartments. Since the concentration of ions in these compartments increases relative to the feed, they are called concentrating compartments, and the water flowing through them is referred to as the concentrate stream (or sometimes, the reject stream).

Ion Exchange MembranesNow let's add some ion exchange membranes to direct the ions into different flow channels as shown in the animation below. The red membranes are cation-selective membranes and the blue membranes are anion-selective membranes. The negatively-charged anions (e.g., Cl-) are attracted to the anode (+) and repelled by the cathode (-). The anions pass through the anion-selective membrane and into the adjacent concentrate stream where they are blocked by the cation-selective membrane on the far side of the chamber, and are thus trapped and carried away by the carrier water in the concentrate stream. The positively-charged cations (e.g., Na+) in the purifying stream are attracted to the cathode (-) and repelled by the anode (+). The cations pass through the cation-selective membrane and into the adjacent concentrate stream where they are blocked by the anion-selective membrane and are carried away.

In the concentrate stream, electrical neutrality is maintained. Transported ions from the two directions neutralize one another's charge. The current draw from the power supply is proportional to the number of ions moved. Both the "split" water (H+ and OH-) and the intended ions are transported and add to the current demand.

Theory of Operation

In an EDI device, the space within the ion depleting compartments (and in some cases in the ion concentrating compartments) is filled with electrically active media such as ion exchange resin. The ion-exchange resin enhances the transport of ions and can also participate as a substrate for electrochemical reactions, such as splitting of water into hydrogen (H+) and hydroxyl (OH-) ions. Different media configurations are possible, such as intimately mixed anion and cation exchange resins (mixed bed or MB) or separate sections of ion-exchange resin, each section substantially comprised of resins of the same polarity: e.g., either anion or cation resin (layered bed or LB and single bed or SB).

The electrodeionization process uses a combination of ion-selective membranes and ion-exchange resins sandwiched between two electrodes (anode (+) and cathode (-)) under a DC voltage potential to remove ions from RO-pretreated water.

Ion-selective membranes operate using the same principle and materials as ion- exchange resins, and they are used to transport specific ions away from their counter ions. Anion-selective membranes are permeable to anions but not to cations; cation-selective membranes are permeable to cations but not to anions. The membranes are not water-permeable.

By spacing alternating layers of anion-selective and cation-selective membranes within a plate-and-frame module, a "stack" of parallel purifying and concentrating compartments are created. The ion-selective membranes are fixed to an inert polymer frame, which is filled with mixed ion-exchange resins to form the purifying chambers. The screens between the purifying chambers form the concentrating chambers.

This basic repeating element of the EDI, called a "cell-pair", is illustrated in Figure 1. The "stack" of cell-pairs is positioned between the two electrodes, which supply the DC potential to the module. Under the influence of the applied DC voltage potential, ions are transported across the membranes from the purifying chambers into the concentrating chambers. Thus, as water moves through the purifying chambers, it becomes free of ions. This stream is the pure water product stream.

Most commercial EDI devices comprise alternating cation- and anion-permeable membranes with spaces in between configured to create liquid flow compartments with inlets and outlets. The compartments bound by a positively charged anion exchange membrane (AEM) facing the positively charged anode (+) and a negatively charged cation exchange membrane (CEM) facing the negatively charged cathode (-) are diluting or purifying compartments. The compartments bound by an anion membrane facing the cathode and a cation membrane facing the anode are concentrating compartments. To facilitate ion transfer in low ionic strength solutions, the dilute compartments, and sometimes the concentrate compartments, are filled with ion exchange resins. A transverse DC electrical field is applied by an external power source using electrodes at the bounds of the compartments such that ions in the liquid are attracted to their respective counter electrodes. The result is that the diluting compartments are depleted of ions and the concentrating compartments are concentrated with ions. Figure 1 is a representation of the process showing two diluting compartments and one concentrating compartment.

FIGURE 1.

NOTES:CEM = Cation exchange membrane.AEM = Anion exchange membrane.The product stream may also be referred to as the dilute stream.The reject stream may also be referred to as the concentrate stream.To facilitate ion transfer in low ionic strength solutions, the dilute compartments, and sometimes the concentrate compartments, are filled with ion exchange resins. A transverse DC electrical field is applied by an external power source using electrodes at the bounds of the compartments such that ions in the liquid are attracted to their respective counter electrodes. The result is that the diluting compartments are depleted of ions and the concentrating compartments are concentrated with ions. The figure above is a representation of the process showing two diluting compartments and one concentrating compartment.

Mechanism of Ion Removal

There are two distinct operating regimes for EDI devices: enhanced transfer and electroregeneration (Ganzi, 1988). In the enhanced transfer regime, the resins within the device remain in the salt forms. In low conductivity solutions the ion exchange resin is orders of magnitude more conductive than the solution, and acts as a medium for transport of ions across the compartments to the surface of the ion exchange membranes. This mode of ion removal is only applicable in devices that allow simultaneous removal of both anions and cations, in order to maintain electroneutrality.

The second operating regime for EDI devices is known as the electroregeneration regime. This regime is characterized by the continuous regeneration of resins by hydrogen and hydroxide ions from the electrically-induced dissociation of water. This dissociation preferentially occurs at bipolar interfaces in the ion-depleting compartment where localized conditions of low solute concentrations are most likely to occur (Simons). The two primary types of bipolar interfaces in EDI devices are resin/resin and resin/membrane. The optimum location for water splitting depends on the configuration of the resin filler. For mixed-bed devices water splitting at both types of interface can result in effective resin regeneration, while in layered bed devices water is dissociated primarily at the resin/membrane interface (Ganzi et. al., 1997).

Regenerating the resins to their H+ and OH- forms allows EDI devices to remove weakly ionized compounds such as carbonic and silicic acids, and to remove weakly ionized organic compounds. This mode of ion removal occurs in all EDI devices that produce ultrapure water. Figure 13-1 is a representation of the process showing two diluting compartments, which illustrates the transport of ions and electrochemical regeneration of ion exchange resins in one type of EDI cell.

To construct a large scale EDI device, many of these cells are assembled together and fed in parallel as shown in Figure 2.

Figure 2

The animation below illustrates the removal of ions and the splitting of water molecules:

Mechanism of Ion Removal

There are two distinct operating regimes for EDI devices: enhanced transfer and electroregeneration (Ganzi, 1988). In the enhanced transfer regime, the resins within the device remain in the salt forms. In low conductivity solutions the ion exchange resin is orders of magnitude more conductive than the solution, and acts as a medium for transport of ions across the compartments to the surface of the ion exchange membranes. This mode of ion removal is only applicable in devices that allow simultaneous removal of both anions and cations, in order to maintain electroneutrality.

