-------
      In 1989, the Water Pollution Control Federation received survey responses from
1003 wastewater treatment plants in the United States.  The survey of "Problem
Technologies and Design Deficiencies at POTWV was intended to identify information
needed to improve wastewater treatment plant design and operation. The WPCF Plant
Survey database was utilized by this study as a means to determine the source of
corrosion problems at wastewater treatment plants.  The survey asked the  1003
participants to rate the level of problems in various areas of the plant and also asked
questions about plant age, daily flow, and methods of operation.

      Several questions dealt with the level of corrosion experienced in major portions
of the plant, as well as the incidence  of influent H2S and odors.  A summary of the
responses related to corrosion problems is shown in Table 2-4.  The approach taken to
determine the source of corrosion was to develop a matrix of responses in an attempt to
correlate the  incidence of corrosion to such factors as influent H2S, odors, age and
recycle streams.

      Overall, it was found  that 68 percent of the plants surveyed experienced some
level of corrosion within the plant The same level of corrosion is generally experienced
in all portions of the plant  For example, if the respondent classified corrosion
problems at preliminary treatment as major but periodic, then the corrosion problems
experienced at the secondary clarifier tended to be rated the same.

      The age of the plant does not explain the consistent level of corrosion problems
in all sections of the plant  When  compared to the level of corrosion at preliminary
treatment,  the same age distribution was seen for all levels of corrosion.  The plants
reporting no  corrosion at preliminary treatment are generally over 20 years old  and have
had major  liquid and sludge train expansions in the last 5 years. The plants showing
some degree  of corrosion are between 1-15 years old but have also had train expansions
in the last 5 years.

      As for plant size, the majority of wastewater treatment plants  without corrosion
problems in preliminary treatment handle 0.1 to 1.0 mgd (42%) or 1 to 5 mgd (31%)
average flow.  The plants exhibiting corrosion are rather large - 34 percent treat
between 1  and 5 mgd.  One interesting finding is that of the largest wastewater
treatment plants (>10 mgd), those reporting corrosion outnumber those without
corrosion by  almost 7 to 1. The distribution of corrosion severity for each size category
is generally the same, but with a tendency for larger plants to show slightly more severe
corrosion problems.

    '  The recycle of filtrate or supernatant streams from digesters, thickeners and
dewatering equipment does  not appear to increase the incidence of corrosion at
preliminary treatment The distribution of responses was similar for those plants which
recycled and  those which did not  There  was also no tendency for an increase in
corrosion with a particular recycle stream. For all recycle streams, the level of corrosion


                                       2-38

-------
ee
             QO   O\
             \0   Tf
                                      ON   «o   i-"   1-1




z
o
oS
0
c£
Otf
SURVEY - CO
NT SYSTEMS
*5
2 ^H
H £ 5
J O «
ea H H
tS MM
u u
c/5 Sx
z <

eu 2
tfi L*
s<
tb ^
O to
> o
§
s
•K







^
g
V
cr
Si
Lu
CM (/J
CQ jj
>^ Q
'*" S
VJ ^
2 «
£ cu

.0
c/l
o
h*
3










£
2
2
cu
u.
.0
S





E

r*.
O
kn
w»
c
c
'i



3
O
3
e
O
U




Periodic

V)
3
c
£
'C
c
5






0
•5
-------
is rated as minor periodic.

      The primary relationship of concern to this study is that of influent H2S to
corrosion.  For preliminary treatment the survey participants were asked to rate the
severity of problems with corrosion and hydrogen sulfide.  The matrix of responses
indicates that for any level of corrosion problem most wastewater treatment plants
generally have minor periodic or no troubles with sulfide.  It is also noted, however, that
the majority of plants reporting major continuous problems with  H2S also report the
same level  of severity for corrosion.

      The severity of odor releases was also compared to the level of corrosion
problems in preliminary treatment  Of those plants reporting major continuous
corrosion problems, the majority of responses were evenly divided among minor  and
major odor problems.  Only a few of these plants reported no odor problems.

      The matrix of responses for corrosion and odor was similar to that for corrosion
and sulfide. It was found that wastewater treatment plants seem to recognize a
relationship between odor releases and the presence of hydrogen sulfide.  In general,
the same response to the severity of H2S was also given to the problems with odor
releases. Approximately 30% of the wastewater treatment plants surveyed reported no
problems with either.

      As part of the selection process for site visits, information was analyzed from 34
cities reported to have hydrogen sulfide corrosion problems. This information is
summarized in Table 2-5.

      Other organizations, manufacturers, and contractors were contacted to gain
additional perspective on the  national extent of hydrogen sulfide corrosion.  These
entities included the Clay Pipe Institute, National Association of Sewer Service
Contractors, Insituform  of North America, Spirolite Corporation, Ameron Corporation,
Sauereisen Cement Co., LaFarge Cement Co., and Specialty Sewer Services, Inc., and
others.  Highlights of information collected from these sources are briefly summarized
in Table 2-6.

      Figure 2-1  is a map pinpointing locations where severe corrosion problems are
judged to exist in  the sewer system or treatment plant This is based on EPA site
investigations, surveys conducted by other organizations and the  experiences of
professionals active in the field  of hydrogen sulfide corrosion control.  This does not
represent all the cities experiencing  severe corrosion problems.

      Figure 2-2  is a map which shows the frequency of use of a proprietary, corrosion-
resistant liner for  concrete pipe. This type of liner is specified during design for
concrete pipes which may be  subjected  to hydrogen sulfide corrosion.  The map  does
not represent actual corrosion problems.


                                       2-40
-------
     UJ
     Ul
*   5
^   s
a   2
     Ul

     u

     a
     U)
                « U|
                a *•
              252

              91s
              1s-
              f- O
              Z Ul
              ui z :
              u *• .
              v- r -»
           0. O ~» —
           'J •*. ft S,
5 Si as S
            at at at s z s :
                         10        mo
                         M        «n •—
                                   o o     oomo — oooooooo-rooooooo   o o o
                                    i O     O 0» O •
                                                                               O Q  —
                                   *rf O  * O  U V O (K
                                                                          '«   M •• J 3 «^
                                                                                      o a — ^-  •
                                                                                   *'"  • O  S
                                                             2-41
-------
> 5
2
m
-1
|S


|,
|1






3

V)
o
1
w



a
Ul
Iv
z
ul
s
&•
s



a
M
Ul <
H" Ul
< at
z





3
u
£Z















Ul  z















i_ — '
fc3
^
a.
s=


55
1*






en a*

o H»
5 u
— ul
3±
U ul




«•
O 

Mt
U I/)

OK 0&






M
O O
t/l »•*
O •«
Is
o "^



ae
>•
f— — •

23
U, I/)



ul









s»
U
*i o o nan

>«0>«-«om«**»«tr*p»\oot<*i»oo»o°>'~
















>. »i » M M Z »•»•>• 1-








c a u







-


















M — »
3— 3= O
35 W U. ••
U) Ul U Ul Ul O
^ui a a ui a a ui








O Ml^flOOWtOOOO








i

* P^ *g ^ 2 M

£s .-a u .§ .-
 — o




a o >» o















»• M









3S ^*








Z Z
















M
^ C3 O
gggg








O O










u*

M —
. f — Z

Z^35 J



O




f^










*








I
^








Z



















o







o












J
•
M
•M
g
a.

mm<*i

O 0 0
"""2



o>m>e
•* * <•»










•*»•>•








5
^ ^ a








» » z

















^^
B
30i
W ^J IH







O m o












!*s
3 - —
— . ^ 3
BOO
s± .
a. BE to

r^
0
m
|A




(H















































=
S
^







>n










^

o
'*
e
e
e
CO
«
«* «««•« at m
«•> «D 0 0 O
iS» <^ «O irt •*
e a o o a



e o o o o
•t a> at -• m
rt -» — •»










*








r.^,1









3£ X 3£ X 31
















S^
M M S
U Uw U 35 U
a a a a a
^ ^ ^ so ^«







•a irt o o -»








5

•
A
5 S
-5 «£
0 U <
M -U. •
« «l - —
— — - a a
«* a. w ••
a « a « fa
CO «1 *- M 3»

fa
S
«•*
«ri
41
U
O
11


u
3
3
fa

a
o



•



•*
o

*•*

4
u
MI

•
o
~
s
s

•*
M


"
e

•
u
w
M
O
•^
fa
*J
«H

H
U

»
V S
M o
o -*
fa 4
a. u
o
—^ u
i r

*rf U
s u
v a,

u
3 X
II S
» o
II

«*
^ f
& 4*
II ««
e» •
>>
• •
IM
O M
a 3
? J
£ ?
o
o" i
II M
M <
<"• N



















U
-3
•«
O
II


•
4*

-
O

•*
2

«*



o
II

•
&

&

*•
^J
1




•*
^
•
•
e
V



w
41


•
^»


U
M
M
•M

O
II
u

•
•
«
u
^


B-2
O w
•* ta
U *«
«•
ta u
it tl
u, ^
II •
s •
a a
o •
v •
— • u
S H

Q
— O
O "•
• o
H W
o e
u

e" «
o •
— 3
•rt II
u S
it

a v
= !"
3 ^

z «

ll •?
•
a f
O 3
u a
^** ^
B
— 8
^

Z*
2 ^

Ul II
_1
-* „
9 M
— it
il •«

e

^
-•• u
• M
a -
a
» o
a •»
O X
~» u
U — 1
<> a
s
?1
u S

a
e ^i

^ •••
U M
V -M
••^ u
a ii
— at



_"•'
TT
a -
a
• o
a —
o •*
•M U
M u
U — I
ti a
•« «
C*J
G±
ii u
< ••
•













































































^^

a*
10
»«
u

2-42
-------
                                  TABLE 2-6

       SUMMARY OF INFORMATION FROM SELECTED ASSOCIATIONS,
                   MANUFACTURERS, AND CONTRACTORS
      Source


Clay Pipe Institute
         Comments
From 1950 to 1969, over 154 miles of
severely corroded concrete pipe in 57
cities was replaced with clay pipe, with
over 5% of production  tonnage used
for this purpose
Spirolite Corporation
From 1986  to 1989, approx. 20 miles
of sewer was sliplined using Spirolite
polyethylene pipe
Insituform of North America
Over  100 miles of  large  diameter
sewer was lined using cured-in-place
inversion lining in more than 36 U.S.
cities between 1977 and 1988
Ameron Corporation
From 1947 to 1988, over 900 miles of
sewer pipe  in 500 U.S. projects was
specified with T-lock liners to prevent
                                             crown corrosion
                                     2-43
-------
w
K
D
O
     CO
  CO  CO

   ^ w

   ^g
   g^

   ^^
   CO
     CO
     
-------
CM

 I
O
         PU
3 0s
^ o
o is
> o
p_ o


£o
         Q

     K
     CO

     D
                                          2-45
-------
2.6   Case Studies

      To gain further insight into the causes and prediction of hydrogen sulfide
generation and subsequent hydrogen sulfide corrosion, case histories were prepared
from previous evaluations  of five wastewater collection systems located in three cities.
The following systems were investigated:

             Sacramento County, California, Central Trunk Sewer
             Sacramento,  California, Regional Interceptor System
             City of Lakeland, Florida, Western Trunk Sewer
             City of Omaha, Nebraska, Papillion Creek Wastewater System
             City of Omaha, Nebraska, South Interceptor Sewer

A brief summary of each of the five case studies is provided below (2)(3):

2.6.1  Sacramento County, California, Central Truck Sewer

2.6.1.1 Description and History

      The Central Trunk  Sewer conveys both domestic and industrial wastewater and,
for several years, conveyed sludge from two upstream wastewater treatment plants. The
trunk is approximately 16 miles long.  About 2 miles of the upper reach is vitrified clay
pipe ranging from 18 to 24 inches in diameter.  The remaining 14 miles is granitic-
aggregate reinforced concrete pipe 27 to 60 inches in diameter.  The slope varies from
0.18 percent in the upper reaches to 0.05 percent in the lower end of the trunk.

      The Central Trunk  Sewer was constructed in 1962. The  following is a brief
history of the system:

       •     1964—The entire line was visually  inspected by County personnel by
             floating on rafts through each reach.  A powder deposit on the inside of
             the pipe was the only evidence of  corrosion.

       •     1968—Anaerobically digested sludge discharge to the Central Truck Sewer
             was initiated.

       •     1968—The entire pipeline was again visually inspected and samples of
             concrete were taken. The pH of the concrete walls was greater than 3.0.
             Maximum corrosion at that time was estimated to be 1/4 inch, and no
             coarse  aggregate was exposed.

       •     1969—Dr. Richard Pomeroy completed an analysis of portions of the
             trunk.  Existing corrosion was estimated to be 1/8 or 1/4 inch.  The useful
             life of the trunk was estimated to be 100 years.


                                       2-46
-------
      •     1972-The County studied the effects of chlorine on sulfide generation in
             the Central  Trunk system.

      •     1972-1973—Vitrified clay plugs were installed at the manholes in the trunk.
             Three sulfide control facilities were designed and one was bid.  Due to the
             high price of the bid, the sulfide control facilities were not constructed.

      •     1976—The clay plug locations were inspected.  Significant corrosion was
             evident, and a hydrogen sulfide corrosion study was initiated in September
             1976.

      •     1979—Three chlorine injection  stations were placed in service along the
             Central Trunk to reduce sulfide levels and to minimize further corrosion.

      •     1982-83~Sludge discharge to the Central Trunk was stopped.

      •     1984—The County completed another field investigation which showed
             continued corrosion, but at a reduced rate.

      •     1984--A 1,800-foot reach of 60-inch-diameter pipe from the Central Trunk
             Sewer was abandoned primarily due to a new alignment, and partially due
             to concern about the structural integrity of the  pipe.

      •     1987-County inspection of the Central Trunk showed reduced sulfide
             concentrations and indications of reduced corrosion rates.

2.6.1.2  Summary of Results

      Average annual  total sulfide concentrations ranged from approximately 0.5 to 1.5
mg/1 from 1965 to 1976, based on weekly or  monthly sampling at 12 noon at one
location. Average 24-hour sulfide concentrations were approximately 40 percent higher.

      Corrosion penetration was as much as 1.5 inches at some locations in the sewer.
Cores taken from the crowns of the pipes showed that the worst corrosion conditions
usually existed in the first 30 ft downstream of the manholes.

      Predicted sulfide concentrations from the  Pomeroy-Parkhurst equation were
compared with measured  values. When the  effect of sewer junctions was considered,
the predictive equation estimated sulfide concentrations with reasonable accuracy.
Predicted corrosion  penetration was also compared with measured values, and showed
an excellent correlation. It was also found that peaking factors  of 2 were justified to
account for minor turbulence at manholes. Higher peaking factors would be required
where high turbulence levels were encountered to account for the increase in off-gassing
of H2S.


                                       2-47
-------
      Dead load testing was conducted on a 12 ft section of 60 inch diameter pipe in
1984. The pipe was highly corroded, but reinforcing steel was not exposed. Tests
showed  that, although concrete loss had occurred, the pipe strength was still significantly
greater  than the original specification for dead load.

      Based on inspections in 1987, the rate of corrosion was reduced from that
observed in the 1970's due to elimination of sludge  discharges and installation of
chlorination stations.

2.6.1.3  Findings and  Conclusions

1.    The observed corrosion information from 1976 is in general agreement with
      Pomeroy's corrosion predictive equation, if conservative assumptions are made
      and input data are based on field measurements and monitoring data.

2.    Valid, positive  measurements of the depth of actual corrosion of in-place pipe for
      the Central Trunk were difficult Accuracy of all methods used was less than
      desirable.

3.    The discharge of anaerobically digested sludge to the Central Trunk for 14 years
      was partly responsible for the higher corrosion rates over those years.  Sludge
      discharges increased the BOD and the temperature of the wastewater  in the
      Central Trunk.

4.    With removal of the sludge discharges and installation of chlorination  stations
      along the Central Trunk Sewer, the sulfide concentration has been reduced in the
      1980's, and the rate of corrosion appears to have dropped significantly.

5.    The Central Trunk is believed to be made from spun RCP.   It seemed to take
      several years before corrosion penetrated the surface layer of this granitic
      aggregate pipe. Once the  high-alkalinity surface layer had been corroded,  the
      corrosion rate increased, since the alkalinity  of the rest of the pipe concrete was
      only 16 percent

6.    The corrosion information on manholes  and  structures with turbulent  flow
      characteristics is particularly interesting.  This information points out the need to
      use conservative  corrosion rate peaking factors in predicting  hydrogen  sulfide
      corrosion rates within close proximity to these locations (within several pipe
      diameters).
                                        2-48
-------
2.6.2  Sacramento, California Regional Interceptor System

2.6.2.1 Description and History

      The Sacramento terrain is quite flat, and the climate features hot summers and
mild winters. The collection system extends 20 miles east of the Sacramento River, and
for more than 20 miles north to  south. The longest interceptor, nearly 30 miles long,
drops less than 200 feet from end to end.  Several pumping stations are included in the
system.

      The Regional Interceptor System is extensive in scope, with a total capital cost of
about $143 million.  It was constructed during the period 1975 to 1982, and
encompasses about 62 miles of gravity sewer, over 25 miles being pipe in the 60- to 120-
inch diameter range.

      In designing the regional  interceptors, a number of passive measures were
implemented:

1.     County Source Control Ordinance. Sacramento County  passed an ordinance in
      1977 controlling the quality of industrial waste discharges into the  Regional
      System.

2.     Calcareous Aggregate.  Use of sacrificial calcareous (rather than granitic)
      aggregate was specified for all concrete pipe construction.

3.     Turbulence Control.  Junction structures were designed for smooth transitions to
      minimize wastewater turbulence.

4.     Lining of Hydraulic Structures. Junction structures, and other  hydraulic elements
      especially vulnerable to corrosion, were lined with plastic (locking  PVC liners).

5.     Slope/Velocity.  Pipe slope and velocity were carefully evaluated to limit solids
      accumulations within the  system.

