ENVIRONMENTAL HEALTH
SERIES
Water Supply
and Pollution Control
METALS AND BIOLOGICAL
SEWAGE TREATMENT
PROCESSES
' U. S. DEPARTMENT OF HEALTH,
EDUCATION, AND WELFARE
Public Health Service
-------
INTERACTION OF HEAVY METALS
AND
BIOLOGICAL SEWAGE TREATMENT PROCESSES
Chemistry and Physics Section
Basic and Applied Sciences Branch
Robert A. Taft Sanitary Engineering Center
U.S. DEPARTMENT OF HEALTH, EDUCATION, AND WELFARE
Public Health Service
Division of Water Supply and Pollution Control
Cincinnati, Ohio
May 1965
-------
The ENVIRONMENTAL HEALTH SERIES of reports was
established to report the results of scientific and engineering
studies of man's environment: The community, whether urban,
suburban, or rural, where he lives, works, and plays; the air,
water, and earth he uses and re-uses; and the water he pro-
duces and must dispose of in a way that preserves these natural
resources. This SERIES of reports provides for professional
users a central source of information on the intramural research
activities of Divisions and Centers within the Public Health Ser-
vice, and on their cooperative activities with State and local
agencies, research institutions, and industrial organizations.
The general subject area of each report is indicated by the two
letters that appear in the publication number; the indicators are
WP - Water Supply
and Pollution Control
AP - Air Pollution
AH - Arctic Health
EE - Environmental Engineering
FP - Food Protection
OH - Occupational Health
RH - Radiological Health
Triplicate tear-out abstract cards are provided with re-
ports in the SERIES to facilitate information retrieval. Space is
provided on the cards for the user's accession number and
key words.
Reports in the SERIES will be distributed to requesters, as
supplies permit. Requests should be directed to the Division
identified on the title page or to the Publications Office, Robert
A. Taft Sanitary Engineering Center, Cincinnati, Ohio 45226.
Public Health Service Publication No. 999-WP-22
-------
FOREWORD
Research investigations concerning the interaction of
metallic wastes on the biological sewage treatment processes
were conducted at the Robert A. Taft Sanitary Engineering
Center, Cincinnati, Ohio. Ten papers describing these results
have been compiled into this report, which is to be used as a
ready reference source.
While these assembled papers represent many man-years
of time and effort, we do not consider the study to be exhaustive
and believe that additional investigations are required. This
volume is simply our best estimate of the response of the
biological purification processes employed by municipalities
to the metallic wastes commonly discharged to the treatment
plants.
The work reported has been under the immediate supervision
of three project leaders: Messrs. W. A. Moore, G. N. McDermott,
and E. F. Earth. Through the investigations the need of careful
pilot-plant operation and sustained analytical observation of
plant performance have required the patient and sustained skills
of many people. The assignments have been difficult, and
we wish to recognize the proficiency and dedicated efforts of
the following crew members: Messrs. J. N. English, B. V.
Salotto, B. N. Jackson, W. E. Tolliver, H. E. Thomas, R. G.
Santangelo, and J. L. Hinchberger.
In the field investigations, the project has been greatly
aided by the vigorous interest of a group of progressive and
cooperative plant operators. With gratitude and deep appreciation,
we acknowledge the assistance of Mrs. D. Voshel, Grand Rapids,
Michigan, Messrs. M. Phillips, Bryan, Ohio; W. E. Ross,
Richmond, Indiana; and P. R. Carlson, Rockford, Illinois.
The entire project was coordinated by M. B. Ettinger,
Chief, Chemistry and Physics Section, Basic and Applied Sciences
Branch, Division of Water Supply and Pollution Control.
We are indebted to the editors of the Journal of Water
Pollution Control Federation and Proceedings of Purdue In-
dustrial Waste Conference in which many of the papers were
published, and to the Natural Resources Institute, Ohio State
University.
Paul W. Kabler, M.D.
Acting Deputy Chief
Basic and Applied Sciences Branch
Division of Water Supply and
Pollution Control
-------
CONTENTS
ABSTRACT ix
INTRODUCTION 1
I. CHROMIUM 3
Plant Description and Operation 4
New Design 6
Operation ?
Sewage Feed 8
Analytical Methods 8
Sample Collection and Preservation 10
Experimental Data 10
Continuous Feeding 10
Slug Dose Feeding 13
Chromium Distribution and Recovery 15
Chromium Removal with Biological Reductor .... 17
Sludge Digestion 17
Test Procedures and Results 17
Filterability of Digested Sludge 23
Discussion of Results 25
II. COPPER 27
Copper Sources and Form 27
Plant Operation 28
Sample Collection and Preservation 29
Analytical Methods 30
Biochemical Oxygen Demand 30
Chemical Oxygen Demand 30
Copper 30
Cyanide 30
Experimental Data 31
Continuous Feeding 32
Copper Sulphate 32
Cyanide Complex 35
Slug Doses 41
Copper Sulphate 41
Cyanide Complex 45
Discussion of Results 46
Anaerobic Sludge Digestion 49
Procedure 49
Results 51
Discussion 58
Summary 58
-------
III. ZINC 61
Plant Description and Operation 61
Zinc Sources and Form in Liquid Wastes 63
Sample Collection and Analysis 64
Zinc and-Activated-Sludge Treatment 65
Continuous Feeding 65
Slug Doses 71
Zinc and Sludge Digestion 73
Summary 77
IV. NICKEL 79
Plant Description and Operation 79
Nickel Source and Form in Liquid Wastes 81
Sample Collection 81
Analytical Methods 82
Biochemical Oxygen Demand 82
Chemical Oxygen Demand 82
Nickel 82
Nickel and Activated-Sludge Treatment 83
Continuous Nickel Addition 83
Slug Dose 90
Nickel and Anaerobic Digestion 92
Discussion 93
Summary 95
V. A MIXTURE OF HEAVY METALS 97
Metal Combinations Employed 97
Experimental Conditions 93
Analytical Methods 100
Results 100
Effects on Aerobic Efficiency 100
Effects on Nitrification 105
Distribution of Metals 108
Anaerobic Digestion of Sludges 110
Discussion 113
Summary 115
VI. SUMMARY OF PILOT-PLANT DATA 117
Effects on Aeration Phase 117
Distribution of Metals Through the Process 123
Effects on Anaerobic Digestion 126
Suitability of Final Effluent 128
Discussion 129
VH. HEAVY METALS IN WASTE-RECEIVING SYSTEMS. . . . . . 131
Sewage Treatment Process Reaction to Metals 135
Harmful Slug Dose 138
VIII. ORGANIC LOAD AND TOXICITY OF COPPER TO
ACTIVATED-SLUDGE PROCESS 139
Procedure [ 13g
Sample Collection and Analytical Methods 140
Results and Discussion 141
Organic Loadings Obtained on Pilot-Plant Units . . 141
Effects on Aerobic Treatment 143
Fate of Copper 147
vi
-------
Summary 149
DC. A SLUG OF CHROMIC ACID PASSES THROUGH
A MUNICIPAL TREATMENT PLANT 151
Conduct of Study 151
Plant Description 151
Arrangements for Slug 154
Sampling Procedure 154
Analytical Methods 154
Results 155
Chromic Acid Slug 155
Background Metals 160
Discussion 163
Summary 166
X. FOUR MUNICIPAL TREATMENT PLANTS RECEIVING
METALLIC WASTES 167
Metal Balances 178
Aerobic Efficiency of Plants 181
Anaerobic Efficiency of Plants 183
Slug of Metals 184
Nitrification 187
Summary 188
REFERENCES 189
SUBJECT INDEX 193
vii
-------
ABSTRACT
This volume, a collection of 10 research papers originating
at the Robert A. Taft Sanitary Engineering Center, describes
the effects of chromium, copper, nickel, and zinc on sewage
treatment processes. Results of pilot plant studies and full-
scale municipal plants are given.
For each of the metals and combinations of metals studied,
the effects on the aerobic and anaerobic treatment processes,
under continuous dosage, are given. The data presented allow
a reasonable estimate to be made of the amount of metallic
wastes that a treatment plant can receive and accomplish
the desired efficiency of treatment. The effects of slug dis-
charges of the metals on the aerobic and anaerobic processes
under pilot plant conditions and at municipal plants are
presented.
The concentrations of the metals in the various sludges
and effluents produced by a treatment plant are given. Metal
balances conducted for each of the studies show the amount of
metal removed by primary and secondary treatment.
-------
INTERACTION OF HEAVY METALS
AND
BIOLOGICAL SEWAGE TREATMENT PROCESSES
INTRODUCTION
At the request of the National Technical Task Committee on
Industrial Wastes, a series of investigations was undertaken because
of the interest of major metal processors in the acceptability of in-
dustrial wastes in municipal sewage treatment. The initial overall
objectives of the study were as stated at the beginning of Chapter I.
The scope of the study has been enlarged to incorporate additional
objectives, which include study of the effects of various metals on
nitrogen transformations and the determination of the effects of the
ratios of organic load to metal content on activated-sludge treatment.
The effects on the aerobic and anaerobic treatment processes,
under continuous dosage, are given for chromium, copper, nickel,
zinc, and combinations of these metals. Eight chapters deal with
studies carried out on a pilot scale; two chapters are concerned with
actual experience at operating municipal plants.
Each chapter has been taken from a completed research paper;
thus, the reader may find repetition of some points common to the
overall research. We hope that this repetition will reinforce those
areas that we feel are important considerations in an investigation of
this nature.
Mention of products and manufacturers is for identification only
and does not imply endorsement by the Public Health Service and the
U.S. Department of Health, Education, and Welfare.
-------
CHAPTER I. CHROMIUM*
The objectives of these studies are:
1. To determine the extent to which sewage treatment processes
can tolerate metallic wastes without losing efficiency in their
treatment of the organic pollutants in sewage.
2. To determine the extent of removal of metallic wastes in
sewage treatment plants and to follow their travel and con-
centration in various conventional sewage process units.
3. To develop modifications of sewage treatment procedures
that will make them more tolerant of metallic wastes or
more efficient in the removal of metals from sewage.
These objectives were established to serve a number of purposes.
The data gathered will assist sewage disposal authorities in determining
the quantity and characteristics of metallic wastes that may be accepted
without fear of disrupting operation. Frequently, using municipal re-
sources is the most efficient and desirable way for industry to dispose of
its wastes. The community and its industries constitute a mutually de-
pendent group; therefore ground rules are obviously desirable for deter-
mining optimum distribution of waste treatment effort between the
municipal and industrial waste treatment facilities.
Numerous studies on the effect of metals on biological processes
have been discussed in the literature in recent years. Unfortunately,
practically all of these have been confined to bench experiments, which
have not reasonably simulated plant situations, and the results obtained
are not necessarily applicable to either the pilot plant or large-scale
treatment of sewage. For this reason, the literature covering these ex-
periments is not reviewed and only those references that apply to the
present study are cited. Previous studies, for instance, do not show the
effects of a given metal on the efficiency of a treatment plant when the
metal is received in the influent sewage continuously or in slug doses.
In view of the incompleteness of data appearing in the literature,
this study on the effects of various metals and combinations of metals on
the activated-sludge process was undertaken on a pilot plant scale.
Chromium, because of wide use in electroplating and tanning indus-
tries, was selected as the first metal to be studied.
*Material in this chapter published previously in Journal Water
Pollution Control Federation, Washington, B.C. 20016. See References.
-------
EFFLUENT
Figure 1. Activated-sludge pilot plant No. 1.
PLANT DESCRIPTION AND OPERATION
The shape and dimensions of the pilot model activated-sludge
units used for studying the effect of continuous doses of chromium
are shown in Figure 1. The units were constructed of thin sheet
steel coated with a nonmetallic paint to minimize corrosion and the
addition of extraneous metal ions to the sewage and to prevent plat-
ing of metal ion being added. The aeration tanks were narrow to
limit short-circuiting and to attain spiral-type flow. The aerators
were carborundum diffuser tubes, 2 inches in diameter.
Sewage was delivered to the plant from a small constant-head
tank, which fed metering pumps of the rubber-tube nursing design.
The constant-head tank was filled from the sewage storage tank
on signal from a capacitor-activated liquid-level control. The con-
tents of the storage tank were mixed continuously by means of a
circulating system, which pumped from the bottom and returned to
the mid-section of the tank. This sewage feeding system was developed
to permit feed of a sewage of relatively constant strength with minimum
loss of strength during feeding. The constant-head tank was kept
mixed with a propeller-type stirrer. Primary sludge was drawn
by gravity flow on a once-per-day schedule. Sludge from the secondary
settler was pumped continuously by nursing-type pumps directly to
the head of the aerator. Excess activated sludge was removed once
daily as aerator liquor.
INTERACTION OF HEAVY METALS
-------
The feed rates, retention periods, loading factors, and other
operating information are listed in Table 1. The quantity of air
used was not measured because oxygen transfer by aeration in full-
scale units is not reproducible in the size of model treatment plant
employed. Sufficient air was supplied to maintain a dissolved oxygen
level of 1 milligram per liter or more in the secondary settler super-
natant. A concentration of suspended matter in the aerator liquor
of 1,800 to 2,000 milligrams per liter was the general objective.
Close control was not maintained in order to permit variation of the
solids level in a random manner over the range of practical operation.
The return sludge rate general objective was 30 percent of the raw
sewage flow. This return was also permitted to vary in a random
manner to cover the range of normal operation.
Table 1. DESIGN DATA AND LOADING FACTORS
ACTIVATED-SLUDGE PILOT PLANT
Item
Primary Settler
Capacity, gal
Detention time, hr
Surface overflow rate,
gpd/ft2 surface area
Aeration Tank
Capacity, gal
BOD loading, lb/day/1,000
ft3 aeration tank volume
BOD loading,
Ib/day/lbVS
Aeration period0, hr
Final Settler
Capacity, gal
Detention time0, hr
Surface overflow rate,
gpd/ft surface area
Design
used for
continuous
dose
studies "
13.8
2.3
125
30.9
21-92
0.60
5.0
9.3
1.5
192
Design
used for
sludge dose
studies
4.6
1.2
142
23.6
42-58
0.65
6.0
7.9
2
102
Feed rate, 147 gpd.
bFeed rate, 95 gpd.
Based on total tank volume and waste flow; retun
sludge not considered.
The metal was fed as a solution of potassium chromate by means
of a capillary tube control on a constant-head vessel. A constant level of
chromate solution in the feeder was attained with a Meriot bottle system,
in which the flow of air into the bottle to replace discharged fluid is
governed by the water level in the constant-head vessel. The feeder de-
livered 0.6 milliliter per minute of chromate solution to the tube carry-
ing sewage to the primary settler just at its entrance to the settler.
Chromium
-------
Variations in the chromium content of the feed were generally less
than 10 percent.
New Design
The activated-sludge units shown in Figure 1 accomplished sat-
isfactory treatment; however, certain operating problems were present.
The hoppers of the primary and secondary settling tanks retained
small portions of sludge, which had to be moved to the drain by
manual scraping. The accumulated sludge in the secondary settler
was removed by hand twice a day and returned to the aerator. To
maintain even airflow distribution along the aeration chamber was
not practical, and this uneven airflow caused a longitudinal roll in
the aerator much of the time and destroyed the desired spiral flow
pattern. The paint blistered and exposed steel surfaces.
SEWAGE FEED
AIR HEADER
CAPACITY: 7.9 gal
DETENTION TIME: 2 hr
SURFACE OVERFLOW RATE:
102 gpd/f!2
PRIMARY
SETTLER
CAPACITY: 4.6 gal
DETENTION TIME: 1 .2 hr
SURFACE OVERFLOW RATE
142 gpd/ft?
SPIRAL-FLOW
AERATOR
CAPACITY: 23.6 gal
AERATION PERIOD: 6 hr
BOD LOADING: 0.5 TO 0.8 Ib/day/lb
OF VOLATILE SOLIDS UNDER
AERATION
FINAL
SETTLER
Figure 2. Plastic activated-sludge pilot plant.
To solve these difficulties, a new plant of the design shown in
Figure 2 was constructed. The units were constructed of acrylic
plastic to avoid corrosion and plating factors. The shape of the set-
tler was altered to create extremely steep sides to prevent prolonged
holdup of any sludge. The concept of the design was to approximate
a pie-shaped section of a circular sedimentation basin, the cir-
cumferential section being short enough to be approximated by a
plane. The three walls were then sloped sharply inward to intersect
at an apex at the bottom.
INTERACTION OF HEAVY METALS
GPO 82O—663—2
-------
The aerator was separated into six chambers by baffles to control
longitudinal mixing and thereby more nearly approximate spiral flow
design, There was one 1/2-inch-diameter circular opening in
each chamber. The holes were placed alternately at the top and
bottom and left and right side of the aerator so that the flow through
each chamber was from the top at one side to the bottom on the other
side, or vice versa. Carborundum air diffusers 2 inches in diameter
were used.
The plastic unit was used for testing chromium slugs. Feed
rates, retention periods, and other information are also listed in
Table 1. The new settler design prevented prolonged sludge ac-
cumulation. A good flow pattern in the primary settler was achieved
with a baffle near the inlet extending into the lower levels of the
settler. The inflow tended to distribute itself in a horizontal layer
near the lower end of the baffle and rise vertically to a thin surface
layer, which moved horizontally to the weir. The achievement of such
a flow pattern in the secondary settler was elusive. Regardless of
how the settler was baffled, the inflow of aerator liquor caused a
pronounced roll in the settler when the feed was near the surface.
Operation with minimum roll was achieved by placing the inlet at
about one-third the depth above the apex. In normal operation, the
incoming sludge could be observed to settle as a layer at the inter-
face of the sludge and the supernatant, and to fall as a slug vertically
down to the return line. A thin film of sludge that accumulated at
times on the sloping walls was squeezed free each day. All studies
of the aeration phase of treatment, in these studies, were at room
temperature, approximately 20° C.
The digesters were 5-gallon glass carboys equipped with gas
collectors of the water displacement type. The digesters and gas
collectors were housed in a constant-temperature room maintained
at 30°C. The digesters were fed once a day with scheduled quan-
tities of primary and excess activated sludge in the ratio of 64 percent
primary sludge and 36 percent excess activated sludge on a volatile
solids (VS) basis. The ratio of volatile solids in the primary and
excess activated sludges was estimated, when the study was started,
as likely to approximate actual production of the pilot plant. This
proportion proved to agree within about 3 percent to the proportions
actually produced in the unit receiving chromium.
Operation
The digesters were thoroughly agitated once a day by hand,
and a volume of mixed digester contents equal to 1/26 of the digester
volume was withdrawn. Thus, the average detention time in the
digester was kept constant at 26 days. The digested sludge withdrawn
was, therefore, what would be the combined supernatant and digested
sludge of normal operation; the solids content was correspondingly low,
Chromium
-------
ranging from about 0.5 to 1 percent. This concentration of solids
was necessary to facilitate sampling.
The withdrawal of sludge and supernatant from the digesters
used for testing the effect of slug doses of chromium was handled in
a different manner. A volume of supernatant exceeding the scheduled
sludge feed was withdrawn before mixing. The sludge was then fed
and the liquid volume made up to a constant level with supernatant
just previously withdrawn. Digested sludge was not removed from
these units during the short test period.
The digesters were fed in the early weeks of operation at a rate
of 5 grams per day of volatile solids; later this was increased to 10
grams per day. It was necessary to estimate the quantity of seed
sludge corresponding to a balance of these feed rates and retention
periods at the start of each study. Several months' operation was
required before the sludge quantity in the digester actually levelled
off. Digester loadings and operating data are listed in Table 2.
Table 2. DESIGN DATA AND LOADING FACTORS FOR DIGESTERS
Capacity, ft3
Detention, days
Volatile solids loading, lb/day/1,000 ft ;
digester volume
0.67
26
33
Sewage Feed
Concurrently with the design and construction of the pilot plants,
an economical and adequate means was sought to fortify the sewage
received in the experimental wing. This sewage, while of domestic
origin, had an average biochemical oxygen demand (BOD) of about
75 milligrams per liter because of high ground-water infiltration.
Various mixtures of supplemental foods were evaluated for use in
changing the BOD and behavior patterns of the sewage to conform to
ordinary domestic sewage characteristics. The cheapest suitable food
was ground dog food. Various brands of dog food were tried before one
was finally selected. Four hundred grams of the meal was allowed
to soak overnight and then homogenized in a large blender. This
homogenate was added to a 325-gallon tank of raw sewage. The ad-
dition of this supplement raised the average BOD to about 260 milligrams
per liter, and the sewage so produced showed characteristic behavior.
Analytical Methods
With the exception of the determination of solids in primary
sludge, aeration liquor, digested sludge, and the determination of
chromium, the methods used in the study conformed to those given
in Standard Methods (1).
8 INTERACTION OF HEAVY METALS
-------
In the case of sludge solids, the aluminum disc was not used
and the sludge was filtered directly on a Buchner funnel with No. 40
Whatman paper.
The determination of total chromium in the raw sewage and
various units of the plant (including digesters) was carried out by
two variations of the basic procedure with essentially the same
results. In both cases the organic matter was destroyed by fuming with
H2 SO4 and HNO3. In this determination it has been postulated that a
loss of chromium occurs (if Cr+« is present) in the presence of chlorides
and, therefore, all chromium should be reduced to the trivalent
state prior to destruction of the organic matter. Just what concentration
of chlorides must be present for this loss to occur, however, has never
been thoroughly investigated. From some unpublished results the
lower limit has been set at about 100 milligrams per liter, this
concentration being higher than that normally encountered in this
study. In the presence of organic matter, Cr+6 is rapidly reduced to
the trivalent state on boiling with an H2SO4 -HNO3 mixture. With
sludges, particularly digester sludge, the sample had to be fumed
several times with this acid mixture before all organic matter was
destroyed and a clear residue obtained. In one modification used, the
Cr*6 was first reduced with sulfite and the sample was taken to fumes
the necessary number of times to destroy the organic matter and then
fumed for 15 minutes without being taken to dryness. The digestate
was then diluted and the trivalent chromium oxidized with KMnO<
(the excess of which was then destroyed with sodium azide). The
solution was filtered through a sintered glass crucible, and the chro-
mium determined on the whole or aliquot (depending on its concentration)
using diphenylcarbazide. Absorbance readings were made on a
spectrophotometer after 5 minutes, but not later than 15 minutes.
The other modification used, consisted of destruction of the organic
matter in the usual way and then fuming to dryness. The residue was
taken up in 1-1 HjSO*, boiled, and then diluted. The silica was filtered
off, and the filtrate made up to volume. The chromium in an aliquot
was then oxidized with KMnC>4, and the color developed in the usual
manner. As stated previously, both modifications gave essentially
the same chromium recovery.
High concentrations of chromium were determined by oxidation
of the trivalent form to Cr*6 after destruction of the organic matter.
The hexavalent chromium was then titrated with a standard solution
of ferrous ammonium sulfate using "ferroin" as the indicator.
In plant operation and special studies, in addition to the deter-
mination of total chromium, it was necessary to determine hexavalent
chromium. This was especially true at the higher chromium feed
rates. In determining hexavalent chromium, the sample was filtered
by means of a membrane filter. This filtrate was perfectly clear,
and hexavalent chromium could be determined with no difficulty in the
usual manner.
Chromium 9
-------
SAMPLE COLLECTION AND PRESERVATION
Samples for efficiency studies of the unit continuously receiving
chromium and the control unit were collected by hand at approximately
hourly intervals and composited for a 7-hour period. The samples
were refrigerated during the compositing period and held in a re-
frigerator overnight before analysis on the following day.
Samples for studies of the effect of slug doses of chromium
were collected by means of a swing-tube solenoid-actuated automatic
sampler. The samplers diverted the stream to be sampled to a
compositing carboy on signal from a timing device. The timing
device was set to collect the flow during about 2 seconds of each
minute. The compositing periods are reported with the experimental
data. These samples were kept in ice chests or refrigerated prior
to analyses.
For purposes of material balances for chromium in the con-
tinuously dosed unit, samples of the effluent were collected continuously,
24 hours per day, 7 days per week, by means of an automatic sampler.
The effluent samples were composited over 7-day periods. The 7-day
accumulation of sludge was placed in a large tank and stirred me-
chanically prior to and during withdrawal of a sample. Samples of the
mixed liquor at the beginning and end of the compositing periods were
collected by dipping equal small quantities at regular intervals along
the length of the aerator.
The digesters were routinely sampled for their mixed contents.
The digesters were shaken vigorously by hand preceding and during a
withdrawal of the samples.
EXPERIMENTAL DATA
Continuous Feeding
Two basic problems were present in this study: (1) the con-
centration at which the effects of a given metal ion are felt when
received continuously in the influent of an activated-sludge treat-
ment plant and (2) the concentration necessary to have an immediate
effect on a plant and the time required for this plant to recover from
a slug dose of the metal in question.
The effects of five concentrations (from 0.5 to 50 mg/1) of
hexavalent chromium were studied. The two lowest concentrations
were fed for approximately 1 month while the other three were studied
for at least a 6-week period. The effects on BOD, chemical oxygen
demand (COD), and suspended solids (SS) removal are given in Table
3. This table shows that the average BOD loads in both the chromium-
10 INTERACTION OF HEAVY METALS
-------
Table 3. EFFECT OF HEXAVALENT CHROMIUM ON PLANT EFFICIENCY
Period
covered
Cr + 6
fed,
mg/
No. of
pies
Raw waste,
mg/liter
Cr*6
fed
unit
Con-
trol
unit
Primary
effluent,
mg/liter
Cr+6
fed
unit
Con-
trol
unit
Reduction, %
Cr+6
fed
unit
Con-
trol
unit
Final
effluent,
mg/liter
Cr+6
fed
unit
Con-
trol
unit
Plant
removal
efficiency,%
Cr*6
fed
unit
Con-
trol
unit
Average BOD
Aug. 29-Oct. 2
Oct. 3-Oct. 31
Oct. 31-Dec. 19
Dec. 19-Feb. 25
Feb. 25-Apr. 24
0.5
2.0
5.0
15.0
50.0
6
7
8
8
8
268
261
311
320
253
259
288
314
296
263
180
199
192
193
138
180
201
173
198
119
33.5
22.4
35.8
39.8
45.5
30.4
28.3
44.4
32.9
54.8
14.8
16.2
10.9
15.9
20.9
14.7
18.9
14.9
12.7
13.0
94.3
93.2
96.8
95.0
91.7
94.3
93.5
94.9
95.7
95.1
Average COD
Aug. 29-Oct. 2
Oct. 3-Oct. 31
Oct. 31-Dec. 19
Dec. 19-Feb. 25
Feb. 25-Apr. 24
0.5
2.0
5.0
15.0
50.0
6
7
8
8
8
452
427
493
458
411
444
447
496
467
406
270
297
312
294
234
266
305
285
277
227
40.3
30.4
36.7
35.8
43.1
40.1
31.8
42.5
40.7
44.1
52.0
70.0
74.0
96.0
67.0
59.0
65.0
75.0
83.0
49.0
88.5
83.6
85.0
79.0
83.7
86.7
85.5
84.9
82.2
87.9
Average SS
Aug. 29-Oct. 31
Oct. 3-Oct 31
Oct. 31-Dec. 19
Dec. 19-Feb. 25
Feb. 25-Apr. 24
0.5
2.0
5.0
15.0
50.0
6
7
8
8
8
323
242
312
267
277
303
254
316
267
270
143
138
157
135
115
146
130
144
119
114
52.5
43.0
49.7
49.3
58.5
55.1
48.8
54.4
55.4
57.8
12.0
20.0
12.0
13.0
12.0
13.0
9.0
13.0
9.0
10.0
96.3
91.7
96.2
95.1
95.7
95.7
96.5
95.9
96.6
96.3
fed and control units were essentially the same, the 5-day BOD com-
paring with what would be expected in a sewage treatment plant
receiving primarily domestic sewage. The BOD removal in the
primary tank is also normal with respect to larger-scale plant opera-
tion. No logical explanation can be offered for the higher BOD re-
moval in the primary settler during the last period of operation since
the data for both the-control and chromium-fed units are uniformly
higher. The overall plant removal of BOD also shows little difference
between the two units during the first four periods. However, when 50
milligrams per liter of Cr*6 was being fed, the average BOD removal
obtained was about 3 percent lower for the chromium-fed unit. Limited
significance can be placed on this figure since during the feeding of
5 milligrams per liter of Cr+6 the unit showed an average BOD removal
2 percent higher than that from the control.
A COD check indicated that the presence of Cr*« had little effect
on the removal of organic matter except possibly during the last
period when the removal by the chromium-fed unit averaged about 4
Chromium
11
-------
percent lower than that of the control. This could be due to the fact
that during this period the final effluent was more turbid, the isolids
contained therein approaching colloidal size and, therefore, not being
determinable as suspended solids.
We had thought that in the presence of organic matter the Cr + 6
might be reduced to Cr*3, which would serve as a coagulant at the
prevailing pH of the sewage. This naturally would lead to higher solids
removal in the primary unit; however, as shown in Table 3 there was
no difference between the control and the chromate-fed units in solids
removal in the primary tanks. Further, the actual solids removed
are in line with those found in larger plants. This suggested that
the Cr+3 was not being precipitated as the hydroxide and was not acting
as a flocculating agent. Udy (2) points out that the hydrous oxide
Cr2C>3-xH2O is formed and may be either a positively or negatively
charged particle, depending on the pH. Only 10 milligrams per liter
of chromium was reduced and precipitated with the highest chromium
feed. Some of the reduced chromium also passed out of the primary in
the effluent. Table 3 indicates that the overall plant removal of
suspended solids was high and that regardless of the chromium feed
concentration, both units were equally efficient.
During the periods in which 0.5 and 2 milligrams per liter of
Cr*4 were being fed, the concentration of soluble chromium in the
final effluent was negligible. With the 5-milligram-per-liter feed, the
hexavalent chromium was usually less than 1.5 milligrams per liter
in the final effluent and the total chromium less than 2.5 milligrams
per liter. When, however, 50 milligrams per liter was being fed
continuously, the primary effluent usually contained about 40 milligrams
per liter of hexavalent chromium and the final effluent, around 30
milligrams per liter. Only occasional spot checks were made of the
hexavalent and total chromium. As will be discussed later, a way
was found to reduce the chromium content of the final effluent to a
much lower level.
In the effluent from the primary tank the chromium was present
in both non-settleable precipitated and soluble forms. A further
reduction of hexavalent chromium took place in the aeration tank.
The activated-sludge particles served to adsorb and reduce a portion
of the soluble chromium as well as the finely divided precipitated
chromium. This caused a buildup of total chromium in the aerator.
Table 4 shows that this buildup increased markedly with increasing
chromium fed to the raw sewage when expressed on a milligram-
per-gram-suspended-solids basis. There naturally was a fluctuation
in this value because of fluctuation in the suspended solids content
of the aerator. The values given in Table 4, therefore, are average
values that express the chromium concentration at the various chromium
feed levels. When the study was concluded, a concentration of 93
milligrams of chromium per gram of suspended solids, or 9.3 percent,
had been reached.
12 INTERACTION OF HEAVY METALS
-------
Table 4. BUILDUP OF CHROMIUM IN
PRIMARY AND AERATOR SLUDGES
Chromium
level in
feed,
mg/liter
0.5
2.0
5.0
15.0
50.0
Avg total
chromium in
primary sludge,
mg/g SS
0.36
1.3
1.5
2.5
5.9
Avg total
chromium in
aeration tank
contents,
mg/g SS
4.0
8.0
26.0
36.0
66.0'
Highest concentration reached was 93.
Slug Dose Feeding
A number of instances of activated-sludge plants receiving chro-
mium in slug doses are recorded in the literature. In one such case,
reported by Jenkins and Hewitt (3), a maximum concentration of
320 milligrams per liter of chromium was received in the influent
sewage. Edwards and Nussberger (4) report the occurrence of two
slug doses of chromium at an activated-sludge plant. During the
first slug dose the influent contained 430 milligrams per liter of
chromium for approximately 30 minutes. The following day a second
slug dose was received for approximately 2 minutes. During this
latter period the influent sewage contained 1,440 milligrams per liter
of chromium. The results of these doses were reflected in a higher
BOD in the effluent and a cessation of nitrification.
In this study of the effects of slug doses of Cr + 6 on the activated-
sludge process, concentrations of chromium of 10, 100, and 500 milli-
grams per liter were used. These concentrations were fed over a
period of 4 hours to activated sludge that had had no previous contact
with chromium. During the first 12 hours, composite samples were
taken over 4-hour periods and at intervals thereafter until the unit
returned to normal operation. The plant efficiency before the slug
dose was fed, had been followed for at least 2 weeks. The data as
given in Table 5 show that the feeding of 10 milligrams per liter over
a 4-hour period had no effect on plant performance. The rise in
suspended solids in the effluent after 4 days was due to bulking in the
secondary settler. When a slug dose of 100 milligrams per liter of
Cr+6 was used, the plant efficiency, as measured by BOD removal,
dropped about 3 percent during the first 24 hours. The drop in COD
removal was greater. Recovery was rapid. This also was true for
the suspended solids removal. When a slug dose of 500 milligrams
per liter was fed over a 4-hour period, the effect was striking and
Chromium
13
-------
Table 5. PLANT PERFORMANCE IN SLUG DOSE STUDIES
Time of sampling
or dosing
Avg data prior
to slug dose3
Feb. f2b
9 a.m. — 1 p.m.
1 p.m.— 5 p.m.
5 p.m. -9 p.m.
9 p.m.— 9 a.m.
Feb. 16 c
Feb. 18C
Avg data prior
to slug dose"
Feb. 25"
8 a.m.— 12 noon
12 noon-4 p.m.
4 p.m. -8 p.m.
Feb. 27 c
Mar. 2C
Avg data prior
to slug dose"
Mar. 31 e
8 a.m.— 12 noon
12 noon— 4 p.m.
4 p.m.— 8 a.m.
8 p.m.— 8 a.m.
8 a.m. -4 p.m.
Apr. 3°
Apr. 6C
BOD
Raw
waste,
mg/liter
291
284
288
221
235
187
269
210
200
179
223
213
201
230
324
249
236
240
205
208
220
189
129
164
Primary
effluent,
mg/liter
177
-
-
_
_
-
126
216
152
137
133
139
147
104
95
Percent
removal
(primary)
36.4
-
-
_
_
:
52.2
8.4
36.6
33.1
36.1
36.8
22.2
19.4
42.1
Plant
effluent,
mg/liter
8.6
8.8
12.9
10.1
7.0
9.4
10.4
6.1
7.1
6.0
11.5
14.8
14.7
8.2
10.8
7.8
21.4
27.0
28.0
27.2
28.7
21.4
10.6
9.3
Percent
removal
(total)
96.5
96.9
95.5
95.4
97.0
95.0
96.1
97.1
96.3
96.6
94.8
93.1
92.7
96.4
96.6
96.9
90.5
<89.0
<86.0
87.0
86.8
88.9
97.9
94.3
COD
Raw
waste,
mg/liter
341
373
422
340
301
304
291
330
292
260
279
299
272
372
375
358
362
352
341
320
350
311
231
224
Primary
effluent,
mg/liter
234
-
_
-
—
-
203
212
206
194
173
189
206
193
165
Percent
removal
(primary)
31.5
:
—
-
-
—
42.9
41.4
41.5
43.1
45.9
46.0
33.8
16.5
26.3
Plant
effluent,
mg/liter
31.9
37.5
45.9
35.7
42.1
43.5
49.1
35.1
44.3
37.6
56.0
74.0
62.0
51.0
58.0
41.2
75.0
64.0
87.0
85.0
69.0
54.0
45.0
51.0
Percent
removal
(total)
89.8
89.9
89.1
89.5
86.0
83.7
83.1
89.4
84.8
85.5
79.9
75.3
77.2
86.3.
84.5
86.7
79.3
81.8
74.5
73.4
80.3
82.6
80.5
77.2
Suspended Solids
Raw
waste,
mg/liter
228
428
342
220
160
258
268
254
154
116
158
115
188
144
258
241
278
256
276
234
262
244
160
158
Plant
effluent,
mg/liter
4.3
4.0
2.0
2.0
1.0
5.0
10.0
38.0
2.0
14.0
3.0
3.0
10.0
4.0
2.0
3.5
20.0
19.0
25.0
11.0
7.0
5.0
7.0
7.0
Percent
removal
(total)
97.1
99.1
99.4
99.1
99.4
98.1
96.3
85.0
98.6
87.9
98.1
97.4
94.6
97.2
99.2
98.6
92.8
92.6
91.0
95.3
97.3
98.0
95.6
95.6
H
rt
i
rt
3
§
H
f
" Plant was operated tor 2 weeks using new activated sludge.
b!0 mg/liter Cr+6fed from 9 a.m. to 1 p.m.
'Denotes 24-hr composite sample.
^ mg/liter Cr**fed from 8a.m. to 12 noon.
1'500 mg/liter fed from 8 a.m. to 12 noon.
-------
was noticeable even during the period of feeding. Plant efficiency as
measured by BOD and COD removal continued to drop for about 32
hours and then started to recover, and after 4 days the unit was again
operating normally. The plant effluent was quite turbid for about 24
hours with a consequent increase in suspended solids.
One effect of hexavalent chromium on sewage systems noted
by many workers is the inhibition of nitrifying bacteria. Jenkins
and Hewitt (3) stress this point in their activated-sludge studies.
Placak et al. (5) also suggest that this fact could be used in deter-
mination of the carbonaceous demand of sewage samples. It was found
in this study that nitrification was inhibited for approximately 10 days
even at the lower hexavalent chromium levels and then proceeded
regardless of the concentration of chromium fed. Clearly, nitrifying
bacteria can acclimatize to the constant presence of chromium.
Chromium Distribution and Recovery
From the weekly composites of primary sludge, excess activated
sludge, plant effluent, and a composite grab sample of the aeration
liquor at the end of the week and knowing the total chromium fed,
the percent recovery of the chromium and its distribution were
determined. The results for the various chromium levels used are
presented in Table 6. For the two higher concentrations of chro-
mium used, more than 50 percent of the chromium appeared in the
Table 6. CHROMIUM DISTRIBUTION AND RECOVERY "
Period
Aug.l8-Sept. 16
Oct. 3-31
Oct. 31 -Dec. 19
Dec. 19 -Feb. 24
Mar. 9-30
Hexa-
valent
mium
fed,
mg/ liter
0.5
2.0
5.0
15.0
50.0
Total
chro-
mium
fed,
1.202
6.23
16.73
56.1
183.0
Total
chro-
mium
primary
sludge,
g
0.089
0.454
0.705
1.45
2.44
Total
chromium
in excess
activated
sludge,
g
0.743
0.862
5.32
15.4
16.8
Total
chro-
mium
plant
effluent,
g
0.258
2.78
6.44
31.5
162.0
Net
change of
chromium
in aerator
solids,
g
+ 0.064
- 0.05
+ 3.17
+13.9
- 6.9
Chro-
covered,
1.103
4.73
13.1
49.7
178.0
Percent
ac-
counted
for
92
76
78
89
97
"All values in table represent average of weekly balance periods.
plant effluent. This was both hexavalent and insoluble trivalent
chromium. In the case of the 50-milligram-per-liter feed, more than
90 percent was in the hexavalent form. In accounting for the amount
of chromium fed it must be recognized that sampling, laboratory
analyses, and flow measurements are involved. Without careful
attention to these factors, wide discrepancies in the percent chro-
mium recovered can be obtained. The results in Table 6 show a
fairly high degree of accuracy, considering the multiplicity of factors
involved.
