The present status of research on the disinfection of drinking water in the USSR

Bull. Org. mond. SantiE 1972, 46, 277-283 Bull. Wld Hlth Org. The present status of research on the disinfection of drinking water in the USSR S. N. tERKINSKIJ 1 & N. TRAHTMAN 2 The article reviews recent research in the USSR aimed at evaluating methods of disinfecting drinking water and at elucidating the mechanisms involved. The use of chlorine, ozone, and gamma rays is discussed, as are their advantages and disadvantages and their effects on Enterobacteriaceae and on enterovirutses. Data on the distribution of communicable diseases throughout the world, including the USSR (Baroyan, 1967, 1968), indicate that disinfection, in order to prevent the transmission of intestinal disease, is still one of the main tasks involved in the provision of community water supplies. This paper reviews the research carried out in the USSR during the last few years to evaluate new and improved methods of disinfecting drinking water and to elucidate the mode of action of disinfectants on Enterobacteriaceae and enteroviruses. A similar review was compiled by (erkinskij et al. (1969). CHLORINATION Up to now, chlorination has been the most widely used method of disinfecting drinking water in the USSR. Moreover, the recent development of a me- thod for obtaining chlorine in the waterworks them- selves, by the electrolysis of sodium chloride, has made it possible to eliminate the difficulties involved in transporting and storing large quantities of liquid chlorine. Electrolytic equipment developed in the K. D. Panfilov Academy of Community Services, Moscow, employing finely divided magnetite or graphite as electrodes, has proved reliable under pilot produc- tion conditions (Maslennikov, 1963). Research has shown that under all operating conditions the main product of electrolysis is sodium hypochlorite (Ser- gunina, 1968). The redox potential of the products of electrolysis is in no way inferior to that of chlorine, which means that these products are effective bac- 1Head, Department of Community Hygiene, First Moscow Medical Institute, USSR. ' Department of Community Hygiene, Central Institute for Advanced Medical Studies, Moscow, USSR. Table 1. Percentage disinfection of tap water contain- ing Escherichia coli and Shigella sonnei, with various concentrations of the products of electrolysis of sodium chloride Concentration of electrolysis products Strain and initial (mg/litre) concentration 0.3 0.5 0.7 E. coil 99 000/litre 95.0 99.76 99.99 S. sonnei 110 000/litre 96.7 99.90 100.0 tericidal and virucidal agents (Table 1). On the basis of laboratory and field research, recommendations were drawn up on the disinfection of drinking water by the above method. A further improvement in the chlorination of drinking water is the trend towards ensuring the presence in the water of so-called free residual chlo- rine (i.e., chlorine in the form of hypochlorous acid), as distinct from combined chlorine, which is bound to ammonium ions to form chloramine. Theoretically, the nature of the chlorine-combin- ing reaction has long been known. However, special attention has been paid recently to the effect of available chlorine under conditions of practical dis- infection of water at waterworks. Experiments have shown a considerable difference between the effective- ness of free chlorine and that of combined chlorine against Enterobacteriaceae and enteroviruses. The experimental work of Gubar' & Kozlova (1967) confirms that free chlorine is far more effective 2802 -277- S. N. tERKINSKIJ & N. TRAHTMAN Fig. 1. Effectiveness of free chlorine (solid line) and combined chlorine (broken line) at a concentration of 1 mg/litre for disinfecting water containing E. coli. against Enterobacteriaceae than is combined chlo- rine (Fig. 1). A comparison of data on the inactivation of vi- ruses, Escherichia coli, and coliphage in practically identical initial concentrations by different levels of residual chlorine (Fig. 2) indicates that higher levels of residual chlorine and longer periods of contact are required to free water from viruses than to destroy either E. coli or coliphage (Lovcevic & Ser- gunina, 1968). A similar assertion in regard to the infectious hepatitis