SEPA
            United States
            Environmental Protection
            Agency
            Municipal Environmental Research EPA-600/2-80-004
            Laboratory         June 1980
            Cincinnati OH 45268
            Research and Development
Critical Review of
Virus  Removal by
Coagulation
Processes and pH
Modifications
i , 

-------
                RESEARCH REPORTING SERIES

Research reports of the Office of Research and Development, U.S. Environmental
Protection Agency, have been grouped into nine series. These nine broad cate-
gories were established to facilitate further development and application of en-
vironmental technology. Elimination of traditional grouping was  consciously
planned to foster technology transfer and a maximum interface in related fields.
The nine series are:

      1.  Environmental Health  Effects Research
      2.  Environmental Protection Technology
      3.  Ecological Research
      4.  Environmental Monitoring
      5.  Socioeconomic Environmental Studies
      6.  Scientific and Technical Assessment Reports (STAR)
      7.  Interagency Energy-Environment Research and Development
      8.  "Special" Reports
      9.  Miscellaneous Reports

This report has been assigned  to the  ENVIRONMENTAL PROTECTION TECH-
NOLOGY series. This series describes research performed to develop and dem-
onstrate instrumentation, equipment, and methodology to repair or prevent en-
vironmental degradation from point and non-point sources of pollution. This work
provides the new or improved technology required for the control and treatment
of pollution sources to meet environmental quality standards.
 This document is available to the public through the National Technical Informa-
 tion Service, Springfield, Virginia 22161.

-------
                                      EPA-600/2-80-OQl|
                                      June 1980
    CRITICAL REVIEW OF VIRUS REMOVAL BY

COAGULATION PROCESSES AND pH MODIFICATIONS
              Otis J. Sproul
      Department of Civil Engineering
         The Ohio State University
           Columbus, Ohio  1*3210
             Grant No. R8057H
              Project Officer

              John N. English
       Wastewater Research Division
Municipal Environmental Research Laboratory
          Cincinnati, Ohio  1*5268
MUNICIPAL ENVIRONMENTAL RESEARCH LABORATORY
    OFFICE OF RESEARCH AND DEVELOPMENT
   U.S. ENVIRONMENTAL PROTECTION AGENCY
          CINCINNATI, OHIO  1*5268

-------
                                DISCLAIMER

     This report has been reviewed by the Municipal Environmental Research
Laboratory, U.S. Environmental Protection Agency, and approved for publica-
tion.  Approval does not signify that the contents necessarily reflect the
views and policies of the U.S. Environmental Protection Agency, nor does
mention of trade names or commercial products constitute endorsement or rec-
ommendation for use.
                                     11

-------
                                 FOREWORD

     The Environmental Protection Agency was created because of increasing
public and government concern about the dangers of pollution to the health
and welfare of the American people.  Noxious air, foul water, and spoiled
land are tragic testimony to the deterioration of our natural environment.
The complexity of that environment and the interplay between its components
require a concentrated and integrated attack on the problem.

     Research and development is that necessary first step in problem
solution and it involves defining the problem, measuring its impact, and
searching for solutions.  The Municipal Environmental Research Laboratory
develops new and improved technology and systems for the prevention, treat-
ment, and management of wastewater and solid and hazardous waste pollutant
discharges from municipal and community sources, for the preservation and
treatment of public drinking water supplies, and to minimize the adverse
economic, social, health, and aesthetic effects of pollution.  This publica-
tion is one of the products of that research; a most vital communications
link between the researcher and the user community.

     This report is a state-of-the-art review of the literature concerned
with the removal and inactivation of virus by chemical coagulation processes
and pH modifications.  Such information is necessary for a rational ap-
proach to the development and standardization of optimum treatment conditions
for the removal and inactivation of viruses.
                                      Francis T. Mayo, Director
                                      Municipal Environmental Research
                                      Laboratory
                                     111

-------
                                 ABSTRACT
     Operation of advanced wastewater and water supply treatment plants to
assure virological safety of the effluent relies on each unit process re-
moving a finite number of viruses.  These treatment plants frequently use
chemical coagulation and precipitation at high pH with hydrated lime as part
of the process.  These treatment methods offer important opportunities for
removal and inactivation of viruses from water and wastewater.  This report
is a literature review which examined the effectiveness of these processes
in removing viruses.

     Coagulation of water and wastewater for enteric virus removal should
provide a removal of 90-99-999 percent of the influent viruses based on ob-
servations with polioviruses and Coxsackie A2.  Either ferric or aluminum
salts provide equal capability for virus removal when used in sufficient
dosage.  The required dosage is related to the conditions of the water and
should be sufficient to provide a maximum removal of turbidity.  The control
of these processes for virus removal can be obtained by monitoring removal
of turbidity.  The pH of the water appears to influence virus removal and
recommended pH values are 5-7 for virus coagulation with metallic coagulants
where hydrated metal oxides are formed.  The optimum pH where aluminum or
ferric phosphates are the sole precipitate is between 5-6.5-  Laboratory
studies have shown a slight reduction in removal of viruses when organics
are present as in treated effluents, but this is not supported by pilot
plant data.  Inorganics in the amounts in fresh waters do not appear to
influence virus removal when floe formation is adequate.

     Virus removal is unsatisfactory when polyelectrolytes are used as the
primary coagulant in the usual concentration ranges of 1 to 2 mg/1.  Virus
removals are not increased when polyelectrolytes are used as coagulant aids
if the coagulation process is otherwise acceptable.  Where polyelectrolytes
are used cationics would be preferred over the nonionics because of their
greater density of positive charges.  Anionic polyelectrolytes are not rec-
ommended since virus removal may be decreased in their presence.

     Enteric viruses are inactivated at high pH and physically removed by
absorption to precipitates formed under alkaline conditions in water.  These
viruses are, however, stable at pH values much lower than are obtained in
either water and wastewater treatment.  Virus inactivation in the absence of
precipitates at high pH is affected by organics, inorganics, pH, contact
time, temperature and type of virus.  The exact effect of each parameter on
virus inactivation cannot be determined from the present evidence, but the
few viruses which have been studied indicate that an inactivation in excess
of 90 percent should be obtained with a contact time of about 90 minutes
at a pH of 11.5-12.  Magnesium hydroxide and calcium hydroxylapatite have
greater absorptive capacities for viruses than does calcium carbonate.
Combined removal and inactivation of viruses in excess of 90-99 percent
should be expected in the lime flocculation of a typical biologically

                                     iv

-------
treated effluent with about 10 mg/1 of phosphorus.  Control of the process
should be maintained by monitoring pH and turbidity with the objective of
maximizing the turbidity removal.

     Recommendations are included for application of these technologies to
field situations.

     This report is submitted in partial fulfillment of Grant No. R805771
by The Ohio State University under the sponsorship of the U.S. Environmental
Protection Agency.  This report covers the period October 1, 1977 to
May 31, 1979-

-------

-------
                                  CONTENTS
                                                                    Page

Foreword	iii
Abstract	iv
Tables	viii
Acknowledgments	ix

     1.  Introduction	1
     2.  Conclusions	2
              Coagulation with metallic coagulants	2
              Coagulation with poly electrolytes	3
              Inactivation with pH extremes without precipitation	3
              Removals with precipitation at high pH	3

     3.  Conventional Coagulation	U
              Background	k
              Metallic coagulants	5
                   Amount of removal	5
                   Type of virus	8
                   Type of coagulant and dosage	9
                   Process control	11
                   Water conditions	13
                        Temperature	lU
                        Organic content	1^-
                        Inorganic material	15
                   Process reliability	15
              Polyelectrolytes	l6
              Mechanism of removal by coagulation	18

     k.  Virus Removal and Inactivation by pH and Lime
         Flocculation	20
              Background	20
              pH effects	20
                   Low pH	20
                   High pH	21
              Lime precipitation	2k
                   Laboratory evaluation	25
                   Pilot plant and full scale evaluation	28
              Process control	29
              Mechanism of Removal by high pH and precipitates	29

     5.  Technology Application for Virus Removal	31
              Background	31
              Coagulation	31
              Removal and inactivation with high pH	32

References	^
                                     vii

-------
                                  TABLES

Number

  1   Coagulation of Enteric Viruses with Metallic Coagulants 	   6

  2   Coagulation of Bacteriophages with Metallic Coagulants  	   7

  3   Virus Removal Versus Metallic Coagulant Dosage  	  10

  k   Dosages for Maximum Removal of f2 Phage and Turbidity
        from Natural Lake Water  	  12

  5   Virus Removal with Polyelectrolytes as Primary  Coagulant  ....  17

  6   Virus Removal at High pH Without Precipitation  	  22

  7   Loss of Infectivity of f2 in Floe-free Ca(OH)?  Treated
        Activated Sludge Effluent in 60 Minutes  	  23

  8   Virus Removal at High pH with Precipitation 	  26 & 27
                                    Vlll

-------
                              ACKNOWLEDGMENTS

     The assistance of Gary S. and Elaine L. Nault in the development of
the literature references is sincerely acknowledged.  John N. English
provided valuable assistance in locating information and reports for review
in the preparation of this report.
                                    IX

-------

-------
                                 SECTION 1

                               INTRODUCTION

     Operation of advanced wastewater and water supply treatment plants to
assure virological safety of the effluent relies on each unit process
removing a finite number of viruses.  The sum of the removal in each of
these barriers assures that safety of the finished water can be assumed.
Principal processes in both advanced wastewater and drinking water treatment
are coagulation, flocculation and sedimentation of the flocculent particles
which are produced.  It is known that viruses are removed in these processes
to varying degrees.  However, there has not been a systematic review of our
present state of knowledge on the removal of viruses by these processes.
This is particularly true in the removal of viruses by coagulation and floc-
culation under conditions where low or high pH conditions are obtained.  Con-
ditions of high pH are obtained during softening of water by precipitation
and in waste water reuse systems for the precipitation of phosphorus, but
the literature presents inconsistent data on virus removal by chemical
coagulation, particularly with organic polymers.  The apparent inactivation
of viruses by pH is very high, however, a few investigators have presented
results which show that the true inactivation may be significantly less than
the apparent inactivation.

     The principal objective of this project is to critically review the
literature on removal or inactivation of viruses in water by chemical co-
agulation and by high and low pH.

-------
                                 SECTION 2

                                 CONCLUSIONS

     This literature review can be summarized and conclusions developed as
shown below:

COAGULATION WITH METALLIC COAGULANTS

     (1)  Enteric virus removal from water and biologically treated effluents,
          based on observations with poliovirus and Coxsackie A2, will be
          from 90 to 99-999 percent.

     (2)  Other representative enteric viruses should be studied to confirm
          the coagulation removal prediction based on the poliovirus and
          Coxsackie A2 data.

     (3)  Either ferric or aluminum salts provide equal capability for virus
          removal when used in sufficient dosage.  An optimum coagulant dose
          for virus removal cannot be determined.  The dosage appears related
          to conditions of the water.

     (k)  Turbidity removal can be used as a process control for virus removal.
          Maximum virus removal occurs at or near the point of maximum tur-
          bidity removal.  Minor reductions in virus removal may occur when
          coagulant dosage is adjusted for maximum turbidity removal rather
          than maximum virus removal.

     (5)  Water conditions will vary the amount of virus removal which may
          be expected.  The few data available indicate that virus removal
          with metallic hydrated oxides is best at a pH of 5-7 and. at a
          pH of 5-6.5 with aluminum or ferric phosphates.  No conclusions can
          be drawn from the few data on virus removal variations caused by
          temperature variations, if in fact there is an effect.  Laboratory
          studies have shown a slight reduction in removals when organics
          are present in moderate amounts (treated effluents) but this is
          not supported in pilot plant results.  Inorganics, in amounts
          existing in fresh waters, do not influence virus removal when floe
          formation is adequate.

