United States
Environmental Protection
Agency
Robert S. Kerr Environmental Research.
Laboratory
Ada OK 74820
Research and Development
EPA-600/S2-83-051 Sept 1983
Project Summary
Removal of Metals in Combined
Treatment Systems
James W. Patterson, Prasad Kodukula, and Toshiro Aratani
This project assessed the variables
influencing 'the removal of metals
through combined industrial-municipal
treatment plants. The metals investi-
gated were: aluminum, cadmium, chro-
mium, copper, iron, lead, nickel, and
zinc. The metals were studied at sub-
toxic influent concentrations, and the
interrelationships which influence
metal removal were assessed.
The research was performed in two
phases. Phase I involved batch studies
on raw sewage and activated sludge, to
identify and define the impact of in-
dividual chemical and physical param-
eters on metals removal. These batch
studies consisted of three parts. In Part
A, metal solubility in filtered raw sew-
age and secondary effluent was deter-
mined as a function of pH. Part B
investigated the equilibrium sorption
of the test metals onto primary sewage
solids and onto activated sludge solids.
In Part C, the effect of sewage variables
such as detergent and ammonia con-
centration on metal sorption was eval-
uated. In Phase II, eight pilot treatment
plants, each consisting of primary clar-
ifier, aeration basin, and secondary
clarifier, were operated at varying in-
fluent metal levels to study the effect
of significant variables indicated from
the Phase I results.,
The results of this project indicate
that the removal of metals in combined
industrial-municipal treatment systems
is influenced by a number of waste-
water and treatment plant operation
characteristics. The principal param-
eters controlling metal partitioning for
each metal investigated were the total
metal and suspended solids concen-
trations. The segregation of influent
metals between the sludge (primary
and secondary) phases and the plant
effluent can be predicted, based upon
the relationships identified in this
study.
This Project Summary was developed
by EPA's Robert S. Kerr Environmental
Research Laboratory, Ada, OK, to an-
nounce key findings of the research
project that is fully documented in a
separate report of the same title (see
Project Report ordering information at
back).
Introduction
In recent years, increasing attention has
been focused on the chemistry, biological
effects, treatment fate, and control of
heavy metals in the environment Findings
include the discovery of heavy metals at
high concentrations in surface waters re-
ceiving municipal and industrial waste
discharges containing such metals, cou-
pled with the recognition of potential
health hazards and adverse environmental
impacts associated with major disposal
methods for metal-laden municipal and
combined sludges. While the manage-
ment of metals originating directly from
industrial discharges has been imple-
mented under effluent limitations guide-
lines and National Pollutant Discharge
Elimination System (NPDES) permits, the
control of industrial plus non-industrial
metals entering combined municipal-
industrial public-owned treatment works
(POTWs) has been found to be much more
difficult As a result heavy metals dis-
charge into the municipal sewage treat-
ment systems and their fate during the
sewage treatment processes have become
subjects of considerable interest
Most studies to date concerning heavy
metals in sewage treatment processes
have involved attempts to perform mass
balances of metals around a POTW and
determine percentage of removal of each
-------
metal of concern across that P07W How-
ever, there is a relative lack of information
on the actual mechanisms affecting the
distribution of heavy metals between liquid
and solid phases through a municipal
sewage treatment plant Such an under-
standing is essential for developing cri-
teria that can be used to predict the
distribution of heavy metals through com-
bined sewage treatment systems.
Methods and Procedures
As indicated above, this investigation
was divided into four parts, and a brief
description of each part of the project is
given here:
IA. Batch studies on tap water, filtered
raw sewage, and filtered conventional acti-
vated sludge mixed liquor, to determine
the solubility limits of the eight metals.
IB. Batch studies on raw sewage and
activated sludge mixed liquor, to develop
sorption data for selected metals.
1C. Batch studies on raw sewage and
conventional activated sludge mixed liquor
to investigate the influence of both domes-
tic and industrial waste constituents on
metals distribution between the soluble
and solid phases.
II. Continuous-flow pilot-scale conven-
tional activated sludge studies to evaluate
the effect of variables such as total metal
concentration, total volatile suspended
solids (TVSS), soluble organic carbon
(SOC), and major inorganic ligands, on
heavy metals distribution in different pro-
cess liquids.
