LIBRARY OF THE UNIVERSITY OF ILLINOIS AT URBANA-CHAMPAIGN G28 oonf.wo* ENGINEERING The person charging this material is re- sponsible for its return on or before the Latest Date stamped below. Theft, mutilation and underlining of books are reasons for disciplinary action and may result in dismissal from the University. UNIVERSITY OF. ILLINOIS. _UBi(A£Y_AT__UR£ANA-aiAMPAIGN El CONFERENCE MAY 8 L161— O-1096 Digitized by the Internet Archive in 2013 http://archive.org/details/virusremovalbych54chau - (k) Survival of Viruses in Water 8 B. Detection of Viruses in Water 8 C. Removal of Viruses by Coagulation and Flocculation 11 D. Removal of Viruses by Other Water Treatment Methods 17 (1) Filtration 17 (2) Disinfection 17 E. Aqueous Chemistry of Aluminum 18 III. SCOPE OF THE INVESTIGATION 22 IV. MATERIALS AND METHODS 2k A. Materials 2k (1) Viruses and Preparation 2k Escherichia coli Bacteriophage Tk 25> Escherichia coli Bacteriophage MS2 27 (2) Biological Media 27 Media for Escherichia coli Bacteriophage Tk 27 Media for Escherichia coli Bacteriophage MS2 29 (3) Coagulant and Polyelectrolytes (Coagulant Aids) 30 (k) Water 30 (£) Organic Materials 31 (6) Glassware 31 B. Chemical and Biological Assay Techniques 31 (1) Determination of Aluminum 31 (2) Assay Procedure for Bacteriophages 32 C . Experimental Technique and Equipment 3k (1) Kinetics of Adsorption of Aluminum by Viruses 3k (2) Virus Inactivation Studies 36 (3) Quantitative Studies on Adsorption of Aluminum by Viruses 36 Studies on Adsorption of Aluminum by Viruses 36 Aluminum Saturation Curves 37 (k) Quantitative Studies on Virus Removal by Chemical Coagulation and Flocculation (Jar Tests) 37 (5) Virus Recovery from Settled Floe ill xv Page V. RESULTS AND DISCUSSION k?- A. Adsorption of Aluminum by Viruses h2 ( 1) Kinetics of Adsorption U2 (2) Quantitative Adsorption ht> B . Virus Inactivation by Aluminum 60 C. Virus Removal by Chemical Coagulation and Flocculation (Jar Tests) ^3 (1) Optimum pH and Aluminum Sulfate Dosages "3 (2) Effect of Calcium and Magnesium on Virus Removal 6U (3) Effect of Organic Matter on Virus Removal 68 (k) Effect of Preformed Floe on Virus Removal 69 (£) Effect of Polyelectrolytes on Virus Removal ?3 D. Qualitative Description of Virus Removal by Chemical Coagulation and Flocculation °1 VI. SUMMARY AND CONCLUSIONS 83 VII. ENGINEERING SIGNIFICANCE 86 VIII. SUGGESTIONS FOR FUTURE WORK 89 LIST OF REFERENCES 90 VITA 98 LIST OF TABLES TABLE Page 1. ENTERIC VIRUSES AND DISEASES WITH WHICH THEY HAVE BEEN CLOSELY ASSOCIATED 6 2 . SURVIVAL OF ENTERIC VIRUSES IN WATER 9 3. EFFECT OF CHLORINE AND IODINE ON VIRUSES AND E. coli 19 k. PROPERTIES OF E. coli BACTERIOPHAGE TU 23' 5. PROPERTIES OF E. coli BACTERIOPHAGE MS 2 2? 6. POLYELECTROLYTES (COAGULANT AIDS) 30 7 . EXPERIMENTAL PROTOCO L FOR KINETIC STUDIES 33' 8. MASS BALANCE ON ALUMINUM BY CENTRIFUGATION METHOD U9 9. AMOUNTS OF ALUMINUM ADSORBED BY A BACTERIOPHAGE TU PARTICLE AT DIFFERENT pH VALUES , 3'3 10 . AMINO ACID COMPOSITION OF BACTERIOPHAGE TU 36 11. RECOVERY OF BACTERIOPHAGES FROM SETTLED FLOC FOLLOWING CHEMICAL COAGULATION AND FLOCCULATION 63 12. REMOVAL OF BACTERIOPHAGE TU IN THE PRESENCE OF CALCIUM AND MAGNESIUM 68 13. REMOVAL OF BACTERIOPHAGE TU BY COAGULATION AND FLOCCULATION IN THE PRESENCE OF ORGANIC MATTER 70 lU. REMOVAL OF BACTERIOPHAGE MS2 BY COAGULATION AND FLOCCULATION IN THE PRESENCE OF WASTEWATER AND WASTEWATER EFFLUENT 71 15. REMOVAL OF BACTERIOPHAGE TU BY PREFORMED FLOGS 72 16. INACTIVATION OF BACTERIOPHAGES TU AND MS2 IN THE PRESENCE OF POLYELECTROLYTES (COAGULANT AIDS) 7U 17. REMOVAL OF BACTERIOPHAGE TU BY COAGULATION AND FLOCCULATION WITH ALUMINUM SULFATE AND CATIONIC POLYELECTROLYTES AS COAGULANT AIDS 75 18. REMOVAL OF BACTERIOPHAGE MS 2 BY COAGULATION AND FLOCCULATION WITH ALUMINUM SULFATE AND CATIONIC POLYELECTROLYTES AS COAGUTANT AIDS 7b VI Vll TABLE Page 19. REMOVAL OF BACTERIOPHAGE Tk BY COAGULATION AND FLOCCULATION WITH CATIONIC POLYELECTROLYTES AS PRIME COAGULANTS 77 20. REMOVAL OF BACTERIOPHAGE MS2 BY COAGULATION AND FLOCCULATION WITH CATIONIC POLYELECTROLYTES AS PRIME COAGULANTS 73 21. REMOVAL OF BACTERIOPHAGE TU BY COAGULATION AND FLOCCULATION WITH ALUMINUM SULFATE AND ANIONIC POLYELECTROLYTES AS COAGULANT AIDS 19 22. REMOVAL OF BACTERIOPHAGE MS 2 BY COAGULATION AND FLOCCULATION WITH ALUMINUM SULFATE AND ANIONIC POLYELECTROLYTES AS COAGULANT AIDS 80 LIST OF FIGURES FIGURE Page 1. SOLUBILITY CURVE FOP. ALUMINUM HYDROXIDE 20 2 . ELECTRON MICROGRAPH OF BACTERIOPHAGE TU 26 3 . ELECTRON MICROGRAPH OF BACTERIOPHAGE MS2 28 U. STANDARD CURVE FOR ALUMINUM 33 5. SCHEMATIC DRAWING OF APPARATUS FOR COAGULATION AND FLOCCULATION STUDIES (JAR TESTS) 38 6. 3CAIE DRAWING OF REACTION VESSEL FOR COAGULATION AND FLOCCULATION STUDIES (JAR TESTS) 39 7. KINETICS OF ADSORPTION OF ALUMINUM BY BACTERIOPHAGE Tk U3 8. KINETICS OF ADSORPTION OF ALUMINUM BY BACTERIOPHAGE MS2 UU 9. QUANTITATIVE ADSORPTION OF ALUMINUM BY BACTERIOPHAGE TU AT pH 5.0 U7 10. QUANTITATIVE ADSORPTION OF ALUMINUM BY BACTERIOPHAGE MS2 AT pll 5.0 U8 11. SATURATION CURVES FOR ALUMINUM ADSORPTION BY BACTERIOPHAGE Tli AT pH 5.0 51 12. SATURATION CURVES FOR ALUMINUM ADSORPTION BY BACTERIOPHAGE TU AT pH 6.0 52 13. SATURATION CURVES FOR ALUMINUM ADSORPPION BY BACTERIOPHAGE TU AT pH 9.0 52 1U. McBAIN - BRITTON PLOT OF ADSORPTION DATA OF BACTERIOPHAGE TU- ALUMIIIUM SYSTEM (pH 5-0) TO FIT LANGMUIR ADSCRPriON EQUATION ... 5U 15. McBAIN - BRITTON PLOT OF ADSORPTION DATA OF BACTERIOPHAGE TU- ALUMINUM SYSTEM (pH 6.0) TO FIT LANGMUIR ADSORPTION EQUATION ... 55 16. McBAIN - BRITTON PLOT OF ADSORPTION DATA OF BACTERIOPHAGE TU- ALUMINUM SYSTEM (pH 9.0) TO FIT LANGMUIR ADSORPTION EQUATION ... 56 17. DEACTIVATION OF BACTERIOPHAGE TU IN THE PRESENCE OF SOLUBLE ALUMINUM AT pH $.0 61 Vlll FIGURE Page 13. INACTIVATION OF BACTERIOPHAGE MS 2 IN THE PRESENCE OF SOLUBLE ALUMINUM AT pH $.0 62 19. REMOVAL OF BACTERIOPHAGE Tli AND CLAY TURBIDITY BY COAGULATION AND FLOCCULATION 65' 20. REMOVAL OF BACTERIOPHAGE MS2 BY COAGULATION AND FLOCCULATION .. 66 21. KINETICS OF ADSORPTION OF ALUMINUM BY BACTERIOPHAGE Tl; IN THE PRESENCE OF CALCIUM AND MAGNESIUM 67 I. INTRODUCTION The need of processes for the removal or inactivation of water-borne viruses has been demonstrated beyond question by virus transmission in drink- ing water and polluted streams. Because of the decrease in the available water supplies, the potential health hazard of viral pollution is increasing, and more knowledge on the survival and removal of viruses in water and waste- water treatment processes is needed. Viruses have been demonstrated to be the causative agents of a wide variety of diseases. Polio, Coxsackie, Infectious Hepatitis, ECHO, and Adenoviruses have all been demonstrated in the feces of infected humans (Clarke et al., 1961;). Of greater significance is the fact that a great many of these viruses may be found in wastewater treatment plant effluents (Kelly and Sanderson, 1959) . Enteric viruses have been found in raw waste- water and treatment plant effluents by a number of investigators (Paul, Trask and Card, 19U0; Paul and Trask, 19^2; Clarke et al., 1951; Kelly, Winsser and Winkelstein, 1957; Mack et al., 1958; Carlson, Ridenour and McKhann 19U3; and Bancroft, Engelhard and Evans, 1957) • The presence of these organisms in water supplies, then, is certainly possible. Although water is far from being an ideal medium for sustaining viruses, they can exist in it for substantial periods of time. In Paris, France, more than six virus isolations have been made from tap water in at least four different sections of the city (Berg, 196U) . Water has been suspected as being the mode of transmission in out- breaks of several virus diseases. At least four epidemics plus numerous lesser outbreaks of viral diseases have been attributed to water-borne viruses (Ban- croft, Engelhard and Evans, 1957; Dennis, 1959; Hayward, 19U6; Little, 195U; and Standly and Eliassen, i960). Hudson (1962), in attempting to correlate 1 drinking water quality with the incidence of infectious hepatitis, found that a water supply that met all existing standards could still transmit these diseases. This would indicate that a specific virus quality control or standard should be established for drinking water supplies (Hartung, 1961) . Before this can logically be accomplished, more knowledge should be gained on the effectiveness of presently used water and wastewater treatment methods for the removal of viruses and the epidemiology of water-borne viral diseases. Although the process of chemical coagulation and flocculation has been used for many years in the treatment of water supplies, little is known of the basic physico-chemical principles involved in the removal of dead, as well as living, organic matter from the treated water. Chemical coagulation and floccu- lation is applied in various forms in a modern water treatment plant to produce a safe and potable water. However, little information is available about the basic mechanisms and kinetics involved in the removal of viruses by this unit process. Virus removals ranging from 25> to 99 percent have been reported by different workers (Senn et al., 196l) . Understanding the mechanism by which viruses, \-fnich behave as typical proteins in water, are removed from water by the process of chemical coagulation and flocculation is important in optimizing its use. The present work was undertaken in order to delineate the basic mechanisms involved in the removal of viruses by chemical coagulation. Studies were also conducted to investigate the quantitative and practical aspects of the process. Various parameters, believed to affect the process, were evaluated. It is be- lieved that the information obtained from this investigation, though it may not be quantitatively applicable to a particular water treatment plant, will con- tribute fundamental knowledge regarding the removal of viruses from water by chemical coagulation and flocculation and the various environmental parameters affecting the process. II. PRESENT STATE OF KNOWLEDGE A. Viruses (l) Physical and Chemical Nature Viruses are the smallest known biological form capable of producing disease in humans and in other living species. They are obligate intra- cellular parasites, incapable of proliferating in an extracellular environ- ment. Viruses are known to infect man, other vertebrates, plants, insects, and bacteria. Those infecting bacteria have been termed bacterial viruses (or bacteriophages), while those infecting man and other vertebrates are known as animal viruses. The Rickettsia, basophilic viruses, and pleuropneumonia- like organisms (PPLO) are usually considered, both physically and biologically, intermediates between the bacteria and viruses. General properties of viruses can be briefly outlined as follows: a. Viruses have definite shapes and structures. Some appear to be spherical, others appear to be brick-shaped, or filamentous. Bacterial viruses may have tails, which give them a sperm-like appearance. In recent years, the surfaces of many viruses have been clearly photographed. These photographs have resulted in a clearer understanding of the viral capsid, the outer cell, which is composed of many subunits, or capsomers. b. Viruses of the same type are characterized by a definite size but as a group they differ in size. Some, such as poliovirus, are on the order of 30 ran. or less in diameter while others may approach hOO mp, - a size approximating that of some bacteria. c. Viruses are composed of known chemical substances, of which protein and nucleic acid are the most important. In solution, viruses behave h as proteins, though the internal structure may be nonprotein in character. In their electrokinetic properties, viruses behave as amphoteric electrolytes. The net charge depends on the pH of the solvent. Most viruses display a net negative charge within the range of pH stability. d. Viruses breed true to form, meaning that the progeny (offspring) derived from a virus particle are like the parent virus. This phenomenon is controlled by the nucleic acid contained in a virus. e. Viruses display specificity with regard to their protein and to the host cell which they infect. f . Viruses can change the life processes of a cell, causing injury, a modification in growth rate, or even death. Some viruses infect a cell without causing any recognizable effect. In such an instance, the virus can remain dormant indefinitely or can be stimulated to multiply. g. Some viruses are extremely sensitive to environmental conditions - temperature, ultraviolet light, etc., while others are fairly resistant to chemi- cal and physical factors. h. Some viruses are transmitted directly to their hosts; others re- quire an intermediate host. (2) Enteric Viruses and Diseases Associated "With Them Unfortunately, a binomial system of nomenclature as is commonly used to classify bacteria is not readily applicable to animal viruses. For a phylo- genic classification there is not, as yet, sufficient knowledge of virus properties nor of the relationships viruses hold to one another. Furthermore, probably only a small fraction of all the animal viruses have yet been dis- covered (Rhodes and VanRooyen, 1962) . Perhaps the most reasonable classifi- cations of animal viruses are those which assign viruses with similar proper- ties to groups designated by distinguishing names and the suffix "virus. " 5 Under this system of nomenclature the viruses of our concern fall within four major groups: the adenovirus group, the reovirus group, the enterovirus group, and the hepatitis group. Collectively, these are referred to as "enteric viruses. " Table 1 lists these viruses together with the diseases with which they have been associated. All enteric viruses share a number of common characteristics (Clarke and Chang, 1959) : a. They are excreted in feces in large numbers. Poliovirus, for example, has been reported in concentrations as high as 100,000 virus units per gram of feces. b. They can be found in municipal wastewater, particularly during the late summer and early fall. c. Infection iri.th many of these viruses is widespread in the normal population; infection rates are highest in infants and young children. d. Illness, when it occurs with infection with one of these viruses, may be so mild as to be mistaken for a slight cold or, in rare cases, may be as severe as paralytic poliomyelitis. (3) Water-borne Viruses and the Hazard of Viruses in Water Undisputably viruses exist in water (Gard, 19U0, and Kelly, Winsser and Winkelstein, 1957). Many are introduced into surface waters and municipal sewer systems through human and animal feces. All enteric viruses occur in excreted feces in considerable numbers. Over 100 different viruses are ex- creted in human feces, including three types of Poliomyelitis, 30 types of Coxsackie, 28 types of ECHO, Infectious Hepatitis and various Adenoviruses (./eibel et al., 196U; Berg, 196k; and Rhodes and VanRooyen, 1962). The typical expected density of enteric viruses in municipal wastewater would average about 7000 per liter of raw, untreated wastewater (Clarke and K abler, 1961;), and in polluted surface water not more than one virus unit per 100 ml TABLE 1 AND DISEASES WITH WHICH THEY HAVE BEEN CLOSELY ASSOCIATED (After Berg, 1966) Disease Virus" CO •H -P o3 co •H •H CD -P H O rt -H CD G rt o 3* i« Oh Jh o co -p CD «J CO u co •H CD a, £ co H CD r-i ffj H CO •H ■P •H CD •P CO CD CO cd 0) CO •H Q CO CD H •H +2 C co 0 episodes of infectious hepatitis reported in the literature in x-jhich the investigators concluded that infectious hepatitis was transmitted by drinking water. Out- breaks of poliomyelitis have been suspected from wastex^ater-contaminated water supplies (Kabler et al., 1961; Little, 195U; and Bancroft, Engelhard and Evans, 1957). Poliomyelitis virus has been isolated from well water supplies in both the United States and Sweden and from a creek in Ohio (Clarke and Chang, 1959). Outbreaks of sore throats and pink eye with fever have occurred in Washington, D. C. and Toronto, Canada, from bathing in areas contaminated by Adenovirus types 3 and 7 (McLean, 196U, and Kabler et al., 196l) . Enteroviruses in swimming pool thus constitute a problem. These two occurrences are not the only incidents associated with bathing areas. Enteroviruses were isolated from wading pools in Albany, ilew York (Kelly and Sanderson, 1961) and in Toronto, Canada (McLean, 1966). In 1957, spread of vesicular exanthema in Toronto, caused by Coxsackie Al6 virus, was sided by backyard swimming pools (Mclean, 196h) . It has been established that viruses are present in surface waters re- ceiving wastewater effluents. There is, therefore, an urgent need for further investigation on the virus content of wastewater effluents, the relative 8 efficacy of different types of wastewater treatment, the consequent virus pollution of rivers, the survival of viruses in natural waters and the effectiveness of water treatment processes in their removal or inactivation. (U) Survival of Viruses in Water The rate of viral inactivation in natural waters depends upon temperature, water quality, predators, and other factors that are not well understood (Berg, Scarpino and Herman, 1966) . Due to the existence of many unknown factors effecting virus survival in water, the differences in the survival of different enteroviruses observed in storage (Clarke et al., 1°6U) are not sufficient to indicate that some enteroviruses survive significantly longer than others. In general, a 99.9 percent reduction requires from a few weeks in the warm season to a few months in the cold season in temperate regions and a shorter time in the tropics and subtropics (Chang, I960). Virus survival data are summarized in Table 2. It may be concluded that the storage time normally available to water treatment plants is not to be depended upon for any significant reduc- tion in the enteric virus concentration. 