     (6)  The mechanism of virus removal appears to be one of a complex with
          the metallic cation at reactive virus sites, probably carboxyl
          groups among others, and incorporation into the floe.

     (?)  Laboratory and pilot plant data substantiate the reliability of
          this process for virus removal but confirming tests on a full
          plant have not been performed.

-------
COAGULATION WITH POLYELECTROLYTES

     (1)  Virus removal is unsatisfactory when polyelectrolytes are used as
          primary coagulants in the usual concentration range of 1-2 mg/1.

     (2)  Virus removals are not increased when polyelectrolytes are used
          as coagulant aids if the coagulation process is otherwise accep-
          table .

INACTIVATION WITH pH EXTREMES WITHOUT PRECIPITATION

     (1)  Enteric viruses are inactivated at pH values of 10 to 12.  The pH
          at which this occurs is evidently specific for each class of
          enteric viruses.

     (2)  Enteric viruses as a class are stable at pH values much lower
          than are obtained in water and wastewater treatment.

     (3)  The inactivation of viruses, particularly in the lower alkaline
          pH range where susceptibility has been observed, may be due to
          aggregation which causes a loss of infectivity, but is not true
          inactivation.

     (k)  Virus removal is affected by organics, inorganics, pH, contact
          time, temperature and type of virus.  The magnitude of each para-
          meter on the virus removal cannot be determined from the present
          evidence.

     (5)  The true inactivation of enteric viruses in this process, based on
          poliovirus 1 data, occurs by disruption of the protein capsid and
          loss of nucleic acid to the water.  The nucleic acid may be
          partially degraded before its release to the water.

REMOVALS WITH PRECIPITATION AT HIGH pH

     (1)  Laboratory investigations have shown removals of poliovirus 1
          up to 99-9 percent during water softening by precipitation and
          up to 99-1 percent during precipitation of calcium hydroxylapatite.
          Increased amounts of precipitates increased the virus removal.

     (2)  Pilot and full scale plant data show high removals of natural
          viruses during treatment of secondary effluents.  Removals up to
          99-9 percent or more have been obtained.

     (3)  The process appears reliable based upon the high removals obtained
          from several field and full scale studies.

     (k)  The questions posed above for removal of viruses by pH effects
          alone have not been answered in the high pH precipitation studies.

     (5)  Process control can be obtained by monitoring pH, turbidity,  and
          suspended solids removal.

-------
                                 SECTION 3

                          CONVENTIONAL COAGULATION

BACKGROUND

     A critical review of the literature on the removal of viruses by coagu-
lation should answer a number of questions on the process.  Among these
questions are those shown below:

     (1)  Reliability of process.  This concern can be examined by considering
          the following:

          (a)  Plant scale data versus laboratory or pilot plant data.
               Similar results in full scale plant operation should be
               obtained under reasonably similar conditions in lab-
               oratory or pilot plants.

          (b)  Reproducibility of data by different investigators.
               Results by various investigators removed from each other
               in time and place should be similar.

          (c)  Delineation of factors affecting process.  The coagula-
               tion process response to variables such as organic
               content, pH, temperature and salt content, among others,
               should be examined.

     (2)  Effectiveness of different coagulants.  Differences between co-
          agulants, if any should be determined.  These differences would
          involve the following:  coagulant dose, response to different
          viruses, effectiveness in waters with varying temperature,
          organic content, salt concentration, etc.

     (3)  Comparative removal of different viruses.  Only a few of the
          more than 100 enteric viruses have been investigated to determine
          their response in the coagulation process.  The ability of these
          tracer viruses to predict removals of all others should be as-
          sessed.

     (4)  Process control.  Routine plant operation uses turbidity measurement
          as a control parameter.  Virus removal may parallel turbidity
          removal and thus be monitored by measuring this parameter.

     (5)  Usefulness of coagulant aids.  Polyelectrolytes have been used as
          coagulant aids for increased removal of turbidity and/or to
          obtain better settling of the floe formed in the process.  They
          may also increase the effectiveness of virus removal.

-------
     (6)  Determination of the mechanism of virus removal by coagulation.
          Knowledge of this mechanism would permit an extrapolation of
          virus removal from the few viruses which have been examined to
          the many which have not.

METALLIC COAGULANTS

     This section of the review is restricted to the metallic coagulants,
generally aluminum and iron salts, primarily used in "clean" waters rep-
resentative of those entering a water treatment plant.  Several cases of
coagulation of an activated sludge effluent have been included because of
their notable interest and careful control of the operation.  The reported
virus removals are those following the sedimentation phase.

     The data from the literature have been divided to permit a general ex-
amination of coagulation conditions; enteric viruses versus bacterial
viruses; and aluminum sulfate versus feric salts.  In general the docu-
mented literature reports are those in which reasonable experimental control
has been exercised.  This has excluded several early reports such as those
of Chang et al. (l), Carlson et al. (2) and Kempf et al. (3).  These early
reports among other things failed to note the importance of adding the
viruses before the coagulation chemicals, failed to report data to evaluate
the effectiveness of the coagulation process or failed to note that phosphate
was a principal precipitate in studies where aluminum sulfate was used and
not the hydrated aluminum oxide.  Many other investigations have used
materials such as aluminum oxides and phosphates and polyelectrolytes in solid
forms as virus absorbers.  Use of such materials in solid form does not
duplicate the method of its use as a coagulant in water.

     The data in Table 1 are for the coagulation of enteric viruses by
metallic coagulants.  Data for the removal of bacteriophages are presented
in Table 2.  Enteric virus data are available only for poliovirus 1,
Coxsackie A2 and naturally occurring viruses (mainly Coxsackie B3 and B5).
The data in Tables 1 and 2 are generally for the best removal results which
were obtained by the various investigators.  Factors which affect the
reported removals will be presented in the following discussion.

Amount of Removal

     The removals obtained for poliovirus 1 and Coxsackie A2 ranged from
48 to 99-999 percent.  The 48 percent reported by Foliguet and Doncoeur (9)
occurred when the clay concentration was low.  These authors commented that
better removals of viruses were obtained when they had an increased op-
portunity to absorb to clay particles.  The virus removal then occurred with
the concommittant removal of the clay.  Lovtsevich et al. (18) in coagulating
poliovirus 1 with AJ^SOiJo also noted better virus removal with higher
initial turbidities.  Other reports in Table 1 indicated higher orders of
removal when the turbidity was at lower levels than that used by Foliguet
and Doncoeur.  The 70 percent removal of poliovirus 1 reported by Boardman
and Sproul (6) was from a study where coagulation for virus removal was not
optimized.  Wolf et al. (4) reported 63 percent removal of poliovirus 1 and
46 percent for f2 phage in an activated sludge effluent when the aluminum

-------
&
&
S.
&
        IA IA  C   I  O
            *  ON     >-H
        cy H
        CO   ON CO  O
        O\
        OJ
            OJ  K  OJ   OJ
         W      0)
               H  OJ  t





                ro  ro  ro


                   SO  O 5
                   CO  tO H

               ^OJ "7\l  OJCJ

               s!  3  ^ +
              11
                               CO   ON    O
                                  CO    CO     CO  r
                                                                     B   B  B  B
                                                                     IA  1A  LA  H   CO
                                                  \O  N-  O\  I"-  CO  ON  LA  VO  IA  O '
                                                   ONONCNONO"\ONO"\CJ\ONBI
                               CM  CM     CM
                               t  CO

                               C  IA
                               CO IA     CO     IA
                               H H     B     OJ


                                  II
                                                                        g  =   g  g
                                                                     IA  IA  IA  IA  W
                                                                        S  5
               ft  *  tj       u  "ft
                                               3Q  O  O  O  O  O  IA
                                               3-  \o  co  o  oj  -^  oj
                                                              ro t   co  to  r   ro co
                                                             H      ^  ^      H  ^-^
                                                             U      C\J   C\J      O   OJ
                                                              V      rj  rj      (b  (U
         .S  .5
         8  8
         I  
                                S?   S?
                    CO
      Si^^!^^^  c

                                                       JS  'S  S  1
                                                       5  s  I  a
                                                       W  -H  h  -P
                                                          +5  8  o
                                                       -P  to  ^  H
                                                          g>rl  ql  -H
                                                          CH  J  E
                                                       I   i    I   I

                                                       g  g  j  P.
                                                                                               i              i,
                                                                                                ft              ft


                                                                                               11 jl-a  I
                                                                                                O   O P  ~-^>   CJ
                                                                                                at   IA o    o  a>
                                                                                               
-------
















M
g
B
1
U
a


l
s
3
w
8

o
M
1
W
&

a
O
M
a


o
o
OJ
1

























i
w





=E

a
o
n
s
s

w

1
O








-p
1
u






M
E
S



w
A)
3
S


4-<
0)
W

f

r


j-
s,|

1


-p
B
CO


Pi
8 u


Sj
3
I



h
1


f?

W H ft
g^ft
oi-g


S,
>
?

||4
M S





!
H
H II
PL, en
^ 5
UN LA
H H =
CO H OJ -d" V) J~ -41 OO LS- r-( CVJ -3"

g g g B
UN ITv O
C-- to in  =V>^P   CP CQ CO H OJ ON
ON S 1 ON ON OJ H UN ON S 1 ON CO
( H f ( i UN
ON
CO Q\
 ON
ON 1
ONONC ON^hONON VDONCOON ON^OONOJ
ONONONCOCOONON jONONON ONCOONON

CO J* J- J- -d-
O OJ C-- C^ C CO CO COONOJ? OJHONUN
vDM>v)vOVlt>-vD vX)vJ3lA^ VOC-M3O-
O
I "^
OOJ H HrOOJ4JOJ OJ fO-d-
 tort 'W ...ft .w
vjDvua coat D UNO v^aco coco


LA UN UN UN J- -d"
OJLALAlTvLAOJlO ONOJOJOJ UNUNOJCM
lOJOJH iSOJOJl' OJOJil
j- m -* -* o oj
OJ OJ OJ OJ OJ OJ

1 1
0 0
H Hp
-^ UN OJ "if 5 O
gEl-3-ojrn H S JSOro



OJ . . V
;* f- "3 Sii ^ * h 5
P > aj 
TH r-j H    -HH
co-dK daJOHHWw -dd
K OJ =tajcan;; d) hr
d BJj^'' 'y4o

ft IS dalpuu ft!3oj
coo SS{5<;<; coSS

r> * I J *
\jp5 HUN M)QO rnoN
SiHiA_ OJM mwojvJD -3-LAOjj-
OQ ~"UNUN OO O OO
OJOO OJH OJUNOJ UNUN

ronm m mm onm
??? 9"?? f?
w-5555 co co corn wmco

ojojoj oj ai ojo ojooj
Hf-jfj rj H rJ4* HWOJ
*,ON
i-t J^OJOJCO OJ Ot*-

II 1
II sl 1. II
i 1 E " Is
gJI2 $is;sif;| 2 siK





c
,33333
^  CO OJ OJ
CO v) c- UN K =

OJ OJ
 CO CO  W
co a a UN s s


 
^ g g ^ g =

H^
u
N. w. > ra
i-t H 0 5 =
OJ A HO
H H H H



T) -d
H m v v

-P H H > H >
IS fiS fiS

t t
-d" IA VO
vo ro rn oj o o
m \ "^-^ i -3- -*
S S 8

ro ro rn m
o* c? 6^ 6*
-* ro CQ w co to

CO O OJ OJ CM OJ
d> n> i) H a> QJ
En fn LE,  s^5
^Q 1 OJ M 1 
iP OJ "UNO- H W


1
d
a S! d  I .
..

-------
to phosphorus ratio was 0.44 to 1.  Under such conditions no hydrated aluminum
oxide would be expected to be formed with the precipitate consisting
principally of aluminum phosphate (19)-  It may be inferred from Brunner and
Sproul (20) that sharply reduced virus removals would be expected when the
aluminum phosphate precipitation was done at A1:P mole ratios less than 1:1.
When Wolf et al. (4) increased the A1:P ratio to 7:1 the poliovirus 1
removal increased to more than 99-7 percent and the f2 phage removal to 99-8
percent.