virus was made by Chang (1968). So far, these investigations have given grounds for assuming that E. coli and coliphage cannot serve as reliable indicators of the effectiveness of chlorine against enteroviruses. However, this conclusion has not gone unchallenged and research on the problem is continuing. In particular, a search is being made for indicator organisms with which to evaluate the virucidal effect of chlorine. In discussions on the use of chlorine for the disinfection of drinking water, it is quite common for its alleged toxicity to be listed among its short- comings and for doubts to be expressed as to the harmlessness of the residual amounts of chlorine. Some authors have considered even small concentra- tions of residual chlorine (from 0.5 mg/litre upwards) to be harmful (Muhin, 1935). Others, however, have not found it to have any adverse effect (Druckrey, 1968). Citovic (1940) refuted the possibility of chlo- rine being absorbed when introduced into the intes- tine or even directly into the blood. In trials on volunteers, Gubar' et al. (1970) estab- lished that the human oral mucosa has a high capa- city for absorbing residual chlorine from water, and that it is highly resistant to the irritant effect of residual chlorine on long-term exposure to concen- trations greatly exceeding those necessary to achieve a bactericidal effect. Histopathological experiments on animals over a 6-month period established that there was no func- tional irritation of the oral or gastric mucosa. The results of these investigations have led to the con- clusion that a concentration of residual chlorine of up to 2.5 mg/litre is practically harmless, and that at a concentration corresponding to the standards based on organoleptic criteria (not exceeding 0.5 mg/ Fig. 2. Course of inactivation of E. coli (solid line), coliphage (dotted line), and poliovirus (broken line) in water (pH 8.3) containing (A) combined residual chlorine at a concentration of 0.4-0.6 mg/litre, (B) total residual chlorine at a concentration of 0.5-0.8 mg/litre together with traces of free residual chlorine, and (C) total residual chlorine at a concentration of 0.9-1.2 mg/litre and free residual chlorine at a concentration of 0.1-0.5 mg/litre. In each figure the arrow indicates the point at which micro-organisms were no longer isolated. 100 101 01 3E o 0.1 % o 0.01 -~0.001 I >~~~~~~~~ 0.0001 mm 0 5 30 120, Duration of contact with chlorine (min) WHO 20021 278 RESEARCH ON DISINFECTION OF DRINKING WATER IN THE USSR litre) there is absolutely no possibility of free residual chlorine entering the body, on account of the neutral- izing effect of the upper part of the alimentary tract. OZONATION There is now no doubt that the most promising method of water treatment for disinfection purposes is ozonation. From the public health point of view this method has substantial advantages. Firstly, as a result of its high redox potential ozone has a stronger bactericidal effect than other chemical agents and is therefore capable of oxidizing many organic sub- stances that give water colour, odour, and taste; and secondly, an excess of ozone does not denature water as does an excess of chlorine. Operational experience at several waterworks in the USSR has shown that the use of ozonation can sometimes simplify the engineering layout of the water treatment plant, since it makes it possible to disinfect, decolorize, and deodorize the water simul- taneously. The most economical method of treating water with ozone in order to obtain a stable water quality from the bacteriological point of view is to ozonate it after clarification in settling tanks and by filtration. When this is done the mean annual dose of ozone required for water without much colour does not exceed 2.0-2.5 mg/litre (Vahler, 1963). x0 .' E 0o _ u lo0 20 40 60 80 100 120 140 160 180 200 220 240 260 Time (seconds) wHo 20023 Fig. 3. Effect of turbidity on the disinfectant action of ozone. Curve A is for water without added kaolin; the other curves are for water containing kaolin at the following concentrations: B, 5 mg/litre; C, 10 mg/litre; D, 50 mg/litre; E, 100 mg/litre. A number of investigations aimed at assessing the public health value of the ozonation of drinking water have revealed