3. Detection of Viruses in Water Studies on the concentration of viruses in water and wastewater and the evaluation of water and wastewater treatment methods in removing viruses have been severely limited due to the lack of suitable quantitative methods for virus detection. Low densities of viruses in natural waters require that large volumes of water be collected and subsequently concentrated before virus detection and isolation. According to Chang (1968), a method suitable for detection of viruses in water will have to fulfill the following qualifica- tions: (l) it can concentrate small numbers of viral units from large volumes of water, (2) the virus material thus concentrated can be conveniently inocu- lated into cell cultures for determination of viral density by the tissue CO CO ^-^ MO o On "LA i^rH CO •s CO On H^ Sh ^s i — i H $H H •H O MO CO XA rH •H CO w .— . ^ P C On P X O o o • CO P CO H CO O -1 MO On rH 1 & CO O | _ P f-) _ _ t3 - _ _ p oo •^1 ^w cu CO CO CD -d CD| CO H fl CO £U CO 9^ ^-s P JL, \A CO nO v ^ CO MO 1A TS & O 1A O MO U O NO M4 o rH On P On CO T3 p On On 43 O •H H i— ! rH C CD rH O H CtS Ph CS v -' >-— fe> rd O cd C O PI o TJ •H rH C CO ^. ^-^ ^-^ ^^. ^-^ CO f-i •H J-, CO r- .•>c*- •^r>- [N- • »\ t — CO 3 H CA ,--> H ^vH CA ^—> rH P P •— N , — ^ ^--> 1A ^-N 1A N ' "LA rH^ M £ RJ fc o O CO • 1 -^f • 1 CO • 1 cd •H h CD CM CM CM H CM H CM rH rH 5L| c-i P v -^-»' *• — ' 1A ^—^ "LA "LA •H CO O a Td o n _Tt CM "LA fA cA 1A CM T! 3 X «: f-. G CO H O rH £2 2 cd co •*--• CD A A ^^ rH CO M C CO h-H ^-»> ^— X , S3 3 c (A ra CA 3^ O CM CM CM H CO o o , s > CO -H P CD ' CM 1 1 o o o u o O o ;— i Pi 3 CV' 1 CM CM S S -3 •4—' > — ' -— ' ^— ' td r- o CM o r-t CD X Sh CM O o £h) cd P CD V ' — ^ rH H W a ■LA h C S • #\ • *- • *\ fin O CD CD •— ^. ,. — s • »\ ^-^ s ^ • — *. O -P o <:o in P rH o rH o" 9 9 k3 CD -d H < CD Ph J 1 1 1 1 i > H 1 H ,0 On 1A CO O CO CO > cd On H ^s CO CM ^^s ^— ' Cti £> CO •-— ' o o * *•— ' 8 CM & O fc ^ 1A o -=! r- t^- CO fn O 2h H H p p •H £s-i co p 2 O O CO cd i CD P ^! o ^ rH jd P o o p P U ,Q P rH p a 2 fn ,o rQ c "S P ,3 JL, •H o g PL, •g.s < P a u ^ rH O o CO 43 CO s — ^ -^-* u C-, a P U h 3 u ^3 c p C -p 5 •H CO p r-i •H i? •H cd ^ o P Td P cO P H p T3 *~~z cd as fn CO U JC O jctJ h »H p fn u rH H •H P CO & 3 f-i O U CO JH H H •H P CO P cd & jd 4-3 £ r» S CO P P § ^ J3 3 rH t> t> CO c6 p !> J> o f> CO CO p CO p H •H •H •H CO •H ^ •H >-i •H •H CO CO _0 cr; O CO W pcj O -d c5 rvJ P. co w CO H '-•* CO CD CO O •H 1 O CO 1 CO g •H rH A*S o 2 O o rt o CO Hl^ -H K P 10 culture technique, and (3) ideally it should separate the virus from bacteria, protozoa, and other microbes, as well as toxic material to avoid the detri- mental effect of the latter on tissue cells. Various concentration methods have been reported in the literature. Older methods include precipitation, ion exchange, centrifugation and combination of these. lie cent methods include hydro-extraction and two-phase separation (Shuvall et al., 1966, and Lund and Hedstrom, 1966), electrophoresis (Bier et al ., 1966), continuous-flow ultracentrifugation (Anderson et al., 1966), passive hemagglutination (Smith and Courtney, 1966), concentration with poly- ethylene glycol (Oliver, 1966a), use of soluble ultrafilters (Gartner, 1966), aluminum phosphate and aluminum hydroxide precipitation (Wallis and Melnick, 1966, 1967a), adsorption on membrane filters (Oliver, 1966b, and Uallis and Melnick, 1967b, 1967c), and adsorption on iron oxide (Rao et al., 1968). An older quantitative technique, grab sampling, has been made more practical by recent improvements in assay technology (Rawal and Godbole, 196U, and Berg et al., 1966), but the amount of water that can be sampled by this technique is limited because the water to be sampled for viruses is used to prepare the medium for nourishing the cell cultures that serve as the assay system. The most commonly used method today is the gauze pad technique. Pads of gauze or sanitary napkins are suspended in flowing waters for several days. Pads are then squeezed after adjusting the pH of the absorbed water to eight, the expressed fluid centrifuged, first at low speed to sediment bacteria and particulate matter, and then at high speed to sediment viruses which are sub- sequently assayed. The major deficiency of this method is that it is not quantitative (Berg, 1968). The method that is most promising at the moment for concentrating small quantities of viruses from large volumes of water is the membrane filter 11 technique of Oliver (Berg, 1968) . The sample of water is filtered through a 0.1i5> p, Ilillipore membrane to which viruses adsorb. Presence of salts and a pH value close to neutrality facilitates virus adsorption (Berg, Dean and Dahling, 1968, and Wallis and Melnick, 1967b). Turbid waters can be cleared by filtra- tion through course filters with relatively little loss of virus. Elution of the viruses is achieved by immersing the filters in three percent beef extract (Berg, Dean and Dahling, 1968). Viral density is then determined by conven- iently inoculating cell cultures with the eluate. This technique, although the most quantitative available presently, suffers in that the beef extract is not a universal elutant. The membrane filter technique is the most quantitative method available within its limitations. The gauze pad procedure is recommended for qualitative studies on the assumption that this technique may be the most sensitive avail- able presently (Berg, 1968) . This method has been recently used in the Santee Recreation Project, Santee, California (Merrell et al., 1967). G. Removal of Viruses by Coagulation and Flocculation The information on the removal of viruses in water by coagulation and flocculation is extremely sketchy and nonquantitative. This is mainly due to the lack of accurate quantitative assay techniques for animal viruses and a reliable method for concentrating water samples containing viruses. Many in- vestigators have used bacterial viruses as models for studying this problem. uhen animal viruses were used, biological assays involving laboratory animals were employed instead of quantitative tissue culture technique. Furthermore, it is difficult to correlate the observations of different workers due to the fact that coagulation and flocculation conditions were different. In most of the studies undertaken in this area, attempts were made to arrive at quantita- tive values like percent virus removal, relationships between virus removal 12 and alum dosage, and settling time, etc. Hone of these studies was directed toward an understanding of the basic physico-chemical processes involved. Using the virus of the mouse -adapted strain of human poliomyelitis, Carlson, Ridenour and IicKhann (l9l|2) found that flocculation using alum (100 ppm) did not render the water noninfective for mice, and that a rapid sand filter heavily impregnated with alum was somewhat more effective in removing the virus than the former process. Using both alum flocculation and filtration on water containing the virus of a monkey- adapted strain of human poliomyelitis, Kempf et al. (l9i|2) freed the supernatant of the virus in two experiments and freed the effluent (sedimented and filtered) of the virus in one, out of the three experiments. Only by greatly increasing the alum dosage were they able to remove all the viruses. Since the data were measured by a biological assay involving the infecting of monkeys, there is some doubt that 100 percent removal was achieved even then. These authors also centrifuged their flocculated water and claimed that the supernatant could be freed of the virus if a floe sediment of 1,$ mg/1 (in a Hopkins tube) was obtained with water flocculated at a dosage of eight grains/gal of alum; but the virus in the floe sediment was not destroyed. They could recover virus activity in the resuspended sediment when none was present in the super- natant, '.forking with the virus of Infectious Hepatitis, Heefe et al. (l Q U7) found that alum flocculation and filtration through a diatomaceous earth filter did not completely remove the infective agent from the treated water. Forty percent of the human volunteers developed the disease after ingesting the treated water. The basic mechanism involved in the removal of viruses by flocculation has been thought to be the formation of a metal cation (coagulant) - protein (virus) complex followed by precipitation and flocculation (Chang et al., 13 1958b; Chang, Isaac and Baine, 1953; Felix, 1965; and Kabler et al., 1963). 'This hypothesis seems quite reasonable from the chemical aspects of coagula- tion. Stumm and Morgan (1962) observed: "The physical or double-layer theory has been developed in great detail and has, in its various forms of simplification, found wide acceptance. .. .This theory has virtually replaced and superceeded the older chemical theory. These two coagu- lation theories are not, however, as mutually exclusive as they might appear to be on first sight, and it is important to call attention to the fact that purely chemical factors must be considered in addition to the theory of the double layer in order to explain, in a more quantitative way, the dependence of the stability of colloids upon the chemical composition of the medium. .. .Occasionally, specific chemical equilibria, such as complex formation, may be more important than double layer compaction through counter-ion adsorption. . . . Complex formation reactions between aluminum or iron coagulant metal ions and carboxylic, phosphato, sulfato, or aromatic hydroxyl functional groups are important in the destabiliza- tion of such naturally occurring colloidal or dissolved impuri- ties as color, proteins, and carbohydrate materials. The marked difference in the response of carboxyl, sulfate and phosphate colloids to coagulation by metal ions is indicative of specific chemical interactions. " By extrapolating the interactions of proteins and salts of metals to that of virus protein and aluminum, Chang, Isaac and Baine (1953) first postulated the aluminum- vir\is complex formation. later, they showed the importance of this complex formation as the first stage in -the removal mechanism by using Gum .Arabic, a substance which interferes with flocculation (Chang et al., 195>Cb). Gummy substances form a protective coating on charged particles and are well known for their stabilizing effect on emulsions. Viruses are not destroyed as a result of the complex formation as was evidenced by the re- covery of viruses from the floe fraction (Chang, Isaac and Baine, 1953, and Gilcreas and Kelly, 1955) . Chang, Isaac and Baine (1953) using bacterial virus showed that virus removal followed a Freundlich adsorption isotherm and that preformed floe had little effect on virus removal. In concentrating Coxsackie virus Ik suspensions by alum f locculation it was again noticed that virus recovery was not high when preformed floe was used (Stevenson et al., 1956). A linear relationship was obtained between percent removal/alum dosage and percent virus remaining in suspension for a defined pH, temperature, f locculation and sedimentation time. The percent removal was a function of coagulant dose below the upper limit of the "zone of f locculation. " Robeck, Clarke and Dostal (1962) also noticed that increasing alum dosage increased the virus removal up to 99 percent. Polyelectrolytes were useful when filtration and coagulation were employed together. Furthermore, Chang, Isaac and Baine (19^3) obtained different values of u in the relationship R = uX (where R = percent removal/alum dose; u and n are constants and X = percent virus re- maining in suspension) for two series of experiments conducted under identical experimental conditions. They concluded that some other unknown factor or factors were responsible apart from the cation-virus complex formation. As to the kinetics, they showed that 20 min were required for the first stage (complex formation). For the second step (aluminum- virus precipitate forma- tion) they obtained a linear relationship between the log of virus remaining active and the square of the contact time. The energy of activation was calculated ajid found to be of the order of magnitude of that of diffusion (S = 6,770 calories). In a recent publication, Berg (196J4.) reported 96 and 9k percent removal of Coxsackie A2 virus by alum and ferric chloride, respectively. Removal was 96.6 percent when both coagulants xrere used. Temperature in the range 5°C to 25°C did not affect the removal significantly. On the other hand, Chang noticed slightly lower removal in cold months (Senn et al., 1961). Virus removal ranging from 25 to 99 percent were reported by different workers (Senn et al., 1961) . It is of interest that high virus renoval by 15 Chang 's group was achieved with low alum dosages, whereas low efficiency was attained by others who used high dosages. A critical comparison of virus removal efficiencies by flocculation cannot be made unless the conditions under which the experiments were conducted are known. As most of the investi- gators did not indicate the relative performance of the flocculation process as judged by visual inspection of turbidity removal, the relatively low virus removal reported might suggest that the flocculation process was inadequate. Using a bacterial virus (bacteriophage against M. pyogenes var. albus) Chang, Isaac and Baine (1953) studied the fate of virus particles removed by flocculation. They found that virus particles were temporarily "inactivated" during the complex formation as a result of the formation of the aluminum salt of protein in the virus and would be "active" again when dissociated from the floe. They were able to 'reactivate" the virus particles by redispersing the flocculated mixture at pi I 7.6 with vigorous stirring. Sixty percent of the removed viruses could be recovered by this technique. Quite contrary to this observation, Puck, Garen and Cline (l95>l) demonstrated that trivalent cations like Al, Cr, and Fe permanently inactivate bacteriophages and their host cells, Later Yamamoto, Hiatt and Haller (196U) reported that A1CL, at concentrations between 10 ^ and 2.10 M failed to inactivate bacteriophage 1-1S2. Carlson et al . (1968) reported a significant increase in the percent inactivation of virus particles (bacteriophage T2) adsorbed on clay particles x^en aluminum salt was present in the system instead of calcium or sodium salts. However, no attempt was made to reactivate the virus particles or to assess the toxic effect of the aluminum on the virus or the effect of aluminum on the surface charge of the virus. Other metal ions may have an important role in the removal of viruses by coagulation and flocculation, e.g. in slowing down the rate of aluminum- 16 virus complex formation. Interfering effects of calcium and magnesium xrere noticed by Chang et al. ( 195Gb) during virus removal by two-stage flocculation with Ohio iiiver water. They believed that the presence of calcium and magne- sium ions in the raw water and the addition of CaO in the second stage inter- fered with the formation of the coagulant-cation bacterial-virus complex. Little information is available regarding the influence of organic matter on the removal of viruses by flocculation. Using alum and artificially con- taminated river water with Coxsackie virus Frank ova, Cervenka and Symon (196U) obtained highest removal of inf ectivity when the virus \ia.s added in the form of a suspension of infected mouse brain and found that the optimum amount of alum for the removal of the virus was dependent on the concentration of organic matter in the water. However, Carlson et al. (1968) observed considerable re- duction in percent virus (bacteriophage T2 and type 1 Poliovirus) inactivation by clays (lllite, Montmorillonite and Kaolinite in presence of sodium chloride or calcium chloride) Xiihen albumin or wastewater was present in the system. This was shown to be due to competition with virus for the adsorption sites on the clay. In connection with their studies on the movement of viruses in ground water, Eliassen et al. (i960) observed the effects of synthetic polyelectrolytes on virus removal by alum flocculation. Virus removal was improved by the addi- tion of varying concentrations of polyelectrolytes. Of the three species of polyelectrolytes tested, the results indicated that for similar dosages the cationic polyelectrolyte was approximately two orders of magnitude more effec- tive than the anionic and neutral polyelectrolytes. Little virus removal was accomplished when the sand phase was absent in their jar tests. Presumably stirring and settling conditions were not sufficient for the floe to settle in the absence of the sorption effect of sand. 17 Quite contradictory observations have been reported regarding the effect of pH on Coxsackie and bacterial virus removal. Chang, Isaac and Baine (1953) found pH 5.5 to be the optimum and obtained very low removal at pH 7.0. In a later paper less removal was observed at pH 5.0 (Chang et al., 1958a). ^uite different results were obtained with bicarbonate and phosphate buffers. He- search conducted at Kings College in England (1956) reported 99 percent removal of bacteriophage with 60 ppm of ferric chloride at pH 6.5. D. Removal of Viruses by Other Water Treatment Methods (1) Filtration A few observations have been made on filtration of viruses through sand and garden soil. Most of the earlier workers, as reviewed in the literature survey by Clarke and Chang (1959), obtained poor virus removal in laboratory- studies in which the virus suspensions were usually prepared from infected animals. Presumably, the poor removal resulted from the high concentrations of extraneous organic matter in the suspensions. The recent study of Robeck, Clarke and Dostal (1962) showed that one to 50 percent of the added Poliovirus was removed by rapid sand filters (2 to 6 gpm/sq ft) and 22 to 90 percent by slow sand filters (0.035 gpm/sq ft). How- ever, rapid sand filtration of coagulated water, after adequate settling of the floe, removed 90 to 99 percent of the virus originally present. Using floccu- lated water, Gilcreas and Kelly (1955) obtained 35 and 98 percent removals for rapid sand (2 gpm/sq ft) and slow sand (0.2 gpm/sq ft) filters, respectively. The observations of Neefe et al. ( 1 9^4 7 ) on the Infectious Hepatitis virus indicate insignificant removal of this virus by diatomaceous earth filtration. (2) Disinfection Viruses are more resistant than bacteria to chlorination. Different types 10 of enteric viruses vary widely in degree of resistance to free chlorine (Kabler et al., 1961, and Chang, 1968) and iodine (Berg, Chang and Harris, 196U) . The data of Weidenkopf (1953) on destruction of Poliovirus 1 by free chlorine at C revealed that HOC1 was considerably more effective than the OCl" ion. This is also supported by the data of Kelly and Sanderson (19^8) on the destruction of enteroviruses. Poynter (1968) prepared a table from the data presented by Clarke et al. (196U) who, by collating data of their own with that of other workers, were able to produce what they considered should be reasonable estimates of the sensitivities of some of the enteroviruses and ari Adenovirus relative to that of E. coli (Table 3) . E. Aqueous Chemistry of Aluminum It is apparent from the solubility considerations of aluminum and iron that coagulation in water and wastewater treatment is carried out under condi- tions of pH and coagulant dosage such that the system is oversaturated with respect to the metal hydroxide (Stumm and 'Melia, 1968). However, it is thought that a brief discussion of the aqueous chemistry of aluminum is warranted here because a great majority of the experiments will involve soluble concentrations of aluminum. Recently, studies on the hydrolysis of the aluminum ion in dilute aqueous solution have received increased attention. A substantial amount of new infor- mation concerning the specific chemical structures of the hydrolysis products of the aluminum ion have been obtained. According to the available literatui on the aluminum hydrolysis reactions, solubility curves for aluminum have been calculated and presented (Black and Chen, 1967; Stumm, 196U; and Stumm and 'Melia, 1968). Figure 1 shows the aluminum solubility curve presented by Black and Chen (1967). On H c * *-. W 0) H o u CD -P S. S E-t O H 1A 1A 19 o 1A CO o o O o o o (A H CM •=3 on < CO to g B CD •H CD o ■g •0 c>~ o o rt «5 C •H 10 (0 o o H X X x! T3 O o o o W KW> Al^OH)^ V\\\W A1(0H)~ \\ / A1 ,3 (OH) 3; 5 Al + l /Al (0H) + 2 /Al (OH), 10 12 1*4 PH FIGURE1. SOLUBILITY CURVE FOR ALUMINUM HYDROXIDE (After Black and Chen, 1967). 21 Figure 1 is useful for visualizing; the specific regions of various aluminum ion species for varying pH values at different total concentrations of aluminum. Within a range of total aluminum ion concentration from 10 " M to 10 M, the pH scale from 3 to 10 can be roughly divided into four different regions in which different predominant aluminum ion species are present. In the region below pH h, the hydrated trivalent aluminum ion is the most active ion species. In the region between pH k and 6, the predominant aluminum, ion species are the hydrolyzed polymeric multivalent cation species: Al 7 (0H),!f and Al,^(0H)-.j . Predominance of other cation species, such as Al^(OH) i and Aln(0H) ? ^ have also been proposed in the literature (Black and Chen, 1967). In the pH range roughly from 6 to 8, insoluble aluminum hydroxide, Al(0H)_, is the predominant species, and the aluminate ion, Al(OH), , is believed to predominate above pH 8. However, it is interesting to note that, in the region between pH 5 and 8.5, the formation of insoluble aluminum hydroxide colloids or precipitates does not take place up to a total aluminum concentration of about 10 M as evidenced by the Tyndall effect in aluminum sulfate solution (Black and Chen, 1967, and Morgan, 1967). III. SCOPE OF THE INVESTIGATION From the preceding discussion, it is evident that there is a great need to obtain more information on the basic mechanisms involved in virus removal by chemical coagulation and flocculation so as to enable the engineer to put the process on a sound scientific base. The removal of viruses from water supplies is extremely important and, becomes more so, with the potential need for wastewater reuse. Information on the basic mechanisms involved In virus removal by coagulation and flocculation should contribute significantly to the solution of the problem and aid in the development of sound design standard for their removal in water treatment facilities. Host of the investigations so far undertaken in this area have been directed towards quantitative results, i.e. gross removal efficiencies. Very limited attention has been given to the mechanisms of virus removal (Chang et al., 19£8b; Chang, Isaac snd Baine, 1953; Felix, 196£; and K abler et al ., 1963) . Considering the observations of the past researchers in this area and the interaction of proteins with metals ions, the virus removal mechanism may be postulated as a two- stage reaction. The first stage is the formation of a virus- aluminum complex, the second stage is the subsequent precipitation and flocculation of the complex. Basically, the first stage is the interaction of virus with metal ion. The present research was initiated to investigate this interaction. The study was undertaken along the following lines: a. demonstration of a "complex formation" between viruses and aluminum, and b. nature of the "complex" and virus inactivation by aluminum. ■he second phase of the study was directed towards a quantitative study of virus removal by alum flocculation under controlled laboratory conditions. 22 23 The effects of the following variables in virus removal by chemical coagulation and flocculation were also investigated: a. pH and coagulant dose, b. bivalent metal ions like calcium and magnesium, c. organic matter, and d. synthetic polyelectrolytes (coagulant aids). IV. MATERIAL MD METHODS A. Materials (l) Viruses and Preparation One of the main criteria in selecting a virus for this study was feasi- bility of culturing and enumeration. Assay techniques for bacterial viruses are better developed than for animal viruses. Bacterial virus assays require about 12 to 2U hr compared with $ to 10 days for animal viruses, and culture procedures are simpler for bacterial hosts. Bacterial and animal viruses have many similar physical, chemical and biological properties, i.e. size, net electric charge, protein coating, etc. (Adams, 1959) . Thus, it may be assumed that their differences in behavior when subjected to coagulation and floccula- tion may not be much greater than the differences in these properties among the animal viruses alone, which could cause significant variations in removal by chemical coagulation and flocculation. Furthermore, there is much more known about the composition and properties of bacterial viruses than is known about the animal viruses. This allows for a more detailed examination of their behavior in chemical coagulation. Two bacterial viruses, bacteriophages ?h and MS2 against Escherichia coli, were selected as the model viruses for this study. Bacteriophage Tlj. is a DNA (deoxyribonucleic acid) containing virus whereas bacteriophage MS2 is a RNA (ribonucleic acid) containing virus. Bacteriophage Th was selected as the orincipal model virus because of its stability in agitated systems and its greater ease of culturing and enumeration (Cookson, 1966, and Drake, 196?). Bacteriophage MS2 was selected as the second model virus in order to confirm the results with bacteriophage TI4. It was thought that use of MS2 along with 25 Tii in some of the major experiments would allow the data to be interpreted in terms of viruses which may be more significant in water supplies. MS2 was selected because of its resemblance in size and shape, and the type of nucleic acid to poliovirus (Kruse', 1968; Hanson, 1969; and Spiegelman, 1969). a. Escherichia coli Bacteriophage Tii Bacteriophage Tii (Figure 2) and its host E. coli BB were obtained from Dr. John VJ, Drake, Department of Microbiology, University of Illinois. Bacteri- ophage Tii has "the following properties (Table k) (Kellenberger, 1962; Stent, 1963; and Putnam, 1953). TABLE h PROPERTIES OF E. coli BACTERIOPHAGE Tii oize head tail Specific weight pH stability range Sedimentation constant nucleic acid 65 x 95 m^j, 20 x 95 mp -16 3.3 x 10 gm/particle 5.0 - 9.0 700 - 1000 Svedberg units (S 2Q ^) DMA (deoxyribonucleic acid) The procedure for growing bacteriophage Tii was also obtained from Dr. Jc % n W. Drake. Stock suspensions were prepared by infecting an early log phase broth culture of E^ coli BB with bacteriophage Tii at a low multiplicity of infection (aoproximately 0.02 phages/ml). The infected culture was incubated at 37°C until the lysis was complete as evidenced by visible reduction of turbidity. Few drops of chloroform were then added to the lysed culture and 26 FIGURE 2. ELECTRON MICROGRAPH OF BACTERIOPHAGE Th . Final magnification 420,000 X. Provided by T. F. Anderson, The Institute for Cancer Research, Philadelphia, Pennsylvania. 27 mixed well in a vortex mixer in order to facilitate lysis of the unlysed cells. The suspension was centrifuged at low speed (5,900 x g for 10 min at k°C) to remove bacterial cells and cell debris followed by high speed centrifugation (3ii,800 x g for one hr at h C) to sediment the virus particles. The sedimented pellet was finally re suspended in phage buffer. The final purification step was repeated twice. The purified stock was then stored at h C. Stocks pre- pared in this way usually titered between 10 - 10 phages/ml. b. Escherichia coli Bacteriophage MS2 A purified stock suspension of bacteriophage MS2 (Figure 3) and its host E. coli A19 were originally obtained from Dr. S. Spiegelman, Department of Microbiology, University of Illinois. Later, purified stock suspension of MS2 in 0.05> M tris buffer (pll 7.6) were purchased from Miles Laboratories, Inc., Elkhart, Indiana. Bacteriophage MS2 has the following properties (Table 5) (Overby et al. , 1966). TABLE $ PROPERTIES OF E. coli BACTERIOPHAGE MS2 Particle diameter 25 th\i Molecular weight 3.7 x 10 gm Isoelectric point (pH) 3.9 Sedimentation constant 79 Svedberg units (S 2Q p Ilucleic acid RNA (ribonucleic acid) (2) Biological Media a. Media for Escherichia coli Bacteriophage Tlj. 28 FIGURE 3. ELECTRON MICROGRAPH OF BACTERIOPHAGE MS2 Final magnification 100,000 X. Provided by S. Spiegelman, University of Illinois, Urbana, Illinois. 