     The removals reported in Tables 1 and 2 were from reports with higher
influent virus concentrations than would be expected from a natural water
source.  A question of some significance is whether the percent removal is
dependent upon the numbers of viruses in the influent stream.  Guy et al.
(10) reported removals in excess of 88 percent with naturally occurring
enteric viruses in a pilot plant evaluation.  Their study was done with
only 0.004 infectious virus particles per ml in the influent water with a
virus concentration technique detection limit of 0.00045 infectious
particles per ml.  Their study with 54.2 PFU/ml of naturally occurring
phages provided removals of 93 percent.  The Pomona virus study (7) also
reported a poliovirus 1 removal of about 95 percent at a virus concentration
of 1.3 x 10^ PFU/ml in an activated sludge effluent using alum with a 1.7
to 1 A1:P ratio.  Wolf et al. (4), on the other hand, obtained more than
99-7 percent removal of poliovirus 1 from an influent concentration of 113
PFU/ml.  Their coagulation conditions were evidently better than that in
the Pomona study since an A1:P ratio of 7:1 was used.

     The laboratory and pilot plant work reporting on poliovirus 1,
Coxsackie A2 and phage indicated that removals of these viruses at influent
levels of 10^ PFU/ml or more will be greater than 90 percent and may be as
high as 99-999 percent.  Sufficient studies have not been made with influent
virus concentrations less than 100 PFU/ml to draw conclusions, but the
available data indicate that the removals will be around 90 percent.

Type of Virus

     Tables 1 and 2 present data on only two enteric viruses, poliovirus 1
and Coxsackie A2 plus, in one study, naturally occurring enteric viruses
(principally Coxsackie B3 and B5) and four phages, MS2, f2, T4 and Micro-
coccus pyogenes phage plus the naturally phages in the Guy et al. (10)
study.  A comparison of the removal data between the two enteric viruses
poliovirus 1 and Coxsackie A2 does not show marked differences between these
viruses.  The Coxsackie virus data for the spiked distilled water samples
with low turbidity do not show the high removals obtained in certain of the
polio studies.  The aluminum sulfate coagulation of the Coxsackie virus with
40 to 100 mg/1 gave removals from 86 to 99 percent, somewhat below the more
than 99 percent removals with poliovirus obtained at lower coagulant doses.
The Coxsackie data in natural river water with higher turbidities gave higher
removals, 95 "to 99 percent at lower coagulant doses.  Data are available on
too few enteric viruses to draw conclusions on whether there are real dif-
ferences in their responses to coagulation.

-------
     Data in Table 2 show 'uniformly high removals by Al2(SOj+)o coagulation
for the f2, M. pyogenes phage and  the MS2 phage.  As discussed above,  the
46 percent removal of the f2 observed by Wolf et al. (4) was  caused by a
known inadequate coagulation condition and is atypical.  The  T^t- phage  also
showed a slightly lowered removal.  The removal data for these phages, when
coagulated using ferric salts,  show few differences.

     Too few enteric viruses have  been examined to determine  whether there
are differences in their removability by the coagulation process.  The
bacteriophage data on four different phages do not show significant dif-
ferences between themselves.  Nor  in fact, were there significant dif-
ferences when their removals are compared with those of the enteric viruses
in Table 1.  The available data suggest that coagulation is relatively
nonselective in virus removal.

Type of Coagulant and Dosage

     The data in Tables 1 and 2 were obtained with Al2(SO^)o, FeClo,
Fe2(SOj+)o and FeSO^ and comparisons can be made of their effectiveness.
These investigators were able to obtain enteric virus removals of 90 percent
or more with each coagulant under  the various conditions of their tests
unless the coagulant dosage was known to be less than optimum.  The phage
data in Table 2 also show similar  results except for the report on ferrous
sulfate and the f2 phage.  In discussing their ferrous sulfate work York
and Drewry (15) and York (lU) reported that virus removal with this co-
agulant was significant below 30 mg/1.  It is also noted that minor turbidity
removals occurred with this coagulant.  These authors concluded that al-
though an acceptable floe was formed and virus removal was in excess of
90 percent that since turbidity removal was poor the coagulant was not
acceptable.

     Coagulant dosages in practice are normally adjusted to minimize the
turbidity leaving the sedimentation tank.  It follows therefore that ac-
ceptable virus removals should  desirably occur at these same  concentrations.
Data from Tables 1 and 2 have been extracted and presented in Table 3  to
facilitate a comparison of the  expected virus removal versus  coagulant
dosage.  A comparison of these  data do not show clearly a positive relation-
ship between coagulant dosage and virus removal.  It does appear, however,
that in a given investigation increasing dosages to some optimum value
increases the virus removal.  This is presented rather clearly by the  data
from Chang et al. (11) when they increased removals of Coxsackie A2 from
86 to 99 percent with Al2(SOi4.)o dosages increasing from kO to 100 mg/1.
York (14) presented similar data but also showed that f2 phage removal
decreased as each of his four coagulant dosages increased beyond the optimum
value.  Other investigators do  not appear to have commented on this fact.
York's observations are not unexpected if one considers that  viruses are
expected to behave, at least in part, as charged colloidal particles.  Re-
movals of such particles are often noted to decrease as the coagulant  con-
centration increases and reversal of the colloidal charge occurs.  He  did
not notice significant decreases in turbidity removals beyond the optimum
concentration for its removal,  however.  The larger variation in coagulant
dosages for the virus removals  noted in Table 3 appears to be related  to the

-------
TABLE 3  VIRUS REMOVAL VERSUS METALLIC COAGULANT DOSAGE*

Coagulant
Type Dosage
mg/1
Al2(S01|)3 10
20
I+O
76
1+0
60
80
100
25
20-26
20-26
15
NS
25
80-100

25
15
FeCl3 66
60
20
1+0
15
50-60
50
1+0
20-1+0

Fe2( SO^K 1+0
50
62
1+0
Type
Virus

Polio 1
ii
it
it
Coxsackie A2
"
"
it
"
MS2
T!+
f2
"
It
M. pyogenes
phage
tt
ii
Polio 1
it
Coxsackie A2
"
"
MS2
f2
"
M. pyogenes
phage
Polio 1, 2, 3
f2
ti
T^
Water
System

Spiked DW
DW
Polluted River
Act. si. effl.
Spiked DW
"
"
"
Ohio River
NS
NS
Natural surface
Act. si. eff.
Natural surface
Spiked DW

Ohio River
Ohio River
Spiked DW
"
it
it
Ohio River
NS
Natural lake
Natural surface
Spiked DW

Polluted river
Natural lake
Natural surface
Polluted river
Removal Reference

%
90
70
99.8
95
86
96
97
99
95-99
99-9
98
99.^
99.8
99-9
99-99-8

81+-95
86
99.7-99.999
1+8-99-7
97
98
95
99-5
99.^
99.1
99.3-99.9

99.8
92
99.91
99-6


5
6
10
7
11
11
11
11
12
13
13
16
l+
ll+ & 15
11

12
12
8
9
11
11
12
17
ll+ & 15
16
11

10
ll+ & 15
16
10

     * Data extracted from Tables 1 & 2

  DW - Distilled water
  NS - Not stated
                            10

-------
condition of the water.  The data of Chang and co-workers (11) (12) show
that dosages of only 25 and 15 mg/1 for A^CSOiJo and FeClg respectively
were required to obtain optimum removals of the Coxsackie A2.  Dosages up
to 100 mg/1 for AJ^C$0^)3 and kO mg/1 for FeCl3 were required for the spiked
distilled water.  Similar results can also be seen for A^(804)3 coagulation
of the poliovirus in the various waters which were used.

     These data support the conclusion that either aluminum sulfate or
ferric salts will provide equal capability for virus removal.  An "optimum"
dosage for virus removal cannot be deduced from the data.  The required
dosage appears related to the characteristics of the suspending water.

Process control

     In practice, dosages of coagulants are normally adjusted to minimize
the turbidity leaving the sedimentation tank.  Virus removals are normally
thought to follow turbidity removal and process control for viruses is ob-
tained by monitoring turbidity.  Several investigators have presented
evidence to show that increased turbidity removals are accompanied by in-
creased virus removals (8) (13) (1*0 (15) (17)  Chaudhuri and Engelbrecht
(13) showed a marked parallelism between turbidity and T^ and MS2 phage
removals using Al2(80^)3 as the coagulant.  Manwaring et al. (17) observed
similar results with coagulation of the MS2 phage with FeCl/j.  Foliguet
and Michelet (8) have also mentioned that high virus reductions were made
in parallel with high turbidity reductions.

     A close examination of certain of these data, however, shows that
maximum turbidity removal may occur before the maximum virus removal is
reached.  Chaudhuri and Engelbrecht (13) showed that the maximum TU bacterio-
phage removal lagged the maximum turbidity removal at pH 5-24 by about 5
mg/1 of Al2(30^)3.  This was not noted, however, at pH values of 6.17, 7.00
and 8.30.  Shelton and Drewry (16) noted a decrease from 98 to 95 percent
in the FeC^dosage required for maximum virus removal compared to the dosage
where the maximum turbidity removal was first reached.

     York (ll|-) and York and Drewry (15) have also shown that their virus
reductions generally coincided with the turbidity reductions.  York (lU)
did note, however, that the optimum coagulant dose for maximum turbidity
removal was slightly lower than that reported for maximum virus removal.
Portions of these data have been reproduced in Table k.  The FeC^ data
showed a rather marked variation from the other coagulants with only 58
percent removal of turbidity when the maximum virus removal occurred.  These
data are confounded by the very low turbidity level of 0.38 TU in the raw
water.  Certain of his FeCl3 coagulation data showed turbidity increases
after coagulation and sedimentation.  Difficulties in obtaining satisfactory
settling of floe particles to yield reproducible data in these turbidity
ranges are well established.  This anomoly in his data may be safely dis-
counted.
                                     11

-------
TABLE k  DOSAGES FOR MAXIMUM REMOVALS OF f2 PHAGE AND TURBIDITY FROM
         NATURAL LAKE WATER*

Coagulant Initial
Turb.


T.U.
Ai2(so^)3 2.3 & 3.6
FeCl3 0.38
TPa ( Qf^l ^ Q Q
r GO ^ &\J}\) o <  O
FeSO^ 1.1 & 1.9

Dosage


mg/1
25
50
50
36
Virus
Max
Virus
Removal
(Ji
IQ
99-9
99 A
92
9^
Turbidity
Turb. Removal
at indi-
cated dosage
of
JQ
91-95
58
92
+
Dosage


mg/1
21
23
^9
-
Max
Turb.
Removal
%
96
92
89
+

          * From York (1*0

          + Minor removals or increases of turbidity
                                  12

-------
     The available data support the general conclusion that maximum virus
removal will occur at or near the maximim turbidity removal.  The evidence
indicates that only minor reduction will occur if the coagulant dosage is
adjusted to that for maxijtium turbidity removal.

Water Conditions

     The water conditions of general concern in virus coagulation appear to
be those of pH, temperature and organic and inorganic chemical content.

pH--
     Chang et al. (11) coagulated Coxsackie A2 in spiked distilled water with
80 ppm of ^2(304)3 at 25C and at final pHs of 5.5, 6.2 and 7.2.  They ob-
tained removals of 95 ? 97 and 99-0 percent respectively.  These authors made
a companion study with M. pyogenes phage at final pHs of 5-5j 6.2, 7-2 and
8.2.  The respective removals were 85, 99 -I? 97 and 80 percent.  They con-
cluded that the optimum pH range was 6.2 to 7-2.