The second operating regime for EDI devices is known as the electroregeneration regime. This regime is characterized by the continuous regeneration of resins by hydrogen and hydroxide ions from the electrically-induced dissociation of water. This dissociation preferentially occurs at bipolar interfaces in the ion-depleting compartment where localized conditions of low solute concentrations are most likely to occur (Simons). The two primary types of bipolar interfaces in EDI devices are resin/resin and resin/membrane. The optimum location for water splitting depends on the configuration of the resin filler. For mixed-bed devices water splitting at both types of interface can result in effective resin regeneration, while in layered bed devices water is dissociated primarily at the resin/membrane interface (Ganzi et. al., 1997).

Regenerating the resins to their H+ and OH- forms allows EDI devices to remove weakly ionized compounds such as carbonic and silicic acids, and to remove weakly ionized organic compounds. This mode of ion removal occurs in all EDI devices that produce ultrapure water. Figure 13-1 is a representation of the process showing two diluting compartments, which illustrates the transport of ions and electrochemical regeneration of ion exchange resins in one type of EDI cell.

To construct a large scale EDI device, many of these cells are assembled together and fed in parallel as shown in Figure 2.

Figure 2

The animation below illustrates the removal of ions and the splitting of water molecules:

There are various types of EDI devices. Due to the nature of the process however, manufacturers to date have taken only two design approaches. These are plate and frame and spiral wound. The plate and frame design was the first to emerge and is similar to a plate and frame filter press or heat exchanger in construction. Alternating diluting and concentrating cells are stacked between the electrodes and sandwiched together with some type of closing mechanism. Increasing the number of cell pairs (one dilute and one concentrate cell) increases the capacity of the unit. The main advantage of this type of construction is the ease of assembly.

In contrast, spiral wound EDI is a bit more complicated and hence more difficult to assemble. We will not attempt to explain the details of spiral EDI construction, however the basic principles of deionization are similar to plate and frame configuration.

Currently, the majority of EDI devices on the market use plate and frame construction. Plate and frame devices can be broken down into two major subsets. These are thin cell and thick cell, designated as such based on the thickness of the diluting compartments. For the purposes of this discussion, thin cell will encompass devices with a diluting cell thickness of 2-3 mm and thick cell will encompass 8-10 mm.

There are various types of EDI devices. Due to the nature of the process however, manufacturers to date have taken only two design approaches. These are plate and frame and spiral wound. The plate and frame design was the first to emerge and is similar to a plate and frame filter press or heat exchanger in construction. Alternating diluting and concentrating cells are stacked between the electrodes and sandwiched together with some type of closing mechanism. Increasing the number of cell pairs (one dilute and one concentrate cell) increases the capacity of the unit. The main advantage of this type of construction is the ease of assembly.

In contrast, spiral wound EDI is a bit more complicated and hence more difficult to assemble. We will not attempt to explain the details of spiral EDI construction, however the basic principles of deionization are similar to plate and frame configuration.

Currently, the majority of EDI devices on the market use plate and frame construction. Plate and frame devices can be broken down into two major subsets. These are thin cell and thick cell, designated as such based on the thickness of the diluting compartments. For the purposes of this discussion, thin cell will encompass devices with a diluting cell thickness of 2-3 mm and thick cell will encompass 8-10 mm.

Intermediate Index

The Intermediate section is comprised of the following EDI topics:

EDI construction

EDI Module Construction

Electrode Reactions and Material Selection

Electroactive Media used in EDI Devices

Ion Exchange Resin Selection

Ion Exchange Membrane Selection

Mixed bed Resin Filler (EDI-MB) - Intermembrane Spacing

Mixed bed Resin Filler (EDI-MB) - Resin Packing

Layered Bed Resin Filler (EDI-LB)

Separate Bed Resin Filler (EDI-SB)

Thin Cell and Thick Cell EDI

Thin Cell EDI

Thick Cell EDI

DC Current and Voltage

Faraday's Law

Current Efficiency and E-factor

Ohm's Law and Module Resistance

Removal of Ionic Species

Series removal of species

Carbon Dioxide Removal

Silica Removal

Water splitting and Module Resistance

Gas Generation in the Electrode Compartment

CEDI Construction

This description of module construction will first discuss the overall device and then the individual cell. Commercially available devices are produced in two main configurations: plate-and-frame or spiral wound. The plate-type devices are similar in concept to a plate-and-frame heat exchanger, with multiple fluid compartments sandwiched between a set of endplates (and electrodes) that are held in compression by bolts or threaded rods. The compartments alternate between diluting and concentrating, and are hydraulically in parallel but electrically in series. An exploded view of a typical plate-and-frame EDI device is shown in Figure 1.

Figure 1Plate-and-frame EDI device

The spiral EDI devices are analogous to a spiral wound membrane element, but with the membrane, resins, and spacers wound spirally around a center electrode rather than a permeate tube. Spiral wound devices must be installed inside a pressure vessel, while plate-and-frame devices incorporate some means of sealing on the individual fluid compartments, essentially making each a pressure vessel. Spiral wound devices are somewhat more difficult to assemble than plate-and-frame units. A typical spiral device is shown in Figure 2.

Figure 2Spiral-wound EDI device

The cells themselves can be classified at either thin cell or thick cell. Thin-cell devices are those with a spacing of approximately 1.5-3.5 mm between the ion exchange membranes in the diluting compartments, while thick cell devices typically use intermembrane spacing of 8-10 mm. Both plate-and-frame and spiral wound configurations are suitable for either thin or thick cell construction. As will be shown below, thin-cell devices allow the use of intimately mixed anion and cation exchange resins in the product compartments, while thick cells work best with separate regions that contain primarily resins of the same polarity.

Flow Spacers

All commercial EDI devices use ion exchange resin in the diluting compartments, and therefore require a component to contain the resin. This 'resin spacer' consists of an inlet port, an inlet distributor, the resin compartment, an outlet distributor, and an outlet port. It is necessary to provide a means of sealing the ion exchange membrane against the spacer to form the sides of the resin compartment. Some designs will also include additional ports to allow slurrying the resin in and out of the cell. A typical dilute spacer for a plate-and-frame EDI device is shown in Figure 3.

Figure3 Dilute spacer from thick cell, layered bed EDI module

All EDI devices will also require flow compartments for the concentrate and electrode streams as well. The two types most commonly used are either a flow-through screen or a resin compartment. The flow-through screen is similar to a sheet-flow electrodialysis spacer. It generally consists of a woven plastic mesh screen (also like an RO feed spacer) that incorporates some sort of sealing mechanism, such as a rubber gasket interpenetrated in the perimeter of the screen.