Chromium
15
-------
Oi
o
f-
U
D
W
K
ac
H
<
0
w
K
S
D
0
K
tc
O
H
2
W
X
W
X
J-
co
H
M
a;
Ofl
6
e"
0
u
cc
0
H
D
O
cn
Q
o
u
*
0
d
S
0)
c
a.
"d
G.
OJ
OJD
CO
OJ
to
CD
i
J5
"u
3
T3
O
"o
3
13
a
o
o
U]
3
.2J
CUD
E
IT)
ro CN
ro t^
CN CN
TT CN
I/) LO
oo ir>
^ oo
CN ,-H
O- CN
CN VD
•<*• ro
£S
O> IT)
CN CN
58
CO CO
CO \O
v£> CO
t--
Tr
I
CN
8
O
CN
LD
O
CN
CN
S3
ro
00
"
CN
O
ro
00
s
IT)
t>
-1
1
«
*
^
OO
00
t>
CN
(^
O
CO
£
00
LO
O
CN
s
CO
CO
o
CN
o
CN
1
&
8
•*r
06
oo
CN
un
ro
CN
CN
in
o
00
00
10
CN
t>
CN
CO
£
Ov
C-I
1
E
in
w
CO
5 >,
13 "^
b ^
i-
0) O
00 u
S 3
> *S
-------
Chromium Removal With Biological Reductor
The large amounts of chromium passing the system with the 50-
milligram-per-liter chromate feed were considered as a possible
menace to receiving waters. A simple modification of the activated-
sludge process to reduce chromium loss was sought. An auxiliary
tank, to which both the primary effluent and return sludge were diverted,
was installed. The effluent of this tank then went to the aeration basin.
The tank was stirred, but not aerated so that it acted as an activated-
sludge plant with chromate serving as the principal source of oxygen.
The pilot plant was operated on this basis for approximately 1 month.
During this period the average hexavalent chromium fed was about
47 milligrams per liter. Previous to the run, the average chromium
in the plant effluent had been about 40 milligrams per liter. Table 7
shows that this method of operation resulted in the removal of more than
90 percent of the chromium fed. The chromium appearing in the plant
effluent was all in the reduced insoluble state. The sludge in the
reductor unit as well as that in the aeration basin became dark gray
in color. Plant efficiency as measured by BOD removal, however,
remained well above 90 percent during the period of operation. The
BOD of the plant effluent decreased as the time of operation increased.
In view of the progressive increase in plant efficiency, it is reasonable
to believe that the plant could have continued operating satisfactorily
indefinitely.
SLUDGE DIGESTION
Test Procedures And Results
In Table 8 the data obtained during the operation of the two
digesters when the various chromium concentration levels were
being fed are summarized. In both the primary and excess activated
sludge the chromium was all in the reduced insoluble state. Coburn
Table 8. GAS PRODUCTION IN TERMS OF VOLATILE SOLIDS FED AND
CHROMIUM RANGES IN DIGESTER
Aug. 29 -Oct. 2
Oct. 3 -31
Oct. 31-Dec. 19
Dec. 19-Feb. 25
Feb. 25-Apr. 24
Chromium
level in
feed of
sludge
plant,
mg/liter
0.5
2.0
5.0
15.0
50.0
Chromium ranges in
digester
mg/liter
11.2"-21.0
21.0- 27.6
27.6- 72.0
72.0-200.0
200.0-420.0
mg/g SS
1.04- 1.69
1.69- 3.16
3.16-14.5
14.5 -28.0
28.0 -34.6
CO 2 content of
digester gas, %
Chro-
mium fed
digester
26.0
27.0
27.0
26.0
-
Control
digester
25.0
26.0
28.0
26.0
-
Gas production,
ml/g VS fed
Digester
fed sludge
from
receiving
unit
769
710
649
666
571
Digester
fed
sludge
from
control
unit
622
680
800
704
617
Unit had operated at this chromium level for approximately 6 weeks before sufficiently reliable
records were obtained.
Chromium
17
-------
(6) in his discussion on toxic wastes stated that the precipitated chro-
mium transferred to the digesters along with the sludge would produce
a toxic action thereby slowing up or stopping digestion. Table 8
shows that regardless of the chromium concentration in the digester,
gas production was not affected, often being higher in the chromium-
loaded digester than in the control. The percent CO2 in both digester
gases was found to be essentially the same. During the entire period
of operation a digester was limed only once, the pH of both being
fairly constant between 6.8 and 7.0. If trivalent chromium in solution
had been added directly to the digester, deleterious effects could
possibly have occurred. This was not done since substantial amounts
of soluble trivalent chromium would require a pH of less than 4, which
is not likely to be encountered in sludge pumped to a digester.
A digester was fed with freshly collected primary and excess
activated sludge from the pilot plant while it was receiving 50 milli-
grams per liter of Cr+« continuously. Both the total and hexavalent
chromium content of these sludges are given in Table 9. This digester
Table 9. CHROMIUM CONTENT OF FRESH SLUDGE FED TO DIGESTER"
Sludge
identification
Primary sludge
Excess acti-
vated sludge
Hexavalent chromium content of feed
Concentration,
mg/liter
38
32
Amount fed,
mg
11
32
Suspended
solids,
mg/g
1.0
3.2
Total chromium content of feed
Concentration,
mg/liter
330
530
Amount fed,
mg
99
530
Suspended
solids,
mg/g
8.7
53
Sludge taken from pilot plant while receiving 50 rag/liter hexavalent chromium continuously.
was fed on the 7th day and then from the 15th through the 21st days.
The hexavalent chromium concentration (based on digester contents)
was about 3 milligrams per liter. Figure 3 shows that this concentration
of hexavalent chromium fed had no effect on gas production.
Another digester with no previous history of receiving chromium
was fed fresh primary and excess activated sludge from the pilot
plant that had received a slug dose of 100 milligrams per liter of
hexavalent chromium. The chromium fed was both in the hexavalent
18
INTERACTION OF HEAVY METALS
-------
140
120
JOO
40
20
DIGESTER FED FRESH SLUDGE FROM 50-mg/]iter
ACTIVATED-SLUDGE UNIT
DIGESTER FED FRESH SLUDGE FROM
CONTROL ACTIVATED-SLUDGE UNIT
I I I I I I I I I I I
12 16
TIME, days
20
24
Figure 3. Effect on digesters of fresh sludges from activated-sludge plant receiving
50 mg/liter chromium continuously.
Chromium
19
-------
90
80
70
60
50
40
30
20
10
II1TlTTIIIIII
DIGESTER RECEIVING C,~
O CONTROL DIGESTER
FED ON v J/
NINTH DAY \, 7>
03 69 12
TIME, days
Figure 4. Effect on digester of sludge from 100-mg/liter
slug dose to activated-sludge plant
and reduced form, and the feed contained 5.8 milligrams chromium
per gram of suspended solids fed. Figure 4 shows that there was no
noticeable difference between the operation of the two digesters. The
average gas production for the chromium fed and control digesters
per gram of volatile solids (VS) fed was 702 and 687 milliliters,
respectively. Still another digester was fed with fresh solids from the
pilot plant receiving a slug dose of 500 milligrams per liter of hexavalent
chromium. No determinations of the hexavalent chromium were made,
but in the case of this digester it would be about 30 milligrams per liter
(based on digester contents). Here again, no effect was noted on
gas production.
20
INTERACTION OF HEAVY METALS
-------
Since the feeding of reduced chromium or a combination of
reduced and hexavalent chromium had no apparent effect on digester
operation, the effect of a slug dose of hexavalent chromium alone was
explored. This was done by adding directly 300 milligrams per liter
(based on digester contents) of hexavalent chromium to a digester
that had previously received chromium-containing sludge. Following
the addition of this slug dose, the mixed liquor contained 700 milligrams
per liter of total chromium. The supernatant, after thorough mixing,
settling, and filtration, was analyzed and found to contain 60 milligrams
per liter of hexavalent chromium. This high reduction in hexavalent
chromium in such a short period seemed questionable; therefore, 300
milligrams per liter of Cr+6 was added to another digester sludge, and
the same manipulative procedure was followed. Again a rapid loss
of Cr+6 occurred; thus, the previous results were verified. A factor
contributing to the rapid reduction of hexavalent chromium is the
amount and condition of the solids present. This very rapid loss
of Cr + 0 can be accounted for largely by oxidation of the easily oxidiz-
able compounds present and the bacterial utilization of oxygen available
in the CrO4~2 ion. The suspended solids content of the digester was
11,000 milligrams per liter. After 2 days the hexavalent chromium
content of the digester had dropped to about 3 milligrams per liter,
In Figure 5 the results of feeding this amount of hexavalent chromium
in a slug dose are shown as well as the curve denoting the normal gas
production obtained by feeding 9 grams of volatile solids (VS) per
day. All gas production stopped for about 7 days and then gradually
resumed, and the digester eventually returned to normal operation.
Since in the feeding of 300 milligrams per liter of hexavalent
chromium the digester was able to recover in a comparatively short
time, a slug dose of 500 milligrams per liter was tried. When this
amount was added to another digester, all gas production stopped and
the digester did not recover. Its contents were finally discarded at
the end of 6 weeks. The likelihood of even 300 milligrams per liter
of hexavalent chromium being pumped to a digester is very remote.
A large percentage of the hexavalent chromium received at a sewage
treatment plant would either be reduced in the primary or aeration
basin, or pass out in the plant effluent.
There are two ways in which a digester could receive hexavalent
Chromium; namely, through the remote possibility of a slug dose or
through receiving smaller amounts in the fresh solids. A further
experiment was tried in which a digester was directly fed 50 milli-
grams per liter of hexavalent chromium (based on the digester con-
Chromium 21
-------
80
70
60
50
0
o
40
O
LU
>
I—
<
30
zo
10
NORMAL GAS PRODUCTION
RATE FROM 9 g/day OF VS
DAILY FEED RATE
HEXAVALENT Cr SLUG DOSE OF
300 mg/liter IN ENTIRE CONTENTS
OF DIGESTER ADDED ON THIRD
DAY &
CUMULATIVE
GAS PRODUCTION
if"
VOLATILE SOLIDS FED
12 ro
Q
10 £
8 9
10
20
30
TIME,days
40
50
60
Figure 5. Effect on digester of 300-mg/liter slug dose of hexavalent chromium
(Based on digester contents).
tents) daily for 42 days. The results obtained are shown in Figure
6. This amount of chromium was fed to one digester starting on the
fifth day of observation. The two digesters were operating almost
identically when the chromium feeding was started. Both digesters
were receiving 10±0.3 grams of volatile solids per day. After about
4 days of feeding 50 milligrams per liter of chromium, gas production
started to decrease and at the end of 42 days the chromium-fed di-
gester was producing about 75 milliliters of gas per gram of volatile
solids fed, whereas the control was producing about 650 milliliters
of gas. The total chromium concentration of the digester contents
at the end of the experiment was 3,046 milligrams per liter. This
effect was no doubt due to the cumulative effect of daily feeding of
this amount of hexavalent chromium. This situation is, of course,
not likely to arise in a sewage treatment plant.
22
INTERACTION OF HEAVY METALS
GPO 820-663-3
-------
320
280
240
200
160
=>
u
120
80
40
O DIGESTER RECEIVING C
• CONTROL DIGESTER
GAS PRODUCTION
GAS PRODUCTION
PER g VS FED
I I I I I I I I I I I I I II I I I I I I I I I I I I I 1
8 12 16 20 24 28 32 36 40 44 48
Figure 6. Effect on digester of 50 mg/liter hexavalent chromium added daily
(based on digester contents)
Filterability of Digested Sludge
In connection with the operation of digesters, the filtering char-
acteristics of the digested sludge are of importance. It was desirable
to learn whether the presence of precipitated chromium would alter
this property. The method of Center (7) was followed in making
the determination. In Figure 7, only a portion of the curves is
shown. Three different sets of conditions were studied. In set 1
digested sludges from the chromium-containing and control digesters
were filtered under vacuum (about 28 in.), and the volume of filtrate
obtained was plotted against time. The "x" on the .graphs denotes the
time for cracking of the sludge cake and loss of vacuum. The chro-
mium-containing sludge filtered much more rapidly than the control.
Chromium
23
-------
240
200
160
120
240
E
III
< 200
a:
o
>
160
I 20
240
160
120
Cr-CONTAINING SLUDGE
CONTROL SLUDGE
(1) UNTREATED SLUDGE
(2) ELUTRIATED SLUDGE
(3) ELUTRIATED AND FeCI3-TR EATED
SLUDGE
10
15 20
TIME, min
25
30
35
Figure 7. Fi Iterobility of digested sludges.
This, no doubt, is due entirely to the presence of precipitated chro-
mium, which increases the permeability of the sludge cake and thus
improves its draining characteristics. In set 2 both sludges were
elutriated once with three times the sludge volume. Tap water was used
to triple the volume. Elutriation shortened the time necessary for
cracking in the chromium-containing sludge, but aided the control
sludge only slightly. When both sludges were elutriated once and
also treated with ferric chloride, as shown in set 3, the time for
cracking was lessened greatly.
24
INTERACTION OF HEAVY METALS
-------
DISCUSSION OF RESULTS
Chromate may enter municipal sewage in many different ways,
Perhaps most frequently it occurs in plating wastes, although it may
have its source in tanning operations, in waters given corrosion
inhibition treatment with chromate, or in aluminum-anodizing wastes.
Since chromate retained in a sewage treatment plant is reduced to
chromic chromium, it would also appear to be pertinent in relation-
ship to the effect of this form of chromium on sewage treatment.
In addition to chromium, many plating wastes contain significant
quantities of copper, nickel, zinc, cadmium, cyanide, and acid or
alkali. Further, the wastes from a metal working industry are likely
to be accompanied by large sewered losses of oil. These factors
and many others that might enhance the toxicity of chromium have
obviously not been considered.
Table 10. SUMMARY OF REACTIONS TO HEXAVALENT CHROMIUM
Chromium
concentra-
tion,
mg/liter
50
100
500
50
300
500
*"•
Feed
methods,
process
Continuous
feed to acti-
vated sludge
plant.
Slug dose to
activated-
sludge plant.
Slug dose to
activated-
sludge plant.
Fed daily to
digester;
based on di-
gester con-
tents
Slug dose to
digester
Slug dose to
digester
Effect on
activated sludge
BOD removal efficiency
dropped about 3%
Plant recovered in abou'
20 hrt as measured by
BOD removal effi-
ciency
Plant recovered within
48 hr, as measured by
BOD removal effi-
ciency
_
_
-
Short-time effects on
digester
—
-
_
Gas production
dropped off rapidly.
At end of 42 days
only 75 ml/g of
volatile solids was
being produced.
Gas production
ceased to 7 Jays.
Digester then
gradually recovered.
-
Sustained damage
No damage noted
No damage noted
No
Yes
No
Yes, digester never
recovered
A summary of results shown in Table 10 is indicative of the
fact that, short of massive slug doses, chromate alone is unlikely
to harm the operation of a sound sewage treatment plant.
Concentrations of hexavalent chromium of up to 0.5 milligram
per liter were almost completely removed under conditions of the
study. At a 2.0-milligram-per-liter feed, hexavalent chromium was
occasionally found in small quantities in the effluent. With the 5.0-
C hromium
25
-------
milligram-per-liter and higher chromate feeds, variable-but-increasing
fractions of the chromium passed through the system to emerge as
either hexavalent or reduced chromium in the effluent. In view of
the present mandatory limit of 0.05 milligram per liter of hexavalent
chromium in drinking water, many situations exist in which total
reduction of hexavalent chromium to insoluble trivalent chromium may
be necessary. Where such chromium reduction and retention are
required, a modification of the activated-sludge system using chromate
as the oxygen source in an unaerated mixed liquor was found to
yield total reduction of chromate accompanied by loss of small amounts
of trivalent chromium dissolved or suspended in the final effluent.
While the systematic presence of chromate will halt nitrification
for short periods, nitrification was evident even when the feed chro-
mate level was 50 milligrams per liter. Chromate noticeably re-
strained the development of odor in the primary and the development
of Sphaerotilus in the mixed liquor. The use of the chemical for
this purpose is definitely not recommended.
The retention of chromium in the system occurred largely in
the activated-sludge solids. The chromium content of the primary
sludge solids (Table 4) invariably was lower than the chromium
content of the mixed liquor solids, with the latter showing more
than 10 times the chromium content of the primary sludge.
During the period when the biological chromate reduction and
removal system was operated, the activated sludge contained up to
18.4 percent chromium on a dry solids basis. Obviously, reduced
chromium has little or no toxicity to activated sludge. The digester
operated well with as much as 3.5 percent chromium in the solids.
Clearly, the total treatment system studied was resistant to and
tolerant of all but the most drastic stresses by chromate.
26 INTERACTION OF HEAVY METALS
-------
CHAPTER II. COPPER*
In the study of the effects of copper on sewage treatment, the
approach consisted of treating domestic sewage from a common
source in three replicate activated-sludge pilot plants. Controlled
additions of copper were made to two of the three plants; the third
served as a control. The effects of copper were measured by differ-
ences in effluent quality. The pilot plants included primary settling,
aeration with continuous sludge return, secondary settling, and an-
aerobic digestion. Thus opportunity existed for precipitation, reduction,
and complexing, as might occur during primary settling in an actual
situation before a biological process is reached. Effects are related
to metal additions to the incoming sewage rather than to metals added
to some specific plant component.
The activated-sludge pilot plants were designed to simulate
standard activated-sludge plants of the spiral-flow type. The shape
and dimensions of the activated sludge units, and a description of
the sewage feed are given in Chapter I.
COPPER SOURCES AND FORM
Copper could be present in domestic sewage and industrial waste
mixtures in several forms, depending upon its source and subsequent
reactions. In the most common electroplating process, copper is
deposited from cyanide baths. Copper in the form of copper sulfate
is used for manufacture of copper articles by deposition from solution
(electrodeposition), for recovery of copper from ore (electrorefining),
and, to a lesser extent, for electroplating. Plating from copper
pyrophosphate solution is also practiced. Plating baths are not simple
solutions of the copp,er compounds, but commonly contain a group of
materials among which are complexing agents. A listing of common
forms of copper and auxiliary bath chemicals is given in Table 11,
taken from Reference 9. In some of the baths the copper compound
would be highly ionized to the simple cation, whereas in others the
copper may be principally a soluble molecular complex or a complex
anion. Copper is used as a catalyst in synthetic chemical manufactur-
ing operations and may become associated with liquid wastes in some
undetermined form. Copper chloride is used in mercaptan removal
processes of the petroleum refining industry.
* Material in this chapter published previously in Journal Water
Pollution Control Federation. Washington, B.C. 20016. See Reference 13.
27
-------
Table 11. INDUSTRIAL PLATING AND
ELECTRO-DEPOSITING BATHS "
Plating
Plating
Plating6
Plating
Electro-
deposition
and plating
Electro-
deposition
Typical bath constituents
and proportions
Compound
CuCN
NaCN
Na2C03
KNaC4H406 • 4H20
Na OH to pH 12.6
CuCN
NaCN
HCN
NaCNS
NaOH
KOH
CuCN
NaCN
NaCNS
NaOH
Copper
Pyrophosphate
Oxalate
Nitrate
Ammonia
PH
CuS04 • 5H20
H2S04
Cu(BF4)
HBF4
H3B03
pH 1.2-1.7
Concentra-
tion,
mg /liter
26
35
30
45
75
45.8
57.8
9.8
15.0
21.0
75
84
9.4
19
22-38
150-250
15-30
5-10
1-3
8.2-8,8
150-250
45-110
224-448
15-30
15-30
Principal
forms
of copper
in bath
NaCu(CN)2
Na2Cu(CN)3
Na3Cu(CN)4
NaCu(CN)2
Na2Cu(CN)3
NasCu(CN)4
NaCu(CN)2
Na2Cu(CN)3
Na.iCu(CN)4
K6Cu(P20-)2
CuS04
Cu(BF4)2
Taken from Reference 9.
bPlating solution used in this study for cyanide plating solution tests.
Investigation of the effects of each of these compounds and
complexes of copper would be a formidable task; therefore, the study
was limited to one copper compound that yields the simple copper
cation in solution, e.g., copper sulphate, and to one copper complex,
e.g., Nan Cu(CN)n, with the expectation that the data would satisfy
the information need. Limitations on the permissible concentration
of cyanides are common. Research on the effects of high concentra-
tion of cyanide complexes would not have practical meaning in such
cases.
PLANT OPERATION
Sewage was fed to the units at a constant rate. Sludge from the
final settler was pumped continuously to the head of the aerator at
28
INTERACTION OF HEAVY METALS
-------
a rate of about 35 percent of the sewage feed flow. An automatic
device, activated once per minute, diverted the return sludge 5 to 7
percent of each minute to a collecting vessel. This procedure wasted
more than 25 percent of the suspended matter in the aeration tank
each day. Capacity and loading factors for the units of the plant are
given in Table 12.
Table 12. PILOT PLANT DESIGN DATA AND LOADING FACTORS
Unit
Primary settler
Aeration tank
Final settler
Loading factor
Capacity
Detention time
Surface overflow rate
Capacity
BOD loading
Aeration period
Capacity
Detention time
Surface overflow rate
4.6 gal
1.2 hr
142 gpd/ft2
23.6 gal
42-58 lb/day/1,000
ft3 aeration tank
volume
0.56 Ib/day/lb VS
under aeration
6 hr
7.9 gal
2hr
102 gpd/ft2
SAMPLE COLLECTION AND PRESERVATION
Samples for the routine measurement of BOD, COD, and SS
removal efficiencies of the units fed copper and the control were
collected by automatic mechanical samplers. The sampler diverted
the stream to be sampled to a compositing carboy for 15 seconds at 15-
minutes intervals. The samples were refrigerated and composited
over 24-hour periods. Analysis was started within 3 hours after the
end of the compositing period. Samples for studies of the effect
of slug doses were collected by the same means. Compositing periods
ranged from 4 to 12 hours to show peaks in effects.
Grab samples were taken for studies of the state of copper
(in solution or suspension) for analysis of the nitrogen forms and
for sulfide measurements. A complex sampling program was in-
volved in making trial balances between copper fed to the unit and
copper in the effluents plus accumulation of copper in the aerator.
The balances were usually made for 1-week periods. Samples of
each withdrawal of primary and excess activated sludge were com-
posited over the balance period. Samples of the final effluent were
collected by automatic sampler at 15-minute intervals and composited.
Grab samples of the aerator liquor at the beginning and end of each
balance period were taken for measurement of copper accumulating in
the system.
Copper
29
-------
ANALYTICAL METHODS
Unless otherwise stated, all procedures were essentially those
outlined in Standard Methods, llth Edition (10). Specific details were
as follows:
Biochemcial Oxygen Demand
The initial and final dissolved oxygen measurements were made
by the Alsterberg azide modification of the Winkler Method. Desired
concentrations of the samples were prepared by the cylinder dilution
technique. All BOD data reported are for samples incubated 5 days at
20° C.
Chemical Oxygen Demand
Primary feed and primary effluent samples were assayed by
use of 0.25 N dichromate. Final effluents were assayed by use of
0.025 N dichromate. Silver sulfate catalyst was not used. No cor-
rection for chloride was made. Chlorides in this sewage were normally
about 40 milligrams per liter.
Copper
Two methods of determining copper were utilized. High con-
centrations were determined by the usual iodimetric titration. Low
concentrations were determined by the colorimetric cuprethol method.
The organic matter was destroyed by fuming with sulfuric and nitric
acid. For determining soluble copper, the sample was passed through
an HA45 Millipore membrane. Copper in some forms will pass through
the Millipore filter, but will not react with the cuprethol. Digestion
to destroy complexes is necessary to determine total soluble copper
in such samples. When copper in the filtrate reacts with cuprethol
without digestion, it is termed reactive soluble copper.
Cyanide
Distillation as described in Reference 10 was used for pre-
liminary screening to separate the cyanide from interfering sub-
stances. Each sample was refluxed for two 1-hour periods. The
sum of the cyanide determined in each of the two 1-hour periods was
reported as the total cyanide. Complexed copper cyanide is slow
to be released and swept from the sample. At low cyanide concentra-
tions the efficiency of recovery of the cyanide from copper complex
in 2 hours of reflux would be especially low. Cyanide in the distillate
at levels above 2 milligrams per liter was determined by AgNO3
titration. Lower cyanide levels were determined by the pyridine-
pyrazolone colorimetric method.
30 INTERACTION OF HEAVY METALS
-------
EXPERIMENTAL DATA
Metals are present in sewage and industrial waste mixtures
because of sporadic receipt of metal-bearing wastes at the treatment
plant, such as those that would result from occasional dumping of a
spent plating bath or continuous losses of metal solutions from routine
operations (slug discharges being avoided). A combination of the
two is most probable. In consideration of these situations, investiga-
tions were made of effects from feeding copper at constant concen-
trations for periods up to 4 months and effects of doses of a few
hours duration. In the continuous feeding studies, periods of 2 weeks
or more were allowed between the initiation of a continuous metal
feed and the collection of the first samples. This permitted time
for possible acclimation or natural selection to occur. For the short-
duration slug-dose tests, 4 hours was selected as the metal feeding
period on the basis that it would approximate the drain time of a vat of
plating solution. The routine slug-dose runs were made on biological
sludges that had not previously received metal-bearing wastes.
During the continuous-feeding runs at each copper level, from
12 to 35 twenty-four-hour composite samples of the feeds, primary
effluents, and final effluents were collected and analyzed. Differ-
ences in sewage feed characteristics among the plants were usually
not greater than variations attributable to sampling and analytical
accuracies. Characteristics of the sewage feed for each series of
experiments are given in subsequent sections. The sewage feed
was generally near or slightly below pH 7, and the final effluent was
about pH 7.5. Removals accomplished in primary settling tanks
were typical of sewage treatment plants.
The results are presented subsequently in four sections differ-
entiated by the copper compound fed and whether a slug dose or a
continuous metal feed was used. Data on the effluents are presented
as cumulative percentage frequency curves. These curves permit
presentation of the complete data and make comparisons of varia-
tions convenient. Comparison of effluent characteristics of a metal
fed unit and the control on a day-by-day basis, that is, on samples
collected simultaneously, showed little coordination in variations.
For example, when the COD of the control plant effluent was high,
that of the metal fed unit was not consistently in the higher levels of
its variation range. This observation is in accord with reported
experience (11) in operation of 12 replicate trickling filters in which
there was no significant correlation between day-to-day variations in
sewage feed and that of the 12 effluents, nor was there significant
correlation among the 12 effluents on a day-to-day basis.
Most of the frequency distribution curves shown are arithmetic
plots. Trials indicated that for much of the data, the use of standard
probability paper, either arithmetic or logarithmic, was not helpful.
Copper 31
-------
In some instances it was necessary to define statistically the ex-
pectancy that differences in the data would result by chance. In
these cases probability paper was employed as part of the statistical
analysis to test the normalcy of distribution of observations.
Consideration was given to the possibility of the cyanide radical
per se exerting a damaging effect. Previous investigations have
shown that activated sludge that has become acclimated and adapted
to a feed containing cyanide would perform at normal efficiency and
would remove the cyanide (12). To be certain that this observation
was applicable to the exact circumstances of the metal studies, a
short study was made with the addition of sodium cyanide alone to
the sewage.
The rate and extent of biological acclimation to the presence of
copper have important bearings on the experiments and actual oc-
currences. In the case of cyanides, not only is acclimation involved,
but the rate of development of bacteria capable of utilizing the cyanide
influences the extent and duration of damaging effects. Theoretically,
the utilization of the cyanide radical could be accompanied by in-
creasing damaging effects as the metal is released from the complex.
It seems probable that the slug discharges would occur where metal-
bearing wastes were normally present to some extent. Whether a
system acclimated or adapted to a low level would be acclimated or
adapted to a slug level five times greater or 10 times greater than
the normal low level is also of interest. Although such questions were
not investigated exhaustively, some data pertinent to the phenomena
were obtained. Biological sludges acclimated to low metal con-
centrations were used in a few slug dose runs. Also in three of the
continuous metal-feeding tests, sampling was initiated simultaneously
with the start of a continuous metal-feeding run; thereby, information
was gathered on treatment efficiency during acclimation.
Checks were made on the concentration of cyanide in the primary
and final effluents. Following the acclimatization period, only traces
of cyanide were present in the final effluents, even at the higher
cyanide feed levels.
CONTINUOUS FEEDING
Copper Sulphate
Three concentrations of copper fed as copper sulphate were
studied, 10, 15, and 25 milligrams per liter. Characteristics of the
sewage feed during these runs are given in Table 13. BOD, COD,
and suspended matter data for the final effluents of the copper fed
units and the control are compared in Figures 8 and 9. The efflu-
ents of the units receiving copper fed as copper sulphate were pre-
dominantly of lower quality than those of the control units. The
average percentage reductions in BOD and COD in treatment are shown
in Table 14.
32 INTERACTION OF HEAVY METALS
-------
Table 13. AVERAGE CHARACTERISTICS
OF SEWAGE TREATED IN COPPER
SULPHATE STUDIES
Loading
BOD
COD
Suspended matter
Primary
settler
feed,
mg/liter
319
513
272
Primary
settler
effluent,
mg/liter
207
363
167
60
40 -
~
Q
O
CD
Q
O
U
D CONTROL
O 10 rug/liter Cu
A 15 mg/liter Cu
• 25 mg/llter Cu
0 10 20 30 40 50 60 70 80 90 100
% OF OBSERVATIONS £ STATED VALUES
Figure 8. Effect of copper fed as copper sulphate continuously
on BOD and COD of final effluents.
Copper
33
-------
3 200
>-
I-
Q
CO
IOO
-i -
CONTROL
""i i'
_L
_L
_L
10 20 3O 40 50 60 70 80 90 100
% OF OBSERVATIONS < STATED VALUE
Figure 9. Effect of copper fed as copper sulphate continuously on
suspended matter and turbidity of final effluents.
Table 14. EFFICIENCY OF ACTIVATED-
SLUDGE TREATMENT OF SEWAGE CON-
TAINING COPPER SULPHATE, COPPER
FED CONTINUOUSLY
Copper,
mg/liter
0
10
15
25
Avg BOD
removal, %
95.3
91.9
90.0
92.8
Avg COD
removal, %
85.2
81.5
80.1
81.1
34
INTERACTION OF HEAVY METALS
-------
The ultimate disposition or fate in the treatment process of
copper fed as copper sulphate during treatment is shown in Table
Table 15. FATE OF COPPER FED AS COPPER SULPHATE IN THE
ACTIVATED-SLUDGE PROCESS
Type and location of check sample
Copper found in outlet
Primary sludge, %
Excess activated sludge, %
Final effluent, %
Unaccounted for, %
Efficiency of copper removal, %
Soluble copper in primary effluent3
Total, mg/liter
Reactive, mg/liter
Soluble copper in primary effluenta
Total, mg/liter
Reactive, mg/liter
Copper in sewage feed
10 mg/liter
9
55
21
15
75
2.06
1.12
0.53
0.31
15 mg/liter
11
58
21
10
79
1.76
1.06
2,32
1.12
25 mg/liter
12
51
21
16
79
3.10
1.96
1.27
0.84
Soluble cop'per is defined as that passing an HA45 Millipore membrane. Total soluble copper is
that determined in the filtrate after acid digestion. Reactive soluble copper is that in filtrate
which reacts with reagents in absence of prior digestion.
15. Information on the average efficiency of the activated-sludge
process in removing copper fed as copper sulphate is included in
Table 15. Soluble copper in the effluents is also shown. The con-
centrations of copper, both total and especially soluble, were highly
variable among samples. In individual samples the concentration
was as much as 100 percent greater than the mean.
Cyanide Complex
Five concentrations of copper fed as copper cyanide complex
were studied, 0.4, 1.2, 2.5, 5, and 10 milligrams per liter. Char-
acteristics of the sewage feed during these runs are given in Table
Table 16. AVERAGE CHARACTERISTICS
OF SEWAGE TREATED IN COPPER
CYANIDE COMPLEX STUDIES
Loadings
BOD
COD
Suspended matter
Primary
settler
feed.
mg/liter
269
460
306
Primary
settler
effluent,
mg/liter
207
318
162
16. BOD, COD, suspended matter, and turbidity data for the final
effluents of the copper fed units and the control are compared in
Figures 10 and 11. The average percentage reductions in BOD and
COD in treatment are shown in Table 17.
35
-------
10 2O 30 40 50 60 70 80 90 100
% OF OBSERVATIONS 5 STATED VALUE
Figure 10. Effect of copper fed as copper cyanide complex continuously
on BOD and COD of final effluents.
36
INTERACTION OF HEAVY METALS
-------
o
I
(B
TURBIDITY, stu
SUSPENDED MATTER, nig/liter
(0 j-" o O
cn to Ji
a n
^
o crc
w CO
-^o
n
§
P) O
»n
i) O
pi TJ
n ^
Sn
ii
r
pi
x
>2
MO
S *
° "
CO
-q
-------
The BOD, COD. suspended matter, and turbidity data for the
final effluents of the units receiving 0.4 and 1.2 milligrams per liter
copper and the control were subject to statistical analysis. At the
0.4-milligram-per-liter copper level the differences between the
metal fed unit and the control were not statistically significant.
At 1.2 milligrams per liter of copper, all parameters differed sig-
nificantly, that is, the likelihood that the differences were due to
chance alone is very low. The data are plotted on logarithmic prob-
ability distribution paper in Figure 12.
120
100
80
Q
P, 60
40
2O
100
90
. 80
| 70
> 6°
cf 50
o
U 40
30
20Li
••• CONTROL
D 0.4 mg liter Cu
A 1.2 mg liter Cu
ARITHMETIC
PROBABILITY
PAPER
T 1 1 r
LOGARITHMIC
PROBABILITY
PAPER
_1_
_1_
U
_1_
_1_
_J_
_J_
10 20 30 40 50 60 70 80 90 95 98 99
% OF OBSERVATION < STATED VALUE
Figure 12. Effect of copper fed as copper cyanide complex
continuously on COD of final effluents.
38
INTERACTION OF HEAVY METALS
GPO 82O—663—4
-------
The ultimate disposition or fate in the treatment process of
copper fed as copper cyanide complex during treatment is shown in
Table 18. The average efficiency of the activated-sludge process in
Table 18 FATE OF COPPER FED AS COPPER CYANIDE COMPLEX IN
ACTIVATED-SLUDGE TREATMENT
Type and location of check sample
Copper fed found in outlet
Primary sludge, %
Excess activated sludge, %
Final effluent, %
Unaccounted for, %
Efficiency of copper removal, %
Soluble copper in primary effluent
Total, mg/liter
Reactive, mg/liter
Soluble copper in final effluent
Total, mg/liter
Reactive, mg/liter
Copper in sewage feed
0.4
mg/liter <
-
43
-
57
0.22
0.12
-
1.2
mg/liter
12.5
43.3
25.1
20
75
0.19
0.10
-
2.5
mg/liter
10.7
25.6
43.3
20
57
_
—
0.67
5
mg/liter
7
23
50
20
50
2.65
_
0.92
removing copper fed in the cyanide complex form is included in Table
18 as well as copper in solution in the effluents. The quantity of
copper in solution was highly variable among samples for each series.
The relationship between plant load and the effect of copper was
studied during a short period of operation of the pilot plants at about
double normal plant loadings.
Copper was fed as copper cyanide complex at 10 milligrams per
liter for this experiment. The data obtained are shown in Table 19.
Table 19. RELATIONSHIP BETWEEN PLANT LOAD AND THE EFFECT OF COPPER
FED AS COPPER CYANIDE COMPLEX"
Pilot plant
A.
Control
Copper fed
B:
Control
Copper fed
Plant loading, BOD
lb/ 1,000 ft 3
aerator
capacity, avg
48
48
100
119
Ib/day/lb VS
under
aeration, avg
0.56
0.56
1.37
0.98
BOD,
mg/liter
11
23
19
33
Final effluent quality5
BOD,
mg/liter
69
98
82
130
Suspended
matter,
mg/liter
6
29
20
32
Turbidity,
stuc
17
100
23
71
1 10 mg/1 fed continuously
1 Median values.
stu indicates standard turbidity units
Copper
39
-------
The increased plant loadings were obtained by doubling the sewage feed
rate; detention times, therefore, were half the values listed in Table
12. The characteristics of the sewage during this special experiment
were about the same as those listed in Table 16.
The normal experimental run with 10 milligrams per liter of
copper fed as the copper cyanide complex was followed by a special
investigation of the effects of cyanide alone. A substitution of sodium
cyanide was made for the copper cyanide in the sewage feed. The
quantity of sodium cyanide added was sufficient to make the con-
centration of cyanide in the sewage the same as that when the metal
complex mixture was fed, a concentration of 12.5 milligrams per liter
as CN-. After a few days the quality of the effluent improved until
it was not significantly different from that of the control.
A direct comparison was made between the effect on acclimated
systems of 1.2 milligrams per liter copper fed as copper sulphate
and the same copper concentration fed as copper cyanide complex.
This comparison was made with parallel operation of two units fed
copper and a control unit so that the sewage fed to each unit and
environmental factors were as alike as reasonably achievable. There
were no significant differences in the quality of the effluents from the
two copper-fed units.
Information on the acclimation or adaption phenomena of activated
sludge is given in Figure 13. Activated sludge with no history of having
received copper-bearing wastes was used for the studies of continuous
feeding of CuCN and CuSO4 at 1.2 milligrams per liter. The COD
of daily composite samples of the final effluents of these units is
shown in Figure 13 for several days immediately after the metal dose
was started. Also, the quality of the final effluent immediately follow-
ing an increase from 2.5 to 10 milligrams per liter of copper fed
as copper cyanide complex is shown.
Microscopic examinations of the activated sludge were made
occasionally for general appearance of the sludge and presence
of protozoa and rotifera. Protozoa and rotifera were absent during
the acclimation period for many of the runs, but even at the 5-milligram
per-liter level of copper a normal population density of the higher forms
eventually was present. The appearance of sludge from a unit receiving
1.2 milligrams per liter copper and greater, differed from that of
sludge in the control unit; it was characterized by dense small sludge
particles. The sludge from the copper-fed units settled rapidly.
The sludge density index for the activated sludge obtained during the
0.4-milligram-per-liter copper run averaged 1.4 and that during
the 1.2-milligram-per-liter copper run averaged 1.3. The copper-fed
unit never was troubled with settling problems caused by filamentous
forms, although filamentous bulking was a frequent problem in the
control unit.
