By A.J. deLange
Brewers who use municipal water for their beers know that it is treated with chlorine for disinfection and that residual chlorine may react with phenols in malt to produce chlorophenols, which lend a plasticlike taste to beer at parts-per-billion levels. Most brewers also remember from their days of keeping pet goldfish that allowing water to stand, aerating it, or boiling it will allow chlorine to escape, thus rendering the water fit for Goldy and for brewing.
In recent years, more water authorities have started to treat water with ammonia in addition to chlorine. This treatment results in the formation of chemicals called chloramines, which are similar to chlorine in that they kill bacteria and aquarium fish and ruin beer.
Standing, aeration, and boiling will remove chloramines from water, but not very effectively. Water in my area (Fairfax County, Virginia) contains the equivalent of 3 mg/L of chlorine in chloramines, a fairly high level. Ten gallons of this water allowed to stand in a 25-gallon stock pot required weeks to lose chloramine down to the <0.1 mg/L level. Almost two hours of boiling is required to get the chloramine in Fairfax County water down to the hundredths of milligrams per liter.
This article explains how to measure chlorine and chloramines in your brew water and how to reduce or eliminate these beer-spoiling chemicals if they are causing you problems.
Get your very own water test kit to test the quality of your water at home!
What Are Chloramines?
Chlorine was first used experimentally for disinfection of a municipal water supply in Louisville, Kentucky, in 1896. The first permanent installation of chlorination equipment in a water treatment plant in this country was made in 1908.
Chlorination is simple, effective, relatively safe, and inexpensive. Its efficacy comes from the strong oxidizing potential of the hypochlorous acid molecule formed when either chlorine gas or salts, such as sodium hypochlorite, are dissolved in water at modestly low pH.
Chlorine oxidizes not only bacteria and viruses but other substances found in water as well. Plants contain phenols, and plant matter finds its way into surface waters so that such waters, when chlorinated, may have the flavors and aromas of chlorophenols, which are regarded no more highly in the water industry than among brewers. If, after the chlorine has done its job of killing microorganisms at the water plant, it can be converted to a less active form, it can still maintain some of its bacteriostatic power but will not be active enough to form chlorophenols during its travel through the distribution mains to the customer. Injecting ammonia into water accomplishes this by converting the chlorine into chloramines, a process known as chloramination.
Chloramination was first done for purposes of flavor improvement in 1926 at Greenville, Tennessee, where the water had objectionable qualities from the presence of phenols. In recent years, many water authorities have begun to chloraminate to limit chlorine’s reaction with other organics from decaying vegetation that are often found in surface water. These decay products (mostly humic and fulvic acids) combine with chlorine to produce a family of chemicals called trihalomethanes (THMs), which are thought to be carcinogenic. Today’s federal and state drinking water regulations regulate allowable THM levels quite strictly, so chloramination is becoming increasingly popular. It is estimated that 25% of the larger and perhaps 5% of smaller water treatment plants in the United States use chloramination today as compared with about 2.6% of plants surveyed in 1963.
Do You Really Need to Remove It?
The reaction between ammonia and chlorine produces monochloramine, dichloramine, and trichloramine. Dichloramine and trichloramine are relatively volatile and escape from the water soon after treatment; by the time the water gets to your house or brewery, only monochloramine and some free chlorine remain. The level remaining depends on how much chlorine and ammonia the treatment plant added, the distance from the treatment plant to your house, the temperature, and the chemical makeup of your water. These factors (except the distance to your house) may vary seasonally.
If your water district is changing over to chloramination, or if you’re moving to a chloraminated-water area, you may be wondering how you can get chloramines out of your brew water. First, however, you should ascertain whether chloramines really are a problem. Does your beer have off-flavors? If you can’t taste anything and are not a trained taster, have it tasted by someone who is. If you aren’t getting the medicinal or plastic-like taste of chlorophenols in your beer, don’t worry about removing chloramines.
Click here to check out our selection of water filters and filtration systems!
If you do decide that you need to dechloraminate your water, you have a couple of options.
Campden tablets: I have found that, for home brewers, the easiest way to dispose of chloramines is to treat the water with Campden tablets, which are typically used as a preservative by winemakers. The Campden tablets I use contain approximately 695 mg of potassium metabisulfite, sometimes referred to as “metabite.” This chemical reduces chloramine to ammonium and chloride ions, both of which are beneficial to beer in the quantities generated by this treatment. (The ammonium ion nourishes the yeast, and the chloride ion enhances the drinker’s sensation of roundness and fullness in the beer.)