The use of screen-type concentrate spacers is quite common in EDI devices, as they are fairly inexpensive and relatively easy to fabricate. Their major disadvantage is that they are not conductive. Since the makeup water feeding the concentrate compartments is normally RO permeate (to avoid scaling and fouling), the concentrate stream is not very conductive, in spite of the ions transferring into the concentrate from the diluting compartments. For example, a EDI system fed RO permeate with a conductivity of 5 S/cm and operating at 90% water recovery would typically have a concentrate outlet conductivity of about 50 S/cm. This is low enough to limit the amount of current that can be passed through the module (see discussion of resistance, below). Several manufacturers recommend injection of salt into the concentrate to raise the conductivity to 300 S/cm or more.

An alternative to the use of screen-type spacers for the concentrate and electrode compartments is to use a resin-filled compartment similar to the ones used for the diluting compartments. By employing a conductive filler, the use of salt injection can be avoided. It has been found that injection of a salt solution into a resin-filled concentrate compartment has little effect on module resistance.

Use of the same size compartments for both diluting and concentrating compartments would typically require a concentrate recirculation pump to maintain adequate velocity in the concentrate while limiting the water sent to drain, to provide high water recovery. An alternative approach is to make the concentrate resin compartments thinner than the dilute resin compartments, to avoid the use of a concentrate recirculation pump.

Electrode Reactions and Material Selection

At the cathode, or negatively charged electrode, electrons are transferred from the external circuit to ions in the solution by the following reaction:

H2O+ e--> H2 + OH- Eq. 6 Therefore an electrode that is stable in the presence of base and hydrogen is required. The most common cathode material for EDI devices is stainless steel.

At the anode, or positively charged electrode, electrons are transferred from ions in solution to the external circuit by one or more of the following reactions:

H2O -> O2 + H+ + e-Eq. 7

Cl- -> Cl2 + e-Eq. 8

Commonly used anode materials include iridium-coated titanium and platinum-coated titanium.

Gases are evolved by the reactions at both the cathode and anode. These must be removed to prevent masking the surface of the electrode, which would result in a voltage drop and reduce the voltage applied to the cells. Removal of the gases is accomplished by maintaining a flow of water across the surface of the electrodes during operation. This requires the use of a flow compartment adjacent to the electrode. Such compartments could be either gasketed screen-type spacers or resin-filled compartments.

Since copper wire is commonly used to conduct electric current to the electrodes, the junction of the copper wire and the non-copper electrode may be subject to corrosion, particularly if it is damp. If possible, it is best to have a projection of the electrode material that passes through the end plate of the module to an external connection that can be kept clean and dry (Schweitzer).

CEDI Construction

This description of module construction will first discuss the overall device and then the individual cell. Commercially available devices are produced in two main configurations: plate-and-frame or spiral wound. The plate-type devices are similar in concept to a plate-and-frame heat exchanger, with multiple fluid compartments sandwiched between a set of endplates (and electrodes) that are held in compression by bolts or threaded rods. The compartments alternate between diluting and concentrating, and are hydraulically in parallel but electrically in series. An exploded view of a typical plate-and-frame EDI device is shown in Figure 1.

Figure 1Plate-and-frame EDI device

The spiral EDI devices are analogous to a spiral wound membrane element, but with the membrane, resins, and spacers wound spirally around a center electrode rather than a permeate tube. Spiral wound devices must be installed inside a pressure vessel, while plate-and-frame devices incorporate some means of sealing on the individual fluid compartments, essentially making each a pressure vessel. Spiral wound devices are somewhat more difficult to assemble than plate-and-frame units. A typical spiral device is shown in Figure 2.

Figure 2Spiral-wound EDI device

The cells themselves can be classified at either thin cell or thick cell. Thin-cell devices are those with a spacing of approximately 1.5-3.5 mm between the ion exchange membranes in the diluting compartments, while thick cell devices typically use intermembrane spacing of 8-10 mm. Both plate-and-frame and spiral wound configurations are suitable for either thin or thick cell construction. As will be shown below, thin-cell devices allow the use of intimately mixed anion and cation exchange resins in the product compartments, while thick cells work best with separate regions that contain primarily resins of the same polarity.

Flow Spacers

All commercial EDI devices use ion exchange resin in the diluting compartments, and therefore require a component to contain the resin. This 'resin spacer' consists of an inlet port, an inlet distributor, the resin compartment, an outlet distributor, and an outlet port. It is necessary to provide a means of sealing the ion exchange membrane against the spacer to form the sides of the resin compartment. Some designs will also include additional ports to allow slurrying the resin in and out of the cell. A typical dilute spacer for a plate-and-frame EDI device is shown in Figure 3.

Figure3 Dilute spacer from thick cell, layered bed EDI module

All EDI devices will also require flow compartments for the concentrate and electrode streams as well. The two types most commonly used are either a flow-through screen or a resin compartment. The flow-through screen is similar to a sheet-flow electrodialysis spacer. It generally consists of a woven plastic mesh screen (also like an RO feed spacer) that incorporates some sort of sealing mechanism, such as a rubber gasket interpenetrated in the perimeter of the screen.

The use of screen-type concentrate spacers is quite common in EDI devices, as they are fairly inexpensive and relatively easy to fabricate. Their major disadvantage is that they are not conductive. Since the makeup water feeding the concentrate compartments is normally RO permeate (to avoid scaling and fouling), the concentrate stream is not very conductive, in spite of the ions transferring into the concentrate from the diluting compartments. For example, a EDI system fed RO permeate with a conductivity of 5 S/cm and operating at 90% water recovery would typically have a concentrate outlet conductivity of about 50 S/cm. This is low enough to limit the amount of current that can be passed through the module (see discussion of resistance, below). Several manufacturers recommend injection of salt into the concentrate to raise the conductivity to 300 S/cm or more.

An alternative to the use of screen-type spacers for the concentrate and electrode compartments is to use a resin-filled compartment similar to the ones used for the diluting compartments. By employing a conductive filler, the use of salt injection can be avoided. It has been found that injection of a salt solution into a resin-filled concentrate compartment has little effect on module resistance.

Use of the same size compartments for both diluting and concentrating compartments would typically require a concentrate recirculation pump to maintain adequate velocity in the concentrate while limiting the water sent to drain, to provide high water recovery. An alternative approach is to make the concentrate resin compartments thinner than the dilute resin compartments, to avoid the use of a concentrate recirculation pump.

Electrode Reactions and Material Selection

At the cathode, or negatively charged electrode, electrons are transferred from the external circuit to ions in the solution by the following reaction:

H2O+ e--> H2 + OH- Eq. 6 Therefore an electrode that is stable in the presence of base and hydrogen is required. The most common cathode material for EDI devices is stainless steel.

