Ion Exchange: Niche Markets for the Common Water Softener
By Peter Meyers
Summary: There are many uses a water conditioner can be put to beyond simply softening water. Some require more care and expertise because of potential health risks if not handled properly. Following is an overview on this concept.
The business of supplying resin and systems to the home softening market is very competitive. Any edge is most welcome. The residential water softener has the ability to remove many unusual contaminants in addition to common hardness ions. Almost any water softener can be outfitted with a variety of specialty resins that are selective for various contaminants. While the unit may not be functioning strictly as a softener, these “unusual” contaminants provide niche opportunities to apply the craft of water treatment, solve problems and make an honest buck or two.
For purposes of this discussion, we’ll limit ourselves to resins regenerated with salt—sodium chloride (NaCl) or potassium chloride (KCl)—and with contaminants found in drinking or other water supplies.
At the total dissolved solids (TDS) levels found in most potable water supplies, cation ion exchange resins prefer divalent ions to monovalent ions—referring to a molecules’ charge as represented by the loss or gain of electrons in its outer shells. Positively charged ions are called cations. Negatively charged ions are called anions. An ion may have entirely different properties than the element from which it was formed. Simply put, though, the ion seeks to balance itself by attracting or giving away elements of a related electron charge. The levels of attraction or subtraction are governed largely by the relationship between the charges of any two ionic substances. This selectivity is why a water softener works—divalent ions such as calcium (Ca+2) and magnesium (Mg+2) are readily adsorbed by cation resin in exchange for monovalent ions such as sodium (Na+) or potassium (K+). Water softeners don’t just remove common hardness ions such as calcium and magnesium, but also remove other bothersome divalent cations such as barium, manganese, iron, copper, strontium and radium. Provided the softener is efficiently regenerated when hardness breaks, all of the above mentioned cations are (almost) completely removed. In fact, most divalent cations, when present in small concentrations are reasonably well removed by sodium from cation exchange resins, provided hardness is also removed.
Anion resins don’t prefer divalent ions over monovalent ions to the same degree as cation resins. This causes anion resins to be susceptible to another type of dumping. If the contaminant we’re trying to remove isn’t the most preferred ion in the feed water, it will be exchanged off the resin and back into the product water if the resin is left in service past the exhaustion point. This dumping of a contaminant occurs whenever the contaminant is less preferred than some other ion and, again, can result in release into product water at a concentration several times higher than in the feed water. Since sulfate is preferred over contaminants such as arsenate and selenate, these contaminants will dump whenever the resin is run to the point of sulfate breakthrough.
So-called “selective resins” are really divalent “de-selective.” Ion exchange materials are commercially available that reduce preference for divalent over monovalent ions. The most common exchanger of this type is “nitrate selective” resin. Another example is “ammonia selective” zeolite. Nitrate and ammonia are monovalent ions and their removal hinges on the selectivity difference between monovalent and divalent ions.
The selectivity of an ion exchange material for divalent ions depends in part on proximity of the exchange groups. Strongly cationic and strongly basic ion exchange resins are really quite concentrated and exchange groups are in close proximity to one another. This proximity creates a preference for divalent ions, because a divalent ion can stabilize itself between two functional groups.
When exchange groups are spread farther apart, the selectivity for divalent ions over monovalent ions decreases. Unfortunately, so does total capacity. However, “de-selection” of divalent ions increases the apparent selectivity for preferred monovalent ions, when divalent ions are present.
This is the case with nitrate selective anion resins. These resins aren’t actually more selective for nitrate, but are less selective for sulfate. They have two advantages over regular anion resins. First, they have high operating capacity for nitrate when feed water contains sulfate in addition to nitrate. Second, they won’t dump nitrate if operated past exhaustion.
A similar phenomenon occurs with exchange of ammonia with respect to hardness. Inorganic zeolites from the “alumino-silicate” family are cation exchangers that are generally less selective for divalent cations than organic “cation resins” normally used in water softeners. They have lower capacity than styrenic strong cation exchange resin; but because they’re de-selective for divalent ions, their selectivity for ammonia in mixed solutions is dramatically improved over strong cation resin. In particular, natural zeolites such as clinoptilolite can be used to remove ammonia in preference to hardness and often have a much higher operating capacity when feed water contains a significant hardness concentration. Natural zeolites are commonly used for ammonia removal in aquaculture and the aquarium industry.
Although ammonia is preferred over sodium by styrenic strong cation resin, the presence of hardness will displace ammonia from resin. In cases where strong cation resin is used for ammonia removal, it’s necessary to have two softeners in series. The first softener removes hardness. The second softener removes ammonia. There are two potential negatives to this process. First, it’s necessary to remove hardness along with ammonia. Second, capacity for ammonia removal is rather low. However, the practice of series softening for ammonia removal has been proven successful, primarily by the pharmaceutical industry.