     Chaudhuri and Engelbrecht (13) coagulating MS2 and T^ phages with
Al2(SOl4.)3 at dosages from 20 to 100 mg/1 at pHs of 5.2U, 6.17, 7-00 and
8.30 found decreased removals as the pH was increased.  They concluded that
the optimum pH for coagulation of these viruses was 5-2.  Using the MS2
phage with FeC^ at dosages from 20 to 100 mg/1 with pHs of 5.0, 6.1, 7.0
and 8.0 Manwaring et al. (17) found very similar results to those of
Chaudhuri and Engelbrecht.  These investigators concluded that the optimum
pH was 5.0 for coagulation of the MS2 with
     Chang et al. (11) have pointed out the importance of distinguishing
the type of floe which is formed in the coagulation process in a virus
study.  They showed that the optimum pH for the removal of viruses in a
system where aluminum phosphate is the precipitate is lower than where
aluminum hydroxide is formed.  They found that the optimum pH was 5-2 under
conditions where AlPO^. was formed when the pH varied over the range from
5-2 to 7 .2.  Their report is unclear on which virus was used in this study.
On the other hand Brunner and Sproul (20) found an optimum pH of 6.h in
the coagulation of poliovirus 1 with Al2vSOi|)o in a distilled water system
when the A1:P mole ratio was 1:1.  They investigated removals at pH values
of 5'lj 6.U and 7-3'  Chang et al. (1) reported on the removal of M. pyogenes
phage with 60 ppm of Al2(SOi|)o in a phosphate buffered distilled water
system.  They reported that the removal decreased from 98 to l4 percent as
the pH was increased from 5-2 to 8.2.  The floe formation was also reported
to decrease from fairly good to very poor over this pH range indicating that
the precipitate formed was actually
     The available data are insufficient to draw conclusions on the optimum
pH for enteric virus removal.  It does appear, however, that the optimum
pH for virus removal where aluminum phosphate is formed may be lower than
where aluminum hydroxide is precipitated.  The importance and usefulness of
precise information on the optimum pH for virus coagulation is debateable
for most situations.  It would appear that in nearly every case coagulation
will be optimized for removal of turbidity.  Adjustment of pH to a value
which might provide better virus removal will not receive high priority.

                                     13

-------
In cases where the raw water is known to have high virus concentrations
control of pH to optimize their removal would be desirable and a laboratory
investigation using the local water would be warranted.

Temperature 

     Data on temperature effects on virus removal during coagulation are
very few.  The only data available appear to be those of Chang et al. (12)
shown in Tables 1 and 2 for the coagulation of Coxsackie A2 and M. pyogenes
phage with A^( ^0^)3.  These investigators noted a decrease in the removal
of the Coxsackie A2 from 99 to 96 percent as the temperature decreased
from 25 to 5C.  The phage removal decreased from 95 "to 84 percent for the
same temperature change.  These authors did not consider the reduced removal
of the Coxsackie virus to be significant.  Reasons for the reduced removal
of the bacterial virus were not given.

     The importance of water temperature on the removal of viruses by co-
agulation cannot be determined with the available evidence.

Organic Content- -

     Chang et al. (12) reported that 20 ppm of gum arable reduced the removal
of Coxsackie A2 from 97 to 17 percent when 15 ppm of PJ-2( S0^)o was used.
The removal of M_. pyogenes phage was reduced from 92 to 0 percent.  The
gum seriously interfered with coagulation since no floe was produced.
Manwaring et al. (17) reported a reduction in the removal of MS2 phage from
99 "to 67 percent when 200 ml/1 of wastewater effluent was added to a clean
suspending medium when 60 mg/1 of FeCl3 was used.  This was accompanied by
a parallel decrease in the turbidity removal.  No decrease was observed
when up to 50 mg/1 of bovine serum albumin was added.  Chaudhuri and
Engelbrecht (13) reported reduction from 99-8 to 9^- percent in MS2 removal
with 200 ml/L of wastewater effluent using 50 mg/1 of AJ^vSO^)^  Parallel
reductions in turbidity removal were also noted.  Reductions or only 2 to 3
percent were noted in the coagulation of T4 with 50 mg/1 of AlgC 80^)3 in
the presence of 50 mg/1 of egg and bovine serum albumin, and 200 ml/1 of
wastewater effluent.  Data in Table 1 from Wolf et al. (4) reported a 99-^
percent removal of poliovirus 1 in activated sludge effluent with a co-
agulant dosage to yield a 7:1 ratio of A1:P.  The Pomona virus study (7)
reported an average removal of 95 percent for poliovirus 1 when their dosage
of Al2(SOl|)o produced an A1:P ratio of 1.7:1.  Foliguet et al. (21) also
found that organic matter reduced poliovirus 1 removal during coagulation
with
     The evidence supports the conclusion that virus removals may be reduced
slightly from their highest values when organics are present, but coagulation
is otherwise satisfactory.  The data also show that chemical dosage modi-
fications may be necessary in a plant to respond to raw water organic content
changes.

-------
Inorganic Material --

     Chang et al. (12) speculated that the presence of calcium (2^-5^ mg/1)
and magnesium (6-l4 mg/l) may have slowed down the formation of the aluminum-
M. pyogenes phage complex and thus reduced the removal of this virus over
that obtained in a water with very low concentrations of these cations.
These data are shown in Table 2 with the Ohio River water sample containing
the higher concentrations of calcium and magnesium and the spiked distilled
water with the low concentrations.  A comparison of their data in Table 1
for the Coxsackie A2 virus for these waters does not show this effect.
Manwaring et al. (17) found minor reduction in the removal of MS2 phage with
FeClo when calcium and magnesium present together were varied from 0-50
mg/l each.  Chaudhuri and Engelbrecht (13) reported no reduction in the
removal of T^ phage with A12( $0^)3 when the calcium and magnesium concentra-
tions were changed in similar fashion.

     Thorup et al. (5) obtained the data shown below in the  coagulation of
poliovirus 1 with kO mg/l of clay using 10 mg/l of A^SOij. or Fe2( 80^)3 at
pH 6.8 in distilled water with the indicated concentrations of CaCl2:
                        Virus and Turbidity Removal

Percent
Ca(from CaCl2)
mg/l
0
1+
ko
Uoo
Uooo
Al;
Virus
11
26
83
86
79
3(3014.) 3
Turbidity
0
18
93
96
9k
Fe2(
Virus
0
0
1
83
85
; so^)3
Turbidity
0
0
0
93
93

Their data do not show significant variations in the removal of the polio-
virus when acceptable turbidity removals were reached with calcium concen-
trations of kO, 400 and U,000 mg/l.

     The available data do not indicate a significant effect on the removals
of viruses by coagulation when the inorganic concentration of the suspending
water is varied.

Process Reliability

     The reliability of the coagulation process using metallic salts cannot
be judged by comparing removals obtained under laboratory and pilot scale
conditions.  Data from actual plant operations are required.  However, there
are no reports in the open literature on full scale conditions.  Tables 1
and 2 cite six reports obtained from four different pilot plants (k) (7) (8)
(9) (10).  These plants were operated for removal of poliovirus 1 (except
for Guy et al. [10] who 'used polio 1, 2 and 3) under a variety of conditions

                                     15

-------
using activated sludge effluent and spiked demineralized water.  The removals
obtained were consistently very high except for two reported instances.  In
these two cases Wolf et al. (U) obtained only 63 percent removal of polio-
virus under known poor coagulation conditions.  Foliguet and Doncoeur (9)
attributed the U8 percent poliovirus removal in one of their runs to low
levels of initial turbidity in the influent water to the pilot plant.  With
the exception of these two cases the removals in the four pilot plants ranged
from 95 to 99-999 percent and are at least as good, and in most cases, better
than the removals obtained from laboratory reports.  It is also to be noted
that these pilot plant reports were obtained by various investigators
throughout the world.  This gives increased evidence to the reliability of
the process.

     The coagulation process appears to be one which when satisfactorily
operated will produce virus removals of up to 99-99 percent or more.
However, this conclusion has not been confirmed by data from a full scale
water treatment plant.

POLYELECTROLYTES

     The reported data on the utilization of polyelectrolytes as primary
coagulants have been summarized in Table 5-  Thorup et al. (5) showed that
the anionic and nonionic polyelectrolytes gave poor removals.  They found
no removals in distilled water but removals up to Uo percent with salt con-
centrations up to 126 mg/1 of Na, K, Ca and Mg.  None to poor or fair floe
formation was observed.  Chaudhuri and Engelbrecht (13) did not observe
removal or inactivation of MS2 or TU phages with anionics in concentrations
of 1 and 1 to 5 flig/1 respectively in deionized water.

     Chaudhuri and Engelbrecht (13) observed a T^ and MS2 phage inactivation
of about 80 percent with the cationics Primafloc C-7 and Catfloc in deionized
water.  They used the polyelectrolytes in concentrations of 1 mg/1 except
the Catfloc and TU when 0.5 to 10 mg/1 was used.  They speculated that the
"inactivation" may have been caused by a virus-host absorption interference
such that one host bacterium cell was infected by more than one phage.  Under
such conditions the inactivation would not be real.

     The data in Table 5 for cationic polyelectrolytes indicate limited
removals of enteric viruses in concentrations of polyelectrolytes from 1 to
2 mg/1.  Thorup et al. (5) and diver (22) reported minor removal of polio-
virus 1 and Coxsackie B3 with Catfloc.  Poor floe formation was noted under
these conditions.  High concentrations of cationics were reported to give
removals of phages as high as 99.9 percent.  Concentrations of the cationic
materials to attain these removals for the MS2 and T^- were as high as 12.5
mg/1.

     Polyelectrolytes have been used as coagulant aids with mixed success (5)
(13) (114.) (15) (21).  Thorup et al. (5) found little improvement in T2 phage
removal when 1 mg/1 of an anionic and nonionic polyelectrolyte were used in
a water which had been deliberately undercoagulated with 5 mg/1 of Al2(S04)o.
Under similar conditions they found that 1 mg/1 of a cationic raised the T2
virus removal from 57 percent to 9^- percent.  They noticed little improvement

                                     16

-------
                      s     2    s
                                          8    S    8
              LA  OJ   OJ
                                     ON
                                                         ON   ON   ON
                                         ^o   cy   LA

                                         00   CO   C^
                                                     ONO ON O
                        oj IA cy IA

                        LA IA LA IA
              CO   t-   t-


              ^O   -*   -*
                                 o   o   w   cy
                                                     o\o
                                         co   co   s
                                                         IA\JD lA IA IAIA
2
s
s
g
                      %     s
                                                     ITv   ITv   LTV   IA
                                 8   8
                                                 H 4)

                                                 28
                                     H   -P   -P

                                     ill
       SV
       Q f >
                                             H   cy   C-   H   tv
               O O O   O
Si
5(
                      -H O   O
                      3   -a-

                  1111   II
SI Sfi 3S  SS  54!  5
III!
                            9    3
                                                 a   cy   cy
              i   i
                 
                                   17

-------
in removal of T2 or poliovirus 1 with these materials when used in situations
where the coagulation with A^SOi^o and Fe2(SOi4.)3 was otherwise acceptable.
Foliguet et al. (21) found that anionics and nonionics used as aids reduced
the removal of poliovirus 1 coagulated with 60 mg/1 of FeClo from 99-7
percent to 92 percent with 0.25 mg/1 of an anio.nic and to 89 and 95 percent
with 0.25 and 1.0 mg/1 of a nonionic.  A cationic material did not change
the removal obtained with FeClo alone.  Chaudhuri and Engelbrecht (13) did
note an increase in T^- phage removal from 98 percent to 99-9 percent when
about 1 mg/1 of cationics were added.  Little effect was noticed in a similar
study with MS2 phage.  York (l4) and York and Drewry (15) observed little
change in removal of the f2 phage when nonionics, anionics and cationics
were used as coagulant aids.  They did observe that utilization of a cationic
might reduce the optimum dosage of Al2(SO^)o for virus removal.  Others
have made similar observations (l6) (21).  York and Drewry also noted that
the effective coagulant dosage range for f2 removal was broadened with their
anionic polyelectrolyte.