At the anode, or positively charged electrode, electrons are transferred from ions in solution to the external circuit by one or more of the following reactions:

H2O -> O2 + H+ + e-Eq. 7

Cl- -> Cl2 + e-Eq. 8

Commonly used anode materials include iridium-coated titanium and platinum-coated titanium.

Gases are evolved by the reactions at both the cathode and anode. These must be removed to prevent masking the surface of the electrode, which would result in a voltage drop and reduce the voltage applied to the cells. Removal of the gases is accomplished by maintaining a flow of water across the surface of the electrodes during operation. This requires the use of a flow compartment adjacent to the electrode. Such compartments could be either gasketed screen-type spacers or resin-filled compartments.

Since copper wire is commonly used to conduct electric current to the electrodes, the junction of the copper wire and the non-copper electrode may be subject to corrosion, particularly if it is damp. If possible, it is best to have a projection of the electrode material that passes through the end plate of the module to an external connection that can be kept clean and dry (Schweitzer).

Electroactive Media used in CEDI Devices

Ion Exchange Resin Selection

Ion exchange resins function much differently in EDI devices than in a conventional demineralizer, or even than in a collection/discharge type EDI device. In EDI, the ability of the resin filler to rapidly transport ions to the surface of the ion exchange membranes is much more important than the ion exchange capacity of the resin. The resins are therefore not optimized for capacity, but for other properties that influence transport, such as water retention and selectivity.

Membrane/resin combinations must also be carefully chosen to selectively catalyze the electrochemical splitting of water at various locations within the EDI device, as mentioned previously. Considerable research has gone into optimization of resin fillers for EDI devices, mostly by the manufacturers of the EDI devices rather than the manufacturers of the ion exchange resins.

Ion Exchange Membrane Selection

Ion exchange membranes are different from the many types of filtration membranes in that they are essentially impermeable to water. They combine the ability to act as a separation wall between two solutions (the diluting and concentrating streams) with the chemical and electrochemical properties of ion exchange resin beads. Ion exchange membranes are selectively permeable, as they will allow the passage of counter ions while excluding co-ions. When placed in a water solution and an electric field, a cation membrane will permit the passage of cations only, while an anion membrane will allow the passage of anions only. An in-depth discussion of the theory and properties of permselective membranes is available elsewhere.

There are two main types of commercially available ion exchange membranes, heterogeneous and homogeneous. Homogeneous membranes consist of thin films of continuous ion exchange material, typically on a fabric support. These are essentially equivalent to an ion exchange resin bead, only in the form of a thin sheet. Heterogeneous membranes consist of small ion exchanger particles embedded in an inert binder, with or without any support.

Some of the more important properties of ion exchange membranes used in EDI devices include the following:

Low water permeability

Low electrical resistance

High permselectivity

High strength

Resistance to contraction or expansion

Resistance to high and low pH

Ion exchange membranes that were developed for electrodialysis may not have sufficient mechanical strength and handling properties for use in assembly of EDI devices, so most manufacturers have developed special ion exchange membranes that are optimized for their EDI devices. Extruded heterogeneous membranes based on a polyolefin binder have become very popular for this application. They are relatively low in cost, offer flexibility in formulation, and have been shown to be fouling resistant.

Mixed bed Resin Filler (EDI-MB) - Intermembrane Spacing

The first commercial EDI devices used mixed-bed ion exchange resin as a conductive media in the diluting compartments. For devices using a mixed-bed resin filler, one of the most important design constraints is the distance between the ion exchange membranes. In order for the resin to transport an ion to the membrane, there must be a continuous path of the appropriate type of ion exchange resin, i.e. cation resin for transfer of cations and anion resin for transfer of anions. For simple cubic packing and equal quantities of equal diameter anion and cation beads, the probability of a direct conductive path can be related to the number of resin beads between the membranes by Equation 1.

Eq. 1

This shows that the probability of a direct conductive path decreases as the intermembrane spacing increases. The effect of intermembrane spacing on salt removal in a EDI-MB device has also been demonstrated experimentally, as shown in Table 1.

Cell Thickness, mmSalt Removal, %Feed, S/cmProduct, S/cmVelocity, cm/sec1.0

99.8

600

1.2

0.86

2.3

99.9

600

0.6

0.86

4.7

94.3

600

34

0.86

7.2

71.7

600

170

0.86

Table 1Relationship between cell thickness and performance for a EDI-MB device

Mixed bed Resin Filler (EDI-MB) - Resin Packing

It has also been shown that the performance of a EDI-MB device can be improved significantly by the use of uniform particle size ion exchange resins instead of conventional resins, which have a Gaussian distribution of bead sizes. The uniform beads allow a higher packing density, approaching a hexagonal close-packed structure. The effect of packing density on salt removal is illustrated by the data in Table 2.

Feed uS/cmProduct, MegOhm-cm non-uniform beadsProduct, MegOhm-cm uniform beads145

0.4

0.7

87

0.8

1.5

65

1.5

4.2

41

3.4

10.5

Table 2Resin particle size distribution and performance for a EDI-MB deviceLayered Bed Resin Filler (EDI-LB)

In the late 1980s and early 1990s there was considerable activity in the development of layered bed (EDI-LB) devices. In this configuration the media comprise separate, sometimes alternating layers (or in one variation, clusters) of ion-exchange resin, each layer containing mainly one type of resin: e.g., either anion or cation resin. Liquid to be deionized flows sequentially through the layers of resins.

For EDI-LB devices there is essentially no "enhanced transfer" regime and less limitation on the intermembrane spacing. This is because transfer of only one type (polarity) ion is enhanced at any given time. In order to maintain electroneutrality, the ion that is transferred out is replaced by a co-ion resulting from splitting of water. This is illustrated in Figure 5. One of the main design constraints is the choice of ion exchange resin, which must catalyze the water splitting reaction at the resin/membrane interface. Resin selection must also ensure that the electrical resistance of the layers is similar, so that the DC current is fairly evenly distributed through the cell instead of preferentially passing through a single type of layer. It is likely that the use of uniform particle size resins will offer some benefit to the performance of thick-cell layered-bed devices, but that the difference will not be as dramatic as it is for a thin-cell mixed-bed.

One of the main advantages to the use of thicker cells is that it greatly reduces the amount of ion exchange membrane used to construct the device, which significantly reduces the assembly cost (both materials and labor). The tradeoff is that the performance for salt removal is lower than for thin cell devices, due to the higher flow per unit membrane area and greater distance that ions need to travel across the cell to reach the ion exchange membrane. The EDI-LB module performance is more sensitive to increases in feed water concentration and to decreases in feed water temperature. However, this is less important now than when EDI was first commercialized, due to improvements in reverse osmosis and gas transfer membranes that have reduced the typical ionic load on the EDI device. The performance of thick-cell EDI devices is sufficient for their use in most ultrapure water applications, given proper system design.