40 INTERACTION OF HEAVY METALS
-------
ISO
^ 100
E
8 50
u
START OF CONTINUOUS FEED
-1
~ i
i
i
OF 1.2
mg/liter Cu AS CuCN COMPLEX
AVERAGE _
0
o
ISU
100
50
START OF CONTINUOUS FEED
1 OF 1.2 mg/liter Cu AS CuSO4
— '
I
-[
1
1
ACCLIMATED
AVERAGE
O
o
£UU
150
100
50
r 5
ACCLIMATED
AVERAGE FOR
- 2. 5 mg/liter
CuCN
1
START OF CONTINUOUS FEED OF 10 mg/liter Cu WITH
.SLUDGE PREVIOUSLY ACCLIMATED TO 2.5 mg/liter
Cu AS CuCN COMPLEX
ACCLIMATED AVERAGE
FOR 10 mg/liter Cu AS
CuCN COMPLEX
-2
TIME, days
Figure 13. COD of final effluents during acclimation to copper.
SLUG DOSES
Copper Sulphate
The effects of four slug doses of copper fed as copper sulphate
on the activated-sludge process were studied. Sludge with no history
of having previously received copper was used. The effects of the
four doses on BOD, COD, and suspended matter content of the final
effluents are shown in Figure 14. The sewage feed during these
runs had an average BOD of 264 milligrams per liter and an average
COD of 407 milligrams per liter.
The fate of the copper fed at 100 milligrams per liter is shown
in Figure 15. The primary sludge withdrawn 7 hours after start of
of the slug dose contained 31 percent of the copper fed. Material
balances in copper fed to and discharged from the primary settler in
the 24 hours after start of the slug accounted for 93 percent of the
copper fed. The aerator liquor solids contained a maximum of copper
8 hours after the beginning of the slug, at which time 46 percent of
Copper
41
-------
66 mg/lit.r Cu
100 mg/liter Cu
120
| 100
f eo
0 *°
S 40
20
O
320
280
J ?40
^200
U 120
80
40
• 0
^i 200
E
oT I6O
UJ
[; 120
1 80
Q
UJ 40
O
£ o
in— r i "
"71"
0. 0 20 40 6(
% TIME, hr
vo
V
" ""' *-w »
J
r
,
f
— i
-Jl
.1
0 20 40 6O
TIME, hr
NOTE: TIME 0 IS
1
1 1
T
j
j~
•
SO 100 0 20 40 60 80 0 20406OBOIOC
TIME, hr TIME, hr
THE START OF 4-hr SLUG DOSE
Figure 14. Effect of slug doses of copper fed os copper sulphate on the BOD, COD,
and suspended matter of final effluents.
the copper fed was associated with the solids in the aerator. At
96 hours the copper in the aerator solids had decreased to 20 percent
of that fed. Copper was slowly released from the aerator and ap-
peared in the final effluent.
Overall material balances including accumulation of copper in
the aerator accounted for about 75 to 80 percent of the copper at
seven sampling times during the 96-hour run. The cumulative quantity
of copper discharged in the final effluent during the 96 hours follow-
ing slugging was 20 percent of the copper fed. Because 20 percent
of the copper fed was associated with the aerator sludge at that time
and the concentration in the final effluent had dropped to low levels,
the process is about 75 percent effective in removing copper as copper
sulphate fed in a slug dose of this magnitude. The copper removed
was associated with the primary and excess activated sludges.
The fate of copper at a slug feed level of 410 milligrams per liter
is shown in Figure 16. The primary sludge withdrawn 7.5 hours
hours after the start of the slug contained 15 percent of the copper
fed. Material balances in copper fed to and discharged from the
42
INTERACTION OF HEAVY METALS
-------
Z E
H-J
OU-
\
en
z E
31
_l O
z^
o n:
i-o
r-
<
Q;
40
30
20
10
0
4
3
2
1
0
30
20
10
0
C
-
-
-
-
i i i i
-
-
-
-
-"
-
-
"
, i ill 1
f
1 1 1 I
) 20 40 60 80 IO
Ull-
-I Z
CD LiJ
10
? • -
3 "^
u |*
LJJ ..
_J \-
CO Z
D UJ
O _J
to U_
N
LL
•J
2
1
Q
1 '
.
-
\
-
*
, — n
0 20 40 5(
TIME, hr
TIME, hr
Figure 15. Fate of copper in activated-sludge treatment, slug dose of
100 mg/Mter fed as copper sulfate.
primary settler in the 24 hours after start of dosing accounted for
83 percent of the copper fed. Much of the copper was adsorbed
on the biological floe in the aerator. Twelve hours after start of
the slug about 50 percent of the copper fed was in the aerator. This
copper was predominantly associated with the biological sludge. The
sorbed copper was slowly released to the final effluent resulting in
an extended period during which copper at relatively low levels was
in the final effluent. Overall material balances 24 hours after start
Copper
43
-------
1 £_
10
8
4
o
1 1
-
[-
-
, 1
0 20 40
ISO
1 ^J W
100
50
o
i
60
i
1
1
80 100 120 140 150
0^0\ ,. mg/ iter IN AERATOR
\5 LIQUOR
_
mg/g (Jh
TOTAL SUSPENDED
./'MATTER IN EXCESS
tflo^'r, .ACTIVATED SLUDGE
i i
0 20 40
~~°-\
i
60
--o
i
240
•>
^ 200
01
E
-,£
~ lu
_l U-
< U-
o^
H£
2
a
80 100
160
120
80
40
I. , :
°0
TIME, hr
20 4O 6C
TIME, hr
Figure 16. Fate of copper in activated-sludge treatment, slug dose of
410 mg/liter fed as copper sulfate.
of the slug dose, including accumulation of copper in the aerator,
accounted for 75 percent of the copper.
Three grab samples taken from the final effluent during the
high copper concentration periods showed that an average of 61
percent of the copper discharged in the final effluent was in solution.
The cumulative quantity of copper discharged in the final effluent
during the 96-hour period following the start of the slug was 23
percent of the copper fed. The process is estimated to be about 65
percent effective in removing a slug dose of copper of this magnitude.
44
INTERACTION OF HEAVY METALS
-------
Cyanide Complex
The effects of three slug doses of copper fed as copper cyanide
complex were studied. The effects of the three doses on BOD, COD,
suspended matter, and turbidity of the final effluents from the activated-
sludge process are shown in Figure 17. The sewage feed during these
10 mg/liter Cu
ACCLIMATED SLUDGE
10 mg/liter Cu
NONACCLIMATED SLUDGE
60
^ 40
D
O
Q
O
(J
20
120
80-
40-
r-R-fT-.
n
i i i
[jfirrrrrr-i
1
r
T-llTHH
-
20 40 60 80 -
TIME, hr
20 40
TIME, hr
60
Figure 17. Effect of slug doses of copper fed as copper cyanide complex
on BOD, COD, suspended matter, and turbidity final effluents.
runs had an average BOD of 258 milligrams per liter and an average
COD of 380 milligrams per liter. Two of these slug doses were
made into activated sludges that had continuously received a 0.4-
milligram-per-liter concentration of copper fed as the copper cyanide
complex for several weeks preceding the slug test.
The maximum copper content of the primary effluent was about
equal to the dose concentration in eacli of three runs, and practically
Copper
45
-------
all of this copper was in solution.
was removed in primary settling.
A negligible part of the copper
Data on the copper concentrations in grab samples of final
effluents are shown in Figure 18. A great part of the copper passed
through the process and remained almost entirely in soluble form.
The efficiency of the process in removing copper was not greater
than 25 percent.
6
1 4
01
I- 2
u
o
10 mg/liter Cu
. . SLUDGE ACCLIMATED TO
A 0.4 mg/liter Cu
/\
A
/ V
6
1 4
01
E
u" 2
0
• | 1 i
~ 1 0 mg/liter Cu
NONACCLIMATED
•
-A
/ v^_
12
0
i
E 6
3-
U
4
2
o'
' 1 1
'A
>
25 mg/liter Cu
SLUDGE ACCLIMATED TO '
0.4 mg/liter Cu
_
• TOTAL Cu
o TOTAL SOLUBLE Cu
-
VA
0 20 40 60 80
TIME, hr
20- 40 60 8O
TIME, hr
Figure 18. Copper in final effluent following slug doses
of copper fed as copper cyanide complex.
DISCUSSION OF RESULTS
Reductions in treatment efficiency caused by copper fed as
copper sulphate at concentrations of 10, 15, and 25 milligrams per
liter were unexpectedly low. The reduction in efficiency averaged
4 percent or less. The data for each of the concentrations were
much alike. In fact, in some cases average reductions in treatment
efficiency were in reverse order of the copper concentrations. Initial
observations led us to think that copper cyanide complex had a much
greater effect than copper sulphate, but the effects proved to be
significantly different only for the acclimation or adaption period,
as discussed later. Reductions in efficiency of similar magnitude
prevailed at copper concentrations from 25 milligrams per liter
down to and including 2.5 milligrams per liter. The direct com-
parison of the effects of the two forms, the sulphate and cyanide com-
plex, in a parallel run at copper concentrations of 1.2 milligrams
per liter showed no significant difference in the effects.
46
INTERACTION OF HEAVY METALS
-------
The use of turbidity as a measure of effluent quality was initiated
after the project was partially completed, when it became obvious
that marked differences in turbidity were occurring, as shown in
Figures 10 and 11. The differences between turbidities of the control
and copper-fed units were greater and more consistent than any
other measurement. Increased turbidity occurred with increased
copper concentration.
The change in efficiency with copper level was so gradual that
a large number of observations and statistical definition of differ-
ences would be needed for bracketing the tolerance level to within
0.1 milligram. Copper may have an effect down to trace levels,
but the accuracy of measurement and sampling errors limit measure-
ment of the effect at very low levels. Small reductions in treatment
efficiency have little practical meaning because of limited accu-
racy in estimation of damaging effects on receiving waters. A per-
missible limit for protection of aerobic treatment efficiency can
be established appropriately, with little caution, since at high con-
centrations only about 7 percent reduction in BOD efficiency is in-
dicated. The maximum concentration of copper that can be received
continuously in normal domestic sewage without having a detectable
effect on common parameters of effluent quality is 1 milligram per
liter. Where turbidity is used as a treatment criterion, a maximum
copper concentration of about 0.8 milligram; per liter appears to be
necessary to obtain an effluent that is not significantly affected.
The activated- sludge process averaged from 50 to 79 percent
efficient in the removal of copper over the concentration range from
0.4 to 25 milligrams per liter based on analyses of the final effluents.
Some 30 to 50 percent of the copper passing through the process was
in soluble form. All the copper in soluble form in the final effluent
would not react with the analytical reagent directly. Digestion of
the filtrate preceding analysis consistently showed more copper.
This indicated that either complexes or colloidal copper passed through
the Millipore membrane. Where the cyanide complex was being fed,
it might have been anticipated that some of the soluble copper in the
final effluent would be cyanide complexes. The cyanide determination,
however, showed only traces of cyanide, and furthermore, consider-
able unreactive filterable copper was present in the effluents when
the copper was fed as copper sulphate.
At the 1-milligram-per-liter level of copper, the protozoa
and rotifera were present in normal numbers. The sludge settles
rapidly; in fact, it appears that sufficient concentration of copper
will prevent the growth of filamentous organisms responsible for
certain types of sludge bulking.
Copper fed as copper sulphate in a slug dose of 4 hours duration
at a concentration of 66 milligrams per liter in the sewage had but
Copper
-------
slight effect on the BOD, COD, or suspended matter content of the
final effluent. The 100-, 210-, and 410-milligram-per-liter doses
of copper as copper sulphate caused severe effects over the first
48 hours, somewhat in proportion to the copper dose. With the 100-
milligram-per-liter dose, efficiency of BOD removal fell off to
about 50 percent and normal operation was restored in about 120
hours. Even with doses as large as 410 milligrams per liter, the
sludge was not destroyed but recovered to normal in about 100 hours.
Slug doses of copper fed as copper cyanide complex had a much
more severe effect than the other slug doses at the same concentration.
The maximum COD and BOD of the effluent from units fed copper as
the copper cyanide complex at 10 milligrams per liter were almost as
high as when copper was fed as copper sulphate at 100 milligrams per
liter. The duration of the damages from the complexed metal fed at
the 10-milligram-per-liter level was about 24 hours. This damage
period is only one-fourth the period that resulted from the copper
sulphate doses. A slug of copper fed as the copper cyanide complex
at 25 milligrams per liter resulted in a very severe upset of the
plant. The COD of the effluent for one 4-hour period was consider-
ably greater than that of the feed. The duration of the effect was also
about 24 hours. The cyanide portion of the complex is apparently
much more toxic than the metal, or perhaps the important factor
is that cyanide keeps the metal in solution. When the amounts of
metal in solution are compared, it is seen that with the 100-milligram-
per-liter copper sulphate slug, copper was present in the final ef-
fluent at a maximum of only 3 milligrams per liter; whereas with a
25-milligram-per-liter dose of copper cyanide complex, soluble copper
was present in the final effluent at 10 milligrams per liter.
An explanation of the more prolonged effect of the copper sulphate
slug is found in the observation that the copper in the copper sulphate
was adsorbed by the sludge and retained in the unit for an extended
period. The copper in the cyanide complex, on the other hand, was
apparently not adsorbed to any appreciable extent and was carried
through the unit rapidly.
Once the biological system has become acclimated to the copper
and cyanide, and adapted to degradation of the cyanide radical, differ-
ences in effects between the two forms of copper disappear. This
accounts for the similar effects the two forms have in acclimated
systems.
Acclimation to a low copper cyanide level, 0.4 milligram per
liter, had no value in reducing the effect of a 10-milligram-per-
liter slug dose.
Massive slug doses are necessary to eliminate the higher
organisms from the sludge.
48 INTERACTION OF HEAVY METALS
-------
Slug observations indicate that slugs of 4-hour duration and
with up to 10-milligram-per-liter concentrations of copper as copper
cyanide complex or up to 50-milligram-per-liter concentrations of
copper as copper sulphate have a minor effect on efficiency and the
effect is not prolonged.
ANAEROBIC SLUDGE DIGESTION*
The digestion of copper-bearing sludges was studied by operation
of bench-scale sludge digesters on sludge feeds obtained from pilot
activated-sludge plants. Sewage from a common source was fed to
three replicate activated-sludge plants. Copper solutions were in-
troduced continuously to the feed of certain pilot plants to produce
selected constant concentrations. One plant was operated in parallel
with no metal addition to the feed. Sludge from this plant was fed
to digesters that served as controls. Differences in gas production
between the control digesters and those receiving copper-bearing
sludges were attributed to the presence of copper in the sludge.
The first part of this chapter, presenting the findings of an
investigation of the effects of copper on the activated-sludge sewage
treatment process, contained information on the sewage feed (13).
Procedure
The sludge digesters were 5-gallon glass carboys, which were
fitted with pumps for mixing the digester contents (Figure 19.) The
pump-mixing arrangement was superior to hand shaking in obtaining
representative samples and waste sludge. The digesters were main-
tained at 30° C in a constant-temperature room. Gas was collected in
a floating-cover gas holder, and the volume was measured daily at
atmospheric pressure and at 30°C. t
Primary sludge for digester feeds was withdrawn once each
day from the pilot plant primary settler. The withdrawn sludge was
settled for 30 minutes and supernatant decanted. This settled sludge
usually contained about 2 percent total suspended matter. The excess
activated sludge to be fed the digesters was withdrawn from the sec-
ondary settler once each day and fed to the digester without delay.
The total suspended matter in the secondary sludge digester feed was
0.5 percent or less.
* Remainder of this chapter published previously in Journal Water
Pollution Control Federation. Washington, D. C. 20016. See Reference
18.
Copper 49
-------
5-gol CARBOY
RECIRCULATING PUMP
GAS COLLECTOR
Figure 19. Experimental digester and gas collection apparatus.
A volume of 470 milliliters of mixed digester contents was
removed once each day from the digester receiving primary sludge
only. An equal volume of primary sludge was then fed. A volume of
8 liters was maintained in the digester, which corresponds to a
detention period of 17 days. The 470 milliliters of feed contained about
10 grams of volatile matter.
The digesters receiving combined primary and excess activated
sludges were fed 300 milliliters of primary sludge and 700 milliliters
of excess activated sludge each day. The two sludge volumes provided
a total of about 10 grams of volatile suspended matter, about 35 percent
of which was contained in the excess activated sludge. This is approxi-
mately the same relationship of primary and excess activated sludge
as was produced in the pilot plants. A sludge volume of 16 liters was
maintained in the digesters receiving the combined sludges, which
corresponds to a detention period of 16 days. One liter of mixed
digester contents was removed once each day and replaced with the
feed sludges. Digester loading information is summarized in Table
20.
50
INTERACTION OF HEAVY METALS
-------
Table 20. LOADING FACTORS FOR DIGESTERS
Item
Capacity, liters
Detention (days)
Loading, Ib VS/day/
1,000 ft3 of
digester volume.
Primary
sludge
digester
8
17
78
Primary and
excess activated-
sludge digester
16
16
39
The digesters were seeded originally with sludge from a municipal
sewage treatment plant. The digesters were fed for a week or more
with sludges from the activated-sludge plant, which had received no
addition of copper to its feed before the feeding of copper-bearing
sludge was started.
e
A sample of each digester feed and digested sludge withdrawn
was collected each day. These daily samples were composited for
weekly periods and analyzed for copper and for total suspended and
volatile suspended matter. Gas production on a per-unit-of-volatile-
solids basis was computed on a weekly basis, with gas production
lagging the feed compositing period by 1 day.
Since the digester was completely mixed when wasting sludge,
the accumulation of copper in the digester would follow the principles
of displacement of one material, A, from a homogeneously mixed
system by continuous addition of a second material, B. Accordingly,
after a period of time equal to the digester volume divided by daily
feed volume (one detention period), the sludge of the new origin (copper-
bearing) would constitute just over 60 percent of the sludge in the
digester. After four periods only a negligible percent of the original
sludge would be left. For this reason the digesters were operated
for over 60 days before an experiment was terminated.
Results
Some data on common digester parameters, such as pH, alkalinity,
volatile acids, and gas composition, were obtained. All such measure-
ments were not routinely made, however, since it was considered
that they would be symptoms of a damaging effect of copper that would
ultimately be reflected in gas production data. In no case were
liming or out-of-the-ordinary steps taken to correct any abnormal
performance condition.
High concentrations of copper were measured by the usual
iodimetric titration after complexing iron present with ammonium
bifluoride; low concentrations were determined by the colorimetric
cuprethol method. The organic matter was destroyed by fuming with
Copper
51
-------
sulfuric and nitric acid. For determining soluble copper, the sample
was passed through an HA45 Millipore membrane. Copper in some
forms will pass through the membrane filter, but will not react with
cuprethol. Digestion to destroy complexes is necessary to determine
total soluble copper in such samples. Colloidal copper in sizes that
pass the membrane filter could have been present. Copper in the
filtrate that reacts with cuprethol without digestion is termed reactive
soluble copper in this paper.
Copper was fed to the sewage in two forms, i.e., copper sulfate
and copper cyanide complex, NanCu(CN)n. Restrictions are usually
placed on the discharge of wastes containing cyanides to sewerage
systems, because of potential hazards through release of HCN. These
restrictions limit the cyanide to low levels so that experimentation
at higher cyanide levels would not ordinarily be of practical value.
In the case of the copper cyanide complex, there is the possibility
that the cyanide per se could affect the anaerobic organisms or that
the cyanide in combination with the copper would have a synergistic
effect. It has been previously reported that 16 milligrams per liter
of cyanide fed as sodium cyanide has no effect on sludge digestion
after an acclimation period (14). This concentration is greater than
maximum concentration of cyanide used in these sludge digestion
studies.
Gas production from digestion of primary sludges obtained
from sewage containing 10, 15, and 25 milligrams per liter of copper
are shown in Figures 20, 21, and 22. Gas production from digestion
I.4OO
1,200
1,000
800
600
RUN
RUN
~~i—i—i—i—i—i—i—[—r~
RUN IV
0 10 20 30 40 0 10 20 30 0 10 20 30 0
'DIGESTION INTERRUPTED BY TIME, days
MECHANICAL DIFFICULTIES
10 20 30 40 50 60 70 80 90 '00
Figure 20. Performance of digester fed primary sludges from unit receiving
10 mg/liter copper fed as copper cyanide complex.
of combined primary and excess activated sludges from sewage fed
5, 10, 15, and 25 milligrams per liter of copper is shown in Figures
23, 24, 25, and 26. The graphs include observations of digesters fed
copper as copper cyanide complex and as copper sulfate, as designated
in the titles.
52
INTERACTION OF HEAVY METALS
-------
•t
3
2
1
0
~l 1 1 1 1 1 1 1 1 1 1 1 1
1
,J '
, f -" Ft
~ r—3t'- \ t/il
Z/
f\'s
T77
',/
,-
1
p
t //
'
' ^
<{<•:
• ",'•
• •'.
i i i i i r ~
1 1 1 FED SLUDGE FROM UNIT
| Lv-l RECEIVING 15 mg/liter Cu
4 P^ FED SLUDGE FROM UNIT
'i ly. RECEIVING 10 mg/liter Cu
,^ —
,\ 1
1
"'
'1
11
xl t
— I
-
i
-
— '
l
—
-
i Ol— i i
-30 -20 -10
30 40 50 60
TIME, days
70 80
Figure 21. Performance of digester fed primary sludge from unit receiving
15 mg/liter copper, fed as copper sulfate.
C£
LU *" 4
^ 3
Q u_ Z
u
- ,471 I
i
. .
f t
7Q^ ——START OF Cu FEED
600
0 500
UJ
u- 400
> 300
0 POO
100
-r™j—i
1
>
f
1
1
1 1 1
• • •
r-NOT FED ON 18
1 [ DAYS OF THIS
L,
1 PERIOD
t
f t
1
.
I
1
t
-10 0 10 2O 30 40 50 60 70 80 90
TIME, days
t
•
-V-
1 1
• •
1
(
SPOT SAMPLES
COMPOSITE
SAMPLES
. ,
1
,
IOO 110 120 13
Figure 22. Performance of digester fed primary sludge from unit receiving
25 mg/liter copper, fedos copper sulfate.
Copper
53
-------
I I I I I I I I
40 50 60
TIME, days
Figure 23. Performance of digester fed combined
sludges from unit receiving 5 mg/liter copper,
fed as copper cyanide complex.
54
INTERACTION OF HEAVY METALS
GPO 82O—663—5
-------
UJ LO
}- O
£2
00
zo
3 6?
u
4
3
2
1
0
900
800
700
600
500
400
300
200
100
n
-
<
™
J
"
1
1
1 .
1
1 1
1 . 1
-
TO
1250
1.
. SPOT
|
--V
SAMPLES
COMPOSITE SAMPLES ~
-
-10
10 20
30 40 50
TIME, days
60 70 80 90 100
Figure 24. Performance of digester fed combined sludges from
unit receiving 10 mg/liter copper, fed as copper sulfate.
6
a:
H 5
UJ- 4
°0
aS 3
^fe *
n
f 1
[,,,.,
1 •
1
-
- BT
"" Ljjjj;
Smi&.
'
11
«K-
i^
f 4
"•''»
'''/ ^
'%'%<
\
1
''/\ i
r__
f
|
I
III 1 1 ! 1
• SPOT SAMPLES
• F ICOMPOSITE SAMPLES
P. -
|
1
1
1
I
1 1
— ^
I
1
-
-
START OF Cu FEED AT 15 i
800
700
6OO
500
400
3OO
200
100
0
1 ' 1 '...' ' l~~! FED SLUDGE FROM UNIT \ ' '
"" ' '
- ^
-f#
r — ;
-^
|
~m
r'l
fl
1 RUN 1 i-Vi RECEIVING 15 mg/liter Cu |— ,
|^'
^
\
\
\
1
•A t
W A FED SLUDGE FROM,
•"^•UNIT RECEIVING
10 mg/liter Cu
I IFED SLUDGE FROM
La
CONTROL UNIT
f
irr^n ,
i
rl
i
1
I
1
RUN II -
—
-
~
-
T
-30 -20 -10
20 30 40 50 60 70 80
TIME, days
10 20 30
Figure 25. Performance of digester fed combined sludges from
unit receiving 15 mg/liter copper, fed as copper sulfate.
Copper
55
-------
Q u-
Z O
6
5
4
3
2
800
700
Q 600
"m
; "- 50O
>•> 400
' ° 300
20O
100
0
1 II 1 1 1 1
-
-
1
i
•
i i
•*- START OF Cu FEED
1 II 1
-
T
-
1 — 1
ii,
i
i
.i ;
•
•
-
i
-*- START OF Cu FEED
1 '
i
1
\
\
i
-
• SPOT SAMPLE
COMPOSITE SAMPLES
hr" -
1,
-10 0 IO 20
-10 0 10
TIME, days
20
Figure 26. Performance of digester fed combined sludges from
unit receiving 25 mg/liter copper, fed as copper sulfate.
Gas production from sludges obtained in parallel operation,
but with no copper added, varied in weekly averages from about
600 to 900 milliliters per gram of volatile solids fed.
As previously described, all digesters were initially fed control
sludges until they functioned satisfactorily; then the feeding of copper-
bearing sludge was started. Gas production data for the development
period are shown in the graphs. In one case a digester was operated
for several weeks on sludge from sewage with 10 milligrams per liter
of copper and then continued on sludge from sewage with a higher
copper level. Periods of subnormal gas production were observed
during the sludge development period. These were followed in
some instances by abnormally high gas production periods, which
resulted from accumulation of undigested material during the preceding
period.
Where serious reduction of gas production occurred, the ex-
periment was repeated and a confirming observation made. In the
digestion of primary sludge from sewage containing 10 milligrams
per liter of copper fed as copper cyanide complex (Figure 19), four
runs were made. In the first run the digester was functioning normally,
but a mechanical failure ended the run prematurely. In runs II and
III a severe effect on the digester was indicated shortly following
initiation of the feeding of the copper-bearing sludge. These runs
were initiated on digesters with short development histories, and
there was a question as to their being in normal operation when
56
INTERACTION OF HEAVY METALS
-------
metal feeding was started. In the fourth run normal digestion con-
tinued throughout an extensive period. It appears that 10 milligrams
per liter of copper is near the level at which digestion is signi-
ficantly affected. The two failures suggest that copper is more likely
to affect the digestion process during the initial development stages
than after the digestion process is well established.
Data on the copper content of the digested sludges are shown
in Figures 21 through 26. The average copper concentration in the
sludge feeds is given in Table 21. Copper concentrations in sludges
are proportional to the concentration of suspended matter in the
Table 21. COPPER IN SLUDGES FED TO DIGESTERS
Copper in
sewage,
rag/liter
5
10
10
15
25
Form of
copper fed
CuCN
Complex
CuCN
Complex
CuS04
CuSO*
CuSO4
Primary sludge
Copper
% of total
mg/liter suspended
matter
73 0.32
140. 0.76
288 0.89
230 0.83
490 2,1
Total
suspended
matter
mg/liter
23,000
19,000
32,000
28,000
23,000
Excess activated sludge
Copper
mg/liter
89
160
210
430
% of total
suspended
.matter
1.8
6,5
6.2
13.1
Total
suspended
matter,
mg/liter
5,000
2,500
3,400
3,300
sludge since copper is predominantly a part of the suspended matter
in sludges. In comparisons of data from several plants, the expression
of copper content on a per-unit-weight-of-suspended-matter basis is
advantageous. The basis could be total suspended matter or volatile
suspended matter; the former was chosen. Any ratio of copper to
solids has disadvantages as a basis for relating digester performance
to copper content. One disadvantage is that as interference with
digestion occurs, less material is destroyed in the digester. Organic
matter accumulates; and as digestion grows progressively worse, the
quantity of copper per unit of suspended matter in the sludge de-
creases. The more critical item is that it seems logical that toxicity
would be a matter of copper in solution in the water surrounding
the microorganisms and, therefore, for soluble copper, the con-
centration basis would be most useful; however, the establishment
of a relationship between effect on digestion and copper in solution
is complex. The concentration of copper in solution in a sludge feed
would be significantly altered in the digester by different environ-
mental conditions, such as pH, alkalinity, and presence or produc-
tion of precipitating or complexing agents. A complicating factor
is that copper may be adsorbed on sludge. It was demonstrated that
portions of the sorbed copper could be brought into solution by pro-
longed agitation. Such sorbed copper could be as important as copper
in solution if it returned to solution as adsorption equilibrium was
established in the digester.
Copper
57
-------
Discussion
This research was intended primarily to relate copper con-
centration in raw sewage to sludge digestion difficulties. It appears
judicious to limit comment on soluble copper in digesters to stating
that the maximum concentration of copper in solution (reactive soluble
copper) in infrequent grab samples of digesters with normal gas
production was 0.7 milligram per liter. This contrasts sharply
with the concentration values of several hundred milligrams per
liter of total copper in many samples. The subject of soluble copper
in digesters will be more extensively considered in later studies.
Data on the relation of copper in sludges and digestion or treat-
ment difficulties may be a useful by-product of the work. A sludge
sample, particularly digested sludge, represents a composite ac-
cumulated over a long period of time. Thus, measurement of copper
in the sludge may provide a means of estimating the average concentra-
tion of copper in sewage received over an extensive period. Sub-
sequently, a judgment of limited certainty on whether or not copper
is responsible for treatment difficulties can be made on the basis
of sludge analysis.
Studies of the effect of slug doses of copper on sludge digestion
were made in association with the studies of the effect of slug loads
on the activated-sludge process (13). Slug doses of 4 hours in
duration and ranging up to 410 milligrams per liter of copper were
fed to the activated-sludge plant. Primary sludge and activated
sludge were collected at the approximate times of maximum con-
centration of copper, as the slug progressed through the process.
These sludges were promptly fed to sludge digesters, which were
operating normally on control sludge. In no case was the gas pro-
duction reduced following feeding of these copper-bearing sludges.
Experiments on the effect of slug doses were limited to the
one form of copper, CuSO4, because the hazard involved with slugs
of cyanide would make their occurrence in practice rare.
Summary
Copper continuously present in concentrations ranging from
0.4 to 25 milligrams per liter in the raw sewage entering a complete
pilot activated-sludge treatment plant reduced BOD removal efficiency
zero to 7 percent, roughly in proportion to metal concentration,
after the plant became acclimated to the metal. Two forms of copper,
copper sulphate and copper cyanide complex, had about the same
effects after the system had become acclimated. The process was
50 to 79 percent efficient in removal of copper. From about 25
to 75 percent of the copper in the final effluent was in solution.
58 INTERACTION OF HEAVY METALS
-------
Four-hour slug doses of copper as copper sulphate in con-
centrations greater than 50 milligrams per liter had severe ef-
fects on the efficiency of an unacclimated system. The system re-
turned to normal in about 100 hours. Slugs of copper cyanide complex
had much more severe maximum effects, but the system returned
to normal in about 24 hours. Copper in slug doses of copper sulphate
is largely adsorbed by the activated sludge and slowly released,
whereas copper cyanide slugs pass through the system in normal
detention times.
The maximum concentration of copper that does not have a
detectable effect on treatment efficiency is concluded to be 1 milli-
gram per liter. Slug doses of a few hours duration with up to 50
milligrams per liter copper as copper sulphate or 10 milligrams
per liter copper as copper cyanide complex have but a slight effect
on treatment efficiency.
The digestion of sludges obtained from sewage to which copper
in known concentrations was fed continuously was observed. Digester
performance, as measured by gas production, is indicated in Table 22.
Table 22. GAS PRODUCTION OF
DIGESTERS FED COPPER
Copper
in
sewage .
mg 'liter:
5
10
15
25
Primary
sludge
Normal
Normal
Subnormal
Subnormal
Combined primary and
excess activated
s ludge
Normal
Subnormal
Subnormal
Subnormal
Slug doses of copper in the sewage did not affect digestion of
the resultant sludges when the sludges were fed as a single feed to
a normally operating digester. The maximum slug dose tested was
410 milligrams per liter of copper fed as copper sulfate.
Copper
59
-------
CHAPTER III. ZINC*
The efficiency of treatment of sewage containing zinc was studied
by operation of pilot activated-sludge plants. Sewage from a common
source was fed to three replicate plants. Zinc solutions were introduced
continuously to the feed of two of the plants to produce selected constant
concentrations; one plant was operated with no metal addition to the
feed. Differences in effluent quality, as measured by BOD, COD,
suspended solids, and turbidity, between the zinc feed units and the
unit receiving the unaltered sewage were attributed to the presence of
zinc in the feed.
The digestion of the zinc-bearing sludges was studied by operation
of bench-scale digesters on sludge feeds obtained from the activated-
sludge plants. Differences in gas production between the digesters
receiving control sludge and those receiving the zinc-bearing sludges
were attributed to the zinc in the sludge.
The objectives of the research were to determine the level of
zinc that can be tolerated in waste waters without reducing the efficiency
of biological processes in removing the organic matter or in stabilizing
the sludges, and to determine the efficiency of the process in removing
zinc.
PLANT DESCRIPTION AND OPERATION
The pilot activated-sludge plants were designed to simulate
standard activated-sludge plants of the spiral-flow type. The acti-
vated-sludge plant included: primary settling, aeration with continuous
sludge return, and secondary settling. Thus opportunity existed for
precipitation, reduction, and complexing such as might occur during
primary settling in an actual situation before a biological process
is reached. Effects were related to metals added to the incoming
sewage rather than to metals added to some specific plant component.
The units are illustrated in Figure 2. Capacity and loading factors for
the plant are given in Table 23. Sewage was fed to the units at a constant
rate. Sludge from the final settler was pumped continuously to the
head of the aerator at a rate of about 35 percent of the sewage feed
flow. An automatic device, activated once per minute, diverted the
return sludge about 5 percent of each minute to a collecting vessel.
*Material in the chapter published previously in Proceedings of 17th
Industrial Waste Conference. Purdue University. See Reference 16.
61
-------
Table 23. PILOT-PLANT DESIGN DATA AND LOADING FACTORS
Unit
Primary settler
Final settler
Loading Factor
Capacity
Detention time
Surface overflow rate
Capacity
BOD loading
Aeration period
Capacity
Detention time
Surface overflow rate
4.6 gal
1,2 hr
142 gpd/ftz
23.6 gal
69 Ib /day/1, 000 ft*
aeration tank volume
0.87 Ib/day/lb VS under
areation
6 hr
7,9 gal
2 hr
102 gpd/ft2
This procedure wasted about
in the aeration tank each day.
25 percent of the suspended matter
The sewage available in the area where the pilot plant was
located, while of essentially domestic origin, had an average BOD
of about 75 milligrams per liter because of high ground-water in-
filtration. Fortification of this sewage with dog food produced a sewage
showing characteristic behavior and adequate strength. Dry granular dog
food was ground, soaked in water overnight, homogenized in a large
blender, and added to the sewage at a rate of 1.2 grams(air dry) per
gallon.
The sludge digesters were 5-gallon glass carboys, fitted with
pumps for mixing the digester contents. The digesters were maintained
at 30° C in a constant temperature room. Gas was collected in a
floating-cover gas holder, and the volume was measured daily at
atmospheric pressure and at 30°C. Digesters were operated on
primary sludge feeds alone and on combined primary plus excess
activated sludge. Sludge from the primary settler was withdrawn once
per day. Excess activated sludge was withdrawn from the secondary
settlers once each day and fed to the digester without delay. A volume
of mixed digester contents equal to the sludge fed was removed once
each day from the digesters prior to feeding. About 35 percent of the
volatile matter in combined sludge feeds was from excess activated
sludge and 65 percent, from primary. This is approximately the
same relationship of primary and excess activated sludge as was pro-
duced in the pilot plants. Digester loading information is summarized
in Table 24.
62
INTERACTION OF HEAVY METALS
-------
Table 24. LOADING FACTORS FOR DIGESTERS
Item
Capacity, liters
Detention, days
Loading, Ib VS /day/1 ,000
ft 3 of digester volume
Primary
s lud ge
digester
8
17
88
Primary and
excess activated-
sludge digester
16
16
40
ZINC SOURCES AND FORM IN LIQUID WASTES
A principal source of liquid wastes containing zinc is the metal
plating industry; two types of electroplating baths are used, acid
baths and alkaline cyanide baths. Acid zinc baths are used most
extensively in galvanizing steel wire and strip. Almost all acid
zinc-plating processes employ sulfate, chloride, or mixed chloride-
sulfate baths. The alkaline cyanide solution used in the major alkaline
processes is a mixture of sodium zincate and zinc cyanide complexes,
with an excess of sodium cyanide and hydroxide. A zinc-mercury
process is also employed. The plating solution is similar to the
conventional alkaline cyanide bath with the addition of a mercury
salt equal to a ratio of mercury to zinc of about 1 to 100.
Zinc is also present In wastes from the manufacture of organic
materials such as acrylic fiber, rayon, cellophane, and special synthetic
rubbers. The historical cases of water pollution by zinc involved
wastes from mining and ore processing. Corrosion of galvanized
iron pipe used in household and factory distribution systems may
contribute a significant amount of zinc to waste waters.
In this investigation zinc was fed to the sewage in two forms,
zinc, sulfate and in the form found in a typical alkaline cyanide
The plating bath formulation was as follows:
as
plating bath (9)
Item
Zinc cyanide, Zn(CN)2
Sodium cyanide, NaCN
Sodium hydroxide, NaOH
Grams per liter
60
23
53
The zinc is present in this bath in the following forms:
+ Zn(CN);
4(CN)~
Na2Zn(CN)4 + 4 NaOH ^ >-Na2ZnO2 + 4NaCN
Na2ZnO2^=±:2Na+ + ZnO?
ZnO2= + 2H.O, » 7n"+ 4(OH)~
2H2 O
Zinc
63
-------
Experimentation with zinc in concentrations greater than a few
milligrams per liter in the form involving cyanide and cyanide com-
plexes has limited practical significance because of limitations usually
imposed on permissible concentrations of cyanides in sewage on
the basis of health hazards.
When zinc and cyanide combinations are used in experiments,
the possibility of effects from the cyanide ion itself must be considered.
Previous investigations have shown that cyanide in sewage is biologically
destroyed in acclimated aerobic biological systems; and once a
system is acclimated, the cyanide has no significant effect on treatment
efficiency (12). The effects of cyanide on anaerobic processes has also
been investigated. A cyanide concentration of 16 milligrams per liter
in the sludge feed was reported to have no effect if the digester is
first acclimated by low initial doses (14).
SAMPLE COLLECTION AND ANALYSIS
Samples for routine removal-efficiency measurements of BOD,
COD, and suspended matter were collected by automatic mechanical
samplers, which diverted the stream to be sampled to a refrigerated
compositing carboy for 15 seconds at 15-minute intervals. The
samples were composited over 24-hour periods. For measurements
of the state of zinc ( in solution or suspension), grab samples were
collected. The filtrates containing the zinc in solution were composited
for weekly periods. An extensive sampling program was used to
balance the zinc fed the unit with the zinc in the effluents plus ac-
cumulation of zinc in the aerator. The balances were usually
made for 1-week periods. Samples of each withdrawal of primary
and excess activated sludge were composited over the balance period.