     Utilization of polyelectrolytes in the usual dosage range of 1-2 mg/1
does not appear to give significant virus removal when used as primary
coagulants.  Virus removals are not significantly increased when these
materials are used in waters which are otherwise coagulated satisfactorily.

MECHANISM OF REMOVAL BY COAGULATION

     Chang et al. (l) postulated that in the first stage of the flocculation
process with alum the aluminum ions form an aluminum-virus precipitate.
Under the "right" conditions the aluminum-virus precipitates are then in-
corporated or aggregated into the floe particles.  They deduced the initial
reaction from knowledge of aluminum-protein chemistry and surmized that the
aluminum-virus complex was actually a salt of aluminum with a protein in the
virus capsid.  As mentioned above Chang et al. (12) found that gum arable,
a compound known to interfere with coagulation by coating the charged
particle, interfered with coagulation and concluded that the formation of
the virus-aluminum complex had been interrupted.  The exact meaning of their
data is in doubt because the gum arabic prevented the formation of any floe
particles thereby preventing any possible incorporation into floe particles
since they were not formed.

     Chaudhuri and Engelbrecht (23) examined the initial aluminum-virus
complex formation in the TU bacteriophage.  They determined the amount of
aluminum which reacted with this virus at pH 55 6 and 9 by exposing the
virus to a known low concentration of aluminum, and measuring the residual
aluminum after removal of the virus.  They determined that 7,370, 6,200 and
6,300 atoms of aluminum were absorbed by the T^ phage at pH 5-0, 6.0 and
9-0 respectively.  They stated without evidence that a satisfactory mass
balance on the aluminum distribution was obtained.  At pH 50 the smaller
MS2 absorbed U,600 atoms per particle.  This initial binding reaction was
found to occur in less than 30-UO seconds, the minimum resolution time in
their system.  They also estimated that the head of the T^ had 7,120
carboxyl groups, 1,980 hydroxyl groups and 6,930 ammonium and guanidinium
groups.  The match of the carboxyl groups with the number of absorbed
aluminum atoms and the known involvement of these sites in binding aluminum

                                     18

-------
in casein and gelatin led them to conclude that this group was the most
probable binding site on the virus.

     The observation by Chang et al. (12) that gum arabic interfered with
coagulation might be explained with the observations of Chaudhuri and
Engelbrecht (23).  Gum arabic has a high density of carboxyl groups which
would act as competitive binding sites for the aluminum.

     Cookson (2U) has discussed the absorption of viruses to metal hydrolysis
species.  He pointed out that adsorption may occur through coordinated hydro-
xyl groups, covalent bonds between the trivalent atom and the virus, and
from electrostatic forces.  He points out that these mechanisms are dependent
upon surface protein chemistry of the virus and the metal hydroxides.
Aluminum or iron phosphate with many fewer hydroxyl groups on its surface
would not be as active in virus removal as the metal hydroxide.  Since dif-
ferent viruses generally have similar protein surface chemistry the con-
clusion that coagulation and removal of viruses is not specific for each
virus is supported.

     The mechanism by which polyelectrolytes coagulate and remove viruses
appears less well understood.  That the electrostatic forces are important
is obvious from the very poor removals obtained from the anionic and non-
ionic polyelectrolytes (5) (13)'  These materials are negatively charged
(anionic) or uncharged (nonionic) and offer little attraction for the virus
which is negatively charged under most water pH conditions.  Addition of
cations in the suspending medium with these materials gave some removal.
Whether this mechanism is one of the charge reduction permitting electro-
static attraction to occur or bridging of the cation which had been absorbed
to either the virus or polyelectrolyte is unknown.

     Adsorption of viruses to cationic polyelectrolytes occurs much more
readily.  Cookson, in discussing adsorption to the cationic polyelectrolyte
FE 60, stated that adsorption to this material would occur at carboxyl or
carboxylate salt groups through hydrogen bonding and by electrostatic at-
traction forces.
                                     19

-------
                                 SECTION k

        VIRUS REMOVAL AND INACTIVATION BY pH AND LIME FLOCCULATION

BACKGROUND

     Viruses are affected by changes in the pH of the suspending water and
by any precipitates that may be formed during these pH changes.  It is
important to note that these effects may proceed simultaneously as the pH
is changed.  Viruses are inactivated or removed in some fashion as the pH
is changed to that outside their stability range.  This effect is time
dependent.  When precipitates are also formed the virus may absorb to the
surfaces and be removed with the precipitate when it is removed from the
water by sedimentation.  These two effects will be considered separately
in this review.

     A critical review of the literature on removal and inactivation of
viruses by pH and lime flocculation should address the following questions:

     (1)  Reliability of process.  Reliability can be examined by con-
          sidering the following:
          (a) Plant scale data versus laboratory or pilot plant data.
          (b) Reproducibility of data by different investigators.

     (2)  Delineation of factors affecting the process.  It would appear
          that at least the following factors would be significant: (l)
          organic and inorganic content; (2) agglomeration of viruses as
          a loss of infectivity mechanism versus true inactivation; (3)
          pH level; (k) type of precipitate formed; (5) contact time;
          and (6) susceptibility of different viruses to the process.

     (3)  Process control.  The utilization of pH, level of solids pro-
          duction and/or other mechanisms should be considered.

     (k}  Determination of the mechanism of virus removal.

pH EFFECTS

Low pH

     Viruses tend to be more stable at pH values around neutrality.  The
enteric viruses which are of interest in water transmission also tend to
be very acid stable.  Salo and Cliver (25) showed that 1 log per week of
Coxsackie A9 and poliovirus 1 was lost at pH 3-0 at 2C.  This inactivation
increased to 1 log per 2k hours at pH 3-5 at 30C.  They also found that
echovirus 7 underwent a 2 log loss in kQ hours over the pH range from 3
to 5 at 30C.  Robinson (26) found that several Coxsackie A and B strains
survived for one day over the pH range from 2.3 to 9^ at room temperature.
It was reported by Ginsberg (27) that adenoviruses 1 and 2 lost 1 log of


                                     20

-------
titer in 30 minutes at 22-23C at pH 3.  He also found that adenovirus 3
lost "between 2 and 3 logs under similar conditions.  Using adenovirus 5
Fields and Metcalf (28) found a 30 percent loss of titer at pH 3.5 in 2
minutes.  No further loss was noted in 180 minutes.

     The lowest pH reached in water treatment is probably not lower than
k to 5.  The resistance of enteric viruses to inactivation at this pH
range precludes consideration of low pH as a potential inactivation process.

High pH

     Data on virus inactivation are presented in Table 6.  Sproul et al. (29)
Berg et al. (31), and Boardman (30) showed that little inactivation of
poliovirus 1 was obtained at pH values up to about 10.5 at 20-25C for ex-
posure times up to 100 minutes.  Sproul (29) and Berg et al. (31) found a
poliovirus 1 loss of infectivity of 99 percent or more over the pH range from
11 to 12 with contact times of 90 "to 100 minutes.  Seemingly lower polio-
virus 1 loss was noted by Sproul (32) using NaOH for pH adjustment in distil-
led water although he did show a 9^ percent loss at pH 11.9.  The PH at
which significant loss of infectivity occurs for viruses other than polio-
virus 1 cannot be readily determined.  Ginsberg (27) found that adenovirus
1, 2, 3 and. 4 were stable up to a pH of about 9.  Adenovirus 5 was totally
inactivated at a pH of 10.5 in 10 minutes.  Robinson (26) found that several
strains of Coxsackie A and B were stable for one day at pH 9^'

     The data in Table 6 for poliovirus 1 indicate that the loss of infectiv-
ity increases as the pH increases and as the contact time increases.  The max-
imum inactivation observed was 99-8 percent in 5 minutes at 22-23C.  Echo-
virus 7 appears to behave similarly to poliovirus 1 although the data are very
limited.  The data on the adenovirus appear to indicate that these viruses may
be more sensitive to alkaline pH values than either poliovirus 1 or echovirus 7-

     Salo and Oliver (25) noted that the type of buffer system used and the
salt content (in the molar concentration range) influenced the rate of in-
activation by pH over the range from 3 to 9.  Sproul et al. (29) noted that
in their distilled water systems the cation associated with the hydroxide af-
fected the loss of infectivity of the poliovirus 1.  Loss of infectivity
started at pH 10.5 when NaOH or Ca(OH)2 was used but with KOH a pH of 11.5 was
required.  Berg and Berman (33) found that the time for 99 percent inactiva-
tion of echovirus 7 was increased from 1.3 minutes at pH 11.92 in BOD dilution
water to 25.6 minutes at pH 11.87 in a 5 percent peptone solution.  Inactiva-
tion comparisons at other pH values were also given.

     Donovan (36) presented an interesting study on the loss of infectivity of
f2 and T2 phages in settled activated sludge effluent under high pH conditions.
His data shown in Table 7 indicate that the f2 virus was very sensitive to the
pH of the system with virus reductions of 99'99 percent noted at a pH of 11.0
and above.  These and other data which he reported tend to show that the virus
reduction was temperature dependent when the pH was held constant since lower
reductions were observed at 5C than 22-27C at pH 11.5.
                                     21

-------
&i  c\j ro  1    mm
                                 ojojojcy
                              CO  ON ON ON ON

                                                            g\       +
                                                            ON  IA  ON  O\
QOCMO    coo\t--ON-4-cr\(gNONgNgNCQ-HgN    ON  ON  ON  ON  oj
ON  ON     (    U"\  ON     lAONONONCf\Cr\ONvCO    ON    ON  ON  ON  ON  .4"

                                                   3
 R&38
 g  g  g  3"  s  =  g  g  g  g  a   a   a  fi  s g  g
          3-d                    o   o   o  o
           C\J    PJ  CVl
 S  x  I S   88
 a  S  d;  oj      )
 O  W  S5  O    O  O
 ONOJHH
                                             a a a
                                                                              333
                                                                              W  -H  h
3  -r

SS-i

381
 K!  Sl  81
in-
'o
1
W  CO  EQ  W  W  CO

a  B5  B;  55  S5  S5
                     c\j  H c\j en
 oooo    oooooo

 3333    333333
 &  S. S, &    
                                  22

-------
 Table 7  LOSS OF INFECTIVITY OF
          f2 IN FLOG-FREE Ca(OH)2
          TREATED ACTIVATED SLUDGE
          EFFLUENT IN SIXTY MINUTES*
               Temperature
              Loss of Infectivity
  PH                 %
            10C       22-27C
7-0
8.0
9.0
10.0
11.0
11.2
12.0
W
30
61
56
99.99
-
>99-9999
58
56
78
98
~>99-999l
99-9999
>99.9999
*From Donovan (36)
                 23

-------
     When Donovan added EDTA to chelate the divalent cations in the pH ad-
justed effluent the f2 inactivation was four logs less than when it was not
present.  A similar test with Ca(OH)2 in distilled water at pH 11.5 had
less than one log of reduction in 5 minutes in the sample to which the EDTA
had been added versus over 5 logs in less than 1 minute in the sample
without EDTA.  Studies in which NaOH, KOH, Ba(OH)2 and Theorell-Stenhagen
buffer were used to raise the pH to 11.5 showed that the valence of the
cation, except for calcium, had no apparent effect on the inactivation.  He
concluded that the f2 virus clumped in the presence of the calcium and that
a significant part of the "inactivation" was actually a loss of infectivity
since a large portion of the viruses were no longer present as discrete
particles.  He also presented electron micrographs showing the aggregates of
viruses in the calcium treated sample.