How much? The required dose is simple to calculate: Take twice the chloramine level in the water, add the chlorine level, and divide by 6. This is the number of tablets required to treat 20 gallons. Scale this value according to how many gallons need to be treated. For example, if I were to brew with the local water, which has 3 mg/L chloramine, I would need one tablet per 20 gallons. It is rare that chloramine levels will be above 3 mg/L, but they occasionally are. It is best to test (see the box “Testing the Waters,” below); what the water authority tells you may be a target or average value, not the actual level.
My experiments have shown that perhaps 20–30% more potassium metabisulfite than calculated should be used to be on the safe side. This represents a modest increase in the amount of by-products. Also be aware that not all Campden tablets weigh 695 mg, nor are they all, apparently, potassium metabisulfite; some are sodium metabisulfite. Weigh the Campden tablets. (If a lab balance is not available, weigh lots of them on a kitchen scale or reloading scale and divide by the number of tablets.) If you are uncertain as to whether they are the potassium or sodium salt, have the supplier check with the wholesaler, or just assume they are potassium. If you guess wrong, you will be adding 17% more bisulfite than you need — not a significant amount.
Testing the Waters
Chlorine is such a simple thing to test for that there is no reason why any brewer concerned about the levels of chlorine in his or her water and the effectiveness of a remediation method should not be checking levels of both free and bound chlorine (chloramine). Test kits of many levels of sophistication (and cost) are available from several sources.
The simplest and quickest place to get a test kit is the local pet store. These kits, sold for use by amateur aquarists, contain poisonous orthotolidine, used to measure free chlorine, and the even more poisonous Nessler’s reagent, used to estimate the ammonia released by chloramine. If you obtain such a kit, treat the chemicals in it with respect. (One such kit, the Doc Wellfish #54 from Aquarium Pharmaceuticals, is no longer available, but others that use orthotolodine may still exist.)
Analysis laboratories have abandoned orthotolidine in favor of DPD (N,N-diethyl-p-phenylenediamine), in part because of orthotolidine’s toxicity but also because of accuracy. Most kits obtained from a lab supply source will be based on DPD. DPD is colorless in its reduced state, but when oxidized it turns magenta, forming what is known as Würster dye. When DPD is added to a sample containing free chlorine (or another oxidizing agent), the depth of magenta color formed depends on the amount of the oxidizer present. When combined chlorine is to be measured, potassium iodide is added to the sample with the DPD. The chlorine in chloramine oxidizes the iodide ion to free iodine, and the iodine then oxidizes the DPD, causing the conversion to Würster dye. Thus, to get a complete picture of the status of a water sample with respect to chlorine using DPD, you would run two tests. The first, using iodide and DPD, gives an indication of the total chlorine (free plus bound in chloramine). The second uses DPD alone and measures the free chlorine only. The estimated chloramine level is taken as the total chlorine test result minus the free chlorine test result.
The depth of the color developed when DPD is oxidized can be judged in several ways, depending on the sophistication of the test and accuracy required. Brewers need not strive for great accuracy unless they want to take several measurements for the purpose of estimating half-life. More typically, brewers are looking for a 0 reading — an indication that they have successfully rendered their water chlorine/chloramine-free.
If more accurate results are desired than can be obtained with test kits, the depth of color can be measured by a spectrophotometer or colorimeter configured to measure light absorbance at 530 nm. The chlorine concentration (free or total) is read from a calibration curve that is determined by the analyst (using chlorine solutions of standard strength), supplied by the manufacturer, or programmed into the instrument.
The most sensitive and accurate measurements are made using amperometric titration, in which current flowing between a pair of biased inert electrodes is monitored while a reducing agent is gradually added to a test sample. This equipment is expensive and requires a skilled operator. Brewers would be unlikely to encounter this method of measurement.
Full details of these and other methods of chlorine measurement can be found in reference 4. Most of the available kit chemistries are based on the procedures set forth in this reference.
I used test kits from Hach Company. Cube-type kits good for 50 analyses cost about $ 13; disc-type kits good for 50 analyses cost $ 35. Other manufacturers of chlorine test kits include Hanna Instruments, CHEMetrics, and LaMotte Company.