      An extensive study to determine needed sulfide  controls  commenced in 1974 and
was completed in 1976. Several sulfide control measures appeared feasible from this
study for use in the Regional Interceptor System. Except for chlorination, no reliable
performance data existed prior to the mid-1970's for the control measures being
considered.  To develop this information, a field testing program of several sulfide
control measures was undertaken in  the summer of 1974.  Information was obtained for
the following:

      1.     Air injection in a force main (injected at  pumping station)
      2.     Hydrogen peroxide injection in a force main


                                       2-49
-------
      3.     Oxygen injection in a force main
      4.     Chlorination in a force main and gravity sewer
      5.     U-tube aeration (air) at the end of a force main
      6.     U-tube aeration (oxygen) at the end of a force main

      For upstream pumping stations, chlorine was selected to provide control of
hydrogen sulfide generation.  However, for a 60-inch diameter and a 72-inch diameter
force main, injection of high purity oxygen into a fall structure was selected as the most
cost-effective alternative. The injection of high purity oxygen enriches the atmosphere
of the fall structure with oxygen  to allow entrainment of oxygen through turbulence
induced by the fall, and elevation of dissolved oxygen levels in the wastewater, thus
allowing oxidation of existing sulfide and prevention of further sulfide generation.

      Attention was also focused on the potential problems of solids deposits,
especially  during the early years  of system operation.  Since solids deposits can generate
sulfide,  the objective of the design was to eliminate the possibility of solids
accumulation, or otherwise plan  for removal of any such deposition.

      The predicted lack of adequate velocities during early years of operation in
several interceptors led the designers to a decision to construct flushing facilities at two
locations.  In this manner, flows could be greatly increased for a few hours at a time
during the long periods of low dry weather flows, flushing solids  down the system of
interceptors.

      The Pomeroy-Parkhurst sulfide prediction equations and the  Pomeroy corrosion
prediction equation were used extensively during design of all  regional interceptors.
Many different assumptions were used to determine the best mix of passive and active
sulfide controls to provide assurance of long pipe life (100-year minimum) and lack of
odor problems.

2.6.2.2  Summary of Results

      Chlorine gas is used at upstream pumping stations and  at three locations along
the Central Trunk. The chlorination stations have been successful in maintaining low
dissolved sulfide levels (generally less than 0.3 mg/1)  at the force main discharge points.

      Solids deposition in the interceptor system has not been a problem.  Peak wet
weather flows have provided  adequate flushing  of the lines, and  use of the flushing
stations has not been required in the northern portion of the system.

      Two oxygen injection stations were constructed and operated at fall structures
(4).  In one system, DO concentrations of 13 mg/1  have been achieved in the wastewater
at a point nearly 4000  ft downstream of the fall structure.  Similar results have been
reported for the other  oxygen injection system.  Dissolved sulfide concentrations


                                       2-50
-------
downstream of the fall structures are consistently less than 0.5 mg/1.

      The primary problem with these stations has been that hydrocarbon compounds
are stripped from the wastewater under the turbulent conditions found in these
structures.  Explosion meters that monitor the  Lower Explosive Limit (LEL) are used in
each station. When  25 percent LEL is reached, pure oxygen addition is shut down.
This shutdown event occurs often.  In the summer months, 25 percent LEL is reached
on almost a continuous basis.  Investigation by the District into the source of the
hydrocarbons has shown that it may be methane gas generated under the anaerobic
conditions in the upstream force mains and gravity sewers, and within the combined
sewer system of the City of Sacramento.

      Vitrified clay plugs were installed at many locations in the pipe of the Regional
Interceptor System to allow future measurement of the corrosion of the pipe
surrounding the plug.  The District has had difficulty obtaining measurements from
these plugs due to a  variety of problems including: problems in finding the plugs in  the
dark, uncomfortable  environment  of the sewers, and, if found, difficulty achieving highly
accurate measurements due to the conditions.  Collected data have been analyzed and,
in general, indicate that corrosion to date is very low.

2.6.2_3   Findings and Conclusions

1.    Although only a few years of monitoring data are available, the information
      suggests that the Regional Interceptor System design, and its sulfide control
      systems, are providing  the level of protection anticipated.

2.    District monitoring work has confirmed  the need for PVC, or other type of non-
      corrodible lining material, for all junctions  and structures where even limited
      turbulence occurs and  where wastewater contains minor amounts of dissolved
      sulfide.

3.    The lack of accurate measurements on the  amount of corrosion in  the early years
      of operation of the Regional Interceptor System has been frustrating.  The level
      of accuracy needed is in  the range of hundredths of an inch. The level of
      accuracy for corrosion  measurements needs to be improved.

4.    The stripping  of hydrocarbon compounds in the fall structures has caused
      additional safety considerations that were not initially anticipated during design.
      In the summer,  one of the fall structures is now bypassed, and oxygen is injected
      in the bypass.  This results  in satisfactory oxidation of sulfide at the ends of the
      two long, large-diameter force mains.

5.    Solids deposits have not been a problem, and do not appear to produce any
      significant sulfide.


                                       2-51
-------
6.     Chlorination in upstream force mains has performed satisfactorily to minimize
      sulfide at the discharge points of these force mains.

7.     Data on wastewater sulfide concentrations and sewer atmospheric H2S levels have
      been difficult to correlate at specific locations in the Regional Interceptor
      System.  This is probably due to analytical inaccuracy at low dissolved sulfide
      concentrations (typically less than 0.3 mg/1), slight wastewater pH variations, and
      the degree of wastewater turbulence, all of which  are critical to H2S off-gassing
      rates.

8.     The operating data from the Regional Interceptor System show that careful and
      conservative use of the sulfide and corrosion predictive equations can be of major
      assistance in designing long interceptor systems in warm climates to meet
      stringent corrosion standards.

2.6.3  City of Lakeland,  Florida, Western Trunk Sewer

2.6J.I Description and History

      The City of Lakeland is a growing community of approximately 66,000
population.  Lakeland has experienced corrosion  problems in portions of its sewer
system due to sulfide generation.  An engineering study was undertaken in early 1988 to
assess the existing conditions and  to develop a plan to renovate portions of the  Western
Trunk Sewer.

      The Western Trunk Sewer receives wastewater discharges from food processing,
other industrial, commercial, and  residential areas. The  collection system  consists of
both force mains  and gravity sewers.

      The gravity portion of the Western Trunk  consists of about 27,300 lineal feet
(LF) of primarily reinforced concrete pipe (RCP) and vitrified clay pipe (VCP)  ranging
in size from 24- to 48-inch diameter. There are variable slopes on most reaches which
cause changes in  velocity. Most dry weather velocities are greater than 2 fps, and some
reaches have velocities of 7 fps.

      The Western Trunk Sewer was constructed in 1960 and 1961.   Lakeland  has
rehabilitated or replaced portions of this sewer in recent years because of  pipe collapses.
An odor control study conducted  by the  City of Lakeland in 1987 confirmed high levels
of H2S gas and high wastewater sulfide concentrations.  The City undertook a sewer
system evaluation study to determine the extent of hydrogen sulfide corrosion and
appropriate solutions to correct the  deficiencies.
                                       2-52
-------
2.6.3.2 Sewer System Corrosion Evaluation

      The corrosion evaluation consisted of detailed inspection and analysis of the
conditions of both manholes and truck sewer.  A total of 79 manholes were inspected.
Internal television inspections were conducted on nearly all the gravity portions of the
trunk sewer.  Core borings of concrete pipe were taken at eight representative locations
along the trunk.

      In general, the manholes were found to be corroded and in need of
rehabilitation.  Seventy-seven manholes had corroded barrels, and 66 had deteriorated
frames.

      In the trunk sewer system, all reaches of reinforced concrete pipe (RCP) had lost
from one to four inches of concrete due to corrosion. Reinforcing steel was found
exposed or missing in numerous  locations, and aggregate was exposed in all reaches of
RCP. A section of 30-inch diameter ductile iron pipe, installed in 1966, was severely
blistered and brittle enough to break by hand. While the vitrified clay pipes did not
suffer from corrosion damage, there were numerous cracks and leaking joints in  the
lower portion of that segment; and manhole corrosion was worse in the VCP reach.

      Predictive models were utilized to estimate sulfide generation, corrosion rates,
and remaining useful lifetimes of pipes. The  model predicted sulfide build-up of 1.5 to
2.0 mg/1. Recent field data show levels of 1.0 to 1.5 mg/1.  Approximate correlation was
shown by the model.  Similarly, predicted corrosion rates approximated estimated
corrosion rates based on field measurements.   Predicted average corrosion rates were
0.03 to 0.15 inches/year, with peak rates approximately double  these values.  It was
estimated that, at turbulent structures, corrosion rates could be five times predicted
average rates. This underscores  the importance of turbulence  on  hydrogen sulfide
corrosion, and the difficulty in estimating corrosion rate based on  sulfide levels.

      Based on existing depth of corrosion and estimated corrosion rates, forty-six
percent of the RCP was determined to have no remaining useful life, and  54 percent
was determined to have a useful life of 1 to 8 years.

      Alternatives were analyzed for rehabilitation of the  corroded pipe.  The following
five methods of rehabilitation were considered feasible for the Western Trunk sewer:

1.     Slip lining with high density polyethylene pipe (with  fusion  joints, or bell and
      spigot joints) or with fiberglass pipe (filament  wound, or centrifugally  cast).

2.     Inversion lining with polyester, resin-impregnated fabric.

3.     Removal of existing pipe and replacement with one of the following:
      1) reinforced concrete pipe with PVC liner, 2) fiberglass pipe (centrifugally cast)


                                       2-53
-------
      designed for direct burial.

4.     Parallel replacement of pipe using the same two options listed above.

5.     Chemical grouting for specific locations (limited on this project to grouting
      structurally sound vitrified clay pipe for infiltration control).

2.6.3.3 Findings and Conclusions

1.     Conditions which contribute to the high corrosion rates in the Western Trunk
      Sewer include at least the following:  high wastewater BOD, high soluble BOD
      fractions, high wastewater temperature, low wastewater pH, variable slopes,
      formation of deposits, use of drop structures, and existence of upstream force
      mains.

2.     The Pomeroy-Parkhurst sulfide predictive equation and the Pomeroy corrosion
      prediction equations were used with a series of conservative assumptions for the
      Western Trunk evaluation.  In doing so, and by assuming that historical system
      flows, characteristics and operation were similar to current situations, the
      "modeled" corrosion approximated the  actual corrosion.  This tends to indicate
      the quantity of sulfide and corrosion expected  in similar situations. However, it
      also points out that the equations should be used cautiously and  with substantial
      conservative assumptions and  safety factors.

3.     The velocities in the Western  Trunk system are insufficient, in some reaches, to
      transport the grit and heavier  solids. Additional sulfide production can be
      expected in these reaches due to the ability of high soluble organic and sulfate
      concentrations  to penetrate into these deposits.

4.     The high likelihood of substantial sulfide generation  in the Western Trunk  in the
      future was judged to preclude the use of corrodible materials, or at least
      minimize use of these materials, in the rehabilitation/replacement of the Western
      Trunk Sewer.

2.6.4  Omaha Nebraska, Papillion Creek Wastewater System

2.6.4.1 Description and  History

      Corrosion and  odor problems  have occurred in Omaha's Papillion Creek
Wastewater System over the past decade. The start-up of the expanded Papillion Creek
Interceptor System in the mid-1970's brought  new dischargers into the system, and
substantially increased the transit time of wastewater  to reach treatment facilities.  The
new Papillion Creek Wastewater Treatment Plant has experienced corrosion and odor
problems which are partly related to  interceptor sulfide problems. Safety has been an


                                       2-54
-------
additional problem due to high concentrations of hydrogen sulfide in the confined
spaces of interceptors and treatment facilities.

       Omaha's Papillion Creek Wastewater System has evolved over several decades of
growth and urban expansion.  It now includes a service area bringing wastewater flows
over 25 miles to the Papillion Creek Wastewater Treatment Planr (Papio Plant).

       Only the largest Papillion Creek interceptors were evaluated in the 1984/85
corrosion and sulfide study because these were the interceptors suspected of corrosion
damage. The interceptor downstream from the old Papio Plant was a primary target of
the study. These interceptors were put into service in the mid-1970's to transport raw
wastewater to the new Papio Plant This system of interceptors allowed several
treatment plants in the Papillion Creek drainage area to be abandoned.

       In 1971, a limited study was completed on potential sulfide and corrosion
problems in this interceptor system, which was then being designed. This report  made
predictions of sulfide levels in  the interceptors and confirmed the need for lining the
sewer with either sacrificial concrete or plastic.

       Between 1973 and the start-up of the new Papio Plant  in August,  1977,
communities and industries were allowed to discharge treated effluent and, in some
cases, raw wastewater to the new Papio Interceptor System. Many odor complaints
from residents living near manholes and structures were received during  this period.
The worst odor conditions occurred in the last five miles of the system, where high
BOD wastes and flows of only a few mgd probably caused very high sulfide production
in at least 1974 and 1975.

       Significant corrosion of the unlined outfall portion (last 3,200 feet to the
Missouri River) was noted in 1977. This corrosion was evidently a result of these raw
and partially treated flows.  The unlined outfall portion of the interceptor was estimated
to have 1/2-inch  of corrosion in August 1977. Extensive corrosion of unlined manhole
risers in  the lower interceptor  reaches also occurred prior to 1977, to the extent that
some risers were replaced during the mid-1970's.

       Odor problems became a major issue in Sarpy County, and the City of Omaha
decided to seal all manhole covers from the old Papio Plant south to the new Papio
Plant in  1975.  Little ventilation of the system can occur, and  only one siphon has
ventilation stacks.  The West Branch has little ventilation since manhole  covers are
solid.

       Data collected at the Papio Plant show influent total sulfide levels of 4 to  5 mg/1
from July through February, dropping to between 1 and 3 mg/1 during spring months.
                                       2-55
-------
2.6.4.2 Summary of Results

      Inspection of the interceptor was conducted in 1984. Manhole risers were badly
corroded, in some cases, to over 1 inch. The risers were constructed of concrete with
non-calcareous aggregate. Measured pipe corrosion in calcareous aggregate reaches
ranged from zero to 0.5 inches.  At points of high turbulence  such as discharge
structures, up to 0.75 inches of calcareous/granitic aggregate concrete had been lost over
a 10 year period.

      Use of predictive models was unsuccessful in that actual sulfide concentrations at
the Papio plant were double those predicted by the model.  The reasons are judged to
be 1)  high fraction of soluble BOD, 2) low oxygen content in the sewer atmosphere due
to sealing of manhole lids, 3) need for more conservative coefficients in the predictive
equation.

      The Papio plant, commissioned  in 1977, was also inspected for corrosion in 1984.
Corrosion at the headworks was significant, with up to 0.75 inches of concrete lost over
a 6-1/2 year period. Several locations within  the plant had little or no concrete cover
remaining.  Reinforcing steel was exposed in the sludge storage tank, decant tank, drain
manholes, and trickling filter walls. Hydrogen sulfide corrosion problems  were caused by
three  basic factors:

      1.     High sulfide concentrations and low pH of influent wastewater

      2.     Long storage times for sludge and recycle streams

      3.     Recycle of streams from anaerobic digestion process

2.6.4.3 Findings and  Conclusions

1.    The large Papillion Creek interceptors were designed with attention to sulfide
      and corrosion issues, and based on 1984 inspection work, most of the interceptor
      system was in good condition.  Use of sacrificial concrete, coatings, PVC linings,
      and calcareous aggregate prevented serious corrosion problems in  the interceptor
      system.

2.    There have been a few specific corrosion problems in the interceptor system.
      These include metal gates and  aluminum grating which was subject to destructive
      acid attack.  H2S off-gassing from highly turbulent wastewater in certain
      structures has caused corrosion rates that are largely unpredictable.  Unprotected
      manholes have been subject to extensive corrosion.

3.    If structural rehabilitation work is required on any of the primary interceptor
      structures, costs could be extensive for diversion and rerouting of raw wastewater


                                        2-56
-------
      flows.

4.    Assessment of concrete corrosion rates over time is difficult without historical
      sulfide data and/or historical inspection information identifying the depth of
      corrosion from the original surface.

5.    Actual sulfide buildup in the interceptor system seems to be at least twice that
      predicted by the Pomeroy-Parkhurst equation.

6.    Actual concrete corrosion rates are greater than predicted in the design for
      portions of the interceptor system due to:  1)  greater sulfide production than
      planned, and 2) reduction in pH of wastewater as it travels down the long
      anaerobic interceptor system.

7.    Concrete, metal, and instrumentation corrosion in the Papio Plant occurred at
      many locations over the period of 1977 to the mid-1980's.  The corrosion was
      caused by high wastewater sulfide levels, and  high H2S off-gassing rates.   High
      dissolved sulfide levels in the wastewater were partially caused by high influent
      sulfide concentrations, and partially by sulfide produced within the  plant

8.    Solutions to plant  hydrogen  sulfide corrosion  problems are now mostly in-place.
      These solutions encompassed rehabilitation, process changes, chemical addition,
      and ventilation improvements.  The solutions implemented thus far appear to be
      working satisfactorily.

9.    There remains an  odor problem at the Papio  Plant caused in large  part by the
      high influent sulfide levels.  High sulfide production in the interceptor system
      and high plant influent sulfide concentrations constitute the  most significant
      unresolved sulfide issues in the Papillion Creek Wastewater  System.

2.6.5  Omaha, Nebraska, South  Interceptor Sewer

2.6.5.1 Description and History

      In the late 1950's,  planning was initiated for collection, diversion, and treatment
of raw waste discharges to the Missouri River.  By 1965, the system of diversion
structures, interceptors, pumping stations, and primary treatment facilities had been
constructed and placed in operation.  The system involves a series of structures along
the west bank  of the Missouri River to intercept flows and pump them to  the Missouri
River Wastewater Treatment Plant (MRWTP).  The area served, by the MRWTP is the
older and more highly developed portion of Omaha.  It contains Omaha's central
business district and  industrial centers which  are located adjacent to, or near, the
Missouri  River.
                                       2-57
-------
      The South Interceptor Sewer (SIS) is a 4-1/2 mile long force main that brings the
majority of the plant flows to the MRWTP. Flow from the SIS is discharged at the
North Inlet Dry weather flow in the winter is about 20 million gallons per day (mgd)
from the SIS. This flow increases in the warmer months due to infiltration, runoff from
lawn watering, and discharges to the sewer system from drawdown of Carter Lake.