several features that are of great importance in increasing the effectiveness of disinfec- tion. Thus, it has been established that, in water treated with a given dose of ozone, the lower the temperature the fewer the viable bacteria that remain (Vrocinskij, 1963). If the bactericidal dose of ozone at a water temperature of 4-60C is unity, then the corresponding dose at a temperature of 18-21°C is 1.6 and at 36-380C is 3.2. Thus, during the warm part of the year, when the water temperature is higher, the dose of ozone must be considerably increased in order to ensure the reliability of dis- infection. This is evidently caused by a decrease in the amount of ozone that is utilized, as a result of its reduced solubility in the water. Turbidity of up to 5 mg/litre has a negligible effect on the efficiency of water disinfection by ozone, but at higher degrees of turbidity the bactericidal effect is considerably reduced (Fig. 3). An increase in the pH of the water necessitates a higher concentration of ozone for effective disinfection. An increase in colour of I unit raises the bactericidal dose of ozone by 0.1-2 mg/litre. Experimental research on the rate of survival of various strains of typhoid bacilli and E. coli (Suckov, 1964) have shown that the latter is highly resistant to ozone (Fig. 4). 4 a ik WH0 2OO29 Ozone dose (mg/litre) Fig. 4. Destruction of E. coili (solid line) and of typhoid bacilli (broken lines) following ozonation of drinking water. The initial concentration of E. coili was 10 times that of the typhoid bacilli. El - 279 S. N. 6ERKINSKIJ & N. TRAHTMAN Fig. 5. Relationship between the concentration of re- sidual ozone and its disinfectant efficacy: river water containing polioviruses (solid line) and coxsackieviruses (broken line). The course of inactivation of enteroviruses when water is ozonated indicates that the virucidal effect of ozone is linked to the duration of ozonation, the concentration of residual ozone, and the concentra- tion of impurities. Suckov (1964) infected purified and sterilized river water with centrifuged virus- containing liquid. The concentration of active viruses reached 5 x 103 TCD50/ml, which is many times the possible concentration in water under natural condi- tions. Together with the virus-containing liquid, or- ganic substances were introduced into the water; these increased its oxidizability more than ten-fold, so that the concentrations of ozone required in the experiments considerably exceeded those ordinarily employed in waterworks. The results of the experi- ments (Fig. 5) show that during the first 5 minutes there was no inactivation of the viruses, since the ozone was used up completely in oxidizing the or- ganic substances, as evidenced by a decrease in the oxidizability of the water. After 7 minutes of ozona- tion, 90% of poliovirus type 3 and 97% of cox- sackie B virus type 3 had been inactivated; after 15 minutes these figures had risen to 99.7% and 99.9%, respectively. The research carried out has led to the conclusion that ozonation is in general a good method of disinfecting water. GAMMA IRRADIATION In public health research and practice, preference is given to physical methods of disinfection, whose use does not entail the introduction into the water of agents that may result in a change in its chemical composition. It is therefore natural that interest was aroused in the possibility of using such a powerful physical factor as ionizing radiation for the disinfec- tion of drinking water. The effectiveness of this method was established by Rjabcenko (1964). A study of the rate of destruction of Enterobacteriaceae exposed to gamma radiation has shown that in any medium the process proceeds exponentially. Rjab- cenko (1964) found that 90% of the bacteria die at relatively low doses of radiation-i.e., 10 000-15 000 rad (Fig. 6). Any further increase in the degree of disinfection requires a considerable increase in the dose of gamma radiation in order to destroy the few remaining resistant micro-organisms. Shigella is the most sensitive organism to gamma rays, followed by typhoid and paratyphoid bacilli. Pathogenic coliform bacilli show a resistance similar to that of typhoid and paratyphoid bacilli, whereas typical E. coli is the most