The other significant advantage of thick-cell devices is that the thicker resin chambers are considerably stronger than thin spacers. They also offer more flexibility in the design of the intercompartment sealing, such as the use of grooves and O-ring seals. This allows construction of modules without external leaks and with higher pressure rating. The only commercial EDI devices that are capable of operating continuously at 7 bar (100 psig) are thick-cell type. Even the spiral-wound devices in a pressure vessel are limited to 4 bar (60 psig) or less.

Separate Bed Resin Filler (EDI-SB)

Another electrodeionization device uses completely separate compartments for the cation and anion resins, and is somewhat analogous to a two-bed demineralizer. The cation exchange resin is placed in a compartment between a cation membrane and the anode, with the resin in direct contact with the electrode. The anion exchange resin is between an anion membrane and the cathode. The two ion exchange membranes create a concentrate compartment at the center of the cell. This configuration is shown in Figure 5.

Figure 5 Removal mechanism in thick-cell, separate-bed EDI cell

Instead of splitting water at a resin/membrane or resin/resin interface, this process obtains the hydrogen (H+) or hydroxyl (OH-) ions needed to regenerate the resin from the electrode reactions; hydrogen ions being generated at the anode and hydroxyl ions at the cathode

Since the resins are in the electrode compartments, the O2, H2, and Cl2 gas that is created remains in the product water, which may require an additional gas removal process step. It is possible that the electrode reaction could produce enough chlorine to reduce the life of the ion exchange resin, depending upon the amount of chloride in the feed water.

It has been shown that the salt removal by EDI-SB device with 10 mm intermembrane spacing, is not nearly as good as for a EDI-MB device with 2.5 mm spacing. But the main disadvantage of the EDI-SB device is that it requires a set of electrodes for each cell. Since the electrodes are by far the most costly component of a EDI device, this approach is only cost effective for low flow rate applications where a single cell is sufficient. There have been some attempts to produce a multi-cell device using bipolar ion exchange membranes, but these have not been commercialized due to the short life of the bipolar membranes

Thick and Thin Cell CEDI

Thin Cell EDIThe first commercial EDI devices were thin cell with mixed bed ion exchange resin in the diluting cells. Although they have been modified over the years to improve performance, the basic principles have remained constant and the technology has proven to be effective and reliable.

Thin cell, mixed bed electrodeionization devices require a much greater area of ion exchange membrane per unit volume of water processed, and are therefore not as cost effective as thick cell devices. The following discussion will attempt to explain the subtle difference in removal mechanism between the two.

In thin cell, mixed bed EDI, two distinct zones are created inside the diluting compartments. Strongly ionized substances are removed first and then weakly ionized substances are removed as the water continues down through the flow path. We refer to these zones as enhanced transport and electro regeneration respectively.

In the production of ultrapure water, the feed to a EDI device is pretreated with reverse osmosis. This water contains low amounts of dissolved, ionized solids and some weakly ionized substances such as carbon dioxide and silica. Because of the low load, the device is able to remove most of the strongly ionized substances in the enhanced transport zone. Here, the ion exchange resin simply acts as a conductor to speed the passage of ions from the dilute compartment through the respective membrane, and into the concentrating compartment. This is because the ion exchange resin is several orders of magnitude more conductive than the water. This is shown in the top portion of the resin bed in Figure 3.

After most of the strongly ionized substances have been removed at the top of the cell, conductance of the diluting cell is maintained by the ion exchange resins. At locations where the minimum thermodynamic overvoltage for water splitting is applied, the concentrations of hydrogen (H+) and hydroxyl (OH-) ions are increased. This is shown in the figure in the electro-regeneration zone. The water decomposition reaction is catalyzed by conditions at the resin/resin or membrane/resin interfaces of dissimilar polarities. Here, liberated H+ and OH- ions convert the resins into the regenerated state where weakly ionized substances can react, become ionized and be moved into the concentrating stream.

Thick Cell EDIThick cell EDI devices arrived commercially in 1996, and several different types are now available. Besides dilute cell thickness, another thing that differentiates these devices from thin cell EDI is the fact that the dilute cells can use separate resins or a combination of separate resins and mixed bed resins.

Thick diluting channels can be a detriment to performance using mixed bed resin filler due to a lower chance of obtaining a continuous path between membranes. Water splitting can still occur at dissimilar resin/resin and resin/membrane interfaces but much of the split H+ and OH- ions will recombine when they encounter the counter ions traveling in the opposite direction. The basis for ion removal is different in devices that use separate resins in the dilute cells. Because a single type of resin is present at any given point between membranes, the transfer of co-ions is not possible. Therefore, water must decompose to provide H+ and OH- ions for transfer to take place while maintaining electrical neutrality. Therefore, current passage and water splitting are critical for both weak and strong ion removal.

Current and Voltage

Faraday's LawIn a continuous electrodeionization device the DC current is the driving force for the removal of ions, while the applied DC voltage is the means of obtaining the required current. Faraday's Law states that the electric charge required to liberate one gram-equivalent of a substance by electrolysis is 96,487 coulombs (a coulomb is the amount of electric charge that crosses a surface in one second when a steady current of one ampere is flowing across the surface). In both electrodialysis and electrodeionization, Faraday's Law is used to relate the transfer of salts through the membranes and the amount of current flowing through the membranes. A common form of this relationship is given in Equation 2:

Eq. 2This shows that the amount of DC current required is directly proportional to the flow rate through the diluting compartments and the amount of ionic equivalents to be removed, and inversely proportional to the current efficiency.

Current Efficiency and E-factorCurrent efficiency can be defined as the ratio of the theoretical minimum current predicted by Faraday's law (at 100% efficiency) to the actual current applied to the electrodes of the device, as shown in Equation 3:

Eq. 3In a EDI device, current that does not cause the transfer of salt will cause water (HOH) to split into hydrogen (H+) and hydroxyl (OH-) ions, allowing electrochemical regeneration of the ion exchange resins within the device. For example in a module that is operating at 25% current efficiency and drawing 4 amps of DC current, 1 amp is causing the transfer of salt and 3 amps are causing water splitting that is unrelated to ion transfer. Cross-leakage and back diffusion can also cause some current loss, but these are normally small compared to the water splitting.

In order to produce high purity water (over 1 megohm-cm resistivity) with a EDI system, it is generally necessary to feed the system with low TDS water such as RO permeate (normally less than 0.0005 equivalents/liter) and to operate at a current efficiency of less than 35%. For optimal removal of weakly ionized solutes such as silica and boron, current efficiencies as low as 5% are sometimes employed.

