The discovery of chlorine dioxide is generally credited to Sir Humphrey Davy, who reported the results of the reaction of potassium chlorate with sulfuric acid in the early 1800’s. Chlorine dioxide today is generated for smaller applications by the reaction of sodium chlorite with chlorine, via either gaseous chlorination (Equation 1), the reaction of sodium hypochlorite with hydrochloric acid (Equation 2), or the reaction of sodium chlorite with hydrochloric acid. (Equation 3)

Cl2 + 2NaClO2 -> 2ClO2 + 2NaCl (1)

HCl + NaOCl + 2NaClO2 -> 2ClO2 + 2NaCl + NaOH (2) 5NaClO2 + 4HCL -> 4ClO2 + 5NaCL + 2H20 (3)

In the mid to late 70’s, researchers linked chlorination of potable water to increased cancer mortality rates. This increase in cancer mortality was tied to the production of trihalomethanes, THM’s. The USEPA established 0.1 ppm as the maximum THM containment level for drinking water. Research in the area of THM reduction in potable water led to the EPA, in 1983, suggesting the use of chlorine dioxide as an effective means of controlling THM’s.

Chlorine dioxide is being used increasingly to control microbiological growth in a number of different industries including dairy, beverages, pulp and paper, fruit and vegetable

processing, poultry, beef processing and canning plants. It is seeing increased usage in municipal potable water treatment facilities and in industrial waste treatment facilities, because of its selectivity towards specific environmentally objectionable waste materials, including phenols, sulfides, cyanides, thiosulfates, and mercaptans.

Despite the many advantages of chlorine dioxide, relatively few heavy industrial plants have made the switch to this superior microbicide for cooling systems. Historically, unless there was a significant performance or cost advantage, there was no impetus to consider chlorine dioxide as a replacement for gaseous chlorine.

In those instances where chlorine is not effective at microbiological control, such as a cooling tower with a high level of organics, ammonia, or amines, chlorine dioxide has been called on to bring control to these heavily fouled systems.

With the recent trend towards elimination of gaseous chlorine from the industrial plant site, there are increasing interests in exploring all the various alternatives to gaseous chlorine.

Criteria for the Ideal Biocide

What criteria would be expected for an ‘Ideal Biocide’?

Identified criteria are summarized into four basic categories:

Performance

It must exhibit rapid kill of target organisms, with a high LC50 toward non-target organisms. It must be able to keep systems clean of biofilm; ideally it should be able to clean up already fouled systems. It should not be consumed by materials commonly encountered in cooling systems, e.g., hydrocarbons, wood, plastic, or other treatment chemicals. Finally, it must be effective over a wide range of operating conditions.

Environment

Side or by-product reactions should be minimized, and reaction products should be environmentally friendly. Neither it, its by-products, nor its reaction products should persist in the environment.

Safety

It must be safe and easy to handle.

Economics

It must be affordable.

Disinfection Efficiency and pH

Many studies have been conducted comparing the disinfection efficiency of chlorine dioxide to chlorine. In one such study, varying dosages of chlorine dioxide or chlorine

were added to solutions containing 15,000 viable cells/ml of E. coli at pH’s of 6.5 and 8.5.

The abscissa is the time in seconds required to kill 99% of the viable bacterial cells. The ordinate is the initial dosage of oxidant. The results are shown in Figure 1 below.

Figure 1. Comparison of the Disinfection Efficiencies of ClO2 with Cl2 at pH’s of 6.5 and 8.5

These results clearly show the decreasing effectiveness of chlorine as the pH increases. This is to be expected because chlorine resets with water rapidly to form hypochlorous and hydrochloric acids. Hypochlorous acid, the primary biocide, dissociates as a function of pH, forming the hypochlorite ion. Hypochlorite is reported to be from 1/20 to 1/300 as effective at microbiological control as hypochlorous acid.

Unlike chlorine, chlorine dioxide remains a true gas dissolved in solution. The lack of any significant reaction of chlorine dioxide with water is partly responsible for it to retains the biocidal effectiveness over a wide pH range. This property makes it a logical choice for cooling systems operated in the alkaline pH range, or cooling systems with poor pH control.

