Efficiency of Chlorine Dioxide as a Bactericide1
Abstract
We found chlorine dioxide to be a more effective disinfectant than chlorine in sewage effluent at pH 8.5. Chlorine dioxide was also found to be a more stable bactericide in relation to pH in the range studied.
Full text
Full text is available as a scanned copy of the original print version. Get a printable copy (PDF file) of the complete article (1.2M), or click on a page image below to browse page by page. Links to PubMed are also available for Selected References.
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC1058342/pdf/applmicro00361-0144.pdf
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC1058342/pdf/applmicro00361-0144.pdf
Efficiency of Chlorine Dioxide as a Bactericide
MELAIN A. BENARDE, BERNARD M. ISRAEL, VINCENT P. OLIVIERI,AND
MARVIN L. GRANSTROM Bio-Engineering Laboratory, Department of Civil
Engineering, Rutgers, The State University, New Brunswick, New Jersey Received
for publication 17 May 1965
ABSTRACT
BENARDE, MELVIN A. (Rutgers, The State University, New
Brunswick, N.J.),
BERNARD M. ISRAEL, VINCENT P. OLIVIERI, AND MARVIN L.
GRANSTROM. Efficiency of chlorine dioxide as a bactericide. Appl. Microbiol.
13:776-780. 1965-We found chlorine dioxide to be a more effective disinfectant
than chlorine in sewage effluent at pH 8.5.
Chlorine dioxide was also found to be a more stable
bactericide in relation to pH in the range
studied.
Text:
Chlorine dioxide, prepared early in the 19th century, was
used in the treatment of water supplies in Europe after 1850. It was not until
the 1940's, however, that experimental data on ist bactericidal efficiency
became available. Although chlorine and chlorine dioxide are similar in many respects,
including the fact that both are powerful oxidizing agents, Cl02 has 2.5 times
the oxidation capacity of Cl2. It was this feature that recommended C102 for
the control of odors and tastes in water supplies. McCarthy (1944) reported
C102 to be an effective germicide in water with low organic content. However,
he found it to be less effective than chlorine at equal concentrations when the
organic content was high.
Shortly thereafter, Ridenour and Ingols (1947) reported C102
to be at least as effective as chlorine. Their conclusions were based upon the
absence of growth after 30 min of contact with the disinfectant. Efficiency was
based on residual values, and these did not represent equal initial
dosages. Trakhtman (1946) and Bedulevich, Svetlakova, and
Trakhtman (1953) found C102 to exceed or at least equal chlorine in
bactericidal efficiency. They reported decreased effectiveness of C102 under
alkaline conditions.
Behause physicochemical characteristics and kinetics of C102
were unavailable at the time these studies were performed, the data obtained are
of questionable value.
Preparation of chlorine dioxide from chlorine, or by the
action of acid on NaCl02, as is the genlPresented in part at the 65th Annual
Meeting of the American Society for Microbiology, Atlantic
City, N.J., April, 1965. eral procedure, would of necessity
introduce interfering substances which would not only hasten the decomposition
of C102, but would also yield erroneously high values on iodometric analysis. This
would suggest that the initial concentrations of C102 in the studies noted were
probably lower than the reported values; the bactericidal efficiency of C102,
when compared with chlorine, would suffer accordingly. Preparation of C102 by addition
of alkali results in major losses of C102 almost instantaneously. Granstrom and
Lee (unpublished data) have shown the disproportionation that occurs above pH
10. Additionally, these older methods probably did not account for the high
degree of volatility exhibited by C102. Depending upon the concentration and
length of exposure, 7 to 30% loss can occur within 1 hr. This means that
dilutions prepared from stock solutions cannot be assumed to be quantitative.
Spectrophotometric analysis of each dilution immediately prior to use is
required. Such a procedure has only recently become available.
The general analytical procedure for detection of C102
residuals employed o-tolidine in a colorimetric determination. This procedure
is subject to similar interferences, as noted for the iodometric titration.