Removal of naturally occurring organics by ion exchange resins is gaining in popularity, because chlorinated organics are a health concern in drinking water and ion exchange produces a smaller waste volume and costs less than reverse osmosis systems. The so-called natural organics are decomposition products from various types of vegetation. The most common natural organics are tannic acid (from pine needles) and humic acid (from leaves). These organics are poorly ionized—relatively large molecular substances with poorly defined structures. However, they all contain carboxylic functional groups, which are weak acids and have a chemical structure similar to carbon dioxide. These groups are important because they provide the hook or ionized linkage that allows an organic to be exchanged. Because organics containing carboxylic groups are at least partially ionized, strong base anion resins remove them.
Every resin manufacturer has its special organic scavenger resins (see Figure 1). Regardless of manufacturers claims, keep in mind that within a given polymer structure, increasing moisture content will always increase the resin’s ability to remove organics (above about 80 percent H2O, a resin becomes so soft water won’t flow through it). Low moisture resins of any kind (less than about 55 percent H2O) are poor candidates for organic traps. This is because spaces between carbon atoms through which ions enter and leave the bead become smaller with less moisture, reducing the ability to exchange larger organic molecules. Also, keep in mind, since natural organics are large ions that are absorbed slowly, increasing surface area also increases a resin’s ability to remove organics. This means fine mesh—or smaller—resins work better than standard mesh resins at the same flow rate. Macroporous resins work better than gel resins, provided the gel-phase porosity of the macroporous resin is similar to that of the gel resin.
Organic acids are generally more soluble in high pH solutions. Organic scavenger systems work better at elevated pH and also when following softeners. Because resins used in these systems are removing natural food substances for bacteria and other bio-organisms, they’re very susceptible to biofilms and biofouling. For instance, the acrylic matrix is particularly susceptible to growth of mold and yeast.
The key point when designing an organic scavenger is the relatively slow diffusion rate of large organic ions. This leads to kinetic sensitivity, both during the service cycle and especially during the regeneration cycle. Regeneration should be performed over as long a period of time as practical given restraints of the equipment used. The brine cycle should be a minimum of 60 minutes, and longer if possible. It’s helpful to increase the pH of the brine to around 10, as this increases solubility of the organic acids. It’s also helpful to heat the brine to around 110º F. The benefit of alkaline chemicals and/or hot water during regeneration must be weighed against the difficulty of including these features, as well as skill of the end-user to operate a more complicated system successfully. Most industrial users will operate a complicated system without problems, while many residential users won’t have a clue. The most common pitfalls associated with organic traps are operating them at too high a flow rate and not regenerating them frequently enough. In the first case, removal of organic ions is poor; in the second, the resin fouls and stops working. Selectivity of strong base resins for organic anions is similar to that of sulfate; thus sulfate concentration in the feed water has an important effect on the resin’s ability to remove organics. Above about 2,000 parts per million (ppm) of sulfate, an organic trap won’t work. This is because scavenger resins prefer sulfate about the same as organics—therefore, a high concentration of sulfate out-competes the organic ions.
Conventional ion exchange resins readily remove natural radioactive metals such as uranium and radium. Radium is a divalent cation with extremely high selectivity for sodium-form cation resin. Uranium is almost always present as an anion and has very high selectivity for strong base anion resin. Radon however is an inert gas and isn’t removed by any ion exchange material because it isn’t present as an ion. Radon and radium are frequently confused because their names are so similar. The resin will concentrate the radioactivity during service exchange. This could present a health risk or potential liability to the manufacturer of such systems. Some of the radioactivity will wind up in the brine discharge to waste, another potential liability. Fortunately neither uranium nor radium will dump.
Removal of most ions listed in this discussion is straightforward—no more difficult than a water softener. However, there are some “tricks of the trade” and subtle problems that can prevent such systems from operating successfully. It’s strongly recommended someone who’s familiar with the particular contaminant—and who has other expertise with specialty uses of ion exchange resins—should review all unusual designs.
Keep in mind, removal of any ion depends on the mix of other ions in the water. This is especially true with less well-known contaminants. It’s quite important to obtain a good water analysis that includes all common ions as well as unusual contaminants. With reasonable care, common conditioners can often be used as a specialty device to remove an objectionable contaminant. These niche applications can be very rewarding from the satisfaction of helping someone solve a problem as well as the higher selling price such systems generally command.
