By Chris Gallagher
Electrodeionization (EDI) has matured and grown in popularity since its first commercial introduction over 18 years ago. The technology can replace mixed ion exchange in many applications using a chemical-free process to produce high-quality water; however, the water quality fed to the EDI system is generally reverse osmosis (RO) permeate. Because of the competitive nature of the business, better manufacturing and improved designs, the price of EDI has dropped over 60 percent in the years since its inception, making it a competitive option.
Life of a water system
With any purchase of equipment, life cycle, maintenance and cost are important factors in the decision-making process. For EDI modules there are no moving parts. EDI systems require power supplies, valves and piping. The simplicity of EDI systems has been greatly enhanced over the years. By maintaining the pretreatment of the water generation system, EDI can have an expected life cycle greater than five years. (Check with individual manufacturers for warranty terms.)
EDI acts as a very sensitive indicator of changes in feedwater and pretreatment. EDI is usually the last piece of process equipment in a water generation system, so if the temperature, TDS or flow changes, EDI performance will change.
Types of units and process principles
Many new EDI products have entered the market. The first commercial EDI was a plate and frame device that had thin purifying spacers of ~0.100 inches; these systems are still offered today. Thick purifying cell EDI entered the market in the mid ‘90s, offering a thickness of > 0.200 inches. Spiral EDI premiered at the same time, offering a different cell pair configuration outside of the plate and frame.
Regardless of the manufacturer, the same fundamental principles are common among all suppliers. The EDI device uses cell pairs: one cell (chamber) is for ion depletion, the purifying cell. Juxtaposed to the purifying cell is a concentrate cell. One purifying cell and one concentrating cell comprise a cell pair and there are many cells in a module. Each cell is separated by alternating anion and cation exchange membranes. The purifying cell is filled with ion exchange resin and some designs incorporate conductive material in the concentrate and electrode chambers. There are two electrodes (where voltage is applied): an anode that attracts anions and a cathode that attracts cations.
When DC voltage is applied, ions in the purifying compartment move to their respective nodes. The ions move out of the bulk solutions and through the ion exchange membranes, where they are captured in the concentrating compartment and leave the EDI module.
One of the unique principles of EDI is when enough ions are transferred, water splitting or polarization occurs. This is when H2O breaks down into H+ (acid component) and OH-(caustic component). The presence of H+ and OH- serves to ionize weakly ionized constituents such as CO2, silica and boron; then allows these weakly ionized constituents to be transferred from the bulk solution through their respective membranes and into the concentrating compartment.
Electrodialysis (ED) has been commercially available for over 50 years. ED uses the same principles as EDI; however, to overcome the concentration polarization effects that occur in dilute solutions, conductive material (i.e., ion exchange resin) was introduced into the purifying chamber.
In the ‘90s, products were introduced where conductive material was used in the concentrate cell and electrodes. This was to lower the overall electrical resistance of the EDI module and to use the current more efficiently.
One spiral EDI device is unique in that the concentrate flow runs tangentially, or spiral to the product or purifying compartment. The anode is on the outer layer and the cathode is in the center. Concentrate recirculation is recommended for this design. The product chamber runs parallel to the nodes.
Concentrate recirculation is used in some designs. The intent is to keep a good flow distribution in the compartment, support the purifying chamber and to increase water recovery. Some manufacturers have a targeted flow rate in the concentrate, similar to an RO reject recycle, hence requiring the recycle. Again, these are unique designs created to make optimal efficiency of the current. The flow rates of the recycle depend on the manufacturer and will change with the EDI supplier.
One of the first things suppliers strove for was to create a standard for EDI design. Today, there are standards of 50 gpm, 15 gpm, 12.5 gpm, 10 gpm and less, based on flow per cell pair; more cell pairs, the higher the flow. Having multiple modules allows more flexibility to isolate problems, but may be more costly for additional power supplies, piping and instrumentation. Single stack design, or higher capacity designs with fewer stacks increase risk and the system may need to be fully shut down for maintenance.
Many recirculation designs also use brine injection. The brine injection is intended to lower the overall stack resistance and make better use of the applied DC voltage.