In Part IA of the study, solubility of
metals at different pH levels was deter-
mined for tap water, raw sewage, and
activated sludge mixed liquor.
Tap water used in this study came from
Chicago's city water distribution system,
while the raw sewage and activated sludge
mixed liquor were obtained from the West-
Southwest Wastewater Treatment Plant
operated by the Metropolitan Sanitary Dis-
trict of Greater Chicago. Batch experi-
ments were performed for each test liquid
(tap water, raw sewage, and mixed liquor).
Initially, the test liquid was filtered using
a0.45-micron membrane filter. Raw sew-
age and activated sludge mixed liquor
were settled prior to membrane filtration,
to enhance membrane filtration efficiency.
Initial pH levels of 6, 7. 8, and 9 (± 0.3
units) were established in units of each
group by pipetting sodium hydroxide or
nitric acid into the test liquid, as required,
with constant stirring. Following pH ad-
justment the appropriate concentrated
metal solution was pipetted into the test
liquid. Simultaneously, pH adjustment
was made to maintain the target test pH
level. Metal solution was added until a
visible precipitate formed and remained
after one minute of continuous stirring.
The sample was continuously stirred
during the metal addition step, and the pH
was monitored.
The test vessels were sealed with para-
film and placed on a shaker with con-
tinuous snaking at ambient temperature.
After two hours, all units were readjusted
to correct for any pH change. Aliquots of
the test liquids were taken at 6,12, and 24
hours for measurement of pH, soluble
metal, and SOC Background analyses on
the test liquids included pH, total dis-
solved solids (TDS), total volatile dissolved
solids (TVDS), initial SOC, background
metals, sulfide, sulfate, total phosphorus,
orthophosphate, ammonia, hardness, and
alkalinity.
In Part IB, sorption of metals onto
sludge was studied by measuring the
amount of metal associated with the sludge
fraction after the metal is added to the test
liquid, at a level below its solubility limit as
determined in Part IA
Batch experiments were set up in a
similar fashion to that described in Part IA
In this component of the project unfiltered
samples were taken, their pH adjusted to
the desired levels, and the selected metals
added. The amount of metal added was
below its solubility limit, to avoid precipi-
tation. After the metal addition, the sam-
ples were constantly stirred, and aliquots
of samples were taken at 0.25,0.50,1,3,
6, and 24-hour time intervals, to measure
pH and soluble metal concentration. The
samples from the 24-hour test period
were also analyzed for total organic carbon
(TOC), SOC, inorganic carbon, total sus-
pended solids (TSS), TVSS, TDS, and
TVDS, total phosphorus, orthophosphate,
and alkalinity.
Part 1C was designed to investigate, in
depth, the influence of domestic and in-
dustrial waste constituents on the dis-
tribution of heavy metals between the
soluble and solid phases of raw sewage
and activated sludge mixed liquor. This
objective was accomplished by spiking
aliquots of test liquids with each selected
waste constituent and determining how
the phase distribution of metals was
affected.
Seven different domestic/industrial
waste parameters, at three levels for each
parameter, were tested. For each param-
eter tested, a series of seven different
metals combinations were evaluated. Metal
Combinations 1 through 4 consisted of
mixtures of eight metals at low (Com-
bination 1) to high (Combination 4) rel-
ative concentrations. Metal Combinations
5 and 6 were replicates of Combination 3,
providing a statistical basis for the eval-
uation of experimental results. In Metal
Combinations 7 and 8, the metal levels
were varied randomly (i.e., some metals
were at high and others at low concen-
trations). Random metal combinations
were incorporated in the studies in order
to determine whether interactive effects
upon metal removal result from preferen-
tial removal of specific metals by the
sludge phase. There was a control group
to which no metal was added. All metals
concentrations fell within the range of
typical influent values reported for POTWs.