resistant organism. A study of the process of inactivation of poliovirus type 1, echovirus type 7, and E. coli showed that the latter were killed more quickly and at lower doses of gamma radiation than were the viruses (Rjab6enko & Lovcevic, 1965). The disinfection of water con- taining enteroviruses requires 2-3 times the dose of Fig. 6. Disinfection of water containing enteric bacte- ria by gamma irradiation. Solid lines, saprophytic bac- teria; broken lines, pathogenic bacteria; A, Shigella sonnei 714; B, Aerobacter aerogenes 1324; C, Salmo- nella typhi; D, Escherichia coil var. communis 39/2; E, Salmonella paratyphi B 627; F, E. coi 675; G, E. coi 734. 0 o 99.7 09 i 3 5 7 10 15 Duration of ozonation (min) trace 0.1 0.2 Concentration of residual ozone (mg/litre) *HO 20025 1 . 0 01. I I I I 1 05 a.001 .90.0X xO0.05 -- a. a. 0* ,=0.005 -- 9.00., u ~~~~~~~~~~~~~~0 50.0015 C A' ,E~~~CI. -0 -oF 0.0001 -99 -9--i- 0.00005------ - 0.00001 -2500 10000 20000 30000 40000 50000 Radiation dose (rad) WHO 20026 280 RESEARCH ON DISINFECTION OF DRINKING WATER IN THE USSR radiation needed to disinfect water containing bac- teria in a similar concentration. It has also been shown that the disinfectant pro- perties of gamma radiation are not affected by the presence of iron impurities and salts producing hard- ness. Neither are they affected by the degree of turbidity or colour, even when these are several times the normal permissible levels in drinking water. Thus experiments have confirmed that gamma radia- tion can be used, and is relatively highly effective, for the disinfection of water. MECHANISM OF ANTIBACTERIAL ACTION Together with a public health evaluation of the efficacy of modified or new methods of disinfecting drinking water, investigations have been carried out on the mechanism of antibacterial processes. Modern research methods using electron micro- scopy have shown that each cell component is adapted to a particular function. At the same time, investiga- tions of the chemical activity of substances obtained from disintegrated cells have made it possible to determine the biochemical reactions that occur in the cell and that underlie its vital processes. Concepts of the submicroscopic structure of the cell have now been confirmed in practice, and this has determined the direction taken by public health research on the mode of action of disinfectants on bacteria. Studies have related in particular to the effect of chlorine, ultraviolet rays, ozone, and gamma radiation on the process of biological oxidation in the cell, as estimated from dehydrogenase activity (Trahtman, 1958; Lipinska, 1963; Vrocinskij, 1963; Suckov, 1964; lerkinskij & Rjabcenko, 1968; Ski- dal'skaja, 1969). The activity of the dehydrogenases in representa- tives of the Enterobacteriaceae was determined on the basis of colour changes in the dye methylene blue, resulting from its reduction, under anaerobic conditions, by hydrogen liberated from a substrate by the action of a dehydrogenase. At the same time the degree of bactericidal effect was determined. It has been shown that the degree of inhibition of dehydrogenase activity is directly related to the chlo- rine concentration. The activity of glucose and glyce- rol dehydrogenases is inhibited completely even when a considerable number of bacteria are still viable. Formaldehyde and alcohol dehydrogenases are even less resistant to chlorine. The above-mentioned enzymes are more resistant to gamma radiation than to chlorine; massive doses of radiation (1 000 krad/min) inhibited the dehydro- genase activity of E. coli by not more than 60-70% (Fig. 7). The activity of Shigella sonnei enzymes following exposure to chlorine and to gamma rays was inhibited somewhat more than that of the same dehydrogenases of E. coUl. Thus, chlorine and gam- ma radiation produce their bactericidal effects by different modes of action on the enzymes that cata- lyse oxidation-reduction processes. The effects of chlorine and gamma radiation on the decarboxylation of amino acids have been as- sessed from changes in the activity of the enzymes concerned (decarboxylases), by identifying the amino acids and their corresponding decarboxylation pro- ducts after separating them by paper electrophoresis. It was established that enzyme activity is reduced Fig. 7. Effect of gamma radiation on the survival of Escherichia coil (A) and on the activity of its dehydrogenases (B, glucose; C, ethanol; D, formaldehyde; E, glycerol; F, formic acid; G, succinic acid; H, lactic acid; 1, glutamic acid). 