About the author
Peter Meyers is technical manager for ResinTech Inc., a resin manufacturer and distributor in Cherry Hill, N.J. He has nearly 30 years industry experience covering a wide range of applications from demineralizers, polishers and softeners to industrial process design and hardware operation. He also is a member of the WC&P Technical Review Committee. Meyers can be reached at (856) 354-1152, (856) 354-6165 (fax) or email: email@example.com
SIDEBAR: Important Disclaimer for All Ion Exchange Systems
Before getting into details of various contaminants to be removed using a water softener, it’s essential water treatment professionals understand the most important limitation of ion exchange—the contaminant must be present as an ion (an atom or group of atoms that carries an electrical charge as the result of having lost or gained electrons) to be removed. When dealing with unusual contaminants we’re not intimately familiar with, we must be certain this is true.
For example, barium sulfate isn’t very soluble and only a few parts per billion (ppb) of barium is ionized when sulfate ions are present. Since a cation exchange resin can only remove the portion of barium that’s ionized, there may be little or no removal by a softener of the total barium in feed water. Another important example occurs with lead. At neutral to low pH, lead is usually present as an ion; but as the pH increases above 7, a larger portion of the total lead will be present as lead carbonate, which isn’t present as an ion. Since a softener can only remove ionized lead, its ability to reduce lead is poor when the inlet pH is high; and, above a pH of 9, practically no removal can be expected—except by filtration.
From a liability standpoint, it’s essential to know the contaminant is present in a form that can be removed by ion exchange. There are several ways to resolve this. One is to analyze the water to determine what form the contaminant is present. This can be difficult or impossible. Grab sample analysis is susceptible to various types of distortion caused by the act of taking the sample. A better way of determining if ion exchange will remove a contaminant is to try a small scale system and analyze the product water quality. It may be necessary to analyze several samples over a period of time to ensure initial results are sustainable. This is particularly true of potentially toxic contaminants—such as arsenic or nitrate—that could be a significant health risk if they’re not removed.
Effect of TDS on Removal of Ions
One of the characteristics of ion exchange resins is that at low TDS they prefer divalent ions, but at high TDS they prefer monovalent ions. During regeneration, high TDS in salt brine reduces the resin’s preference for divalent cations. The resin begins to favor monovalent ions. Thus, resin is efficiently regenerated. We can take advantage of selectivity reduction (or even reversal) to design systems that efficiently remove divalent ions during service exchange, yet are efficiently regenerated by common salt brine.
Preference for one ion over another isn’t by itself sufficient to predict success of an ion exchange system. As previously mentioned, only ions can be exchanged. This isn’t only true for service exchange, but also for the regeneration exchange. One problem during regeneration is a slightly soluble contaminant (such as iron or lead), can be concentrated on the resin during service, and then can precipitate into and/or onto the resin during regeneration. This leads to fouling of resin that can cause them to stop working after several cycles. It can also lead to a buildup of non-ionized contaminants trapped in the resin bed that can periodically “breakthrough” and “dump” into the product water at a concentration greater than in the feed water. Increasing the salt dose used during regeneration may or may not be effective to prevent this type of fouling from occurring. Since most water softeners backwash before but not after regeneration, dumping of suspended contaminants released during regeneration can be a concern, particularly if the contaminant has a high toxicity potential.
The principal of selectivity reversal—or changes caused by TDS in a resin’s preference—also applies to strong base anion resins. If ordinary strong base anion resin is used in a softener instead of cation resin, it will remove divalent contaminants such as uranium (generally present as an anion), molybdate, arsenate and selenate as well as other divalent anions—provided the resin is regenerated well before it’s exhausted by the most preferred species (usually sulfate).
Figure 2. Commercially available organic scavenger resins*
Resin Type Advantages Disadvantages
Acrylic strong base gel Very good in both loading and also during regeneration Can smell fishy at high pH. Very susceptible to mold and other biogrowths
Acrylic strong base macroporous As good as its gel counterpart, perhaps even better. Lower capacity and higher cost than other scavenger resins
Very high-moisture gel Type 1 strong base Almost as good as the acrylics at removal and regeneration; Does not add as much chloride to the water Resin is very soft and elastic, and is prone to high pressure loss
High-moisture macroporous Type 1 strong base The strongest and most chemical resistant of all the choices; Works well, provided there’s enough macroporosity Lower capacity than other scavenger resins; Might not work quite as well as the other choices
Fine mesh resins Greater surface area improves kinetics, leading to as much as 10-15% more capacity and also lower leakage More expensive, higher pressure loss, more prone to physical plugging and requires special slot or screen size
Uniform particle size resins If uniformity allows average bead size to be made smaller, will work like a fine mesh resin, otherwise almost no benefit; Slightly lower pressure loss, slightly shorter rinse and slightly lower leakage Slightly lower capacity due to increased void spaces; Easier to backwash out of the tank
* Anion ion exchange resins Source: ResinTech Inc.