Grab samples of final effluent were collected once per day for
turbidity analyses. These samples were settled for an additional 30
minutes in beakers, and turbidity measurements were made on the
decanted supernatant. Thus the efficiency of the final settlers was
not involved in the measurement. The turbidity reported is that from
material not removable by practical sedimentation methods.
Each digester feed and digested sludge withdrawn was sampled
each day. These daily samples were composited for weekly periods
and analyzed for zinc and for total suspended and volatile suspended
matter. Gas production per gram of volatile solids fed was computed
on a weekly basis with gas produced for a weekly period lagging by
1 day the feed compositing period. Grab samples of digested sludge
were also collected for metal analyses.
All analytical procedures with the exception of zinc were essen-
tially those outlined in Standard Methods, 10th and llth Editions
(1,10). In the BOD test the dissolved oxygen measurements were
64 INTERACTION OF HEAVY METALS
-------
made by the Alsterberg azide modification of the Winkler method.
Desired concentrations of the samples were prepared by the cylinder
dilution technique. All BOD data reported are for samples incubated
5 days at 20° C. The COD determinations of plant feed and primary
effluent were assayed using 0.25 N dichromate; final effluents were
assayed using 0.025 N dichromate. Following 1 hour of refluxing,
silver sulfate was added. The COD's were all corrected for chloride
oxidation.
The zinc content of the various samples was determined with
a recording polarograph. A dropping mercury electrode, with a
3-second drop time, in conjunction with a saturated calomel electrode
was used to record the current voltage curve. The samples for
polarographic assay were wet-ashed with a nitric-sulfuric acid mix-
ture; resistant samples were treated with perchloric acid. The
samples were taken to cessation of fumes on an electric hot plate.
The polarographic assay consisted essentially of dissolving the acid-
digested sample in 1-M NILiCl-NH-iOH electrolyte, filtering, and
recording the current voltage curve between -1.2 and -1.6 volts.
The height of the diffusion curve at-1.4 volts was used as the measure-
ment of zinc. Triton X-100 was used as maximum suppressor.
Addition of zinc to various samples gave satisfactory recovery.
The background zinc content of the sewage used in this study was
approximately 0.1 milligram per liter. To determine soluble zinc,
the samples were passed through an HA45 Millipore membrane.
Samples for cyanide determinations were first treated to separate
interfering substances; then each sample was refluxed for two 1-hour
periods. The sum of the cyanide determined in each of the two 1-hour
periods was reported as total cyanide. Good recoveries in the first
hour were demonstrated.
ZINC AND ACTIVATED-SLUDGE TREATMENT
Continuous Feeding
An experimental run of the activated-sludge plant was made at
each of three levels of zinc in the sewage feed, 2.5 10, and 20 milli-
grams per liter. The zinc was fed in the form of zinc sulfate in each
concentration; in addition, zinc in the form of the alkaline cyanide
plating bath formula, referred to as complexed zinc, was fed at the
10-milligram-per-liter zinc level. These three runs were made with
continuous addition of zinc to the sewage. For each zinc level, 2
weeks or more was allowed to pass between the initiation of feeding
the metal to a normal sludge and the collection of the first samples
for use in obtaining data on effluent quality for an acclimated system.
Fifteen to thirty-four 24-hour composite samples of the feeds, primary
effluents, and final effluents were analyzed. The average BOD, COD,
suspended matter, and turbidity values for the final effluent of the
Zinc 65
-------
Table 25. QUALITY OF FINAL EFFLUENTS FROM CONTROL AND ZINC-FED UNITS
Zinc
in
sewage,
mg/liter
0
2.5
0
10
10
0
20
Form
of
zinc
added
Control
ZnSO<
Control
ZnS04
Complexed
zinc
Control
ZnS04
Avg
BOD,
mg/liter
13
15
13
18
22
11
15
Avg
COD,
mg/liter
39
40
44
49
57
58
68
Avg
suspended
matter,
mg/liter
7
8
10
17
16
7
16
Avg
turbidity,
stu
18
22
16
17
27
18
46
control and zinc fed units are presented in Table 25. Differences in
sewage feed characteristics among the plants were predominantly no
greater than differences attributable to sampling and analytical varia-
Table 26. CHARACTERISTICS OF SEWAGE FEEDS AND PRIMARY EFFLUENTS
Zinc
mg/liter
0
2.5
0
10
10
0
20
Zinc
Control
ZnSO*
Control
ZnSO,
Complex-
ed zinc
Control
ZnSO
BOD
Primary
feed,
mg/liter
289
313
228
271
245
262
268
Primary
effluent,
mg/liter
184
195
162
182
179
199
236
COD
Primary
feed,
mg/liter
483
492
469
555
552
512
517
Primary
effluent,
mg/liter
306
315
323
342
376
390
426
Suspended
matter
Primary
feed,
mg/liter
299
326
240
394
331
265
274
Primary
effluent,
mg/liter
157
169
148
195
208
179
179
tions. Characteristics of the sewage feed are given in Table 26. The
sewage feed was generally near or slightly below pH 7.0, and the final
effluent near 7.5.
Analytical data for the run with a zinc concentration of 2.5
milligrams per liter are presented as cumulative percent frequency
plots on logarithmic probability paper in Figure 27. This presentation
66
INTERACTION OF HEAVY METALS
-------
Q
O
3O
20
Q
O
90
8O
70
60
5O
4O
3O
2O
BO
70
60
50
4O
3OI
I I I l
i I I I
2.5 mg/liter Zn
CONTROL
5 10 EO 30 40 50 6O 70 80 90 95 98
% OF OBSERVATIONS < STATED VALUE
Figure 27. Cumulative frequency data on quality
of final effluents with zinc concentration
of 2.5 mg/liter fed as zinc sulfate.
of the complete data makes comparisons of variations convenient.
From Table 25 and Figure 27 a slightly lower quality of effluent
in the unit fed zinc may be infer red. Statistical analysis of the variations
Zinc
67
-------
in the data indicates a strong likelihood that these differences could
have occurred by chance alone in randomly selecting 30 some values
from an infinite number of measurements; therefore, from both
statistical inference and practical considerations, the indicated dif-
ferences are considered insignificant.
6O
5O
4O
30
« 2O
0
o
90
SO
70
60
5O
40
30
o
u
20
IO
1 1 1 1 1 T
I I I I I
I
I
I I I I I
I
I
5 10 20 30 40 50 60 70 80 90 95
% OF OBSERVATIONS^ STATED VALUES
Figure 28. Cumulative frequency data on quality of final
effluents with zinc concentration of 10 mg/liter in sewage feed.
98
68
INTERACTION OF HEAVY METALS
-------
Some of the runs with zinc concentrations of 10 milligrams per
liter fed as zinc sulfate and the same concentration fed as the alkaline
cyanide bath were made simultaneously in parallel. These data are
shown as cumulative frequency plots in Figure 28. The differences
between the data for the two forms of zinc were not significant. The
differences between the data with either form of zinc and the control
are significant, statistically speaking. They have a low likelihood of
having occurred by chance alone.
Correlation coefficients were computed for aerator loading,
aerator suspended-matter level, temperature, COD, and BOD values
of final effluents for each of the runs. The correlations were not
significant, since no substantial relation between the variables was
indicated. The differences that occurred in aerator solids level
and loadings among the units did not, therefore, appreciably affect
results.
The ultimate fate of zinc in the treatment process during the
runs with zinc sulfate concentrations of 2.5 and 10 milligrams per liter
is shown in Table 27. A minor part of the zinc was removed in
Table 27. ZINC DISPOSITION IN PROCESS OUTLETS
Zinc
in
sewage,
ing/liter
2.5
10
2.5
10
Form
of
zinc
fed
ZnSO4
ZnS04
ZnSO4
ZnSO4
Primary
sludge
Excess
activated
sludge
Final
effluent
Imbalance
Zinc, mg/liter
64
375
120
328
0.12
0.88
-
Zinc fed in outlet, %
13
14
85
63
5
9
+12
-14
primary settling; a large part became associated with the aerator
liquor suspended matter. These figures are the mean values of the
weekly material balances in zinc described previously. The im-
balance data refer to the average degree of success in the balances
between zinc fed and zinc accounted for. Zinc in the sewage supply
was not included in the balance because measurements indicate a
level of about 0.1 milligram per liter, which, at the most, is 4 percent
of the zinc fed. This background zinc may in part account for the
positive imbalance at the 2.5-milligram-per-liter level.
The average efficiencies of the processes in removing zinc,
based on zinc determinations in effluents, are given in Table 28.
The complexed zinc was as easily removed by an acclimated system
Zinc
69
-------
Table 28. EFFICIENCY OF PROCESS IN ZINC REMOVAL
Zinc
in
sewage,
mg/liter
2,5
10
10
16
20
Form
of
zinc
fed
ZnS04
ZnS04
Complexed
zinc
Complexed
zinc
ZnS04
Removed by
primary
treatment, %
13
14
-
8
-
Removed by
complete
activated -
sludge
treatment, %
95
89
96
_
74
as was the zinc sulfate form. A distinct difference in the effect of
the two forms was demonstrated in the acclimation period. In the
experiments with continuous feeding of Complexed zinc at a concen-
tration of 10 milligrams per liter and zinc sulfate at 20 milligrams per
liter, data on effluent quality were obtained during the first few days
following introduction of zinc to the unit feeds. The sludges were
developed on feeds to which no zinc had been added. The acclimation
phenomena of these sludges is demonstrated by the turbidity data shown
in Figure 29. The first sample of the feed containing Complexed
1 1—i 1—i—r
Zn CN
COMPLEX, 10 mg/liter
I 1 I I I
~i 1 1 r
Zn S04, 20 mg/
J 1 I I I I I I I
3456789 10 0 I 334567 89 10
TIME, days TIME, days
Figure 29. Comparison of acclimation to Complexed zinc and zinc sulphate.
zinc, collected about 30 hours after initiation of the metal feed, had
a turbidity of almost 80 standard units. The turbidity declined in
subsequent samples and, after about 5 days, reached a level that
prevailed during the remainder of the run. This improvement in
turbidity paralleled the decrease in cyanide in the final effluent.
In the first samples cyanide in the final effluent was practically
70
INTERACTION OF HEAVY METALS
GPO 82O—663—6
-------
equal to that fed. In subsequent samples the cyanide level was pro-
gressively lower until samples collected on the fifth and succeeding
days had at the most only traces of cyanide. Acclimation to the
alkaline cyanide bath is apparently a phenomena of adaption of the
system to degradation of the cyanide. Feeding of zinc sulfate, on the
other hand, resulted in a turbidity in the first sample of final effluent
collected of about the same level as that prevailing during the run.
Thus, acclimation of this system to zinc occurred in a few hours.
The state of the zinc, whether in solution or in insoluble form,
was of interest. Zinc not in solution would not be expected to exert
a toxic action. Data on zinc in solution in the primary and final
effluents are presented in Table 29. The values given are average;
the range about these averages was great.
Table 29. ZINC CONCENTRATION AND FORM IN PROCESS
Zinc
in
sewage,
mg/liter
2.5
10
10
20
Form
of
zinc
fed
ZnSOi
ZnSO-i
Complexed
zinc
ZnS04
Primary effluent, mg/liter
In
solution
0,05
0,64
0.94
10.4
Total
2.05
8.9
9.8
19.8
Final effluent, mg/liter
In
solution
0.02
0.18
0.09
4.29
Total
0.12
0.88
0.39
5.16
Slug Doses*
The reaction of the activated-sludge process to a 160-milligram-
per-liter slug dose of zinc was studied. The system was acclimated
to 5 milligrams per liter of zinc for 1 month prior to the slug. The
slug consisted of zinc in the form of zinc sulfate, and lasted for 4
hours in the influent sewage. After the slug, the 5-milligram-per-
liter zinc dose was continued. Eight hours after the slug, the solids
in the final settler showed slight bulking, but after 24 hours the
settling characteristics of the sludge were satisfactory. Microscopic
examination of the mixed liquor before, during, and several days after
the slug showed that the higher forms were not affected by this con-
centration of zinc.
The response of the system is shown on Figure 30. There was
a serious upset of the plant for about 30 hours after the start of the
slug but the plant was producing effluent of pre-slug quality 40 hours
after the start of the slug.
*The material in this section was completed after original publication
of Reference 16.
Zinc
71
-------
10 15 20 25 30 35 40 45 50 55 60 65
TIME AFTER 160-mg/liter SLUG OF ZINC SULFATE, hr
Figure 30. COD, turbidity, and suspended solids of final effluent.
The primary and waste activated sludges removed from the
plant in a 3-day period after the slug accounted for 67 percent of
the zinc added; approximately 33 percent of the zinc from the slug
went out with the final effluent. A material balance for zinc in the
slug accounted for 104 percent of the metal. The zinc discharged
in the final effluent was predominantly in an insoluble form and did
not exceed 9 milligrams per liter at any time, as shown on Figure 31.
72
INTERACTION OF HEAVY METALS
-------
10 15 20 25 30
TIME, hr
75
Figure 31. Zinc in sewage and effluents, slug of
160 mg/liter for 4 hours.
This figure also shows an unexpected behavior of the zinc during the
slug study; the soluble zinc content of the influent sewage was quite
low in contrast to the soluble zinc content of the primary effluent.
This can probably be explained by dilution of the slug in the primary
settler causing re-solution of insoluble zinc in the influent sewage,
and complexing reactions occurring during the detention period.
ZINC AND SLUDGE DIGESTION
The average gas production from digestion of zinc-bearing sludges
for 7-day periods, the interval of daily feed compositing, is shown in
Figures 32 through 35. The digesters were seeded originally with
sludge from a municipal sewage treatment plant. Sludges from the
activated-sludge plant that received no addition of zinc to its feed
were fed for a week or more before the feeding of zinc-bearing sludge
was started. Production of gas during this normalization period is
shown on the graphs. Gas produced by sludges from the control unit
ranged from approximately 600 to 900 milliliters per gram of volatile
solids fed.
Zinc
73
-------
KXDO
800
i 600
400
200
START OF FEEDING
Zn-BEARING SLUDGES
20 3O 40
TIME, days
Figure 32. Gas production of combined sludges from sewage fed
10 mg/liter zinc as zinc sulfate.
a
o
I I 1 1 1 1
L- START OF FEEDING
Zn-BEARING
SLUDGE
-30 -20 -10
10 ZO 30
TIME, days
10 20 30 40
TIME, days
Figure 33. Gas production of combined sludges from sewage fed
20 mg/liter zinc as zinc sulfate.
74
INTERACTION OF HEAVY METALS
-------
N
5'
GAS PRODUCTION, ml/g OF VS FED
1 2
•»* >
T
3 -D
2. 3
o S.
GAS PRODUCTION, ml/g OF VS FED
I O
-q
Ul
-------
Where a serious reduction in gas production occurred, the
experiment was repeated and a confirming observation made. The
experiments with effects of zinc on digesters were essentially limited
to a zinc sulfate feed because high cyanide levels are prohibited, as
discussed previously. One exception was a run made with digestion of
primary sludge from sewage fed complexed zinc in a concentration
of 16 milligrams per liter. This concentration of zinc in the form of
the alkaline cyanide bath corresponds to a concentration of CN~of
18.3 milligrams per liter. This experiment showed that sludge from
sewage with a zinc level of 16 milligrams per liter together with a
CN level of 18.3 milligrams per liter would not digest at normal
rates when introduced to an unacclimated normally functioning digester.
In the second experiment at this concentration the feeding of zinc-
bearing sludge was inadvertently initiated before the digester was
producing normal volumes of gas. The gas production rate was
affected much more rapidly in this case.
Data on the concentration of zinc in the primary sludges, excess
activated sludges, and digested sludges for three levels of zinc in the
sewage are given in Table 30. Zinc concentrations in sludges are
Table 30. TOTAL ZINC IN SLUDGES
Zinc
sewage,
mg/liter
2.5
10
16
Form
of
zinc
fed
ZnSOi
ZnS04
Complexed
zinc
Primary sludge,
mg/
liter
64
375
548
%of
total
residue
0.22
0.95
2.0
%of
volatile
residue
0.27
1.6
3.0
Excess activated
sludge,
mg/
liter
119
328
_
% of
total
residue
2.5
6.0
_
%of
volatile
residue
3.7
12
_
Digested primary
sludge.
mg/
liter
_
545
%of
total
residue
_
a
% of
volatile
residue
_
a
Digested combined sludges,
mg/
liter
341
_
%of
total
residue
3.16
_
% of
residue
8.0
_
Digestion subnormal, values change as undigested material accumulates.
proportional to the concentration of suspended matter in the sludge
since the zinc is predominantly a part of the suspended matter. In
order to compare zinc data among sludges with varying solids con-
centrations, the zinc is expressed as a percent of the total and volatile
residues in the sludges.
Sludge in the digesters was completely mixed at times of its
removal; therefore, the accumulation of zinc in the digester would
follow the principle of displacement of one material, A, from a homo-
geneously mixed system by continuous addition of a second material,
B. Theoretically, after a number of feedings equal to the digester volume
divided by the daily feed volume (one detention period), the sludge of
the new origin (zinc-bearing) would constitute just over 60 percent
of the sludge in the digester. After four periods only a negligible
percent of the original sludge would be left. In order to reach the
maximum zinc concentration in the digesters, they were operated
76
INTERACTION OF HEAVY METALS
-------
for over 60 days before terminating an experiment, if a digester
continued with normal gas production. The concentration of zinc
reported for digested combined sludges at the zinc level of 10 milli-
grams per liter is an average value after the zinc concentration in
the digester had leveled off.
Data on the relation of zinc in sludges to digestion or treatment
difficulties may also be a useful by-product of the work. A sludge
sample, particularly digested sludge, represents a composite ac-
cumulated for long periods of time. Thus measurement of zinc in
the sludge may provide a means of estimating the average concentration
of zinc in sewage received over an extensive preceding period. Sub-
sequently, a limited judgment as to whether or not zinc is responsible
for subnormal treatment can be made from a few sludge analyses.
It seems logical that toxicity of the liquid surrounding micro-
organisms would result from zinc in solution. For this reason, in two
of the runs, measurements were periodically made of zinc in solution
in the sludge feeds and the digester. The data obtained are given
in Table 31. The quantity of zinc in solution did not appear to follow
any pattern of increase corresponding to decreases in gas production.
Table 31. ZINC IN SOLUTION IN SLUDGES
Zinc
in
sewage,
mg/liter
10
16
Form
of
zinc
fed
ZnS04
Complexed
zinc
Zinc in solution
Sludge
Primary
sludge
Excess activated
sludge
Digested combined
primary and excess
activated
Primary sludge
Digested primary
sludge
Avg, mg/liter
0.31
0.06
0.17
1.33
0.34
Range, mg/liter
0-1.18
0-0.21
0-0.67
0-4.39
0-1.44
SUMMARY
Zinc fed continuously in concentrations ranging from 2.5 to 20
milligrams per liter of sewage entering a complete pilot activated-
sludge treatment plant reduced the BOD removal efficiency a maximum
of about 2 percent. Two forms of zinc, zinc sulfate and complexed
zinc such as that which occurs in an alkaline cyanide plating bath,
had about the same effects after the sludge became acclimated. The
maximum level of zinc that will not produce a significant effect on
Zinc
77
-------
treatment efficiency was indicated as being>2.5 and< 10 milligrams
per liter.
Primary treatment is not efficient in removing zinc; however,
the microbial floe of secondary treatment adsorbs much zinc. The
overall process is from 95 to 74 percent efficient in removing zinc
at the feed levels of 2.5 and 20 milligrams per liter, respectively.
A 160-milligram-per-liter slug dose of zinc, lasting for 4
hours, caused a serious reduction in treatment efficiency for about
1 day. Forty hours after the slug the plant recovered and produced
suitable effluent.
Sludges from sewage fed zinc, as zinc sulfate in a concentration
of 10 milligrams per liter digested at normal rates. The combined
sludges from sewage fed zinc, as ZnSCU in a 20-milligram-per-liter
concentration, caused rapid failure of the digestion process. For
normal digestion of the primary or combined sludges, the maximum
level of zinc in sewage is between 10 and 20 milligrams per liter.
78 INTERACTION OF HEAVY METALS
-------
CHAPTER IV. NICKEL*
The efficiency of activated-sludge plants in the treatment of
sewage containing nickel was studied. Sewage from a common source
was fed to three replicate pilot plants. Nickel solutions were added
to the sewage entering certain pilot plants to produce selected constant
concentrations. No metal was added to the sewage entering one of
the units. Differences in the quality of the effluents, as measured
by BOD, COD, suspended solids, and turbidity, from the nickel-fed
units and the unfed unit were attributed to nickel in the feed.
The anaerobic digestion of the nickel-bearing sludges was studied
by operation of bench-scale digesters on sludge feeds obtained from
the activated-sludge plants. Any differences in gas production in
the digesters receiving control sludge and those receiving the nickel-
bearing sludges were also attributed to nickel in the sludge.
The objectives of the research were (1) to determine how much
nickel in waste waters can be tolerated without reducing the efficiency
of biological processes in removing the organic matter or in sta-
bilizing the sludges and (2) to determine the efficiency of the process
in removing nickel.
PLANT DESCRIPTION AND OPERATION
The activated-sludge pilot plants were designed to simulate
standard activated-sludge plants of the spiral flew types. The shape
and dimensions of the activated-sludge units are illustrated in Figure 2.
The nickel solutions were fed to the sewage before it entered the
primary settler. Thus precipitation, reduction, and complexing that
might occur during primary settling before a biological process is
reached were included in the experimental conditions. Effects are
related to metal additions to the incoming sewage rather than to
metals added to some specific plant component. The sewage fed to
the plants was either a weak sewage obtained from the Eastern Ave-
nue interceptor or a more normal sewage from the Beechmont inter-
ceptor of the city of Cincinnati. The latter sewage was used only
for a short time at one nickel concentration.
* Material in this chapter published previously in Journal Water
Pollution Control Federation. Washington 25, D.C. 20016. See Refer-
ence 17.
79
-------
The sewage from the Eastern Avenue source was of low strength
because of dilution by ground water infiltration. The weak sewage
was supplemented with either fish meal or dog food to bring its
organic content to a level found in strong domestic sewage since
an adverse effect of metal on effluent strength probably would be
exhibited at the high-level end of the range of domestic sewage
strength. Dog food of the granular, dry type was ground to a fine
powder and soaked over night in water; the resultant slurry was
blended for 5 minutes and then mixed with sewage at a concentration
of 1.2 grams (air dry) per gallon. Chemically, the dog food was
considered to approach the organic matter composition of domestic
sewage except for its low nitrogen content. To raise the nitrogen
content of the strengthened sewage to near that found in strong
domestic sewage, urine was added at the rate of 1.4 milliliters per
liter. The nitrogen content of the fish food was relatively high, and
no supplemental nitrogen was needed.
The sludge digesters were 5-gallon glass carboys, which were
connected to pumps for mixing the digester contents. Single-stage
digestion without continuous agitation was employed. The digesters
were maintained at 30° C in a constant-temperature room. Gas
was collected in a floating-cover gas holder; its volume was measured
daily at atmospheric pressure and 30°C.
Sewage was fed to the units at a constant rate. Sludge was
returned from the final settler continuously at a rate of about 35
percent of the sewage feed flow. An automatic device was activated
once per minute to divert about 5 percent of the return sludge to
a waste-excess activated-sludge-collecting carboy. Capacity and
loading factors for the units of the plant are given in Table 32.
Table 32. PILOT-PLANT DESIGN DATA AND LOADING FACTORS
Unit
Primary
settler
Aeration
tank
Final
settler
Loading factor
Capacity
Detention time
Surface overflow rate
Capacity
BOD loading
Aeration period
Capacity
Detention time
Surface overflow rate
4.6 gal
1.2 hr
142 gpd/ft2
23.6 gal
41-63 lb/day/1,000
tank volume.
0.50-0.75 Ib/day/lb
aeration
6 hr
7.9 gal
2 hr
102 gpd/ft*
ft aeration
VS under
80
INTERACTION OF HEAVY METALS
-------
Digesters were operated on primary sludge feeds alone and
on primary combined with excess activated sludge. Primary sludge
was withdrawn once per day from the primary settler. Excess acti-
vated sludge for digester feed was collected once per day from the
final settler. The sludge accumulating in the waste-excess activated-
sludge carboy was not used for digester feed because of the possi-
bility that septic conditions during the collection period might cause
nontypical metal reactions. To minimize changes of metal in solution,
the sludges were allowed to consolidate only 30 minutes or less after
collection. For this reason solids concentrations in the sludges were
lower than those usually found in sewage treatment practice.
The digesters were thoroughly agitated once daily by means
of a pump, which withdrew sludge from the bottom of the digester
and returned it at the top. Digesters were fed once each day. The
digesters receiving a combination of primary and excess activated
sludge were fed 300 milliliters of primary sludge and 700 milliliters
of excess activated sludge. This daily feed contained about 10 grams
of volatile matter, approximately 60 percent of which was from
the primary sludge. This is the approximate ratio of production
of primary and excess activated sludge in the pilot plant. The
digesters receiving only primary sludge were fed 470 milliliters
of primary sludge, which contained about 10 grams of volatile matter.
Digester capacity and loading parameters are given in Table 24.
NICKEL SOURCE AND FORM IN LIQUID WASTES
A common source of liquid wastes containing nickel is the
metal plating industry. Usually, nickel used in plating baths is prin-
cipally nickel sulphate with smaller quantities of nickel chloride
and boric acid. A solution of nickel (II) sulfate was added to the
sewage in this investigation.
SAMPLE COLLECTION
Samples for measurement of effects of nickel on treatment
efficiency were collected by mechanical samplers. The sampler
was activated by a timer at 15-minute intervals and diverted the
stream to be sampled to a refrigerated collecting vessel for a period
of about 12 seconds each time. The samples were composited over
24-hour periods. The analytical procedures were started within
a few hours after the compositing period. Samples for studies of
the effect of slug doses were collected in the same manner except
that compositing periods were limited to as little as 4 hours to show
peaks in effects. Grab samples were taken for some nickel con-
centration measurements. Samples of sludges and the final effluents
were collected for making material balances between nickel fed
the units and nickel in the effluents plus nickel accumulated in the
aerator. Samples of each withdrawal of primary and excess activated
Nickel 81
-------
sludge were collected and composited over the balance period of
1 week. Samples of the final effluent were collected by automatic
samples at 15-minute intervals and composited for the week. Grab
samples of the aerator contents were collected at the beginning and
end of each balance period for nickel accumulation measurements.
ANALYTICAL METHODS
The procedures used were those outlined in Standard Methods,
llth edition (10). Details of alternatives selected and procedures
utilized are described in the following discussion.
Biochemical Oxygen Demand
In the BOD test, the initial and final dissolved oxygen measure-
ments were made by the Alsterberg azide modification of the Winkler
method. Sample dilutions were prepared by the cylinder dilution
techniques. All BOD data are for incubation at 20°C for 5 days.
Chemical Oxygen Demand
In the determination of COD, primary feed and primary effluent
samples were oxidized by use of 0.25 N dichromate. For final effluent
samples, 0.025 n dichromate was used. Silver sulfate catalyst was
not used. No correction for chloride was made. Chloride concentrations
in the sewage were normally about 40 milligrams per liter.
Nickel
Nickel analyses were made by two methods. In one method,
samples relatively high in nickel content, such as sludges, were
assayed by selectively precipitating nickel from an ammoniacal slurry
of an acid-digested sample with dimethylglyoxime. The nickel
dimethylglyoxime precipitate was then eluted from the filter paper
with concentrated HC1; the filtrate was made ammoniacal; and nickel
reprecipitated with dimethylglyoxime. The nickel complex was then
redissolved in concentrated HC1, and the excess HC1 removed by
evaporation on a hot plate. The residue was made ammoniacal and,
titrated with a standardized solution of Versenate in the presence
of the purple dye Murexide.
In the other method, low concentrations of nickel, found in feed
and final effluent samples, were assayed by the alpha-furildioxime
colorimetric procedure (15). The only interfering ions usually en-
countered in sewage are Cu-++ andFe++. The assay consisted essential-
ly of adding dilute K2Cr2O7(to oxidize Fe++) and sodium citrate
(to complex Fe+++) to an aliquot of the sample diluted to 100 milliliters,
(adding 1 gram sodium thiosulfate, to complex Cu++), adjusting to pH
8-9, and adding alpha-furildioxime dissolved in ethyl alcohol. The
82 INTERACTION OF HEAVY METALS
-------
colored complex is extracted with 3x7 milliliters of CHC13 and diluted
to 25 milliliters. The optical density at 435 millimicrons is proportional
to the nickel concentration.
For both methods outlined above, recovery tests, in which
standard nickel additions were made to samples, were satisfactory.
Both methods are sensitive and specific for nickel.
Many samples were analyzed for both total nickel and nickel
in solution. Sample aliquots for determining nickel in solution were
filtered with an HA45 Millipore membrane. The filtrate was digested
in a nitric and sulfuric acid mixture, and nickel was determined by
the alpha-furildioxime colorimetric procedure. Many analyses showed
that a portion of the nickel in the filtrate would not react with the reagent
without prior acid digestion. All results reported for nickel in solution
are for samples subjected to acid digestion.
NICKEL AND ACTIVATED-SLUDGE TREATMENT
Industrial wastes containing heavy metals may be discharged
more or less continuously in, for example, drainage and rinse water
wastes from metal plating operations, or wastes with high metal
concentrations may be discharged over short periods from, for
example, a plating bath dump or spill. Observations under conditions
simulating both occurrences were made, that is, with nickel con-
tinuously present at constant concentration in the influent sewage
and with nickel introduced as a slug dose.
Continuous Nickel Addition
Runs were made with 1-, 2.5-, 5-, and 10-milligram-per-liter
concentrations of nickel continuously present in the sewage feed.
A control plant received no metal and was operated in parallel during
each run. The 5- and 10-milligram-per-liter runs were made con-
currently with a common control unit. The average characteristics
of the sewage feed during each of the runs are given in Table 33.
The analytical data on sewage feed to the control and to the experimental
units during each run generally were in agreement within a range
attributable to sampling and analytical variations.
Before samples for efficiency studies were collected, the metal-
fed units were allowed to acclimate for a 2-week period after initiation
of the metal feed. Thirteen to twenty-six 24-hour composite samples of
the experimental and control units were collected and analyzed during
each run.
Nickel 83
-------
Table 33. AVERAGE CHARACTERISTICS OF SEWAGE FEEDS AND
PRIMARY EFFLUENTS FOR CONTROL AND
NICKEL-FED UNITS
Nickel(asNiS04)
addition, mg/liter
0
10
0
5
0
2.5
0
1.0
BOD,
mg/liter
Primary
feed
217
207
217
255
247
260
172
186
Primary
effluent
145
148
145
190
178
192
123
117
COD,
mg/liter
Primary
feed
326
342
326
393
396
409
272
287
Primary
effluent
236
238
236
267
269
301
228
235
Suspended
matter,
mg/liter
Primary
feed
257
304
257
314
337
303
178
177
Prima ry
effluent
155
187
155
175
143
184
121
125
The quality of the final effluents from the nickel-fed units and
the control units are presented in cumulative frequency distribution
curves on logarithmic probability paper in Figures 36, 37, and 38.
Such curves enable presentation of the complete data and rapid com-
parisons. Table 34 gives a brief summation of the results in arith-
metic averages of final effluent determinations.
Table 34- AVERAGE CHARACTERISTICS OF FINAL EFFLUENTS
FROM CONTROL AND NICKEL-FED UNITS
Nickel (asNiSO-0
addition, mg/liter
0
10
0
5
0
2.5
0
1
BOD,
mg/liter
9
14
9
13
13
26
21
23
COD,
mg/liter
40
54
40
51
59
63
48
51
Suspended
matter,
mg/liter
8
17
8
16
5
9
11
8
Turbidity,
stu
4
28
4
15
10
29
25
34
The BOD's of the final effluents for the various runs are given
in Figure 36. The data show that nickel concentrations of 2.5, 5, and
10 milligrams per liter significantly affected treatment efficiency.
At 1 milligram per liter, however, there was no significant difference
84
INTERACTION OF HEAVY METALS
-------
Z 5 10 20 30 40 SO 60 70 80 90 95 98
% OF OBSERVATIONS^ STATED VALUE
Figure 36. Effect of nickel on BOD of
final effluents.
Nickel
85
-------
in the efficiency of the nickel-fed unit and the control unit. Data
at this low nickel concentration are for sewage supplemented with
fish food and for strong domestic sewage. Nickel at this level did
not have a significant effect on BOD removal with either sewage.
The BOD data for the 2.5 -milligram-per-liter run is interesting
because a greater effect was shown than would be expected from
the 5- and 10-milligram-per-liter runs.
100
90
80
70
I 60
? 50
30
20
CONTROL
J I L
J L
86
5 10 20 3O 40 50 60 70 80 90 95 98
% OF OBSERVATIONS £ STATED VALUE
Figure 37. Effect of nickel on COD of
final effluents.
INTERACTION OF HEAVY METALS
GPO 820-663-7
-------
CONTROL
I .. I.
1 mg/lite'
CONTROL
25 10 20 30 10 5O 60 7O 8O 90 95 98
% OF OBSERVATIONS 5 STATED VALUE
Figure 38. Effect of nickel on turbidity
of final effluents.
Nickel
87
-------
COD was routinely determined because of the possibility that
nickel would inhibit the BOD analysis. In no case was there any
indication that such inhibition occurred. Figure 37 shows that 5-
and 10-milligram-per-liter concentrations of nickel significantly affect
treatment efficiency, based on COD analysis. The COD curves for
concentrations of 2.5 and 1 milligram per liter show no significant
effect, whereas the BOD curve for the 2.5-milligram-per-liter con-
centration shows a greater effect than the curve for the 5- or 10-
milligram-per-liter concentrations.
The turbidity plots shown in Figure 38 reflect the nickel con-
centrations in magnitude of effect; each nickel dosage caused impair-
ment of effluent clarity in the order of nickel concentration. Each
concentration had a significant effect on .turbidity. From the data
in Table 34 and Figures 36, 37, and 38, it is concluded that nickel
concentrations of 10, 5, and 2.5 milligrams per liter definitely affect
the treatment efficiency of an activated-sludge process. The effect
of 1 milligram per liter is subtle and is considered near the threshold
limit for nickel.
An explanation is not apparent for the anomalous results of
the BOD and COD data for the 2.5-milligram-per-liter run, which
remain a variant in an otherwise orderly series of observations.
Table 35. NICKEL DISTRIBUTION IN PROCESS OUTLETS
Nickel in
sewage,
mg/liter
10
2.5
1
10
2.5
1
Primary
sludge
Excess
activated
sludge
Final
effluent
Imbalance
Nickel, mg/liter
62
-
15
89
-
26
72
1.4
0.8
-
_
-
Nickel fed in outlet, %
2.5
_
5.4
14.8
_
7.2
72.1
(52-90)a
58
(62-80)a
72.5
(56-87)a
-11
_
-15
Range of observations.
During the 10- and 1-milligram-per-liter runs the apportionment
of the nickel fed among the various sludges and final effluent was
88
INTERACTION OF HEAVY METALS
-------
traced by material balances. During the 5- and 2.5-milligram-per-
liter runs the sampling and analytical program for metal balances was
eliminated. The nickel distribution during activated-sludge treatment
is given in Table 35. It can be seen that only a small amount of
nickel precipitates with the primary sludge. The activated sludge
showed no great affinity for nickel; consequently, the major portion
of the nickel passed out with the final effluent. The imbalance figure
on Table 35 refers to the average degree of success in the balance
between the nickel fed and nickel accounted for.
The efficiency of primary and complete activated-sludge treat-
ment in removing nickel from sewage is given in Table 36. Primary
Table 36. PERCENT EFFICIENCY OF TREATMENT PROCESSES IN REMOVING NICKEL
Nickel in
sewage,
mg / liter
10
2.5
1
Primary
treatment,
%
3
-
5
Complete activated-
sludge treatment,
%
28
42
28
treatment in the range studied removes approximately 5 percent of
the influent nickel; activated-sludge treatment removes approximately
30 percent of the influent nickel.
Table 37. NICKEL CONCENTRATIONS IN EFFLUENTS
Nickel (asNiSCU)
in sewage,
mg / liter
10
2.5
1
Primary effluent, mg/liter
In
solution
-
0.9
0.78
Total
8.2
2.0
0.97
Final effluent, mg/liter
In
solution
-
1.1
0.70
Total
7.2
1.4
0.75
Table 37 presents the results of analyses for nickel in solution
during the 2.5- and 1-milligram-per-liter runs. Grab samples were
collected, filtered immediately with a membrane filter, and com-
posited over a 5-day period. The primary effluents have slightly
greater total nickel content than nickel in solution; however, the
nickel in the final effluent is almost entirely in solution. The decision
to classify material passing through the 0.45-micron membrane as
soluble was strictly arbitrary. The differentiation of total and soluble
metal was made because soluble metal, rather than total metal,
Nickel
89
-------
content would be expected to be more indicative of physiological
response of recipient organisms, both in the treatment plant and the
receiving stream.
Table 38. EFFECT OF SULFIDE ON ACTIVATED-SLUDGE TREATMENT
EFFICIENCY OF NICKEL REMOVAL
Period
1
2
3
Average sulfide
content of sewage fed,
mg/liter
6
0,8
11
Final effluent ratio:
total nickel
soluble nickel
1
1.1
1.1
1.2
1.9
1.5
1.9
1.1
1.2
1.3
1
Efficiency of
overall nickel
removal, %
52
28
44
Table 38 is a summation of results gathered in an attempt
to correlate nickel removals with sulfide content of the sewage fed
during the 2.5-milligram-per-liter run. The various sulfide levels
were obtained by manipulation of standard pilot-plant operation.
Period 1 was the sulfide level normally occurring with routine op-
eration. Period 2 sulfide levels were obtained by mild aeration
of the sewage in the holding tank. Period 3 levels were obtained
by reserving part of each day's sewage in the holding tank for the
next day. From the limited data obtained, no correlation of sulfide
content and nickel removal could be deduced, considering the wide
variation in efficiency of nickel removal encountered in each of the
runs (Table 35).
Slug Dose
The reactions of the activated-sludge process to 4-hour nickel
slug doses of 25, 50, and 200 milligrams per liter of sewage were
studied. In each case the activated sludge was acclimated to a con-
tinuous 2.5-milligram-per-liter nickel dose before the slug-dose
test. The logic in using the acclimated system was that routine
slight losses are likely to prevail wherever a slug dose occurs.
The slug doses of 25 and 50 milligrams per liter did not impose
a very great stress on the system; therefore, only the data obtained
during the 200- milligram -per-liter run are presented. Figure 39
90
INTERACTION OF HEAVY METALS
-------
100
50
30O
200 -
1— 1~
TOTAL
SUSPENDED
MATTER
^-.