120 0 Control Control100 0 F 100 200 300 400 500 *1000 100 200 300 400 500 1000 Radiation dose (krod/min) WHO 20027 281 S. N. 6ERKrNSKIJ & N. TRAHTMAN Table 2. Effect of gamma radiation on the base ratios in DNA of Escherichia coil Dose of Amount of base in DNA (in mol %) Ratios gamma G radiation Guanine Adenine Cytosine I Thiamine Purine G/C A/T G+C(kR) (G) (A) (C) (T) Pyrimidine A+G 50 25.17 22.25 24.95 27.33 0.91 1.01 0.82 1.00 300 25.79 24.78 22.39 26.04 1.06 1.19 0.95 0.95 1 000 29.61 23.01 22.76 24.62 1.11 1.30 0.93 1.10 Control 25.52 23.82 26.88 23.78 0.97 0.95 1.00 1.00 following exposure to chlorine but not following exposure to gamma radiation. Studies of the effects produced by various concen- trations of active chlorine lead to the conclusion that the respiratory enzymes of bacteria are more sensi- tive to chlorine than are those that are involved in the breakdown of amino acids. It has been shown, using partition paper chromato- graphy with subsequent spectrophotometric determi- nation of nucleotide spots, that exposure of E. coli to chlorine has no effect on the nucleotide composition of its DNA. On the other hand, a study of the base ratios in the DNA of irradiated E. coil showed that the DNA molecules disintegrate on exposure to gamma rays (Table 2). Electron microscopy of E. coil exposed to chlorine has shown that the blocking or complete inactivation of enzymes is accompanied by varying degrees of cytological change (reduction in the size of cells, coagulation of protoplasm, damage to cell walls). By studying the effects of disinfectant substances and agents on the enzymatic activity of bacteria, it has been shown that chlorine and ozone have a similar mode of action, whereas gamma rays act differently. The effects of chlorine and of gamma radiation on the structure of bacterial DNA have also been shown to differ. In the case of chlorine the main effect is a disturbance of the enzymatic processes, whereas fol- lowing exposure to gamma radiation the most marked changes occur in the structure of thebacterial DNA. The above investigations throw some light on the mechanism of antibacterial action of a number of substances and agents. However, no similar work has yet been carried out in regard to viruses. At the same time, experiments at the level of bacterial DNA are opening up prospects for studying the mechanism of the virucidal effect also. CONCLUSION Although active research is currently being carried out on the disinfection of drinking water, many problems remain to be studied, the nature of which has to some extent been outlined above. The study of all aspects of the virucidal effect of water disinfection, by methods that are already in use or that may be adopted in the future, is of paramount importance. The next task should be research aimed at the public health evaluation of new methods when these are put into practice on a large scale. RE-SUME' ETAT ACTUEL DE LA RECHERCHE SUR LA DESINFECTION DE L'EAU DE BOISSON EN URSS Les auteurs passent en revue les recherches menees recemment en URSS en vue d'evaluer des methodes nou- velles et plus efficaces de traitement de l'eau de boisson et de preciser le mecanisme d'action des desinfectants sur les enterobacteries et les enterovirus. La chloration connait un regain de faveur, grace a la mise au point d'un procede permettant d'introduire des derives chlores, obtenus par electrolyse, directement dans les reseaux de distribution d'eau. Le pouvoir oxydo- reducteur de ces produits, et leurs effets bactericides et virulicides, ne le cedent en rien a ceux du chlore. Pour detruire les virus, on doit recourir 'a des concentrations de chlore residuel plus elevees et a une duree de contact plus longue que pour 6liminer Escherichia coli et les coli- 282 RESEARCH ON DISINFECTION OF DRINKING WATER IN THE USSR 283 phages. Des essais sur des volontaires et l'experimenta- tion sur l'animal ont montre qu'une concentration de chlore residuel ne depassant pas 2,5 mg/I, n'a