Some authors prefer to use the term E-factor. This is defined as the ratio of the applied current to the theoretical current, and is therefore the reciprocal of the current efficiency:

Eq. 4Ohm's Law and Module ResistanceOhm's law states that the direct current flowing in an electric circuit is directly proportional to the voltage applied, and inversely proportional to the resistance of the element:

Eq. 5Most manufacturers of EDI devices limit the applied voltage to 600 VDC, in order to avoid the need for the more expensive wiring construction that is required for higher voltages. Given such a voltage limitation, the electrical resistance of the module therefore controls how much current can be passed through the cells. Since the DC current determines how much water can be processed for a given product quality (or what the quality will be for a given flow rate), it is important to optimize the electrical resistance of the module.

The overall resistance of the EDI module can be affected by the following:

Resistance of the anion-selective membranes

Resistance of the cation-selective membranes

Resistance of the ion exchange resins

Resistance of the concentrate stream

Resistance of the anolyte

Resistance of the catholyte

Feed water temperature

Ionic composition of the feed water

In addition to proper selection of resin and membranes, there are several methods that reduce the electrical resistance of the cell and therefore allow greater passage of DC current. The first technique used to accomplish this was to increase the water recovery and therefore the amount of salt in the concentrate compartments. This is generally done by incorporating a feed-and-bleed arrangement, using a pump to recirculate the concentrate stream and ensure adequate flow distribution, while decreasing the flow rate of the bleed that is sent to drain.

An alternative method of reducing the cell resistance is to inject a conductive salt such as NaCl into the feed to the concentrate compartments using a dosing pump. There are several possible drawbacks to this method. Increasing the TDS may prevent reclaiming the concentrate stream for other uses, and may increase the possibility of salt bridging and stray DC currents. If the concentrate is used to feed the electrode compartment, this can also lead to generation of chlorine gas at the anode.

A third method is to incorporate resin filler into the concentrate (and in some cases, electrode) compartments, which eliminates the need for injection of a conductive salt. It has also been seen that the resin helps ions transfer away from the surface of the concentrate side of the ion exchange membrane. This reduces the ion concentration in the boundary layer, reducing the driving force for back-diffusion and improving salt removal.

Ion Removal

Ionic species are not all removed by the EDI process with equal efficiency. This fact impacts the quality and purity of the product water.

Easy ions are removed first.

The ions with the strongest charge, the smallest mass, and the highest adsorption to the resins are removed with the highest efficiency. These typically include: H+, OH-, Na+, Cl-, Ca+2, and SO4-2 (and similar ions).

In the first section of the EDI module, these ions are removed preferentially to other ions. The relative quantity of these ions affect the removal of the other ions. The pH approaches 7.0 in this section since the H+ and 0H- ions become balanced.

The first section of the EDI module is known as the "working bed".

Moderately ionized and polarizable ions are removed next (e.g., CO2).

Carbon Dioxide Removal

This graph shows the effect of CO2 in relatively pure deionized water. CO2 is the next most common EDI feedwater constituent. CO2 has complex chemistry depending on the local concentration of protons, and is considered moderately ionized:

CO2 + H2O => H2CO3 => H+ + HCO3- => 2H+ + CO3-2Eq. 8

Since the pH is forced to be near 7.0 in this section, most of the CO2 is forced into the bicarbonate (HCO3-) form. Bicarbonate is weakly adsorbed by the anion resin, so cannot compete with "easy" ions such as Cl- and SO4-2.

In the second section ("Polishing Section") of the EDI module, CO2 (in all of its forms) is removed preferentially to weaker ions. The amount of CO2 and HCO3- in the EDI feed strongly effects the final resistivity of the product water and the efficiency of silica and boron removal.

It is found that as long as CO2(in all forms) is less than 5 mg/l, high quality ultrapure water can be achieved. If the CO2 concentration is greater than 10 mg/l, it can interfere with the total removal of ions and strongly impacts the EDI product quality and the silica removal.

Weakly ionized species are removed last (e.g., dissolved silica and boron).

Silica Removal

Since species such as molecular silica are very weakly ionized, and difficult to adsorb on ion-exchange resin, they are the most difficult to remove using any deionization process.

If all of the "easy" ions are removed, and all of the CO2 is removed, the EDI module can focus its force on removing these weakly ionized species. The residence time available in this third section of the module is important. The longer the residence time available in the module, the higher the removal efficiency. A long third-section residence time can be achieved by minimizing the conductivity of the RO product (the quantity of "easy" ions to be removed) and minimizing the quantity of CO2 in the RO product.

The second and third sections of the EDI module are known as the "polishing bed".

Silica is one of the more important minerals to remove from water for power generation and semiconductor applications. It also one of the most difficult to remove.

Silica chemistry is complex. On the most basic level, silica comes in "colloidal" and "reactive" forms. silica level in feedwater depends upon the geology of the region and whether the source is surface water or well water. Silica in raw water will range from less than 2 ppm to over 100 ppm.

Physical processes such as reverse osmosis (RO) will remove colloidal silica. EDI will only remove reactive silica.

Removal of reactive silica depends upon its charge. silica has little, if any, charge at neutral pH near 7 since the pK1 of silicic acid is 9.8 This makes it difficult to exchange with ion exchange resin, or to remove with RO or EDI. Raising the pH to above 9.8 helps with the driving force.

Silica scaling is also an issue. The solubility of silica at pH 6-8 is only 120 ppm at 25oC. this means that 30 ppm of silica in an RO feedstream with 75% recovery will begin to scale. There are two prevention techniques for silica scaling. One is the use of an antiscalant in the RO process, which will delay the precipitation of solid silica. The other is raising the pH, which increases the solubility limit of silica. At pH 10 silica is soluble up to 310 ppm. Of course, high pH will cause hardness scaling if the feed is not adequately softened.

It is important to maintain the silica content of the EDI feed stream to under 0.5 ppm as silica in order to:

Avoid scaling in the EDI concentrate stream

Minimize silica levels in the product water

Typical commercial RO modules will reject silica at only twice the passage of chloride ion. Most spiral RO module manufacturers claim 99.7% rejection for individual high quality elements. 99.0% - 99.5% is a reasonable silica rejection for a well designed RO system.

With a 20 ppm silica feed and 75% recovery, a 99.0% rejection element will maintain the RO product at 0.5 ppm silica.

For higher levels of silica in the feed, the RO system should be designed with higher quality RO elements and/or lower recovery. Using 99.7% rejection elements and 65% recovery, the RO feed can approach 90 ppm and still maintain 0.5 ppm silica in the effluent.

