By Richard A. Hayden, Andrew F. McClure and Neal E. Megonnell

Summary: Activated carbon is used routinely for dechlorination in municipal water systems, however, an activated carbon bed properly sized for dechlorination may not be adequate for dechloramination. Since the ability of activated carbon to destroy chloramines is lower than that to destroy chlorine, premature breakthrough of chloramine may result from using an activated carbon bed designed for dechlorination. In Part 2 of this series next month, an author from Parker-Hannifin Corp.’s O-Ring Division will comment on chloramines’ affect on seals for tubing, fittings and connectors.

Historically, municipal drinking water treatment facilities have used chlorination as the most common means of bacterial control in their treatment systems. However, a commonly used alternative growing in popularity is chloramination, where the weaker oxidant chloramine is used for water disinfection. Both disinfectants can impart taste and odor problems to treated drinking water and many end users desire removal of these compounds for various reasons.

Activated carbon has been the technology of choice for years for removal of both chlorine and chloramine. It’s important to note, however,  performance of carbon differs greatly between these two applications, and carbon systems designed for dechlorination may not be adequately sized for chloramine removal. Water treatment professionals need to be aware of differences in activated carbon performance between chlorine and chloramine removal, especially in locations where municipalities switch from chlorine to chloramine during summer months.

Chlorine is still the most common disinfectant used in drinking water treatment today. When added to water, chlorine will dissociate to hypochlorous acid (HOCl) and hypochlorite ion (OCl-). These two species act as strong oxidants that serve as very effective disinfectants, but can impart taste and odor to the treated water. Both hypochlorous acid and hypochlorite can be removed through chemical reactions with activated carbon. The carbon surface is represented as C* and the oxidized carbon surface is represented as C*O in the following equations:

  • HOCl + C* ↔ HCl + C*O
  • OCl- + C* ↔ Cl- + C*O

These reactions occur very quickly and slowly oxidize the carbon over time. Eventually, the carbon is consumed by this reaction and needs to be replaced. Due to the rapid nature of these reactions, a relatively short contact time for a 12×40-mesh carbon is normally adequate to remove the chlorine from typical concentrations of 1-to-2 parts per million (ppm)—the U.S. limit on chlorine and chloramine in potable water is 4 ppm under the federal Disenfectants/Disinfection By-Product Rule—down to levels where taste and odor problems are alleviated. All grades of carbon regardless of starting material will generally provide equal levels of performance and require the same contact time for efficient dechlorination, although denser carbons can provide a longer bed life since they allow for more carbon by weight to be put on-line in a given bed volume.

As mentioned above, chlorine is a strong oxidant. While this makes it an effective disinfectant, there’s a downside in that residual chlorine can react with naturally occurring organic matter in the water to form trihalomethanes (THMs), which have been classified as carcinogens. To avoid trihalomethane formation, many municipalities have begun to use chloramine as a disinfectant. Chloramine isn’t as strong an oxidant and thus is a much weaker disinfectant, but results in much lower THM formation.

Chloramination is accomplished through addition of ammonia (NH3) before or after the chlorine addition step, which forms monochloramine (NH2Cl) at neutral pH levels, i.e., 6.5-7.5:

  • NH3 + HOCl ↔ NH2Cl + H2O

For nearly all drinking water applications, the pH of the water is above 6.0 and the predominant chloramine species is monochloramine. At lower pH levels, other species (dichloramine and trichloramine) can form, however, these species are much less stable, far less prevalent and far more reactive with carbon than monochloramine, and therefore won’t be discussed further in this article.

Similar to dechlorination, activated carbon removes chloramine through a chemical reaction. The main difference is that the reaction involving chloramine removal is a pure catalytic reaction, in which the carbon is not consumed as part of the reaction. The exact chemistry involved isn’t known but, generally, the following reactions are believed to predominate:1

  • NH2Cl + H2O + C* ↔ NH3 + Cl- + H+ + C*O
  • 2NH2Cl + C*O ↔ N2 + 2Cl- + 2H+ + H2O + C*

Of note in these reactions is the fact the carbon surface (C*) emerges unchanged from the set of reactions and ammonia, chloride, and nitrogen are formed from the catalysis. If necessary, these reaction products can be removed by reverse osmosis or ion exchange.

Because the chloramine removal reaction with activated carbon is catalytic in nature, the degree of catalytic activity possessed by the carbon is the crucial parameter for measuring performance in chloramine removal applications. Traditional measures used with activated carbon such as iodine number and BET surface area (an acronym based on the names of developers of the test method) don’t provide any indication of performance for chloramine removal. In recent years, the peroxide number has been developed as a more accurate measure of catalytic function in activated carbon, with lower peroxide numbers indicating greater levels of catalytic activity. Traditional activated carbons tend to vary in terms of natural catalytic function; coconut- and lignite-based carbons possess very little catalytic activity, while bituminous coal-based products generally possess a moderate level of catalytic functionality. Certain grades of carbon, which are manufactured with enhanced catalytic function, have been on the market for the past few years.