     The f2 virus concentrations used by Donovan were evidently on the order
of 10 to IQlO PFU/ml.  The opportunity for collisions to permit aggregation
of particles of similar diameters is proportional to the square of the
number of particles.  In Donovan's work the opportunity for collisions and
possible aggregation of viruses was much larger than there would be in a
field situation.  In a biologically treated effluent or a contaminated
surface water the opportunity for aggregation would be greatly reduced since
virus concentrations in these cases would, at most, be on the order of 10^
to 10  PFU/ml.  The aggregation effect noted by Donovan is probably not
important in the real world, or at least is of minor importance.

     If the animal viruses are accurately modelled by the f2 bacteriophage,
however, then a portion of the "inactivation" under high pH conditions may
actually be only an aggregation of discrete particles.  The actual inac-
tivation or removal will be less than is otherwise thought.  Additionally,
the inactivation of viruses in a clumped state is more difficult than in-
activation of discrete virus particles.

     The few available data on virus inactivation by high pH are presently
inadequate to determine the exact conditions under which this process will
give high inactivations.  It is readily obvious that viruses do undergo a
loss of infectivity when a certain pH, which appears to be virus specific,
is reached.  The amount of loss to be attributed to aggregation, or true
inactivation has not been determined.  The salt content, particularly
calcium, may be very significant.  The influence of temperature, contact
time, organic content, though significant, cannot be deduced from the data.
The necessary pH for inactivation of a wide spectrum of viruses under any
set of conditions is unknown.

LIME PRECIPITATION

     The addition of Ca(OH)2 "to water will form various precipitates
depending on the water characteristics.  The precipitates of major importance
are CaCOj, Mg(OH)2, and calcium hydroxylapatite, Ca(OH)2(POj^)6, or some
other precipitate of phosphate.  The CaCOo and Mg(OH)2 are formed in the
softening of drinking water by the precipitation process.  Calcium hydro-
xylapatite is formed during phosphate removal with Ca(OH)2  CaCO-a and


                                     2k

-------
Mg(OH)2 would also be formed in this latter reaction depending upon the
concentrations of Ca and Mg.

Laboratory Evaluation

     Virus removal and inactivation data under high pH conditions with the
formation of a precipitate are presented in Table 8.  In a laboratory water
softening study using poliovirus 1 Wentworth et al. (37) found a maximum
removal of 8l percent with 726 mg/1 of CaCC^, and 99.92 percent when U86
mg/1 of Mg(OH)2 as CaCCb were precipitated.  Increased removals of the virus
were noted as the pH increased although the virus removal with CaCC>3 was
clearly more dependent upon the amount of precipitate formed.  They also
found that poliovirus removals were in excess of 99-9 percent at all tested
levels when both CaC03 and Mg(OH)2 were formed together.  They attributed
all of the virus removal to the incorporation of the virus into the pre-
cipitate and its removal by settling with the precipitate since their work
was done at pH levels of 11.2 or below, a point where their poliovirus was
not affected by pH alone.  Their work with CaCOo was confirmed by that of
Boardman (30).  In a laboratory study Brunner and Sproul (20) were able to
remove up to 99.1 percent of poliovirus 1 during the formation of calcium
hydroxylapatite at a pH of 11.0 using a defined chemical medium.  Using an
activated sludge effluent they obtained a removal of 9^- percent with a pH of
10.9.  Berg et al. (31) using Ca(OH)2 flocculation of a secondary effluent
with a dosage of 200 to 500 mg/1 of Ca(OH)2 which gave pH values up to 11.0
obtained slightly higher poliovirus 1 removals, up to 99-9 percent.  Removals
of 92 percent were obtained at a pH of 9.3 using 200 mg/1 of Ca(OH)2

     These laboratory data indicate that poliovirus 1 removals in excess of
90 to 99 percent are possible with increased removals associated with in-
creased amounts of precipitate.  It would appear from the pH removal data
without precipitate formation that a portion of the removals obtained in
these studies at pH values of 11 or more was due to the pH effect and not
by incorporation into the precipitate.  Larger removals were associated
with the precipitation of Mg(OH)2 and calcium hydroxylapatite rather than
the CaCOo.  Thayer and Sproul (3?) felt that the negative charge of the
CaC03 precluded high removals with this material.  The presence of the
organics in the treated effluents used in several studies did not appear to
interfere with the poliovirus removal when compared to that obtained in a
clean water system.  Sattar et al. (39) using Ca(OH)2 at pH 11.5 obtained
high poliovirus 1 removals in the presence of the organics in raw sewage.
In the Berg and Berman (33) study with echovirus 1, however, the very high
concentration of organics (5 percent peptone) did interfere with the virus
removal when no precipitates were formed.  These data would indicate that
organics may interfere in this process when present in sufficient concen-
tration.

     Insufficient data have been reported to determine whether inorganic
concentrations affected the virus removal.  Additionally data are not
available from these studies to judge whether a portion of the observed
removals can be ascribed to aggregation associated with calcium (or other
ions) as discussed by Donovan (36).
                                     25

-------
B.
       it:
                      JU~S  8
                      r-i d w   m
                        H  w
                      n d v R>  vi
                                             g
                                           3

                                           1*
             -3    J   J
             ON   ON   OX
             g    a   s
              c-   t    t
              000
                                                       CJ irv
P O -P O  ON
                  T   t



                  ON   ON       TO
         CO   C^   S    H


         ON   ON   ON ON ON  ON
                                                                J  .!
                                                               g a * 
                                                               i   i
                                                               H "o H "O
                                  P   -P    O
                                                       H
                                                             PJ C\J   C\J   CU

                                                            s  I  s   I
                                                            I 5   5   I
                                   ON   ro    

                                   H   H*   ^
         ^^   ^SO^'O^^
         ON^      d^s^^o;
                                            moo
                                       H    M   r-4 a    H  H
                               o   o    o   o
                                            rt   3 I
                                            5   -g I

                                               Si
                       ii
                                                                         ,   a

                                                                   1   sal
i
                                                                                  IT* IfN

                                                                                  2 d
                                                                                  vo irv
                                                                                   . i
                                                                                  d d
                                        26

-------
3  .
          pj!:  15'
            SO
           Tf
                                                                     I -H O
                                                                      P O
                                                                     } -H IT\
                                                                     J H  A
                                                                     ) -H  t
          m < >   f4 < P>  t-H h   l-ULbp  QU     00    D   -H  O  \  m
         H K b  -              V
                          .45             . .p
         31    s      3
$      5    8^
A      p  a    A
                 ir\      u\
         CO       W       CO
                                 o
                                 ITv
P  O  -P O


31  i
T

                                              OJ     O\
                                              7
                                         OJ    OJ    OJ
                                        S    5
                                        ON    CO     ro
                                             J-    ^      v)
                                              I     i       I
                                                                   *
                                                                                     IS  ^  5  i
                                                                                 3    -      $

                                                                                 cT I  S  3  S

-------
     In general, contact times were not varied in these studies so no con-
clusions can be made on this parameter.  It would be expected, however, that
increased contact time at higher pH values would increase the observed
inactivation from the pH effect based on Berg and Herman's data.

Pilot Plant and Full Scale Evaluation

     Data from several pilot plant and full scale evaluations of virus
removal under high pH precipitation conditions have been presented in Table
8.  Wolf et al. (4), using Ca(OH)2 flocculation of an. activated sludge
effluent, were not able to detect any of the k-Q x 1CP PFU/ml of influent
poliovirus 1 in their settled effluent.  It should be noted that FeClq was
used in their two tests.  They also were not able to detect any of the 0.2
to 7.2 x 10^ PFU/ml of f2 in the effluent under the same test conditions.
Their pH values of 11.0 - 11.3 and 11.7 - 11-9 were well within the range
when removals from the pH effect alone would be expected.  Hiser (40),
flocculating raw screened sewage with Ca(OH)23 found increasing removals
of filtrable viruses with increasing pH, reaching 99-6 percent at pH 11.5.
His study showed that naturally occurring enteric veruses underwent the
same percentage removal as did a spiked poliovirus 1.  In this investigation
the suspended solids were removed from the influent and effluent samples
before concentration of the viruses for plaguing.  Since the virus removal
was a logarithmic function of the pH Hiser felt that the pH was the primary
factor in the removal of the filtrable virus fraction.

     Full scale test data from the Water Factory 21 investigations showed
that 98 percent of the naturally occurring enteric viruses were removed
based on the Buffalo Green Monkey (BGM) cell line and 99-9 percent on the
Primary African Green Monkey cell line (38).  These data were based on the
influent and effluent virus titers occurring 50 percent of the time.  The
titer occurring 50 percent of the time was obtained from a probability dis-
tribution plot of the number of viruses in each of their samples taken over
the entire sampling period.  This report is of particular interest since it
is from a 15 mgd plant over a 17 month sampling period.  One hundred in-
fluent samples (some evidently were parallel samples) and 32 effluent samples
were examined for virus content.  The authors pointed out that only 5 percent
of the 0.027 PFU/ml isolated in the influent on the BGM cell line could be
confirmed upon passage in this cell line.  In their study effluent samples
were analyzed for viruses by two independent laboratories.  One laboratory
found only 3 positive samples while the other found 28 positives.  Un-
fortunately none of these positives were confirmed.  All of their percentage
removal data are based on unconfirmed plaques, i.e. plaques which were not
regrown in the same or another cell line to assure that the plaques had been
caused by viruses and not by some extraneous effect.  Their data also show
that removals based on the virus titers equalling or exceeding that occurring
90 percent of the time were 99-3 and 96 percent based on the PAG and BGM
cell lines respectively.  The confirmed plaquing study done on the influent
indicated k A and 5 B strains of Coxsackie, 10 strains of Echo, 3 strains
of polio, 3 strains of Reo and 5 unknown plaques in ^5 positive samples.
Only 3 positive effluent samples were confirmed indicating 2 echo strains
and polio 1.  Unfortunately, while a rather broad spectrum of viruses were
present in the influent insufficient effluent data were reported to draw

                                     28

-------
conclusions on whether all of the strains were equally susceptible to removal
by this process.
     Grabow et al. ^4l) in a full scale test (1.2 mgd) using an activated
sludge effluent did not find any enteric viruses in the effluent when the
pH was 11.1  0.2.  Removals in excess of 9998 percent were obtained when
the pH was as low as 9-6  0.1.  Parallel coliphage removal was 99-95
percent at pH 11.2  0.2 decreasing to 57 percent at pH 9-6  0.1.  The
99.98 percent removal of enteric viruses at pH 9-6  0.1 is surprisingly
high and unexpected based upon the laboratory data.  No information was
presented on the amount of precipitate formed during the treatment.

     The removals obtained in these reports indicate good reliability of
the process under field operating conditions.

PROCESS CONTROL

     The control of this process to maximize virus removal most clearly
relates to the pH.  The evidence shows that increasing pH causes increased
inactivation .  The point has been made above, however, that the pH necessary
for inactivation of a wide spectrum of viruses under any set of conditions
is unknown.  Development of this information would permit pH to be used as
a control parameter.

     High removal of the suspended solids resulting from the precipitates
which form in this process is also required.  This phase of the process can
be adequately monitored by suspended solids and turbidity measurements .

MECHANISM OF REMOVAL BY HIGH pH AND PRECIPITATES

     Removal of viruses by high pH conditions is a two fold process in which
irreversible inactivation occurs due to the pH effect and a physical removal
by incorporation into the precipitates.  It is well to point out at this
point that Donovan (36) has shown that part of the so called inactivation
of high concentrations of the f2 bacteriophage which may be caused by pH
alone is really only a loss of infectivity caused by aggregation in the
presence of calcium.  Each aggregate then represents only one infectious
unit rather than the number of infectious units that each virus in a deag-
gregated state would have.  When these aggregates are subsequently exposed
to a lower pH they may deaggregate and manifest themselves singly.