Each set of raw sewage test samples
was mixed on a shaker table for four hours
at ambient temperature. At the termina-
tion of the mixing period, an aliquot of the
whole fraction of each sample was taken
for analyses. An additional aliquot was
filtered through a 0.45-micron filter to
obtain soluble samples. Since metals in-
fluent to activated sludge units have had
extended contact periods with raw sew-
age, the settled supernatant resulting
from the raw sewage experiments was
utilized as the media for introduction of
metals to the activated sludge samples.
Part 11 of this investigation was designed
to study the distribution of metals in
different process liquids of continuous-
flow pilot-scale conventional activated
sludge systems receiving raw sewage,
spiked with heavy metals at different
concentrations.
The continuous-flow studies of Part II
were divided into several runs, each run
involving eight separate parallel pilot-scale
activated sludge treatment systems. There
were 39 different activated sludge treat-
ments contained in this phase of the
project Table 1 presents the concentra-
tions of different heavy metals in the raw
sewage fed during the 39 different acti-
vated sludge runs. These individual metals
concentrations and combinations were
selected on a random basis, to simulate
low, high, and mixed levels of metals in
raw sewage.
Municipal sewage was pumped from a
City of Chicago sewer line to a laboratory
grit chamber on a continuous basis. Settled
grit was discharged. Raw sewage over-
flowed from the grit chamber into a 300-
gallon stirred holding tank, having an
average six-hour detention time. The hold-
ing tank was equipped with a low level
alarm, to cut off all downstream pumps
and valves (except for return activated
sludge pumps and excess sludge wastage
valves), in the event that the raw sewage
flow was interrupted. The raw sewage was
pumped into a common header, and then
-------
Table 1. Average Influent Metals Concentrations (fig/1) In Raw Sewage Fed to 39 Different
Activated Sludge Systems
Treatment
No. Aluminum Cadmium Chromium Copper Iron Lead Nickel Zinc
1
2
3
4
5
6
7
8
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
783
433
1003
1310
678
298
375
372
932
383
495
500
295
710
678
678
295
677
520
661
983
655
385
785
240
834
890
669
278
567
216
740
778
1574
1193
678
337
693
25
42
140
80
12
124
63
143
60
28
77
105
154
93
59
12
137
88
138
146
53
24
157
135
128
77
57
11
63
69
98
22
222
87
102
11
87
81
135
143
174
630
113
84
150
128
600
155
159
153
122
1062
460
113
97
183
144
500
420
106
137
109
124
513
530
113
.62
90
128
144
253
140
100
113
84
124
393
359
274
280
90
161
177
530
150
429
479
271
460
338
240
90
173
453
625
425
270
308
460
367
325
363
350
90
162
213
180
302
756
1071
210
98
170
269
1265
1542
1750
1460
1399
1247
1292
1610
2675
1439
1641
1521
2220
3360
1534
1399
1576
636
1510
3225
2510
.1378
2243
2492
1488
3200
2350
1399
1527
936
650
1385
2322
2117
1510
1399
1483
671
81
93
293
140
35
37
75
320
150
57
88
158
170
150
90
35
154
267
475
150
140
41
75
221
190
175
180
35
100
143
120
66
200
260
160
35
97
100
672
756
1629
2740
334
369
1780
1220
838
795
1002
869
986
1220
1615
245
352
2983
3263
1678
1263
68.0
653
4008
6075
2050
2132
330
366
490
2050
603
1522
708
319
245
373
619
482
413
1114
826
409
383
481
830
1583
510
617
643
553
1003
1575
409
450
1114
694
1025
1860
564
477
514
766
1463
2160
409
440
429
644
520
536
540
463
409
413
450
into eight parallel dosing tanks of two-
hour detention time each. Selected metals
mixtures were metered into each chemical
dosing tank, in accordance with the ex-
periment underway for that particular treat-
ment system.
Each dosing tank overflowed to a pri-
mary clarifier of a system. The flow rate
was about 130 ml/mn. Primary clarifier
overflow was through a flow splitter, to
control hydraulic loading to the activated
sludge unit Each activated sludge unit
was constructed as a five-chamber, 100-
liter total capacity unit, with removable
partitions to convert from a plug to com-
plete mixed flow mode.