29 T Agar (TU) : (Constituents per liter of water) Bacto-Tryptone (Difco) 10.0 gm Bacto-Agar (Difco) 12.0 gm NaCl 5.0 gm L Br o tli (Ti|) : (Constituents per liter of water) Bacto-Tiyptone (Difco) 10.0 gm Yeast Extract (Difco) 5.0 gm NaCl 10.0 gm Glucose 1.0 gm Adjusted to pH 7.0 with 1 N IlaOH Soft Agar (Tk) : 200 ml T agar (Tk) plus 175 ml L broth (Tit) Phage Buffer (Tk) : (Constituents per liter of water) Tris(hydroxymethyl)aminomethane 1.21 gm MgSO, 0.60 gm IlaCl 5.35 gm Adjusted to pH 7.U with 1 N IIC1 b. Media for Escherichia coli Bacteriophage 1IS2 L Broth (MS2): Same as L broth (Tk) plus 1.0 ml 2 M CaCl per liter L Agar (HS2): L broth (MS2) plus 15 g/1 Bacto Agar (Difco) Soft Agar (MS2) : L broth (MS2) plus 10 g/1 Bacto Agar (Difco) 30 Phage Buffer (MS2): (Constituents per liter of water) Tris(hydro:xymethyl)aminome thane 6.06 gm NaCl £.8$ gm Adjusted to pll 7.6 with 1 N HC1 (3) Coagulant and Polyelectrolytes (Coagulant Aids) The coagulant used in this study was aluminum sulfate [Alp (SO, ) _ .18 H p 0] marketed by Allied Chem. Corp., Morristown, N. J. The polyelectrolytes (coagulant aids) used are listed in Table 6. Two cationic and two anionic polyelectrolytes were selected. Stock and working solutions were prepared according to the instructions giren in the product literature . TABLE 6 POIZELKCTR0IXTES (COAGULANT AIDS) llame Type Manufacturer Primafloc C-7 Catfloc Primafloc A- 10 Coagulant Aid #2l;3 Cationic (polyamine) Cationic Anionic (polycarboxylic) Anionic Rohm and Haas Calgon Corp. Rohm and Haas Calgon Corp. (h) Water Distilled water demineralized through two Illco-Way Research Model Demin- eralizer (Illinois Water Treatment Co., Rockford, Illinois), and then 31 sterilized by autoclaving at 1$ psi for l£ min was used in all soluble aluminum studies. Sterilized laboratory deionized water was used in preparing the raw water for the coagulation and flocculation studies (jar tests). (5) Organic Materials The following organic materials were used in this study: a. Bovine Serum Albumin ($ percent). Prepared by Pentex Inc., Kankakee, Illinois, and distributed by Calibiochem, Los Angeles, California. b. Egg Albumin. Prepared by Allied Ghem. Corp., Morristown, N. J. c. Beef Extract. Prepared by Difco Laboratories, Detroit, Michigan. d. Wastewater and wastewater effluents. Collected fresh from the Urbana-Champaign Sanitary Distrist waste treatment plant and prepared by fil- tering through four layers of cheesecloth so as to remove the larger size particles. (6) Glassware All glassware used for the soluble aluminum studies were cleaned by soaking overnight in chromic acid f ol lowed by rinsing in tap water and deionized water. Glassware used for the coagulation and flocculation studies (jar tests) were cleaned by soaking overnight in Haemo-sol (Meinecke and Co., New York, N. Y.) followed by rinsing in tap water and deionized water, as suggested by Chang, Isaac and Baine (1953) • Sterilization was accomplished in a hot air oven at 200 C for one hr or longer. B. Chemical and Biological Assay Techniques (1) Determination of Aluminum A simple, rapid and sensitive method for determining soluble aluminum concentrations in the microgram range was required. The eriochrome cyanine R method originally introduced by Knight (i960) and later modified by Shu 11 and Guthan (196?), and Chaudhuri and Engelbrecht (1968) was used. The standard 32 aluminum curve she,™ in Figure h was used to determine the aluminum concentra- tion in unknown samples. In the procedure a 1$ ml sample was used. The sample was first titrated to pll U.5 with N/50 sulfuric acid and then one ml added in excess. This was followed by the addition of one ml of ascorbic acid (l g/l) and 10 ml of buffer solution (l 3 6 gm sodium acetate plus If) ml of 1.0 M acetic acid in one liter of deionized water). Finally 5 ml of 0.03 percent eriochrome cyamne R was added, mixed and immediately made up to a volume of $0 m i with deionized water. The sample was mixed again and allowed to stand for 5 to 10 min. The absorbance was read against a suitable blank on a spectrophotometer (Universal Spectrophotometer, Model Ifc, Coleman Instruments, Inc., Maywood, Illinois) using a wave length of 533' inn and a k cm light path. All readings were corrected for dilution due to added solutions. (2) Assay Procedure for Bacteriophages The assay procedure for bacteriophage Th was obtained from Dr. John W. Drake. Before assaying, the sample was diluted in L broth (Tit) to yield about 200 plaques per plate. A liquid top-agar mixture was prepared from 2.5 ml of soft agar (Tl*) at 1£°C, 0.25 ml of a log growth phase culture of E. coli BB cells, and 0.1 ml of the diluted virus sample. The top-agar mixture was then poured on solidified bottom agar (T agar) plates and in- cubated at 37°C for 12 to 2U hr. -Plaques" were counted with the aid of a Quebec colony counter. Triplicate plates were prepared from each sample to increase accuracy. The assay procedure used for bacteriophage MS2 was very similar to that of bacteriophage 1%. E. coli A19 was the indicator bacteria. The top agar was soft agar (KS2) a*d the bottom agar was L agar (MS2). Plates were incu- bated at 37°C for 6 to 8 hr before counting. 33 0.2 E LA < CO o 00 CD < 0.5 .0 2.0 ALUMINUM CONCENTRATION (p,g in 50 ml Reaction Volume) FIGURE h. STANDARD CURVE FOR ALUMINUM. 3U C. Experimental Technique and Equipment (l) Kinetics of Adsorption of Aluminum by Viruses All studies on the kinetics of adsorption of aluminum involved soluble concentrations of aluminum. Concentrations of aluminum were always kept below the solubility at the pH of the test (Figure 1) . Experiments v/ere carried out at three different pH values: pH £.0 (0.2M acetate buffer), pH 6.0 (0.2M succinate-sodium hydroxide buffer), and pH 9.0 (0.2M borate- potassium chloride-sodium hydroxide buffer) . All experiments were performed at room temperature (2k C to 2$ C) . In the procedure the reaction mixture (total volume 150 ml) was prepared by adding 95> ml of deionized water, 20 ml of buffer solution, 2$ ml of alumi- num solution (l ml = 0.2^ M min. In the filtration procedure, the first 5 ml of the filtrate were wasted; the remaining 1$ ml were saved for the determination of aluminum. Aluminum concentrations in these filtrate samples were determined by the eriochrome cyanine R method using a filtered reagent blank. Aluminum concentrations were also determined in duplicate 15 ml filtered samples from the control. The difference in aluminum concentrations between this and that in a filtered sample of the 35 reaction mixture was used to compute the quantity of aluminum adsorbed by the virus particles. The average of duplicate aluminum determinations on the unfiltered 15 ml samples from the control flask was denoted as "available aluminum. " This available aluminum concentration was always found to be slightly lower than the calculated concentration added. This indicated that some loss of aluminum from the liquid phase was talcing place. It is believed that this was due to the adsorption on glassware. A typical experimental protocol with accompanying calculations is shown in Table 7. TABLE 7 EXPERIMENTAL PROTOCOL FOR KINETIC STUDIES Deionixed pll Buffer Aluminum Virus Phage Water (ml) Solution Suspension Buffer (ml) (ml) (ml) (ml) Reagent Blank Reaction Mixture Control 120 95 95 20 20 20 10 25 25 10 10 Aluminum adsorbed = (Aluminum concentration in filtered sample from by the virus the control) - (Aluminum concentration in particles filtered sample of the reaction mixture) Available aluminum = Aluminum concentration in unfiltered sample from the control Filtration through a 0.U5n- Millipore membrane filter was used as a means for rapid physical separation of the virus particles from the reaction mixture. Viruses are known to adsorb on 0.U5m- Millipore membrane filters (Oliver, 1966b, and Wallis and Melnick, 1967b, 1967c) . Further, it has been demonstra- ted that E. coli bacteriophages adsorb on 0.U5M- Millipore membranes (Hoff and 36 Jakubowski, 1965, and Loehr and Schwegler, 1965) . Preliminary tests indicated that under the experimental conditions used in this study, not more than 9 to 10 percent of the virus particles were passing through the membrane filters. (2) Virus Inactivation Studies Experiments to evaluate inactivation of bacteriophages Tk and MS2 in the presence of soluble concentrations of aluminum were carried out at pH 5.0 (0.2 M acetate buffer) and at room temperature (2k C to 25 C) . Experimental tubes were prepared by adding 3.3 ml of buffer solution, 1.0, 5.0 or 7.0 ml of aluminum solution (l ml - 0.25 u.g aluminum), 0.5 ml of a suitable dilu- tion of the virus stock suspension to yield a final virus concentration of 10-10 /ml, and deionized water to make up to a final volume of 25 ml. A control tube was also prepared without any aluminum. Samples for virus assay (0.1 ml) xrere withdrawn at 0, 3, 6, 12, and 2k hr and the virus titer deter- mined by the standard technique. Duplicate experimental tubes were also pre- pared in which 0.5 ml of phage buffer was added instead of the virus dilution. Samples were withdrawn from each of these tubes for determination of "available aluminum. " (3) Quantitative Studies on Adsorption of Aluminum by Viruses a. Studies on Adsorption of Aluminum by Viruses These studies were carried out at pH 5.0 (0.2 M acetate buffer) and at room temperature (2k C to 25 G) . A procedure similar to that used in the kinetic studies was followed for preparation of reaction mixtures, blanks, and control flasks. Virus concentrations were varied from 2 x 10 /ml to 1.8 x 10 /ml. Since it was evident from the kinetic studies that adsorption of aluminum by the virus particles was very rapid and took place at least within the first 30 to l£) sec, samples from the reaction flasks were withdrawn and filtered 5 to 10 min after the start of the reaction. The quantity of aluminum 37 adsorbed by virus particles and "available aluminum" were computed in the same manner as before. b. Aluminum Saturation Curves Experiments were also carried out to obtain saturation curves for aluminum adsorption by different concentrations of bacteriophage TU. Studies were made at pH values of 5.0 (0.2 M acetate buffer), 6.0 (0.2 H succinate-sodium hy- droxide buffer), and 9.0 (0.2 M borate-potassium chloride- sodium hydroxide buffer), and at room temperature (2ii C to 25 C). Using a procedure similar to the one used in the studies on adsorption of aluminum by viruses, quantities of aluminum adsorbed by a particular concentration of bacteriophage TJj. particles at varying "available aluminum" concentrations were computed and plotted, (h) Quantitative Studies on Virus Removal by Chemical Coagulation and Flocculation (Jar Tests) A bench-scale apparatus employing a carbonic acid-bicarbonate buffer system for pH control was set up in the laboratory for coagulation and floccu- lation studies (Figure 5) . Bicarbonate buffer was used because this is the main buffering system present in natural surface waters (Chang et al., 1958a) . All experiments were performed at room temperature (2I4. C to 25 C) . liixing was provided with a six-place multiple stirrer (Phipps and 3yrd, Richmond, Va.). The tachometer readings were found to agree with paddle rpm values, specially constructed paddles (Figure 6), made from 0.25 in. stainless steel rod, were used in an attempt to provide more uniform velocity gradients than is thought to occur with the conventional 1x3 in. paddles. The paddles were cleaned and sterilized before each experiment using the standard procedure for cleaning glassware. Control of pli, over the range of 8.3 to 5.0, was obtained by introducing various mixtures of air and C0 ? into the partially confined atmosphere above 38 CM -- O CJ < CD O ra i — (D CL 1 o X !_ •— ■M CO i_ u cu 0) "D i_ r— QJ i_ U-l 0) •— T3 a ■m C HD oo co o dJ • — _l 0) (U •t-i 1 — ^— m mm J3 O. c VO IT] •— •— • — ■M J3 x: !_ « — F +J ra 3 O •— . > z: <_> 3 en \ o h- < < o c_> q; o < < < CO \- co LU 1- cc CO LU o ID h- co S2 < i- cc < Q —I r> o c_> — <_> H O < _J CO < LU 39 C0 2 -Ai r 1 1/V Stainless Steel Rod 1 Liter Pyrex Beaker FIGURE 6. SCALE DRAWING OF REACTION VESSEL FOR COAGULATION AND FLOCCULATION STUDIES (JAR TESTS). ko each beaker. By measuring the individual gas-flow rates prior to blending, various mixtures of the two gases could be obtained. The blended gas was distributed from a central manifold to the individual beakers. Each experi- ment was initiated when an equilibrium pH was achieved; this usually required 0.5 to one hr. Values of pH were determined in individual beakers before the addition of coagulant and after the settling period. A Beckman Electromate pH meter was used. It was standardized daily using commercial buffer solutions of pH h, 7, sJid 9. The "raw" water was made up in 10 liters batches in a covered poly- ethylene vat. The raw water for all experiments contained 150 mg/1 of sodium bicarbonate, 120 mg/1 of Montmorillonite (Wyoming Bentonite), and k-5 x 10 /ml of bacteriophage Ti| or MS2. The clay was supplied by Ward's Natural Science Establishments, Inc., Rochester, N. Y. Montmorillonite was selected because it has been shown to be present in natural waters (Packham, 1962) . Samples for turbidity measurement, 30 ml each, were taken with a k mm bore pipet, 1.5 in. below the air-water interface and delivered to 50 ml beakers. All turbidity measurements were made after the samples had been in the instrument (Model i860, Hach Chemical Co., Ames, Iowa) for two min. Samples for virus enumeration, 0.1 ml each, were taken with 0.1 ml serological pipets, 1.5 in. below the air-water interface and immediately diluted in L broth for subsequent assay by the standard assay procedure for bacteriophages. The following method was used in performing an experiment: (l) six 975 ml aliquots of the "raw" water were placed in individual beakers situated on the mixing apparatus, (2) the six beakers were mixed simultaneously as pH was adjusted, (3) a calculated quantity of water was added to each beaker to Ill give a final total volume of one liter after the addition of chemicals, (k) while mixing rapidly at 100 rpm, a stoichiometric amount of sodium bicarbonate was added to each experimental beaker to neutralize the acidity due to alumi- num sulfate added in the next step and then chemicals (aluminum sulfate, polyelectrolytes, etc.) were added into all but one beaker which served as a blank, and (5) mixing was continued at 100 rpm for one min followed by 30 min of slow mixing at 20 rpm. Mixing was stopped gradually over a J4O sec period and the mixing paddles were left in place during the settling period. Samples for virus assay and turbidity measurements were x-jithdrawn after 30 min of quiescent settling. The pH of each system was determined, also. In experi- ments using coagulant aids, rapid mixing was extended for a period of one min after the polyelectro]yte addition, as recommended by the manufacturer. (5) Virus Recovery from Settled Floe After performing a jar test at the optimum pH and aluminum sulfate dosage for the particular virus, contents of both the blank and the experimental beakers were stirred at a high speed for l£ min using a magnetic stirrer so as to disperse the floe completely. While being stirred, duplicate i? ml aliquots were withdrawn from each beaker wi th a broken-tip pipet and poured into screw- cap tubes containing 1$ ml of an elutant. Deionized water, 3 percent beef extract, one percent bovine serum albumine, 0.1 M tris buffer (pH 8.0), and 0.2 M phosphate buffer (pH 8.0) were used as elutants. Tubes were kept at U°C with the contents mixed every 15 min in a vortex mixer. After 6 hr the content of each tube was centrifuged at 5,900 x g for 10 min and the super- natant assayed for the virus titer. Recoveries were calculated on the basis of the virus titer of the supernatant from the blank tube. V. RESULTS AMD DISCUSSION Results of all experiments are presented in graphical or tabular form. In order to facilitate the presentation, a discussion of the results follows each phase of the experimental work. For the most