      Velocities in the 66-inch diameter South Interceptor Sewer are about 0.7 foot per
second (fps) or less at night and typically average 1.2 to 1.6 fps in dry weather.
Deposition of solids is no doubt occurring under these conditions.  Dissolved sulfide
levels in the plant influent are typically 3 to 6 mg/1 during warm months.

      Additions to the MRWTP completed in 1980 included covering open tankage at
the North Inlet, as well as other locations throughout the plant  Fans were installed to
exhaust foul air from under most of these covered areas and scrub it in chemical mist
units prior to discharge to the atmosphere. The City has not found the chemical
scrubber at the North Inlet to be effective, and does not use it Hypochlorites and
permanganate solutions were attempted in the scrubber with little success.  The covers
have remained in place, but the ventilation system is not used, since  very high  H2S
concentrations would be discharged to the atmosphere. Corrosion of the concrete is
occurring faster since installation of the covers, due to high levels of hydrogen sulfide  in
the atmosphere of the tanks and channels of the North Inlet  Gaseous hydrogen  sulfide
levels under the North Inlet covers have reached the 50 to 300 ppm  range during
summer and fall months.

      In  1984, a  study was initiated to evaluate odor and corrosion problems in the
South Interceptor Sewer and  at the Missouri River  Wastewater Treatment Plant  The
results are summarized below.

2.6.5.2 Summary of Results

      Hydrogen sulfide corrosion has occurred at a number of locations  in the
MRWTP, but the North Inlet is the location where the problem is most apparent at this
tune.  Besides the North  Inlet, concrete corrosion has  occurred at the South Inlet and in
various wastewater channels and boxes prior to, and following, primary clarification.
Corrosion in foul air ducts from the biological filters has also occurred, as well as
corrosion in the anaerobic digestion portion of the plant

      Inspection work in 1984  showed that concrete spalling was occurring above the
water line in the Parshall flume area and in other channels. The worst corrosion was  at
the drop structure where at least 1/2- to 3/4-inch of concrete was estimated to  have
become corroded from walls and structural support members.  Most  of this corrosion
probably occurred in the 1980 to 1984 period after the tanks were covered. No
reinforcing bars were exposed.  Metal corrosion at gates was also significant  In the bar
screen building, continued corrosive environmental conditions had deteriorated many


                                       2-58
-------
metal components, and the exposed electrical and instrumentation equipment had
suffered irreparable damage.

      The recommended approach to sulfide control at North Inlet was as follows:

1.     Reduce the high sulfide and H2S concentrations  through sulfide control methods
      in the South Interceptor Sewer.

2.     Leave the covers in place at the North Inlet and implement improved ventilation
      and odor scrubbing to reduce corrosive atmospheres and treat residual H2S and
      other odorants more reliably.

3.     Once improved ventilation and HjS control is in place, conduct structural
      rehabilitation and  replace equipment as necessary.

      Alternatives evaluated for the SIS included chlorination, iron chloride addition
upstream from the plant, and caustic slugging. Costs favored caustic slugging,  although
the performance was not expected to be as consistent as other alternatives.  Full-scale
testing of caustic slugging was undertaken  in late 1984 to confirm its effectiveness and
define costs more accurately.

      The objectives of a caustic injection program is to inactivate  sulfide-producing
bacteria which grow in the slime layer on the walls of the pipe.  High pH  (12 to 12.5
and above) is toxic to these bacteria, and interim application of a strong alkaline
solution has proven effective in depressing sulfide production within sewers. The test
program was successful, and the  technique has been used on the SIS each summer since
that time.  Caustic slugging is effective for reducing dissolved sulfide levels in the plant
influent from over 3 mg/1 to an average of 0.4 mg/1. In 1985 the City spent $44,000 on
17 truckloads of caustic, estimated to eliminate 47,000 pounds of sulfide.  In the
summer of 1988, the effectiveness of caustic slugging was reduced.  This was believed to
be due to the dry weather conditions which led to higher wastewater strengths,  and
lower velocities and greater solids deposition in the SIS. Evidently, the  caustic  slug is
unable to penetrate sludge deposits and inactivate the bacteria.

      The City is currently evaluating alternatives to caustic slugging for control  of
sulfide generation in the  South Interceptor Sewer.

2.6.53 Findings and Conclusions

1.     Significant corrosion and odor problems have occurred at the North Inlet,  due to
      sulfide production within the SIS. Attempts to control the odor problem in 1980
      by covering tankage and scrubbing the foul air resulted in worsening of the
      corrosion problem due to failure of the odor scrubbing system, and subsequent
      shutdown of the ventilation  system.


                                       2-59
-------
2.     Rapid rates of corrosion can be expected for concrete and metal exposed to the
      high H2S concentrations (50-300 ppm) and moist environment which exist
      beneath the covers at the North Inlet The life of some of these facilities appears
      to be 10 years or less under the severe conditions which have existed in the
      1980's. The high H2S concentrations constitute a safety hazard.

3.     Caustic slugging has provided a cost-effective reduction in high sulfide
      concentrations from  the SIS since late 1984 when it was initiated.  Its
      performance is variable, primarily because of the solids deposition problem in the
      SIS, caused by low wastewater velocities.  Atmospheric H2S levels beneath the
      primary clarifier domes have dropped from over 50 ppm to below 10 ppm.

4.     Other sulfide control methods are likely to be more reliable and have better
      overall performance  than caustic slugging; however, the costs for other control
      methods appear to be substantially higher than caustic slugging.

5.     The low velocity and resulting deposition problem in the SIS is likely to continue
      to plague  attempts at reliable sulfide control in this pipeline. Velocities of about
      4 to 5 fps are needed on a  regular basis to scour grit deposits from a force main
      of this size. The periodic scouring of the SIS causes considerable solids loading
      fluctuations to the MRWTP which may affect treatment performance.

6.     The City,  with the use of caustic slugging, has implemented the first phase  of a
      program  to solve the sulfide problems at  the North Inlet  Substantially more
      work is needed to restore the North Inlet to a  sound, safe, and acceptable
      condition.

2.7   Hydrogen Sulfide Corrosion in  Other Countries

      Hydrogen sulfide corrosion has been reported in the literature of many countries
including France, Germany, Italy, United Kingdom, the Netherlands, Denmark,
Czechoslovakia,  Iraq, India, China, the Soviet Union, Japan, Saudi Arabia, Kuwait,
Egypt, South Africa, Venezuela, Brazil and Australia.  This observation is based on a
series of literature searches conducted for EPA  in 1982 and  1988 on the subjects of
odor and corrosion in wastewater systems.

      Several severe cases of hydrogen sulfide corrosion are briefly summarized below
1.    Venezuela:  Reinforced concrete pipe was corroded to a depth of 2.8 inches
      within eighteen months of construction in an area downstream from the
      discharge of a force main.  Vitrified clay pipes used in the same system were not
      affected.
                                       2-60
-------
2.    Cairo, Egypt: As early as 1920, hydrogen sulfide corrosion of sewer pipes was
      recognized as a problem in the Cairo system.  By 1922, in spite of aeration and
      regular flushing, the crown of the original main outfall sewer, 60 inches in
      diameter and made of local cement concrete, was corroded to a depth of 3.9
      inches over a length of 8 miles.  By 1930, the depth of corrosion had increased to
      5.9 inches, nearly half the thickness of the pipe.

3.    Baghdad, Iraq: A system of reinforced concrete interceptors  was put in
      operation in 1963, and by 1977 the maximum depth of corrosion in extended
      sections of the interceptors varied from three to  four inches at the crown where
      reinforcing steel, which was designed  to be protected by a two-inch concrete
      cover, was exposed and had corroded away in places.  The walls were also
      damaged to a depth of two inches and the access manholes showed severe
      corrosion.  The rate of internal corrosion was estimated at 0.3 inches per year,
      which would have led to a critical situation in  approximately  15 years.
      Rehabilitation had begun in certain sections of the system, and new extensions
      have been made with PVC or fiberglass-reinforced plastic lining.

4.    Singapore:  The sewer of concern was 71 inch diameter concrete pipe internally
      lined with 0.5 inches of high alumina cement mortar.  The line was over 1600 ft
      long with a design capacity of 81 mgd. The sewer was commissioned in 1961.
      This portion of the line received almost entirely  pumped sewage, much of which
      was pumped more than once.

      Inspections since  1970 revealed extensive and continuing hydrogen sulfide
      corrosion.  The high alumina lining was completely corroded  away or reduced to
      a soft past above  the waterline.  Corrosion in some areas had proceeded to a
      depth of over 1.5 inches, and had gone beyond the reinforcing steel,  in less than
      15 years.

      Several rehabilitation options were under consideration  including sliplining with
      glass-reinforced polyester or high-density polyethylene, installation of corrosion-
      resistant panel liners, and cured-in-place inversion liners.

2.8   Conclusions

      Attempts to gain  a thorough understanding of the severity and extent of
hydrogen sulfide corrosion problems in U.S. were thwarted by the lack of historical data
on sewer corrosion, the lack of a standardized technique to measure corrosion, and the
poor documentation by municipalities of sewer corrosion and the expenditures for sewer
rehabilitation or replacement Upon review of information gathered, the following
findings and conclusions are presented:

      •      Severe hydrogen sulfide corrosion problems are not limited to CSDLAC.


                                       2-61
-------
            Extensive corrosion damage requiring immediate repair or rehabilitation
            has been observed in sewers in other cities.  In some cases, corrosion
            damage is so extensive as to compromise the structural integrity of the
            pipe, which could lead to collapse.

      •     Hydrogen sulfide corrosion problems in operating systems are often not
            recognized early enough to take corrective action before considerable
            damage has occurred.

      •     In a 1984 survey, approximately 30 percent of the 89 cities reported sewer
            collapses that were judged to be due to hydrogen sulfide corrosion.

      •     In two independent surveys, 60 to 70 percent of the municipalities
            reported hydrogen sulfide corrosion at their wastewater treatment plants.
            Of those plants experiencing corrosion, about 20 percent  are considered to
            have severe problems.

      •     Hydrogen sulfide corrosion problems have been documented in the
            literature of at least 20 foreign countries.

      •     Due to lack of historical data, corrosion rate is estimated based on depth
            of corrosion and age of pipe.  This may not reflect the true corrosion rate,
            which may be substantially higher at a given time and condition.

      •     No entities other than CSDLAC  had sufficient data on corrosion rate to
            establish whether the rate of corrosion had changed over time.

      •     Due to changing  alkalinity in spun vs. cast concrete pipes, corrosion rate
            can change over time.

      Evidence of severe corrosion  may be found in cities throughout  the United States
and other countries.  Cases of "high rate" corrosion are also common.  However, at this
time EPA has  been unable to document other cases of "accelerated" corrosion of the
type that has been experienced in the sewers of CSDLAC.
                                       2-62
-------
                                 REFERENCES

1.     Jin, Calvin, "Sulfide Control with Sodium Hydroxide in Large Diameter Sewers,"
      internal report, CSDLAC, March  1987.

2.     Witzgall, R.A., Homer, I.S., and P.L. Schafer, "Sulfide Corrosion in Collection
      and Treatment Systems - Case Histories and Evaluation of Predictive Equations,"
      prepared for the 62nd Annual WPCF Conference, San Francisco, CA,  October,
      19, 1989.

3.     Witzgall, R.A., Homer, I.S., and P.L, Schafer, "Case Histories of Sulfide
      Corrosion:  Problems and Treatment,"  Water Environment and Technology,
      Water Pollution Control Federation, 4247, July, 1990.

4.     EPA, "Odor and Corrosion Control in Sanitary Sewerage Systems and Treatment
      Plants,"  EPA/625/1-85/018, Cincinnati,  OH, October, 1985.

5.     Prevost, R.C., "Corrosion Protection of Pipelines Conveying Water and
      Wastewater - Guidelines," World Bank Technical Paper No.  69, Water Supply
      Operations Management Series, The World Bank, Washington,  D.C.,  1987.

6.     Nadarajah, A., and J. Richardson, "Prevention and Protection of Sewerage
      Systems Against Sulphide Attack with  Reference to Experience in Singapore,"
      Prog. Wat Tech. Vol 9: 585-598, Pergamon Press, Great Britain, 1977.
                                      2-63
-------
3.0   EFFECTS OF INDUSTRIAL PRETREATMENT

3.1   Overview

      Section 522 of the Water Quality Act of 1987 requires EPA to study the extent
to which implementation of the categorical pretreatment standards will affect hydrogen
sulfide corrosion in wastewater collection and treatment systems in the United States
(U.S.).  EPA's industrial pretreatment program regulates the discharge of certain
constituents such as metals and toxic materials into municipal sewer systems. The
County  Sanitation Districts of Los Angeles County (CSDLAC)  implemented industrial
pretreatment  standards in 1975-1977 to meet ocean discharge requirements, and
additional controls starting in  1983 to comply with the EPA-mandated industrial
pretreatment  program.

      Metals and other constituent levels in CSDLAC wastewater dropped substantially
between the early 1970's and the mid 1980's as a result of implementation  of these
industrial pretreatment programs.  A concomitant rise in both total and dissolved sulfide
levels in the wastewater occurred over this same time period. Further, CSDLAC
observed an increase in the rate of corrosion in their concrete sewers.

      Two hypotheses have been  set forth to explain increased corrosion rates in
CSDLAC sewers due to the reduction in levels of metals. The first is that at the higher
levels of metals, a significant amount of sulfide was rendered insoluble in metal-sulfide
compounds, reducing the amount of dissolved H:S available for release to  the sewer
atmosphere.  The second is that the higher levels of various metals and other
compounds in the wastewater had a toxic effect on the sulfate-reducing bacteria
responsible for the generation of sulfide. When the metal concentrations were
significantly reduced, the sulfate-reducing bacteria flourished, increasing sulfide levels in
the wastewater, which generated more dissolved H2S.  Both  phenomenon would increase
the amount of H2S available for release  to the  sewer atmosphere, and subsequent
corrosion of the sewer crown due to increased  sulfuric acid production.

      A thorough search of the literature and contacts with municipalities throughout
the U.S. revealed that no data existed from other cities to show a correlation between
implementation of industrial pretreatment standards and increased sulfide generation
and corrosion. Municipalities simply do not  have historical data on corrosion rates or
sulfide levels  that would allow establishing a correlation such as was  found in CSDLAC.

      Given  the unavailability of full-scale data to support the  theory proposed by
CSDLAC, the study objectives were determined to be as follows:

      1.     Investigate the theoretical  impacts of metals on sulfide  levels.

      2.     Review and analyze research conducted or supported by CSDLAC.


                                       3-1
-------
      3.     Compare metals levels of CSDLAC with other cities to assess whether
             other municipal sewerage systems could potentially experience a similar
             phenomenon (decrease in industrial wastewater constituents and increase
             in corrosion).

      4.     Review data from site visits to industrial cities to determine if corrosion
             rates differed in sewers with high industrial contributions vs. those with
             predominately residential contributions.

      5.     Review other potential impacts of implementing pretreatment standards.

3.2   Theoretical Impacts of Sulfide Reaction with Metals

      It is well known that many metals "bind" with sulfide to produce a precipitate
which is insoluble, effectively preventing release of hydrogen  sulfide gas to the sewer
atmosphere and preventing formation of corrosive sulfuric acid.  As discussed in Section
4,  salts of metals such as iron and zinc are routinely added to wastewater to prevent
odors and corrosion associated with hydrogen sulfide.

      The weight of metal required to precipitate a given weight of sulfide can be
predicted theoretically using chemical reaction equations.  Table 3-1 shows the probable
precipitation  reactions of metals with sulfide in wastewater devoid of oxygen.  Based on
the stoichiometry of the reactions, the necessary concentration of each metal required to
precipitate 1  mg/1 of sulfide has been calculated.  The last column shows the inverse of
this value, or the  theoretical concentration of sulfide that would be precipitated by 1
mg/1 of metal. However, in wastewater containing a complex mix of organic and
inorganic compounds  which interfere with such reactions, the amount of metal required
to precipitate a given  weight of sulfide may be much greater than what would be
predicted from the equations.

      Table  3-2 shows the theoretical stoichiometric increase in dissolved sulfide
concentration based on the reduction in metals concentration experienced in CSDLAC
between the periods 1971-1974 and 1983-1986. The total theoretical increase in
dissolved sulfide due to reduced availability of metals to precipitate the sulfide is
approximately 4 mg/1. The measured increase in dissolved  sulfide during that same
period was approximately 1 mg/1.

      Reduction in iron alone accounts for 69 percent of the theoretical increase in
sulfide.  Zinc accounts for 16 percent, and chromium 10 percent.  The reduction in
these three metals accounts for 95% percent of the theoretical increase in dissolved
sulfide levels.

      Because of its toxicity, chromium is not used for sulfide control. However, data
are available  on dosage requirements for iron and zinc to precipitate sulfide that are


                                        3-2
-------
                   TABLE 3-1

PROBABLE METAL - SULFIDE PRECIPITATION REACTIONS
        IN WASTEWATER DEVOID OF OXYGEN
Reactions
Fe*2 + S-2
Zn*2 + S-2
Nr2 + s-2
Cd*2 + S'2
Pb*2 + S'2
Cu"1"1 + S'2
Cr*2 + S'2
	 > FeS
— > ZnS
	 > NiS
— > CdS
— > PbS
•> Pit s
— > CrS
Theoretical mg/1
of Metal to
Precipitate
1 mg/1 of Sulfide
1.74
2.04
1.83
3.51
6.48
3.97
1.63
Theoretical mg/1
of Sulfide Precipitated
bv 1 mg/1 of Metal
0.57
0.49
0.55
0.28
0.15
0.25
0.61
                        3-3
-------
                                TABLE 3-2

             THEORETICAL INCREASE IN DISSOLVED SULFTDE
              BASED ON METAL PRECIPITATION; LA COUNTY
Metal

Chromium
Copper
Lead
Zinc
Nickel
Iron
Cadmium
TOTAL
Reduction
in Metals1
mg/1
0.68
0.38
0.17
1.34
0.14
4.92
0.01

Theoretical
Increase in Dissolved
Sulfide Concentration2
mg/1
0.42
0.10
0.03
0.66
0.08
2.83
0.00
4.12
Expected
Increase Based
on Field Studies3
mg/1
~
—
—
0.06 - 0.1
-
0.1 - 0.7
~

1  Difference in average values for the periods 1971 - 1974 and 1983 - 1986.

2  Based on stoichiometry of chemical precipation reactions

3  Based on field dosages  required to precipitate dissolved sulfide;
  LA County research data.
                                    3-4
-------
useful to estimate the actual increase in sulfide levels that might be expected by
reduction in metals.