. ^VlJp
I
1
f
\
N
— i i i
\, TURBIDITY
^"~-
}— i 1 1 r*TB"n
3
loot
CO
tz
H
0
D
O
CD
100 -
-20
Figure 39. COD, BOD, suspended solids, and turbidity of final effluents,
unit fed 200 mcj liter nickel for 4 hours.
depicts the reaction of the activated-sludge process to a 4-hour slug
dose of 200 milligrams per liter of influent sewage. The final effluent
showed a marked increase in BOD, COD, suspended matter, and
turbidity 10 hours after initiation of the nickel slug. These effects
diminished in a rather linear manner for the next 30 hours, and the
system was producing effluent of preslug quality 40 hours after the
slug. This sludge had been acclimated to the continuous 2.5-milli-
gram-per-liter addition, and this addition was continued after the
slug dose.
Figure 40 shows the distribution of nickel after the 200-milli-
gram-per-liter slug dose. The nickel concentration reached its peak
in the primary effluent in 4 hours and then rapidly decreased; by the
end of 8 hours the nickel content was equal to that during the 2.5-
milligram-per-liter continuous feed. At the peak of nickel concentration
in the primary effluent, 60 percent of the nickel was in solution.
The nickel content of the final effluent reached its peak 10 hours
after the slug. Eight hours after slugging, 60 percent of the nickel
in the final effluent was in solution. After 20 hours, the soluble
and total nickel contents were equal. The nickel content of the final
effluent gradually dropped to its normal level after 60 hours.
Nickel
91
-------
50
0
100
50
Ni IN
SOLUTION
20
60
30
TIME, hr
Figure 40. Nickel in primary and final effluents, total and
solution; unit fed 200 mg/liter nickel for 4 hours.
NICKEL AND ANAEROBIC DIGESTION
The sludges produced during the 10-milligram-per-liter nickel
run were digested anaerobically. A digester fed primary sludge and
a digester fed primary and excess activated sludge were studied.
These digesters were operated on the nickel-bearing sludge for
at least 60 days, almost four detention periods, without any signs
of interference with gas production or volatile solids destruction.
The gas production averages for weekly periods for the combined
primary and excess activated sludge were within the range of gas
production of sludges from the control unit.
Since sludges from the 10-milligram-per-liter run did not
interfere with anaerobic digestion, primary sludges produced by
primary settling of sewages containing nickel doses of 20 and 40
milligrams per liter were studied. The digester receiving sludge
from the 20-milligram-per-liter run digested normally for 60 days.
At the end of this time the digester was fed sludge produced from
sewage containing a 40-milligram-per-liter nickel dose. Digestion
proceeded normally for an additional 60-day period. For each anaerobic
digestion study, with the exception of the 20-milligram-per-liter run,
complete material balances for nickel and solids were obtained. Each
sludge added to the digesters was assayed for both total nickel and
soluble nickel. Table 39 summarizes the results of the analyses.
An interesting feature of these data is that although the sludges fed
to the digesters contained considerable nickel in solution, the digested
sludges had a very low soluble nickel content. The concentration
92
INTERACTION OF HEAVY METALS
-------
Table 39. NICKEL IN SLUDGES
Nickel
in
sewage,
mg/liter
10
20
40
Primary sludge,
mg/liter
Total
62
_
308
In solution
Avg
9.8
12.8
13.2
Range
8.3-13.75
6.2-17.4
6.8-21.4
Excess activated sludge,
mg/liter
Total
8.9
-
-
In solution
Avg
8.9
-
-
Range
7.8-11.5
-
-
Digested primary sludge,
mg/liter
44
-
277
In solution
Avg
1.6
1.90
1.47
Range
0.7-3.6
-
0.7-5.0
Digested combined sludge,
mg/liter
Total
70
-
-
In solution
Avg
1.6
-
-
Range
0.4-6.5
-
-
of nickel reported for digested sludges is an average value after the
nickel concentration in the digester had become rather constant.
Table 40 shows the amount of nickel
basis. The purpose of this calculation is
sludges of various solids contents. The
the digested sludges can be used to make
nickel content of the influent sewage to
fleets the average concentration of nickel
sludge accumulation period.
in the sludges, on a dry
to allow comparisons of
nickel concentrations in
an approximation of the
a plant. The sludge re-
in the sewage over the
Table 40. NICKEL CONCENTRATIONS IN DRIED SLUDGES
Nickel
concentration,
mg/liter
40
10
1
Nickel in total suspended solids, rag/g
Primary sludge
8.3
2,2
1,1
Excess
activated
sludges
-
10.6
4.6
Digested
primary
sludges
15
2.8
-
Digested
combined
sludges
-
7.1
-
During the run with the 200-milligram-per-liter nickel slug
dose, primary sludge and a sample of excess activated sludge were
collected at peak nickel concentrations and fed to a normal digester.
No effect on gas production resulted.
During each anaerobic digestion run a control digester was
also operated in parallel with the experimental digester. This digester
was operated under identical conditions, but was fed sludges bearing
no nickel.
DISCUSSION
This study has demonstrated that the effects of nickel on the
activated-sludge process are not linear with concentration, but dis-
play decreasing response to increasing concentration. The increased
BOD and COD of the final effluents from nickel-fed units, above
Nickel
93
-------
those of the control units, were about the same for doses of 10 and
5 milligrams per liter. This behavior was noted with chromium
(8), copper (16), and zinc (16), all of which have been studied at
this laboratory. Biological systems frequently show this behavior,
for instance, vitamin and antibiotic assays show linear relation-
ships between, dose and effect over only a very narrow range of
concentrations.
In the slug dose study, doses of nickel under 200 milligrams
per liter did not seriously upset the system. This dose is at least
100 times the amount needed to affect the continuous-dose runs
significantly. Even the 200-milligram-per-liter slug dose caused
only a temporary decrease in effluent quality, and by the end of
40 hours the system was producing preslug quality effluent.
Table 41. AVERAGE VALUES FOR OVERALL BOD REMOVAL
Continuous
nickel
concentration,
mg/liter
0
10
0
5
0
2.5
0
1
BOD
remaining in
final effluent,
%
4
7
4
5
5
10
12
12
Overall
removal,
%
96
93
96
95
95
90
88
88
Reduction in
overall removal,
%
3
_
1
5
0
Table 41 tabulates the overall plant efficiencies based on BOD,
for the various continuous nickel-feeding runs. Nickel doses of
10, 5, and 2.5 milligrams per liter had only a slight effect on over-
all plant performance. No effect on BOD removal was shown by
a continuous nickel dose of 1 milligram per liter.
These observations support the conclusion that the aerobic
phase of activated-sludge treatment can tolerate, without reduced
efficiency, the continuous presence of nickel at concentrations no
greater than 1 milligram per liter, but can satisfactorily recover
from slugs of at least 200 milligrams per liter.
Primary treatment removed only a small amount of nickel.
The majority of the nickel reaching the aeration chamber is passed
through to the effluent in soluble form. Complete activated-sludge
treatment is approximately 30 percent efficient in reducing the nickel
content of influent sewage. Considerations of the effect the final
94
INTERACTION OF HEAVY METALS
-------
effluent will have on the receiving stream should include the ob-
servation that 70 percent of the influent nickel reaches the final
effluent.
The possibility of increasing nickel removal efficiency by in-
creasing the sulfide content of the sewage was investigated. The
formation of insoluble nickel sulfide was expected to cause the nickel
to be sorbed on the biological floe and be removed from the effluent,
but the assumption proved unwarranted.
The anaerobic digestion process proved to be very resistant
to any effects caused by nickel in the sludges. Digestion of the mixed
sludges produced during the 10-milligram-per-liter run proceeded
normally. No difficulty with anaerobic digestion of primary sludge
produced during the 40-milligram-per-liter run was encountered,
nor did a slug of nickel-bearing sludge have a noticeable effect. This
nickel-bearing sludge was obtained during a slug dose of 200 milligrams
per liter to the sewage feed. An interesting feature of the anaerobic
digestion of nickel-bearing sludges is the fact that soluble nickel
introduced with the feed sludges is converted to an insoluble form
during digestion. The long detention time, high alkalinity, sulfide
content, and hydroxyl ion concentration offer a favorable environ-
ment for the formation of insoluble nickel compounds.
SUMMARY
Nickel, present continuously, in concentrations ranging from
2.5 to 10 milligrams per liter in the sewage entering a complete
activated-sludge pilot plant reduced the BOD removal efficiency a
maximum of about 5 percent. Increased turbidity in the final ef-
fluent is the most objectionable feature. The maximum level of
nickel that will not produce a detectable effect on treatment efficiency
was indicated as being greater than 1 and less than 2.5 milligrams
per liter.
A 200-milligram-per-liter slug dose of nickel caused a serious
reduction in treatment efficiency for a few hours, but the plant returned
to normal performance within 40 hours.
Combined primary and excess activated sludge from a plant
receiving 10 milligrams of nickel per liter continuously digested
satisfactorily. Primary sludge from sewage containing 40 milligrams
of nickel per liter digested satisfactorily.
A small percentage of nickel is removed in primary settling.
The complete activated-sludge process is about 30 percent efficient
in removing nickel. The sulfide content of the influent sewage has
no correlation with efficiency of nickel removal.
Nickel 95
-------
CHAPTER V. A MIXTURE OF HEAVY METALS*
The effects of copper, chromium, nickel, and zinc introduced
individually to the sewage feed of complete activated-sludge pilot
plants have been discussed in the previous chapters. The results
obtained from the individual studies were used as background in-
formation to investigate the effects of a mixture of these four metals
on the activated-sludge and anaerobic digestion processes. The plant
design and operation can be found in Chapter I.
METAL COMBINATIONS EMPLOYED
The response of the activated-sludge process was measured
with each of three metal combinations. The combinations, denoted
as MC No. 1, 2, and 3, are given in Table 42. Two of the metals,
Table 42. METAL COMBINATIONS USED TO MEASURE RESPONSE OF
ACTIVATED-SLUDGE PROCESS
Metal combination
MC No. I
MC No. 2
MC No. 3
Metal in influent sewage, mg/liter
Copper3
0.4
0.4
0.3
Chromium
4.0
-
-
Nickel
2.0
2.0
0.5
Zinc a
2.5
2.5
1.2
Total heavy
metals, mg/liter
8.9
4.9
2.0
Total CN~
mg/Iiter
4.3
4.3
2.0
Fed as complex cyanides.
zinc and copper, were fed as soluble complex cyanides. Chromium
(VI) was introduced as potassium dichromate, and nickel, in the
form of nickel (II) sulfate.
Previous studies have shown that once the activated-sludge
process acclimates to cyanide, no difference in effects on treatment
exists between the metal as a complex cyanide or free cation (13, 16).
Metal combination No. 1 was based on data from the previous
studies on the individual metals and represents the approximate
threshold limit for each metal. Chromium was deleted from MC No. 2
because chromium at 4 milligrams per liter should have no effect
*Material in this chapter published previously in Proceedings of 18th
Industrial Waste Conference, Purdue University. See Reference 30.
97
-------
on the activated-sludge process (8); therefore, MC No. 2 should show
the same reaction as MC No. 1 if there was no interaction of the
metals. MC No. 3 was chosen in order to observe miniumum effects
and because the ratios of metals reflected average analyses obtained
from field samples.
EXPERIMENTAL CONDITIONS
The run in which MC No. 1 was used was divided into two parts.
In the first part a supplemented weak domestic sewage was used as
sewage feed (8,13,16,17). In the second part a strong domestic sewage
was employed with no supplement. The studies with MC No. 2 and
MC No. 3 were carried out entirely with the strong domestic sewage.
The sewage was fed from a common storage tank to duplicate
pilot-plant units operating at 350 liters per day capacity. One unit
received no metal and served as a control. The experimental unit
received the combination of metals by constant-head, calibrated,
capillary tubes, from which the metal solutions dripped into the
sewage feed line immediately ahead of the primary settler. The
sewage and metals were added continuously to the units at a constant
rate throughout the entire run.
Loading factors for the various unit operations during the
aeration phase are given in Table 43. The characteristics of the
sewage used during each run are given in Table 44.
Table 43. PILOT-PLANT DESIGN DATA AND LOADING FACTORS
Unit
Primary
settler
Aeration
tank
Final
settler
Loading
Capacity
Detention time
Surface overflow rate
Capacity
BOD loading
Aeration period
Capacity
Detention time
Surface overflow rate
factor
4.6 gal
1.2 hr
142 gpd/ft2
23.6 gal
0.34 lb/day/1,000 It3
aeration tank volume
0.60 Ib/day/lb VS under
aeration
6 hr
7. 9 gal
2 hr
102 gpd/ft2
INTERACTION OF HEAVY METALS
-------
Table 44. CHARACTERISTICS OF SEWAGE FEEDS AND PRIMARY EFFLUENTS
FOR CONTROL AND METAL-FED UNITS
Unit
MC No. 1
Control
MC No. 2
Control
MC No. 3
Control
COD, mg/liter
Primary
feed
463
366
456
407
498
409
Primary
effluent
284
269
275
284
290
286
BOD, mg/liter
Primary
feed
232
203
186
180
243
223
Primary
effluent
145
140
128
132
147
149
Suspended solids,
mg/liter
Primary
feed
342
252
350
269
348
287
Primary
effluent
159
150
180
183
155
158
Each week four 24-hour composite samples of feed, primary
effluent, and final effluent from the control and experimental units
•were analyzed for COD, BOD, and suspended solids. Daily turbidity
measurements were made on final effluent grab samples from the
control and experimental units. Analyses of total metals at all outlets
were performed on 7-day composite samples. Soluble metal analyses
were made on daily grab samples that had been filtered immediately
through a membrane filter and then composited for 5 days.
The run with MC No. 1 lasted 6 months. The MC No. 2 and
No. 3 runs were each of a 3-month duration. All units were ac-
climated to the experimental feed for 2 weeks before data were
collected. Analysis of the final effluent for cyanide, at this time,
showed virtually complete destruction of the cyanide.
Details of the anaerobic digestion procedures are given in the
literature (18). The loading factors for the digesters are given in
Table 45.
Table 45. DIGESTERS, CAPACITY, AND LOADING FACTORS
Digester data
Denoted as
Capacity, liters
Detention time, days
Organic loadingf
Ib VS/day/
1000 ft3
Primary sludge
digesters
Control
s lud ge
CP
8
17
67
Metal-bearing
sludge
MC No. IP
8
17
65
Primary and excess-activated-sludge
digesters
Control
sludge
CPE
16
16
37
Metal-bearing
s lud ge
MC No. 1 PE
16
16
39
"Average loading during test period.
Metal Mixture
99
-------
ANALYTICAL METHODS
The procedures used to determine BOD, COD, suspended solids,
turbidity, chromium, nickel, zinc, and cyanide have been previously
described in Chapters I through IV.
Copper was determined with neocuproine; volatile acids, by
the distillation method (tentative); alkalinity, by titration to pH 4.50
with a pH meter; and COj, by absorption in 30 percent KOH. These
procedures are all outlined in Standard Methods (10). The three
forms of nitrogen, NH3-N, NO2-N, and NO3-N, were determined also
according to procedures in Standard Methods (10).
The analytical method for each of the four individual metals
in a mixture of metals in sewage and sludges was tested for inter-
ference by the method of standard addition. In each case the assay
employed proved specific enough to eliminate interference by the
other metals in the ranges encountered.
RESULTS
Effects on Aerobic Efficiency
Data from the analyses performed on final effluent samples
during the runs were plotted on probability paper as frequency distri-
bution curves. The COD, BOD, and turbidity data for MC No. 1,2,
and 3 are given in Figures 41 through 49. Each figure includes
data on the proper control unit. Table 46 shows the arithmetic averages
of BOD, COD, suspended solids, and turbidity.
MC No. 1
0.01 0.05 .2 0.5 I 2 5 10 20 30 40 50 60 70 80 90 95 98 99 998 99.99
% OF OBSERVATIONS 5 STATED VALUE
Figure 41. COD of final effluents.
100
INTERACTION OF HEAVY METALS
-------
D
O
U
I I I I I I I
MC No. 2
Figure 42. COD oi final effluents.
a
o
U 40
MC No. 3
CONTROL UNIT
0
0.01
0.05 0.2 0.5
Figure 43. COD of final effluents.
0.01 0.05 .2 .5 I 2 5 10 20 30 40 50 60 70 60 90 95 98 99 99.8 9999
% OF OBSERVATIONS < STATED VALUE
5 10 20 30 40 50 60 70 80 90 95 98 99 998 99.99
OF OBSERVATIONS £ STATED VALUE
Metal Mixture
101
-------
QOI 0-05.1 -2 .5 I 2
10 20 30 40 50 60 70 80 90 95 98 99 99.8 99.9 99.99
; OF OBSERVATIONS
-------
0.01 0.05.1 .2 .5 I
2 5 10 20 30 40 50 GO 70 80 90 95 98 99 99.8 99.9 99.99
% OF OBSERVATIONS 5 STATED VALUE
Figure 46. BOD of final effluents.
MC No. 1
CONTROL UNIT ~
.01 0.05 0.2 0.5 I 2 5 10 2O 30 40 50 60 60 90 95 98 99
% OF OBSERVATIONS < STATED VALUE
Figure 47. Turbidity of final effluents.
Metal Mixture
103
-------
OOI 0.05 0.2 0.5 I 2 5 10 ZO 30 40 50 60 TO 80 90 95 98 9999.5 99~a
% OF OBSERVATIONS < STATED VALUE
Figure 48. Turbidity of final effluents.
n—i—r
MC No. 3 •
CONTROL UNIT
J 1 I I I I L_
0.01 0.05.1 0.2 0.5 I 2 5 10 20 30 10 50 60 70 80 90 95 98 99 99.8 99.99
% OF OBSERVATIONS 5 STATED VALUES
Figure 49. Turbidity of final effluents.
Table 46. CHARACTERISTICS OF FINAL EFFLUENTS FROM
CONTROL AND METAL-FED UNITS
Unit
MC No. 1
Control
MC No. 2
Control
MC No. 3
Control
BOD,
mg/liter
27
18
21
21
16
21
COD,
mg/liter
66
45
63
48
57
52
Suspended
solids ,
mg/liter
15
10
16
13
•)
12
Turbidity,
stu
39
26
74
32
22
16
104
INTERACTION OF HEAVY METALS
-------
These figures and Table 46 show that MC No. 1 and No. 2 had
a significant effect on the process; the effect of MC No. 3 combination
was of a borderline nature.
The fact that the MC No. 2 BOD data (Figure 45) showed no
effect is not surprising. The precision of the routine BOD test is
not as good as that of the COD procedure, and small differences
in effluent quality may not be apparent from the BOD test. The data
on COD, turbidity, suspended solids, and inhibition of nitrification
during the MC No. 2 run are sufficient to indicate an overall effect
as great as that of the MC No. 1 run,
The effects recorded are significant on the basis of COD, but
the overall reduction in plant efficiency for both the MC No. 1 and 2
runs is only about 5 percent. The turbidity curves in Figures 47,
48, and 49 include several points of very high turbidity and, therefore,
the entire data did not plot as a straight line. This is due to the
occasional receipt in the pilot plants of sewage containing excessive
collodial clay turbidity, and the sensitive nature of the turbidity
assay.
An interesting observation is that the BOD and suspended solids
were actually lower for the MC No. 3 run than for the control unit
(Table 46). This may be due to the heavier weight of the metal-fed
sludge, resulting in more efficient settling in the final settler. The
influence of the metals on sludge density index and volatile solids
content of the mixed liquor is shown in Table 47.
Table 47. EFFECTS OF METALS ON MIXED LIQUOR
Sludge density
index
Volatile
solids, %
Control
1.5
66.7
MC No. 1
3.2
57.9
Control
1.5
66,7
MC No. 2
3.4
61.8
Control
1.5
66.7
MC No. 3
2.4
63.8
Effects on Nitrification
Three forms of nitrogen were determined on final effluent
samples of MC No. 1 and No. 2 runs. The results for the MC No. 1
run are shown in Figures 50, 51, and 52. Similar results were ob-
tained for the MC No. 2 run. Data for the MC No. 3 run were not
obtained.
Metal Mixture
105
-------
~i r
n r
FINAL EFFLUENT OF PILOT
PLANT UNIT RECEIVING
MC No. 1 METAL SOLUTION
FINAL EFFLUENT OF
PILOT PLANT
CONTROL UNIT
DO OF FINAL
A*-EFFLUENT OF
/ PILOT PLANT UNIT -
\ / RECEIVING MC No. 1
^—J METAL SOLUTION
DO OF FINAL EFFLUENT OF
PILOT PLANT CONTROL UNIT
0 10 20 30
40 50 60
TIME, days
70 80 90
Figure 50. Ammonia nitrogen in final effluents.
FINAL EFFLUENT OF PILOT PLANT UNIT
RECEIVING MC No. 1 METAL SOLUTION
A
DO OF FINAL EFFLUENT OF PILOT
HI-PLANT UNIT RECEIVING MC No. 1
METAL SOLUTION
FINAL EFFLUENT OF
PILOT PLANT
CONTROL UNIT
DO OF FINAL EFFLUENT
OF PILOT PLANT
CONTROL UNIT
10 20 30 40 50 60 TO 80 90 100
TIME, days
Figure 51. Nitrite nitrogen in final effluents.
106
INTERACTION OF HEAVY METALS
-------
o
r
•FINAL EFFLUENT OF PILOT PLANT
CONTROL UNIT
FINAL EFFLUENT OF
— PILOT PLANT UNIT \i
\RECEIVING MC No. 1 METAL
\ , -•...... . SOLUTION
[/•-DO OF FINAL
EFFLUENT OF PILOT
PLANT UNIT RECEIVING
MC No. 1 METAL SOLUTION
W— DO'OF FINAL EFFLUENT OF
>. PILOT PLANT CONTROL UNIT
i..—
o
Q
TIME, days
Figure 52. Nitrate nitrogen in final effluents.
Pertinent observations made during the runs with MC No. 1
and 2 are as follows:
1. Ammonia-nitrogen concentrations (Figure 50) in the final
effluents of the metal-fed units were consistently higher
than those of the control for the same compositing period.
2. Nitrite-nitrogen concentrations (Figure 51) were erratic,
particularly in samples from the metal-fed unit, in which
they averaged slightly higher than in the control.
3, Nitrate-nitrogen concentrations (Figure 52) in the final ef-
fluents of the metal-fed units were usually less than 1 milli-
gram per liter, as contrasted to much higher values found
in samples from the control unit.
4. Dissolved oxygen concentrations in the mixed liquors and
final effluents of the metal-fed units were consistently higher
than those of the control.
The conversion of nitrite to nitrate is virtually completely
inhibited, the oxidation of ammonia to nitrite is erratic, and the
air requirement of the experimental sludge is not as great as that
of the control sludge. Equal amounts of air were introduced to the
units through rotometers, and the dissolved oxygen was measured
with a galvanic lead/silver probe (19). The presence of dissolved
oxygen in the experimental unit indicates that the inhibition of nitrifi-
cation was independent of oxygen concentration.
Metal Mixture
107
-------
Denitrification, with rising sludge, was visually evident in the
final settler of the control unit. No such activity was noted in the
metal-fed units.
Bozich (20), working with these metals individually, also re-
ported inhibition of nitrification, which appears to be a general symptom
of the toxicity of the heavy metals.
Distribution of Metals
Complete material balances for the four metals were performed
during the MC No. 1 run. Distribution of the metals in the various
process outlets is shown in Table 48. The range of observations
show considerable variation in the balance periods; however, the
average values agree well with those reported in previous studies
of the individual metals.
Table 48. METALS IN PROCESS OUTLETS
Metal
Copper
Chromium
Nickel
Zinc
mg/liter
% of
metal fed
mg/liter
% of
metal fed
mg/liter
% of
metal fed
mg/liter
% of
metal fed
Primary
sludge
18.2
19
44
5.4
9.8
2.3
59
11
Excess
activated
sludge
22
39.5
97
20.6
33
13.3
141
67.5
Final
effluent
0.22
45.8
2.6
63
1.2
69.4
0.26
9.7
Imbalance,
%
+ 4
-11
-15
-12
Overall
removal, %
54
37
31
90
Range
of observa-
tion, %
32-89
18-58
12-76
74-97
The importance of using a complete activated-sludge pilot plant
in the study of metal toxicity is shown in Table 49. Two important
factors are shown here. First, primary settling reduces slightly
the metal burden going to the aerator by removing some metals with
the primary sludge. Second, the chemical and physical characteristics
108
INTERACTION OF HEAVY METALS
-------
Table 49. EFFECT OF PRIMARY SETTLING ON METALS
MC No. 1
Metal
Copper
Chromium
Nickel
Zinc
Soluble metal
introduced in sewage, mg/liter
0.48
4.1
2.0
2.7
Total metal in
primary effluent, mg/liter
0.40
3.5
1.6
2.0
Soluble metal in
primary effluent, mg/liter
0.22
3.6
1.3
0.15
of the sewage may drastically alter the form of the metal originally
introduced. This is emphasized in the case of zinc. Ninety percent
of the added soluble zinc is converted to an insoluble form.
These results are similar to earlier observations reported by
Masselli (21). Soluble metal is defined in the present study as that
portion passing through an HA45 Millipore membrane, followed
by acid digestion of the filtrate before analysis. Table 50 shows
that metal in the final effluent, 'with the exception of zinc, is pre-
dominantly in soluble form and not associated with the suspended
solids fraction.
Table 50. SOLUBLE AND TOTAL METAL CONTENT OF FINAL EFFLUENT
MC No. 1
Total
Soluble
Metal, mg/liter
Chromium
2.7
2.4
Copper
0.19
0-14
Nickel
1.3
0.9
Zinc
0.26
0.04
The average metal content of the mixed liquor during the MC
No. 1 run is given in Table 51. If the influent metal concentration
Table 51. AVERAGE METAL CONTENT OF MIXED LIQUOR
MC No. 1
Metal in total solids, wt %
Chromium
1.00
Copper
0.22
Nickel
033
Zinc
1.40
3.0
Metal Mixture
109
-------
is considered, the mixed liquor has an affinity for the metals in the
following order in the system studied: Zn, Cu, Cr, Ni. This affinity
reflects accurately the overall removal pattern shown in Table 48.
ANAEROBIC DIGESTION OF SLUDGES
The sludges produced by the control unit and the experimental
metal-fed unit, MC No. 1, were digested anaerobically. Both primary
and mixed primary and excess activated-sludge digesters were operated
for each unit. The mixed digesters received a 3:7 (volume) ratio
of primary and excess activated sludge. No difficulties attributable
to the metals were encountered. The gas production and metal content
for each of the MC No. 1 sludges are shown in Figures 53 and 54 and
volatile acids, alkalinity, and pH in Figures 55 and 56. The control
digesters, receiving no metal-bearing sludges, gave results almost
identical with those shown in the figures. Volatile solids destruction
in all cases was satisfactory. Digestion of the MC No. 2 or No. 3
metal-bearing sludges was not studied, in view of the satisfactory
digestions of metal-bearing sludges of MC No. 1.
Initially, maintaining both control and experimental digesters on
the municipal sewage was difficult because of great fluctuations in
< — o
o
TIME, days
Figure 53. Digester receiving metal-bearing primary sludge.
110
INTERACTION OF HEAVY METALS
-------
20
TIME, days
Figure 54. Digester receiving metal-bearing primary
and excess-activated sludge.
START FEEDING OF METAL-BEARING SLUDGE
20 30 40 50 €0 70 80 90
Figure 55. Dig*«trr receiving metal-bearing primary sludge.
Metal Mixture
111
-------
Q
U
0) 4000
CO, OF GAS MIXTURE
J L
START OF MC NO. 1 FEEDING DIGESTION AT 30°C
— ipoo
<
-PH
TOTAL ALKALINITY AS CaC03
T~
0 10 20 30 40 50 60 7O 80 90 IOO
TEST PERIOD, days
Figure 56. Digester receiving metal-bearing primary and excess activated sludge.
the solids content of the sewage. The problem was solved by adding
to the feed sewage a domestic primary sludge from a reserve supply,
according to initial daily Imhoff cone measurements of sewage as
collected.
The digesters were started with digesting sludge from a municipal
primary treatment plant. Figure 53 shows that the initial total heavy-
metal content of this sludge was about the same as the metal content
of MC No. 1 in the experimental run. Very little change in the total
metal content of digested sludge was noted throughout the run. Figure
54 reveals a rising metal content of the digested sludge, eventually
leveling off, caused by the higher concentration of metals in the excess
activated sludge mixed with the primary sludge. The separate and
total metal content of the digested sludges are given in Table 52.
Table 52. TOTAL AND INDIVIDUAL METAL CONTENT OF DIGESTED SLUDGES
MC No. I
Digested
primary sludge
Digested primary
and excess
activated sludge
Metal , mg/liter
Chromium
47,0
88.1
Copper
18.5
22.3
Nickel
10.2
35.7
Zinc
55
122
Total
metals
130
268
Total metals,
% of solids
0.93
2.73
112
INTERACTION OF HEAVY METALS
-------
Each week the digester contents were also analyzed for soluble
metals; in no case was the soluble metal content above 1 milligram
per liter for any of the four metals. The previous studies of the
individual metals had shown that anaerobic digesters are efficient
in converting introduced soluble metal to an insoluble form. Coinci-
dental with this finding was the observation that weekly analyses of
the digesting sludges for free H2S* always showed a detectable amount
in both the control and experimental digesters.
DISCUSSION
Operation of the two phases in the run with MC No. 1 caused
no difference in the distribution of the metals throughout the process
or the effect of the metal on the efficiency of treatment. One phase
was supplemented with weak domestic sewage, and the other with
strong domestic sewage. With the two types of sewage used, the
described control pilot plant (8) treated the supplemented weak sewage
to 95 percent BOD removal; however, the strong domestic sewage was
treated to only 91 percent BOD removal.
Although the effect of the metal dosage in MC No. 1 on treatment
efficiency (Figure 41) is significant, it is not striking and is not more
than the effect one metal alone would have at this concentration.
No synergistic action was found in the combinations employed. MC
No. 2 with a total metal concentration of 4.9 milligrams per liter
gave an effect as large as that of MC No. 1 with a total metal
content of 8.9 milligrams per liter. A previous study (17) showed
that both 5- and 10-milligram-per-liter concentrations of nickel gave
about the same reduction in efficiency as MC No. 1 and No. 2. This
is shown in Figure 57. The nonlinear response with increasing
10 20 30 40 50 60 70 80 90
% OF OBSERVATIONS^ STATED VALUE
Figure 57. Effect of nickel on COD of final effluents.
95
98
* By evolution with CO2 into zinc acetate.
Metal Mixture
113
-------
metal dosage was also characteristic in the studies on chromium,
copper, and zinc (8, 13, 16). MC No. 3 with a mixed metal content of
2 milligrams per liter showed a borderline effect when all the measured
parameters were considered.
The nonsynergistic effect of multiple-metal dosage on activated-
sludge organisms was recognized by Dawson and Jenkins (22)
and Jenkins (23). Tarvin (24) reported no deleterious effects on the
aerobic or anaerobic systems of an actual plant receiving heavy
metals in concentrations approximating those of MC No. 1 in the
present study. The distribution of metals throughout the processes
was also similar to that reported here.
The inhibition of nitrification, while only a general symptom
of heavy-metal toxicity, is significant because the final effluent
from a plant so affected could contain excessive ammonia. A high
ammonia content can be a potential toxicant to fish in the receiving
stream, create a high chlorine demand if breakpoint chlorination
processes are employed, and possibly cause a large oxygen usage
because of stream nitrification after dilution.
The oxygen requirement of the metal-loaded system was not so
large as that of the control because less oxygen was used for nitrifi-
cation. Because nitrifying organisms did not acclimatize to the
metals during the entire test period, nitrification was never observed
in a metal-loaded system.
Material balances and tracking of the metals in the process
outlets for the MC No. 1 run agree well with earlier studies of the
individual metals, which indicates that there was no appreciable
interaction of the metals in combination. Stones, in a series of
studies on the distribution of metals in actual treatment plants (25,
26, 27, 28), observed values similar to those reported here. Of
the four metals studied, chromium (as chromate) can be expected
to be the most variable in efficiency of removal from the influent
sewage because the amount of removal is to a large extent controlled
by the amount of dissolved oxygen present in the system (8).
Even with complete conventional activated-sludge treatment,
considerable metal passes out of the plant with the final effluent.
The effects the metal content of the final effluent will have on the
receiving stream have not been considered in this study. Pettet
(29) has commented on this aspect, but reports no definitive research.
The metal removed from the influent sewage is concentrated
in the primary and excess activated sludges. Anaerobic digestion of
these metal-bearing sludges produced by the experimental unit was
satisfactory. The results were similar to results of previous studies
114 INTERACTION OF HEAVY METALS
-------
(8, 13, 16, 17). Anaerobic digestion of sludges is not interfered with
when the individual metals Cr, Ni, and Zn are present continuously
in the influent sewage at concentrations of 10 milligrams per liter.
Copper, continuously present in the influent sewage at a concentration
of 10 milligrams per liter, allows normal digestion of primary sludge;
difficulty with mixed digestion, however, may occur (18).
SUMMARY
A combination of four metals, with a total concentration of
8.9 milligrams per liter, had no great effect on the overall efficiency
of a pilot-scale activated-sludge plant. No synergistic action was
noted. No difficulty with the anaerobic digestion of the sludges pro-
duced by the plant was encountered. Approximately 90 percent of the
zinc, 54 percent of the copper, 37 percent of the chromium, and 31
percent of the nickel were removed from the influent sewage. The
metals, in combination, behaved independently in their distribution
throughout the process.
Nitrification in the experimental units was almost completely
inhibited. This was shown to be a general symptom of heavy-metal
toxicity. The ramifications of this inhibition are discussed.
With the exception of zinc, the metals passing through the
activated-sludge process and discharged with the final effluent are
predominantly in a soluble form. The effects of the metals discharged
to the receiving stream were not considered in this study.
Metal Mixture 115
-------
CHAPTER VI. SUMMARY OF PILOT-PLANT DATA*
The effects of copper, chromium, nickel, and zinc, individually
and in combination, on biological treatment processes studied at the
Robert A. Taft Sanitary Engineering Center have been reviewed in
the previous chapters. This study resulted from a suggestion by
the National Technical Task Committee on Industrial Wastes that the
Center study the metallic wastes from the plating industry from the
standpoint of their effects on biological treatment.
The work was conducted in pilot plants that were good simulants
of a sewage disposal system and were operated under sustained an-
alytical supervision. Sufficient observations were made to establish
statistically valid evidence of performance in systems with metal
input and metal withdrawal in general working balance. This chapter
summarizes these data.
EFFECTS ON AERATION PHASE
To relate results to the metal content of the influent sewage and
to duplicate field conditions as closely as possible, a complete
activated-sludge pilot plant (Figure 2)was used. The design and loading
factors of the pilot plant are representative of many municipal con-
ventional activated-sludge plants. The sewage feed to the plant
during the various studies was either a weak supplemented domestic
sewage (17) or a strong nonsupplemented domestic sewage. Both
type feeds give results indistinguishable by usual analytical measures.
Many investigators of metal toxicity have employed batch op-
eration or direct dosing of the metal to the aeration chamber. Data
from individual studies (Table 53) show that primary settling has
two effects on the metals before entry into the aeration tank. First,
the total metal content of the primary effluent is less than that of
the influent sewage, because some metal is removed with the primary
sludge. Second, the chemical and physical characteristics of the
sewage alter the form of the soluble metal introduced. This was
especially true in the case of zinc where 90 percent of the added
soluble zinc was converted to an insoluble form. The differentiation
of soluble and insoluble metal in all studies was made by filtration
*Material in this chapter published previously in Journal Water
Pollution Control Federation, Washington D. C. 20016. See Ref-
erence 33.
117
-------
of the sample through an HA45 membrane filter and by acid digestion
of the filtrate before analysis. Masselli (21) working with domestic
sewage, reported findings similar to those in Table 53.
Table 53. METALS IN PRIMARY EFFLUENTS
Metal
Chromium (VI)
Copper
Nickel
Zinc
Soluble metal
introduced in
sewage feed,
mg/liter
50
10
2.5
10
Metal in primary
effluent, mg/liter
Total
47
9
2.0
9
Soluble
38
3.0
1.0
0.6
The procedure used to determine the concentration of metal
in the influent sewage that would give a barely detectable reduction
in efficiency during the aeration phase of treatment can best be
explained by Figure 58, which shows the results of a study of copper
200
150
100
Q
o
u
50
••• CONTROL (NO METAL)
• 0.4 mg/liter Cu
X 1.2 mg/liter Cu
O 2.5 mg/llter Cu
D 5 mg/liter Cu
• 10 mg/liter Cu
10 20 30 40 50 60 70 80
% OF OBSERVATIONS < STATED VALUE
_L
90 100
Figure 58. Effect of copper, fed continuously as copper cyanide
complex, on COD of final effluents.
118
INTERACTION OF HEAVY METALS
GPO S2O—663—9
-------
(13). During each run data from an experimental pilot-plant unit
and a control unit receiving no metal were compared. The metal
was added continuously to a constant sewage feed of the experimental
unit. Two weeks of acclimation was allowed before data on the quality
of the final effluent were collected. This time interval is also required
for the metal content of the activated sludge to build up to a condition
of operating equilibrium. Final effluents from both units were assayed
daily for BOD, COD, suspended solids, and turbidity. The run for
any selected metal dosage was continued for 60 days to obtain sufficient
data. The values for the two units were then compared as frequency
distribution curves. The parameter of effluent quality in Figure 58
is COD; this is plotted as frequency distribution on arithmetic paper.
As shown on the figure copper present continuously at 0.4 milligram
per liter did not noticeably increase the COD of the experimental unit.
A copper concentration of 1.2 milligrams per liter, however, showed
a significant increase in COD. From this and the other parameters
measured, copper present continuously at 1 milligram per liter
in the influent sewage is concluded to be the threshold dose for the
aeration phase.
Another type plot in which frequency distribution curves were
plotted on probability paper was found to be useful. Readily available
statistical measurements are given by this type plot. If a straight
line is obtained with arithmetic probability paper, normal distribution
of data is verified. The 50 percent point is very close to the true
arithmetic mean of the observations, and the slope of the line is a
measure of the standard deviation. Figure 59 is such a plot of data
Q 40
O
U
I I I
METAL MIXTURE: Cr, Cu,
Ni, Zn; TOTALING 8.9 mg/liter
METAL
MIXTURE
0.01 05 2 0.5 I 2 5 10 20 30 40 50 60 70 80 90 95 96 99 998 99.99
% OF OBSERVATIONS < STATED VALUE
Figure 59. COD of final effluents.
collected during a study of the effects of a mixture of four metals on
the activated-sludge process. The need for extensive sampling is
shown here. The control unit had an average final effluent COD of
Summary
119
-------
45 milligrams per liter; however, continuation of this point to the
experimental unit shows that 12 percent of the time the experimental
unit final effluent had a COD of 45 milligrams per liter or less.