Note: Hypobromous acid also dissociates with pH. The dissociation curve is essentially equivalent to that of chlorine; its curve is offset by about 1 pH unit toward the alkaline range from that of hypochlorous acid. For example, the pH of 50% dissociation of the hypohalous acid to the hypohalite anion is about 7.5 and 8.7 for chlorine and bromine, respectively.

Kinetics of Disinfection

Figure 2 shows a comparison of dosages of several commonly used biocides used as hard surface sanitizers, i.e., for hospitals, food applications, dental offices, etc. Comparisons are made for dosages required to achieve a five-log reduction in bacterial populations for several strains of bacteria with a 60 second contact time. It is clear from this figure that chlorine dioxide reacts very rapidly, achieving a 99.999% reduction in viable bacterial cells in one minute.

Figure 2. Comparison of the Dosage Required to Achieve a 5-Log Reduction in Viable Bacteria at 60 Seconds

Contact Time

These results confirm those shown in Figure 1 along with those of earlier workers, who demonstrated the very rapid kill of bacterial populations by chlorine dioxide at times much less than the 30-minute period normally used in disinfection studies.

Selectivity

 In potable and waste treatment applications, a number of researchers have commented on the significantly lower demand of the water for chlorine dioxide than for chlorine. An example is shown in Figure 3.

Figure 3. Comparison of the Measured Residual vs. Dosage of ClO2 with Cl2 for a Heavily Contaminated Water.

Equivalent amounts of chlorine dioxide and chlorine were added to water streams with various levels of contamination. This figure shows the results for a highly polluted water stream. Residuals were measured after 30 minutes of contact, and the results were plotted against the initial dosage. The chlorine was largely consumed, regardless of how much was added. Chlorine dioxide, after some initial consumption, remained mostly unreacted. This characteristic is indicative of the much greater selectivity of chlorine dioxide than chlorine.

Chlorine is known to react with a wide variety of compounds. It reacts primarily through oxidation, although it can react by both substitution and addition reactions.

There are many reports of the limited reaction of chlorine dioxide with organics. This indicates that much more of the chlorine dioxide added to a system is available as a biocidal agent and is not consumed to the degree that chlorine would be under the same circumstances.

In addition, chlorine will react with ammonia or any amine, while chlorine dioxide reacts very slowly with secondary amines, and not at all with primary amines or ammonia.

Bacterial Recovery after Disinfection

The phenomenon of rapid regrowth of bacteria in a highly organically loaded system after high chlorination is well established. How rapidly a bacterial population re-establishes itself after sterilization is an interesting phenomenon and is called bacterial recovery. The results of one investigation is demonstrated in Figure 4. This figure shows that after sterilization of a wastewater stream with chlorine, the bacterial population re-establishes itself relatively rapidly. For the same stream sterilized with chlorine dioxide bacterial recovery is somewhat slower.

Figure 4. Recovery of Bacterial Activity After Shock Dosage with ClO2 and Cl2

Relatively, little information on bacterial recovery after disinfection with bromine was found. However, it is likely that recovery after disinfection with bromine would be comparable to that with chlorine. For ozone, due to the rapid oxidation of organics into smaller, more easily metabolized fragments, rapid regrowth of bacteria is expected if the ozone residual is lost.

Effectiveness on Biofilm Control/Removal

The disinfection requirements of an open recirculating industrial cooling system are markedly different from those of a potable water treatment facility. The disinfection goal of potable water facilities is the sterilization of water as measured by specific water borne pathogens. The goal of disinfection for industrial cooling systems is the removal or minimization of any biofilm, which retards heat transfer, causes biofouling, provides a place of agglomeration for marginally soluble or insoluble salts, and provides a place which nurtures and promotes the growth of highly corrosive anaerobic bacteria.

One possible reason for the relatively slow re-growth of bacteria after sterilization by chlorine dioxide (Figure 4) lies in its superior ability to penetrate and disperse a biomass.