Post and Moore (1959) reported that o-tolidine and o-tolidine arsenite methods
made no distinction between Cl2 and C102. Consequently, it
is nonspecific and allows only a range of values for any given concentration. Thus,
critical comparison of chlorine versus chlorine dioxide was impossible. As a
result, C102 is sparingly used in water and waste water treatment. The study
reported herein used as its point of departure the detailed physicochemical
findings of Granstrom and Lee (1958) and Granstrom et al. (uinpublished data)
to avoid the pitfalls of the past and to obtain more reliable and specific
data.
MATERIALS AND METHODS
Organic-free distilled water. Organic-free water was
prepared by distilling distilled water in the
presenice of an acid-permanganate solution (1%concentrated
H2SO4 and 1% KMnO4). The distillate traveled through a vertical column 6 ft (183
cm) long. It was necessary to use a heating tape at the top of the column to
break the film of water from the distillation flask; otherwise, creeping of the
permanganate occurred and would appear in the distillate. Extremely small
amounts of permanganate can interfere in the o-tolidine determinationi of
chlorine by giving itself a positive test (yellow color). This, however, provides
a means of checking the quality of the distillate
(Israel, 1961). Organic-free water was used in preparing the
phosphate buffers needed for the various trials. Glassware. To obtain chlorine
demand-free glassware, tubes, pipettes, reaction vessels, and syringes were
soaked overniight in chlorine water of approximately 5,000 mg per liter. Prior
to use, these were rinsed several times in distilled water and heat-sterilized
when necessary.
Neutralization of disinfectant. Tubes (16 by 150 mm)
containiing several crystals of Na2S203 were sterilized at 121 C for 15 min.
These were used for quenchinig further disinfection action of both chlorinie
anid chlorine dioxide solutions.
Preparation of chlorine dioxide. A 4.0-g amount of NaCO1 was
dissolved in 50 ml of distilled water in a reaction vessel. To this was added a
solution of 2.0 g of K2S208 in 100 ml of distilled water. High-purity nitrogen
gas (Linde Division, Union Carbide Corp., New York, N.Y.) was used to sweep out
the C102 formed; this N2-C102 gas mixture was passed through a dry NaCl02
column to remove any traces of HOCI that might have been formed. The gases were
then passed into a trap to remove any NaCl02 dust that might have been carried
over. C102 was then collected in cool (O to 10 C) organic-free water.
Preparation of chlorine. Aqueous chlorine stock solutions
were prepared by bubbling chlorine gas into distilled water for 15 to 20 min.
This yielded a 10,000 mg per liter concentration.
Concentrations of C102 and Cl2 were prepared on an
equivalent chlorine basis: 1 mg per liter of chlorine is equal to 1.4 X 10-s
moles per liter; 1 mg per liter of C102 is equal to 1.5 X 10-6 moles per liter.
For this study, C102 was prepared to equal 1.4 X 10-1 moles per liter. Thus,
although equal in molar concentration, C102 was low in per cent concentration.
Since the high molar absorbtivity of C102 lends itself to spectral analysis, we
were able to measure the initial and residual concentrations of C102 directly
at 357 m,u. This was not the case with chlorine, whose molar absorbtivity is approximately
one-tenth that of C102. Although we were able to prepare stock solutions of
chlorine of 20 ppm, using the spectrophotometer (238 m,u), the very low initial
coincentrations used in this study could not be so measured. For this reason, dilutions
of chlorine were prepared in organic-free water, since it had been shown in our
laboratory that quantitative dilutions of C12 could be made. For this study,
initial dosages of both chlorine and chlorine dioxide of 0.25, 0.5, 0.75, 2.0,
and 5.0 mg per liter were used. Suspensions of washed cells (freashly isolated fecal
strain of Escherichia coli) were mixed by magnetic stirrer in a reaction vessel
with an equal volume of disinfectant solution to yield the final desired
concentrations of C12 or C102, or both, and a biacterial density of
approximately 19,000 cells per milliliter. The density of washed cells had been
standardized spectrophotometrically by. opticaldensity measurement at 525 mp.