     Poliovirus has been shown to be inactivated under alkaline conditions
by a disruption of the capsid and a loss of the RNA to the water (42) (43)
(44).  Boeye and Van Elsen (42) suggested that the KNA when released from
the capsid was either already degraded or that it was quickly degraded upon
its release.  Their work was done at pH 10, and at 30C and higher temper-
atures.  Information does not appear to be available to determine exactly
how the proteins in the capsid are attacked to bring about this disruption.
Extrapolation of this mechanism to enteric viruses other than polio should
be done with care.  Anderson and Stephens (45) have shown that T6 bacterio-
phage underwent a general loss of structural integrity following exposure
to pH 9.6.  Protein sheaths covering the tail structures were "triggered"

                                    29

-------
in some cases and had become disintegrated from the base to the head.  Sepa-
rated heads, tails and tail base plates were observed in electron micrographs.

     Satisfactory explanations for the mechanisms of virus removal by
calcium carbonate and Mg(OH).2 precipitates have not been developed.  Ca and
Mg will react with carboxylate ions of proteins which would provide a virus
cation complex which could then be incorporated into the precipitate (3^-).
The negative charge of the calcium carbonate would minimize the electro-
static attractive forces between it and the virus particle.

     Calcium hydroxylapatite probably removes viruses by some of the same
mechanisms as aluminum hydroxide (24).  The hydroxylapatite is hydrolyzed
but possesses a negative charge.  The importance of the reactions at the
hydrolysis groups is probably pointed up by the lack of such sites on
calcium carbonate which removes viruses to a lower degree.
                                     30

-------
                                 SECTION 5

                 TECHNOLOGY APPLICATION FOR VIRUS REMOVAL

BACKGROUND

     Virus removal or inactivation is complicated by several virus phenomena:
their small size - as low as 15 Mm; the many different types which may be
present - over 100 enteric viruses are knovni; the state in which they exist -
as discrete particles or embedded within or attached to solid particles;
and the difficulty of detecting them in very small concentrations.  When
viruses are embedded within certain solids, such as their host cells, they
are extremely resistant to inactivation by chemical disinfectants (46) (4?).
In this case one must depend primarily upon the physical removal of the
cell by coagulation and sedimentation and/or filtration since the required
dosage for inactivation by ozone and chlorine are well beyond that normally
used.

     This section will discuss the utilization and application of coagula-
tion and high pH for virus removal and inactivation under field operating
conditions.  In cases where our present state of knowledge is incomplete
best estimates will be applied to the situation.

COAGULATION

     The coagulation process is relatively non specific in terms of its
ability to remove viruses when the principal precipitate formed is hydrated
metallic oxide.  This is the usual material formed in waters low in phos-
phate.  In wastewater effluent phosphate is present in larger amounts and
aluminum and ferric iron will react preferentially with the phosphate before
hydrated oxides are formed.  Viruses will absorb to aluminum or ferric phos-
phates and are subsequently removed with the phosphate although they are
less well absorbed on phosphate than on hydrated oxides.  Molar ratios of
aluminum or iron to phosphate in excess of 1:1 are desirable to provide
excess metal for the formation of the hydrated oxide.  Enteric virus removal
from 90 to 99-999 percent should be obtained in a water which has been well
coagulated, flocculated and settled.

     The coagulants of choice are either alum, ferric chloride or ferric
sulfate.  Insufficient experience has been gained with ferrous sulfate to
recommend its use at this time.  Coagulant dosages which will give the
minimum turbidity level after treatment are recommended.  Polyelectrolytes
are not generally useful in increasing the virus removal which would other-
wise be obtained when a water is properly coagulated and flocculated with
metallic coagulants.  In cases where the coagulation with the metallic
coagulant is inadequate due to insufficient dosage or poorly settling floe
a polyelectrolyte which will cause the floe to settle should give satisfactory
virus removal.  Cationic polyelectrolytes are preferred over the nonionic form
because of their greater density of positive charges.  Anionic polyelectro-
lytes are not recommended since a report has shown that virus removal is

                                     31

-------
decreased in their presence.  Polyelectrolyte addition may decrease the
coagulant dosage needed for virus removal.  It would appear desirable to
verify this conclusion in each specific application, however.

     Better and easier removals of viruses generally occur when turbidity
is present.  Turbidity in excess of 10 to 15 turbidity units would appear
desirable to obtain this effect.  This effect appears related to two
factors:  (l) the coagulation process and subsequent settling of the floe
is facilitated when turbidity is present, and (2) the turbidity presents an
increased surface area for adsorption of the virus which would" not otherwise
be present.  Viruses are known to absorb readily to turbidity surfaces in
water and are thereby removed when the turbidity is removed.  Maximum
removal of turbidity is also desirable in order to best prepare the water
for subsequent disinfection.  As mentioned above, viruses embedded within
body cells, fecal material, etc., are difficult to inactivate with chemical
disinfectants and a water with very low residual turbidity facilitates the
disinfection process.

     Virus removal by coagulation is slightly affected by water conditions.
The temperature and inorganic content probably affect virus removal in a
minor way provided that the coagulation and floe formation are otherwise
acceptable.  The affect of pH has not been well established where metallic
hydroxides are formed but a pH on the acid side, between 5~7j appears desir-
able.  Where aluminum phosphate is formed the optimum pH is lower, between
5 to 6.5.  Organic content, as in a treated secondary effluent, may reduce
slightly the virus removal otherwise expected.  In such cases there does
not appear to be an acceptable remedy except to optimize the coagulant
dosage to assure the lowest residual turbidity possible.

REMOVAL AMD INACTIVATION WITH HIGH pH

     Viruses are inactivated at high pH and are physically removed by ad-
sorption to precipitates formed at alkaline pHs in water.  Most enteric
viruses, however, are stable under acid conditions at pH values below those
of any interest in water or wastewater treatment.  The inactivation obtained
at high pH is a function of both the pH and the contact time so the higher
the pH within the lethal range and the longer the contact time the greater
the expected inactivation.  Removal by adsorption to precipitates is a
function of type of precipitate and the amount available.

     The pH for virus inactivation varies with the type of virus.  The few
which have been studied indicate that this may range from pH 10 to 12 for
an inactivation in excess of 90 percent in 60 to 100 minutes of contact
time.  A pH of 11.5 to 12 with contact times of about 90 minutes should
produce inactivations in excess of 90 percent for most viruses.  It is to
be emphasized that it is necessary for the virus itself to be exposed to
this pH condition and that for viruses embedded with solid particles this
probably will not occur.  Such particles and their viruses would require
removal by sedimentation.  The exact effect of increasing contact time has
not been determined but longer times will increase inactivation.  In the
absence of other information a 90-minute contact time appears acceptable.
                                     32

-------
     Hie principal precipitates formed in water and biologically treated
wastewater when lime is added to elevate the pH are calcium carbonate,
magnesium hydroxide and calcium hydroxylapatite.  Magnesium hydroxide and
calcium hydroxylapatite have greater adsorptive capabilities for viruses
than calcium carbonate.  Adsorption of viruses to these precipitates with
subsequent removal will occur well below the lethal pH inactivation range.
Increased volumes of precipitates increases the virus removal.

     Combined removals and inactivation of viruses in excess of 90 to 99
percent should be expected in the lime flocculation of a typical biologically
treated effluent with about 10 mg/1 of P.  Removals of about 90 percent or
more of the viruses should be expected from a typical drinking water excess
lime softening plant with a raw water which contains about 25 mg/1 or more
of magnesium as Mg.  Control of the process can be obtained by monitoring
pH and turbidity.  Turbidity removal should be maximized.  Organics in the
concentrations expected in a biological effluent should not interfere with
the process.
                                     33

-------
                                 REFERENCES

 1.  Chang,  S.L. ,  P.C.G.  Isaac and N. Baine.  Studies on Destruction of
     Bacterial Viruses III.  Dynamics of the Removal of Bacterial Virus
     (Bacteriophage Against Micrococcus Pyogenes var. ATbus)  in Water by
     Flocculation with Aluminum Sulfate.  Am. J. Hyg. 57 (3):  253-266, 1953.

 2.  Carlson, H.J., G.M.  Ridenour, and C.F. McKhann.  Efficacy of Standard
     Purification Methods in Removing Poliomyelitis Virus from Water.  Am.
     J. Publ. Hlth, 32 (11): 1256-1262, 19U2.

 3.  Kempf,  J.E.,  M.G. Wilson, M.E. Pierce, and M.H. Soule.  Effect of
     Aluminum Hydroxide Sedimentation, Sand Filtration and Chlorination on
     the Virus of Poliomyelitis.  Am. J. Publ. Hlth., 32 (12): 1366-1370,
     Wolf, H.W., R.S. Safferman, A.R. Mixson, and C.E. Stringer.  Virus
     Inactivation During Tertiary Treatment.  J. Am. Water Works Assn.,
     66 (9): 526-531,
 5.  Thorup, R.T. , F.P. Nixon, D.F. Wentworth, and O.J.  Sproul.  Virus
     Removal by Coagulation with Polyelectrolytes .  J. Am. Water Works Assn.
     62 (2): 97-101, 1970.

 6.  Boardman, G.D. and O.J. Sproul.  Absorption as a Protective Mechanism
     for Poliovirus.  In: Proceedings 97th Am. Water Works Assn. Conf.
     Denver, Colorado, 1977.  Part 1.  lUA-^.

 7.  Sanitation Districts of Los Angeles County.  Pomona Virus Study Final
     Report.  Los Angeles, California, 1977, 230 pp.

 8.  Foliguet, J.M. and S. Michelet.  The Elimination of Enteroviruses During
     the Treatment of the Public Water Supply,  (in French) Rev. Hyg. et
     Med. Soc., 17 (2): 175-206, 1969.

 9.  Foliguet, J.M. and F. Doncoeur.  Elimination of Enteroviruses During
     Treatment of Drinking Water by Coagulation-Flocculation-Filtration.
     (In French) Water Res. 9 (11): 953-961, 1975.

10.  Guy, M.D., J.D. Mclver and M.J. Lewis.  The Removal of Virus by a
     Pilot Treatment Plant. Water Res. 11 (5): 1+21-^28,  1977.

-------
11.  Chang, S.L., R.E. Stevenson, A.R. Bryant, R.L. Woodward and P.W. Kabler.
     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 or Ferric Chloride under
     Various Testing Conditions.  Am. J. Public Health k8 (l): 51-61, 1958.

12.  Chang, S.L., R.E. Stevenson, A.R. Bryant, R.L. Woodward, and P.W.
     Kabler.  Removal of Coxsackie and Bacterial Viruses in Water by Floccula-
     tion, 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. Publ. Hlth., 3 (2): 159-169, 1958.

13.  Chaudhuri, M. and R.S. Engelbrecht.  Removal of Viruses from Water by
     Chemical Coagulation and Flocculation.  J. Am. Water Works Assn., 62 (9):
     563-567, 1970.

Ik.  York, D.W.  Virus Removal by Chemical Coagulation.  M.S. Thesis,
     Univ. of Tennessee, 1970.  Ilk pp.

15.  York, D.W. and W.A. Drewry.  Virus Removal by Chemical Coagulation.
     J. Am. Water Works Assn. 66 (12): 711-716, 197^.

16.  Shelton, S.P., and W.A. Drewry.  Tests of Coagulants for the Reduction
     of Viruses, Turbidity, and Chemical Oxygen Demand.  J. Am. Water Works
     Assn., 65 (10): 627-635, 1973.

17.  Manwaring, J.F., M. Chaudhuri, and R.S. Engelbrecht.  Removal of Viruses
     by Coagulation and Flocculation.  J. Am. Water Works Assn., 63 (5):
     298-300, 1971.