part, only typical results are shown and discussed although all results were substantiated by two or more replicate experiments. The terms "coagulation" and "f locculation" often denote different mean- ings. Recently there has been shown an awareness of the long existing general confusion in the literature over the meaning of these two terms. It should be recognized that in the overall process of coagulation and flocculation there is a cause-and-effect relationship. For reactions of the type which are en- countered in water treatment practice, destabilization and particle collision opportunity can be viewed as "causes. " Aggregation of the destabilized, collid- ing oar tide is then an "effect. " In order for a conventional water treatment plant to operate effectively both destabilization (accomplished by chemical addition) and particle collisions (accomplished primarily by mixing) must be provided. In presenting the results of this study, "coagulation" will refer to destabilization and "flocculation" will refer to aggregation. When used together, these two terms will denote destabilization and aggregation as pro- vided in a water treatment plant and will be referred as "the process of coagu- lation and flocculation. " A. Adsorption of Aluminum by Viruses (l) Kinetics of Adsorption Kinetics of adsorption of aluminum at pH values 5, 6, and 9 by bacterio- ohages Tk and MS2 are shown in Figures 7 and 8, respectively. Attempts were k2 k3 en :± Q LU 00 CC O co Q < 25 20 15 7 o D A Available Aluminum: 33 ng/1 6 pH 5.0; Virus Cone: ^.00 x 10 /ml 6 -c rhpH 6.0; Virus Cone: 3.67 x 10 /ml pH 9.0; Virus Conr: 3.67 x 10 10 /ml 7S X 1 5 10 TIME IN MINUTES 15 FIGURE 7- KINETICS OF ADSORPTION OF ALUMINUM BY BACTERIOPHAGE Jh en CO C£. O CO Q < 25 20 15 10 ^ Available Aluminum: 3*+ M-9/ Average Virus Cone: 6.25 x 10 /ml n. -± pH 5-0 A TIME IN MINUTES SE pH 6.0 5 X -c 2L pH 9.0 15 111* FIGURE 8. KINETICS OF ADSORPTION OF ALUMINUM BY BACTERIOPHAGE MS2 16 made to keep the virus concentrations constant for a particular bacteriophage in each system. However, it is to be borne in mind that the virus concen- trations reported here are as determined by the plaque count method and do not indicate the total number of actual virus particles in the sample. The latter can be determined only by an electron micrographic count (Luria, Williams and Backus, 195>l) • Although specific information is lacking, it is generally accepted that the precision of the plaque count method is on the order of 5 to 10 percent (Drake, 1967). Since processing of the sample took approximately 30 to U0 sec before the virus particles were physically separated from the reaction mixture, the data for the samples withdrawn from the reaction mixture immediately after mixing were plotted against 30 sec. It is evident from Figures 7 and 8 that adsorption of aluminum by virus particles is very rapid and takes place at least within the first 30 to k0 sec. Maximum adsorption is also attained at this time. For all practical purposes it may be regarded that adsorption of aluminum by virus particles is instantaneous. However, Chang, Isaac, and Baine (1953) estimated that 20 min would be required for aluminum -virus precipitate formation. It is to be noted that their experimental conditions were entirely different and that they attempted to separate the first-stage reaction (aluminum- virus complex for- mation) from the second-stage (precipitation and flocculation) by assuming that the second-stage reaction would not start in the absence of Si0„ which was found to be required for the formation of good floes under their experimental conditions. ' Time required for the first-stage reaction was assumed to be the longest contact time between aluminum and virus particles at which there were no significant differences in percentages of recovery of the virus between the tests performed in the presence of Si0 ? and those in the absence of Si0 ? . ho It is also seen from Figure 8 thai; the amount of aluminum adsorbed by the MS2 particles does not change appreciably with pH in the range $.0 to 9.0 and the variations are well within the limits of experimental error. However, the variations are more pronounced in the case of the Tit particles (Figure 7). It is not possible to provide a completely satisfactory explana- tion based upon the ionization of the charged groups on the coat protein of the bacteriophage Tiu Thus, the differences may be due to the variations in the amount of aluminum adsorbed on the surfaces of the glassware. (2) Quantitative Adsorption Figures 9 and 10 show the quantitative adsorption of aluminum by bacterio- phages TU and MS2, respectively. For studies with bacteriophage TU, three different virus stock suspensions and two different aluminum concentrations were used. The three virus stocks were prepared at different times and were stored for varying periods of time ranging from 1 to 6 months at h C. These figures show that a linear relationship exists between virus con- centration and the amount of aluminum adsorbed. In other words, aluminum is reacting stoichiometrically with the virus particles. It can also be con- cluded from observations with bacteriophage TU that the amount of aluminum adsorbed is not affected by the period and conditions of storage of the virus particles used. The amount of aluminum available also does not exert an effect providing the amount available is greater than the limiting concentration. To substantiate the reliability of the Millipore membrane technique used so far in these studies, a comparison was made between this method and another method involving high speed centrifugation. Virus particles along with adsorbed aluminum were separated from the reaction mixture by centri- fuging at 3U,800 x g for one hr. The difference in aluminum concentrations in supernatants of the control and the reaction mixture (Table 7) was used to hi "V. "-s. "V. CXI CD CT> i 3. 3. ro — o • • * LPi CNI vD c^ ro vD < < > > < < < us Stock 1 ; us Stock 1 1 ; us Stock 111; i_ i_ i_ r> > > □ < O — c i o 00 — ^o o X < DC o LA < o_ o DC UJ h- < CD >- cc o Q_ o cc o oo CO x> Q DC < < en CC C3 O o oa O CM (i/6t1) QBdciOsav wnNiwniv ua en CSJ < 0) 05 > < \° o\ CO CVJ O CM O la x £ CNJ CO CD < X Q. O CC LU h" <_> < >- CD O O X t- < o o CO X) ct: H- a. cc o CO Q < CO < CC X) CD Ll_ o o o CM (i/6t1) a3ayosa\/ wnNiwmv k9 compute the quantity of aluminum adsorbed. Results agreed very well. For a bacteriophage Tl± concentration of 7 x 10 /ml, the amount of aluminum adsorbed was 20 n,g/l. Using the phosphate modification method (Shull and Guthan, 196?) concentration of aluminum in the sedimented pellet was also determined for a mass balance on aluminum. The pellet was eluted with 0.2 M tris buffer (pH 8.0), digested in 1 N sulfuric acid at 15 psi for 30 min, neutralized to pH I4.5 with 1 N sodium hydroxide and the concentration of aluminum determined (Table 8). TABLE 8 MASS BALANCE ON ALUMINUM BY CENTRIFUGATION METHOD Virus concentration = 7 x 10 /ml; pH = £.0 (0.2 M Acetate Buffer) Initial concentration of aluminum in the control and in the reaction mixture =37.5 [i,g/l Control Concentration of aluminum in the control after centrifugation = 27.5 |ig/l Percent loss = % 1 '%" ' /' ~ x 100 = 26.7 37o Reaction Mixture Concentration of aluminum in the supernatant of the reaction mixture after centrifugation = 7.5 p,g/l Concentration of aluminum in the sedimented pellet = 22.5 |j,g/l Total aluminum recovered = 7.5 + 22.5 = 30.0 |j,g/l In this example (Table 8), the initial concentration of aluminum in the control and in the reaction mixture was 37.5 u.g/1. The concentration of aluminum in the control after centrifugation was 27.5 p,g/l« This was the amount of aluminum available for reaction with the Ti| particles. The 26.7 50 percent loss is assumed to be due to adsorption on the glassware and on the wall of the centrifuge tube. A total recovery of 30.0 p.g/1 of aluminum was made from the supernatant and the sedimented pellet from the reaction mixture. This is within 9 percent of the amount of aluminum available (27.5 u.g/1). Hence excellent mass balance was obtained. In order to further investigate the quantitative nature of aluminum adsorption, saturation curves for aluminum adsorption for different bacterio- phage Tlj. concentrations at pH values 5.0, 6.0, and 9.0 were obtained. These curves are shown in Figures 11, 12, and 13. The saturation curves follow the same pattern at pH values 5.0, 6.0, and 9.0. It can be seen that when the concentration of available aluminum is less than the limiting concentration, almost complete adsorption of aluminum from the solution takes place. The plateau regions of the saturation curves indicate complete saturation of the adsorption sites on the virus coat protein. Consequently, the quantity of aluminum that can be adsorbed on the virus coat protein surface is constant,. These adsorption data were then plotted according to the linear form of the Langmuir adsorption equation. This was done using the least squares method of curvefitting to facilitate the calculation of the amount of aluminum adsorbed by a Tli particle at different pH values. The basic assumptions involved are that (l) all adsorbent sites are identical and that (2) no interaction takes place between molecules adsorbed on adjacent sites (Graham, 1959). The Langmuir equation can be expressed by n/ - 1 x C ^ - — + Z where C = concentration of aluminum in the solution at equilibrium, p,g/l q = aluminum adsorbed per unit of adsorbent at equilibrium, ^g/particle o CO o r>- O LTV X D. h- < o LU CD < X a. cr LU o CJ la < >- 00 ,-v 2 1 — O en H 75. o_ o s — ' cc -cr 2: 00 Z < O _J _l m < < _J cc — O < Ll. > < CO LU > CC O CSJ < cc < CO z> o J- o CO o CNl ( 1/61) Q3gyoSQV WflNIWfnV en DC O CO o < 30 20 10 Virus Cone: 5-67 x 10 /ml Vi rus Cone : 3 -67 x 10 /ml 20 30 ko AVAILABLE ALUMINUM (y,g/l) FIGURE 12. SATURATION CURVES FOR ALUMINUM ADSORPTION BY BACTERIOPHAGE T*+ AT pH 6.0. 50 $2 30 en CO O Q < 20 Vi rus Cone: 7-33 x 10 /ml. 20 30 ^+0 AVAILABLE ALUMINUM (yg/1) FIGURE 13. SATURATION CURVES FOR ALUMINUM ADSORPTION BY BACTERIOPHAGE T1+ AT pH 9.0. 53 Z = saturation ratio, pg/particle K = constant, particle/1 Adsorption data of Figures 11, 12, and 13 fit well to the linear form of the isotherm (Figures ll;, 15>, and 16) . The amounts of aluminum adsorbed by a Th particle at different pH values were calculated from these plots. These data are given in Table 9. TABLE 9 MOUNTS OF ALUMINUM ADSORBED BY A BACTERIOPHAGE ?k PARTICLE AT DIFFERENT pH VALUES Aluminum adsorbed PH |i g/particle atoms/particle 5.0 3.31 x 10" 13 7.37 x 10 3 6.0 2.79 x 10" 13 6.21 x 10 3 9.0 2.83 x 10" 13 6.30 x 10 3 Even though saturation curves were not obtained for the bacteriophage MS2 - aluminum system, an estimate can be made of the amount of aluminum adsorbed by a MS 2 particle at pH 5.0. Taking values from Figure 10, this -13 3 appears to be 2.05> x 10 (ig/particle or k*6 x 10 atoms/particle. Due to the smaller size of a MS2 particle comoared to a fh particle, lesser amount of aluminum is adsorbed. It is evident that the amounts of aluminum adsorbed by a Tl| particle at pH values 6.0 and 9.0 are same and the variations reported are well within the limits of experimental error. However, it is not possible to provide a com- pletely satisfactory explanation for the difference in the amounts of aluminum ft < E 'i 1 n — o o o o o ' — r^A LA V Vi rus Cone : 1 V i rus Cone : 6 Vi rus Cone : 3 o\ o □ <\ °\ o \ CM \ X CM O " 1 d\ + O X LA CM -4" — II CT I] >o \ o o O LA O J" o OA O CM < o o < I -cr CD < Q- O * O LU _ •~ I- < =3 CD <-y LL W O Z O < — I— h- < 0- o ^ 00 O Q — < h- Q- a: C£ — o => oo s: Q CD < 2 < U I o O Ll _l O- O o -—v I— o CD < I- CD OO u >- X. oo ID CD LA CM O CM i\ 01 * b/3 # v^3 LA ca u u c c O 1 — 1 \ (/> (/I D 3 !_ 1_ > > D O LI \ \ p c_> CM o X o> LA CO + CO o X LA || □\ cr \o \ \ <_> \ LA on o CJ < O- o — z QC o LA *" s LU CM * (- h- \ <-> < CD < => :i ccj cy < Q- o 1- Q CC cm < O OC" Z CO 1— O Q 2 — < LU h- (_> Q- cc z cc — o O ZD o Q CJ 2: < -z. cc CD o — o 1 O Ll _l Q- o I— o oc *> CO 3Z Q. < h- CO LO O >- DC CD O CM £[01 x b/0 i>6 E E \ \ \ ° o o o o X X fv-, r-^ r»~. \D r-^ on o > s Cone s Cone 3 D L- L. > > T D O o \ rsl y o X CO rA + CO O X • II cr o \ \ \ LT\ CO o CM < o 1 CO J- \- LU < X Q_ O . — z oc LA LU — CM „ — .. 1- 1- *— ^ < s < => CD QD o" d. LU z CJ> •• < — z 1- h- o < Q- — Q CC O 1— CM < Z 00 QC 1- — < z 1- LU Q- cc o cc — z O => o 00 2: <_> < Z LA 2: < Q- o => o --^ — h- o QC I- • O CD — CT\ — — CC _l CD X — Q. o> 1 < h- CO CO o >- 2: 00 vD a: £[ 0[ * b/ 3 57 adsorbed at pH values 5.0 and 6.0. The ionization of the charged groups on the bacteriophage Tit coat protein may not be appreciably different at these two pH values. The fact that a stoichiometric amount of aluminum is adsorbed by a TU particle at pH values between 5.0 and 9.0 leads to the likelihood of a reaction of the "complex formation" type. Coordination complexes between aluminum and some ionic group(s) on the coat protein of bacteriophage is quite likely. Physical adsorption involving electrostatic attraction is ruled out because of the excellent stoichiometry between aluminum and TU particles in spite of the fact that the predominant aqueous aluminum species at pH values 3.0 and 6.0 are positively charged whereas the species at pK 9.0 is negatively charged (Figure l) . The positively charged species are either Al (0H)*!± and Alj-CCH)*? or A1 6 (0H)*3 and AlgCOH)^. The negatively charged species is Al(0H). . It seems quite likely that aluminum bound in these hy- droxo complexes is complexing with the ionic group(s) on the coat protein of the bacteriophage. By considering the protein coat of bacteriophage TU, it is possible to make some approximate calculations as to the availability of ionic groups for the formation of coordination complexes with aluminum. Detail amino acid composition of the TU coat protein is available in the literature (Table 10) . The head of the phage particle is made up of about 300 identical protein sub- units (Edgar and Epstein, 1965), which is the major protein constituent of the phage (Stent, 1963) . A large protein generally consists of subunits which are the minimal physical units. Assuming about 200 amino acid units in a protein subunit, the total number of amino acid units in the head protein are about 6 x 10 . This estimate and the mean amino acid composition of the TU coat protein lead to the follox>ring approximate calculations. £8 TABLE 10 AMINO ACID COMPOSITION OF BACTERIOPHAGE Tk Amino Acid Mole Percent Fitch and Poison and Mean Susman (1965) Uyckoff (19U0) Lysin 6.k 7.1 6.75 Histidine 0.0 2.0 1.00 Arginine U.1 U.5 1+.30 Aspartate and Aspargine n.i n.i 11.10 Threonine 7.0 7.1 7.05 Serine 5.9 5.5 5.70 Glutamate and glutamine io. U 9.9 10.15 iroline U.i 5.3 U.70 Glycine 9.1i 11.8 10.60 Alanine 11.8 12.8 12.30 Valine 6.9 6.8 6.85 hethionine 2.1 1.1 1.60 Isoleucine 6.5 3.6 5.05 Leucine o.O 6.0 6.00 Tyrosine U.i 2.5 3.30 ihenylalanine U.3 3.1 3.70 'jnide Content ho . 6" — - 'Hole percent of total dicarboxylic acids. 59 Number of carboxyl groups = (Number of terminal carboxyl groups) + (Total number of iD-carboxyl groups due to aspartate and glutamate residues - number of id- carboxyl groups combined with ammonia in amide linkage) = 300 + {(11.10 + 10.15) x (100 1 ^ U6 ' 6) } x percent were never obtained. This is presumably due to the fact that the elutants used were not able to completely dissociate the 61 H < lo- ci: ZD GO 1.0 0.9 0,8 0.7 0.6 0.5 0.4 0.3 0.1 i i ^ Average Virus Cone: 6.23 x 10 /ml A — Cont rol — i+.7 p-g/1 available Al /\ 36.7 jig/1 available Al Q — 62.6 ^g/1 available Al 12 24 30 TIME IN HOURS FIGURE 17. INACTIVATION OF BACTERIOPHAGE Tk IN THE PRESENCE OF SOLUBLE ALUMINUM AT pH 5.0. 