      Studies by CSDLAC in 1985-1988 on the use of iron addition to control sulfide
showed that when the dissolved sulfide levels were between 1 and 4 mg/1, a dosage ratio
of six to seven parts iron to 1 part dissolved sulfide was required to achieve 90 percent
removal. When the dissolved sulfide was less than 1 mg/1,  a dosage ratio of 44 to 1 was
required (1). The theoretical dosage ratio is 1.7 to 1.  Thus, four to  25 times the
theoretical  dosage was required for iron.  Other studies conducted in 1971-1972 in
CSDLAC have showed that five to seven times the theoretical dosage is  required to
remove sulfide using zinc.

      Based on this analysis, it is possible that the reduction in metals experienced by
CSDLAC could account for some portion of the observed increase in dissolved sulfide,
considering that iron and zinc can account for between 0.2 to 0.8 mg/1 of the increase in
dissolved sulfide. However, it is unlikely that precipitation could  account for all of the
measured increase in dissolved sulfide (over 1 mg/1) in CSDLAC  wastewater.

      During the period 1971 to 1986, total sulfide  increased from 0.4 mg/1 to 3.0 mg/1,
and dissolved sulfide from 0.1 mg/1 to 1.4 mg/1.  If sulfide precipitation with metals was
the only mechanism, the fraction of dissolved sulfide would increase, but the total
sulfide level would remain essentially constant as metals were reduced.  This is because
the insoluble metal-sulfide precipitates are still detected in the total sulfide test
However, research by Pomeroy found that when iron was added to wastewater
containing  sulfide, a reduction in total sulfide was observed. Two explanations were
suggested.  The  first is that one of the products of the  reaction is iron disulfide, which
when treated with acid in the sulfide test, forms H2S and elemental sulfur.  Elemental
sulfur is not measured in the  test The second is that the iron may act as a catalyst to
oxidize sulfide to a product which is not detected in the sulfide test (3).  Thus,  it is
unlikely that chemical precipitation or the presence or absence of iron could account for
the increase in total sulfide between 1971 and 1986.

3.3   Biological Inhibition by Metals and Toxic Compounds

      CSDLAC has conducted in-house experiments to investigate inhibition of
microbial sulfide generation by constituents present  in  wastewater. In addition,
CSDLAC is partially funding  research at the University of Arizona to investigate the
toxic effects of metals on the sulfide-oxidizing  bacteria Thiobacilli. which are responsible
for the production of sulfuric  acid on the sewer crown.  Other research at the University
of California at Los Angeles funded by CSDLAC has considered  the effects of metals
concentrations on both sulfate-reducing bacteria and sulfide-oxidizing bacteria.

      The  first in-house  experiments conducted by CSDLAC were bench-scale
laboratory studies designed to determine the acute toxicity  levels of selected metals and


                                        3-5
-------
cyanide on sulfate-reducing bacteria.  Growth medium for sulfate reducers was dosed
with varying concentrations of metals and cyanide.  Tubes were inoculated with primary
effluent containing the sulfate-reducing bacteria and incubated for three weeks.  The
tests involved dosing with individual metals and cyanide as well as with a stock solution
containing nickel, chromium, zinc, copper, lead, cadmium, and cyanide in  ratios
approximating that in  sewage to determine if a synergistic effect existed.

      Table 3-3 shows the results of the experiment with the individual metals and
cyanide.  Copper was the most toxic at 6 mg/1, while cyanide was the least toxic at 50-55
mg/l(4).

      The stock solutions containing the mixtures of metals and cyanide were added to
the tubes at various dilutions to simulate total constituent concentrations of 1, 5, 10, 15,
20, and 25 mg/1. The  distribution of constituents was as follows:  nickel - 6.2%,
chromium - 20.4%, zinc  • 44.3%, copper - 13.3%, lead - 6.6%, cadmium -  0.7%,  and
cyanide - 8.5%.  The results of this experiment showed that growth  was completely
inhibited at a total constituent concentration of 10 mgA, but was not visibly affected at a
concentration of 1 mg/1.  At 5 mg/1, growth was notably retarded. A synergistic effect
apparently existed when  the combination of metals and cyanide were added to the
growth medium.

      CSDLAC performed a second experiment using column tests to determine on a
larger scale how changes in wastewater metal concentrations affect  the generation of
sulfide.  The testing was carried out at the Joint Water Pollution Control  Plant
(JWPCP)  in Carson, California.

      The test apparatus consisted of three 8-inch diameter, 6-feet high, polyvinyl
chloride (PVC) pipes that functioned as test columns. The characteristics of the
columns are shown in Figure 3-1. Each column was filled with 4 feet of hand-cut
polystyrene cubes measuring 1.5 inches on each side.  This provided a substrate  for the
attached growth of sulfide-generating bacteria.  Primary effluent from the  JWPCP was
pumped through each column at a rate of 0.75 liters per minute at  a temperature of
80°F. One column was used as a control while  solutions with known metal
concentrations were added to the influent of the other two columns. Dissolved sulfide
and total sulfide levels were measured at the influent and effluent ends of each column.
After each solution was  tested, all the columns were cleaned, the polystyrene was
replaced, and the system was flushed.  As of April 1, 1989, seven different metal
solutions had been tested.

      The first of the seven trials was performed with a cocktail of chromium, copper,
zinc, and cyanide.  The feed rate of the cocktail was controlled so that the
concentrations  of metals in the wastewater approximated  those of the early 1970's.  The
second trial was performed with five times the concentration of metals and cyanide in
the wastewater. In the remaining trials, chromium, copper, nickel,  and  cyanide were


                                       3-6
-------
                             TABLE 3-3






TOXICITY OF WASTEWATER CONSTITUENTS ON SULFATE-REDUCING BACTERIA1
Compound
Copper
Nickel
Chromium
Lead
Zinc
Cyanide
Toxic Cone, mg/1
6
13
23
25
25
50-55
    In-house experiment conducted by CSDLAC
                                3-7
-------
 I

CO
LJL
   21
CO
O T3


'"§!


8|
CD =£
    ®D
   03
                              3-8
-------
each tested separately at wastewater concentrations approximating those of the
early 1970's.  The target concentrations for the trials are listed in Table 3-4.

      Results of the column experiments are summarized in Table 3-5 and 3-6.
Examination  of the data yields several pertinent conclusions.  Clearly, when the
"cocktail" of metals and cyanide were added at five times the 1971-1974 concentration,
sulfide generation was  significantly inhibited.  These results are supported by bench
scale studies at the  CSDLAC laboratory which showed that inhibition of sulfide-
reducing bacteria requires metals concentrations much higher than that observed in
wastewater.  However,  the concentration of metals causing sulfide generation inhibition
in these pilot scale studies cannot be used to predict the increased levels of total and
dissolved sulfide observed by CSDLAC in  their sewers.

      Of greater significance is the comparison of performance when  the "cocktail"
containing metals and cyanide was fed to the columns at the 1971-1974 levels.  Total
sulfide concentrations in the effluent from the control column and test column were
16.6 mg/1 and 12.4 mg/1, respectively. Dissolved sulfide  levels were 13.5 mg/1 and 9.7
mg/1 for the control and test columns, respectively. A statistical analysis was conducted
of the data from the experiment in which metals and cyanide were added at the 1971-
1974 levels, assuming identical influent metals concentrations  in both control and test
columns. The analysis indicated that at a 0.01 level of significance  (highly significant),
the total and  dissolved  sulfide concentrations in the effluent from the control columns
were higher than the levels in  the effluent from the test column.  Further, it was
determined that, at the 90% confidence level, the  actual difference in effluent total
sulfide concentrations was between 2.3  and 5.6 mg/1.  For effluent dissolved sulfide, the
actual difference was between  2.7 and 5.1 mg/1.

      Based  on the difference in sulfide levels in  the effluent and influent for both the
control column and the test column, CSDLAC staff prepared  Figure 3-2 showing the
percent change in sulfide generation upon addition of metals and cyanide.  At the Ix
levels (corresponding to the early 1970's), generation of total and dissolved sulfide was
reduced by 34 percent   At the 5x levels, generation of total and dissolved sulfide was
reduced by 114 and 106 percent, respectively.   Interestingly, addition of zinc alone at the
1970's levels resulted in reduction of sulfide generation  by approximately one third.

      Another pilot study conducted by CSDLAC involved the construction of two 3/4-
inch diameter, PVC "force mains," each 150 feet in length.  One served as a control
pipeline, receiving primary effluent, and the other served  as a test pipeline receiving a
"cocktail" of metals and cyanide similar to  that used in the column experiment One
difference in  the feed material to the pipeline pilot system was that iron was also
included in the metals  cocktail at a level approximately  that of the early 1970's (10.7
mg/1).  Only the  Ix  levels of metals and cyanide were used during the pipeline
experiments.  Figure 3-3 is a diagram of the pipeline pilot plant system.
                                        3-9
-------
                                      TABLE 3-4

          CONCENTRATION OF AGENTS ADDED TO UPFLOW PACKED COLUMNS (5)
                   1971-74           1987               Ix               5x
Agent            Concentration      Concentration      Concentration      Concentration
^•B^^PMM^B*
Chromium
Copper
Cyanide
Nickel
Zinc
mg/1
0.92
0.60
0.32
0.285
2.17
mg/1
0.178
0.18
0.02
0.087
0.60
mg/1
1.00
0.50
0.40
0.25
2.00
mg/1
5.00
2.50
2.00
2.25
10.00
                                         3-10
-------

















1

Ed
^
ea
•<
H


























a
z
C/5
a
Q S
C/J W5
-i
w S
a s
z^
u] O
eu u
M Q
a u
o £
u *•

•< o

S £
r T "^
— • • •*
ta Q
-! E
3 -J
C/3 ^

5 C—
fid Z
Zfe
•• b*
Cd U
O
u
^





^ p
V p^
3
s
UJ K/3
: Q
C/3
Z
D
800
C^J
& S
UJ *• C
3
(S
• ri

Q






!
c J2
« p"
3
S
S • c
S ' u
u
o
0

-J
O v)

sj i

C
is

en
Q


 2
£ 7


E »

*S> m
E K

ha
3
.a
X
,_, Tf IO *-l
ts P* PJ jb

W^ ON oj 00
o en Q^ en



O ^« CO i
t- vo CO £
(S vo O O

p vo ON oo

00 P^ ON i-l
ca «n -^ vb






^O OO OO ^^
^^ ^^ ^^\ ^^

en O P* i^
« vri p4 in






-« O 00 P-
p~ oo >n p~
O «-" ON ON
(S OO CO CO
^f en oi *o


rs P q ^
06 jj o 06

c^ ^^ *o *o
\o oo r** vo
>^
^ C
u, C
3 D 3
•B 3 1 -5
5 c o Ji
** * b o
X_^^ M? *"*
r \ f ^ *"y
>o U U Z
en P^ >n
vb ^ en

•* p ^



ON v-> co
^ en en
en P* P~
T- ' O en
T)

ON oq p
vb o ^f

~* f^ O
*^^ f^ ^4






^. \o ^
^ °° en
So »-
•V CO


•^ p co
oo P~ vb

m c^i ts

^
E
3
SL 'g
1 = 1
8 N S
-------
                         TABLE 3-6

COMPARISON OF CONTROL AND TEST COLUMNS' SULFIDE GENERATION;
                 UPFLOW PACKED COLUMNS (5)
EFFLUENT MINUS INFLUENT SULFIDE
Control Test
Additive DS TS DS TS

Ix Mixture
5x Mixture
Cyanide
Chromium (VI)
Nickel
Copper
Zinc
Chromium (III)
mg/1
6.2
5.1
6.1
5.2
8.6
9.6
8.9
5.8
mg/1
7.1
6.4
6.8
5.8
9.0
9.0
8.8
7.3
mg/1
4.1
-0.3
8.2
5.0
7.7
8.8
6.1
7.9
mg/1
4.7
-0.9
9.8
5.6
7.3
9.4
5.9
8.5
PERCENT CHANGE
Control vs. Test
DS TS
%
-34.0
-106.0
34.0
-4.0
-10.0
2.0
-32.0
36.0
%
-34.0
-114.0
44.0
-3.0
-19.0
4.0
-33.0
16.0
                            3-12
-------
   o>
   3
   "5
      o
     cu
co 73
3   ^^
so"§
"- « •£
   2  8
   o  a
   o

   
-------
     c
     cc
CO  .2
 1   £  §
     C  a
     £%•
     CO  2
CO

 CD
0)
LJL



u.
CD
C
CD
CD
CD
T3
_u
"S
^M
a
Q_
Wb»


                                                             CO
                                                           II
                                                           U_ LU
                                                             UJ
                                                             CO
                                   3-14
-------
      Tables 3-7 and 3-8 provide the results of the experiment At the Ix metals and
cyanide dosage without supplemental iron, the generation of total and dissolved sulfide
was reduced by 36 to 25 percent, respectively.  When iron was added to simulate
concentrations in the early 1970's, total dissolved sulfide generation was reduced by 51
and 77 percent, respectively. These  results are depicted graphically in Figure 3-4.

      The results of these experiments strongly suggest that the generation of hydrogen
sulfide in the wastewater of CSDLAC was suppressed due to the presence of
constituents associated with industrial discharges of the early 1970's.  Higher levels of
sulfide in the wastewater would be expected to result in higher concentrations of
hydrogen sulfide gas in the sewer atmosphere and higher sewer corrosion rates.
However, the relationship between wastewater sulfide levels and corrosion rate is not
well established.

3.4   Comparison of Metals at CSDLAC with Other Cities Before Pretreatment

      Using available data, pre-1975 levels of metals and cyanide entering the main
CSDLAC wastewater treatment plant were compared with levels in the wastewater of
other municipalities across the  U.S..  Data were analyzed for 50 cities from the EPA
report, "Fate of Priority Pollutants in Publicly Owned Treatment Works"(6)(7).  These
data were collected in 1978-1979 prior to any significant implementation  of industrial
pretreatment standards.  The fifty cities typically had estimated industrial flow
contributions ranging from ten to fifty percent of the total flow. Analysis of these data
allowed determination of the number of cities with metals and cyanide levels similar to
those of CSDLAC prior to pretreatment, and assessment of whether other cities may
have had the potential to experience suppression of sulfide  generation and corrosion
due to the presence of these constituents.

      Table 3-9 shows a ranking of the 50 cities plus CSDLAC based on the
concentrations of selected metals and cyanide in the wastewater.  This was developed
from the sum of the equivalent weight concentrations of each of the constituents, and
does not account for the relative toxicity of the constituents on sulfide-producing
bacteria. Of the 50 other municipalities, only three (six percent) are  ranked higher than
CSDLAC, while 47 (94 percent) are ranked lower. The total concentration of metals
and cyanide in CSDLAC wastewater was approximately three times the median
concentration for the 51 cities.   Table 3-10 shows the actual constituent concentrations,
in ug/1, for  the 51 cities.

      The  total metals levels in CSDLAC wastewater in 1986 are also shown in Table
3-9.  On an equivalent weight basis,  1986 levels were 42 percent of  1971 - 1974 levels.
Comparing 1986 CSDLAC levels with 1978 - 1979 levels of 50 other cities, 16 cities (32
percent) were higher than CSDLAC, and 34 cities (68 percent) were  lower.