Copper and zinc are frequently used by the plating industry
as cyanide complexes. These two metals were studied in both the
soluble cation form (as sulfate) and as soluble cyanide complexes
(13, 16). Results show that once the activated sludge acclimates to
the continuous presence of either form of the metal, there is no
difference in effects on treatment efficiency. Figure 60-A shows
too
z 8O
0 20
Zn (CN)7
COMPLEX, 10 rng/liter
I I I I I I
ZnS04, 20 mg/liter
-1 1 1
01 23456789 10 0123456789 10
TIME, days TIME, days
Figure 60. Comparison of acclimation to complexed zinc and zinc sulfate.
that where turbidity of the final effluent was used as the measure
of treatment efficiency, after 8 days the system receiving a 20-
milligram-per-liter concentration of zinc cyanide complex had ac-
climated to cyanide and was producing effluent of stable turbidity.
The cyanide content of the effluent followed a similar pattern, with
almost complete removal of cyanide at the end of 7 days. Figure
60-B, from a run with a 10-milligram-per-liter concentration of zinc
as the sulfate, showed no such acclimation. Direct comparison of
the 10-milligram-per-liter concentration of zinc sulfate and a 10-
miiligram-per-liter concentration of zinc cyanide complex versus
the same control unit is shown in Figure 61. The BOD data were
collected after 2 weeks' acclimation, and no significant difference
between the two forms of zinc exists.
The reaction pattern of the activated-sludge process was the
same for each of the metals studied. A small dose of metal gives
120
INTERACTION OF HEAVY METALS
-------
20
Q
O
m
_1_
I I I
I
I
10 20 30 40 50 60 70 80 90
% OF OBSERVATIONS^ STATED VALUE
95
98
Figure 61. Cumulative frequency data on quality of final effluents
with zinc concentration of 10 mg/liter in sewage feed.
a significant reduction in treatment efficiency, but substantially larger
doses do not further decrease the efficiency greatly. Figure 62
graphically illustrates this situation.
100
u
,z, 80
60
40
20
CONCENTRATION OF METAL IN INFLUENT SEWAGE
Figure 62. Response of system to metal dosage.
Summary
121
-------
Table 54. CONTINUOUS DOSE OF METAL THAT
WILL GIVE SIGNIFICANT REDUCTION IN
AEROBIC TREATMENT EFFICIENCY
Metal
Concentration in
influent sewage,
mg/liter
Chromium (VI)
Copper
.Nickel
Zinc
10
1
1-2.5
5-10
Table 54 lists the concentration of metals that give a signifi-
cant increase in the usual parameters of judging treatment efficiency.
These may be considered threshold concentrations; it should be
borne in mind, however, that these limits were obtained under carefully
controlled laboratory operation. The significance of Figure 62 is
that the threshold concentration is mainly of academic interest and
actual plant situations are concerned with the plateau region of metal
dosage and response.
The results of these studies (30) show that the aeration phase
of biological treatment can tolerate in the influent sewage, chromium,
copper, nickel, and zinc up to a total heavy-metal concentration
of 10 milligrams per liter, either singly or in combination, with about
a 5 percent reduction in overall plant efficiency. Tarvin (24) working
in municipal plants reported conclusions similar to the above. Dawson
and Jenkins (22) from laboratory investigations and Jenkins (23)
from field experience also indicate this range of concentration.
Slug doses of metals to the activated-sludge process were also
studied. The concentration of metal that constitutes a harmful slug
dose is determined by the waste volume, the volume and characteristics
of the dilution water, the specific form of the metal, and the usage
of the stream below the point of effluent discharge. For convenience
only a single measure of effluent quality, such as an increase in
organic material passing through the plant, has been used to judge
a harmful slug dose. As an example, in Figure 59, the control unit
has a COD of 70 milligrams per liter or less 98 percent of the time;
then a harmful slug dose can be defined as that concentration of
metal that will yield an effluent COD in excess of this value for the
subsequent 24 hours of performance. The effects of slug doses
were observed on 4-hour metal doses to the influent sewage.
122
INTERACTION OF HEAVY METALS
-------
Table 55. METAL CONCENTRATION IN 4-HOUR SLUG
DOSE THAT WILL PRODUCE HARMFUL SLUG,
AS MEASURED BY COD
Metal
Chromium (VI)
Copper
Nickel
Zinc
Concentration
influent sewage, mg/liter
> 500
75
> 50<200
160
Table 55 gives the results obtained. To more accurately fix
these concentrations would require an inordinate amount of time
and expense. Table 55, therefore, is the best estimate of what con-
centration of metals causes an exceptional displacement of treat-
ment-plant performance as the result of a slug dose.
Not reported in the table are the results of slug studies in
which the metals were added as cyanide complexes. In these cases
the cyanide toxicity completely obscured the toxic effect of the metal.
In general, acclimation of the system to low concentrations of metals
or cyanide did not offer protection from slug doses.
The inhibition of nitrification by heavy metals has been previously
studied with regard to individual metals (20,29). A pilot plant that
received a combination of four metals also showed inhibition of
nitrification. There was no evidence of acclimation of the nitrifying
organisms to the metals. The oxygen requirement of this metal-
loaded sludge was less than that of the control unit because oxygen
for the biological transformation of ammonia to nitrate was not
utilized (30). Figure 63 shows the nitrate content of the final effluents
of a control and metal-fed unit. Inhibition of nitrification is regarded
as an important effect of metal toxicity. A plant so affected would
discharge all the influent nitrogen in excess of that needed for synthesis,
predominantly in the form of ammonia. Such an effluent would require
considerable chlorine if downstream breakpoint chlorination were used,
and nitrification in the receiving stream would use large amounts of
oxygen.
DISTRIBUTION OF METALS THROUGH THE PROCESS
Complete material balances of the metals were made during
each study. Table 56 summarizes these studies. The table is based on
the amount of metal fed toaunitduring a compositing period. Variation
between compositing period was common, as indicated by the range
of observations for the efficiency of the process in removing metals.
The percent metal unaccounted for in Table 56 is not a firm figure,
Summary 123
-------
5, 20
E
FINAL EFFLUENT OF PILOT PLANT
CONTROL UNIT
FINAL EFFLUENT OF PILOT PLANT
UNIT RECEIVING METAL MIXTURE;
DO OF FINAL EFFLUENT
OF PILOT PLANT UNIT
RECEIVING METAL MIXTURE
P DO OF FINAL EFFLUENT OF
PILOT PLANT CONTROL UNIT
4 E
o"
40 50
TIME, days
Figure 63. Nitrate nitrogen in final effluents.
Table 56. DISTRIBUTION OF METALS THROUGH ACTIVATED-SLUDGE
PROCESS WITH CONTINUOUS DOSAGE
% of metal fed
Outlet
Primary sludge
Excess activated sludge
Final effluent
Metal unaccounted for
Average efficiency of
process in removing
metal
Range of observations
Chromium (VI)
(15 mg/liter)
2 4
27
56
15
44
18-58
Copper
(10 mg/liter)
9
55
25
15
75
50-80
Nickel,
(10 mg/liter)
2.5
15
72
11
28
12-76
Zinc
(10 mg/liter)
14
63
11
12
89
74-97
but represents the cumulative errors involved in sampling sludges,
flow measurements, and analytical methods.
Metal balances were performed for each selected concentration
of the metals studied, not only those shown in Table 56. Each metal
124
INTERACTION OF HEAVY METALS
-------
was studied in about five increments over the range of 1 to 20 milli-
grams per liter. In addition, four metals were simultaneously traced
during a combination study (30). Over che concentration ranges studied,
no great difference in the efficiency of the process in removing the
metals was noted. Zinc and copper studied as the cyanide complexes
showed the same overall removal as when studied as the sulfates.
Stones, in a series of articles dealing with the fate of copper,
chromium, nickel, and zinc through municipal plants (25, 26, 27, 28),
records metal removals in general agreement with those reported
here. The distribution of metals given by Tarvin is also similar (24).
Chromium, introduced to an activated-sludge process as hexavalent
chromate, can show wide variation in concentrations at the various
process outlets. Reducing substances in the raw sewage can cause
precipitation of trivalent chromium with the primary sludge. Also,
under anaerobic conditions the organisms in the return sludge entering
the primary settler can utilize the oxygen of the chromate radical
and adsorb the trivalent chromium on the biological floe (8). Under
these conditions chromium removal can reach 90 percent.
Table 56 shows that a considerable portion of the metal introduced
is removed in the secondary sludge. The effects of the metals on the
mixed liquor are apparent even in the 1- to 2-milligram-per-liter
range. During 5 years of study no bulking was encountered in a metal-
fed system. The floe in the final settler quickly settled. Control
units frequently bulked. Table 57 shows the effects of a combination
of four metals on the sludge density index and volatile solids content
of mixed liquor.
Table 57. EFFECTS OF METALS ON MIXED LIQUOR SOLIDS
Analysis
Sludge density
index
% volatile
solids
Mixed liquor from
Control
unit
1.5
66.7
Metal mixture
8.9 mg/liter
3.2
57.9
4.9 mg/liter
3.4
61.8
2.2 mg/liter
2.4
63.8
With the exception of zinc, the conventional activated-sludge pro-
cess is not very efficient in the removal of metals from the influent
stream. The metal removed is concentrated at two points. In the
primary sludge, a maximum concentration would occur if all the
metal were removed with this sludge. Here the ratio of total flow
Summary
125
-------
volume to primary sludge volume is a limiting factor. Another
point of concentration is in the secondary sludge. Since the volume
of secondary sludge removed from the process may be small com-
pared to the flow through the process, concentration may be high at
this point.
There is no net removal of metal if the primary and secondary
sludges containing the metals are not permanently removed from the
line of flow. For instance, an extended aeration plant passes all
the metal to the receiving stream unless secondary sludge is removed.
The copper, chromium, and nickel discharged with the final
effluent from an activated-sludge plant receiving these metals are
predominantly in a soluble form. At an influent concentration of
10 milligrams per liter, only a small amount of zinc is discharged,
and this is an insoluble zinc. At higher influent concentrations greater
amounts of zinc are discharged as soluble zinc (16).
EFFECTS ON ANAEROBIC DIGESTION
The metal-bearing sludges produced by the pilot plants were
digested in single-stage nonmixed digesters. Organic loading was
for nonmixed operation. A small circulating pump was used once
each day to obtain representative samples of sludge for material
balances. In each metal study both primary sludge and combined
primary and secondary sludges were digested (18). The metal content
of the sludges fed to the digesters during several of the runs is
given in Table 58. The primary sludges were about 2 percent solids,
Table 58. METAL CONTENT OF SLUDGES FED TO DIGESTERS
Metal
Chromium (VI)
Copper
Nickel
Zinc
Continuous
dose in
influent sewage,
mg/liter
50
10
10
10
Primary sludge,
mg/liter
330
280
62
375
Excess activated
sludge, mg/liter
530
160
89
328
and the secondary sludges, about 0.5 percent solids during these
studies. On a percentage-of-solid basis, the metals in the secondary
sludge are concentrated to a much greater extent than in the primary
sludge.
126
INTERACTION OF HEAVY METALS
-------
A digester receiving combined sludges will contain more metal
on a percent-of-solids basis than a digester receiving primary sludge,
when operated at the same influent sewage metal concentration.
Digester failure due to heavy metals occurs at a lower influent metal
concentration in a combined-sludge digester than in a primary-sludge
digester (18).
The maximum continuous influent sewage metal concentrations
for satisfactory anaerobic digestion are given in Table 59.
Table 59. HIGHEST DOSE OF METAL THAT WILL ALLOW
SATISFACTORY ANAEROBIC DIGESTION OF SLUDGES
CONTINUOUS DOSAGE
Metal
Chromium (VI)
Copper
Nickel
Zinc
Concentration in
influent sewage, mg/liter
Primary sludge
digestion
>50
10
>40
10
Combined sludge
digestion
>50"
5
>10a
10
Higher dose not studied.
The response of the anaerobic system to metal dosage does not
exhibit a plateau region as does the aeration phase; it is an all or
none reaction. Digestion either proceeds normally or ceases en-
tirely. This may be more apparent than real, however, because
the analytical measures of assessing digester performance are not
as direct as those for the aeration phase.
The results of these metal studies show that in the cases of
chromium, nickel, and zinc an influent sewage metal concentration
of 10 milligrams per liter, either singly or combined, will not affect
digestion. Copper continuously present at 10 milligrams per liter
causes failure of combined-sludge digestion.
The prevailing conditions of anaerobic digestion are such that
soluble metal introduced with the feed sludges is efficiently converted
to an insoluble form. This is shown in Table 60.
During these studies no correlation of digester failure with
soluble metal in the digesting sludge could be found. The soluble
sulfide content of digesting sludge offered no direct measure of
the digester's ability to tolerate metals.
Summary
127
-------
Table 60. SOLUBLE METAL CONTENT OF SLUDGES COMPARED
WITH TOTAL METAL CONTENT OF DIGESTED SLUDGE
Metal
Chromium (VI)
Copper
Nickel
Zinc
Concentration
in influent
sewage,
rag/liter
50
10
10
10
Soluble metal
Feed sludges
Primary,
mg/ liter
38
2
10
0.3
Excess
activate d,
ing/liter
32
0.5
9
0.1
Digested
combined,
mg/liter
3
0.7
1.6
0.1
Total metal
Digested
combined,
mg/liter
420
196
70
341
A few slug doses to anaerobic digesters were studied. The
slugs to the digesters were in conjunction with the aeration slugs.
The sludges produced by the activated-sludge process during a metal
slug were collected and fed to satisfactorily operating digesters.
In no case was there any interruption of digestion caused by the
metal-bearing sludges. Concentrations of metals in the influent
sewage during these slug studies are given in Table 61.
Table 61. DIGESTERS FED COMBINED SLUDGES PRODUCED
DURING METAL SLUG TO ACTIVATED-SLUDGE PLANT
Metal
Chromium (VI)
Copper
Nickel
Concentration of metal in
sewage feed, mg/liter
500
410
200
Effect on
digestion
None
None
None
More-detailed studies were not conducted because the logistics
of digester operation make it unlikely that an operating digester
would be upset by the sludges produced during a slug period. This
belief is based on the facts that a digester is not on the main flow
stream and only a small part of the total flow through the plant reaches
it, and the daily additions to a digester are only a fraction of the
total digester volume.
SUITABILITY OF FINAL EFFLUENT
These studies have dealt with the effects of metals on the bio-
logical sewage treatment processes; however, with the increasing
reuse of surface water, the metal content of the final effluent becomes
128
INTERACTION OF HEAVY METALS
-------
important. Pettet(29) and Ettinger (31) have commented on this point.
Table 62 shows that each of the ions considered in our metal studies
have definite maximum limits either for drinking water or protection
of game fish.
Table 62. RECOMMENDED MAXIMUM CONTENT OF METALLIC
TOXICANTS AND ASSOCIATED IONS
Chromium (III)
Chromium (VI)
Copper
Nickel
Zinc
Cyanide
Ammonia, free
Maximum concentration, mg/liter
For drinking water
_
0.05b
1.0b
_
5.0b
0.01b
0.5 e
For game fisha
1°
20C
0.04d
0.8 c
0.8d
0.02d
2d
The values given for game fish may not be pertinent to any given situation, since there are wide
ranges of values depending upon the characteristics of the water, aquatic species involved, and
chronic versus acute exposure conditions. (Post-publication comment)
b Recommended by USPHS Drinking Water Standards (1962).
Estimated from data presented by Doudoroff and Katz, Sew. and Ind. Wastes 25:802. (1953).
d Estimated by C. M. Tarzwell, Robert A. Taft Sanitary Engineering Center.
International Commission.
The importance of considering the metal content of the final
effluent is illustrated by chromium. Moore (8) pointed out that 10
milligrams per liter would not interfere with conventional activated-
sludge treatment or anaerobic digestion of the sludges and that about
50 percent of the metal would reach the final effluent. Chromium
at this concentration would be acceptable from the standpoint of
plant performance; however, the low concentration of chromium
allowable in drinking water indicates that if downstream use of the
final effluent from a plant receiving a 10-milligram-per-liter con-
centration of chromium was for this purpose, considerable dilution
would be required before the supply would be acceptable.
DISCUSSION
The results of these studies show that for each phase of treat-
ment, aerobic, anaerobic, and discharge of final effluent, there are
different bases for judging the concentration of metals acceptable
in the influent sewage.
The plateau-type response of the aeration phase shows that
concentrations of metal many times higher than the threshold con-
centration can be received without greatly reducing efficiency. In
a situation in which removal of organic matter is not critical, the
most sensitive performance criterion may be the ability of the digester
Summary
129
-------
to handle the sludges produced. Since there are alternatives to
anaerobic digestion, this need not be a bottleneck.
In other cases the amounts of metals passing through the plant
to the receiving stream may be the factor that determines the con-
centration of metals permissible in the plant influent.
130 INTERACTION OF HEAVY METALS
-------
CHAPTER VII.
HEAVY METALS IN WASTE-RECEIVING SYSTEMS*
In zones of concentrated population in inland areas, industrial
and domestic waste water enters a system that automatically places
this discarded material where people must use it again. This system
started out very primitively with the invention of water-borne disposal,
but the amount of waste water reconditioning included in the system
is continually increasing, essentially because it must.
The most common elements of a contemporary waste-water
transportation and reconditioning system are shown schematically
in Figure 64. Clearly the inland sewage treatment plant is only one
STORM WATER
OVERFLOW
SEPARATE SEWER!
COMBINED SEWER
PUMPING
STATION
DIGESTFD
DIGESTER
SOLIDS ,
NDARY
TMENT
PF
DIGESTER
ANAEROBIC
SUPERN/
IMARY SLUDGE
k 1 PRIMARY
** -' TREATMENT
'
^TAN
\
SECONDARY
JSLUDGE
RETURN
CHLORINE
FEEDER
^>—
FINAL
EFFLUENT
BYPASSED
STORM WATER
AND SEWAGE
TO FURTHER USE
RECEIVING STREAM
(DILUTION, SEDIMENTATION, AEROBIC TREATMENT,
ANAEROBIC TREATMENT)
Figure 64. Common elements of municipal system for preparing
and sending sewage to reuse
(Main flow channel shown in heavy lines).
*Paper presented at Interdepartmental Natural Resources Seminar,
Columbus, Ohio, March 1963. See Reference 31.
131
-------
of the way stations along a system that converts waste water into
a portion of a general-purpose water resource to furnish water for
the following missions:
1. Drinking and culinary purposes.
2. Recreation and aesthetic enjoyment.
3. Work.
Examples of work performed by water include such domestic
chores as flushing the toilet or conveying ground garbage from the
home, and the complete array of industrial water usages including
irrigating, cooling, industrial processing, growing fish for commercial
harvesting, generating power, and economical moving of freight.
Many characteristics of waters receiving wastes are altered
by contaminants brought in by treated waste water. In the case of
the metallic wastes, factors requiring examination include the effects
of the metallic element or compound on the utility of receiving water
for growing fish or for reconversion to drinking water. Measures
of water quality required or recommended for drinking water and
estimates of the quantities of metals tolerated by desirable fishes
are shown in Table 62. Values such as those cited define permissible
boundaries where treated and diluted mixtures of sewage and metallic
wastes enter multifunctional bodies of water.
The literature contains a mass of conflicting information based
largely on shallow academic studies and fragmentary observation
by disposal plant operators. Answers were desired concerning
the amounts of metallic wastes that could be regularly discharged
as a normal part of manufacturing operations. In addition, there was
general interest concerning the effect of sudden slug doses such
as those that result from manufacturing accidents or haphazard
dumping. Our task clearly was to operate a good simulant of a sewage
disposal system under close and sustained analytical supervision,
and to make sufficient observations to establish statistically valid
evidence of performance in systems with metal input and metal
withdrawal in general working balance. In some cases there was
need for studying both simple salts and cyanide complexes of the
metals of interest.
As stated previously, the objectives of our studies have been:
1. To determine the extent to which sewage treatment processes
can tolerate metallic wastes without losing efficiency in
their treatment of organic pollutants in sewage.
132 INTERACTION OF HEAVY METALS
-------
2. To determine the extent of removal of metallic wastes in
sewage treatment plants and to follow their travel and con-
centration in various conventional sewage process units.
3. To develop modifications of sewage treatment procedures
that will make them more tolerant of metallic wastes or
more efficient in the removal of metals from sewage.
Additional objectives now include study of the effects of various
metals on nitrogen transformations, and the determination of the
effects of ratios of organic load to metal content on activated-sludge
behavior.
Some of the characteristics of the system that receives, and
reconditions, waste water and submits it for reuse are shown in
Table 63. Metals may be exposed to reducing conditions and to
Table 63. CHARACTERISTICS OF SEWAGE COLLECTION, TREATMENT,
AND REUSE SYSTEMS
Element
Collection system
Primary treatment
Secondary treat-
ment
Anaerobic
digestion
Receiving water
Common
ranges of
time in transit
0.2-24 hr
0,5-3 hr
1-12 hr
7-28 days
Reuse
may start
immediately
Environmental characteristics
Aerobic and anaerobic, 2 ft/sec
velocity. Movement may be in-
termittent.
Anaerobic frequently, although
it may be aerobic. Both condi-
tions may occur on 1 day.
Aerobic with limited but im-
portant micro and macro
anaerobic environments.
Anaerobic
Hopefully, aerobic; may be
aerobic with anaerobic bottom
areas.
Remarks
Sulfides frequently generated,
at least in portions of system.
May be omitted or very brief,
ahead of activated-sludge
treatment
Likely to concentrate biologi-
cally resistant materials by
inclusion and adsorption.
May eventually be superseded
as a frequent system com-
ponent.
Drinking water has been based
on water from sewage given
multiple recycling.
sulfides in the sewer, pumping stations, primary treatment tanks,
sludge thickeners, and even secondary treatment units as well as
the anaerobic digester. As will be pointed out later, we have found
reduction and sulfide formation important in determining the inter-
action of metals and treatment processes.
The pilot plant operated has been described in some detail
in Chapter I. The aerobic part of the system consists of a reservoir
Waste-Receiving Systems
133
-------
of sewage, metal and sewage feeding devices, and -replicate model
treatment plants designed to treat approximately 100 gallons of
sewage per day. The waste supply, metal addition procedures, feeding
schedules, analytical procedures, etc., (8, 13, 16, 17) are important
factors, but time does not permit going into these essentials of effective
investigation.
A diagram of the activated-sludge unit used is shown in Figure 2.
Digestion has been studied in conventional digesters using primary
sludge and excess secondary sludge produced by the primary and
secondary treatment of the waste flow. The detention periods have
varied from 16 to 26 days. A diagram of one of the digesters used
is shown in Figure 19. The mixing apparatus was operated only
once per day to thoroughly mix the sludge before withdrawing digester
sludge, and after the feed of raw sludge. Sludges studied have con-
sisted of primary sludge alone and mixtures made up of primary and
secondary sludge containing 70 percent secondary sludge by volume.
An anaerobic digester is relatively slug-resistant. The contents
are not part of the main stream of flow; a slug of waste can reach
the digester only through sludge fed, and the daily feed is usually
a small fraction of the digester contents. The digester is affected,
however, by the concentration of metallic wastes in the part of the
system where the main stream of flow occurs because the material
removed therefrom is sent to the digester. In an operating plant
most of the metallic wastes reaching a digester are metallic com-
ponents of the sludges.
Concentration of metals in the waste treatment system occurs
at two principal points. The first is in the primary sludge, where
the maximum concentration would occur if all the metal were removed
with this sludge. Here, the ratio of flow volume to primary sludge
volume is a limiting factor. The secondary sludge is a second point
where concentration can occur. Since the volume of secondary sludge
removed from the system may be small compared to the flow through
the system, concentration may be large at this point.
There is no net removal of metal if the primary and secondary
sludges containing the metal are not removed permanently from the
line of flow. For instance the extended aeration plant, common in
Ohio, passes all the metal to the river unless secondary sludge is
removed. If all the sludge is burned and the ash dumped in the river,
there is little or no net removal of metal.
In an extended aeration system, sludge concentration of metal
would be limited by losses of sludge, with the ratio of sludge con-
centration to influent metal concentration as the upper limit of the
number of concentrations possible. Such a limit might be approached
134 INTERACTION OF HEAVY METALS
GPO 820-663-10
-------
by an extended aeration plant
tent and does not waste sludge.
that treats wastes of high metal con-
Figure 65 shows patterns of metal retention encountered. Curve
1 represents the type of retention curve shown by chromium introduced
100
CONCENTRATION OF METAL IN RAW WASTE
Figure 65. Patterns of heavy metal retention by treatment systems.
as chromate. Here, relatively inefficient reduction of hexavalent
chromium in the system to more removable forms of chromium
causes the behavior shown. Type 2 curves have been shown by most
other metals. Curve 3 occurs where a limited amount of metal is
removed by a mechanism with a limited capacity such as adsorption,
coprecipitation, precipitation by a minor or inefficient precipitating
agent, or some combination of such actions.
SEWAGE TREATMENT PROCESS REACTION TO METAL,S
Figure 66 shows general modes of reaction of sewage treatment
processes to the continual presence of metals. In the aerobic part
of the plant small amounts of metal make no detectable difference
in the overall efficiency of treatment; larger amounts decrease
treatment efficiency slightly. Over a very considerable range of
metal increment, little increase in the organic content of the ef-
fluent occurs. While biological destruction or removal of organic
Waste-Receiving Systems
135
-------
matter is only slightly disturbed, the amount of oxygen used may be
reduced because biological oxidation of nitrogen may be halted or
minimized. Oxygen usage is obviously an observation of limited
value in appraising organic treatment effectiveness where varying
amounts of several kinds of sludge are withdrawn or where nitrification
may vary.
As shown in Figure 66, in the anaerobic digester we have en-
countered only two sets of conditions: the digester performs adequately,
100
so
z
UJ
y eo
40
UNITS OF METALLIC TOXICANT IN SEWAGE
Figure 66. Patterns of performance depreciation in sewage
treatment processes.
or the digester grinds to a halt and produces no gas. We have not
observed any case in which the digester continues to function with
reduced efficiency. As an additional observation, where digesters
have contained dissolved sulfides, they have always worked; however,
they have not always failed in the absence of sulfides.
Table 64 gives the tolerances of the complete activated-sludge
process and anaerobic digestion to metals in the plant influent. Table
65 outlines some of the impairments of receiving waters related
to the metal content of wastes. Where removal of organic matter
is not critical, the most sensitive performance criterion may be
the ability of the digester to handle the sludge. Since there are
136
INTERACTION OF HEAVY METALS
-------
Table 64. TOLERANCES OF SEWAGE TREATMENT PROCESSES TO
METALS FED AT CONSTANT CONCENTRATION
Metal in
raw sewage
Copper
Chromium
(as chromate)
Nickel
Zinc
Concentration in feed, ppm
Aerobic treatment
Mo observable
effect on COD
0.4
>50
1
10
Increases final
effluent COD 20 ppm
2.0
-
>10
>20
Anaerobic treatment
Digestion of
primary sludge
10
>50
>40
10
Digesti on of primary + excess
secondary sludge (3-7)
5
>50
10"
10
"Higher dose not studied.
alternatives to digestion, this need not be a bottleneck. In some
situations metallic materials passing through the process may be
highly objectionable. In our studies of chromate, we devised a means
of biochemical reduction of this material to less objectionable and
Table 65. EFFECT OF METALS IN RAW SEWAGE ON SUITABILITY OF SECONDARY EF-
FLUENTS (CONVENTIONAL ACTIVATED SLUDGE) FOR DISCHARGE TO WATERS TO BE RE-
USED
Metal
Copper
Chromium
(as chromate;
Zinc
Nickel
Cosmetic
factors
Ugly turbidity
above 2.5 ppm.
Color at about
2 ppm in waste
Satisfactory at
2,5 ppm. Ugly
turbidity at 10
ppm.
Ugly turbidity
above 1.0 ppm
Dilution required to
meet Drinking Water Std
When Cu content of raw waste
exceeds 1.3 ppm.
High dilution required.
Standard is 0.05 ppm.
Satisfactory up to 10 ppm.
Dilution required to
protect fishes
Tolerance of fishes estimated
at 0,05 ppm, A high dilution
may be required.
Only with very high chromate
(>20 ppm.)
5 ppm is upper limit. Is re-
duced below 0,8 ppm by
process.
Waste exceeding 1.0 ppm
requires dilution to be non-
toxic to fishes.
largely removed chromic chromium. This was done by modifying
the activated-sludge process to force it to take oxygen from chromates
(8).
Waste-Receiving Systems
137
-------
In our examination of aerobic systems it has been necessary
to base conclusions on replicate observations. The lower section of
Figure 28 illustrates the effect of two forms of zinc on the over-
all performance of the sewage purification processes observed. Both
sources of zinc caused a significant alteration of performance; and
it is also clear that a small number of observations could not yield
reliable data on the interplay of the metal and the system. The
observations on complexed zinc were observations of systems pre-
viously acclimatized to cyanide so that it was destroyed in the course
of aerobic treatment. Time is also required to enable the zinc content
of the activated sludge to build up to a condition of operating equilibrium,
HARMFUL SLUG DOSE
Another of our objectives has been to define the amount of metal
that, in a specific form, constitutes a harmful slug dose. What con-
stitutes a harmful slug dose is determined by the waste volume, the
volume and characteristic of the dilution water, and the usage of the
stream below the point of waste entrance. As an example, we can,
however, examine one facet of waste quality, the amount of organic
matter penetrating waste treatment processes. Figure 28 also shows
that the COD of the effluent from our plant exceeded 70 milligrams
per liter only about 2 percent of the time. If we then take a COD
average in excess of 70 milligrams per liter over a 24-hour period
•as a criterion of a harmful slug dose we arrive at the limits shown
in Table 66. These conclusions are based on 4-hour slugs of metal
fed to unacclimated sludges, and the worst 24 hours of subsequent
performance.
Table 66. HARMFUL SLUG DOSES FOR UPPER 2 PERCENT
OF CONTROL VALUES FOR 24 HOURS,
AS MEASURED BY COD
Metal
Chromate
Copper
Nickel
Concentration (4-hour slug),
ppm
>500
75
>50 <200
Such an evaluation should be based upon examination of per-
formance disturbances caused by enough slug doses to give a suitable
collection of reactions so that the behavior of large groups of slug
doses could be compared to the range of behavior shown by normal
substrates. Because these data cannot be procured without inordinatfe
expense, we can only offer an insecure estimate of what constitutes
an exceptional displacement of treatment plant performance as the
result of a slug dose of metal.
138 INTERACTION OF HEAVY METALS
-------
CHAPTER VIII. ORGANIC LOAD AND TOXICITY OF
COPPER TO ACTIVATED-SLUDGE PROCESS*
The Robert A. Taft Sanitary Engineering Center has studied
the effects of heavy metals, singly and in combination, on the activated-
sludge process. In these studies the organic load was approximately
the same. The present investigation considers the effect of organic
loading on the toxicity and the distribution of copper in the activated-
sludge process. Copper was chosen because its toxicity is significant
at a relatively low concentration and because preliminary work on the
effect of organic loading had been reported previously (16).
PROCEDURE
The acitvated-sludge pilot plants used (Figure 2) were designed
for complete treatment of sewage employing primary settling, aeration
with continuous sludge return, and secondary settling. Sewage was fed
at a constant rate. Sludge from the secondary settler was pumped
continuously to the first chamber of the aerator at a rate of about 35
percent of the sewage flow. Sludge wasting rates were adjusted to hold
mixed-liquor volatile-suspended-solids (MLVSS) concentrations between
1,000 and 1,200 milligrams per liter. Total detention time was 9 hours
with a flow of 100 gallons per day.
The sewage used was domestic in origin and of relatively low
strength because of ground-water infiltration. The sewage was
fortified each day by addition of homogenized fish meal to the main
holding tank. Average COD values of the sewage before and after
additions were 200 and 380 milligrams per liter. The nutrient char-
acter of the fortified sewage was found to be adequate. The calculated
BOD:N:P ratio was 100:6:3.
Three identical pilot plants were operated in parallel. One
unit received undiluted fortified sewage at a constant rate. Copper
sulfate solution was introduced at the sewage feed inlet continuously.
The other two units were fed the same sewage diluted approximately
2:3 with tap water. Copper sulfate solution of the same concentration
was introduced continuously in the influent sewage of one of these
units; the other unit received no copper and served as a control.
Since the MLVSS in the three units were maintained at approximately
*Paper presented at 19th Annual Purdue Industrial Waste Conference,
Lafayette, Ind. See Reference 40.
139
-------
the same levels, the organic load on the units receiving diluted sewage
was about one-half that of the unit receiving the strong sewage. The
effects of two copper concentrations were studied at each organic load
level. The concentrations selected were 1 and 5 milligrams per liter,
present continuously in the influent sewage. The 5-milligram-per-liter
copper run lasted about 6 months followed by the 1-milligram-per-liter
copper run, which lasted about 2-1/2 months. The low-organic-load
control was converted to a high-organic-load control during the last
2 weeks of the 5-milligram-per-liter copper run. The conversion
was brought about by feeding this unit undiluted sewage. Experimental
conditions with respect to units, load conditions, and copper runs
are summarized in Table 67.
Table 67. SUMMARIZED EXPERIMENTAL CONDITIONS
Pilot-plant
type
Control
Control
Copper fed
Organic loading
condition
Low
High
Low
High
Low
High
Copper level
and run
No copper
1 mg / liter
5 mg/iiter
COD, suspended solids, BOD, turbidity, and copper determinations
of sampled outlets were used to measure the effects or organic loading
and copper on the activated-sludge process.
SAMPLE COLLECTION AND ANALYTICAL METHODS
The sampling program for the study consisted of taking four
24-hour composites of feeds and primary and final effluents from the
three units per week of operation. COD, suspended solids, and pH
were determined on each of the 24-hour composites. BOD deter-
minations were made twice a week. The samples were collected by
means of an automatic sampler. The sewage flow was diverted for
a few seconds every 15 minutes to gallon containers set in refrigerated
chests. The 24-hour composites were homogenized by a blender,
and analysis started within 3 hours after final collection. Daily
140
INTERACTION OF HEAVY METALS
-------
turbidity determinations were made on grab samples of the final
effluents from each unit. Mixed-liquor suspended solids and volatile
suspendid solids were determined daily on grab samples taken from
the aerator tanks.
Copper balances were calculated from total copper analysis of
7-day composites of sludges, feeds, and final effluents and grab samples
from the aerators. Grab samples of mixed liquor were taken at the
beginning and ending of each compositing period to deter mine the weekly
accumulation of copper in the aerators.
Daily grab samples taken from primary and final effluent outlets
were passed through an HA45 membrane filter and composited for 5
days. These samples were analyzed for soluble copper, which is
arbitrarily defined as that portion of the total copper passing through
the filter.
As has been our practice in previous metal toxicity studies, the
units were acclimated to copper for 2 weeks prior to any sampling.
Analytical methods used are identical to those described in
Chapter III, and generally follow Standard Methods (10) procedures.
Exceptions were in the determination of COD, in which the mercuric
sulfate modification (32) was used, and in the polarographic* deter-
mination of copper. By the latter, samples for either total- or soluble-
copper analysis were wet-ashed with mixtures of nitric and sulfuric
acids. Last traces of organic matter were destroyed by small additions
of a mixture of perchloric and nitric acids. The dried residue was
treated with an electrolyte solution, approximately 6 M in NH.jOH
and 1 M in NH4C1. The solution was filtered through a sintered glass
funnel, treated with small additions of powdered Na2SO3 to remove
dissolved Oj, and then placed in the polarographic cell. The copper
polarogram was recorded between 0.0 and -1.0 volts versus a
saturated calomel electrode. The concentration of copper in the
unknown was calculated from a determination of the height of the
diffusion current trace at a half-wave potential of-0.45 volt against
copper standards treated similarly. Tests run on several sewage
samples with added amounts of copper gave 95 to 100 percent re-
coveries. The background concentration of copper in the raw sewage
was less than 0.1 milligram per liter.
RESULTS AND DISCUSSION
Organic Loadings Obtained on Pilot-Plant Units
Daily organic loadings were calculated for high- and low-load
units from determinations of COD of primary effluents and mixed-
* Sargent Model XV Polarograph.
Organic Load 141
-------
liquor volatile suspended solids of the aerators. The daily values
were plotted as a frequency distribution on arithmetic probability
paper. A straight-line plot indicates normal distribution of the
data. The arithmetic mean point on the curve is directly above the
50 percent abscissa point, and the slope of the line is a measure
of the standard deviation. The plots, means, and standard deviations
are shown in Figure 67. By comparing the means, it can be seen
1.05
. 0.95 -
MEAN = 0.73 ORGANIC LOAD UNITS
STANDARD DEVIATION ± 0.19
° MEAN=0.36 ORGANIC LOAD UNITS
STANDARD DEVIATION ± 0.09
METAL-FED
HIGH ORGANIC LOADING
• MEAN = 0.31 ORGANIC LOAD UNITS ,
STANDARD DEVIATION ± 0.08 /
METAL-FED
LOW ORGANIC LOADING
CONTROL
LOW ORGANIC LOADING
0.1 0.5 1 2 5 10 20 40 60 80 90
% OF OBSERVATIONS < STATED VALUE
98 99
99.9 99.99
Figure 67. Daily variation of organic loading on activated-sludge
pilot-plant units, 5-mg/liter copper run.
that the organic load of the unit receiving the undiluted sewage was
about twice that of the units receiving diluted sewage. The effect of
dilution is shown by the smaller slopes of the lines of the low-loaded
units. Daily organic loadings were not determined for the 1-milli-
gram-per-liter runs because pilot plant operation was not changed,
and the loadings shown on Figure 67 were not expected to change.
There is no standard way of expressing organic loading. For
convenience, the organic loadings are expressed in various equivalent
forms in Table 68.
142
INTERACTION OF HEAVY METALS
-------
Table 68. AVERAGE ORGANIC LOAD OF PILOT PLANTS
Pilot
plant
Low organic
load
Low organic
load
(5 mg Cu/liter)
High organic load
(5 mg Cu/liter
COD units
lb/day/
Ib MLVSS
under
aeration
0.31
0.36
0.73
lb/day/
1,000 fts
areation
capacity
22
24
57
BOD units
lb/day/
Ib MLVSS
under
aeration
0.14
0.16
0.32
lb/day/
1,000 ft3
aeration
capacity
10
11
25
Effects On Aerobic Treatment
The COD of the final effluents from the runs at 5 milligrams per
liter copper are shown in Figure 68. The mean COD and standard
deviation values were determined from the frequency distribution
plots of the data. The curves for the two control units show that
increasing the organic load increased the COD of the final effluent.
Copper at 5 milligrams per liter had about the same effect on COD of
the final effluent of the low-organic-load unit as doubling the organic
load Curves of the two metal-fed units show that effects of 5 milli-
grams per liter copper and organic loading on COD were roughly
additive.