Confiraation of this was obtained by spot-plate counts onr nutrient agar, using
a Unopette (Becton, Dickinson and Co., Rutherford, N.J.) for delivery of 0.025
ml. We found this technique to be more accurate and reproducible, and a great
deal easier, than standard plate counts. Figure 1 shows the type of spot-plate counts
obtained. The sampling apparatus consisted of two automatic hypodermic
syringes, each with a two-way flap valve assembly, permitting rapid drawing and
dispensing of the solution from the reaction vessel. Figure 2 shows the
apparatus ready for use. At 15-sec intervals, samples were removed for
estimation of remaining population. Sterile thiosulfate crystals were used to
quench further disinfectant effects. In the case of chlorine, residuals were determined
by the o-tolidine method. Chlorine dioxide residuals, however, were determined spectrophotometrically.
A 30-ml sample was withdrawn from the reaction vessel and placed in a 10-cm
cuvette. Absorbance values at 357 mi were obtained at intervals, and were used
to calculate concentrations.
Plate counts were made on nutrient agar, after 18 and 48 hr
of incubation at 35.5 C. To prevent clumping and spreading of colonies, it was
necessary to add the sample to a layer of melted agar and swirl. Between
trials, the sampling apparatus was sterilized by drawing a strong chlorine
solution into it and allowing it to remain in contact for 10 min. Rinsing with
sterile organic-free distilled water was repeated until negative o-tolidine tests
were obtained. Sterility was determined by plate counts of the rinse water.
This procedure additionally provided a chlorine demand-free
system.
RESULTS AND DiscussioN
It is generally understood that the germicidal activity of
chlorine results from its hydrolysis in aqueous solutions to form hypochlorous
acid, the disinfectant constituent:
Cl2 + H20 =
HOCI + H+Cl- (1)
hydrolyze in aqueous solutions and, therefore, is not
subject to dissociation in the manner of HOCI (Mellor, 1922). A stock solution
of 10.2 mg per liter of Cl02 was diluted and its absorbance at 357 m,u was
measured. At pH 4.0, 6.45, and 8.42, the C102 was unaltered (1.39 mg per liter,
100%). Thus, the intact Cl02 molecule appears to be the bactericidal compound. Figure
3 shows the effect of pH on the disinfectant efficiency of chlorine and chlorine
dioxide in organic-free buffer. It can be seen that at pH 6.5 chlorine was
somewhat more efficient than Cl02. Based on our sampling times of 30 and 60 sec,
it would appear that half the time is needed for equal kill at an initial
dosage of 0.25 mg per liter. At 0.5 mg per liter, killing time was reduced, but
chlorine remained more effective. Both compounds were equally efficient at an
initial dose of 0.75 mg per liter. Increasing the pH to 8.5 altered the
comparative disinfectant properties dramatically. It is seen that at 0.25 mg
per liter Cl02 obtained 99+% kill in 15 sec. Chlorine did not achieve this
until 300 sec-20 times less effective. At 0.5
mg per liter of Cl02, kill was also completed within 15 sec,
whereas chlorine requires 60 sec to achieve equal effects. Equal kill occured
at 0.75 mg per liter for both compounds.
From Fig. 3 it would seem that pH affects C102 efficiency.
Our data and the data of Mellor (1922)
and Granstrom et al. (unpublished data) indicate that Cl02
is not directly affected by pH in aqueous solutions; other factors are probably
contributing to the loss of efficiency. Although the chemical reactions of Cl02
with many materials are yet to be studied, it may well be that the rates of reaction
of Cl02 with substances found in our system are pH-dependent.
To establish the comparative bactericidal efficiencies that
would more nearly resemble inplant conditions, tests were conducted in sterile,
unchlorinated sewage effluent to which a known cell density of E. coli was
added. Since there were materials in the effluent that absorbed at 357 m,u, spectral
analysis of Cl02 residuals, in this instance, was impractical. Thus, both
chlorine and Cl02 residuals were determined by o-tolidine.