Activated sludge unit mixed liquor over-
flowed by gravity to a secondary clarifier,
where settled sludge was returned by a
peristaltic pump to the activated sludge
unit The recycle ratio used for all activated
sludge units in this study was 1:1. Excess
sludge was wasted directly from the sec-
ondary clarifier, or by intermittent interval
wasting of activated sludge unit overflow,
as was most appropriate for control of
sludge age. Sampling from each unit was
by timer activated solenoid switch flow
diverters, to yield eight-hour composite
samples.
Composite samples of the raw sewage,
primary effluent activated sludge mixed
liquor, secondary effluent primary sludge,
and secondary sludge were collected sev-
eral times each week. Total and soluble
metal analyses were performed on all
process liquid samples, while the sludge
samples were analyzed for total metals. In
addition, pH, suspended solids, andTVSS
were also measured on these samples.
Soluble samples of the four process liquids
were analyzed for SOC, inorganic carbon,
phosphate, sulfate, chloride and ammonia
nitrogen.
Metal analyses were performed by
atomic absorption spectrometry. Total phos-
phorus, orthophosphate, sulfate, chloride,
ammonia calcium, hardness, and alkalin-
ity determinations were performed ac-
cording to procedures described in EPA
Methods. TVSS values are reported as the
weight of the dry solids per liter of sample
retained by a 0.45-micron membrane fil-
ter. Total dissolved solids represented the
dry solids present in the filtrate of one liter
of original sample. Volatile solids are re-
ported as the weight of residue lost upon
ignition at 600°C of one liter of the original
sampla
Results and Conclusions
The following conclusions were drawn
from the Phase IA studies on metals
solubility in filtered raw sewage and aera-
tion basin mixed liquor.
1. At all pH values tested, equilibrium
solubility conditions were achieved within
six to 12 hours. Levels of metal solubility
were equivalent at 24 hours to those
observed at 12 hours.
2. High correlations were observed be-
tween metal solubility and process liquid
pH, for all metals investigated.
3. Within each process liquid, over the
24-hour period of the solubility tests, the
pH in each case shifted from the more
extreme high or low initial pH values
toward a final pH value of about 8. This pH
shift suggests that the process liquids are
well buffered, and the occurrence of more
extreme pH conditions in full-scale treat-
ment systems would indicate the presence
of strong acid or basic industrial wastes
which would influence metals solubility.
4. A comparison of metals solubility in
filtered process liquids with that in tap
water revealed that in most instances the
process liquids yielded higher metals sol-
ubility than did the tap water. This re-
sponse is probably due to the metal com-
plexation effects of organic and inorganic
ligands in the process liquids.
5. The pH range of minimum metals
solubility, for all metals tested and in both
process liquids, was in the pH range of 8 to
9, except for aluminum in mixed liquor
where a pH of minimum solubility of 6.8
was observed.
The distribution of metals between the
soluble and solids (sludge) phases in raw
sewage and mixed liquor was studied,
with metals added to the test liquids at
concentrations below the metals solubility
limita The following results were observed.
1. A major portion of each added metal
was removed from the soluble phase onto
the solid phase in each test liquid. The
distribution was essentially completed
within a 15-minute contact time although
-------
some minor additional redistribution con-
tinued for up to six hours.
2. Since the metals were added to the
process liquids at concentrations below
their solubility limits, removal from the
liquid phase could not be by precipitation
of metal salts, and therefore was due to
accumulation by sorption onto the raw
sewage and activated sludge solids.
3. The sorption behavior of each metal
could be described by an adsorption iso-
therm relating ftg of sludge metal sorbed
per mg of TVSS, versus total metal present
Although the sorption data generally fol-
lowed this isotherm, the data for most
metals did not fit a standard Freundlich
isotherm, based upon residual metal in
solution.
4. Sorption of added metal in raw sew-
age ranged from 0 to 99%, with the
following ranking of metals from least to
most completely sorbed: iron, nickel, cad-
mium, copper, zinc, lead, chromium. Sorp-
tion of added metal in activated sludge
mixed liquor ranged from 8 to 98%, with
the following ranking of metals from least
to most sorbed: iron, nickel, zinc, cad-
mium, chromium, copper, and lead.