18.  Lovtsevich, E.L., V.A. Riabchenko, A.I. Mel'nikova and L.G. Mel'nikova.
     Sanitary-Virological Efficiency Evaluation of Pilot Facilities Methods
     for Treating Potable Water,  (in Russian.)  In Gig. Vody Okhr. Vodoemov.
     G.I. Sidorenko, ed.  Akademiya Meditisinskikh Wauk SSSR Institute,
     Moscow. 1973. PP- 19-22.

19.  Stumm, W.  Discussion of "Methods for Removal of Phosphorus and Nitrogen
     from Sewage Plant Effluents".  In: Advances in Water Pollution Research,
     Pergamon Press, New York, 196^, Vol. II, pp. 207-

20.  Brunner, D.R., and O.J. Sproul.  Virus Inactivation During Phosphate
     Precipitation.  J. San. Engr. Div., ASCE, 96 (SA2): 365-379, 1970.

21.  Foliguet, J.M., F. Doncoeur and J.C. Block.  A Contribution to Study
     on the Removal of Enteroviruses During Drinking Water Treatment,  (in
     French.)  J. Francais D'Hydrologie, lU: 11-2U, 197^.

22.  Cliver, D.O.  Viruses in Water and Wastewater:  Effects of Some Treat-
     ment Methods.  In: Virus and Water Quality: Occurrence and Control,
     V. Snoeyink, ed. Proceedings of the 30th Water Quality Conference,
     1971.  Univ. of 111. Bull 69, 1971.  pp. 1^8-158.
                                     35

-------
23.  Chaudhuri, M. and R.S. Engelbrecht.  Virus Removal in Wastewater
     Renovation by Chemical and Coagulation and Flocculation.  In:  Advances
     in Water Pollution Research, Pergamon Press, New York.  Proc. 5th
     International Conf. on Water Pollution Research, San Francisco, Calif.,
     1970.

2k.  Cookson, J.T.  The Chemistry of Virus Concentration by Chemical Methods.
     In: Developments in Industrial Microbiology, Vol. 15.  Am. Inst. of
     the Soc. for Ind. Microbiology, 197^- pp. 160-173.

25.  Salo, R.J., and D.O. Cliver.  Effect of Acid pH, Salts, and Temperature
     on the Infectivity and Physical Integrity of Enteroviruses.  Arch, of
     Virology 52 (4): 269-282, 1976.

26.  Robinson, L.K.  Effect of Heat and of pH on Strains of Coxsackie Virus.
     Proc. Soc. Exp. Biol. Med. 75 (2): 580-582, 1950.

27.  Ginsberg, H.S.  Characteristics of the New Respiratory Viruses (Adeno-
     viruses) II.  Stability to Temperature and pH Alterations.  Proc. Soc.
     Exp. Biol. Med., 93 (1): 48-52, 1956.

28.  Fields, H.A. and T.G. Metcalf.  Concentration of Adenovirus from
     Seawater.  Water Res. 9 (4): 357-364, 1975.

29.  Sproul, O.J., R.T. Thorup, D.F. Wentworth, and J.S. Atwell.  Salt
     and Virus Inactivation by Chlorine and High pH.  In: Proceedings of the
     Specialty Conference on Disinfection, Am. Soc. of Civil Engineers,
     New York, N.Y. 1970. pp. 385-396.

30.  Boardman, G.D.  Protection of Water-Borne Viruses by Virtue of their
     Affiliation with Particulate Matter.  Ph.D. Thesis, University of
     Maine at Orono, Orono, ME 1976.  196 pp.

31.  Berg, G., R.B. Dean, and D.R. Dahling.  Removal of Poliovirus 1 from
     Secondary Effluents by Lime Flocculation and Rapid Sand Filtration.
     J. Am. Water Works Assn., 60 (2): 193-198, 1968.

32.  Sproul, O.J.  Investigation to Increase the Viricidal Capacity of
     Disinfectant, Germicidal, and Fungicidal, Phenolic, Dry-Type.  Technical
     Report TR 75-90 FSL, U.S. Army, Natick Dev. Center, Natick, Mass.,
     1975- 38 pp.

33.  Berg, G. and D. Berman.  Effectiveness of Military Phenolic Dry-Type
     Disinfectant (MIL-D-51061) Against Viruses.  Final Report to Fungicides
     and Germicides Laboratory, Pioneering Research Division, U.S. Army
     Natick Laboratories, Natick, Mass. 1967. pp. 18 plus 3 tables and
     55 figures.

34.  Thayer, S.E., and O.J. Sproul.  Virus Inactivation in Water-Softening
     Precipitation Processes.  J. Am. Water Works Assn., 58 (8): 1063-1074,
     1966.


                                     36

-------
35-  Boardman, G.D. and O.J. Sproul.  Protection of Viruses During Disin-
     fection by Adsorption to Particulate Matter.  J. Water Poll. Control
     Fed., If9 (8): 1857-1861, 1977.

36.  Donovan, T.K.  Virus Inactivation Associated with Lime Precipitation
     of Phosphate From Sewage.  Ph.D. Thesis, Johns Hopkins University,
     Baltimore, Md., 1972. 133 pp.

37.  Wentworth, D.F., R.T. Thorup, and O.J. Sproul.  Poliovirus Inactivation
     in Water-Softening Precipitation Processes.  J. Am. Water Works Assn.,
     60 (8): 939-9^6, 1968.

38.  McCarty, P.L., M. Reinhard, C. Dolce, H. Nguyen andD.G. Argo.  Water
     Factory 21: Reclaimed Water, Volatile Organics, Virus and Treatment
     Performance.  EPA-600/.2-78-076, U.S. Environmental Protection Agency,
     Cincinnati, Ohio, 1978. 88 pp.

39.  Sattar, S.A., S. Ramia, and J.C.N. Westwood.  Calcium Hydroxide (Lime)
     and the Elimination of Human Pathogenic Viruses from Sewage:  Studies
     with Experimentally-Contaminated (Poliovirus type I, Sabin) and Pilot
     Plant Samples.  Can. J. Publ. Health, 67 (3): 221-225, 1976.

kO.  Hiser, L.L.  Demonstration of Virus Removal from Municipal Sewage.
     Final Report SwRI Project No. 01-2831, Southwest Research Institute,
     San Antonio, Texas, 197^. 75 pp.

Ul.  Grabow, W.D.K., I.G. Middendorff and N.C. Bassow.  Role of Lime
     Treatment in the Removal of Bacteria, Enteric Viruses and Coliphages
     in a Wastewater Reclamation Plant.  Appl. and Environ. Microb. 35 W'
     663-669, 1978.

k2.  Boeye, A. and A. Van Elsen.  Alkaline Disruption of Poliovirus:  Kinetics
     and Purification of RNA-Free Particles.  Virology, 33 (2): 335-3^3,
     1967.

^3-  Van Elsen, A. and A. Boeye.  Disruption of Type 1 Poliovirus under
     Alkaline Conditions:  Role of pH, Temperature, and Sodium Dodecyl
     Sulfate (SDS).  Virology, 28 (3): ^81-1+83, 1966.

kk.  Maizel, J.V., Jr., B.A. Phillips, and D.F. Summers.  Composition of
     Artificially Produced and Naturally Occurring Empty Capsids of Polio-
     virus Type 1.  Virology, 32 (4): 692-699, 1967.

^5-  Anderson, T.F. and R. Stephens.  Decomposition of T6 Bacteriophage in
     Alkaline Solutions.  Virology, 23 (1): 113-116, 196^.

h6.  Hoff, J.C.  The Relationship of Turbidity to Disinfection of Potable
     Water, pp. 103-117.  In: Evaluation of the Microbiology Standards for
     Drinking Water.  C.W. Hendricks, ed.  EPA-570/9-78-006, U.S. Environ-
     mental Protection Agency, Office of Drinking Water, Washington, D.C.
     235 PP.


                                     37

-------
4 7.  Sproul, O.J., C.E. Buck, M.A. Emerson, D.S. Boyce, D.S. Walsh and D.M.
     Howser.  Effect of Particulates on Ozone Disinfection of Bacteria and
     Viruses in Water.  EPA-600/2-79-089, U.S. Environmental Protection
     Agency, Cincinnati, OH. 86 pp.
                                     38

-------
                                    TECHNICAL REPORT DATA
                            (Please read Instructions on the reverse before completing)
 1. REPORT NO.
     EPA-600/2-80-004
                              2.
                                                            3. RECIPIENT'S ACCESSION1 NO.
4. TITLE AND SUBTITLE
     CRITICAL REVIEW OF VIRUS REMOVAL BY COAGULATION
     PROCESSES AND  pH MODIFICATIONS
                                                            5. REPORT DATE
                                         June 1980 (Issuing Date)
                                      6. PERFORMING ORGANIZATION CODE
7. AUTHOR(S)
     Otis J.  Sproul
                                                            8. PERFORMING ORGANIZATION REPORT NO.
9. PERFORMING ORGANIZATION NAME AND ADDRESS
     Department  of Civil Engineering
     The Ohio  State University
     Columbus, Ohio   43210
                                       10. PROGRAM ELEMENT NO.
                                        35B1C,AP C611A, SOS 4,  Task 13
                                       11. CONTRACT/GRANT NO.
                                                                 R805771
 12. SPONSORING AGENCY NAME AND ADDRESS
     Municipal  Environmental Research LaboratoryGin.,OH
     Office of  Research and Development
     U.S. Environmental Protection  Agency
     Cincinnati,  Ohio    45268	
                                       13. TYPE OF REPORT AND PERIOD COVERED
                                       Final  10-77 to 5-79
                                       14. SPONSORING AGENCY CODE
                                        EPA/600/14
 15. SUPPLEMENTARY NOTES
     Project Officer - John N. English  513/684-7613
 16. ABSTRACT
     Operation  of  advanced wastewater and water supply treatment plants  to assure
     virological safety of the effluent relies on  each unit process  removing a
     finite number of viruses.   These treatment plants frequently use  chemical
     coagulation and precipitation at high pH with hydrated lime as  part of the
     process.   These treatment methods offer important opportunities  for removal and
     inactivation  of viruses  from  water and wastewater.   This report  is  a literature
     review which  examined the effectiveness of these  processes in removing viruses.
17.
                                KEY WORDS AND DOCUMENT ANALYSIS
                  DESCRIPTORS
                                               b.lDENTIFIERS/OPEN ENDED TERMS
                                                    c.  COS AT I Field/Group
     Viruses
     Coagulation
     pH
     Polyelectrolytes
     Lime
     Metallic Coagulants
Wastewater  Treatment
Water Treatment
                                                        13B
13. DISTRIBUTION STATEMENT
     Release to  Public
                                               19. SECURITY CLASS (ThisReport)
                                               Unclassified	
                                                    21. NO. OF PAGES
                                                        49
                         20. SECURITY CLASS (Thispage)

                         Unclassified	
                                                                          22. PRICE
EPA Form 2220-1 (9-73)
                                             39
                                                                   ? U.S. GOVERNMENT PRINTING OFFICE: 1980-657-146/5694

-------

-------

-------

-------

-------
           .

            o O


              s
en
o
o
oo
o
o
o
2 S
C -,

3S

5 S

9 "
re o

              m
             .
            CL O
            D 31
                   
                   .  Q, cil
                   IT O V>
                   Qj J Ct)
  -1

  9-
_  fo
3  Q,
  CT
D-  o
  <

c  


1  s-
to  CC
                                                                                              u :?
                                                                                              QJ O
                                                                                             -< 


                                                                                              CD
                                                                                             CO

                                                                                             O

                                                                                             O
                                                                                                 O
en
T3
CD
O
Q)

-n
O
CD C
O 3
o =r
rn i> ~D rn ~n "D
TltO 3 3 5 0
>   <.  
1 D CD - QJ
CO O O O -D(Q
CO -< CT .3 a) CD
01 O 3 =
-i 3 Q- Q}
J CD 3
2 O.
QJ


-------