Figure 4 presents the results of the effluent study. It can
be seen that the combination of organic matter and pH sharply curtails
disinfectant activity. For initial Cl2 dosages under 1 ppm, less than 30% of
the initial population is removed. At the same initial dosages, chlorine
dioxide achieves up to 70% removal. An even more dramatic example of the
relative efficiencies of the two compounds is seen with chlorine at 5.0 ppm,
which obtains 90%/ reduction after 5 min, compared with chlorine dioxide at 2.0
ppm, which obtains approximately 100% kill in 30 see. pproximately 100% kill in
30 see.
Generally, reports of chlorine efficiency are stated in
terms of residual chlorine values. We found that residuals are not in
themselves indicative of the concentration required for kill. Similarly, the
initial concentration is not the killing dose. We feel that the actual value
lies between the two. The amount of material used is probably closer to the
true killing value.
Figure 5 presents residual values obtained for chlorine and
chlorine dioxide in the effluent studies. For complete removal of organisms in
30 see, about 0.9 ppm of Cl02 was used when dosed with an initial concentration
of 2.0 ppm. In the case of Cl2, 2.25 ppm were actually used of an initial dose of
5.0 ppm to obtain 90%/o removal in 5 min. Our data indicate that, even though
residuals are present after 5 min, major reductions occur within the first
minute of contact and do not occur appreciably thereafter.
This would suggest that residuals actually have little
disinfectant value. Thus, reports of killing concentrations based on residual
values seem unrealistic. It should be borne in mind that the plateau effects in
the case of chlorine are probably due to the formation of monochloramines (pH
8.5) and other chloro compounds which are of little value as disinfectants. In
the case of chlorine dioxide, however, reaction with ammonia to form
chloramines has not been reported.The fact that 100% kill has not been obtained
with residuals up to 0.5 mg per liter, as measured by o-tolidine, which is
nonspecific, would suggest that CIO2 undergoes reaction to form organo-chloro
compounds that have little disinfectant ability.
Additionally, Fig. 5 shows that less chlorine dioxide was
used to obtain greater bacterial reductions. This may be due to fewer
side-reactions at pH 8.5 that tie up the essential disinfecting material. As
fewer molecules of chlorine dioxide were present initially, 1.4 X 10-5 as
opposed to its usual 1.5 X 10-s moles per liter, the results would tend to
support even more strongly the conclusion that Cl02 exhibits greater bactericidal
activity than chlorine.
ACKNOWLEDGMENT
This investigation was supported by Public Health Service
grant PHT 1-61A-61.
LITERATURE CITED
BEDULEVICH, T. S., M. N. SVETLAKOVA, AND N. N. TRAKHTMAN.
1953. New data on the use of chlorine dioxide in water purification. Gigiena i Sanit.
19:14-17.
GRANSTROM, M. L., AND G. F. LEE. 1958. Generation and use of
chlorine dioxide in water treatment. J. Am. Water Works Assoc. 50:1453-1466.
ISRAEL, B. M. 1961. Hydrolysis of organo-halogenating
agents. Ph.D. Thesis, University of Wisconsin, Madison.
MCCARTHY, J. A. 1944. Bromine and chlorine as water
disinfectants. J. New Engl. Water Works
Assoc. 58:55-68.
MELLOR, J. W. 1922. Comprehensive treatise on inorganic and
theoretical chemistry. Longmans,
Green & Co., New York.
MOORE, E. W. 1951. Fundamentals of chlorination of sewage
and waste. Water Sewage Works
98:No. 3.
POST, M. A., AND W. A. MOORE. 1959. Determination of
chlorine dioxide in treated surface
waters. Anal. Chem. 31:1872-1874.
RIDENOUR, G. M., AND R. S. INGOLS. 1947. Bactericidal
properties of chlorine dioxide. J. Am. Water Works Assoc. 39:561-567.
TRAKHTMAN, N. N. 1946. Chlorine dioxide in water
disinfection. Giegiena i Sanit. 11:10-13