It has been suggested in the literature
that various waste parameters might inr
fluence the distribution of metals in raw
sewage and mixed liquor between the
soluble and solid phases. Constituents
evaluated in Phase 1C were inorganics plus
hardness, detergents, SOC, pH, cyanide,
and ammonia The following conclusions
were drawn, based upon statistical anal-
ysis of the experimental data.
1. Few of the waste constituents, at the
levels tested, had a statistically significant
effect on metals distribution between the
soluble and solid phases.
2. At the 99% confidence level, SOC
influenced aluminum distribution in raw
sewage; pH influenced iron and nickel
distribution in raw sewage; and ammonia
influenced aluminum in mixed liquor.
3. At the 95% confidence level, inor-
ganics and hardness influenced the dis-
tribution of aluminum and lead in raw
sewage, and cadmium and lead in mixed
liquor. Detergent strength influenced the
distribution of chromium and nickel in raw
sewage. In mixed liquor, chromium, iron,
lead, and nickel were indicated to be
influenced. At this confidence level, pH
influenced the distribution of aluminum in
raw sewage and mixed liquor. Ammonia
was indicated to influence the distribution
of cadmium in raw sewage.
During Phase II of the project eight
parallel continuous-flow pilo.t activated
sludge systems were monitored around
each unit process, during a total of 39
runs. Each run was approximately 30 days
in length. Raw domestic sewage, spiked
during each run with random levels of a
mixture of test metals was treated (Table
1). Composite process liquid samples
were collected several times weekly during
each run, for raw sewage, primary clarifier
effluent, mixed liquor, secondary clarifier
effluent and settled primary and secon-
dary sludge analysis Total and filtered
fractions of each sample were analyzed for
metals plus other constituents including
SOC and TVSS. The conclusions devel-
oped from this phase of the project are
comprehensive and are only briefly sum-
marized here.
1. The removal of metals across the
treatment system was directly related to
the degree of distribution of each metal in
the raw sewage and mixed liquor, and the
efficiency of removal of the suspended
solids (and associated metals) in the pri-
mary and secondary clarifiers. Thus, there,
are two principal classifications of vari-
ables which influence metals removal in
combined treatment systems: those as-
sociated with the metals distribution in
each process liquid; and those associated
with the performance of the clarifiers in
solids separation.
2. In some experimental runs, negative
removals of the metals were observed
across the primary clarifiers, and/or the
full treatment systems. These negative
metals removals always resulted from
negative removals of suspended solids in
the primary clarifier. Intermittent negative
removals of suspended solids in primary
clarifier are observed in full-scale systems
as well as pilot units. This negative per-
formance of the primary clarifier in sus-
pended solids removal suggests why some
short-term mass balance studies on full-
scale systems have resulted in negative
full-system removals of metals.
3. Over the course of the 39 experi-
mental runs, a wide range of concen-
trations of influent SOC, TVSS and metals
were observed, reflecting the combination
of natural fluctuations in the raw sewage
composition, plus the spiking of the raw
sewage with metals. Average performance
of the system in solids removal was 76%,
and removal of SOC averaged 61%. Ranges
and averaged values are presented in
Table 2.
4. Ranges of total effluent metals were
also broad, although less so than the
influent metals ranges. However, an eval-
uation of the soluble metals levels re-
vealed that the average soluble concen-
tration, for each metal, remained essen-
tially constant across each unit process
and the entire treatment system (Table 2).
Thus, the reduction of total metals across
the unit processes was due to the sedi-
mentation of solid-bound metal.
5. The total metal concentrations in the
activated sludge aeration basin were much
higher than was observed in any other
process liquid. However, the soluble metals
levels in all process liquids were equiva-
lent and the higher total metals levels in
the mixed liquor resulted due to higher
levels of suspended solids and their asso-
ciated metals
6. Relatively wide variation in the total
metals discharged in the secondary efflu-
ent resulted from variation in effluent
suspended solids; the effluent soluble
level of each metal was comparable to the
raw sewage soluble level of that metal.
7. The relative contribution of the sol-
uble fraction of the effluent metals ranged
from a low 2.9% for chromium up to
34.1% for nickel. Increased secondary
clarifier efficiency in suspended solids
removal would reduce only the non-soluble
portion of the effluent metals.