      Clearly, sulfide generation and corrosion in CSDLAC sewers increased


                                       3-15
-------
                                 TABLE 3-7

                  AVERAGE INFLUENT AND EFFLUENT SULFIDE;
                          PIPELINE PILOT PLANT (5)
Additive
                   CONTROL PIPELINE
               -Influent-       -Effluent-
DS   TS
DS   TS
                                  TEST PIPELINE
                              -Influent-       -Effluent-
DS   TS
PS   TS

Ix
Ix

Mixture
Mixture -t-Fe
mg/1
2.5
2.5
mg/1
3.8
3.9
mg/1
7.9
7.8
mg/1
9.7
9.8
mg/1
2.7
2.5
mg/1
3.7
3.6
mg/1
6.4
3.9
mg/1
7.5
6.4
                                 TABLE 3-8

       COMPARISON OF CONTROL AND TEST PIPELINE SULFIDE GENERATION:
                          PIPELINE PILOT PLANT (5)
EFFLUENT MINUS INFLUENT SULFIDE
Control
Additive

Ix Mixture
Ix Mixture + Fe
DS
mg/1
5.2
5.3
TS
mg/1
5.9
5.9
Test
DS
mg/1
3.8
1.4
TS
mg/1
3.8
2.8
PERCENT CHANGE
Control vs. Test
DS
%
-25
-77
TS
%
-36
-51
                                    3-16
-------
t
CO 
-------
                               TABLE 3-9

      COMPARISON OF CSDLAC METALS LEVELS BEFORE AND AFTER
      PRETREATMENT WITH METALS LEVELS OF 50 CITIES IN 1978-1979





CADMIUM CHROMIUM
PLANT
1
2
3
LA County(A)
4
5
6
7
8
9
10
11
12
13
U
15
16
LA County(B)
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
ueq/l
0.30
17.19
19.14
0.57
1.78
0.07
0.91
0.02
0.04
0.18
0.02
0.04
0.07
0.07
0.04
3.10
0.21
0.25
0.02
0.04
0.09
0.04
0.02
0.04
0.05
1.78
0.07
0.04
0.04
0.27
0.07
0.18
0.04
0.09
0.04
0.14
1.19
0.04
0.16
0.44
0.16
0.04
0.04
0.07
0.05
0.48
0.05
0.04
0.11
0.04
0.02
0.02
ueq/l
33.00
82.04
80.20
52.79
72.24
1.62
9.81
9.17
240.88
13.04
14.77
4.62
24.64
26.54
8.94
23.89
24.17
10.90
13.79
23.66
16.67
3.17
9.40
2.94
2.83
26.43
0.87
2.25
7.21
9.92
4.62
6.17
0.06
6.29
0.06
5.54
7.62
2.65
8.77
4.73
5.77
5.83
1.90
4.10
0.69
6.23
3.17
0.92
5.89
0.06
0.52
0.75


COPPER
utq/l
8.40
25.27
23.92
18.76
4.34
8.47
53.63
10.39
3.27
3.87
10.61
1.48
4.78
11.90
9.16
29.02
5.76
5.63
1.86
0.63
7.02
3.68
2.58
3.68
3.21
5.35
0.66
5.82
7.59
5.19
3.37
3.08
0.63
3.30
1.23
7.93
11.27
10.58
7.71
1.95
4.47
6.96
3.46
1.70
3.75
5.82
2.20
2.27
2.20
0.72
1.83
6.23
• Equivalent ut. •
CYANIDE
ueq/l
2.35
4.73
3.81
12.38
3.23
3.85
15.92
0.19
2.08
182.58
0.42
1.42
50.88
5.42
12.96
34.27
0.65
0.85
3.19
60.31
1.62
81.62
67.42
6.50
2.73
1.46
0.08
9.35
1.08
9.81
46.73
27.46
6.73
17.38
77.04
6.31
10.27
8.23
27.42
7.85
11.12
19.23
14.23
2.96
3.92
10.65
3.15
31.62
0.46
0.77
4.81
4.65
LEAD
ueq/l
5.63
1.58
1.92
3.01
2.09
0.95
11.81
1.87
2.51
1.31
3.18
1.23
1.27
2.71
0.07
0.56
0.46
1.50
0.48
0.48
0.69
0.56
0.43
1.31
0.34
1.93
1.19
0.25
0.48
1.01
0.53
0.45
0.48
0.49
0.48
1.93
0.88
0.37
1.53
0.09
1.30
0.15
0.48
0.15
0.28
0.78
0.88
0.33
0.65
0.39
0.09
0.23


MERCURY
ueq/l
0.00
0.05
0.03
0.01
0.00
0.01
0.01
0.02
0.02
0.00
0.00
0.01
0.01
0.00
0.00
0.01
0.00
0.00
0.01
0.00
0.01
0.00
0.01
0.01
0.00
0.00
0.01
0.01
0.01
0.01
0.01
0.01
0.00
0.00
0.00
0.00
0.01
0.01
0.01
0.00
0.01
0.01
0.01
0.00
0.01
0.00
0.01
0.01
0.00
0.01
0.00
0.00


NICKEL
ueq/l
6.44
14.99
23.89
9.75
7.43
2.08
3.10
1.57
4.77
3.34
14.55
0.68
37.38
10.22
13.43
5.59
3.17
3.37
0.85
1.70
11.76
0.34
0.34
0.82
0.72
5.79
0.37
2.15
1.87
2.35
0.20
1.84
0.34
2.93
0.34
9.03
3.68
2.11
2.18
0.75
2.73
0.17
0.44
1.02
3.99
1.16
1.29
0.14
0.00
0.27
3.27
0.14


ZINC
ueq/l
205.44
119.19
150.94
66.19
28.38
9.70
219.30
24.65
46.49
14.86
52.67
15.11
8.66
28.35
24.44
49.40
7.95
22.98
5.78
84.69
18.93
7.13
7.98
10.09
11.65
24.25
3.49
11.32
12.08
18.08
5.47
6.35
2.84
7.10
3.15
5.54
10.77
6.91
5.99
8.38
7.40
3.67
3.58
8.50
8.99
8.32
4.89
6.36
7.59
2.72
4.43
3.67


IRON
ueq/l
5440.47
313.00
263.68
383.40
423.55
468.21
110.01
364.28
91.18
152.81
195.53
263.51
134.30
170.29
180.60
103.43
190.88
183.99
199.40
45.27
124.02
79.22
84.98
146.94
135.73
73.88
132.50
104.75
96.16
79.57
63.32
72.88
103.50
71.52
25.18
65.50
46.16
60.63
34.42
60.09
49.06
45.34
53.79
58.73
52.36
39.50
53.90
25.50
42.69
50.14
26.21
24.32

TOTAL
ueq/l
5702.04
578.05
567.54
546.87
543.05
494.94
424.50
412.16
391.24
372.00
291.75
288.09
261.99
255.51
249.63
249.24
233.27
229.47
225.38
216.78
180.81
175.75
173.18
172.33
157.26
140.88
139.24
135.93
126.51
126.22
124.32
118.93
114.62
109.11
107.51
101.92
91.34
91.52
88.19
84.28
82.02
81.39
77.93
77.24
74.04
72.95
69.56
67.17
59.58
55.12
41.18
40.01
LA County (A) - Average levels during 1971 - 1974
LA County (B) - Average levels during 1986
                                  3-18
-------
                            TABLE 3-10

   METALS AND CYANIDE CONCENTRATIONS IN WASTEWATER
                         FROM 51 CITIES1

PLANT
1
2
LACounty*
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51

AVG:
STO:
CADMIUM
ug/l
17
10
32
966
1076
100
4
51
2
1
4
2
1
174
2
2
1
2
2
4
4
1
12
10
2
5
9
3
2
2
5
15
4
2
2
100
2
2
67
8
2
9
25
27
3
4
3
6
2
1
1
CADMIUM
55
198
CHROMIUM
ug/l
572
226
915
1422
1390
1252
28
170
55
159
427
410
163
414
1
80
256
155
4175
80
460
239
419
107
51
289
152
49
39
16
109
172
15
101
1
458
125
33
132
96
46
100
82
108
12
71
55
102
1
9
13
CHROMIUM
314
638
COPPER
ug/l
267
123
596
803
760
138
269
1704
117
330
152
20
82
922
39
47
337
291
104
107
378
59
183
98
117
223
245
102
185
72
105
165
21
221
20
170
241
110
358
252
336
142
62
185
119
54
70
70
23
58
198
COPPER
232
284
>- toncentration
CYANIDE LEAD
ug/l
61
4747
322
123
99
34
100
414
2122
5
1323
1568
1753
891
2003
37
11
337
54
1215
141
83
17
714
169
42
713
71
243
822
452
255
2
500
175
38
28
370
267
164
214
289
204
277
102
77
82
12
20
125
121
CYANIDE
472
799
ug/l
583
136
312
164
199
217
98
1223
58
194
132
50
45
58
50
127
329
7
260
55
281
50
48
47
136
72
158
35
26
34
51
105
123
16
50
200
50
50
91
200
38
135
9
81
29
16
91
67
40
9
24
LEAD
131
187
MERCURY
ng/l
133
333
1400
5000
3233
300
1000
617
67
1667
983
295
1050
517
200
1250
350
333
2000
600
117
900
305
600
517
1000
667
283
767
633
300
933
833
833
67
400
617
533
550
333
683
817
50
350
833
214
999
12
1167
483
500
MERCURY
757
812
NICKEL
ug/l
189
98
286
440
701
218
61
91
10
46
1097
50
10
164
10
20
427
394
140
6
300
25
93
54
24
345
64
21
63
4
86
69
11
5
10
170
55
13
108
265
62
80
22
34
117
30
38

a
96
4
NICKEL
135
198
ZINC
ug/l
6717
486
2164
3897
4935
928
317
7170
233
806
283
2769
261
1615
103
494
1722
799
1520
179
927
189
260
224
330
619
196
381
370
208
232
591
114
120
93
793
395
117
352
181
226
242
274
272
294
278
160
248
89
145
120
ZINC
911
1539
IRON
ug/l
151917
4267
10706
8740
7363
11827
13074
3072
2212
10172
3750
1264
2373
2888
703
7358
5460
5043
2546
1768
4755
5568
5330
2035
4103
3463
961
3790
2925
712
1997
2222
3700
1266
2890
2063
2685
1502
1289
1829
1693
1370
1678
1103
1462
1640
1505
1192
1400
732
679
IRON
6393
20791
Without PLANT #1
AVG:
STD:
56
200
299
644
226
287
477
805
120
177
770
816
133
199
771
1315
3416
2948
1  Compiled froa data contained in Refs. 5,6,7
2  LA County - average levels during 1971-74.  All other data for 1978-79.
                                   3-19
-------
dramatically between 1971 and 1986, and metals and cyanide levels dropped
significantly.  Unfortunately, it is not known what levels of specific metals, cyanide, and
combinations cause suppression of hydrogen sulfide corrosion.  Although CSDLAC
appears to have passed a threshold level of metals and cyanide which resulted in
increased sulfide levels and corrosion, it is difficult to predict whether other cities could
experience a similar increase in sulfide generation and corrosion upon reduction in
metals levels resulting from industrial pretreatment

3.5    Site Visits to Industrialized Cities

       EPA  conducted site visits to three cities having portions of the sewer system with
high industrial contributions.  Initially, it was believed that comparison of corrosion in
residential vs. industrial sewers might show differences attributable to the metals and
other constituents present in  the wastewater. The cities were Charlotte, NC;
Milwaukee,  WI; and Tempe, AZ.

3.5.1  Charlotte, North Carolina (see also Section 2.2.7)

       In the Charlotte sewer system, four sewers conveying primarily residential
wastewater and six sewers with a large industrial flow contribution were  inspected. All
of the sewers were 20 to 25 years old.

       Total sulfide levels in  the residential sewers ranged from 0.2 to 0.6 mg/1.  Pipe
surface pH measurements were generally 6.0, with the  exception of one  site where
surface pH levels ranged from 4.5 to 6.0.  At that site,  corrosion penetration was
approximately 0.25  inches, exposing aggregate.  No measurable H2S was detected in the
sewer head space at any of the four sites.

       Two of the six industrial sites showed signs of shallow hydrogen sulfide corrosion,
with penetration estimated to be up to 0.12 inches.  Wastewater sulfide  levels at the six
sites ranged from 0.0 to 0.3 mg/1. At sites where corrosion was observed, high
turbulence levels were noted. Wastewater pH measurements were 6.0 at four of the
industrial sites, 5.5  at one site, and 10.0 at the remaining site.  No H2S was detected in
the headspace at any of the six industrial sites.

3.5.2  Milwaukee, Wisconsin (see also Section 2.2.8)

       In Milwaukee, observations included  five sewers conveying primarily residential
wastewater and five sewers with a heavy industrial flow contribution. Three of the
residential sites were located in the Jones Island WWTP service area, and were concrete
pipes ranging in age from 50 to 70 years.  No corrosion was observed, and  the surface
pH of the concrete was measured to be 6.5. Two other residential sites were sewers in
the South Shore WWTP service area.  One site was a sewer less than 20 years old; the
other was 50 years old.  The 20 year old sewer was similar to the first three - no


                                        3-20
-------
corrosion and high surface pH.  However, at the other site wastewater sulfide content
was 0.5 mg/1 and the crown pH was 3.5.  Severe corrosion was observed, with up to one
inch of concrete lost

       All five industrial sites are at least 6 miles downstream from the beginning of the
collection system and at least 40  years old. No corrosion was observed at any of these
sites.  However, two sites had measurable wastewater sulfide levels of 0.18 and 0.40
mg/1.  Crown pH levels were between 6.0 and 7.0 at all of the industrial sites.  One site
was less than 0.5 miles downstream of a tannery.

3.53  Tempe, Arizona

       A site visit was made to Tempe because one area of the City generated
wastewater primarily of industrial origin. These industries included four circuit board
manufacturers,  two  electroplaters, two metal  finishers, one coating operation, and two
dry cleaners.  It was believed  that results of detailed monitoring data from industrial vs.
residential areas could provide some  insight as to the impact of industrial discharges on
the extent of hydrogen sulfide corrosion.

       Data from  previous monitoring by City of Tempe staff were reviewed, and
inspections were made of industrial and  residential sewers. The City had measured
sulfide, pH, temperature, ORP, and D.O. of  the wastewater, and had contracted with a
local laboratory for  metals analysis. At the industrial site, dissolved sulfide levels ranged
from 0.05 to 1.9 mg/1 on five separate days (grab samples).  Dissolved sulfide levels
averaged 0.6 mg/1. Atmospheric  H2S levels were 1 to 2 ppm.  Wastewater pH ranged
from 8.0 to 8.9. Inspection of a manhole in the 27 inch sewer conveying the majority of
wastewater from the industrialized  area showed no evidence of corrosion.  The sewer
system in this area was approximately 20 years old, and appeared to be in excellent
condition with no sign of deterioration.  On the day of inspection, no atmospheric H2S
was detected in the  manhole near the liquid surface.

       A large trunk sewer conveying residential wastewater from the adjoining City of
Mesa was inspected at several locations. Previous monitoring by the  City of Tempe had
indicated dissolved sulfide levels  of up to 9 mg/1, and  atmospheric H2S levels of up to 68
ppm.  The concrete  manholes which were unprotected exhibited severe corrosion above
the waterline, with abundant quantities of corrosion product  Several manhole
chambers had been  lined with plastic, and those manholes that were sources of odor
complaints were equipped with carbon canisters to control odor emissions.

       No conclusions could be drawn from existing data regarding the impact of
industrial discharges on hydrogen sulfide corrosion. A further,  detailed monitoring
program was abandoned because of the high  pH (8.0  to 8.9) of the wastewater
emanating from the industrial zone.  This high pH prevents significant  amounts of H2S
from being  released from solution,  since over 90 percent of the dissolved sulfide  is


                                       3-21
-------
present as the hydrosulfide ion, not as the dissolved gas. In general, it is believed that
attempts to assess the impacts of metals and other industrial constituents on hydrogen
sulfide corrosion by monitoring industrial vs. residential sewers are futile due to the
many factors which  affect sulfide generation and corrosion.

3.6   Beneficial Effects of Local Industrial Pretreatment Programs

      It is important to recognize that several aspects of the industrial pretreatment
standards may actually lower the potential for sulfide generation and corrosion  in sewer
systems.  Among the more important of these are 1) reduction of sulfide-bearing wastes,
2) reduction of high strength organic waste discharges, 3) reduction of high temperature
discharges, 4) reduction in fats, oils, and grease, and 5) reduction in acidic wastes.
Because of the complex interaction of all the factors that affect sulfide generation, it is
very difficult to quantify these effects for a broad base of sewer systems.  Beneficial
impacts of local regulation of industrial waste discharges on sulfide  generation in
municipal sewers are summarized in Table 3-11.  In this table, sulfide is the only
parameter specifically regulated by the EPA Categorical Pretreatment Standards.

3.7   Conclusions

      The national effects of industrial pretreatment on hydrogen sulfide corrosion are
impossible to ascertain since no municipalities other than CSDLAC were found to have
sufficient data to establish a correlation.  Based on theoretical analysis, review of full
scale and pilot scale research data from CSDLAC, and a series of site investigations, the
following conclusions are presented.

       •     The reduction in metals and other industrial constituents in CSDLAC
             wastewater apparently caused an acceleration in corrosion rate, possibly
             due to biological inhibition  and/or chemical precipitation.

       •     Two pilot studies conducted by CSDLAC demonstrated that sulfide
             generation was reduced when metals were added to the wastewater at
             levels approximating those in the early 1970's.

       •     When comparing 1970's data from 50 other cities having 10 to 50 percent
             industrial flow input, total metals and cyanide levels in CSDLAC
             wastewater were higher than levels in 94 percent of 50 U.S. cities.

       •     If current (1986) CSDLAC  data are compared with 1970's data from 50
             cities, CSDLAC levels would be lower than 32 percent of the cities.

       •     It is difficult to project how many cities could potentially be adversely
             affected by industrial pretreatment since it is not known at what levels
             industrial constituents begin to suppress sulfide generation.


                                        3-22
-------
                                 TABLE 3-11
                  BENEFICIAL IMPACTS OF CONTROLLING
             INDUSTRIAL DISCHARGES ON SULFIDE CORROSION
Type of Discharge Controlled
Sulfide-bearing wastes
Benefit

Lowers sulfide levels,
corrosion potential
High organic strength
(BOD) wastes
Sulfide   generation   rate
proportional to BOD; reduction in
organic  strength reduces oxygen
uptake and depression of dissolved
oxygen
High temperature wastes
Lower temperature reduces sulfide
generation  rate;      increases
solubility of H2S, reducing release
of  H2S;  increases  solubility  of
oxygen
Wastes containing fats,
oils, and grease
Reduces   potential   for   sewer
clogging, reduced velocities, solids
deposition, and sulfide generation
Low pH wastes
Maintaining   pH  at  or  above
neutral reduces release of H2S to
the sewer atmosphere
                                     3-23
-------
Site visits to inspect corrosion in residential vs. industrial sewers were
inconclusive regarding the impacts of metals and other industrial
constituents on hydrogen sulfide corrosion.

Local regulation of certain non-toxic constituents in industrial waste
discharges has likely had a beneficial impact in reducing the potential for
sulfide generation and corrosion.

Additional research is necessary to establish the constituents and their
associated levels at which sulfide generation is suppressed or accelerated.
                           3-24
-------
                                REFERENCES
1.     Won, D.L., "Sulfide Control with Ferrous Chloride in Large Diameter Sewers,"
      internal report, County Sanitation Districts of Los Angeles County, November,
      1988.

2.     Pomeroy, R.D., Parkhurst, J.D., Livingston, J., and H.H. Bailey, "Sulfide
      Occurrence and Control in Sewage Collection Systems," U.S. EPA, EPA 600/X-
      85-052, Cincinnati, OH, 1985.

3.     Pomeroy, R.D., and F.D. Bowlus, "Progress Report on Sulfide Control Research,"
      Sewage Works Journal, Vol. 18, No. 4, pp 597-640, July, 1946.