pratique- ment aucun effet nocif; si la concentration est infdrieure A 0,5 mg/I, le risque de p6ne'tration du chlore residuel libre dans l'organisme est absolument nul, en raison de sa neutralisation dans les portions hautes du tractus digestif. La methode la plus prometteuse pour la desinfection de l'eau de boisson est l'ozonisation. Outre son action sterilisante, l'ozone a pour effet de debarrasser 1'eau des gouts et des odeurs desagreables. On doit tenir compte de certains facteurs qui peuvent influer sur I'efficacite du procedd. Ainsi, pour une teneur en ozone identique, le nombre des bacteries resistant au traitement diminue a mesure que la tempdrature du milieu s'abaisse. Une faible turbidite n'a que peu d'influence sur l'action de l'ozone. Par contre, si l'eau est tres trouble ou tres coloree, le dosage doit etre augmente. Esch. coli est beaucoup plus r6sistant A l'ozone que diverses souches de bacilles typhiques. Quant aux effets virulicides du traitement, ils sont fonction de la duree du contact, de la concentration d'ozone residuel et de la teneur de 1'eau en impuretes. La desinfection de l'eau par les rayonnements gamma presente un interet certain. L'etude du rythme de destruc- tion des enterobacteries par cette methode revele que le processus, independant du milieu oiu il se deroule, est du type exponentiel. L'analyse du mecanisme de l'action bactericide du chlore montre que les enzymes bacteriennes catalysant les processus respiratoires sont plus sensibles au produit que Jes enzymes intervenant dans la degradation des amino- acides. Les rayons gamma, a l'inverse du chlore, n'inhi- bent que faiblement l'activite des deshydrogenases et des decarboxylases. En revanche, ils modifient profondement la structure de l'ADN bacterien. La microscopie electro- nique revele que chez Esch. coli traite par le chlore l'inactivation des enzymes s'accompagne de diverses alterations de la morphologie cellulaire. REFERENCES Baroyan, 0. V. (1967) Oc'erki po mirovomu rasprostra- neniju vafnejSih zaraznyh boleznej c'eloveka [Essays on the world distribution of the most important com- municable diseases of man], Moscow, Medicina Baroyan, 0. V. (1968) Itogi poluvekovoj bor'by s infek- cijami v SSSR i nekotorye aktual'nye voprosy sovre- mennoj epidemilogii [The results of half a century of communicable disease control in the USSR and some topical problems of modern epidemiology], Moscow, Medicina Cerkinskij, S. N. & Trahtman, N. N. (1962) Obezfara- zivaniepit'evoj vody [The disinfection ofdrinking water], Moscow, Medgiz Cerkinskij, S. N. & Rjabcenko, V. A. (1968) Gig. i Sanit., 33, No. 5, pp. 10-14 Cerkinskij, S. N. et al. (1969) Materialy vsesojuznoj kon- ferencii po gigiene vody i sanitarnoj ohrane vodoemov [Documentation of an All-Union Conference on Water Hygiene and the Protection of Watercourses against Pollution), Moscow, 87-90 Chang, S. L. (1968) Bull. Wld Hlth Org., 38, 401-414 Citovic, I. S. (1940) In: Trudy Rostovskogo medicinskogo instituta [Transactions of the Rostov Medical Institute], Vol. 7, pp. 11-20 Druckrey, H. (1968) Food Cosmet. Toxicol., 6, 147-154 Gubar', M. A. & Kozlova, N. D. (1967) Gig. i Sanit., 32, No. 5, pp. 11-14 Gubar', M. A. et al. (1970) Gig. i Sanit., 35, No. 8, pp. 18-22 Lipinska, R. K. (1963) Gig. i Sanit., 28, No. 10, pp. 19-23 Lovcevic, E. L. & Sergunina, L. A. (1968) Gig. i Sanit., 33, No. 9, pp. 22-27 Maslennikov, N. A. (1963) Gorodskaja kanalizacja, 2 [Urban sewerage, 2]. In: Nau6nye trudy Akademii kom- munal'nogo hozjajstva im. K.D. Panfilova [Scientific Transactions ofthe K. D. Panfilov Academy ofMunicipal Services], Vol. 20, pp. 57-60 Muhin, A. I. (1935) Nekotorye gigieniceskie voprosy central'nogo vodosnabtenija v Jksperimental'noj ih obrabotke [Experimental work on some aspects of hygiene in central water-supply systems], Smolensk Rjabcenko, V. A. (1964) Gig. i Sanit., 29, No. 9, pp. 23-27 Rjabcenko, V. A. & Lovcevic, E. L. (1965) Gig. i Sanit., 30, No. 8, pp. 26-30 Sergunina, L. A. (1968) Gig. i Sanit., 33, No. 4, pp. 16-21 Skidal'skaja, A. M. (1969) Gig. i Sanit., 34, No. 11, pp. 11-17 Suckov, B. P. (1964) Gig. i Sanit., 29, No. 6, pp. 22-29 Vahler, B. L. (1963) Gig. i Sanit., 28, No. 3, pp. 8-13 Vrocinskij, K. K. (1963) Gig. i Sanit., 28, No. 12, pp. 11-17