8. The averaged removal of metals in
the primary clarifier ranged from 14.0%
for zinc to 41.1 % for iron, and the metals
ranked from lowest to highest removal in
the primary clarifier were: zinc, copper,
cadmium, aluminum, chromium, lead, iron,
nickel.
9. The averaged removal of metals
across the activated sludge process plus
secondary clarifier ranged from 1.3% for
aluminum to 38.9% for cadmium, and the
metals ranked from lowest to highest as
follows: aluminum, chromium, nickel, iron,
zinc, copper, lead, cadmium.
10. The averaged overall removals of
metal across the entire treatment system
ranged from 27.6 for aluminum to 54.996
for lead, with the metals ranked from
lowest to highest removal as follows:
aluminum, zinc, chromium, copper, iron,
nickel, cadmium, lead.
11. For the metals aluminum, chro-
mium, iron, and nickel, the bulk of overall
removal occurred in the primary clarifier.
For the metals cadmium and copper, the
secondary processes accounted for the
majority of overall removal. Removals of
lead and zinc were about equally distrib-
uted between the primary and secondary
stages.
1 2. Although the experimental data of
the 39 runs can be fitted to adsorption
isotherms, a more striking and significant
relationship was identified on the basis of
the data generated from the continuous-
run pilot units. This relationship reveals
that the concentration of each metal sorbed
on the solids of each process liquid was
directly related to total metal, and was
-------
Table 2. Overall Averages and Ranges for Different Parameters in Different Test Liquids*
Raw Primary
Parameter Sewage Effluent
VSS
SOC
Aluminum
Cadmium
Chromium
Copper
Iron
Lead
Nickel
Zinc
Total
Soluble
% Soluble
Total
Soluble
% Soluble
Total
Soluble
% Soluble
Total
Soluble
% Soluble
Total
Soluble
% Soluble
Total
Soluble
% Soluble
Total
Soluble
% Soluble
Total
Soluble
% Soluble
Avg.
Range
Avg.
Range
Avg.
Range
Avg.
Range
Avg.
Avg.
Range
Avg.
Range
Avg.
Avg.
Range
Avg.
Range
Avg.
Avg.
Range
Avg.
Range
Avg.
Avg.
Range
Avg.
flange
Avg.
Avg.
flange
Avg.
Range
Avg.
Avg.
flange
Avg.
Range
Avg.
Avg.
Range
Avg.
Range
Avg.
62
2-460
28
3-294
652
63-5100
81
1 1-425
12.4
85
3-650
16
1-305
18.8
241
18-1700
4.2
2-17
1.7
330
1 1-2900
17
'1-157
5.2
1778
200-7000
118
5-783
6.6
142
0-1069
24
2-197
16.9
1349
22-8500
319
8-1168
23.6
741
100-5000
90
2-1000
12.1
36
1-196
19
1-106
478
24-3032
79
8-375
16.5
72
2-514
14
1-295
19.4
170
5-650
4.0
2-9
2.4
281
3-913
12
1-100
4.3
1247
200-3500
97
5-842
7.8
100
0-600
27
2-248
27.1
794
5-15000
297
9-1479
37.4
637
80-3400
74
1-430
11.6
Mixed
Liquor
1307
150-8106
14
1-200
7179
526-21000
61
0-325
0.8
411
4-810
15
1-98
3.6
1292
10-3150
4.0
2-9
0.3
3215
4-8500
14
1-96
0.4
28184
1048-8400
70
3-885
0.2
1971
1 1-9000
24
2-474
1.2
6602
77-23000
290
5-975
4.4
11589
1000-36000
79
2-900
0.7
Secondary
Effluent
15
1-220
11
1-38
472
67-2732
83
5-350
17.6
44
2-382
13
1-67
29.5
162
31-1600
3.9
2-5
2.4
210
11-1866
14
1-50
6.9
1089
100-5800
52
3-580
4.7
64
0-1200
18
2-211
28.3
733
10-5000
250
3-849
34.1
514
100-4100
65
1-900
12.6
* VSS and SOC expressed as mg/l,
metals concentrations as ftg/l.
inversely related to TVSS present In other
words, at constant suspended solids, the
metal per unit of solids increased with
increasing total metal. However, at con-
stant total metal, the metal per unit of
solids increased with decreasing sus-
pended solids concentration. Figure 1
presents one example, for nickel distribu-
tion in raw sewage.