62 1.0 0.9 0.8 0.7 0.6 0.5 < a: ib 0.3 0.2 0.1 Average Virus Cone : 9.25 x 10 5 /ml ^ — Cont ro 1 — ^.7 u-g/1 available Al f\^ —36.7 M-g/1 available Al 12 2k 30 TIME IN HOURS FIGURE 18. INACTIVATION OF BACTERIOPHAGE MS2 IN THE PRESENCE OF SOLUBLE ALUMINUM AT pH 5.0. 63 TABLE 11 RECOVERY OF BACTERIOPHAGES FROM SETTLED FLOC FOLLOWING CHEMICAL COAGULATION AND FLOCCULATION Elutant Percent Recovered Bacteriophage Tii Bacteriophage MS2 Deionized Water 28. 50 3p Beef Extract 52.27 52.8b' 1% Bovine Serum Albumine 53 - 75 0.1 M Tris Buffer (pll 8.0) 1+2.05 0.2 M Phosphate Buffer (pH 8.0) 1+2.95 U5.25 bacteriophage particles from the floe resulting in infection of a host bacterial cell by more than one bacteriophage. Chang, Isaac and Baine (1953) were able to recover 60 percent of the removed viruses by redispersing the flocculated mixture at pH 7.6 with vigor- ous stirring. It is evident from these observations that disposal of water treatment plant sludges treating water containing pathogenic viruses might constitute a public health hazard. Especially in the case of land disposal of water treatment plant sludges, this might constitute a ground water con- tamination problem. C. Virus Removal by Chemical Coagulation and Flocculation (Jar Tests) (l) Optimum pll and Aluminum Sulfate Dosages In presenting the results of the coagulation and flocculation studies (jar tests), virus removals are reported as percent removals. It should be noted that for an initial input virus concentration of k x 10 /ml, a removal of 99 oercent would mean a reduction in the virus titer from k x 10 /ml to 6k 3 k x 10 /ml in the supernatant. Correspondingly a removal of 98 percent 5 3 would mean a reduction in the virus titer from k x 10 /ml to 8 x 10 /ml and so on. Figure 19 shows the removal of bacteriophage Tij. ajid clay turbidity at different pJI values and coagulant dosages. It is interesting to note that virus removal closely parallels turbidity removal. It can be seen from this figure that based on both virus and turbidity removal, the optimum aluminum sulfate dosage and pH for bacteriophage Tl; removal was i|0 to 50 rag/1 (1.2 x 10~ M to 1.5 x 10 M as aluminum) and the lowest pH studied (5.2U). The highest removal obtained was 98 percent. All subsequent experiments with bacteriophage Tk were performed at pH values ranging from 5.2 to 5.3 with 50 mg/1 of aluminum sulfate. Removal of bacteriophage MS2 is shown in Figure 20. The optimum aluminum sulfate dosage and pH was U0 to 50 mg/1 (1.2 x 10 M to 1.5 x 10 M as alumi- num) at pH 6.0. Removal was 99.9 percent. Turbidity removal curves for IIS 2 were similar to those shown in Figure 19. Removals higher than 99.3 percent were never obtained at pH 5.10 even with higher aluminum sulfate dosage. However, all subsequent experiments with bacteriophage MS2 were performed at pH 6.0 with 50 mg/1 of aluminum sulfate. (2) Effect of Calcium and Magnesium on Virus Removal Chang et al. (1958b) believed that the presence of calcium and magnesium ions in raw water interfered with virus removal by reducing the rate of coagulant-cation bacterial-virus complex formation. Experiments were per- formed to investigate this phenomenon in further detail. Figure 21 shows the kinetics of adsorption of aluminum by bacteriophage Tk at pH values 5.0 and 9.0 in the presence of 50 mg/1 of each of the cations calcium and mag- nesium. Effect of these two cations on the removal of bacteriophage Tk by 100 90 £ 80 > o z: LLl IX. 00 en Average Input Virus Cone: h.S x 10 /ml >- CD a: ALUMINUM SULFATE DOSE (mg/1) FIGURE lg. REMOVAL OF BACTERIOPHAGE Jh AND CLAY TURBIDITY BY COAGULATION AND FLOCCULAT I ON . 66 00 CO 95 90 85 pH 6.00\^ ' \pH 7-0 \ pH 8.50 * pH 5-10 Average Input Virus Cone: 3-9 x 10 /ml 20 U0 60 80 00 ALUMINUM SULFATE DOSE (mg/1) FIGURE 20. REMOVAL OF BACTERIOPHAGE MS2 BY COAGULATION AND FLOCCULATION. 67 on CD o CO < 25 20 15 10 Available Aluminum: 39 jig / 1 Average Virus Cone: 2.5 x 10 /m O A No Ca and M 9 1 A a A — 50 mg/1 Ca and Mg 1 pH 9-0 1 4 A * i _____ J i 1 2 4 ^ c > ! 1 15 TIME IN MINUTES FIGURE 21. KINETICS OF ADSORPTION OF ALUMINUM BY BACTERIOPHAGE Jk IN THE PRESENCE OF CALCIUM AND MAGNESIUM. 68 coagulation and flocculation (jar tests) is shown in Table 12. It was not possible to conduct jar tests at pH 9.0 due to the limitation of the carbonic acid-bicarbonate buffer system, the upper limit of pH for this system being 8.3. TABLE 12 REMOVAL OF BACTERIOPHAGE Tk IN THE PRESENCE OF CALCIUM AND MAGNESIUM Average Input Virus Concentration: 1|.05 x 10 /ml; pH range: 5.1 - 5.U Aluminum Sulfate: 50 mg/1; Average Turbidity: 12.5 JTU Calcium Magnesium Percent Removal mg/1 mg/1 97.86 25 25 98.02 50 50 97.91 It is evident that presence of calcium and magnesium ions up to a con- centration of 50 mg/1 does not interfere either with kinetics of adsorption of aluminum or with removal of bacteriophage Tl| by coagulation and floccula- tion. The fact that presence of bivalent cations like calcium and magnesium does not change the kinetics and stoichiometry of aluminum virus interaction is further evidence for the formation of coordination complexes between alumi- num and virus coat protein. (3) Effect of Organic Matter on Virus Removal It is logical to assume that the presence of extraneous organic matter of proteinaceous character will interfere with virus removal by coagulation and flocculation. This may occur because of the competitive action of the organic 69 matter in question with the virus particles in the coagulation and floccula- tion reaction. To study this effect, virus removal in the presence of al- bumin, wastewater and wastewater effluent was studied. Table 13 shows the removal of bacteriophage TI4. in the presence of egg albumine, bovine serum albumin, settled wastewater and wastewater effluent. Removal of bacteriophage MS2 in the presence of settled wastewater and wastewater effluent is shown in Table Hi. It is seen that egg albumin, bovine serum albumin and settled wastewater interfered with bacteriophage Tli removal in the pH range 5>.1-5.1| (Table 13). The albumins also interfered with the flocculation process as evidenced by lower turbidity removals. The effect of wastewater effluent was not very much pronounced either on turbidity removal or on the removal of bacteriophage Tlj. in the pH range £.1 - $.h* However, considerably lower removals were ob- tained for turbidity and bacteriophage MS2 in the pH range 5.9-6.1 in the presence of settled xvastewater or wastewater effluent (Table lU) . These observations indicate that the process of coagulation and flocculation may not be expected to operate with high efficiency if the raw water contains organic matter. ik) Effect of Preformed Floe on Virus Removal At this stage it is clear from the discussions in the preceding sections that the nature of the interaction between aluminum and virus is most probably that of a coordination complex formation. Consequently it follows that inti- mate contact' between the virus particles and the soluble aluminum species is necessary before the formation of any hydra ted aluminum oxide precipitate in a coagulation and flocculation system. Experiments were performed to study this aspect by adding the virus stock suspension to the water at various times after the addition of aluminum sulfate. Table 1$ shows the results of this study. 70 TABLE 13 REMOVAL OF BACTERIOPHAGE Tk BY COAGULATION AND FLOCCUIATION IN THE PRESENCE OF ORGANIC MATTER Average Input Virus Concentration: 3.58 x 10 /ml; pH range: 5.1 Aluminum Sulfate: 50 mg/1; Average Turbidity; 12.5 JTlT - $.k Organic Hatter Concentration Percent Removal Bacteriophage Tk Turbidity Egg Albumin Bovine Serum Albumin Settled VJastewater' 'Wastewater Effluent' mg/1 10 it 20 ii 30 it 5o ii mg/1 10 ii 20 ii 30 ii 5o ii ml/1 200 ii ml/1 200 u 98.02 96.89 96.38 95.29 9U.83 98.65 97.67 97.63 96.85 96.00 97. hO 95.71 97. UO 97.03 99.00 98.21; 98.20 98.08 95. 8U 99.12 98.80 98.60 90.UO 9$. kk 98.37 98.63 98.37 98.77 Initial turbidity values ranged from 18 to 32 JTU when settled wastewater was added. Characteristics: 5-day BOD (mg/1) Total suspended solids (mg/l) Volatile suspended solids (mg/l) Raw 'Wastewater Wastewater Effluent U6 21 210 37 19k 11 71 TABLE 1U REMOVAL OF BACTERIOPHAGE MS2 BY COAGULATION AND FLOCCULATION IN THE PRESENCE OF WASTEWATER AND WASTEWATER EFFLUENT Average Input Virus Concentration: 3.87 x 10 /ml; pli range: 5.9 Aluminum Sulfate: $0 mg/1; Average Turbidity: 12.5 JTU - 6.1 Percent Removal Bacteriophage MS 2 Turbidity Ho Settled Wastewater or Wastewater Effluent 200 ml/1 Settled Wastewater 200 ml/1 Wastewater Effluent** 99.82 98.10 89.80 92.10 9U.00 93.60 'Initial turbidity was 19.0 JTU when settled wastewater was added. Characteristics: 5-day BOD (mg/l) Total suspended solids (mg/l) Volatile suspended solids (mg/l) Raw Wastewater Wastewater Effluent 181 5o 253 10 178 C.) 72 TABLE 1$ REMOVAL OF BACTERIOPHAGE TI4. BY PREFORMED FLOCS Average Input Virus Concentration: q..O x 10 /ml; pFI range: $.1 Aluminum Sulfate: 50 mg/1; Average Turbidity: 12.5 JTU - 5.U Bacteriophage Tii added (minutes after aluminum sulfate addition) Percent Removal Remarks Bacteriophage TJU Turbidity 1 5 15 98. Oil 99.17 Control 80.60 97.0 Preformed floe not in situ. Bacteriophage Tii added one min after addi tion of pre- formed floe. 76.1*0 99.0 60.50 99. OU 37.25 99.17 73 It can be seen that preformed floe was not very effective in removing bacteriophage Tk from the water. Similar observations also have been re- ported in the literature (Chang, Isaac, and Baine, 1953). These observations demonstrate that intimate contact between the virus particles and aluminum is necessary before their incorporation into the floe masses of hydrated alumi- num oxide precipitates and subsequent removal by settling. In contrast, removal of clay turbidity by coagulation and flocculation with aluminum sul- fate is due to a physical interaction between aluminum and the clay particles resulting in adsorption of polynuclear aluminum hydrolysis species to the clay particles and consequent aggregation by interparticle bridging involving particle transport and chemical interaction (Stumm and 'Melia, I960). (5>) Effect of Polyelectrolytes on Virus Removal Commercially available synthetic polyelectrolytes (coagulant aids) are being used extensively in the water treatment industry for better coagulation and flocculation and longer filter runs. It was thought appropriate to study the effect of these polyelectrolytes on virus removal by chemical coagulation and flocculation (jar tests). Inactivation of bacteriophages Tk and MS 2 in the presence of the polyelectrolytes in deionized water was studied in order to be able to interpret the results of the jar tests using the polyelectrolytes, Samples for bacteriophage assay were withdrawn after a contact time of 1 hr. Table 16 shows the results of this study. Tables 17 to 22 show the results of the jar tests performed at pH values of approximately 5>.2 and 6.0 for bacteriophages Tk and MS2, respectively. These were the best pH values for their removal in jar tests. It is evident from Table 16 that both the cationic polyelectrolytes, Primafloc C-7 and Catfloc, inactivated bacteriophages Tk and MS2. This i^as presumably due to adsorption of the phage particles to the cationic sites on 71* TABLE 16 INACTIVATION OF BACTERIOPHAGES Tii AND MS2 IN THE PRESENCE OF POLYELECTROIZTES (COAGUIANT AIDS) Average Input Virus Concentration: 3.UU x 10 /ml Po lye le c tro ly te Concentration mg/1 pH Bacteriophage Percent Inactivation Primafloc C-7 0.5-1.5 0.5 5.2 6.0 Tli MS2 81* 85 Catfloc o.5-io.o 1.0 5.2 6.0 Til FiS2 77* 82 Primafloc A- 10 1.0-5.0 1.0 5.2 6.0 Ti| MS2 None None Coagulant Aid #2U3 1.0-5.0 1.0 5.2 6.0 Tit MS2 None None "Average value for the concentration range of the polyelectrolyte. 75 TABLE 17 REMOVAL OF BACTERIOPHAGE TI4. BY COAGULATION AND FLOCCULATION WITH ALUMINUM SULFATE AND CATIONIC POLYELECTROLITES AS COAGULANT AIDS Average Input Virus Concentration: h-h9 x 10 /ml; pH range: 3.1 Aluminum Sulfate: 50 mg/lj Average Turbidity: 12.5 JTU - $.k Cationic Po lye le c tr o ly te Percent Removal Concentration mg/1 Bacteriophage Tij. Turbidity Primafloc C-7 0.00 0.50 1.00 1.50 2.00 97.71 99.82 99.90 99.95 99.99 99.08 98.87 97.^6 95.50 9li.30 Catfloc 0.00 0.25 0.50 1.00 i.5o 97. Qh 98.U1 99.50 99.96 99.93 99.32 99.32 99.36 98.98 98. hi 76 TABLE 18 REMOVAL OF BACTERIOPHAGE MS2 BY COAGULATION AND FLOCCULATION WITH ALUMINUM SULFATE AND CATIONIC POLYELECTROLYTES AS COAGULANT AIDS Average Input Virus Concentrations: 2 .7U x 10 /ml; pH range: 5.9 - 6.0 Aluminum Sulfate: 50 mg/1; Average Turbidity: 12.5 JTU Cationic Po lye le c tr o lyte Concentration mg/1 Percent Removal Bacteriophage MS2 Turbidity 0.0 99. 6U 98.25 1.0 99.5U 98.25 Primafloc C-7 2.0 99.23 96.58 3.0 99.38 93.75 0.0 99. 7U 98.56 1.0 99.71 98. UO Catfloc 2.0 99.60 97.67 3.0 98.87 96. 2h 77 TABLE 19 REMOVAL OF BACTERIOPHAGE Tl* BY COAGULATION AND FLOCCULATION WITH CATIONIC POLYELECTROLYTES AS PRIME COAGULANTS Average Input Virus Concentration: U.98 x 10 /ml pH range: 5.2 - 5.5; Average Turbidity: 12.5 JTU Cationic Po lye le c tr o ly te Percent Removal Concentration mg/1 Bacteriophage Tli Turbidity Primafloc C-7 Catfloc 2.5 5.0 7.5 10.0 20.0 1.0 2.5 5.0 7.5 10.0 12.5 15.0 20.0 25.0 8U.U0 99.27 99.93 99.98 99.99 17.00 52. 1+0 85.70 97.22 99. Ok 99.52 99.81; 99.93 99.9k 89.10 98. U3 98.95 79.70 6.50 0.00 62.60 88.70 98.17 98.96 99.11 96.75 80.00 72.90 TABLE 20 REMOVAL OF BACTERIOPHAGE MS2 BY COAGULATION AND FLOCCULATIOM WITH CATIONIC POLIELECTROLYTES AS PRIME COAGULANTS Average Input Virus Concentration: 2.8 x 10 /ml pH range: 5.8 - 5.9; Average Turbidity: 12.5 JTU 78 Cationic Polyelectrolyte Concentration mg/1 Percent Removal Bacteriophage MS2 Turbidity 2.5 93.05 89.25 5.0 98.72 97.81; PrijTiafloc C-7 7.5 99.15 97.81; 10.0 97.30 80.00 20.0 95. UO 15.00 5.0 98.02 86.65 10.0 99.57 97.08 Catfloc 12.5 99.21 96.75 i5.o 98.7U 96.30 20.0 98.76 76.70 TABLE 21 REMOVAL OF BACTERIOPHAGE TU BY COAGULATION AND FLOCCULATION WITH ALUMINUM SULFATE AND ANIONIC POLIELECTROLYTES AS COAGULANT AIDS Average Input Virus Concentration: 2.78 x 10 /ml; pH range: 5.0 - 5.U Aluminum Sulfate: 50 mg/1; Average Turbidity: 12.5 JTU 79 Anionic Po lye 1 e c tr o ly te Concentration mg/1 Percent Removal Bacteriophage Th Turbidity 0.0 97.13 99.00 0.5 97.20 98.82 Primafloc A- 10 1.0 97.20 97.91 2.5 97.33 97.73 5.0 97.88 96.36 0.0 98.03 99. 