By effectively killing and stripping off any biofilm, bacteria are much slower to re- establish than when the biofilm is left intact.

Many researchers have cited the excellent biofilm removing properties of chlorine dioxide. In at least one previously reported case history, the introduction of chlorine dioxide into a heavily fouled cooling system resulted in an increase in both turbidity and calcium. These were explained by a dispersing of the biofilm which both increased turbidity and released small calcium carbonate particulates which had been trapped in the biofilm.

Other industries have made use of the excellent biofilm removal properties of chlorine dioxide, particularly the food industry. Small cooling towers, frequently contaminated by food products or by-products, have tremendous slime forming potential. Chlorine dioxide has achieved widespread usage in such systems, due to its excellent biofilm dispersing/bacterial disinfecting properties.

Effects of System Contamination

Chlorine dioxide has a history of working exceptionally well in systems which are contaminated with ammonia. It has also been very effective in cooling systems with a high level of organic contamination. Figure 5 shows a rough relationship between oxidant demand and increasing system contamination. It is clear from this figure that as system contamination increases HOCl/NaBr becomes more economical than chlorine. As contamination continues to increase, chlorine dioxide becomes economically favored.

Figure 5. General Overview of Relation Economics of Several Oxidants with Increasing System Contamination.

In summary, for ‘stressed’ systems, e.g., systems which are contaminated with hydrocarbons or other contaminants, there is generally much less of a demand increase for chlorine dioxide than for other oxidizing biocides which may be used for microbiological control.

Disinfection Effectiveness Comparison

The question has arisen as to how chlorine dioxide can be so effective at such low dosages. Figure 6 shows an overview of oxidizing biocides used as disinfectants in cooling systems. The left most column, labeled Oxidizing Strength, is the oxidation potential in volts. This parameter describes how strongly the oxidant reacts with an oxidizable substance. Thus ozone, which is the strongest agent listed, has a higher oxidation potential and will react strongly with everything that is oxidizable.

Hypochlorous acid, the active biocidal agent which is formed from the reaction of water and chlorine gas, is weaker than ozone, but stronger than hypobromous acid and chlorine dioxide. Finally, chlorine dioxide is the weakest agent shown. Hypochlorite and hypobromite are shown for comparison, but since these are in equilibrium with their counterparts, these should not be considered as separate entities.

Figure 6. Comparison of Oxidation Potential and Oxidation Capacity of Several Oxidants Along with Relative oxidation of materials commonly found in cooling water.

Oxidation of Materials Commonly Found in Cooling Water.

Column 2 of Fig 6. Below shows the Oxidation Capacity of the various oxidizing biocides. That is, how many electrons are transferred during an oxidation reduction reaction. As the chlorine atom in chlorine dioxide has an oxidation number of +4, chlorine dioxide can accept 5 electrons in an oxidation-reduction reaction, if it is reduced to chloride.

Accounting for molecular weights, chlorine dioxide has 263 % ‘available chlorine’, or slightly more than 2.5 times the oxidizing capacity of chlorine.

Finally, column 3 of fig 6. shows a subjective overview of the “oxidizability” of certain materials commonly encountered in cooling systems, or in other words, how easily these materials are oxidized. As one goes down this table, the materials shown are less easily oxidized. For example, ammonia, known to be oxidized by ozone and hypochlorous acid, reacts reversibility, with bromine and does not react with chlorine dioxide at all.

Wood is another material commonly found in cooling systems which is attacked by ozone and chlorine. Delignification is a well-known phenomenon which occurs by chlorine attack at relatively high pH’s. Chlorine dioxide will react with only the phenolic-type compounds in wood. In other words, chlorine and ozone react with far more of the compounds in wood than does chlorine dioxide.

In summary, Figure 6 shows that of the materials commonly found in open recirculating cooling systems, chlorine dioxide reacts with much fewer than chlorine or ozone, i.e., it is a much more ‘selective’ oxidant than either chlorine or ozone. Thus, when added to a cooling system, far more of the chlorine dioxide is available to react with the intended target, e.g., bacteria and bacterial slimes than either chlorine or ozone.