13. A number of models were assessed
for their accuracy in predicting the dis-
tribution of metals in each process liquid,
between the soluble and solid phases. An
investigation of the influence for the total
metal concentration of the parameters
VSS, SOC, and pH revealed that a model
•vhich related total metal to sludge-bound
metal per unit weight of VSS and to VSS
solids in the process liquid provided an
accurate prediction tool for metals dis-
tribution. This model has been designated
as Metals Distribution Model 3 in this
report and model coefficients for each
metal in each process liquid were derived.
Model 3 yielded high correlation coef-
ficients with the experimental data on each
process liquid and each metal, ranging
from a squared coefficient of 0.80 for
nickel to 0.99 for chromium in raw sew-
age, and coefficients of 0.99 for all metals
in mixed liquor (Table 3). At moderate to
high suspended solids levels, a simplified
model (termed Model 4) which directly
relates total metal to sludge-bound metal is
equally accurate, and Models 3 and 4 have
been utilized as the basis fora model of the
full-treatment system.
This project has resulted in the develop-
ment of a full-system model to predict the
removal of metals at each unit process
across a combined treatment system. The
full-system model relies upon submodels
for (1) metals distribution, by process
liquid, (2) primary clarifier performance in
suspended solids removal, and (3) sec-
ondary treatment system performance in
terms of sludge yield, and secondary
clarifier performance.
A comparison of the full-system model
to pilot-plant experimental data revealed
that where the full-system model was
-------
7201-
I
I
"5
1
I
,3
to
80
40
31r
24,
Figure 1.
100mg/l
0.8
3.2
1.6 2.4
Total Metal Concentration, mg/l
Adsorption isotherms of nickel in raw sewage at different VSS concentrations.
4.0
Table 3. Squared Correlation Coefficients for Metals Distribution Model 3
Process Liquid
Metal
Aluminum
Cadmium
Chromium
Copper
Iron
Lead
Nickel
Zinc
Raw
Sewage
0.959
0.970
0.999
0.996
0.989
0.877
0.803
0.953
Primary
Effluent
0.749
0.837
0.999
0.989
0.984
0.840
0.560
0.914
Mixed
Liquor
0.999
0.997
0.999
0.999
0.998
0.997
0.986
0.999
Secondary
Effluent
0.852
0.720
0.999
0.992
0.949
0.826
0.909
0.814
inaccurate, it failed through an inability to
track the short-term solids balance around
each unit process. These unit processes,
while performing in a predictable fashion
on a long-term average basis, perform in a
more erratic fashion over short periods of
days to weeks, sometimes exhibiting, for
example, negative suspended solids re-
moval in the primary clarifier or short-term
interruptions in activated sludge yield.
Metals removals are closely tied to the
solids balances around the unit processes
of the treatment system, and improved
models to predict the short-term behavior
of the systems in terms of solids are
necessary before more accurate short-
term modeling of metals dynamics will be
possible.
James W. Patterson, Prasad Kodukula, and Toshiro Aratani are with the Illinois
Institute of Technology. Chicago, IL 60616.
Thomas £. Short. Jr., is the EPA Project Officer (see below).
The complete report, entitled "Removal of Metals in Combined Treatment
Systems," (Order No. PB 83-226 076; Cost: $22.00, subject to change) will be
available only from:
National Technical Information Service
5285 Port Royal Road
Springfield, VA 22161
Telephone: 703-487-4650
The EPA Project Officer can be contacted at:
Robert S. Kerr Environmental Research Laboratory
U.S. Environmental Protection Agency
P.O. Box 1198
Ada, OK 74820
-------
United States
Environmental Protection
Agency
Center for Environmental Research
Information
Cincinnati OH 45268
Postage and
Fees Paid
Environmental
Protection
Agency
EPA 335
Official Business
Penalty for Private Use $300
------- |