1U 1.0 97.80 99.1k Coagulant Aid #2U3 2.0 97.22 9Q.95 5.0 96.93 98.18 10.0 96.60 96.57 80 TABLE 22 REMOVAL OF BACTERIOPHAGE MS 2 BY COAGULATION AND FLOCCUIATION WITH ALUMINUM SULFATE AND ANIONIC POLYELECTROLYTES AS COAGULANT AIDS Average Input Virus Concentration: 2.76 x 10 /ml; pH range: 5.9 - 6.0 Aluminum Sulfate: 50 mg/1; Average Turbidity: 12.5 JTU Anionic Po lye lee tr o ly te Primafloc A- 10 Concentration mg/1 0.0 0.5 1.0 2.5 5.0 Percent Removal Bacteriophage MS 2 Turbidity 99.78 99.70 99.3k 98.75 97.68 97.50 96.58 9k. k2 88.30 83. UO Coagulant Aid #2k3 0.0 1.0 2.0 5.0 10.0 99.61 98.55 97.53 98.36 95.22 97.20 89.60 90. UO 89.60 88.00 81 the poly-electrolyte molecules, which resulted in subsequent infection of one host bacterium by more than one phage particle during phage assay. However, no attempt was made to reactivate or to free the phage particles from the polyelectrolyte. The anionic polyelectrolytes, Primafloc A- 10 and Coagulant Aid #2U3, did not inactivate Tl\. and MS2 particles. This was due to the absence of cationic adsorption sites on the polyelectrolyte molecules. It is seen from Table 17 that both the cationic polyelectrolytes used were quite effective as coagulant aids for bacteriophage Tij. removal in the dosage range 0.5 to 1.0 mg/1. Turbidity removal xjas less efficient at higher dosages even though virus removal was higher. This was presumably due to virus inactivation by the polyelectrolyte per se . No improvement was noticed in the removal of bacteriophage MS2 (Table 18). Both the cationic poly- electrolytes were quite effective as prime coagulants (Tables 19 and 20) . Primafloc C-7 and Catfloc in the dosage ranges 5.0 to 7.5 mg/1 and 10.0 to 12.5 mg/1, respectively, gave the best results from the viewpoint of both virus and turbidity removal. Neither of the anionic polyelectrolytes used were effective as coagulant aids (Tables 21 and 22). This was presumably due to the absence of a sufficient concentration of calcium ions in the system which has been thought to be necessary for the action of anionic polyelectro- lytes (Packham, 1967) . This was also the reason for not using anionic poly- electrolytes as prime coagulants in this study. D. Qualitative Description of Virus Removal by Chemical Coagulation and Flocculation For purposes of attaining a better understanding of the removal of viruses from water by chemical coagulation and flocculation, it seems appropriate to present a qualitative description of the process. On the basis of the 82 results obtained in this study and the information available in the literature, it is possible to visualize the entire process. In a natural surface water, depending on the concentrations of cations like sodium and calcium, a certain fraction of the virus particles present remain reversibly adsorbed to the clay particles constituting turbidity due to the formation of a clay- cat ion- virus bridge (Carlson et al., 1968). The other fraction may be assumed to be free. Addition of a coagulant like aluminum sulfate to this water immediately results in the formation of certain hydro- lyzed polymeric multivalent aluminum species depending on the pH of the water (Black and Chen, 1967) . Interaction between aluminum species and viruses, other organic matter, and clay particles proceeds immediately. The interaction between aluminum and viruses and other organic matter is a very rapid one and presumably results in the formation of coordination complexes. On the other hand, the interaction between aluminum and the clay particles constituting turbidity results in adsorption of polynuclear aluminum hydrolysis species to the clay particles and consequent aggregation of the destabilized particles by interparticle bridging involving particle transport and chemical interaction (Stumm and 'Melia, 1968) . Furthermore, precipitation of hydrated aluminum oxide species proceeds simultaneously, incorporating the complexed virus particles and the aggregating clay particles which then groxtf into "floes" and ultimately settle down resulting in a clear supernatant. Presence of organic matter in the water can have considerable effect on the overall efficiency of the process by interfering with virus removal. This may occur because of the competitive action of the organic matter with the virus particles in the coagulation and flocculation reaction. VI. SUMMARY AND CONCLUSIONS It has been shown that removal of bacteriophages TU and MS2 from water by chemical coagulation and flocculation with aluminum sulfate consists of a primary reaction step which involves interaction between aluminum and virus coat protein. The reaction was found to be instantaneous and proceeded accord- ing to a definite stoichiometry. The kinetics and the stoichiometry of the reaction were not affected by the pH, period and condition of storage of the virus particles, the quantity of available aluminum or the presence of bivalent cations like calcium and magnesium as studied in this investigation. Aluminum adsorption data were found to fit Langmuir adsorption equation. Amounts of aluminum adsorbed by a single virus particle at pH values 5.0, 6.0, and 9.0 were calculated and found to be comparable. Considering the aqueous chemistry of aluminum, amino acid composition of the virus coat protein and the evidences of an aluminum- protein interaction as reported in the literature, it was concluded that the interaction between aluminum and virus possibly re- sulted in the formation of coordination complexes between aluminum and the carboxyl groups associated with the virus coat protein. The complexed viruses were not inactivated and active viruses could be recovered from the settled floe following their removal from water by coagulation and flocculation. The process of chemical coagulation and flocculation was found quite effective in removing bacteriophages Th and MS2 from water. The optimum coagu- lant dosages and pH values were J4O to 5>0 mg/1 of aluminum sulfate at pH £.2li for bacteriophage TU and at pH 6.0 for bacteriophage MS2. The highest re- movals attained were 98.0 and 99.9 percent, respectively. Presence of biva- lent cations like calcium and magnesium up to a concentration of 50 mg/1 each did not interfere with the efficiency of the process. Organic matter like 83 albumins and that associated with wastewater lowered the removal efficiency considerably. Commercially available cationic polyelectrolytes were found effective both as coagulant aids and as prime coagulants. Based on the findings of this investigation, the following conclusions may be drawn: (i) Removal of viruses by chemical coagulation and flocculation with aluminum sulfate comprises of a primary instantaneous reaction step which results possibly in the formation of coordination complexes between aluminum and the carboxyl groups of the virus coat protein. (ii) Virus Dar tides are not inactivated as a result of this interaction between aluminum and the virus particles and remain active in the settled sludge following their removal from water by coagulation and flocculation with aluminum sulfate. (iii) Chemical coagulation and flocculation is an effective process in removing viruses from water. Removals in the range 93.0 to 99.9 percent can be expected. (iv) Presence of bivalent cations like calcium and magnesium up to a concentration of $0 mg/1 each does not interfere with the efficiency of the process. (v) The efficiency of virus removal is reduced when the raw water con- tains organic matter. (vi) Intelligent use of commercially available cationic polyelectrolytes with or without hydrolyzed metal ions may markedly increase the efficiency of the coagulation and flocculation process. It seems appropriate to note the recent observations made by Stumm and O'Kelia (1968): "It is important to reemphasize that coagulation phenomena in natural systems are quite specific. This specificity arises 8£ from the fact that colloid stability is affected by colloid- solvent, coagulant- solvent, and colloid-coagulant interactions.... Overemphasis on electrostatic phenomena in studies of coagulation in natural systems can produce results that are inefficient, un- economical, or both. " VII. ENGINEERING SIGNIFICANCE The most significant result of this study is that a more complete under- standing of the removal of viruses from water by chemical coagulation and flocculation has been attained. This is very pertinent in view of the more stringent water quality standards which are foreseen in the near future. Con- sidering the rapid population growth and accompanying urbanization, there is going to be a greater demand for water for public consumption and other uses which, in turn, will require more water reuse. Thus, the removal of viruses from water supplies becomes extremely important. A good understanding of the basic mechanisms involved in the removal of viruses from water by chemical coagulation and flocculation and the role of various other parameters affecting the process should aid in developining design standards for water treatment facilities. It will enable such standards to be developed on a sound, realistic and rational basis. From the experimental results it is possible to extrapolate some generali- zations which are of practical significance. The foremost among the generaliza- tions is the interpretation of Tli. and M32 removal data in terms of viruses which may be more significant in water supplies, viz., human enteric viruses. Con- firmation of Ti| removal data by MS 2 which is very similar to picornaviruses (enteroviruses of man and other animals) in shape, size and the nucleic acid contained, shows that the process of chemical coagulation and flocculation may be quite effective in removing enteroviruses from water. From the results of jar tests using polyelectrolytes it is apparent that intelligent use of commer- cially available cationic polyelectrolytes with or without hydrolyzed metal ions may markedly increase the efficiency of the process. However, from economic 86 87 considerations, the use of the cationic polyelectrolytes as coagulant aids with hydrolyzed metal ions seem more favorable. Based on today's market, polyelectrolytes are rather expensive. From the results of virus inactivation studies and virus recovery from settled floe it is possible to extrapolate some useful information. The observation that viruses are not inactivated as a result of the complex for- mation and remain viable in the settled sludge immediately leads to the po- tential hazard for ground water contamination during land disposal of sludges from water treatment plants treating water contaminated with pathogenic viruses. However, more information should be gained in this area such as the fate in- cluding the survival of the viable viruses in sludge during land disposal before any definite conclusion can be reached. On the basis of the findings of Chang, Isaac and Baine (1953) that approx- imately 20 min would be required for the completion of the first-stage reaction (aluminum- virus complex formation) it may be speculated that the incremental addition of the coagulant may be advantageous in optimizing the process for virus removal. In such a process the procedure might be to first add aluminum sulfate to the raw water in an amount equivalent to or less than its solu- bility, talcing into accounL the pH of the system; second, provide sufficient contact time (20 min) for the formation of the aluminum-virus complex; and, third, add sufficient aluminum sulfate to bring the total amount added up to the predetermined optimum dosage for coagulation and flocculation. Following the second addition of coagulant, the usual period of flocculation and sedi- mentation would be included. However, the findings of the current study that the interaction between the aluminum and the virus is instantaneous immediately rules out such a speculation. 08 The observation that the presence of organic matter interferes with virus removal by coagulation and flocculation leads to another generalization. For virus removal, the process may be more reliable as practiced at a water treatment plant than at a wastewater renovation plant because of the presence of higher concentration of organic matter in a wastewater effluent. The findings of this study indicate that virus removal by coagulation and flocculation parallel turbidity removal. Robeck, Clarke and Dostal (1962) observed that effective coagulation and flocculation was an essential pre- requisite for effective virus removal by rapid sand filtration. They also observed that any breakthrough in turbidity through the filter was accompanied by a breakthrough in virus. Consequently, it may be suggested that in a water treatment plant care should be taken to produce a high quality effluent in terms of turbidity. Any breakthrough in turbidity should serve as a warning to the operator. This is particularly important during heavy pollution of the raw water and may become more critical when marginal chlorination is practiced. Finally, the results of this study suggest that the process of chemical coagulation and flocculation as it is practiced today can be quite effective in removing pathogenic viruses from water if proper care is taken to control pll and other parameters which affect the process. It should be noted that pll was not found to be an inroortant variable with respect to the formation of the aluminum-virus complex; however, such is not the case when both coagu- lation and flocculation are considered. VIII. SUGGESTIONS FOR FUTURE WORK On the basis of the results of the current study it is felt that further investigations should be pursued in the following areas. (i) Studies should be made to evaluate the efficiency of the chemical coagulation and flocculation process in removing viruses which may be signi- ficant in water supplies, viz., polioviruses. Initial virus concentrations to be used in these studies should be comparable to the expected density of enteric viruses in a oolluted surface water. This would require the use of concentration techniques for virus detection following their removal from water by chemical coagulation and flocculation. (ii) Virus removal efficiency of the process of chemical coagulation and flocculation should be evaluated when iron salts are used as coagulant. (iii) More studies should be made to find the effect of pH on virus removal by chemical coagulation and flocculation using commercial poly- electrolytes as prime coagulants and coagulant aids. (iv) Investigations should be undertaken to find the fate of viable viruses in water treatment plant sludge during land disposal in order to assess its potential as a source of ground water contamination, if any. 89 LIST OF REFERENCES Adams, M. H. 1959. Bacteriophag es. Interscience Publishers, Inc., New York, N. Y. Anderson, N. G., Cline, G. B., Harris, W. W., and Green, J. G. 1966. Isolation of Viral Particles from Large Fluid Volumes. In: Transmission of Viruses by the Water Route , G. Berg (ed.), John Wiley and Sons, New York, N. Y. Bancroft, P. M., Engelhard, W. E., and Evans, C. A. 1957. Poliomyelitis in Huskerville, Nebraska. J. Am. Med. Assoc. 161j:836. Berg, G. 196k > The Virus Hazard in Water Supplies. J. New England Water Works Assoc. 78:79. Berg, G. 1966. Virus Transmission by Water Vehicle. I. Viruses. Health Lab. Sc. 3:86. Berg, G. 1968. Standard Methods for Water and Wastewater - Detection of Enteric Viruses in Water and Wastewater. Prepared at the request of the Am. Pub. Health Assoc. Com. on Standard Methods (Unpublished). Berg, G., Berman, D., Chang, S. L. , and Clarke, N. A. 1966. A Sensitive Quanti- tative Method for Detecting Small Quantities of Virus in Large Volumes of Water. Am. J. Epidemiol. 83:196. Berg, G., Chang, S. L. , and Harris, E. K. 196U. Devitalization of Microorgan- isms by Iodine. I. Dynamics of the Devitalization of Enteroviruses by Elemental Iodine. Virology. 22:U69. Berg, G., Dean, R. B., and Dahling, D. R. 1968. Removal of Poliovirus 1 from Secondary Effluents by Lime Flocculation and Rapid Sand Filtration. J. Am. Water Works Assoc. 60:193. Berg, G., Scarpino, P. V., and Berman, D. 1966. Survival of Bacteria and Viruses in Natural Waters. Proc. Natl. Sym. on Quality Standards for Natural Waters, The University of Michigan School of Public Health, Ann Arbor, Michigan. Bier, M. , Bruckner, G. C, Cooper, F. C, and Roy, H. E. 1966. Concentration of Bacteriophage by Electrophoresis. In: Transmission of Viruses by the Water Route , G. Berg (ed.), John Wiley and Sons, New York, N. Y. Black, A. P., and Chen, C. 1967. Electrokinetic Behavior of Aluminum Species in Dilute Dispersed Kaolinite Systems. J. Am. Water Works Assoc. 59:1173. Carlson, H. J., Ridenour, G. M. , and McKhann, C. F. 19U2. Efficacy of Standard Purification Methods in Removing Poliomyelitis Virus from Water. Am. J. Pub. Health. 32:1256. 