Water splitting and Module Resistance

As described above, water splitting is critical for the removal of weakly ionized species like silica, carbon dioxide, and boron. It is also critical for the removal of strong ions in thick cell, separate bed EDI. The amount of water splitting can be quantified using Faraday's law to compare the theoretical amount of current needed to transfer a given amount of ionized species out of one electrochemical cell to the actual applied current through that cell. Faraday's law states that

Eq. 9

Where:I = theoretical current, ampsEq = number of equivalents transferred per cellt = time, secondsF = the Faraday constant = 96,500 coulombs/equivalent

If we define the current efficiency as the theoretical current given by Faraday's law, divided by the actual applied current, we obtain the following equation:

Eq. 10

Where:h = current efficiency, %Ia = applied current, amps

To make this equation easier to use, we can substitute flow rate and feed concentration for the equivalents removed per unit time per cell so that:

Eq. 11

Where:TDS = total dissolved solids, mg/l as CaCO3Q/n = product flow rate per cell, l/min/cell3.22 = conversion factor

In Eq. 11 we show TDS as the total feed load to the cell but the total feed load also includes weakly ionized species, such as carbon dioxide and silica, which are also removed. To take these into account, Ionpure uses a term called Feed Conductivity Equivalent (FCE) and E-Cell uses a term called Total Exchangeable Anions (TEA). To calculate true current efficiency, TEA can be substituted directly for TDS in Eq. 11. However, to calculate current efficiency using FCE the conversion factor changes as shown:

Eq. 12

We can make a couple of important conclusions from Eq. 10 and 11. First, thick cell EDI, which operates at higher flow per cell than thin cell EDI, will require higher current to maintain the same current efficiency. Second, current efficiency can be lowered, and hence water splitting increased, by increasing the applied current at a constant flow rate and feed concentration. Current efficiency is especially important with regard to weak ion removal. It is not uncommon to operate below 10% current efficiency for improved removal of silica, boron, etc.

E-Cell (General Electric corp.) uses a term, which is a variation of current efficiency. They actually refer to it as the E-factor, which is the inverse of the fractional current efficiency. Therefore, increasing the E-factor is analogous to lowering the current efficiency.

In any case, increasing the current requires either increasing the voltage or lowering the module electrical resistance. Increasing the voltage imparts several drawbacks including greater power consumption and increased safety risk. So reducing the module resistance becomes paramount in EDI module design.

Because all EDI devices use ion exchange resin in the diluting cells, the concentrating stream is really the limiting resistance in a module. For modules that don't have resin in the concentrate, the only way to increase the conductivity is to increase the conductivity of the water. This is done by increasing the recovery or by direct injection of a salt, such as sodium chloride, to the concentrate. In addition, many manufacturers use recirculation of the concentrate stream along with high recovery or salt injection to provide a less variable concentration along the length of the module.

There are several drawbacks to concentrate recirculation. Concentrate recirculation requires the use of a pump and additional ancillary equipment for control such as motor starters and throttling valves. This adds complexity to the system design and increases overall cost. In systems with fluctuating operating conditions, operation or adjustment of the pump can make the process significantly more labor intensive. Also, the power required to operate the pump can be a large portion of the total power consumption of the system. A typical industrial EDI system with concentrate recirculation would consume about 1 to 2 kilowatt-hours per thousand gallons of product water (kWh/kgal), where about 0.5 kWh/kgal is for the recirculation pump alone.

Salt injection also has several drawbacks. Increasing the salt concentration in the EDI concentrate stream:

Limits the ability to recover that stream

Increases the concentration gradient between the diluting and concentrating cells facilitating co-ion back diffusion if the membranes are not ideally permselective

Increases the possibility of salt bridging and electrical shorting, and

Leads to the formation of chlorine at the anode when fed with recirculated concentrate water.

Gas Generation in CEDI Cells

Chlorine Gas Production

Normally the reaction at the anode will result in the formation of oxygen, but with sufficient voltage and with chloride ion present, chlorine gas can be formed. In particular this is an issue with electrodeionization systems which inject sodium chloride into the concentrate and electrode streams in order to reduce module resistance and improve performance. Because many EDI devices feed the reject water to the electrode chambers, concentrate salt injection increases the chloride concentration at the electrode. As shown below in Eq. 13, chloride at the anode is converted to chlorine, which is a strong oxidant and a toxic gas that can damage ion exchange membranes and create a safety hazard for those near the device.

2Cl- Cl2 + 2e-

Eq. 13

To investigate the extent of chlorine formation, an E-Cell MK-1 module was operated at 600 volts DC and 3.0 amps. The temperature of the feed water was varied between 15 and 25oC such that the resistance of the module changed proportionally. Then the concentration of sodium chloride was varied at the inlet of the concentrate cells to overcome this change in resistance and maintain 3.0 amps. Free chlorine was measured in the electrode product water. As shown in Figure 7, a significant amount of free chlorine was detected. There was also a large off-gassing of chlorine evidenced by the odor near the device. Concentrations in the range shown in Figure 7 can cause significant damage to ion exchange membranes or resins in contact with the electrode stream. In particular this is an issue with electrodeionization systems which inject sodium chloride into the concentrate and electrode streams in order to reduce module resistance and improve performance.

Figure 7. Chlorine Generation at the Electrodes in an E-Cell Module Other systems do not require brine injection because they use an ion-exchange resin filler in the concentrate compartments in order to minimize the electrical resistance of the module. This in itself limits the possibility of chlorine generation at the anode by reducing the chloride concentration. Because the resin is orders of magnitude more conductive than typical RO product water, it removes the conductivity of the water as a contributing factor in the overall resistance. This alleviates the need for salt injection or recirculation. In addition, chlorine production is virtually eliminated. An additional advantage of the this design is that with elimination of salt injection and chlorine production, the EDI concentrate will generally be better quality than the raw water and can often be recycled.

To show the effect of concentrate resin filling, two modules were tested with and without salt injection under similar conditions. Shown in Figure 8 is the resistance per cell pair for a module with concentrate resin filling compared to a module without concentrate resin filling, as a function of temperature. The electrical resistance of the module with a resin filled concentrate was lower by two orders of magnitude.

Figure 8. The Effect of Concentrate Resin Filling on Module Resistance

Hydrogen Gas Production

People often express concern about the hydrogen gas produced at the cathode, since it is known that under certain conditions hydrogen can be explosive. However, the amount of gas that is produced by an EDI module is so small that it does not present a safety hazard when the EDI system is installed in an area with normal ventilation. Codes require that buildings have multiple air changes every day, one air change being a turnover of air equivalent to the building's internal volume. The number of air changes varies depending upon the building's use and the local codes, but a widely accepted value is half an air change per hour.