I I I—I—I—I—I—I
i MEAN = 24 COD UNITS
STANDARD DEVIATION±6
' MEAN = 46 COD UNITS
STANDARD DEVIATION± 1 1
• MEAN = 44 COD UNITS
STANDARD DEVIATION + 7
• MEAN = 58 COD UNITS
STANDARD DEVIATIONJ13
.»-« CONTROL
LOW LOAD
0.01 0.1
99.99
1 5 20 40 60 80 90 95 99 99.
% OF OBSERVATIONS 2 STATED VALUE
Figure 68. Statistical comparison of COD of final effluents in 5-mg/liter copper run.
Table 69 gives the characteristics of sewage feeds and primary
and final effluents of each of the runs as arithmetic averages. Some
of the data of Table 69 are shown graphically in Figures 69, 70, and 71
to make comparisons easier. Figure 69 shows that at the 1-milligram-
per-liter concentration, the effect of copper was marginal at either
load condition and that the increase in COD was almost entirely due
to doubling the organic load. It was thought that, by choosing the run
Organic Load
143
-------
Table 69. CHARACTERISTICS OF SEWAGE FEEDS, AND PRIMARY AND FINAL EFFLUENTS"
Load
condition
Sampling
outlet
COD,
mg/liter
BOD,
mg/
liter
ss,
mg/liter
Turbidity,
stu
COD
removal
feed
to final
effluent,
%
Controls
Low organic load
High organic load
Feed
Primary
effluent
Final
effluent
Feed
Primary
effluent
Final
effluent
163
92
24
390
277
47
83
44
4
202
140
20
112
54
8
270
163
12
-
-
10
-
-
24
86
88
1-mg/liter copper run
Low organic load
High organic load
Feed
Primary
effluent
Final
effluent
Feed
Primary
effluent
Final
effluent
226
126
27
418
24?
50
115
63
5
195
137
11
161
92
9
273
136
9
-
-
9
-
-
16
88
88
5-mg/liter copper run
Low organic load
High organic load
Feed
Primary
effluent
Final
effluent
Feed
Primary
effluent
Final
effluent
134
92
44
355
217
58
54
36
10
147
96
21
84
55
24
238
123
23
-
-
33
_
-
44
67
84
3 Arithmetic average.
with 1 milligram per liter copper, any enhancement of the toxicity
as a result of the organic loading would be in sharp contrast to the
previously determined marginal toxicity of copper at high organic
144
INTERACTION OF HEAVY METALS
-------
1
I
jf 60
z
LU
13
_J
£ 4°
_j
<
z
El
u. 20
o
Q
O
u
ui 0
-
5 mg/liter
Cu
-
1 mg/liter
CONTROL
-
Cu
5 mg/liter
1 mg/liter
CO
NTF
OL
Cu
Cu
-
-
-
LOAD CONDITION
Figure 69. Comparison of effluent quality i n terms of COD.
03 £ 20
0 E
y, j2
uj z 10
< :D
o; _i
UJ U_
> u- n
5 mg/l
Cu
-
1 mg/liter
l~ CONTROL ,. Cu,
n II
ter
5 mg/l i ter
CONTROL
. 1 mg/liter
JCu
II
_
-
- LOW LOAD -
HIGH LOAD-
LOAD CONDITION
Figure 70. Comparison of effluent quality in terms of suspended solids.
1 mg/li
CONTROL "=u
ter
5 mg/liter
Cu
n
1 mg/liter
CONTROL .Cu 5 mg/liter
r^~i
1 1
Lr 1 nw 1 CIAD i . •-*!* HIGH 1 OAD »i
100 -
80 -
LOAD CONDITION
Figure 71. Comparison of aerobic efficiency.
load (16). This was not the case, and no differences in the effects
of copper at 1 milligram per liter were noted under either load
condition.
The data for the 5-milligram-per-liter copper runs in Figure 69
show that at the low-load condition COD in the final effluent increased
more than the COD in the final effluent from the high-load metal-fed
Organic Load
145
-------
unit, compared with their respective controls. This should not be
considered an indication of increased toxicity of copper at the low-
organic-load condition and can be explained by reference to the solids
data in Figure 70. These data show that the effluent of the low-loaded
unit receiving 5 milligrams per liter copper contained a larger
proportionate amount of suspended solids than any of the other ef-
fluents. This is probably related to the fact, as shown in Table
70 that the mixed liquor in this unit was high in ash content and,
Table 70. AVERAGE CHARACTERISTICS OF MIXED LIQUOR
IN5-mg/liter COPPER RUN
Characteristic
Suspended solids,
% volatile
matter
Sludge density
index
Average concentration
of copper, mg/liter
Low organic
loading control
74
2.0
-
Metal fed
Low organic
loading
62
2.1
151
High organic
loading
72
1.4
81
because of infrequent sludge wasting, was more typical of extended-
aeration sludge, which produces a more turbid effluent than con-
ventional activated sludge. The physical design of the final-settler
did not effectively remove this type of suspended matter from the
effluent. The increased COD of this effluent is related to the in-
creased solids present.
The data presented for the 1-milligram-per-liter copper runs
at both load conditions in Figure 70 show that, at this copper concen-
tration, no pronounced effect on the suspended solids of the effluents
was found.
The most notable effect of the relation of organic load and metal
toxicity is shown by the 5-milligram-per-liter copper run at low
organic load in Figure 71 and Table 69. The percent COD removal
was greatly reduced in this run, which by implication indicates an
increase in metal toxicity; however, this is only an apparent in-
crease. Direct comparison of the effluents from the two metal-fed
units shows that the effluent from the unit removing only 67 percent of
the influent COD was of better quality than that from the unit removing
84 percent. When a low-strength sewage, such as that used for the
low-organic-load studies, is treated, any increment of extra material
in the final effluent greatly influences the percent-overall-removal
calculation.
146
INTERACTION OF HEAVY METALS
-------
Because of the variability of the data collected, as indicated
by the standard deviations in Figures 67 and 68, it would be necessary
to study a wider range of organic loads, aerator solids, and metal
concentrations than those presented here, to establish firmly the
entire relation of organic load and metal toxicity. These data, however,
do show that moderate variation in organic loading does not markedly
alter the effect of copper on the treatment process.
Fate Of Copper
The effects of organic loading on distribution of copper through
process outlets during the runs with 5 milligrams per liter copper
are shown in Tables 70 through 73. Table 71 shows that the total
Table 71, AVERAGE CONCENTRATIONS AND FORMS OF COPPER
IN PRIMARY AND FINAL EFFLUENT COMPOSITES IN
5-mg/liter COPPER RUN
Process outlet
Primary effluent
Soluble copper1, mg'liter
Total copper1"', rng' liter
„, soluble copper
total copper
Final effluent
Soluble copper's rag/liter
Total copperb, mg/liter
^ soluble copper
/c total copper
Low organic
loading
1.2
4.4
27
0.5
2.2
23
High organic
loading
1.8
4.6
39
0.7
1.3
39
a Soluble copper determined on the acid-digested filtrate.
bTotal copper determined on the acid-digested unfiltered sample.
copper content in the two primaries was about the same. A small
difference, which appears in the final effluents, indicated less total
copper going out of the high-organic-load unit. The ratio of soluble
copper to total copper shows that the high organic load produced
a higher soluble-copper content in the primary and secondary ef-
fluents. This was probably due to the fact that the stronger sewage
contained a larger concentration of complexing agents than the dil-
uted sewage used to obtain the low organic loading did.
Copper balances were made at each load condition during the
5-milligram-per-liter copper run. Table 72 shows the average
results obtained after combining 13 weekly balances for each metal-
fed unit. The sludges removed from the high-organic-load unit
contained more copper than those removed from the low-organic-
Organic Load
147
-------
Table 72. AVERAGE COPPER BALANCES IN
5-mg/liter COPPER RUN
Copper fed
found in
outlets, %
Process
outlet
Primary
sludge
Excess
activated
sludge
Final
effluent
Unaccounted
Efficiency of copper removal,
feed to final effluent, %
Low organic
loading
8
33
50
-9
50
High organic
loading
12
43
39
-6
61
load unit, and, consequently, the efficiency of copper removal, ex-
pressed in percent, was greater in this unit. More copper was lost
in the final effluent of the low-organic-load unit in association with
the suspended solids of this effluent.
The average copper content of these sludges on a concentration
basis is given in Table 73. The primary sludge from the low-organic-
Table 73. AVERAGE CONCENTRATIONS OF COPPER IN SLUDGES IN
5-mg/liter COPPER RUN
Process outlet
Primary
sludge
Excess
activated
sludge
Low organic loading
mg/'Uter
71
388
Total dry
solids,
mg/g
11
77
High organic loading
mg/liter
116
180
Total dry
solids ,
mg/g
5.5
36
load unit had a copper concentration about one-half that of the high-
organic-load unit on milligram-per-liter basis. On a milligram-per-
gram total-solids basis, however, it contained more copper because
there were less solids associated with primary sludge of the low-
148
INTERACTION OF HEAVY METALS
-------
organic-load unit. The excess activated sludge produced by the low-
organic-load unit contained considerably more copper than that of
the other unit. This condition was brought about by infrequent sludge
wasting, which caused a buildup of copper in the aeration solids.
At the higher load condition, frequent sludge wasting wag necessary
to maintain a constant aerator solids level, and, consequently, copper
did not build up to the same extent.
The characteristics of the mixed liquor during the 5-milligram-
per-liter copper runs are given in Table 70. As a result of the greater
ash content of the sludge, the percent MLVSS value of the low-organic-
load unit was lower and the sludge density index value was higher
than the corresponding values of the high-organic-load unit.
SUMMARY
Moderate variations of organic loading did not markedly affect
the toxicity of copper to the activated-sludge process, under the
conditions employed in this study.
Increasing the organic load increases the COD of the effluent.
The effect of 5 milligrams per liter copper fed continuously
to a low-organic-loaded unit had about the same effect on COD of
the effluent as doubling the organic load. The suspended solids in
the final effluent of this unit were more than would be expected from
conventional activated-sludge treatment.
Organic loading altered the distribution and form of metal during
the 5-milligram-per-liter copper studies. The unit receiving the
high organic load was more efficient in removal of copper and pro-
duced a higher ratio of soluble to total copper in the process effluents.
In order to observe effects due to copper toxicity alone, reason-
able control of organic loading and aerator solids is necessary, and a
parallel control unit is recommended.
Organic load 149
GPO 820—663—1 1
-------
CHAPTER IX. A SLUG OF CHROMIC ACID PASSES
THROUGH A MUNICIPAL TREATMENT PLANT*
Previous studies conducted at the Robert A. Taft Sanitary
Engineering Center using pilot-scale, activated-sludge sewage treat-
ment plants indicated that a 10-milligram-per-liter slug of chromium
lasting 4 hours had no effect whatsoever on plant performance, while
100- and 500-milligram-per-liter slugs caused the plant efficiency,
as measured by BOD removal, to drop by 3 and 10 percent, respec-
tively, during the first 24 hours (8).
A field study was undertaken to complement the pilot-scale
study. The objective was to determine the effects of passage of a
prearranged slug of chromic acid on the efficiency of the Bryan,
Ohio, municipal sewage treatment plant. This chapter presents the
results of this study. In addition to the levels of chromium in the
plant processes attributable to the chromic acid slug, background
data on the concentrations of chromium, copper, zinc, and nickel
are presented also.
CONDUCT OF STUDY
Plant Description
The Bryan, Ohio, sewage treatment plant serves a community
of 7,400 persons with an equivalent population of 12,000. The plant
has conventional activated-sludge secondary treatment, an average
sewage flow of 0.8 mgd, and a design flow of 1.9 mgd. A flow chart
for the plant is shown in Figure 72.
The sewage, after passing through the grit chamber and com-
minutors, enters a wet well from which it is pumped into the primary
clarifiers. The effluent from the four primary clarifiers passes
through the first of the three aeration tanks. The mixed liquor from
this tank is divided equally between the remaining two tanks, from which
it flows into the final clarifiers. The effluent from the clarifiers is
discharged to drainage ditch number 40 and comprises 90 percent of
the flow. The ditch discharges into Lick Creek, a tributary of the
Tiffin River, 11-1/2 miles below the plant outfall sewer.
*Paper presented at 19th Annual Purdue Industrial Waste Conference
Lafayette, Ind. See Reference 35.
151
-------
DIGESTED SLUDGE
TO DRYING BEDS
Ul
(S3
H
M
o
H
FINAL EFFLUENT TO DITCH 40
Figure 72. Flow chart of sewage treatment plant Bryan, Ohio.
-------
Ninety-two percent of the sludge pumped from the final clarifiers
is returned to the first aeration unit, and the remaining 8 percent is
wasted to the raw sewage in the wet well. The return sludge flow
averages 87 percent of the raw sewage flow at Bryan.
The sludge in the primary clarifiers is pumped to the primary
digester twice each day from 8 to 9 in the morning and 3 to 4 in the
afternoon. Supernatant is withdrawn from the secondary digester
and returned to the sewage wet well. Digested sludge is pumped to
sand drying beds and ultimately disposed of in fill areas. Design
data and loading factors for the various plant processes previously
mentioned are presented in Table 74.
Table 74. PLANT DATA AND LOADING FACTORS
Proc
Primary
clarifier
Aeration
unit
Final
Anaerobic
digestion
Design data and loading factors
Capacity
Detention at 0.8 mgd
Surface overflow rate
Weir overflow rate
Capacity
Detention at 0.8 mg
Loading
BOD
COD
MLVSS
Air supplied
Capacity
Detention at 0.8 mgd
Surface overflow rate
Weir overflow rate
4 § 4,440 ft 3ea = 17,800 ft3
4hr
390 gpd/ft2
13,800 gal/ft of weir/day
2(3 23,300 ft3)
1 @ 18,000 ft3/
14hr
64,600ft2
0.5 Ib/day/lb VSS
221b/day/l,000ft3
l.Olb/day/lb VSS
47 lb/day/1,000 ft 3
720 ing/liter
2.4 ftVgal of sewage
4 J8 5,840 ft3 ea = 23,300 ft3
Shr
390 gpd/ft2
2,700 gal/ft of weir/day
Capacity
Detention time
Loading
Primary - 45,700 ft3
Secondary - 33,200 ft3
10 days
100 Ib VS/day/1,000 ft3
Chromic Slug
153
-------
Arrangements For Slug
The industries in Bryan, Ohio, discharging metallic wastes into
the municipal sewers include the ARO Corporation and a small-scale
job plater. The ARO Corporation designs and manufactures pneumatic
tools, industrial pumping units, lubricating equipment, and aircraft
products and has a metal-plating department that plates the various
components. The job plater handles orders from local industries and
organizations that have materials to be plated, but do not have plating
equipment.
With the cooperation of the officials of the city of Bryan, the ARO
Corporation, and the Department of Health of the State of Ohio, ar-
rangements were made to dump 150 gallons of chromic acid anodizing
solution into the municipal sewers, where it would ultimately enter
the sewage treatment plant.
The ARO Corporation furnished chromic acid solution that had
been used to anodize aluminum parts by the electrochemical conversion
of their surfaces to aluminum oxide, which increased their corrosion
protection. The chromic acid in the anodizing bath is inactivated by
neutralization because the alumina dissolves as the anodic film forms.
The procedure used at ARO to renew a bath is to withdraw a portion
of the bath periodically and replace it with fresh chromic acid. Of
one of these portions, 150 gallons was saved. The chromic acid
solution had a pH of 0.8, and the 150 gallons contained 50 pounds of
hexavalent chromium. The personnel at ARO dumped the solution
into their sewer in 9 minutes, beginning at 8:00 a.m., December 4, 1963.
Sampling Procedures
An around-the-clock 5-day sampling program, to determine the
concentration and distribution of heavy metals in the aerobic and
anaerobic treatment processes, before and after the chromic acid
entered the plant, was carried out with specific emphasis on chromium.
Grab samples of the sewage were taken every 5 minutes during
the height of the slug. Grab samples of the primary clarifier effluent
and mixed liquor were taken at various time intervals to establish
the variation of chromium in these steps of the process. Daily com-
posites of raw sewage, primary clarifier effluent, final clarifier
effluent, and return sludge were made up of hourly grab samples.
All sampling was done by hand, and where applicable, proportioned
to flow.
Analytical Methods
Chromium was determined by using the permanganate-azide
method outlined in Standard Methods (10). A polarographic procedure
154 INTERACTION OF HEAVY METALS
-------
was used to analyze samples for copper, zinc, and nickel. The samples
for metal analyses were evaporated and digested with acid to remove
the organic matter according to the procedure outlined in Standard
Methods. The other methods used in this study also conform to those
given in Standard Methods. Soluble metals are defined as those passing
through an HA45 membrane filter.
RESULTS
Chromic Acid Slug
The slug of chromic acid began arriving at the plant at 10:43
a.m., 2 hours and 40 minutes after it was dumped at the ARO Cor-
poration. It lasted for 1 hour in the incoming sewage. The con-
centrations of chromium in the raw sewage and the primary clarifier
effluent during the slug and several hours thereafter are presented
in Figure 73. The pH values of the sewage and primary effluent
during the slug are also included.
The sewage had a chromium concentration of 500 milligrams per
liter and a pH value of 5.7 during the height of the slug. Traces of
chromium were noted in the effluent at the weir of one of the four
primary clarifiers 15 minutes after the peak concentration occurred
in the sewage. The peak concentration of chromium in the primary
clarifier effluent was 65 milligrams per liter with a corresponding
pH value of 6.8. Eighty percent of the chromium in both the sewage
and primary effluent was in solution.
500
400
:= 300
200
- SEWAGE
^PRIMARY EFFLUENT
rPRIMARY EFFLUENT
D.....
•" -P-"
10AM 11AM 12 NOON 1PM 2PM 3PM 4PM
Figure 73. Concentration of Chromium in sewage and primary effluent.
5 PM
Chromic Slug
155
-------
The variation in concentration of chromium in the primary
sludge on the basis of milligram per liter and milligram per gram
of total solids (TS), the variation in quantity of chromium, and the
variation in the volume of primary sludge during the study are all
illustrated in Figure 74. The quantity of chromium in the sludge
pumped to the digester after the slug was similar to that pumped
routinely before the slug.
40
20
g
260
160
120
80
40
n
U
10
9
e
i
i 6
3 4
2
1
0
1
O
z
c
—
--
c
UJ
0
Q
=}
—1
g g 88
— o o °. o
"~
—
-
-
-
-
—
-
_
_
~-
1
1
1 I
1
I
1
1 I ' '" "T
t
O
^
_j
oO
0
o
o
o
i 8
1
,
i
i
i
\
i
j_
1
1
1
1
1
1
i
i
1
80
o
0. 0
"^ -O
1
0 0
g 8
o" -o
r
i
T
1
o
8
—
—
—
1
—
--
1
—
—
—
|
—
—
MON TUE WED THUR FRI SAT SU
Figure 74. Chromium in primary clarUier sludge.
in Figure 75 are the concentrations of chromium
solids in mixed-liquor samples taken before and
entered the plant. The chromium increased from
3 milligrams per liter just prior to the slug to 13
liter 3 hours after and remained at an average con-
centration of 10 milligrams per liter during the remainder of the
study. The concentration of chromium on a basis of milligrams per
gram of total solids in the return sludge is shown in Figure 76. The
chromium is significantly higher after the slug.
Presented
and suspended
after the slug
an average of
milligrams per
156
INTERACTION OF HEAVY METALS
-------
o
I/)
s*
z-
Q- E
^
_J
<
1-
1200
800
400
0
' ' . ' L-.
^ — • —
- r^**-
_ 0 ' '
—
-
L. 'II
. SLUG
THUR Ffll SAT
Figure 75. Chromium in aeration unit.
1
1
Figure 76. Chromium in return sludge.
The variation in the concentration and quantity of chromium
in the final effluent is presented in Figure 77. Ninety percent of
the chromium leaving the plant in the 1-day period following the
slug was in a soluble form. Chromium was detected in Lick Creek
in a concentration of 0.3 milligram per liter approximately 22 hours
after the slug entered the plant. The concentration before the slug
was less than 0.1 milligram per liter.
The concentration of chromium in the primary digester and the
variation in the daily gas production are illustrated in Figure 78.
The cubic feet of gas produced per pound of volatile solids fed to
the digesters for the 3-day period following the slug is also included
Chromic Slug
157
-------
5
4
£ 3
_- 2
u
0
^ 5
£ ,
~v, 3
E Z
U '„
1 1 | 1 1 1
- | 1 1 1 <, 11 1 1 ,11-1 1 n 1 U
|- SLUG
1 1 | 1 1 '
! , . !^^ . . i
Figure 77. Chromium in final clarifier effluent.
in Figure 78. The digester contents had the following characteristics;
pH, 6.8; volatile acids, 800 milligrams per liter; alkalinity (CaCO3),
1,800 milligrams per liter; temperature, 77° F.
»-- o
-10 -5 0
DAYS BEFORE SLUG
10 15 20
DAYS AFTER SLUG
30
Figure 78. Primary digester, chromium and gas production.
A material balance was made to account for the chromium
dumped into the municipal sewers at the ARO Corporation. Itemized
in Table 75 are quantities of chromium attributed to the slug in the
various plant processes and process outlets. The quantity of chromium
in the primary sludge is the difference between the sewage and primary
clarifier effluent quantities.
The quantity of chromium entering the plant in the sewage was
calculated by plotting the concentration of chromium during the slug
versus the volume of sewage and determining the area enclosed by
this curve. This calculated quantity (43 pounds) and the 50 pounds
dumped at ARO were averaged to give the value in Table 75. The
quantity of chromium in the primary effluent was determined by
158
INTERACTION OF HEAVY METALS
-------
Table 75. DISTRIBUTION OF CHROMIUM IN THE PLANT
Source
Chromium, Ib
Sewage
Primary effluent
Primary sludge (by difference)
Aeration tanks
Waste sludge
Final effluent
47
37
10
25
4
10
using concentration and flow data. The quantities of chromium in the
aeration tank, final effluent, and waste sludge were determined by
subtracting quantities of chromium in a 1-day period prior to the
slug from the 1-day period immediately after.
Since the waste sludge is returned to the primary clarifier, a
portion of the 4 pounds of chromium from the slug leaves via the
primary sludge while the remainder enters the aeration chamber via
the primary clarifier effluent, and some of the 4 pounds of chromium
in the waste sludge is, therefore, included in the quantity of the
aeration tank. The sum of the quantities of the chromium in the
aeration chamber, the final effluent, and waste sludge is similar
to that calculated in the primary effluent.
COD, BOD, SS, and turbidity data for 12-, 8-, and 4-hour com-
posite samples of final effluent taken before and after the slug entered
the plant are presented in Figure 79.
Qi
-1 r-
frr^-^
H--'1-"1
1
SUN MON TUE WED THUR FRI SAT SUN
Figure 79. COD, BOD, Suspended solids, and turbidity in final clarifier effluent.
Chromic Slug
159
-------
Background Metals
Illustrated in Figure 80 are the daily average concentrations of
metals and cyanide in the sewage (solid lines) and primary clarifier
effluent (dashed lines). The significant increase in the chromium
concentration of the 24-hour composite beginning Wednesday was due
to the chromium slug. A slug of copper entered the plant on Friday,
but was not detected until later analysis of samples. It is evident
from Figure 80 that the copper slug was associated with cyanide in
a copper cyanide complex, since a significantly higher concentration
of cyanide was also detected in the Friday sewage composite than
in previous composite samples. The concentration of chromium in
the 24-hour sewage composite on the day of the slug was less than that
in the primary effluent sample for the same period. In the case of
the copper slug, the sewage composite had a higher concentration than
the primary effluent had.
1 1
-
1 1
1 1
1 1
-
-
1 1
-
f — •-.—-— -1
I , 1«_^_
' 1 1
SEWAGE ~
PRIMARY EFFLUENT —
I
I f™— «r™|
—T
• SLUG
1
- r i ~
i
i
1 ' '
-} , H
j_Jrr
i
i....
-
Figure 80. Background metals in sewage and primary effluent.
160 INTERACTION OF HEAVY METALS
-------
3
en?
Source
Sewage
Metal
type
Total
Primary
effluent 1 Total
Final
effluent
Soluble
Total
Soluble
t
1
l Concentration in 24-hour composites, mg/liter
Chromium
Avg
0.8
0.9
0.5
i
1 0.2
0.2
!
Range
0.6-1.1
Copper
Avg
0.2
0.6-1,5 0.2
0.1-1.2
0.2-0.3
0.1-0.4
0.09
Range
0.17-0.25
0.1 -0.3
Trace_0.2
|
0.1
0,1
0.04-0.1
0.05-0.2
Zinc
Avg
2.2
1.8
0,3
0,2
0.2
Range
1.4-2.9
1.2-2.7
0.1-0.5
0.17-0.29
O.H-0.30
Concentrations in primary clarifier influent attributed to digester supernatant and waste sludge, mg/liter
Digeste.
supernatant
Waste-
activated
O5 sludge
h-1
Total
Total
2.6
1.2
i
i
0,1
0.2
-
2.7
1.5
-
Nickel
Avg
0.05
0.06
0.06
0.05
0.05
n
Kange
0.03-0.07
0.05-0.09
0.03-0.10
0.03-0.07
0.04-0.14
Total
of 4
metals
3.3
3.0
1.0
0,5
0.5
0.06
0.01
-
5.5
2,9
-------
A summary of the averages and ranges of concentrations of
metals in the sewage, and the primary and final clarifier effluents
for the 5-day study is presented in Table 76. The 24-hour com-
posite samples reflecting increases in metal concentrations due
to the slugs of chromium and copper are not included in the table.
Zinc was present in larger average continuous concentrations in
the sewage than any other metals. Zinc averaged 2.2 milligrams per
liter while chromium, copper, and nickel, with slug concentrations
omitted, averaged 0.8, 0.2, and 0.05 milligram per liter, respectively.
Since the digester supernatant and waste-activated sludge are
returned to the sewage prior to its entering the primary clarifiers,
the concentrations of metals in the primary clarifier influent con-
tributed by these sources are also included in Table 76. The data
were obtained by dividing the quantity of each returned to the sewage
during the 5-day study by the total volume of sewage during the
same period. The concentrations of chromium and zinc attributed
to supernatant and waste sludge are greater than those in the sew-
age. The data in Table 76 indicate that approximately 100 percent
of the metals in the final effluent were soluble as compared with
35 percent in the primary effluent.
Presented in Table 77 are the concentrations of metals in the
mixed liquor, and primary and return sludges. The average con-
Table 77. METALS IN MIXED LIQUOR, AND PRIMARY AND RETURN SLUDGES
Sludges
Primary
Return
Mixed
liquor
Units
mg/liter
mg/g
total solids
mg/liter
mg/g
total solids
mg/liter
Total metals
Chromium
125
4.7
17.2
5.3
9.3
Copper
36.1
1.3
2.2
0.7
1.2
Zinc
242
11.7
20.4
6.3
9.9
Nickel
1.7
0.05
0.1
0.04
0.07
centrations of metals in the primary digester and digester supernatant
are shown in Table 78. These tables are based on the average of 5
days of analyses.
162
INTERACTION OF HEAVY METALS
-------
Table 78. METALS IN DIGESTER AND DIGESTER SUPERNATANT
Source
Digester
Digester
supernatant
Digester
Units
mg/liter
mg/g
total solids
mg/liter
mg/liter
Total metals
Chromium
88
4.3
77
Copper
27
1.3
4,0
Zinc
220
10.6
82
Nickel
2
0.1
1.9
Soluble metals
0.09
0.13
0.16
0.05
The BOD, COD, and suspended solids data for 24-hour composite
samples of raw sewage, primary clarifier effluent, and final clarifier
effluent collected during the study are presented in Table 79. The
sewage contained 24 milligrams per liter of total nitrogen and the
primary effluent 38 milligrams per liter. The higher nitrogen in
the primary effluent is caused by recycling digester supernatant.
No nitrate nitrogen was present in the final effluent, either before or
after the chromic acid slug. Ammonia nitrogen was present in the
final effluent in a concentration of 19 milligrams per liter during the
study. The dissolved oxygen levels in the first aerator and the final
clarifiers were 0.4 and 1.1 milligrams per liter, respectively.
Table 79. SEWAGE AND PROCESS EFFLUENT CHARACTERISTICS
Analysis
BOD
COD
Sus-
pended
solids
Raw sewage,
Avg,
mg/liter
325
603
164
Range,
mg/liter
275-359
481-754
100-216
Primary clarifier effluent
Avg,
mg/liter
216
451
141
Range,
mg/liter
182-256
391-517
98-166
Removed,
33
24
13
Final clarifier effluent
Avg,
mg/liter
25
90
25
Range,
mg/liter
20-30
85-96
21-30
Removed,
%
92
85
84
DISCUSSION
A controlled slug of chromic acid entered the treatment plant and
lasted for 1 hour in the sewage, with 95 percent of the chromium
Chromic Slug
163
-------
entering the primary clarifier in 25 minutes. The chromium slug
short-circuited along the bottom of one of the primary clarifiers and
appeared in the primary effluent before any trace of the characteristic
greenish-yellow color was noted in the clarifier itself. Owing to
detention in the clarifiers, chromium from the slug appeared in the
effluent up to 12 hours after the slug entered.
Thirty-seven pounds of chromium was accounted for in the
primary effluent. With a value of 47 pounds, which is an average of
the quantity of chromium dumped at the ARO Corporation and that
calculated in the sewage at the plant, 10 pounds had to be accounted
for in the primary sludge. Since the background quantity of chromium
was unknown for the primary sludge removed at 3:00 p.m., the day of
the slug, this value arrived at by difference could not be confirmed.
The 37 pounds of chromium leaving the primary clarifiers is
reasonably accounted for in the aeration units, final effluent, and waste
sludge, as is shown in Table 75. If the figures of 47 pounds of chromium
entering the plant and 10 pounds leaving via the final effluent 1 day
after the slug are used as a basis, then approximately 80 percent was
retained by the plant processes.
Since the return sludge rate is significantly greater than that of
the typical average activated-sludge plant, more metal is recirculated
back to the aeration chambers to be detained and dribbled out in
small quantities via the final effluent. This is evident from the
chromium concentration in the mixed liquor shown in Figure 76.
Three days after the slug entered the plant, the concentration was
10 milligrams per liter in the aeration chambers, which is only 3
milligrams per liter less than the peak value several hours after
the slug.
The reason for the significant increase in the concentration of
chromium in the digester contents after the slug is obscure (Figure
78), since the 10 pounds in the primary sludge attributed to the slug
was only 5 percent of the quantity of chromium in the digester prior
to the slug. The quantity of chromium pumped to the digester after
the slug was similar to that pumped before the slug. Gas production
was not adversely affected. An anaerobic digester is not on the
main flow stream, and coupled with its large retention volume, it
is greatly less subject to shock than is commonly thought.
In view of the precision of the parameters used to evaluate
effluent quality, and the normal variation from hour to hour and
day to day that can be expected when dealing with a municipal treat-
ment plant, there was no significant difference in overall effluent
quality attributable to the slug of chromic acid.
164 INTERACTION OF HEAVY METALS
-------
The difference between the 24-hour composite concentrations of
chromium in the sewage and primary effluent was due to the method
of sampling. The 24-hour composite samples were made up of 24
hourly grabs taken on the hour. A sample of raw sewage was taken
while the slug was in progress, since 95 percent of the chromium
entered the plant in 25 minutes and the entire slug lasted only 1 hour,
and this was the only grab sample of the sewage containing chromium
from the slug. In the case of the primary effluent, in which the
slug lasted about 12 hours, 12 samples that contained chromium were
taken from the slug. This would account fox- the higher 24-hour
concentration in the primary effluent.
The same reasoning applies to the copper slug, except that the
sewage sample must have been taken near the time the peak con-
centration occurred in the sewage, which introduced more copper
into the sewage composite in one grab sample than several grabs
of primary effluent did for its 24-hour composite.
If both slugs had occurred shortly after the hour and ended
shortly before, the 24-hour raw sewage samples would not have in-
dicated the presence of the slugs.
If the unannounced slug of copper and cyanide entering the plant
on the last day of the study had occurred sooner, and a significant
effect on plant efficiency had taken place, an effect attributable to
chromium would have been difficult to determine. This incident
illustrates one of the differences in pilot-scale and full-scale studies.
In pilot-scale work, full control can be exercised over the influent
to the plant, but a municipal plant operator has no such control.
The slug of copper cyanide was not detected during the sampling
program by the project personnel since there was no visible in-
dication of its presence as is the case with chromium, which is readily
identified by its greenish-yellow color. No data are available on the
effect of the copper cyanide slug on the quality of the plant effluent.
In view of the gas production shown in Figure 78, however, the digester
was not significantly affected by the copper slug.
The sum of the background metals in the sewage, when the
chromium and copper slug data are omitted, was approximately 3
milligrams per liter, with zinc responsible for two-thirds of this
concentration.
Since the Bryan plant receives less than one-half of its design
flow, the detention times in the processes are longer than those in
the average plant. The return sludge rate is also higher at Bryan,
averaging between 80 to 90 percent of the sewage flow. The percent
removal of chromium in a plant with a shorter detention period and
a 20 percent return sludge rate may not be so high, and, consequently,
Chromic Slug 165
-------
higher concentrations of chromium in the receiving water would be
expected.
Even though this study has demonstrated that the biological
processes employed by municipal sewage plants are reasonably toler-
ant to slugs of chromium, it is not recommended that chromium wastes
be indiscriminately dumped into the sewers. With the increasing
reuse of surface water, the metal content of a plant effluent becomes
important.
SUMMARY
A field study to determine the effects of a controlled slug of
chromic acid on the efficiency of the Bryan, Ohio, municipal sewage
treatment plant was undertaken. The slug contained 47 pounds of
chromium, and concentrations reached 500 milligrams per liter
in the sewage. Ninety-five percent of the chromium entered in a
25-minute period and the entire slug lasted 1 hour. Approximately
80 percent of the chromium in the slug was retained by the plant
processes, and no significant adverse effects on these processes,
both aerobic and anaerobic, were noted.
166 INTERACTION OF HEAVY METALS
GPO 82O—663—12
-------
CHAPTER X. FOUR MUNICIPAL TREATMENT PLANTS
RECEIVING METALLIC WASTES*
Four municipal sewage treatment plants were selected for a
field survey concerning the receipt of copper, chromium, nickel,
and zinc; distribution of the metals in the plant processes; and effects
of the metals on the biological phases of treatment.
The plants were chosen on the bases of a history of receiving
metallic wastes, good management by cooperative officials, and prox-
imity to the Robert A. Taft Sanitary Engineering Center.
Activated-sludge plants in Bryan, Ohio; Grand Rapids, Michigan;
and Richmond, Indiana, and a high-rate trickling filter plant in
Rockford, Illinois, were each visited by a field crew for a 2-week
period. A brief synopsis of the plants is shown in Table 80. The
Table 80. MUNICIPAL SEWAGE TREATMENT PLANTS
Location
Bryan,
Ohio
Richmond,
Indiana
Rockford,
Illinois
Grand Rapids,
Michigan
Population
served
7,400
46,000
175,000
225,000
Average
daily
flow,
mgd
0.8
6.8
28.5
35.0
Design
flow,
mgd
1,9
18.0
(Including
recycle)
45.0
44.0
Average
sewage
BOD,
mg/liter
325
113
128
96
Type of
treatment
Activated
sludge
Activated
sludge
High-rate
trickling
filter
Activated
sludge
design and loading factors for each plant can be found in Tables 74,
81, 82, and 83.
The objective of this study was to find whether field observations
would substantiate earlier pilot-scale studies and enable this large
reservoir of data to be used to advise municipalities on the effects
that metallic wastes would be expected to have on their treatment
procedures.
*Submitted to Journal of Water Pollution Control Federation,
ington, D. C. 20016.
Wash-
167
-------
Table 81. UNIT DIMENSIONS AND OBSERVED LOADING FACTORS
GRAND RAPIDS, MICHIGAN
Process
Primary
clarifiers
Design data and loading factors
Capacity
Detention time
at 35 mgd
Surface overflow rate
Weir overflow rate
12@16,OOOft3L70000ft3
4 § 69,500 ft3 /
2 4 hr
745 gal/ft 2/day
90,200 gal/ft/day
Aeration
units
Capacity
Detention time
at 35 mgd
Loading
BOD
COD
MLVSS
Air supplied
6 @ 173,000 ft3= 1.04 x 106ft3
5.3 hr
0.5 Ib/day/lb VSS
16.5 lb/day/1,000 ft3 aeration volume
1.6 Ib/day/lb VSS
56.5 lb/day/1,000 ft3 aeration volume
540 mg/liter
0.5 ft3/gal sewage
Final
clarifiers3
Capacity
Detention time
at 35 mgd
Surface overflow
rate
Weir overflow rate
6 @ 94,000 ft3 = 565,000 ft3
2.4 hr
892 gal/ft2/day
12,600 gal/ft/day
Anaerobic
digestion
completely
mixed by gas
recirculation
Capacity
Detention time
in completely
mixed unit
Loading
Digested sludge
1 completely mixed § 150,000 ft3
5 storage
13 days
170 Ib VS/day/1,000 ft3
Dewatered on sand and removed to fill
area or filtered and incinerated
"Only five clarifiers in use at time of study.
The plants were sampled for the usual measures of treatment
efficiency by analyses of 24-hour composites, collected hourly by
hand. An additional 8-hour composite was kept during the expected
peak of flow and concentration for the 9:00 a.m. to 5:00 p.m. working
day. Samples were •proportioned to flow. Compositing periods for
the various processes were staggered in time, on the basis of theoretical
168
INTERACTION OF HEAVY METALS
-------
detention times, in order to follow the flow through the plant. Sludges,
because of their nonuniform nature, cyclic pumping, and hand sampling,'
were the most difficult to sample. All numbers reported in the text are
reported to two significant figures, because the combination of sampling
Table 82. UNIT DIMENSIONS AND OBSERVED LOADING FACTORS
RICHMOND, INDIANA
Proce
Primary
clarifier
Aeration
units
Final
clarifier
Design data and loading factors
Capacity
Detention time
at 7 mgd
Surface overflow rate
Weir overflow rate
4 @ 22,000 ft3 ea = 88,000 ft3
2.3hr
1,200 gpd/ft2
133,000 gal/ft of weir/day
Capacity
Detention time
at 7 mgd
Loading
BOD
COD
MLVSS
7 @ 79,800 ft3 ea = 558,000 ft3
14.3 hr
0.09 Ib/day/lb VSS
10 lb/day/1,000 ft3 aeration volume
0.29 Ib/day/lb VSS
28 lb/day/1,000 ft 3 aeration volume
1,650 mg/liter
Capacity
Detention time
at 7 mgd
Surface overflow rate
Weir overflow rate
3 IctrculT131^ 153,000 ft3
3.9 hr
600 gpd/ft2
5,200 gal/ft of weir/day
Anaerobic
digestion
with one
primary
digester
in opera-
tion
Capacity
Detention time
in primary
digester
Loading
2 primary digesters - 200,000 ft3
3 secondary digesters - 196,000 ft3
20 days
112 Ib VS/day/1,000 ft3
Municipal Treatment
169
-------
error, short sampling period in relation to a year of plant operation,
and analytical error indicates that greater accuracy is not justified.