4.     Internal Monthly Reports, County Sanitation Districts of Los Angeles County,
      1988.

5.     Morton, R., Caballero, R., Chen, C.L., and J. Redner, "Study of Sulfide
      Generation and Concrete Corrosion of Sanitary Sewers," prepared for the 62nd
      Annual Conference of the Water Pollution Control Federation, San Francisco,
      October, 1989.

6.     "Fate of Priority Pollutants in Publicly Owned Treatment Works - Volume I,"
      EPA 440/1-82/303, USEPA, Washington, D.C., Sept, 1982.

7.     "Fate of Priority Pollutants in Publicly Owned Treatment Works - Volume II,"
      EPA 440/1-82/303, USEPA, Washington, D.C., Sept, 1982.
                                     3-25
-------
4.0   DETECTION, PREVENTION AND REPAIR OF HYDROGEN SULFIDE
      CORROSION DAMAGE

      Alternatives for the detection and prevention of hydrogen sulfide corrosion in
both existing and new wastewater systems, and techniques for repairing hydrogen sulfide
corrosion damage, are summarized in this  section. Additional detailed information  may
be found in publications prepared by the Environmental Protection Agency, The
American Society of Civil Engineers,  the American Concrete  Pipe Association, and the
U.S. Department of Housing and Urban Development (1)(2)(10)(7).

4.1   Detection and Monitoring of Hydrogen Sulfide Corrosion

      One of the most useful "early warning1' indicators of potential hydrogen sulfide
corrosion problems is pH of the pipe crown or structure wall.  This is a simple test using
color-sensitive pH paper which is applied to the moist crown of the pipe.  New pipe has
a high pH of 10 to 11.  After aging  the pH of the crown under non-corrosive conditions
may drop to near neutral.  Pipe experiencing severe hydrogen  sulfide corrosion may
have a pH of 2  or lower.

      Dissolved sulfide levels in the wastewater and hydrogen sulfide  levels in  sewer
headspaces can  be checked to determine if sulfide is being generated in the sewers  and
where and to what extent it is being released from solution. Routine monitoring may be
justified at the lift stations, junction structures, discharges  of force mains, treatment
plant headworks, or other locations in the  collection and treatment system.  Such tests
indicate whether conditions are present  for hydrogen sulfide corrosion to occur.

      Routine visual inspections are essential. Where accessible by a worker, this can
be done by entering manholes or sewers and checking the soundness of the pipe
material. A screwdriver or other sharp tool can be used to determine  the depth of
penetration into soft corrosion product  Since corrosion products occupy greater
volume than the original concrete, depth of penetration is not an accurate  measurement
of concrete lost to corrosion (section 1.4).  Concrete loss can be  approximated by
measuring the depth of aggregate protrusion from the surface. Sewer  pipe may also be
inspected remotely through the use of television cameras.  With  improvements  in the
resolution of camera equipment, TV inspections can often identify corrosion problems,
although considerable damage may already have been done.

      A relatively recent development in remote sewer inspections is the use of "sonic
caliper" technology to measure the inside dimensions of the pipe. Sonic signals  are
transmitted from a floating raft to the pipe walls, and the signal is detected after
reflecting off the wall. Software developed by a proprietor is used to process the signals
and determine the variation in pipe diameter along its length.  Areas where loss of  pipe
material has occurred can thus be detected. The technique was successfully used to
inspect over 40,000 feet of 36 in. to 54 in. diameter pipe in Tampa, Florida (9).


                                       4-1
-------
      Core borings of the pipe crown and submerged pipe may be taken to calculate
the extent of the corrosion loss.  Some municipalities only take cores of corroded
portions of pipes to determine how much pipe remains.  Expandable rods have also
been used to measure the inside pipe diameter, rather than taking core borings, as a
means of estimating the extent of corrosion, although errors are introduced due to
variation in wall thickness and pipe "roundness.11

      To determine the rate of corrosion, the thickness of corroded pipe must be
compared at two different points in time, since no instantaneous technique  has been
proposed for monitoring corrosion rates.  CSDLAC is the only entity known that has
determined corrosion rate on a regular basis using core borings.  Some sewers may be
installed with vitrified clay "plugs" or stainless steel rods in the crown at accessible
locations, providing a direct visual comparison of adjacent corroded and non-corroded
material.

4.2   Prevention of Hydrogen Sulfide Corrosion in Existing Systems

      A number of techniques have been used to control corrosion and odors
associated with hydrogen sulfide generation  in existing systems. The most common
techniques can be divided into the general categories of oxidants, precipitants, or pH
elevators.   Oxidants control  sulfide by chemically or biologically causing the oxidation of
sulfide to thiosulfate or sulfate.  Such techniques include air or oxygen injection, or
addition of chemicals such as hydrogen peroxide, chlorine, or potassium permanganate.
Precipitants control sulfide by precipitation with a metal salt such as ferrous chloride,
ferrous sulfate, or zinc salts.  The dissolved sulfide is converted to an insoluble
precipitate, preventing  release of gaseous H2S.  Elevation of the pH through shock
dosing of caustic controls  sulfide generation by inactivation of sulfide-producing slimes
present on the wall of the sewer pipe.  A summary of sulfide control techniques is
provided in Table 4-1.

      All of the above control techniques are oriented towards reducing  the  levels of
dissolved sulfide in solution  such that less sulfide is released to the sewer atmosphere.
Work conducted by CSDLAC indicated that, although significant reductions (75 to
95%) in dissolved sulfide could be obtained with chemical addition, only modest
reductions (50 to 60%) in H2S levels in the sewer atmosphere were  realized (4).  Thus,
a 90% reduction in dissolved sulfide  does not necessarily indicate that the rate of
corrosion will be reduced  by 90%. Although empirical predictive corrosion equations
assume that corrosion rate is directly proportional to the rate of H2S flux  from the
wastewater to the sewer walls, this relationship is very difficult to verify with field  data.

      No one sulfide control technique can be generalized as being the most cost-
effective.  Dosages of chemicals to control sulfide vary widely from one wastewater
system to another, and are dependent on wastewater characteristics and other site-
specific factors.  Sulfide control options must be considered on a case-by-case basis.


                                        4-2
-------
             I
             T3
                                                      cTa

                                                    £oj
                                                    — 00 3
                                                    .2 .S "5
                                                    *; a) 35
                                                O 4>
                                               D- T3
                                                                        V)
                                                                        O
                                                                        O
                                                                        CM
                                                                        C/3
                                                         w  3 W
                                                         .G .w 73
                                                         .2P!tS g
                                                           K<«M Cv
                                                           T3 A
Ed
w
6-
                  c
                  cs;

                 T5I
                         — SO
                       »> § -S -a
                       T3 ^ &  C8

                       "2 - ^  «
                         3 -  a
                       «x 5 e  j>
                       (« » 4)  00
|o ?1
•So 2  3
                                                              O 7!^

                                                              w  c °  «
                                                              I-  ft at TT
                                                              «  P <*
                                                              »P  O *•
                                                              <*<    4)
                                                  2 S
                                                  00 U
                                                  o g
                                   -° tS «
                                   « «« .°*
                                   O C tf)
                                   IS 1) ""
                                   •s. & a
                                   »• £? ^
                                   < O S
S                    a
                 — 2
           > « 4) M '5

          I'sJii
          fi fe « C8 u?
                                             <-    M °

                                             =   .§2
V
I
o.

Sf-
•c e
o «
                                                                               W
                                                                                       X
                                                                                       o
Tf
U
Z
O
u
w
a
           
                                   oo

                                   Q.S
                                               S
                                               CQ
                                                  0
•3-Si,
35 .S
                                                              
-------
            •o
            re

u
.0
— >»
S w-1
§ 5 2 2
!^P
§2*^
i *sli
Q §32^5
2-a
re «
8 «
a •?
Not economical
compared to iro
discharge is regi
                                                   12
                                            if I-
                                             v-2 2 3
                                             o.<~ 
u
U
U
H
H
Z
O
u
            re

            T3I

                      is*
                        O m
                             -o a
                             3--0 i

                                » « a
                                   —
                                       2
2                 e
               .- o

               II

               II
                                                              I «

                                                              «* 5
                                                              3 £
                                                              •O 3
                                                              V M
                                                           g a
83
    C/3
    b
    O

    g
      u

      'I ~
      re Q
      "u 0
                                                                 oo
    1/1
            
                                                           08
                                                           •frrf

                                                           '5

                                                           S
                                                           3


                                                           1
                                                           1/1
                                    4-4
-------
Brief descriptions of sulfide control methods applicable to existing wastewater systems
are provided below.  More detailed information may be found in  references (1) and (2).

4.2.1   Air Injection

       Injection of compressed air into the wastewater is most applicable to force mains,
siphons, and pipes flowing under pressure. Often, air is injected on the discharge side
of sewage pumps to provide dissolved oxygen which promotes oxidation of existing
sulfide and prevention of further sulfide build-up.  The pressure in the pipe, being
greater than atmospheric, allows dissolution of greater quantities  of oxygen. Air
injection is an economical alternative for sulfide control in pressurized lines. Because of
the large quantities of air injected, potential exists for gas accumulation and increased
head  losses.  Although research has been conducted on pressure tank dissolvers and U-
tubes for use in aerating gravity sewers, such devices appear to be marginal for this
purpose.

4.2.2   Oxygen Injection

       Oxygen is five times more soluble in water than air, and thus it is possible to
achieve higher  DO levels in the wastewater.  As with compressed air injection, oxygen is
most  applicable to sulfide control in force mains and pipes under  pressure.  However, it
is currently being used in Sacramento to oxygenate wastewater in  a fall structure, and
CSDLAC has conducted demonstrations of pressurized sidestream dissolution for
oxygenation of  gravity sewers.  In the sidestream dissolution  system, a portion of the
flow is directed through a pressurized pipe into which oxygen is injected. The  oxygen-
saturated  sidestream is then introduced back into the gravity main.

       Oxygen is generally an economical technique for sulfide control. However, the
annual costs for purchased oxygen are highly dependent on how efficiently  the oxygen is
transferred into solution.

4.23   Hydrogen Peroxide

       Hydrogen peroxide is widely used  for sulfide control in force  mains and gravity
sewers. At neutral and acidic pH, H2O2 oxidizes H2S to elemental sulfur. Dosage
weight ratios of H2O2 to H2S vary from near stoichiometric (1:1) to over 5:1, depending
on degree of control desired, wastewater characteristics, initial sulfide level, and
wastewater travel time between injection station and control point  Costs for sulfide
control using hydrogen peroxide are competitive with other sulfide control chemicals.

4.2.4   Chlorine

       Chlorine  oxidizes sulfide to sulfate or elemental sulfur.  Chlorine can be
purchased as a  gas or as hypochlorite solution.  In practice, C12:H2S  dosage  weight ratios


                                        4-5
-------
are typically in the range of 10:1 to 15:1.  Although commonly used for sulfide control,
dosage requirements and unit chemical costs often make chlorine uneconomical
compared to other chemicals  used for sulfide control.  The hazardous nature of chlorine
gas make it less attractive for use near residential areas.

4.2L5  Potassium Permanganate

      Potassium permanganate is a powerful oxidant that is effective for sulfide control.
In general, dosage weight of KMnO4 to H2S are approximately 6:1 to 7:1. Potassium
permanganate is purchased as dry crystals.  Chemical costs are high, and use of KMnO4
for wastewater applications is generally not cost-effective.

4.2.6  Metal Salts

      Iron salts such as ferrous sulfate and ferrous chloride are widely used,
economical chemicals for sulfide control.  Iron reacts with H2S to form an insoluble
precipitate, preventing release of H2S from solution.  In practice, dosage weight ratios of
FeSO4 to  H2S are approximately 5:1, although higher dosage ratios may be required
depending on wastewater characteristics, initial sulfide levels, and degree of sulfide
control required.

4.2.7  Sodium Hydroxide

      Sodium hydroxide is added in "shock doses" to sewers for sulfide control. Caustic
soda (NaOH) is added over a period of 20  to 30 minutes at sufficient dosages to elevate
the pH to between 12.5 and 13.0.  The high pH slug temporarily inactivates sulfate
reducing bacteria and greatly reduces hydrogen sulfide generation.  Within a period of
several days to two weeks, the sulfate reducing bacteria become re-established, and
caustic dosing must be repeated.  If this approach is employed in the collection system
near the treatment plant such that dilution of the high pH slug does not occur,
provisions must be made to store the  high pH wastewater and gradually release it to the
plant to avoid biological upset

4.2.8  Other Chemicals

      Other chemicals have  been used for sulfide control with varying degrees of
success. Sodium nitrate has been used for  H2S control in lagoons,  trickling filters, and
carbon columns, but has not been widely used for sulfide control in sewers. Nitrate
prevents sulfide generation by acting as a hydrogen acceptor which is used preferentially
by bacteria over sulfate.

      Several proprietary bacterial cultures and  enzyme preparations are claimed to be
effective for sulfide and odor control although their effectiveness has yet to be
demonstrated.


                                        4-6
-------
4 J   Prevention of Hydrogen Sulflde Corrosion in the Design of New Systems

      Consideration of sulfide generation and corrosion is critical in the design of
wastewater collection systems. While it is possible, and sometimes necessary, to
incorporate chemical addition stations for sulfide control as part of the overall system
design, the most cost-effective and rational engineering approach is to develop  a
hydraulic design that minimizes sulfide generation.  In general, such an approach strives
to maintain aerobic conditions in the wastewater by providing adequate wastewater by
providing adequate wastewater velocities, and by minimizing the use of force mains,
inverted siphons, and surcharged sewers in which anaerobic conditions can develop,
resulting in sulfide generation.

      Under certain conditions, sulfide generation may be unavoidable.  Empirical
equations have been developed to allow prediction of sulfide build-up and rates of
corrosion. Where  sulfide generation is anticipated, corrosion resistant materials can be
selected, or the alkalinity and thickness of concrete pipe can be specified to help reduce
the effects of hydrogen sulfide corrosion.

      Table 4-2 summarizes various approaches used to minimize sulfide generation
and corrosion  during the design  of wastewater collection and treatment facilities.
Several key design elements are  summarized below.  More detailed discussions of
corrosion prevention during design may be found in references (1),(2), and (10).

4.3.1  Wastewater Velocity

      Wastewater velocity is critical in designing sewer systems to prevent or minimize
sulfide generation. Adequate velocity 1)  prevents deposition of solids which can cause
flow obstructions and increase sulfide generation and 2) provides surface reaeration
which helps to maintain aerobic  conditions and prevent sulfide generation.  Although a
minimum scouring velocity of 2 ft/sec has been historically used by engineers designing
gravity sewers, large diameter sewers require much higher scouring velocities, on the
order of 3 to 4.5 ft/sec.  Minimum scouring velocities for force mains are typically  3 to 5
ft/sec depending on pipe size.

      The impact of velocity on reaeration rate is significant For a 24 inch diameter
flowing half full, increasing the velocity from 2 to 3 ft/sec increases the reaeration  rate
by a factor of  2. The effect is somewhat less dramatic for sewers larger than 36 inches.

      Velocity in sewers is controlled by flowrate, slope, and pipe diameter.  Figure 4-1
is a generalized guide showing the potential for sulfide generation as a function of flow
and slope.
                                        4-7
-------
                    «»
                    «
                         0
             S t5
             !»
          2
          V
          a
         N

         |
         5

         *S

         2
                                         a
Cd
a
o

2
3
a
z
o
55
O
atf
fig
O
u
u
a
E
              •o
              u
             a.3
             o o'
            • 5 c
            2
            S S
       J2
       .2
       *C
       u
       **
       «
       S

         •«
       If
       S • 5
       .1
       ••=
       5S
     .•»§
     eg

     ^ 9
     eo 3
     00 —
     V A
     Sh'G
     06 -0
                                   SI
                     S
                     «
    Z
    u
    u
    fig
    ft.

    g
    CA
    ft.
    ft.
N "3
Itl
•** ^s
             <« c
             S.2
            '
       I
       0)
       •a

       Is
       0..0

       2|

       
-------
               0.6
               0.5
        o
        o
        5

        o
        o
        c/5
               0.4
0.3
               0.2
               0.1
                                    Effective BOO
                                    500 mg/l
                      Little suiflde
                      generation potential
                             Moderate suiflde
                             generation potential
                                                    Severe suiflde
                                                    generation potential
                  0  0.1 0.51.0    5.0   10  1520  3040

                              Row, cu ft/s
FIGURE 4-1  GUIDE FOR ESTIMATING SULFIDE GENERATION POTENTIAL
                                    4-9
-------
43.2  Design of Junction and Drop Structures

      Turbulence created by junctions and drop structures can have opposite effects on
sulfide build-up and corrosion, depending on the characteristics of the wastewater when
it reaches the structure.  If no sulfide is present, turbulence will increase reaeration
rates, thereby adding dissolved oxygen and maintaining aerobic conditions.  On the
other hand, if dissolved sulfide is present, turbulence will increase the release of
hydrogen sulfide to the atmosphere and increase the rate of corrosion.  Therefore, if
potential for sulfide generation exists, designs of such structures should be such that
turbulence is minimized.

433  Force Mains, Siphons, and  Surcharged Sewers

      Force mains, siphons, and surcharged sewers have one thing in common:  the
sewer flows full with no opportunity for reaeration.  This results in anaerobic conditions,
generation of H2S, and often severe corrosion at the outlet  In general, use of force
mains, siphons, and surcharged sewers should be avoided whenever possible.

      Where required, design velocities should be selected to avoid solids deposition,
and detention times should be minimized. Where long force mains or siphons are
necessary, consideration should be given to positive sulfide control systems (e.g.,
chemical addition) and/or use of corrosion resistant materials such as PVC liners at
discharge points.

43.4  Sewer Ventilation

      Sewers are naturally ventilated through  building vents  and manholes,  occurring
from factors such as changes in barometric pressure, wind, air density differences, and
flow conditions.  Wet wells at pumping stations have 12-30 changes/hour, and dry wells
have 6-30 changes/hr.  Ventilation is often practiced at wastewater treatment plants,
where air is withdrawn at the headworks and either treated separately or piped to
existing biological processes (1).