To show how little risk is presented by the electrode gases, let's perform an example calculation assuming that an EDI module is installed in a 4m x 4m x 4m office which has the normal ventilation of half an air change per hour. This is equivalent to an air flow of 32 m3/h. Assume the EDI module is operating with a DC current of 10 amps, therefore producing hydrogen at the rate of 74.6 ml/min (equivalent to 0.0045 m3/h). If all the hydrogen gas leaves the water and enters the room air, then the resulting concentration of hydrogen in the air would be about 0.014% (v/v), or about 141 ppm. This is well below the explosive limit of a hydrogen/air mixture, which is 4.2% v/v at STP, and also well below the concentration at which asphyxiation would occur. Even at the maximum current output of an EDI DC power supply there is no safety risk as long as the ventilation meets typical building code requirements.

Advanced Index

The advanced section is comprised of the following EDI topics:

Pretreatment to EDI systems - EDI Feedwater Requirements Critical contaminants that adversely affect the EDI process include hardness (calcium, magnesium), organics (TOC), particulates and suspended solids (SDI), active metals (iron, manganese), oxidants (chlorine, ozone), and carbon dioxide (CO2). The pretreatment process designed for the RO/EDI system should remove these contaminants from the feed water as much as possible. Proper pretreatment design will greatly enhance EDI performance. Suggested water treatment strategies are listed throughout this section. The main areas that affect EDI operation and performance are:

Dechlorination

CO2 - Carbon Dioxide

Particulates, Metals and TOC

EDI Process Considerations

EDI system Safeguards

EDI Process Design

Feed Water Hardness

Design to prevent EDI scaling

Avoiding a common mistake

Water Recovery

Product Water Quality

Reclaim of EDI Reject Water

Boiler Makeup Water (Power Plant)

Other EDI "Watchouts" Feed Water

Other EDI "Watchouts" Hydraulic

Feed Water Conductivity Equivalent (FCE)

DC Volts and Amps

Definition of EDI Recovery

Calculating Reject Flow

Suggested Test Kits

Factors Affecting EDI Performance

Voltage

Current

Ionic Species

Temperature

Flow

Feed Conductivity

Effects of contaminants

Optimizing EDI Performance

Voltage Driving Force

Current density

Ionic Balance and pH

Troubleshooting EDI Systems

Some Very Important Points to Remember!

Requirements for EDI troubleshooting

Questions To Ask

EDI Troubleshooting: Data Required

EDI Troubleshooting: Low Product Quality

Mass Balance Calculation

More going in than coming out

EDI Troubleshooting: Resistance Increase

EDI Troubleshooting: Low Flow

Flowcharts

Oxidants

Scale

Organic Fouling

Biofilm

Excessive Feed Pressure

High Temperature

Particulates

EDI Cleaning and Sanitization

Purpose of cleaning

When to Clean?

Chemicals Used on EDI

For Any Cleaning/Sanitization

Typical Cleaning System

General Considerations

Multiple Cleanings

HCl Cleaning Procedure

Brine/Caustic Cleaning

Sanitization: Definitions

Caustic Cleaning/Sanitization

Sodium Percarbonate Cleaning/Sanitization

Multi-Agent Cleaning ("3 Step")

Peracetic Acid Sanitization

Hot Water Sanitization

Hot Water Sanitization "watch-outs"

Which Cleaning for High d/p?

Which Cleaning for High Module Resistance?

Action for Quality Decline with no Other Symptoms?

Shutdown/Preservation

Pretreatment - Dechlorination

The main cause of EDI failure is oxidation of the module, either by chlorine. Oxidation decrosslinking also causes the resins to disintegrate which results in an increased pressure drop across the module. For this very reason, dechlorination is extremely important to put the proper measures in place in the pretreatment to both the RO and EDI system. In addition to chlorine, chloramine or ozone attacks ion-exchange resins and membrane and cause decrosslinking, which results in reduced capacity. If Chloramines are present, further methods of removal may be required. Oxidation will increase apparent TOC, and the byproducts cause fouling of the anionic resin and membrane, reducing the ion-transfer kinetics.

The ideal concentration level for oxidants is zero. RO membranes are more resistant to chlorine that EDI modules so do not measure the decrease of RO rejection as an indication that there is no chlorine present!

Rule #1: There should be no detectable chlorine in the EDI feed water. Oxidative attack on a EDI module is not recoverable by cleaning!There are three principal means of achieving the above requirement:

Granular activated carbon

Injection of sodium bisulfite or sodium sulfite

UV dechlorination

INCLUDEPICTURE "http://www.cediuniversity.com/images/stories/CEDI-U/Lmi_pump.jpg" \* MERGEFORMATINET

INCLUDEPICTURE "http://www.cediuniversity.com/images/stories/CEDI-U/UV_sterilizer_03.jpg" \* MERGEFORMATINET Recommendation: Activated carbon dechlorination is preferred over chemical dechlorination because it is both more reliable and also less likely to cause biological fouling problems. UV dechlorination is a relatively new process, which may have difficulty achieving complete chlorine removal, especially on chlorinated feed water.

Feed water chlorine can be measured with a test kit. The smallest increment for this test kit is 0.02 mg/l.

A) Dechlorination by activated carbonIt is the responsibility of the system supplier to ensure that the activated carbon filter is sized properly for complete removal of the feed water chlorine, taking into account the concentration of chlorine in the raw feed water. Ionpure would usually design dechlorination carbon filters for a minimum empty bed contact time of 6 minutes.

B) Dechlorination by chemical injection (reducing agent)It is the responsibility of the system supplier to ensure that the chemical injection system is sized properly for complete removal of the feed water chlorine, and includes adequate instrumentation and control safeguard to ensure that the EDI system is not fed water containing chlorine should there be a change in the feed water conditions or an upset in the operation of the chemical injection system.

In systems that contain injection of multiple pretreatment chemicals (for example antiscalant and sodium sulfite or sodium bisulfite and sodium hydroxide) it is recommended that there be a separate injection point and static mixer for each chemical.

C) Ultraviolet DechlorinationIt is the responsibility of the system supplier to ensure that the UV dechlorination system is sized properly for complete removal of the feed water chlorine, taking into account the concentration of chlorine and chloramine in the raw feed water.

SUMMARY There must be no detectable chlorine in EDI feed water!

Chlorine damage is irreversible

Chlorine is a common cause of EDI failure

Take Cl2 readings at RO inlet and EDI inlet

RO permeate can contain more free chlorine than RO feed

Suggested Test Kit: Use Cl2 Hach Model CN-70 (#1454200)

Pretreatment - Hardness

Hardness

Hardness can