In simple inexpensive kits, water and DPD (plus iodide if a total test is being done) are placed along with suitable pH buffers in a small test tube and allowed to react. The color of the test tube is then compared to colored patches printed in the instructions or on the side of the carton in which the kit was shipped. In other inexpensive implementations, the test tube includes a fixture that contains built-in color comparison patches. In either case the value is read from the patch that most closely matches the color of the liquid in the tube.
More expensive kits will contain a transparent wheel around whose edge a colored strip of variable density Würster dye has been printed. A small portion of this wheel is viewed through an aperture adjacent to the tube containing the test water plus DPD. The user rotates the wheel until the best color match is obtained and reads the chlorine amount from a scale on the periphery of the wheel. In some cases, the light passing through the wheel first enters a second tube containing test water to which no chemical has been added. This removes the effects of any color in the water itself.
Your beer will easily tolerate two or three times the required dose (vintners use one or two tablets per gallon), so if your answer contains a fraction of a tablet, you can just round up to the nearest whole tablet. It probably makes sense to round up to the nearest half tablet or whole tablet to be sure you get all the chlorine and chloramine if actual levels are higher than the reported (or measured) ones or if the particular tablets you use weigh less than 695 mg.
If you did want to add fractional tablets, you could dissolve, or rather suspend, a tablet in 100 mL of water and, just after agitating, measure out the number of milliliters that corresponds to the percentage fraction required. For example, if a third of a tablet is required, measure out 33 mL.
The reaction takes place fairly quickly and is essentially complete in a couple of minutes.
What’s being added to your beer? The amount of metabite required to neutralize the levels of chlorine and chloramine typically found in municipal tap water results in by-products at concentrations that are insignificant when brewing all styles of beer except those that require very soft water (for example, Bohemian Pilsener). You can easily figure out how much of these by-products is produced. If exactly the correct dose of metabite is used, each milligram per liter of chlorine or chloramine will be converted to 1 mg/L chloride ion. The chlorine conversion produces some hydrogen ions so that alkalinity is reduced slightly: 2.1 ppm of hydrogen ion for each milligram per liter of chlorine and 1.4 ppm for each milligram per liter of chloramine. Each milligram per liter chlorine converted will also result in 1.3 mg/L sulfate and 0.55 mg/L potassium. Each milligram per liter chloramine will produce 2.7 mg/L sulfate, 0.5 mg/L ammonium ion, and 1.1 mg/L potassium. If sodium metabisulfite is used, sodium in the amount of 1.34 mg/L will be added to the water for each milligram per liter chlorine neutralized and 2.67 mg/L for each milligram per liter chloramine. For the Fairfax County example with its 3 mg/L chloramine, treatment with potassium metabite should result in 3.3 mg/L more potassium, 3 mg/L more chloride, 1.5 mg/L more ammonium, and 8.1 mg/L more sulfate.
Carbon Filtration Systems for Professional Brewers
by Michael Davis
Many large and small commercial breweries are installing granular activated carbon (GAC) filter systems to remove chloramines and chlorine. GAC is the industry standard for reducing levels not only of chloramines and chlorine, but also pesticides, industrial chemicals, trihalomethanes, and other halogenated organic compounds, as well as bad tastes and odors. Designing a GAC system for chloramine removal is not rocket science, but when you compare suggestions made in published articles or advice given by different suppliers or manufacturers, you are likely to get some conflicting and confusing information. This is because the chemistry involved is subject to site-specific variables, such as pH and water temperature. Thus, a system that works in one brewery might fall short in another. Although it is difficult to generalize about system requirements, the following guidelines should help you to shop for a water treatment system for your brewery.
Cartridges or Backwashing Tank Systems?
First, you need to determine whether you are going to use a cartridge-based system or a backwashing tank–type system. If you are using water at a slow flow rate (that is, less than 2 gallons per minute), a cartridge system will work if sized correctly. Unfortunately, because there is no regulatory standard or protocol yet in place for the removal of chloramines, small cartridge system manufacturers are not yet testing or making claims for chloramine reduction. You can, however, use the industry standard required contact time for chloramine reduction as a guideline; that is, you need at least two to three times the contact time required for chlorine reduction. So, if you are going to use a cartridge-based system, size it to at least two to three times what is being claimed for chlorine reduction, then test to make sure that you are getting a sufficient reduction in chloramines. (See the box “Testing the Waters” on page 60.) Although they are the least expensive option, cartridge systems may be limited in their capabilities because of their low flow capacity.