The amount of contact time required for adequate decomposition of chloramine to occur will vary depending on the type of activated carbon used, along with the particle size, catalytic activity, chloramine concentration and the pH and composition of the water contacting the carbon. Generally speaking, monochloramine removal efficiency will increase with decreasing carbon particle size, increasing carbon catalytic activity, increasing monochloramine concentrations and decreasing solution pH and total organic carbon (TOC) content. Because in most applications water composition is fixed, and in many applications the parameters of a system design (i.e., pressure drop, contact time, etc., may also be limited), it’s important to select a carbon having those properties that are the most beneficial to chloramine removal. For a typical point-of-entry application utilizing a 12×40-mesh size and an influent chloramine concentration of 2-to-4 milligrams per liter (mg/L), the contact time required to reduce the chloramine to 0.1 mg/L or less is dependent above all on the catalytic nature of the activated carbon used.

A comparison
The following table shows characterization data for six activated carbons. For these carbons, the catalytic activity is measured by the ability of the carbon to catalytically decompose hydrogen peroxide under a controlled set of conditions. The peroxide number is defined as the time a fixed amount of activated carbon destroys three-quarters of the peroxide utilized in the test. The lower the peroxide number, the more catalytically active the carbon (see Table 1).

Table 2 summarizes the contact time required to remove chlorine or chloramine from 3 ppm to 0.1 ppm under ideal conditions (deionized water and no background organic carbon) utilizing the six carbons shown in the characterization table. All carbons were sized to 16×20 mesh to eliminate the effects of particle size on performance, except for the 8×12-mesh carbon, which was utilized to show the effect of particle size on performance.

For dechlorination, the contact times required to reduce 3 ppm chlorine to 0.1 ppm are similar with the exception of the Coconut #2 and Catalytic Bituminous 8×12-mesh materials. Coconut carbons are known to vary in quality, therefore it’s possible to obtain a relatively active material for both chlorine and chloramine; however, the consistency from batch to batch is an issue that’s not controlled during the manufacturing process.

For dechloramination, the contact times required to reduce 3 ppm chlorine to 0.1 ppm vary depending mainly on the catalytic activity as measured by the peroxide number. The coconut carbons vary widely again showing the variation in coconut carbons. For dechloramination, the catalytic bituminous carbon shows superior performance when comparing equal particle size materials.

When comparing the six carbons for dechloramination, it should be noted that the contact time required for the same carbon used for dechlorination can increase by as little as 58 percent for the catalytic bituminous carbon or as much as 804 percent for the Coconut #2 carbon. Note the variation again with the coconut carbon in that the increase in contact time required between dechlorination and dechloramination can be as little as 106 percent or as much as 804 percent.

Particle size will play a major role in performance for dechloramination. The necessary increase in contact time by using a larger particle size carbon is also not constant since the contact times increased by 58 percent for the 16×20-mesh carbon and by 115 percent with the 8×12-mesh carbon. The increased contact time for the 8×12-mesh carbon is expected since it’s known that particle size is the major controlling factor for dechlorination and a major controlling factor for dechloramination.

Surface area
Parameters such as iodine number, BET surface area and apparent density do not control the ability of a carbon to dechlorinate or dechlora-minate. Although the iodine number and BET surface area for the carbons tested ranged from the middle 600s to close to 1,200, the contact times required for dechlorination varied only by a few seconds. Therefore, a high iodine number or high BET surface area doesn’t indicate a carbon will perform well for dechlorination or dechloramination.

Since the contact time required for dechloramination can vary greatly, it’s extremely important that the type of carbon selected be taken into account since a carbon that dechlorinates well won’t necessarily dechloraminate similarly. As mentioned, typical carbon parameters such as iodine number, BET surface area and apparent density do not control a carbon’s ability to dechlorinate or dechloraminate. Although other as yet unidentified factors are clearly in play, the peroxide number is a reasonable measure of a carbon’s ability to dechloraminate but has no bearing on a carbon’s ability to dechlorinate. The particle size of the carbon is the main controlling factor for dechlorination.

Carbon variability is also a major concern for dechloramination since only the catalytic bituminous carbons are manufactured to control the catalytic property necessary for dechloramination. Although it’s possible to obtain samples of various activated carbons that will perform acceptably for dechloramination, this is purely coincidence since the catalytic property isn’t controlled. In general, catalytic ability is higher for bituminous coal-based carbons over coconut- and lignite-based carbons, with catalytic bituminous coal-based carbons being superior for dechloramination.

The contact time required for dechlorination is controlled mainly by the particle size of the carbon. Typical activated carbon properties such as iodine number, BET surface area and apparent density don’t control the ability of a carbon to dechlorinate. Dechloramination is controlled mainly by the catalytic activity of the carbon as measured by the peroxide number and somewhat by the carbon particle size but, again, not by typical properties such as iodine number, BET surface area and apparent density. Finally, carbon variability is also an issue when selecting an activated carbon that may be used for both dechlorination and dechloramination. Variability in activated carbons made from materials such as coconut not only occurs with respect to adsorption properties, but also to dechlorination and dechloramination.


  1. Kim, B.R., and V.L. Snoeyink, Journal AWWA, Vol. 72(8), October 1980, pp. 488-490.

About the authors
Richard A. Hayden is senior research associate at Calgon Carbon Corp. of Pittsburgh. Andrew F. McClure is Calgon’s market manager. And Neal E. Megonnell is senior research engineer at Calgon. For more information on this article, contact McClure at (412) 787-6761, (412) 787-6713 (fax) or email: [email protected]


Comments are closed.