43.5  Local Control of Industrial  Discharges

      Local control of industrial discharges  as a means of minimizing sulfide generation
is applicable to both existing and new wastewater systems. Beneficial impacts of
controlling industrial discharges  on sulfide generation were summarized in Table 3-11.
Adequate industrial pretreatment to make such wastes compatible with municipal
wastewater can eliminate their contribution to sulfide generation potential.
                                       4-10
-------
4.3.6  Design Considerations When Sulflde Generation is Anticipated
4.3.6.1 Prediction of Sulfide Generation and  Corrosion
      In some cases, it is difficult or not cost-effective to design a wastewater collection
system that will be free of sulfide problems.  It is then useful to know what levels of
sulfide can be expected.
      Empirical equations have been developed to allow prediction of sulfide levels.  In
addition, a model has been developed to allow prediction of corrosion rates where H2S
is present
      The Pomeroy-Parkhurst equations that predict sulfide build-up are given below:
      Pies Flowin   Less than Full
           log'1
where:
                         2.31 dn
                   = predicted sulfide concentration at time tj
                   = sulfide concentration at time tt
                   = theoretical upper limit of sulfide concentration
            s      = slope of the pipe
            u      = stream velocity
            t      = (tj-t,) flow time
            m     = empirical coefficient for sulfide loss
            dm    = mean hydraulic depth
      Pipes Flowing Full
      S2 = Sj + (M)(t)[EBOD (4/d) + 1.57)]
      where:
            M    = experimentally determined empirical constant
                     representing the sulfide flux
            EBOD = BODfl.O?™] (T = temperature, °C)

                                      4-11
-------
The rate of corrosion of concrete pipe can be predicted using the following equation:

      Cwg = 11.5 k 0,,
                A
      where,                        —

             Cmg   = average rate of penetration, mm/yr
             k     = Coefficient of efficiency for acid reaction considering the
                     estimated fraction  of acid remaining on the wall. May be as low
                     as 0.3 and will approach  1.0 for a complete acid reaction.

             0n   = flux of H2S to the  pipe wall, gm/m2-hr

             A     = Alkalinity of the cement  bonded material, expressed as CaCO3
                     equivalent  Approximately 0.18 to 0.23 for granitic aggregate
                     concrete, 0.9 for calcareous aggregate, 0.4 for mortar linings, and
                     0.5 for asbestos cement

             11.5   = constant
             0W   = 0.69(su)3/8j[DS](b/P')


             where,

             s     = energy gradient of wastewater stream, m/m

             u     = stream velocity, m/s

             j      = fraction of dissolved sulfide present as H2S as a function of pH

             [DS]  = average annual concentration of dissolved sulfide in the
                     wastewater, mg/1
             b/P'   = ratio of width of wastewater stream at surface to exposed
                     perimeter of the pipe wall above the water surface.

      Peak corrosion rates occur at the sewer crown, and may be higher than the
average corrosion rate by a factor of  1.5 to 2.0.  This is the crown corrosion factor
(CCF).  Another factor, the turbulence corrosion factor (TCP) is used to account for
greater  flux of H2S to the pipe wall, and may vary from 1 to 2.5 for well-designed
junction structures or other areas with nonuniform flow conditions. At drops or
turbulent junctions, the  turbulence corrosion factor may be 5 to 10.

Thus, the peak crown corrosion rate  is given as follows:

      C^ = Cavg x CCF x TCP
                                       4-12
-------
4 J.6.2  Selection of Materials

       When it is anticipated that sulfide will be present, consideration must be given to
use of corrosion-resistant materials.  Pipe materials such as PVC, PE, and vitrified clay
are virtually unaffected by sulfuric acid produced by the biological oxidation of hydrogen
sulfide, and should be specified if hydrogen sulfide corrosion is anticipated.  For larger
diameter sewers, PVC or vitrified clay pipes are not available, and concrete  pipe is often
used.  However, PE pipe is available in diameters up to 120 inches.  Two alternatives
are frequently used to protect the concrete pipe from corrosion or to extend its useful
design life.  These are 1) use of PVC liners which are imbedded into the concrete pipe
during manufacture, and 2) use of high alkalinity calcareous aggregate and/or additional
sacrificial concrete cover over the reinforcing bars.

       The proprietary T-lock" PVC liner has been successfully used for concrete pipe
protection for many years.  Although  the cost of the pipe is increased, follow-up tests
have shown that the PVC liner provides excellent protection of the underlying concrete.
The PVC is immune to sulfuric acid attack, and proper installation of the liner prevents
migration of acid to the concrete.  Surface-applied synthetic coatings have yet to
demonstrate the longevity and acid resistance of PVC liners.

       Use of calcareous aggregate  in concrete  pipe increases  the alkalinity of the
concrete and thus increases the resistance to sulfuric acid attack.  An additional
sacrificial layer of concrete over the reinforcing steel increases the useful design  life of
the pipe.  An  equation has been developed to assist in proper selection  of pipe materials
(alkalinity) and thickness. This is the "Az" or "life factor" equation, shown below.

       Az = 0.45k 0W L(CCF)(TCF)

       where,

             Az   = life factor, equal to the product of alkalinity and thickness of
                   allowable concrete loss

             L     = desired design lifetime, years

       Thus, assuming a design lifetime of the pipe, it is possible for the engineer to
specify the desired combination of pipe wall thickness and concrete alkalinity to achieve
the target lifetime.

4.4 Repair of Damage Caused by Hydrogen Sulfide Corrosion

       Once corrosion damage has occurred, it  may be necessary to repair a structure to
reduce the potential for failure or collapse.  In  the past, excavation and  replacement was
a common repair solution to corroded pipes and structures.  However, due to the


                                        4-13
-------
expense, the disruption to traffic, the potential for damage to other underground
utilities, and the interruption to the service itself, in-line rehabilitation techniques have
become more attractive.  Rehabilitation techniques are those methods and repairs
applied to an existing structure to prolong its useful life.  With such techniques,
municipalities can repair existing structures at a lower cost than replacement, and
without public inconveniences due to traffic disruptions and service interruptions.  In
many situations, pipelines can be rehabilitated at somewhat less than the cost of
replacement Rehabilitation techniques are not acceptable under the following
conditions:

       •    where significant additional capacity is needed
       •    where rehabilitation methods that are adequate to restore pipeline
            structural integrity  would produce an unacceptable reduction in service
            capacity
       •    for point repair where short lengths of pipeline are too seriously damaged
            to be effectively rehabilitated by any means
       •    where entire reaches of pipeline are too seriously damaged to be
            rehabilitated
       •    where removal and replacement is less costly in dollars and urban
            disruption than other rehabilitation methods

       Numerous rehabilitation techniques  exist, but not all are applicable to corrosion
repair.  The selection of a particular method  of rehabilitation depends on many factors
such as economics, extent of damage and structural integrity, disruption of traffic and
excavation requirements.  High concentrations of sulfuric acid may be detected on the
walls of sewers where H2S is being generated. Sulfuric acid  can quickly deteriorate
crown and sidewall concrete, thereby exposing aggregate and reinforcing steel,  and
potentially weakening structural  integrity.  Therefore,  corrosion rehabilitation techniques
must focus on internal repairs to ensure structural  integrity and provide a protective
barrier against  subsequent acid attack, rather than  on external repairs (e.g., soil
stabilization).  After surveying consulting engineers, municipal engineers, and
manufacturers,  the following seven generic rehabilitation measures were identified as
appropriate for acid corrosion repair:

             insertion renewal (sliplining)
            deformed pipe insertion
            cured-in-place pipe
             specialty concrete
            coatings
             liners
             spot replacement
       Table 4-3 describes various methods of pipeline rehabilitation, indicating their


                                        4-14
-------
ori
                                          .5   .§
Limite
applica
diamet
Bypassi
                                                -,
                                          a e 8 -a?
                                          I
                                          fill!

O

a
a
ja


E
2
c
0
>
•o
i
u
i
•a
^
,2

cement
n
a.
u
•o
<•
1
tt
>
S
X
w

S-

§
a.
.2
•a
"3
B
a
'§

w
e
S
a

8
«j
ka
•o
>»
J3
W
S
a.
S
s
!
VI
U
s
S
Provides
*

d
i.
w
8.
o
a.
e
u
-a
rcement w
5
a
'u
h*
•a
"a-
w
h«
O
a
oo
a
5
s.
>,
CO
•o
ii
3
o
&
                              frJ
                              S a-
                              .. u
                              2L<*
                              Ijf
                              5 I
                              a >-.
                              2 00
                              *« w
                              O k*
                              si
                              §5
                              as
                                              -9
no mixing of resins, curing.
May not require excavation.
May improve hydraulics.
                                          u   o
                                            "•
                                                B >«
                                                o ,2
                                                     S
                                                     t»
                                                    i
mp
f£
Cd
z
T   a.
1   «
S   2
    V)
    a
    o
                                               a  o
                                               S '3
                                               o  r?  v
                                               1U--  S-
                                               -0 o.'a.
                                           oSs
                                           s.i-
                                           w « U
                                           u u a
                                           i oo—
                                           « S «
                                           co .2 w
                                                           00
                                                           a
                                                           •3
                                                           u
                                                           u
                                                           a
                o .±
                2? a
                32
u
z
2
a.
u
1*8
§ 5
e
_o
0
•c
X
41
-S«l
•a:5
1s
0 «
0 B
a. a
'5. '-3
1- U
11
< §


a
u
.d
2L
s.
BO
a
w)
•3
W
o
.2
^i
ia
h*
iJ
«
i
X
2
•o
s
conne
V
a
u
5
l-o
£s
8.5
11
Is
»> -H
•3 i-
'C 3
H
la
s
<2
a
u
h«
O*
1
D.
OO
_s
S
k*
U
a


: mortar.
VI
fl
Q.

VI
a
5.
o
vt
«
O.
O
U
.£
<

^
O
oo
a
•3
£
>,
J3
•a
u
s
£
o
•o
o
.3
^
0
I
•a
a
«
§
3
2
a.
S
8
1 s
^3 w
e -
ag pipe and expa
ally, hydraulically
11
W B


^J.
1
e
ra
_a
u
u
F
in
st
b
After placement of steel
reinforcement, a mixture
                                                                   a
                                                                   !3
                                                                   4s
                                                                   B 3
                                                                   « U!
                                                                   s S.
                                                                   U O.
                                                                   E .b
                                                                   S *
                                                                   s&
                                                                   a -a
                                                                   2"!
                                                                   eb a*
                                                                   52 a.
                                                                   « 0)
             g
             a
             u
             OS'S
             a 3
         •wl   o 'B
         2   P a
         •£   o *
         V|   M —
                           a
                           o
                           •8
                            u
                           Q
00
°a

a
o
'5
w
§

8
.2
a,
.3
                                4-15
-------

                                             a. « a Q.
u
e
o
U
_1
03
I'
3

3
at
u)
z
3
u
a.
53
2
V)
a
O
   2  8
   -.3§1
    Jo .. o
    is?
   •aS"a-§
   !a-'s.
   lllH
   1I.I1J
     r, g-O
     1 3 -
     •2 a o
      o — -"*
      ?! a. *•
      a £ *
      M • 5
                    •  s
                     **
                    .3 * 2
                    •a a 8
                     •9-S*
                     s a u
                     • o C
                                     00 UJ
                                     •I"
                              •3
                                 a ~S S J3 a.
                                 «j .. 2 <«
                                 S S-i'5
                                 3*23
                                             •  -a
                                             w u
                                        0
                                        a.
                                        '
                                     11 r
iirti
                                 O-3
                        «» .2
                      -
1

                      >-3   -  >,.o
                     1:1 2 ill
                            a

                           .2.
                              ._ u
                              > -a

                              S.2
                                                -2
                                  S -c a 3 JJ '= § §
                                  3**--2  oSSu
                                            -a
                                           8 as
                                           '2 Ifl
                                           «a«a § 2
         « S-H g
         as a. S 3
               w —•
               ^ -O
               .8-1-

               -S§
               •2 " C

               ill

               i!l
             s: - .2 — "3
             o o .2 S o
             h» hM u '^ »•
             au a. M vs 5
'« «

§ 5
5 8

8 2
a. 2
3 **
•, -a
•3 a
<• "
J3 O
M O
u '2
•O O
   a.
      U
      .o
      Q
      r
      5(
      «
      a
      II
      ^ 5
      •c >•
      «i 2
      « a.
      s ts
      •5-3
      '•Ok.
      !C "5. a
      Q s-a
                •o
               a^
               u ••
               ii^
               ^ '!! g -2 •=
                     J3 "
                     ii
                  a 3 'u  .
                  M31
                                 a
                                 u
in o
Replacem
                                              •S
                                              5
                                           U 3 3
                                           > '= •« u
                                           fc tt^ u
                                           -o s « c
                                           u •» •» u


                                           !i«
                                           ^3-28

                                           •3 a"8|
                                             2*1
                                               h«
Pan
plac

corr
                                  a
                                  8
                                  <•
                                  "H.
                                  o
                                  tt

                                  I
                                  V)
                         4-16
                                           •a
                                           a

                                           o.
                                           S


                                           1
-------
applications, advantages, and disadvantages.  More detailed information may be found
in references (5),(6),(7), and (8).

      Sliplining or insertion renewal is the most widely used rehabilitation technique.
Sliplining involves inserting a new, continuous length of pipe or segments of pipe inside
an existing pipe.  Pipes are made by joining individual lengths with heat fusion or
various types of bell and spigot joints.  Materials used for sliplining pipes  include
polyethylene (PE), polybutylene, fiberglass-reinforced polyesters, reinforced
thermosetting resins, and PVC.  Material selection is based on application, design needs,
economics, and to some extent,  space availability at the installation working area.
Sliplining is capable of dealing with  a variety of serious structural problems if the
annular space between the existing pipe and the sliplining pipe has been properly
grouted.  Exceptions are severely crushed or collapsed pipes.  Installation  is done by
excavating an  access pit and pushing or pulling the slipline pipe into the existing pipe.
It is widely used for cracked or deteriorated sewer pipes and to a lesser extent, water
pipes.  It takes less time for installation, has lower cost than excavation and
replacement,  and requires minimal excavation and disruption.   It has the ability to
accommodate large radius bends and may improve hydraulics  in spite of the reduction
in the overall  pipe diameter. Most of the sliplining materials  available are acid-resistant
and provide good hydraulics.

      Deformed pipe  insertion  lining is a new process with a  limited experience base.
Several variations exist In general,  folded or compressed plastic pipe is inserted into
the existing pipe.  Pipe material may be polyethylene or plasticized PVC.  The pipe may
be heated to increase flexibility  prior to installation. In one process, the heated pipe is
pulled through a die to reduce the diameter prior  to insertion. In processes using
folded pipe, steam is added after insertion and a ball is propelled through the folded
pipe to expand it to conform with the existing pipe, or the pipe is expanded
hydraulically using steam pressure.   In the process using the mechanical die to reduce
the diameter,  the pipe naturally  reverts to its original diameter within several hours.
After insertion and expansion, the ends of the liner are cut off, trimmed, and likely
sealed.  Proprietors claim that the process results in a  tight fit of the liner to the pipe,
eliminating the need for grouting of the annular space. Mixing of resins and curing, as
required with  cured-in-place systems, are  unnecessary. The folded liner systems are
supplied  in rolls, and insertion pit excavation is not required.  For the system which
reduces pipe diameter using a die, an insertion pit is apparently required as with
conventional sliplining. Because of the limited experience with deformed  pipe insertion
methods for rehabilitating sewers, little information is  available on the applicable size
ranges, ease of installation,  and  cost-effectiveness of these systems.

      Cured-in-place pipe is formed from a resin-impregnated felt tube which is
inverted into  an existing pipe and allowed to cure.  After curing, the felt tube becomes a
smooth, hard pipe of slightly less diameter and of the same shape as the original pipe.
Cured-in-place pipe can be  installed in  pipes of all shapes up  to diameters of 96 inches.


                                        4-17
-------
It can also adhere to bends present in the original pipe. Inversion lining is successful in
dealing with a number of structural problems. Caution must be used in the application
of this method to any structural problems involving major loss of pipe wall, reinforcing
steel, or exterior pipe and bedding support  Within limits, the liners can be designed to
deal with these more serious structural problems. Precautions should be taken in
determining the selection of an inversion lining method for sewer rehabilitation.
Various resins can be used to provide different degrees of acid resistance. Standard
polyester resins are suitable for most sewer applications. Cured-in-place  inversion
linings have found more acceptance where minimal excavation and traffic disruption is
required.  This technique is one of the most widely used rehabilitation methods.  The
long-term acid resistance of the liner is unknown.  Observations made within  the past 15
years have not indicated corrosion problems.

       Specialty concretes are sulfate-resistant cements applied to corroded surfaces
primarily for structural reasons and to resist corrosion. Sulfate resistant cements
include potassium silicate.  Typically, reinforcing steel is added for additional  support
Specialty concretes include cement mortar, shotcrete, and cast concrete.  Cement mortar
is applied with a hand trowel for spot repairs in man-entry size (i.e., greater than 32-
inch-diameter) sewers or with a centrifugal lining machine for complete coverage within
a stretch of pipe.   Shotcrete is applied with a special nozzle using compressed air.  Cast
concrete is installed using prefabricated or hand-built  interior pipe forms. The
development of mechanical, in-line application methods (e.g., centrifugal and mandrel)
has established mortar lining as a viable rehabilitation technique. Both shotcrete and .
cast concrete are  used in large-diameter sewers where adequate space is available to
handle materials and  equipment The specialty concrete applications depend on the
degree of corrosion present and the structural integrity of the unit in question.
Generally, thin film concrete will perform best on relatively non-corroded concrete
whereas an elastic membrane concrete system will work for all cases. Testing performed
by CSDLAC revealed that seven out of eight supposedly acid-resistant specialty
concretes failed after  submersion from 0.1 to 488 days in a 10-percent sulfuric acid
solution.  Only one concrete has maintained its acid resistance even after 605 days of
testing. Specialty concretes are used mostly for spot repairs such  as manhole  barrels,
wet wells, junction chambers, and sections of pipe.