The other option is a backwashing tank system. This is a tank filled with bulk GAC granules through which water trickles. Opening a valve back-flushes the granules to clean them.
With either method, you must determine how much carbon you need to provide empty bed contact time (EBCT) sufficient to remove all chlorine and chloramine. Multiply the volume of granular activated carbon (GAC) in cubic feet times 7.48 to determine the volume in gallons. Then divide that by your maximum water flow rate. As long as you are using a high-quality carbon with a maximum mesh size of 12 X 40 and a minimum iodine (a quality specification) of 900, you need to size the carbon bed or cartridge for a minimum EBCT of 10 minutes. Ten minutes may be more than you need (it is the standard for chloramine removal for hemodialysis treatment), but other variables that can affect removal capability (actual level of chloramine in the water, pH, temperature, and so forth) may make it prudent to oversize your system.
Your Mileage May Vary
You will need to replace the cartridge or bulk GAC periodically. Again, cartridge manufacturers are not yet publishing replacement timing specifications for chloramine reduction. They are recommending replacement based on the cartridges’ performance in chlorine removal and the possibility of bacteria fouling. Assume that your cartridges will be able to remove chloramines for only one-half to one-third of the life they would have removing chlorine alone. To be sure, regularly test for chloramines and make sure that you change any regular cartridge at least every 6 months and any backwashing GAC system at least every 12 months. When the cartridge or bulk GAC is changed, make sure you thoroughly clean the housing or tank before replacing the media.
How Much Will This Cost?
The answer is a firm, “It depends.” The greater the flow rate required, the more expensive the system. But just because you produce more barrels of beer than your cross-town rival doesn’t necessarily mean that you need a bigger carbon filtration system than your rival’s. You can minimize the necessary flow rate by buying a big storage tank and filling it up through the filter between brews; even a cartridge filter used in this way will provide enough water for a brewpub. If chlorine-free water is left in the tank, however, the tank will begin to develop a bacterial film, which can cause off-flavors. Some breweries use ultraviolet light to control biofilm in the pipes.
If you have any concerns about designing or obtaining water treatment equipment, refer to a water treatment specialist who has had experience in chloramine reduction.
Any excess metabite winds up as potassium or sodium plus sulfur dioxide in the beer. The L.D. Carlson tablets are potassium metabisulfite, which is about 35% potassium and about 55% sulfur dioxide (the rest is oxygen). Because these tablets weigh 695 mg, this means an extra tablet (that is, one that has no chlorine or chloramine to react with) would leave 243 mg (3.2 mg/L in 20 gallons) of extra potassium and 382 mg (5 mg/L in 20 gallons) of sulfur dioxide. The sulfur dioxide will either reduce something in the mash to a reductone (a reduced-state organic substance thought to prevent staling reactions in beer) and become sulfate in the process, or be driven off as a gas during the boil. If it all converted to sulfate, 1 tablet in 20 gallons would increase sulfate by about 9 mg/L. As metabite, in either form, is a basic salt, excess metabite will increase the alkalinity of the water slightly.
Activated carbon: For commercial brewing operations, granulated activated carbon (GAC) is the best way to remove chlorine and chloramines. Home brewers also can use this method. There are three major types of GAC systems — those that use GAC only, those that use GAC with a material called KDF, and those that use GAC with ion-exchange resins.
GAC only. GAC-only units will remove chlorine, chloramine, and organics, which are often responsible for colors, tastes, and odors in water.
GAC and KDF. KDF is a copper-zinc alloy developed by KDF Fluid Treatment Inc. (Three Rivers, Michigan). The alloy can reduce heavy metal ions such as copper, lead, mercury, and chromium to metal atoms, which electroplate onto the medium and are thus removed from the water. Beneficial brewing ions, such as calcium and magnesium, are not affected. The zinc in KDF will also reduce dissolved chlorine gas to chloride ions, relieving the GAC of some of the chlorine reduction burden. I have seen no data concerning KDF’s ability to reduce chloramines, but, of course, the GAC will take care of that.