90 91 Carlson, H. J., Ridenour, G. M., and McKhann, C. F. 191+3. Effect of the Activated Sludge Process of Sewage Treatment on Poliomyelitis Virus. Am. J. Pub. Health. 33:1083. Carlson, G. F., Woodard, F. E., Wentworth, D. F., and Sproul, 0. J. 1968. Virus Inactivation on Clay Particles in Natural Waters. J. Water Poll. Control Fed. hO:R89. Chang, S. L. 1968. Waterborne Viral Infections and their Prevention. Bull. World Health Org. 38:U01. Chang, S. L. , Isaac, P. C. G., and Baine, N. 1953. Studies on Destruction of Bacterial Viruses in Water by Flocculation. III. Dynamics of the Removal of Bacterial Virus (Bacteriophage against Micrococcus pyogenes var. albus) in Water by Flocculation with Aluminum Sulfate. Am. J. Hygiene. FF2^3. Chang, S. L. , Stevenson, R. E., Bryant, A. R., "Woodward, R. L. , and Kabler, P. W. 1958a. Removal of Coxsackie and Bacterial Viruses in Water by Flocculation. I. Removal of Coxsackie and Bacterial Viruses in Water of Known Chemical Content by Flocculation with Aluminum Sulfate and Ferric Chloride Under Various Testing Conditions. Am. J. Pub. Health. 1*8:51. Chang, S. L. , Stevenson, R. E., Bryant, A. R., Woodward, R. L. , and Kabler, P. W. 1958b. Removal of Coxsackie and Bacterial Viruses in Water by Flocculation. II. Removal of Coxsackie and Bacterial Viruses and the Native Bacteria in Raw Ohio River Water by Flocculation with Aluminum Sulfate and Ferric Chloride. Am. J. Pub. Health. Ij.8:l59. Chaudhuri, M. , and Engelbrecht, R. S. 1968. Eriochrome Cyanine R Method for Aluminum Determination (Notes and Comment). J. Am. Water Works Assoc. 60: 618. Clarke, E. M. , Knowles, D. S., Shimada, F. T., Rhodes, A. J., Ritchie, R. C, and Donohue, W. L. 1951 • Coxsackie Virus in Urban Sewage. Can. J. Pub. Health. 1*2:103. Clarke, N. A., Berg, G., Kabler, P. W., and Chang, S. L. 1961*. Human Enteric Viruses in Water: Source, Survival and Removability. In: Advances in Water Pollution Research , Vol. II, W. W. Eckenf elder, Jr. (ed.), The Macmillan Company, New York, N. Y. Clarke, N. A., and Chang, S. L. 1959. Enteric Viruses in Water. J. Am. Water Works Assoc. 51:1299. Clarke, N. A., and Kabler, P. W. 1961*. Human Enteric Viruses in Sewage. Health Lab. Sc. 1:1*1*. Clarke, N. A., Stevenson, R. E. , and Kabler, P. W. 1956. Survival of Coxsackie Virus in Water and Sewage. J. Am. Water Works Assoc. lR3:677. Cliver, D. 0. 1966a. Detection of Enteric Viruses by Concentration with Polyethylene Glycol. In: Transmission of Viruses by the Water Route , G. Berg (ed.), John Wiley and Sons, New York, N. Y. 92 Cliver, D. 0. 1966b. Enterovirus Detection by Membrane Chromatography. In: Transmission of Viruses by the Water Route , G. Berg (ed.), John V/i ley and Sons, New York, N. I. Coin, L. , Mene trier, M. L. , Labonde, J., and Hannoun, M. C. 1965. Modern Microbiological and Virological Aspects of Water Pollution. In: Advances in Water Pollution Research , Vol. 1, 0. Jaag (ed.), Pergamon Press, Inc., New York, N. Y. Cookson, J. T., Jr. 1966. Kinetics and Mechanisms of Adsorption of Escherichia coli Bacteriophage Tii to Activated Carbon. Doctoral Dissertation, Calif. Inst, of Tech., Pasadena, Calif. (Unpublished). Dennis, J. M. 1959. 1955-1956 Infectious Hepatitis Epidemic in Delhi, India. J. Am. Waterworks Assoc. 51:1288. Drake, J. W. 1967. Personal communication. Edgar, R. S., and Epstein, R. II . 1965. The Genetics of a Bacterial Virus. Sc. Amer. 212:2,70. Eliassen, R., Drewry, W. , Chen, 0., and Tchobanoglous, G. 1966. Studies on the Movement of Viruses in Ground Water (Annual Progress Report) . 'Water Quality Control Research Laboratory, Department of Civil Engineering, Stanford University, Stanford, Calif. Felix, S. P. 1965. U. S. Patent 3,21^,369. Fitch, W. M., and Susman, M. 1965. Amino Acid Composition of Some T Phages. Virology. 26:75U. Frankova, V., Cervenka, R., and Symon, K. 196U. Studium Vlivu Upravy Vody Koagulaci Siranem Hlinitym Na Odstraneni Virove Kontaminace. Ceskoslov. Hyg. (Prague). 9:U89. Also Bull.Hyg. UO:h32 (1965). Fruton, J. S., and Simmonds, S. 1958. General Biochemistry . John Wiley and Sons, Inc., New York, N. Y. Gard, S. 19U0. Method for Detecting Poliomyelitis Virus in Sewage. J. Exptl. Med. 71:799. Gartner, H. 1966. Retention and Recovery of Polioviruses on a Soluble Ultra- filter. In: Transmission of Viruses by the Water Route , G. Berg (ed.), John Wiley and Sons, New York, N. Y. Gilcreas, F. W., and Kelly, S. M. 1955. Relation of Coliform-Organism Test to Enteric-Virus Pollution. J. Am. Water Works Assoc. Ii7:683. Graham, D. 1959. Adsorption Equilibria. Chem. Eng. Prog. Sym. Sr. 2U: 17 - Hanson, L. E. 1969. Personal communication. Hartung, H. 0. 1961. Physiologic and Health Aspects of Water Quality (Task Group Report). J. Am. Water Works Assoc. 53:135U. 93 Hayward, M. L. 19U6. Epidemiological Study of Outbreak of Infectious Hepatitis. Gastroenterology. 6:50l±. Hoff, J. C, and Jakubowski, W. 1965. Studies on Membrane Filtration of Bacteri- ophage. Proc. Northwest Shellfish Sanitation Research Planning Conference, Gig Harbor, Washington. Hudson, H. E., Jr. 1962. High Quality Water Production and Viral Disease. J. Am. Water Works Assoc. 5U:1265. Kabler, P. W. , Chang, S. L. , Clarke, N. A., and Clark, H. F. 1963. Pathogenic Bacteria and Viruses in Water Supplies. Proc. 5th. San. Eng. Conf., Univ. of 111., Urbana, 111. Kabler, P. W. , Clarke, N. A., Berg, G., and Chang, S. L. 1961. Virucidal Effi- ciency of Disinfectants in Water. Pub. Health Reports. 76:565. Kellenberger, E. 1962. Vegetative Bacteriophage and the Maturation of Virus Particles. Adv. Virus Res. 8:1. Kelly, S. M. , and Sanderson, W. W. 1959. The Effect of Sewage Treatment on Viruses. Sewage and Ind. Wastes. 31:683. Kelly, S. M. , and Sanderson, W. W. 1961. Enteric Viruses in Wading Pools. Pub. Health Reports. 76:199. Kelly, S. M., Winsser, J., and Winkelstein, W. 1957. Poliomyelitis and Other Enteric Viruses in Sewage. Am. J. Pub. Health. 1;7:72. Kempf, J. E., Wilson, M. G., Pierce, M. E., and Soule, M. H. 19U2. Effect of Aluminum Hydroxide Sedimentation, Sand Filtration and Chlorination on the Virus of Poliomyelitis. Am. J. Pub. Health. 32:1366. King's College, New Castle, England. 1956. Report on the Research Carried Out by the P. H. Engineering Section (19U8 - 1955). (5). Removal of Viruses by Water Treatment. Bull. No. 6. Knight, A. G. I960. Residual Aluminum: Its Estimation and Control. Proc. Soc. Water Treat, and Exam. 9:72. Kruse, C. W. 1968. Mode of Action of Halogens on Bacteria and Viruses and Protozoa in "Water Systems (Annual Progress Report) . The Johns Hopkins University, Baltimore, Maryland. Little, G. M. 195U. Poliomyelitis and Water Supply. Can. J Pub. Health. U5:100. Loehr, R. C, and Schwegler, D. T. 1965. Filtration Method for Bacteriophage Detection. J. Appl. Microbiol. 13:1005. Lund, E., and Hedstrom, C-E. 1966. The Use of an Aqueous Polymer Phase System for Enterovirus Isolations from Sewage. Am. J. Epidemiol. 8^:287. 9k Luria, S. E., Williams, R. C., and Backus, R. C. 1951. Electron Mi orographic Counts of Bacteriophage Particles. J. Bact. 61:179. Mack, W. N. , Mailman, W. L. , Bloom, H., and Kruger, B. B. 1958. Isolation of Enteric Viruses and Salmonellae from Sewage. I. Comparison of Coliform Enterococci Incidence to the Isolation of Viruses. Sewage and Ind. Wastes. 30:957. Malik, W. U., and Muzaffaruddin, M. 1965. The Interaction of Metals and their Hydrous Oxide Sols with Proteins - Interaction of Aluminum, Chromium, Iron, and their Hydrous Oxide Sols with Casein. Kolloid-Z. Z. Polymere. 206:55. McLean, D. M. 1961;. Contamination of Water by Viruses. J. Am. Water Works Assoc. 56:585. McLean, D. M. 1966. Transmission of Viral Infections by Recreational Water. In: Transmission of Viruses by the Water Route , G. Berg (ed.), John Wiley and Sons, New York, N. Y. Merrell, J. C. Jr., Jopling, W. F., Bott, R. F., Katko, A., and Pintler, H. E. 1967. The Santee Recreation Project, Santee, California. Final Report. U. S. Dept. of Int., FWPCA, Cincinnati, Ohio. Metcalf, T. G., and Stiles, V*. C. 1966. Survival of Enteric Viruses in Estuary Waters and Shellfish. In: Transmission of Viruses by the Water Route , G. Berg (ed.), John Wiley and Sons, New York, N. Y. Morgan, J. J. 1967. Personal communication. Mosley, J. ./. i960. Transmission of Viral Diseases by Drinking Water. In: Transmission of Viruses by the Water Route , G. Berg (ed.), John Wiley and Sons, New York, N. Y. Neefe, J. R., Baty, J. B., Reinhold, J. G., and Stokes, J., Jr. 19U7. Inacti- vation of the Virus of Infectious Hepatitis in Dr inking Water. Am. J. Pub. Health. 37:365. Neefe, J. R., and Stokes, J., Jr. 19U5. An Epidemic of Infectious Hepatitis Apparently Due to a Waterborne Agent. J. Am. Med. Assoc. 128:1063. Overby, L. R., Barlow, G. H., Doi, R. H., Jacob, M., and SpiegeLman, 3. 1966. Comparison of Two Serologically Distinct Ribonucleic Acid Bacteriophages. I. Properties of the Viral Particles. J. Bact. 91:kk2- Packham, R. F. 1962. The Theory of the Coagulation Process - A Survey of the Literature. I. The Stability of Colloids. Proc. Soc. Water Treat, and Exam. 11: 50. Packham, R. F. 1967. Polyelectrolytes in Water Clarification. Proc. Soc. 'Water Treat, and Exam. 16:86. Paul, J. R., and Trask, J. D. 19l|2. Occurrence and Recovery of the Virus of Infantile Paralysis from Sewage. Am. J. Pub. Health. 32:235. 95 Paul, J. R., Trask, J. D., and Gard, S. I9I4O. Poliomyelitis in Urban Sewage. J. Exptl. Med. 71:76^. Plotkin, 3. A., and Katz, H. i960. Minimal Infective Doses of Viruses for Man by the Oral Route. In: Transmission of Viruses by the Water Route , G. Berg (ed.), John Wiley and Sons, New York, N. Y. Poison, A., and Wyckoff, R. W. G. I9J48 • The Amino Acid Content of Bacterio- phage. Science. 10 8:501. Poynter, S. F. B. 1968. The Problem of Viruses in Water. Proc. Soc. Water Treat, and Exam. 17:187. Puck, T. T., Garen, A., and Cline, J. 1951. Mechanism of Virus Attachment to Host Cells. I. The Role of Ions in the Primary Reaction. J. Exptl. Med. 93:65. Putnam, F. W. 1953. Bacteriophage: Nature and Reproduction. Adv. Prot. Chem. 8:175. Rao, V. C, Sullivan, R., Read, R. B., and Clarke, IT. A. 1968. A Simple Method for Concentrating and Detecting Viruses in Water. J. Am. Water Works Assoc. 60:1288. Rawal, B. D., and Godbole, S. H. I96I4. A Simple Tissue Culture Method for Detection of Viruses in Drinking Water: Sample Incorporation Method. Env. Health (India). 6:231;. Rhodes, A. J., Clark, E. M. , Knowles, D. S., Goodfellow, A. M. , and Donohue, W. L. 1950. Prolonged Survival of Human Poliomyelitis in Experimentally Infected River Water. Can. J. Pub. Health. ]RL:lU6. Rhodes, A. J., and VanRooyen, C. E. 1962. Textbook of Virology , Williams and Wilkins, Baltimore, Maryland. Robeck, G. G., Clarke, N. A., and Dostal, K. A. 1962. Effectiveness of Water Treatment Processes in Virus Removal. J. Am. Water Works Assoc. 5U:1275. Salahuddin. 196U. Metal Ion-Binding Capacity of Gelatin and Transfusion Gelatin. Min. Tan. Paper Sym., Madras, India. Salahuddin., and Malik, U. 1962. Studies on the Interaction of Metal Ions and their Hydrous Oxide Sols with Proteins. Part I. Interaction of Aluminum, Iron, and their Hydrous Oxide Sols with Gelatin. J. Ind. Chem. Soc. 39:693. Salahuddin., and Malik, U. 196k. The Interaction of Metal Ions and their Hy- drous Oxide Sols with Proteins. V. Interaction of Aluminum, Iron, Cobalt, Nickel, Silver, and their Hydrous Oxide Sols with Transfusion Gelatin. J. Ind. Chem. Soc. I4I: 365. Schulman, J. II., and Dogan, II. J. 1952. Tanning of Fatty Acid and Protein Monolayers by Metal Ions (Chromium, Aluminum, Iron, and Copper) . Nature. 70:31. 96 Senn, C. L. , Berger, B. B., Jensen, E. C, Ludwig, H., and Shapiro, M. 19ol. Coliform Organisms as Index of Water Safety (Progress Report by the Committee on Public Health Activities). J. San. Eng. Div., Proc. Am. Soc. Civil Engrs. 87:SA6, 1±1. Shull, K. E., and Guthan, C. R. 196?. Rapid Modified Eriochrome Cyanine R Method for Determination of Aluminum in Water. J. Am. Water Works Assoc. 59:ll;56. ohuval, H. I., Cymbalista, S. , Fatal, B., and Goldblum, N. 1966. Concen- tration of Enteric Viruses in Water by Hydro-Extraction and Two-Phase Separation. In: Transmission of Viruses by the Water Route , G. Berg (ed.), John Wiley and Sons, New York, N. Y. Smith, J. E., and Courtney, R. J. 1966. Concentration and Detection of Viral Particles by Passive Hemagglutination. In: Transmission of Viruses by the Water Route , G. Berg (ed.), John Wiley and Sons, New York, N. Y. Spiegelman, 3. 1969. Personal communication. Standly, W, E., and Eliassen, R. I960. Status of Knowledge of Ground Water Conataminants. Report to the Federal Housing Administration. 1;23 p. Stent, G. S. 1963. Molecular Biology of Bacterial Viruses . W. H. Freeman and Company, San Francisco, California. Stevenson, R. E., Chang, S. L. , Clarke, II. A., and K abler, P. W. 195>6. Concen- tration of Dilute Virus Suspensions by Alum Flocculation. Proc. Soc. Exptl. Biol, and Med. 92:76U. Stumm, W. 1961|. Chemistry of Natural Waters in Relation to Water Quality. Proc. Sym. Env. Measurements, Robert A. Taft San. Eng. Center, Cincinnati, Ohio. Stumm, W. , and Morgan, J. J. 1962. Chemical Aspects of Coagulation. J. Am. Water Works Assoc. 5^:971. Stumm, W. , and 'Melia, C. R. 1968. Stoichiometry of Coagulation. J. Am. Water Works Assoc. 60:5lU. Wallis, C, and Melnick, J. L. 19o6. Concentration of Viruses on Aluminum Phosphate and Aluminum Hydroxide Precipitates. In: Transmission of Viruses by the Water Route , G. Berg (ed.), John Wiley and Sons, New York, N. Y. Wallis, C., and Melnick, J. L. 1967a. Concentration of Viruses on Aluminum and Calcium Salts. Am. J. Epidemiol. 85 : U5 9 . Wallis, C, and Melnick, J. L. 1967b. Concentration of Enteroviruses on Membrane Filters. J. Virol. 1:U72. 97 Wallis, C, and Melnick, J. L. 1967c. Concentration of Viruses from Sewage by- Adsorption on Millipore Membranes. Bull. World Health Org. 36:219. Wiebel, S. R, 3 Dixon, F. R., Weidner, R. B., and McCabe, L. J. 196U. Water- borne Disease Outbreaks, 19U6 - I960. J. Am. Water Works Assoc. £6:9U7. Yamamoto, N., Hiatt, C. W. , and Haller, W. 196U. Mechanism of Inactivation of Bacteriophages by Metals. Biochim. Biophys. Acta. 91:25>7. Name: Date of Birth: Education: Professional Experience: Publications: VITA Malay Chaudhuri May 27, 191*2 School Final, 195>6, Kumar Ashutosh Institution, Calcutta, India. West Bengal State Merit Scholarship. Intermediate in Science, 195>8, Presidency College, Calcutta, India. West Bengal State Merit Scholarship. Bachelor of Civil Engineering, 1962, Jadavpur University, Calcutta, India. Graduated in First Class with Honors. Sarada Gold Medal in Public Health Engineering. Master of Technology in Municipal Engineering, 1963, Indian Institute of Technology, Kharagpur, India. Candidate for Ph.D. in Sanitary Engineering, 196U-present, University of Illinois, Urbana, Illinois, U.S.A. Assistant Engineer, Calcutta Metropolitan Planning Organization (WHO Project: INDIA 170), Calcutta, India, 1963-61*. Research Assistant, Department of Civil Engineering, University of Illinois, Urbana, Illinois, U.S.A., 1961*-o9. Instructor, Department of Civil Engineering, University of Illinois, Urbana, Illinois, U.S.A., 1969-present. "Upflow Contact Basin - A Simple and Rapid Method for Turbidity Removal. " Environmental Health. 8:212, 1966. "Staining of Free-Living Nematodes with Alcoholic Eosin-Y Dye. " Co-authors R.S. Engelbrecht, J.H. Austin, and Melvin Goodrich. J. Am. Water Works Assoc. 60:1*18, 1968. "Erio chrome Cyanine R Method for Aluminum Determinations (Notes and Comments) . " Co-author R.S. Engelbrecht. J. Am. Water Works Assoc. 60:618, 1968. 98