Fig. 6. Comparison of Oxidation Potential and Oxidation Capacity of Several Oxidants Along with Relative oxidation of materials commonly found in cooling water.

Environmental

Formation of Trihalomethanes

In the mid to late 70’s researchers linked the chlorination of potable water to increased cancer mortality rates. This increase in cancer mortality was related to increased levels of trihalomethanes, THM’s, primarily chloroform. Researchers found that chlorine reacted with certain naturally occurring organics to produce chloroform. As a result of this work, the USEPA set 0.1 ppm as the maximum THM containment level for drinking water.

Subsequently, research in the area of THM reduction in potable water led to the EPA in 1983 citing chlorine dioxide as an effective means of controlling THM’s. Practical application has shown a significant reduction or absence of THM’s in systems treated with chlorine dioxide. And although THM production has been considered predominantly in the drinking water industry, the production of chloroform is seeing increased pressure from environmental regulators in industrial plants.

Economics

The properties discussed previously make chlorine dioxide an ideal choice as a microbicide for industrial and municipal use. With the pressure to eliminate gaseous chlorine, chlorine dioxide costs become favorable to many of the conventional chlorine replacements. In fact, for heavily contaminated systems, the use cost of chlorine dioxide can rival that of chlorine. The relative costs for common chlorine alternatives are similar to that reported previously, with some notable exceptions.

For clean systems:

pH 6.8 – 8.0: Cl2 < NaOCl < HOCl < NaBr < ClO2 < BCDMH < ozone

pH 8.0 – 9.3: HOCl < NaBr < ClO2 < BCDMH < Cl2 < NaOCl < ozone

For systems with high organic loading:

pH 6.8 – 9.3: ClO2 < HOCl < NaBr < BCDMH < Cl2 < NaOCl < ozone

For systems with ammonia contamination:

pH 6.8 – 9.3: ClO2 < HOCl < NaBr < BCDMH < Cl2 < NaOCl < ozone

Summary

The properties of an ideal microbicide have been described. The commonly used oxidizing biocides, chlorine, bromine, chlorine dioxide and ozone have been reviewed in the light of these properties. The results are summarized in Figure 7.

Figure 7. Relative Rankings for Chlorine Bromine, Chlorine Dioxide in View of Characteristics of an Ideal Biocide.

Although these relative rankings are somewhat subjective, the rankings are based on data from literature. Grades have been assigned to each category, and a cumulative grade point average has been computed. It should be clear from this that although each of the oxidizers listed excel in one or more areas, when reviewed as a whole, chlorine dioxide comes closer to achieving the status of ‘ideal’ biocide than any of the others.

Conclusions

The short-term driving force for the move away from gaseous chlorine to the various alternatives is predominantly the health and safety issues relating to the handling, storage and integrity of one-ton cylinders. If the recent trend continues, the long-term force driving the move away from gaseous chlorine will be the reaction products between chlorine and organics, and the impact that these chemicals have on the environment.

We have seen a move into bromine chemistry, although this will be undoubtedly a relatively short-term measure for the same reason if the trend continues. Not much work environmentally has been done with brominated organics, although they are expected to be more harmful than structurally similar chlorinated organics.

Ozone as a viable long-term replacement for chlorine is being explored, due to its excellent biocidal activity. However, ozone does suffer from several serious drawbacks.

Because of its strong oxidizing ability, it will not differentiate between the target organisms and the corrosion inhibitor packages used with it to control scale and corrosion.

Its effectiveness on biofilm control is questionable.

If the industry trend continues, the useful life of hypohalous acid-based oxidizing biocides will be limited. With the additional factors of increasing water reuse, i.e., increasing system contamination, and regulatory requirements of non-oxidizing biocide re- registrations, i.e., decreasing availability of non-oxidizing biocides, the need for a high- performance chlorine replacement is becoming increasingly apparent. The superior performance characteristics of chlorine dioxide coupled with the mobile feed system makes chlorine dioxide the ‘ideal’ biocide.