Ion-exchange resins. Ion-exchange resins exchange ions of metals such as lead and calcium for hydrogen ions and also exchange cations, such as sulfate and chloride, for hydroxyl ions. Water treated with them is virtually ion-free, and the brewer using an ion-exchange filter will have to use supplemental salts to restore a normal ion profile when brewing most styles of beer.
Of these filters, I have experimented only with the Brita pitcher. I found that 16% of monochloramine remained after a pass through the filter. The Brita literature indicates that only 7.5% of free chlorine will remain. I have not verified their number but have no reason to disbelieve it. If 84% reduction of chloramines isn’t enough for you, run the water through the pitcher again. This leaves 16% of 16% (2.5%), which should be a sufficient reduction in most situations.
Reverse osmosis. Reverse osmosis (RO) units of the type installed in homes usually have one or two activated carbon filters in the path to protect the RO membrane from chlorine and chloramine, which would poison it. RO-processed water is, therefore, chloramine-free, but as with ion-exchange–processed water, brewers must add calcium, carbonate, and sulfate to RO-processed water, according to the requirements of the style being brewed, to get the best flavor in their finished beer.
Other methods: Much speculation has been made about whether boiling will remove chloramines, prompting me to perform my own experiments. I concluded that boiling will work, but it is not very practical for brewers. I also tried a few other things that I found to be even less practical. I will outline the results briefly.
Boiling. I found that an hour of boiling (plus the hour required to bring the water to the boil) removes chloramines from 10 gallons of the local water. A boil of this duration results in a loss of typically 10–20% of the volume of the water. With the time and energy considerations, this isn’t a very practical method of chloramine reduction, unless the boil is being done anyway for carbonate reduction.
Standing. This, too, will work, but only after a very long time. Chloramines can have a very long half-life in standing water. County water in my area exhibits a chloramine half-life of about 155 hours when standing undisturbed in a 15-gallon stock pot. This means that this water, with its nominal 3 mg/L chloramine out of the tap, would be down to 1.5 mg/L after 155 hours, 0.75 mg/L after another 155 hours, and so on. Reducing 3 mg/L chloramine content to 0.025 mg/L takes seven half-lives, or 45 days! The length of time required will vary, of course, depending on the chloramine concentration of the water. The time can be appreciably shortened by aeration or stirring, but standing is still not going to be a practical method of chloramine removal.
Adding bleach. Yes, funny though it sounds, it is possible, theoretically at least, to hasten the departure of chloramine from water by adding household bleach to it. This encourages the monochloramine to convert to dichloramine, which is appreciably more volatile. The excess chlorine from the bleach exits as chlorine gas, which is more volatile than monochloramine. In an experiment with synthetically chloraminated water, this technique was quite successful. In an experiment with Fairfax County water, it was less so. Although the chloramine level dropped rapidly just after the addition of the bleach, its rate of loss eventually slowed to a half-life of 270 hours. I suspect a larger dose of bleach might have been more effective, and I am still experimenting with this technique. Note that bleach also leaves residual substances in the water.
Acidification. It is also possible to remove chloramines by lowering the pH of the water. At low pH monochloramine converts to dichloramine, which, as we saw above, clears quickly (that is, overnight) from even standing water. Reducing the pH of the water to near pH 4 with hydrochloric acid will allow the dichloramine to escape; you could then restore the pH to a higher value suitable for dough-in with some kind of base. Anyone interested in the details can get in touch with me.
I first began to suspect that boiling and aeration might remove chloramine from water in the spring of 1998, and a few simple experiments confirmed my hunch. This finding was posted to the internet’s Home Brew Digest in an article entitled “Chloramine Heresies”. Since then, I have performed more elaborate experiments, though the investigation is far from complete. Space constraints prevent me from presenting many of my discoveries in this article, and I invite readers interested in water chemistry to correspond with me for more details.
To brewers, the most important conclusions from the test data are that chloramine is more difficult to remove than free chlorine, and its behavior is somewhat unpredictable. This unpredictability is due to the complexities of chlorine and ammonia chemistry, which varies depending on pH, temperature, relative chlorine and ammonia concentration, the passage of time, and the presence of minerals normally found in drinking water. Readers contemplating use of one of the methods I tried must verify its effectiveness on their own water.