When designing an EDI system, one must look at the whole water generation system. Each EDI module has a pressure drop across the device that can range from 15 to 40 psi; each supplier also has specifications on back pressure, temperature, flow and TDS. The use of break tanks may be required when RO permeate is being used to feed other processes onsite. If the RO is feeding directly to the EDI, one must take the pressure drop across the device when sizing the high pressure RO pump.
In the System Design (1.) illustration, we see two different process flows for an EDI system. The first shows the reject from both the RO and EDI going to the drain. In this case, the EDI pressure drop and piping should be considered when sizing the high pressure RO pump. In the second design, an RO storage tank feeds the EDI. It is recommended that a one µm filter be used before any EDI that is not directly fed by the RO. The EDI is not a filter and cannot take any particulates. In many cases, the RO storage tank can inadvertently have particles in it as a result of normal use.
The System Design (1.) illustration represents an RO storage tank, but also a booster pump pre- and post- EDI. If the EDI storage tank is not local, this may be necessary.
The System Design (2.) illustration shows a recycle of the EDI reject. In most cases, the EDI reject can have a lower TDS than the RO feed water. However, special attention needs to be given to the ions that the RO does not reject and the EDI does. Silica, boron and CO2 are the three major constituents to look for in the EDI reject. If they are present, there could be a build up in the system.
Current efficiency governs all EDIs. The main factors are flow rate, total feed water equivalents and current. Most manufacturers recommend a nominal flow and consistency in the feed to produce consistent EDI water quality. The current does the work, so changes in current will directly affect the water quality. The major factors that affect all EDIs are: current, temperature, TDS and fouling. The higher flow-through lowers the residence time, thus decreasing the time the fluid is in the path of the current. The lower the amperage, the less work the module is performing.
Variables to be monitored
- Flow rates
Voltage and amperage should be monitored. This can detect an increase in overall stack resistance that can be indicative of fouling or of changes in the feed water. Many thicker cell designs can be more sensitive to changes in feed water compared to thinner cell designs. A decrease in flow and an increase in pressure can be indicative of changes in the feed source or fouling.
The importance of data collection is trending. Some upsets can be seasonal or the result of events occurring in the feed source. If the data is trended, it can predict the performance of the system and give you the information to make informed decisions.
Normalization corrects recorded data and assists with making educated comparisons. For example, temperature can greatly affect the performance of certain EDIs. A decrease of 10°C in temperature could increase the overall stack resistance by threefold, hence reducing the amperage and reducing the water quality; however, it does not mean the EDI stack may have fouled or that it needs cleaning. Temperature effects can be compensated by sizing the power supply correctly to assure there is enough voltage or by tempering the water.
Seasonal effects can include geography, naturally occurring events and feed water source variability (chlorine, chloramines, hardness, pH and temperature). Seasonal activities can also affect the water generation system. Depending on the geography, feed water TDS can change throughout the year. One of the variabilities commonly seen is the addition of ammonia with chlorine to form chloramines. Total chlorine (free +chloramines) can irreversibly damage today’s EDI modules. If a system was designed to remove chlorine and not chloramines, additional processes will be required. Reduced EDI performance is due to symptoms in pretreatment processes.
About the author
Chris Gallagher is the Founder and President of Applied Water Solutions. He has spent over 18 years managing, manufacturing and designing separation and filtration systems within a wide range of industries. In addition to his in-depth experience with conventional treatment, Gallagher has specialized in troubleshooting, upgrading, testing and maintaining all types of EDI systems. He installed some of the first commercial EDI systems and since then has contributed to the growth and development of the technology. He received a BS Degree in mechanical engineering from the University of Massachusetts. He earned a Masters Degree in management technology and operations from the McCallum Graduate School of Business, where he is a member of the Executive Board today. Gallagher is an inventor with a number of US patents relevant to EDI and has written for many industry publications.
About the company
Applied Water Solutions, Inc., a global, independent authorized RO-EDI expert, has over 18 years of experience supporting membrane technology. The company is dedicated exclusively to solving the pure water needs of different industries, with deep expertise in Electropurification (EP™), current Purewater Design Practices (cPDP), RO, ion exchange, WFI and cGMP environments for purified water systems. Visit the company’s website: www.AppliedWaterSolutions.com.
All illustrations in this article are courtesy of Applied Water Solutions, Inc.