       Coatings include a myriad of proprietary epoxies, resins, sealers, silicones,
urethanes, and coal tars applied by spray or brush.  Coatings are experiencing rapid
growth with new products being marketed continually.  Unfortunately, field testing has
not kept up with the rapid growth.  CSDLAC have  been testing new products, and cities
such as Seattle have been using those products that exhibited good results.  Only about
25 percent of coatings tested by CSDLAC exhibited good acid resistance.  The majority
of failures can be attributed to application difficulties  (e.g., pinholes and blowholes).
All coating systems require some form of surface preparation.  The surface preparation
and conditions under which coatings material is applied are extremely critical. The
specification of any acid resistant epoxy should require the minimum application of 1.5


                                        4-18
-------
mm of material for rehabilitation.  Consideration should be given to the use of coatings
that will cure underwater for projects that require either short down times, or where it
is impractical to completely remove the structure from service, thereby requiring coating
application to an intermittently wet area.

      Liners used for rehabilitation may be prefabricated panels or flexible sheets that
are installed manually with anchor bolts or with concrete-penetrating nails; or
continuous, interlocking strips that are installed in a spiral fashion using a special
machine or manually. Manual methods are applicable only to man-entry  sewers.
Common liner materials are fiberglass-reinforced cement, fiberglass-reinforced plastic,
PVC, and PE.  The liner materials themselves are acid-resistant but problems have
occurred due to poor jointing. Liners may be susceptible to acid leakage  due to
numerous joints.  Some panel  systems are time consuming to install and thus prolonged
bypass is required. Hydrogen  sulfide gases have been documented  to pass through poor
joints and cause failure by attacking the concrete substrate behind the liner.  A recent
design introduced to the U.S. involves a continuous, helical, interlocking strip with
improved joints which may overcome such gas-penetration problems. Also, a new acid-
resistant urethylene mastic has shown excellent results in bonding PE and PVC sheets
to concrete surfaces,  and may  eliminate problems with mechanical anchoring and poor
jointing.  Cracking of polyethylene liners has been observed in areas of high turbulence.

4.5    Conclusions

      No material or technique  is effective for controlling sulfide generation or sulfide-
induced corrosion in every situation. The environmental variables that determine the
success of a prevention or repair method (e.g., the characteristics of the wastewater and
the collection system and the severity of corrosion) vary with each system  and within
each reach of sewer.  It is only after a system has been analyzed that these variables can
be taken into account and an effective measure selected. Often,  several different
methods must be used in combination or at different points in the same system to
combat corrosion under various conditions. The effectiveness of a method is deter-
mined by more than  just its physical properties and/or theory of operation. Proper
design, installation, operation, and maintenance are  required to ensure that the material
or technique is  effective.  However, even if studies indicate that a method has the best
long-term cost-  effectiveness, initial costs may exceed the budgetary constraints of a
municipality and force it to use a less expensive and less effective method. Because  of
the variations in effectiveness, affordability, availability, convenience, and  applicability,
sulfide control techniques  and corrosion rehabilitation methods must be evaluated on a
case-by-case basis.
                                       4-19
-------
                                 REFERENCES
1.     "Design Manual - Odor and Corrosion Control in Sanitary Sewers and Treatment
      Plants," EPA/625/1-85/018, EPA, Cincinnati, OH,  1985.

2.     "Sulfide in Wastewater Collection and Treatment  Systems," ASCE Manuals and
      Reports on Engineering Practice - No. 69, ASCE, New York, 1989.

3.     "Sulfide and Corrosion Prediction and Control." American Concrete Pipe
      Association, Vienna, VA, 1984.

4.     Stahl, J.S., Redner, J., and R. Caballero, "Sulfide  Corrosion in the Sewer System
      of Los Angeles County," presented at llth U.S./Japan Conference on Sewage
      Treatment Technology,  the Public Works Research Institute, Tsukuba, Japan,
      October, 1987; and ASCE Conference on Sulfide  Control in Wastewater
      Collection and Treatment Systems, Tucson, AZ, February, 1989.

5.     "No-Dig Technology Outline,"  prepared by National Association of Sewer Service
      Companies, Altamonte, FL,  1989.

6.     Schrock, B.J., "Pipeline  Rehabilitation Seminar," Portland, Maine, August 8,
      1988.

7.     U.S. Department of Housing and Urban Development, Utility Infrastructure
      Rehabilitation: Office of Policy Development and Research  - Building
      Technology, November, 1984.

8.     Redner, John, A., "Evaluation  of Protective Coatings for Concrete," presented at
      the 59th Annual  Water Pollution Control Federation Conference, Los Angeles,
      CA, October 7, 1986.

9.     Cronberg, Andrew  T., Jack P.  Morris, Ted  Price,  "Determination of Pipe Loss
      Due to Hydrogen Sulfide Attack on Concrete  Pipe," prepared for the 62nd
      Annual Water Pollution Control Federation Conference,  San Francisco, October,
      1989.

10.   "Concrete Pipe Handbook," American Concrete Pipe Association, Vienna,
      Virginia.  1988.
                                      4-20
-------
         APPENDIX A

 ANNUAL AVERAGE WASTEWATER
CHARACTERISTICS FOR LA COUNTY
           1971-1986
-------
                                                July 1987






            JOINT WATER POLLUTION CONTROL PLANT



                   RAW SEWAGE PARAMETERS



                1971 - 1986 YEARLY AVERAGES




(Based on Water Quality Characteristics Monitoring Program)
iar
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
Alkalinity
Total (mg/1)
307
302
316
298
289
307
330
322
316
314
317
323
340
338
329
339
Arsenic
(mg/1)
0
0
0.0250
0.0354
0.0155
0.0073
0.0114
0.0135
0.0188
0.0064
0.0067
0.0079
0.0087
0.0257
0.0180
0.0101
Barium
(mg/1)
-
-
0.53
0.55
0.75
1.07
0.91
0.78
0.67
1.02
0.80
0.82
0.91
0.83
1.03
0.97
BOD
Total
(mg/1)
384
319
357
314
302
306
334
324
322
335
322
313
291
317
329
328
Boron
(mg/1)
1.03
1.11
1.14
1.35
1.49
1.54
1.51
1.64
1.50
1.52
1.41
1.66
1.68
1.76
1.72
1.58
-------
JWPCP - Raw Sewage Parameters
1971 - 1986 Yearly Averages
Page 2

Year
1971
1972
1973
1974
1975
1976
1977
1978
1979
1980
1981
1982
1983
1984
1985
1986

Cadmium
(mg/1)
0.0250
0.0320
0.0320
0.0400
0.0390
0.0310
0.0343
0.0385
0.0358
0.0343
0.0244
0.0206
0.0337
0.0199
0.0180
0.0140

Chloride
(mg/1)
560
502
423
365 .
341
345
326
397
387
387
408
434
453
498
460
461
Chromium
Hexavalent
(mg/ij
0
0
0
0
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
0
0
0
0
0
Chromium
Total
(mg/i)
0.780
1.125
0.877
0.887
1.020
0.905
0.720
0.502
0.445
0.563
0.430
0.335
0.278
0.250
0.237
0.189
COD
Soluble
(mg/1)
326
250
310
251
252
265
244
263
273
246
230
266
262
259
246
257
-------
7PCP  - Raw Sewage  Parameters
 71 - 1986 Yearly  Averages
Page 3


ar
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
COD
Total
(mg/1)
680
721
925
344
818
1022
863
923
910
813
759
916
904
900
881
879
Conductance
MMHO
cm
2992
2813
2785
2369
2145
2162
2353
2185
2312
2273
2355
2524
2582
2679
2632
2620

Coooer
(mg/1)
0.450
0.736
0.563
0.635
0.580
0.430
0.430
0.360
0.337
0.334
0.268
0.230
0.245
0.240
0.197
0.179

Cyanide
fmg/^i)
0.200
0.293
0.363
0.430
0.280
0.290
0.240
0.180
0.178
0.118
0.080
0.063
0.042
0.040
0.020
0.022
DDT
Total
(mg/1)
0.01527
0.02132
0.01802
0.00278
0.00172
0.00325
0.00211
0.00273
0.00234
0.00177
0.00172
0.00076
0.00049
0.00096
0.00037
0.00019
-------
JWPCP - Raw Sewage Parameters
1971 - 1986 Yearly Averages
Page 4
Year
1971
1972
1973
1974
1975
1976
1977
1978
1979
1980
1981
1982
1983
1984
1985
1986
Detergents
(MBAS)
(rag/1)
7.23
7.40
7.49
6.60
6.50
6.74
6.31
8.17
7.62
6.97
5.97
6.20
6.67
6.24
6.44
6.10
Flow
(MGD)
372
351
359
347
342
353
330
345
367
374
364
360
353
352
361
364
Fluoride
(rag/1)
1.05
1.24
1.51
1.40
1.37
1.48
1.45
1.56
1.66
1.56
1.51
1.39
1.99
2.61
2.17
2.01
Hardness
Calcium
(mg/1)
273
259
215
215
176
165
197
197
191
191
204
194
210
211
192
183
Hardness
Magnesium
(mg/1)
165
162
143
109
81
79
97
105
94
82
97
94
101
104
99
98
-------
tfPCP - Raw Sewage  Parameters
)71  - 1986 Yearly  Averages
Page 5


ar
71
72
73
74
75
76
77
78
79
80
81
32
83
84
85
86
Hardness
Total
(mg/1)
438
421
371
350
282
259
294
265
298
294
296
289
307
310
292
287

Iron
(mg/1)
13.130
11.340
8.875
9.480
13.950
6.840
8.290
7.562
7.367
6.429
6.106
5.430
7.840
5.290
4.870
5.138

Lead
(mg/1)
0.280
0.306
0.292
0.371
0.370
0.272
0.340
0.270
0.225
0.212
0.16-4
0.160
0.150
0.140
0.129
0.155

Lithium
(mg/1)
0.070
0.070
0.060
0.050
0.042
0.058
0.100
0.059
0.061
0.053
0.063
0.066
0.063
0.071
0.071
0.065

Manganese
(mg/1)
0.11
0.12
0.11
0.12
0.15
0.10
0.12
0.13
0.11
0.10
0.09
0.09
0.12
0.11
0.11
0.10
-------
JWPCP - Raw Sewage Parameters
1971 - 1986 Yearly Averages
Page 6
Year
1971
1972
1973
1974
1975
1976
1977
1978
1979
1980
1981
1982
1983
1984
1985
1986
Mercury
(mg/1)
0.0022
0.0010
0.0012
0.0012
0.0014
0.0015
0.0014
0.0014
0.0014
0.0011
0.0009
0.0011
0.0013
0.0011
0.0010
0.0011
Nickel
(mg/i;
0.230
0.310
0.324
0.280
0.280
0.340
0.310
0.342
0.245
0.245
0.210
0.200
0.220
0.150
0.129
0.099
Nitrogen
Ammonia
(mg/1)
80.6
41.3
58.5
34.4
33.6
32.9
33.6
34.4
34.2
34.2
34.0
34.6
35.0
31.6
31.9
34.1
Nitrogen
Organic
(mg/1)
26.7
17.4
20.8
16.6
18.8
18.9
17.9
21.0
19.6
20.2
20.3
19.6
23.1
21.1
19.5
21.1
Oil
and Grease
(mg/1)
-
-
-
-
91
219
124
91
90
76
62
73
69
79
72
64
-------
fPCP  -  Raw  Sewage Parameters                             Page 7
J71 - 1986  Yearly Averages
iar
»71
)72
)73
>74
»75
>76
>77
)78
J79
)80
)81
)82
)83
84
85
86
PCS Total
(mq/1)
0.02126
0.01077
0.01233
0.01685
0.00531
0.00061
0.00242
0.00130
0.00098
0.00092
0.00070
0.00081
0.00061
0.00064
0.00031
0
Phenols
pH (raq/1)
7.92 3.83
7.61 2.51
7.65 4.64
7.55 3.27
7.51 3.74
3.17
3.02
2.69
2.40
2.28
2.51
2.07
2.37
1.93
1.70
2.30
Phosphate
(mg/1)
45.8
59.5
53.9
52.2
47.5
38.6
24.5
13.3
11.5
11.5
10.7
10.6
11.6
12.6
12.4
13.2
Potassium
(mg/1)
20
20
19
19
19
20
19
19
19
16
18
18
18
18
18
17
-------
JWPCP - Raw Sewage Parameters
1971 - 1986 Yearly Averages
       Page 8
                                   Calculated4
                                    Sludge
Calculated4
 Sludge
Year
1971
1972
1973
1974
1975
1976
1977
1978
1979
1980
1981
1982
1983
1984
1985
1986
Selenium
(rag/1)
-
-
-
-
-
0.015
0.016
0.009
0.012
0.010
0.009
0.014
0.016
0.015
0.013
0.017
Silver
(rag/1)
0.0140
0.0170
0.0126
0.0109
0.0140
0.0140
0.0106
0.0191
0.0197
0.0153
0.0150
0.0175
0.0182
0.0170
0.0200
0.0198
Concentration
(mg/1)
-
-
-
58.57
57.57
74.38
76.47
76.61
72.00
78.52
78.39
92.27
115.44
120.99
108.33
107.54
Mass
(Ibs/d)
-
-
-
169,656
164,320
219,122
210,597
220,572
220,524
245,075
238,127
277,223
340,082
355,420
326,353
326,691
Sodium
(mg/1)
369
381
357
336
308
306
336
357
331
333
338
368
369
407
395
378
-------
PGP - Raw Sewage Parameters                             Page 9
71 - 1986 Yearly Averages
ar
71
72
73
74
75
76
77
78
79
30
31
32
33
34
35
36
Solids
Total
(mg/1)
2112
2040
1975
1828
1681
-
-
-
«•
-
-
-
-
-
-
-
Sulfate
(mg/1)
347
330
349
320
258
224
260
270
275
275
286
240
282
296
268
285
Sulfide
Total
(mg/1)
0.4
0.5
0.5
0.5
0.5
0.6
1.0
1.2
1.3
1.3
1.6
1.6
1.6
2.0
2.6
3.0/2.0
Sulfide
Dissolved
(mg/1)
0.1
0.2
0.2
0.2
0.2
0.3
0.5
0.6
0.6
0.7
0.7
0. 6
0.6
0.8
1.2
1.4/0.6
Suspended
Solids
(mg/1)
397
416
518
459
484
424
463
448
435
442
442
442
463
455
445
454
-------
JWPCP - Raw Sewage Parameters
1971 - 1986 Yearly Averages
Page 10
Year
1971
1972
1973
1974
1975
1976
1977
1978
1979
1980
1981
1982
1983
1984
1985
1986
TDS
(mg/1)
1716
1624
1458
1368
1199
1191
1300
1404
1262
1259
1338
1318
1474
1523
1522
1426
Thiogyanatie
(mg/1)
-
-
2.76
2.00
1.80
2.00
1.89
2.34
2.11
1.18
1.27
1.53
1.50
1.51
1.09
1.42
TICK
(mg/1)
0.03680
0.01986
0.03032
0.02157
0.00724
0.00456
0.00459
0.00470
0.00316
0.0030.2
0.00281
0.00227
0.00166
0.00318
0.00088
0.00040
Zinc
1.930
2.269
2.470
1.990
1.640
1.420
1.460
1.278
1.000
0.952
1.012
•0.960
1.011
0.830
0.720
0.751
-------
?CP  - Raw Sewage Parameters                             Page 11
71 - 1986 Yearly Averages
:es:
   1.   The values in the tables are based on the Water Quality
       Characteristics Monitoring Program and were used for
       the correlation with dissolved sulfide.  The values,
       in some cases,  do not exactly match the graphs and
       slides previously prepared.  Some of the data for the
       graphs and slides were from sampling conducted for the
       Industrial Waste Section.  However, the industrial waste
       data is incomplete,  so to be consistent, only data from
       the Water Quality Characteristics Monitoring Program
       were used for the correlation.

   2.   Total solids and pH  were not correlated with dissolved
       sulfide because of insufficient data.

   3.   Sludge mass was not  correlated with dissolved sulfide
       because it doesn't account for any change in flow.
       Sludge concentration was used instead.
-------
                                                       February  1988
 _nk
 1
 2
 3
 4
 5
 6
 7
 8
 9
 .0
 .1
 .2
 .3
 .4
 L5
 L6
 L7
 L8
 L9
 20
 21
 22
 23
 24
 25
 26
 27
 28
 29
 30
 31
 32
 33
 34
 35
 36
 37
 38
 39
 40
 41
42
43
44
45

ote:
              CORRELATION OF JWPCP RAW SEWAGE PARAMETERS
                WITH DISSOLVED SDLFIDE FROM 1971-1987
     Parameters
     Nickel
     Chromium-Total
     Sludge-Concentration (theoretical)
     Zinc
     Copper
     Cyanide
     Lead
     Iron
     Potassium
     Silver
     Fluoride
     Phenols
     Alkalinity-Total
     Phosphate
     Sodium
     TICH
     PCB-Total
     Barium
     DDT-Total
     Boron
     Oil and Grease
     Mercury
     Nitrogen-Ammonia
     Hardness-Calcium
     COD-Soluble
     Cadmium
     Detergent (MBAS)
     Hardness-Magnesium
     Selenium
     Sulfate
     Flow
     Maganese
     Thiocyanate
     COD-Total
     Hardness-Total
     Chloride
     Chromium-Hexavalent
     Lithium
     Nitrogen-Organic
     Suspended Solids
     Conductance
     Temperature
     TDS
     Arsenic
Correlation
Coefficient
  -0.83
  -0.81
   0.78
  -0.78
  -0.77
  -0.77
  -0.76
  -0.70
  -0.69
   0.69
   0.68
  -0.67
   0.67
  -0.66
   0.63
  -0.61
  -0.60
   0.57
  -0.53
   0.51
  -0.49
  -0.48
  -0.43
  -0.43
  -0.42
  -0.40
  -0.39
  -0.38
   0.37
  -0.33
   0.33
  -0.32
  -0.28
   0.28
  -0.21
   0.20
  -0.13
   0.12
   0.11
   0.07
   0.05
  -0.05
   0.04
   0.02
  -0.02
All parameters correlated with dissolved sulfide without sulfide
control except for Iron which was correlated with dissolved
sulfide with sulfide control in 1986 and 1987.
-------