All analytical procedures employed in this study were according
to Standard Methods (10) with the exception of those for COD, copper,
Table 83.
UNIT DIMENSIONS AND OBSERVED LOADING FACTORS
ROCKFORD, ILLINOIS
Process
Design data and loading factors
Primary
clarifier
Capacity
Detention time
at 28.5 mgd
Surface overflow rate
Weir overflow rate
2 § 134,000 ft3 = 268,000 ft3
1.8 hr
950 gal/ftVday
143,000 gal/lineal ft of weir/day
Trickling
filters
(high rate)
Capacity
Loading
BOD
COD
Hydraulic
4 @ 150 ft dia, 5 ft deep
88,300 ft3 ea = 353,000 ft3
85 lb/1,000 ft3/day
252 lb/1,000 ft3/day
28 x 106 gal/acre/day
Final
clarifiers
Capacity
Detention time
at 28.5 mgd
Surface overflow rate
Weir overflow rate
4 @ 106,000 ft = 423,000 ft3
l.Shr
1,600 gal/ft2/day
47,800 gal/lineal ft of weir/day
Anaerobic
digestion
(completely
mixed by gas
recirculation)
Capacity
Detention time
Loading
Digested sludge
24 days
100 Ib VS/1,000 ft3/day
Digested sludge is pumped to holding
tanks and then to lagoons. Sufficient
holding time is available for the for-
mation of supernatant, which is releas-
ed to the Rock River, The settled
sludge is removed to fill areas when
dry.
170
INTERACTION OF HEAVY METALS
-------
zinc, and nickel. The COD procedure used was the mercuric sulfate
modification to eliminate chloride interference(32). The metals copper,
zinc, and nickel were determined by utilizing a polarographic technique.
In Tables 84 through 87 the metal contents of the various pro-
cess effluents of each of the four plants are given. The 24-hour com-
posites, 8-hour composites, and the soluble metals in the 8-hour
composites are tabulated for three of the plants; no 8-hour composites
were collected at the Bryan plant. Comparison among the average
total metal composites for the 24-and 8-hour periods shows that the
receipt of metals by the municipal plants, on a concentration basis,
is approximately constant throughout the day. Rockford is the only
plant that shows a significantly larger metal concentration during the
9:00 a.m. to 5:00 p.m. workday over the 24-hour composite.
The relatively constant dosage of metal to the plants was an
unexpected finding, but can be explained by holdup in the interceptors,
varied industrial work schedules, and feedback to the process in
the form of digester supernatant and waste activated sludge. For
instance, at Bryan (Table 84), because of limited digester volume
and full drying beds, an excessive amount of poorly settled digester
supernatant was fed back to the process during the study period.
This resulted in a computed metal concentration for the total plant
flow several times that received via the incoming sewage. The full
amount of this computed concentration of metal did not reach the
aerators because most of the excess activated sludge and solids in
the supernatant were recycled back to the digester with the primary
sludge. Feedback is the reason there is no apparent metal removal
through the primaries at Bryan and Richmond. Grand Rapids recycles
a small volume of a good supernatant and a low-solids waste sludge.
Rockford does not recycle digester supernatant. These latter two
plants show metal removal through the primaries. The digesters
and the return activated sludge act as reservoirs of the metals that
continually impress a metal dosage on the biological system.
Tables 84 through 87 are based on average concentration; con-
sequently, no accurate estimate of removal of the metals by the plant
processes can be made from them, because of variable flow patterns
and variable daily metal dosage.
These tables show that zinc at all four plants exists predominantly
in an insoluble form. Nickel passes through the plants almost entirely
in the soluble form. Chromium and copper exhibit erratic solubility
behavior in the primary effluents when compared among the various
plants. The metals in the final effluents are mostly in a soluble form,
except for zinc in the final effluent from Rockford. This solubility
pattern of the metals had been previously demonstrated in pilot plant
studies (33).
Municipal Treatment 171
-------
ro
2
at
03 M
^ D
w s
U <
rTl Ofl
D 6
o "8
K •£
0 g.
< •§
& ^
03 o
t*J
on y]
f-< OJ
2 -^
P D.
b. °
b, °
W ^
tn S
U .5
o. .°
Z 2
c^i "c
^ s
S g
[x] U
S
CO
OJ
2
CO
, yj
5 ^r CD
jS'o |
0)
y
u
G
N
OJ
a.
a.
o
U
E
o
g
OJ
Ofl
1
OJ]
<
o
Ofl
CD
X
00
^J.
.
5D
CD
OS
oc
<
DJ
1
>?
•<
S §
s~°
(D
O
3
O
CO
ro
CD
S
CD*
in
o
o"
o
ro"
tj-
*
CN
CN
CO
O
1
CN
CD
CN
0
^
'
I
CD
00
CD
3
o
D
OJ]
CO
0 0
CO ^
0* 0
1 1
un ro
o o
CD O
s s
o o
t> LO
CN CD
CN ?-i
^H CD
oo ro
^-1 CD
CN
^ 0*
0 |
1 in
r-< O
CD O
CN i— I
CD CD
in CN
i-4 ^H
I 1
CD CD
CTI in
o o
OJ
•3 i
o -o
>1 ^
cd CD
6 ^
VC ^>
0 0
CD O
1 I
CO TJ-
CD O
CD CD
in in
CD 0
CD CD
CO CO
0* 0
I |
CN i— I
CD CD
CN CN
o" o
T— ( CN
CD 0
1 1
s s
CD O
^H — i
CD O
rO ^T
CD* CD*
CN 1-1
CD CD
CN CN
O O
£
_CO 3
"o *3
H c/)
~^
-H (D
CD ^3
1
edback, i
OJ
-a
OJ
1
1
c
o
c
JS
a.
c
c
0
03
C
OJ
u
c
o
u
•D
_QJ
3
a.
e
o
3
1
s
CD
1
^
CN
1
O
I
CN
"Q
^
C
CO
QJ CO
M S
as.
s s
CN
1
CO
O
o*
1
in
t-H
1
CN
CD*
1
CN
I-H
"o
H
jU
o> ra at)
w T! "H
« u ^J
^ to W
CO
1
o
1
CN
<*
1
ro
CD
1
00
ro
03
"o
H
E
172
INTERACTION OF HEAVY METALS
-------
g_
o'
•5'
3
0>
S
Table 85. METALS IN PROCESS EFFLUENTS: GRAND RAPIDS, MICHIGAN, SEPTEMBER 1963
Sewage
Total metal f
Primary effluent
Total metal \
Soluble metal
Final effluent
Total metal V
Soluble metal
Composite
period,
24
8
24
8
8
24
8
8
Chromium
Avg
3.6
3.8
3.2
3.5
2.8
2.5
2.6
1.7
Range
0.7-5.6
0.6-5.1
0.6-6.3
0.6-5.3
0.3-4.0
1.0-3.3
1.0-3.8
0.2-3.1
Copper
Avg
1.4
1.6
1.5
1.4
1.4
1,6
1.6
1,3
Range
0.7-2.4
0.3 3.7
0.6-2.8
0.4-2.3
0.5-2.7
0.4-2.9
0.3-3.2
0.2 2.6
Zinc
Avg
1.5
1.5
1.0
1.0
0.2
0.8
0.7
0.3
Range
0.6-2.5
0.4-2.2
0.4-1.5
0.4-1.6
0.1-0.4
0.6-1.2
0.6-0.9
0.2-0.6
Nickel
Avg
2.0
2.1
1.8
1.9
1.7
1.8
1.8
1.6
Range
1.3-3.4
0.9-2.9
1.0-2.4
0.8-2.2
1.0-2.5
1.0-2.2
0.8-2.1
Total
metals
8.5
7.5
7.8
6.1
6.7
6,7
4.9
Digester
supernatant
Waste
activated
sludge
Sum
—
-
-
Computed concentrations in total plant flow introduced by feedback, mg/liter
0.1
0.5
0.6
-
-
-
0.01
0.05
0.1
-
-
-
0.05
0.2
0.2
-
-
-
0.01
0.05
0.06
-
-
-
0.2
0.8
1.0
-a
oo
-------
Table 86. METALS IN PROCESS EFFLUENTS; RICHMOND, INDIANA, AUGUST 1963
Concentrations for 14-day period, mg/liter
Source
Sewage \
Total metal /
Primary effluent
\
Total metal >
Soluble metal
Final effluent
Total metal >
Soluble metal
Composite
hr
24
8
24
8
8
24
8
8
Chromium
Avg
0.8
0.3
0.8
0.7
0.3
0.2
0.1
0.04
Range
0.2-2.1
0.2 1.2
0.3-1.8
0.4-1.0
0.01-1.2
0.01-0,5
0.01-0.5
0.01-0.1
Copper
Avg
0.2
0.2
0.3
0.3
0.1
0.07
0.05
0.04
Range
0.1-0.4
0.1-0.5
0,2-0.6
0.2-0.3
0.2-0.3
0.04-0.2
0.03-0.1
0,01-0.1
Avg
0.3
0.3
0.4
0,3
0.1
0.1
0.1
0.1
Zinc
Range
0.1-0.5
0.2-0.5
0.3-0.9
0.3-0.5
0.04-0.1
0.1-0.2
0.1-0.2
0.1-0.2
Nickel
Avg
0.03
0,03
0.03
0.1
0.04
0.02
0.02
0.02
Range
0.01-0.1
0.01-0.1
0.01-0,05
0.02-0.2
0.01-0.1
0.01-0.03
0.01-0.04
0.01-0.1
of 4
metals
1.3
0,8
1.5
1.4
0.6
0.4
0.3
0.2
H
M
O
H
i
a
Digester
supernatant
Waste
activated
sludge
Sum
-
-
-
Computed concentrations in total plant flow introduced by feedback, mg/liter
0.4
0.4
0.8
-
-
-
0.4
0.3
0.7
-
-
-
0.4
0.3
0.7
-
-
-
0.05
0.02
0.07
—
-
-
1.3
1.3
2.3
-------
ro
§
OS
w
CQ
§
U
O
z
OS
u
o
H f
8 1
in "^
o E
Z -id"
J o
d 1
§ 1
^ ro
fi< ^
U 0
O
OS g
i I
W c
J c
W
c/>
U
U
O
OS
a,
z
^J
H
W
00
_0)
.Q
03
H
«*"3
"o ^ V
f"" ° S
QJ
y
u
c
N
QJ
a,
a,
c3
S
3
1
U
CD
DO
CO
K
Qfl
^
^
OJ
QJD
C
CO
Qfl
^
dj
WJ
a
a
bJD
QJ
GC
C
ro
Q^
OD
OJ
•r; -a"
% o
o .;; i-
E "
O
U
OJ
CJ
d
0
on
OO i-j
k£i a>
CN
^ 1
1 00
CN O
0 0
o\ o_
O --H
. o
ro i— i
j |
CM O
t-- r--
CN ro
-7 2
1 1
^J;
1 1 !
ro ro CN
O* O O
o> o o>
0 ^ 0
ro ro i-t
1 1 1
a^ oo CN
o o o
O i— I LO
C4 CN O
^O LO LO
CN t^ LO
! 1 1
LO CN CN
o o o
ro l>- LO
CN 0 LO
CN CO CN
1 1 1
L/I ro i— i
0 CD 0
LO 00 --H
^r co oo
CN
Jj _ 3
I£H CO "^J
>> S OJ
03 -t* ^
E .2 =1
£ (2 S
•^ CN O
^ ui ro
1 ! 1
LO LO \O
O O O
O^ Q O
O t-I rH*
I> CO LO
i-H »-< O
1 1 1
00 O r-J
O* r-H O
ro TT CN
i-H ,-*; O
ro (r^ LO
1 1 1
LO M- CN
O CD O
O »^O CN
LO
LO OO ^
^ CN O
O CD O
CN CN ^O
rH ^ CD
T OO OO
CN
C CO
a re CD
S ~s e
a> S cu
CO 03 D
.S "o "o
Municipal Treatment
175
-------
The concentrations of the metals in the various sludges pro-
duced by the plants are given in Tables 88 through 91 along with
the average metal content of the raw sewages. Modest amounts of
the metals in the raw sewage can produce sludges containing several
percent metal on a total solids basis. The mixed liquor at Richmond,
the plant receiving the least amount of metal, has about the same
concentration of metals as Grand Rapids, the plant receiving the
most metal. This is because Richmond carries a higher mixed-
liquor solids.
A unique feature of the Richmond plant is the practice of feeding
the municipal garbage to the anaerobic digesters (34). The garbage
amounts to 40 percent of the total volatile matter added, but the metal
content is so low that only 1 percent of the metals added to the digester
comes from the garbage (Table 90).
Table 88-A. METAL CONTENT OF SLUDGES PRODUCED BY BRYAN PLANT
Type
Primary
sludge
Excess
activated
sludge
Mixed
liquor
Digesting
sludge
Digester
supernatant
Digesting
sludge
Chromium
mg/liter
125
17
9
88
77
mg/g
5
5
-
4
4
Copper
mg/liter
36
2
1
27
4
mg/g
1
0.7
-
1
0.6
Zinc
mg/liter
242
20
10
220
82
mg/g
12
6
-
11
7
Nickel
mg/liter
2
0.1
0.07
2
2
mg/g
0.05
0.04
-
0.1
0.1
Soluble metals
0.1
-
0.1
-
0.2
-
0.05
-
Table 88-B. METALS IN RAW SEWAGE
Metal
Average concentration of
metal entering plant, mg/liter
Chromium
0.8
Copper
0.2
Zinc
2.2
Nickel
0.05
176
INTERACTION OF HEAVY METALS
-------
Table 89-A. METAL CONTENT OF SLUDGES PRODUCED BY GRAND RAPIDS PLANT
Primary
sludge
Excess
activated
sludge
Mixed liquor
Digesting
sludge
Digester
supernatant
Digesting
sludge
Chromium
mg/ liter
510
152
20
386
50
mg/g
9
17
-
11
10
0.2
-
Copper
mg/liter
107
16
4
88
12
mg/g
2
2
-
3
2
Zinc
mg/liter
317
48
7
232
38
mg/g
5
5
-
7
8
Soluble metals
0.3
—
0.2
—
Nickel
mg/liter
125
14
4
97
8
mg/g
2
2
-
3
2
0.8
—
Table 89-B. METALS IN RAW SEWAGE
Metal
Average concentration of
metal entering plant, mg/liter
Chromium
3.6
Copper
1.4
Zinc
1.5
Nickel
2.0
Table 90-A. METAL CONTENT OF SLUDGES PRODUCED BY RICHMOND PLANT
Type
Primary
sludges
Excess
activated
sludge
Mixed liquor
Digesting
sludge
Digester
supernatant
Garbage
Digesting
sludge
Chromium
mg/liter
103
34
12
95
39
-
mg/g
3
3
-
3
-
0.005
Copper
mg/liter
116
22
8
88
40
-
mg/g
3
2.6
-
3
-
0.02
Zinc
mg/liter
94
23
9
73
40
-
mg/g
3
3
-
3
-
0.2
Nickel
mg/liter
8
1.5
0.4
4
5
-
mg/g
0.2
0.1
-
0.2
-
0.008
Soluble metals
0.03
-
0.4
-
0.1
-
0.1
-
Table 90-B. METALS IN RAW SEWAGE
Metal
Average concentration of
metal entering plant, mg/liter
Chromium
0.8
Copper
0.2
Zinc
0.3
Nickel
0.03
Municipal Treatment
177
-------
Table 91-A. METAL CONTENT OF SLUDGES PRODUCED BY ROCKFORD PLANT
Type
Primary
sludge
Secondary
sludge
Trickling
filter
slime
Digesting
sludge
Digester
supernatant
Digesting
sludge
Chromium
mg/liter
271
9
-
358
14
mg/g
5
7
18
8
-
Copper
mg/liter
108
6
-
105
3
mg/g
2
4
13
2
-
Zinc
mg/liter
395
12
-
390
24
mg/g
11
8
17
10
-
Nickel
mg/liter
27
1
-
28
2
mg/g
0.5
0.8
3
0.5
-
Soluble metals
0.8
~
1.0
—
0.7
-
-
Nil
Table 91-B. METALS IN RAW SEWAGE
Metal
Average concentration of
metal entering plant, mg/liter
Chromium
1.8
Copper
1.4
Zinc
2.7
Nickel
0.9
METAL BALANCES
At each of the plants metal balances were performed. The
balances shown in Tables 92 through 95 are for 2-week periods,
except that of Bryan, which is from a special study of a prearranged
slug of chromium to this plant (35). One-hundred and fifty gallons
of a spent plating bath containing 50 pounds of hexavalent chromium
as CrOa was dumped to the municipal sewer and traced through the
Bryan plant.
The balance figures for each of the plants are given in pounds
to indicate the actual quantities of metals that are handled by municipal
plants.
178
INTERACTION OF HEAVY METALS
-------
Table 92. BRYAN, OHIO
METAL BALANCE FOR DAY OF CHROMIUM SLUG
50 POUNDS AS CrO3 DUMPED TO SEWER
Source
Sewage
Primary effluent
Primary sludge (by difference)
Aeration tanks
Waste sludge
Final effluent
% of chromium accounted for
% of chromium retained by plant
(1-day basis)
Chromium,
ib
47
37
10
25
4
10
94
80
Table 93. GRAND RAPIDS, MICHIGAN,
METAL BALANCE FOR 14 DAYS
Source
of
metals
Sewage
Digester
supernatant
Waste activated
sludge
Aeration tanks
Total quantity added
Quantity of metal in 14 days, Ib
Chromium
15,500
250
2,040
-20
17,770
Copper
6,240
60
216
+100
6,516
Zinc
6,540
190
650
-40
7,340
Nickel
8,440
40
194
-5
8,669
Final effluent
Primary sludge
Total in outlets
10,600
4,970
15,570
6,440
1,040
7,480
3,110
3,090
6,200
7,580
1,220
8,800
Outlet
for
metal
Final effluent
Primary sludge
% accounted for
% removal by plant, from
total quantity added
Percent of metal added
Chromium
60
28
88
40
Copper
99
16
115
= 16
Zinc
43
42
85
58
Nickel
87
14
101
12
Municipal Treatment
179
-------
Table 94. RICHMOND, INDIANA,
METAL BALANCE FOR 14 DAYS ON PRIMARY (A)
OVERALL REMOVAL BY PLANT FOR SAME PERIOD (B)
p
A
R
T
A
P
A
R
T
B
Source
of
metals
Sewage
Digester
supernatant
Waste activated
sludge
Total quantity added
Quantity of metals in 14 days, Ib
Chromium
559
287
326
Copper
180
288
218
1,172
686
Balance
Primary effluent
Primary sludge
630
406
Total quantity found
% accounted for
1,036
88
255
460
Zinc
262
274
229
765
Nickel
20
33
15
68
on primary
326
372
24
33
715
104
698
91
57
84
Overall removal by plant
Ib in final
effluent
% removal by plant,
from total quantity
added
210
82
50
73
112
85
15
78
Table 95. ROCKFORD, ILLINOIS,
METAL BALANCE FOR 13 DAYS
Source
of
metals
(In)
Sewage
(Out)
Final effluent
Primary sludge
Total
Outlet
for
metal
Final effluent
Primary sludge
% accounted for
% removal
by plant
Quantity of metals in 13 days, Ib
Chromium
5,837
3,662
2,294
5,956
Chromium
63
39
102
37
Copper
4,502
3,483
927
4,410
Zinc
8,458
3,968
3,360
7,328
Nickel
2,860
2,630
230
2,860
Percent of metal added
Copper
77
21
98
23
Zinc
47
40
87
53
Nickel
92
8
100
8
180
INTERACTION OF HEAVY METALS
-------
The balance for Grand Rapids (Table 93) is very striking when
it is realized that in the 2-week study period approximately 18 tons
of the metals entered the plant.
The balance for Richmond (Table 94) shows very clearly the
amount of metal feedback to the system by the digester supernatant
and waste sludge. A balance on the complete plant could not be made
because of the loss of one of the aerator samples; however, the
balance through the primary gives a satisfactory account of the metals.
The percent removals of the individual metals at Rockford
(Table 95), and Grand Rapids are similar. Richmond shows higher
removals than the other plants; however, the quantity received by
this plant is much less than that by the others. At each of the plants
zinc is most effectively removed and nickel least. This was expected
from pilot plant studies(33). Copper and chromium are less effectively
removed than expected. In general, all the removals are less than
indicated by the pilot studies, with the exception of the Richmond
removals.
AEROBIC EFFICIENCY OF PLANTS
Tables 96 through 99 give the average characteristics of the
various raw sewages and plant effluents; also tabulated are the percent
removals from raw sewage to primary and final effluents. Richmond
and Bryan show excellent overall efficiency. The removals in the
primaries at these plants are not so good as the primary removals
at Grand Rapids and Rockford. This is another reflection of the
higher-than-usual feedback of digester supernatant to the primaries
at Richmond and Bryan.
The lower overall removal efficiency at the Grand Rapids plant
is believed to be due to its inability to maintain a suitable mixed
liquor1 solids because of a limited return-sludge pumping capacity.
This results in a young, nonflocculant sludge and a turbid effluent
Table 96. BRYAN, OHIO
AVERAGE CHARACTERISTICS OF SEWAGE AND EFFLUENTS
FOR 5 DAYS
Analysis
BOD
COD
Sus-
pended
solids
Turbidity
Raw sewage
Avg,
m g/li ter
325
603
164
-
Range,
mg/liter
275-359
481-754
100-216
-
Primary clarifier effluent
Avg,
mg/liter
216
451
141
-
Range,
mg/liter
182-256
391-517
98-166
-
Removed,
%
33
25
14
-
Final clarifier effluent
Avg,
mg/liter
25
90
25
49"
Range,
mg/liter
20-30
85-96
21-30
32-60"
Removed,
92
85
.85
-
1 In stu.
Municipal Treatment
181
-------
with high suspended solids (Table 97). Moreover, the Grand Rapids
plant is the only one of the three activated-sludge plants studied that
has a hydraulic load approaching design capacity.
Table 97. GRAND RAPIDS, MICHIGAN,
AVERAGE CHARACTERISTICS OF SEWAGE AND AFFLUENTS
FOR 14 DAYS
Parameter
BOD
COD
Sus-
pended
solids
Turbidity
Raw sewage
Avg,
mg/ liter
96
314
163
-
Range,
mg/ liter
65-147
276-415
124-244
-
Primary clarifier effluent
Avg,
mg/ liter
61
202
91
-
Range,
mg/liter
45-80
152-303
46-156
-
Removed,
%
36
36
44
-
Final clarifier effluent
Avg,
mg/liter
24
103
62
92"
Range,
mg/liter
19-26
70-125
26-94
71-124"
Removed,
75
67
62
-
In stu.
Table 98. RICHMOND, INDIANA,
AVERAGE CHARACTERISTICS OF SEWAGE AND EFFLUENTS
FOR 14 DAYS
Parameter
BOD
COD
Sus-
pended
solids
Turbiditv
Raw sewage
Avg,
mg/liter
113
258
194
-
Range,
mg/liter
68-179
178-380
124-282
-
Primary effluent
Avg,
mg/liter
95
266
166
-
Range,
mg/liter
66-153
163-374
84-334
-
Removed,
%
16
0
14
-
Final effluent
Avg,
mg/liter
9
33
19
12"
Range,
mg/liter
3-16
19-47
6-28
8-16"
Removed,
92
87
90
-
In stu.
Table 99. ROCKFORD, ILLINOIS
AVERAGE CHARACTERISTICS OF SEWAGE AND EFFLUENTS
FOR 13 DAYS
Parameter
BOD
COD
Sus-
pended
solids
Turbidity
Raw sewage
Avg,
mg/liter
128
370
189
-
Range,
mg/liter
105-166
330-490
118-286
-
Primary clarifier effluent
Avg,
mg/liter
98
293
105
-
Range,
mg/liter
78-126
226-512
60-120
-
Removed,
23
21
44
-
Final clarifier effluent
Avg,
mg/liter
48
153
71
75"
Range,
mg/liter
38-67
121-231
36-104
53-139 '
Removed,
Of
63
59
62
-
In stu.
182
INTERACTION OF HEAVY METALS
GPO 82O—663—13
-------
The high-rate trickling filter plant at Rockford shows the lowest
overall organic treatment efficiency, as would be expected.
The removal efficiencies of all the plants were calculated on
the basis of raw sewage, and the extra loads imposed by digester
supernatant and waste activated sludge were not considered.
These efficiencies are based only on a limited sampling period,
and the efficiencies of the plants on a yearly basis may be signifi-
cantly different than reported here. Previous pilot studies have shown
that the concentrations of metals encountered in these studies would
cause only about a 5 percent reduction in overall efficiency (33).
ANAEROBIC EFFICIENCY OF PLANTS
During this series of field surveys, three of the plants were
encountering varying degrees of operating difficulties with their
anaerobic treatment. As previously mentioned, Bryan had full drying
beds and limited digester volume. One of the two primary digesters
at Richmond was out of operation because of fouling by plastic wrappers
introduced with the garbage feed (36). Owing to this reduced capacity,
sufficient time for formation of good digester supernatant in the
secondary digesters was not available. One of the digesters at Rock-
ford was undergoing modification for conversion to gas mixed operation,
and a second had just begun operation; consequently, part of the gas
produced was vented unmetered. The other three mixed digesters at
Rockford -were producing sludge with satisfactory drainability, but the
high volatile acid content indicates a condition requiring close control.
The digesters at Rockford have been followed for 4 years by this
laboratory, and a volatile acid content of 2,000 milligrams per liter
is characteristic of their behavior.
The digesters at Grand Rapids, which contained the most metals
of all the plants studied (Table 89), were functioning in an excellent
manner, producing a workable sludge with a low volatile content,
good gas production, and a satisfactory relation among volatile acids,
alkalinity, and pH. The Grand Rapids digesters have been followed for
3 years by this laboratory, and the plant chemist has reported on the
composition and use of the gas produced (37).
On the bases of experience with the Grand Rapids digesters and
pilot investigations (33), the concentrations of metals encountered in
these field studies cannot alone be responsible for .any difficulties
with anaerobic digestion of the sludges.
Tables 88 through 91, which give the metal content of the various
sludges at these plants, indicate that even though the digesting sludges
contain several percent metal, not one of them has a soluble metal
Municipal Treatment 183
-------
content above 1 milligram per liter. Table 100 shows the character-
istics of the digesting sludges at the various plants.
Table 100. CHARACTERISTICS OF DIGESTING SLUDGES
Digester
location
Rockford,
Illinois
Grand Rapids,
Michigan
Richmond,
Indiana
Bryan,
Ohio
PH
6.2
7.4
7.0
6.8
Alkalinity
(CaCOa),
mg/liter
6,500
2,500
630
1,800
Volatile
acids
(Acetic),
mg/liter
3,500
445
400
800
°?
volatile
matter
59
51
58
50
Gas,
ft3/lb VS added/
day
Part unmetered
9,5
8.3
Part unmetered
SLUGS OF METALS
At each of the plants, slugs of metals or metal cyanide complexes
were encountered. At Richmond, during the 2-week study, several
yellowish-green slugs were noted and sampled. The grab samples
were then correlated with the amount of chromium received daily
by the plant. Figure 81 shows the pounds of chromium entering the
plant each day. The greater daily quantities correspond to the detected
slugs of chromium. The concentration of chromium in the grab sample
and the time of the slug are also shown. The slugs always occurred
in the evening and after normal working hours. During the first
week, the largest daily quantities entered the plant on the weekend.
Analyses of these grab samples showed that chromium was the major
metallic constituent. Analyses of the 24-hour composite samples for
cyanide indicated that, during this study period, only a small amount
of cyanide was received by the plant.
The receipt of a planned slug of chromic acid to the Bryan plant
is recorded in Figure 73, graphed from the data compiled on the day
of the slug. Analyses of the daily composites for the entire 5 days
of the study, given in Figure 80, showed that in addition to this planned
slug of chromium, a nonplanned copper cyanide complex slug oc-
curred 2 days later. The complex imparted no color to the sewage
and went unnoticed until later laboratory analysis. In Figure 80 the
difference between the 24-hour composite concentrations of chromium
in the primary effluent and the raw sewage was due to the method of
sampling. The 24-hour composite samples were made up of 24
184
INTERACTION OF HEAVY METALS
-------
l?o
no
90
80
70
60
50
40
30
20
10
n
1
-
-
-
_
-
-
1
1
^
1
1
.*
'
1.4 5 P.M.
3> 2.2 6 P.M.
$> 2.4 10 P.M.
-
QJ w _
0- (Jt
<
to —
0
-1 - -P— ^
-t 1. . t i v i - < r i
MONTUE WEDTHURFRI SAT SUNMON TUE WED THU FRI SAT SUN MONTUE WEDTHU
DAY OF WEEK
Figure 81. Daily variation of chromium, Richmond, Ind.
hourly grabs taken on the hour. A sample of raw sewage was taken
while the slug was in progress. Since 95 percent of the chromium
entered the plant in 25 minutes (Figure 73) and the entire slug lasted
only 1 hour, this was the only grab sample of the sewage containing
chromium from the slug. In the case of the primary effluent, in which
the slug lasted for 12 hours, 12 samples that had chromium from the
slug were taken for compositing. This would account for the higher
2 4-hour-composite concentration in the primary effluent.
The same reasoning applies to the copper slug, except that the
sewage sample must have been taken at the peak concentration of the
complex in the sewage, which introduced more copper into the sewage
composite in one grab than several grabs of primary effluent did for
its 24-hour composite.
If both slugs had occurred shortly after the hour and ended
shortly before, the 24-hour raw sewage composites would not have
indicated their presence.
The recorded slugs did not produce any significant effect on the
treatment efficiency of the Bryan plant. This agrees well with the
past experience of the plant superintendent and pilot investigations
(8, 13, 38).
Municipal Treatment
185
-------
Several slugs were caught at the Grand Rapids plant. Since the
color of the sewage was the signal to collect a grab for analysis,
again at Grand Rapids, chromium was the most common slug detected.
Table 101 shows the record of these slugs. One analysis shows, how-
ever, that in addition to the chromium there was a respectable con-
centration of zinc in the September 25 sample. Samples of composites
analyzed for cyanide showed an average concentration of 1 milligram
per liter during this 2-week period.
Table 101. SLUGS OF METALS IN INFLUEN f SEWAGE.
GRAND RAPIDS, MICHIGAN
Date
(1963)
9-20
9-25
9-28
9-28
10-1
Time
10 p.m.
3 a.m.
10 a.m.
11 p.m.
/ a m.
Metal, mg 'li'.er
Chromium
12.6
Copper
1.2
3,2 • 1.2
25.8
3,1
14.6
0,5
1.0
0.6
Zinc
2.0
9.3
1,0
1.6
10
Nickel
2,6
2.1
0.7
1.2
1.7
Rockford has a history of receiving cyanide and metal cyanide
complexes (39). While this study was in progress, a slug of metal
cyanide complexes hit the plant and is recorded in Table 102, Part
A. The officials of the plant had informed the study team that the
plant had been receiving cyanide slugs for several months previous
to the study. The 11:00 a.m. grab sample was taken for routine
analysis. When a fish kill in the Rock River was reported in the
early evening of October 23, downstream from the plant outfall, the
additional grabs and composites listed in the table were analyzed
for cyanide by the plant chemist. Part B of this table shows the
concentrations of cyanide detected in the Rock River at the time of
the fish kill. Judged by the concentrations of metals and cyanide in
the 11:00 a.m. grab and the 8-hour composite, a slug containing a
mixture of copper and zinc, probably as the cyanide complexes, entered
the plant in the late morning period. During this portion of the year,
the Rock River drainage area was in the midst of a dry spell, and the
usual dilution afforded by the river was not available. The slug
caused no significant decrease in the efficiency of the trickling filter
performance for that day, as compared with the average efficiency of
the filter for the 2-week period.
The slug data for Richmond and Grand Rapids point out that even
though the presence of a slug of chromium can be detected by the
yellowish-green color of the sewage, the actual concentration of the
metal may not be very large. Apparently 1.5 to 3 milligrams of
chromium per liter can be noted by eye.
186
INTERACTION OF HEAVY METALS
GPO 820-663-14
-------
Table 102. ROCKFORD, ILLINOIS
SLUG OF METAL COMPLEXES
Part A-Plant samples
Date
(1963) Time
10-22 to
10-23
10-23
10-23
0-23
10-23 to
10-24
24-hr composite
11 a.m. grab
8-hr composite
12-8 p.m.
8 p.m. gra-b
24-hr composite
Location
Sewage
Sewage
Primary
effluent
Sewage
Final
effluent
Metals and cyanide, rag/liter
Chromium
2.2
2.9
2.2
5.8
1.4
Copper
1.7
0.7
7.5
1.5
1.1
Zinc
3.4
7.9
3.4
3.1
1.6
Nickel
0.9
2.2
1.3
1.1
1.4
Cyanide
_
9.8
16.3
3.0
3.8
Part B-Rock River samples
Date
10-23
10-23
10-23
10-24
10-24
8 p.m.
7:15 p.m.
7:45 p.m.
2:30 p.m.
3 p.m.
1.9 miles
upstream
0.3 mile
downstream
2.2 miles
downstream
0.3 mile
downstream
2.2 miles
downstream
Metals and cyanide, mg/liter
Chromium
0.09
0.14
0.24
0.16
0.59
Copper
0.05
0.46
1.19
0.17
0.05
Zinc
0.07
0.14
0.24
0.25
0,01
Nickel
Nil
0.01
0.10
0.04
0.02
Cyanide
Nil
0.3
1.4
Nil
Nil
The slug data for all four plants show that the biological systems
are tolerant to moderate slug conditions; however, the Rockford study
points out the necessity of considering the effects that discharge of
the final effluent will have on the receiving water.
NITRIFICATION
Only one of the four plants, Richmond, produced a nitrified ef-
fluent. The other plants discharged their effluent nitrogen largely as
ammonia. Complete analyses for the various nitrogen forms, or nitrogen
balances, were not carried out. Table 103 clearly indicates that nit-
rification was active only at Richmond. Heavy metals at a concentration
of approximately 5 milligrams per liter have previously been shown to
produce a pronounced inhibition of nitrification (30).
Municipal Treatment
187
-------
Table 103. NITROGEN FORMS IN FINAL EFFLUENTS"
Location
Bryan, Ohio
Grand Rapids ,
Michigan
Richmond,
Indiana
Rockford,
Illinois
Total
Kjeldahl
nitrogen
Nil
18
4
19
Ammonia
nitrogen
20
10
2
11
Nitrate
nitrogen
Nil
Nil
8
Nil
Dissolved
oxygen
final settler
0,5
0.5
1.9
2.6
In mg/liter.
The lack of nitrification at the other three plants cannot be con-
clusively correlated with inhibition of the metals, because of the many
variables between these plants. Of the four plants, judged solely by
observation of operating conditions, Richmond would be the one expected
to nitrify. This is because of the high aerator solids, adequate detention
time, small amount of heavy metals, satisfactory DO aided in part by
cascade flow of mixed liquor through drop pipes down three tiers of
aerators (34) and the warm season of the year at the time of the study
at Richmond.
SUMMARY
A survey of four municipal treatment plants, concerning the
receipt of heavy metals, distribution of the metals in the various
process outlets, and effects of the metals on the treatment efficiency,
has shown satisfactory correlation with pilot-plant investigations.
The results show that the plants receive metallic constituents
on an almost continuous concentration basis. Several slug discharges
of metals to each of the plants were also recorded. At two of the
plants, digester supernatant accounted for a considerable portion
of the metal in the process. The findings indicate that in the range
of 1 to 9 milligrams per liter heavy metals cause no serious reduction
in efficiency of the aerobic or anaerobic treatment of sewage.
In these studies at operating municipal plants many uncon-
trollable variables were encountered, but the pattern of response
of the plants was similar to the 100-gallon-a-day pilot studies.
188
INTERACTION OF HEAVY METALS
-------
REFERENCES
1. Standard Methods for the Examination of Water, Sewage, and
Industrial Wastes. 10th ed. Am. Public Health Assn. New York
1955.
2. Udy, M. J. Chemistry of Chromium and Its Compounds. In: Chro-
mium, Vol. 1. Reinhold Publishing Corp. New York 1956 p
120.
3. Jenkins, S.H., and Hewitt, C.H. The Effect of Chromium Com-
pounds on the Purification of Sewage by the Activated-Sludge
Processes. J. Inst. Sewage Purif. (Midland Branch). 222. 1942.
4. Edwards, G. P., and Nussberger, F. E. The Effect of Chromate
Wastes on the Activated-Sludge Process at the Tallmans Island
Plant. Sewage Works J. 19 (4):598. July 1947.
5. Placak, O. R., Ruchhoft, C. C., and Snapp, R. G. Copper and
Chromate Ions in Sewage Dilutions. Ind. Eng. Chem. 41-2238
1949.
6. Coburn, S. E. Limits for Toxic Wastes in Sewage Treatment
Sewage Works J. 21(3):522. May 1949.
7. Center, A. L. Adsorption and Flocculation as Applied to Sewage
Sludges. Sewage Works J. 6(4):689. July 1934.
8. Moore, W. A., McDermott, G. N., Post, M. A. et al. Effects of
Chromium on the Activated-Sludge Process. JWPCF 33-54
Jan. 1961.
9. Gray, A. G. Modern Electroplating. John Wiley & Sons Inc
New York. 1953.
10. Standard Methods for the Examination of Water and Wastewater.
llth ed. Am. Public Health Assn. New York. 1960.
11. Gameson, A. L. H., Truesdale, G. A., and Van Overdijk, M. J.
Variation in Performance of Twelve Replicate Small-Scale Per-
colating Filters. J. Inst. Sewage Purif. Part 4. 342. 1961.
12. Ludzack, F. J., Schaffer, R. B., and Bloomhuff, R. N., Ex-
perimental Treatment of Organic Cyanides by Conventional
Processes. JWPCF. 33:492. May 1961.
189
-------
13. McDermott, G. N., Moore, W. A., Post, M. A., and Ettinger, M. B.
Effects of Copper on Aerobic Biological Sewage Treatment.
JWPCF. 35:227. Feb. 1963.
14. Ridenour, G. M., Backus, R. D., and Sherron, C. Effect of Poly-
sulfide Treated Cyanide Case Hardening, Copper and Zinc Plating
Wastes on Sludge Digestion. Sewage Works J. 17 (5):966. Sept
1945.
15. Taylor, C. G. Determination of Small Quantities of Nickel with
a - Furildioxime. Analyst. 81:369. June 1956.
16. McDermott, G. N., Earth, E. F., Salotto, B. V., and Ettinger,
M. B. Zinc in Relation to Activated-Sludge and Anaerobic
Digestion Processes. Proc. 17th Ind., Waste Conf., Lafayette,
Ind. May 1-3, 1962. Eng. Ext. Ser. 112. Eng. Bull., Purdue Univ.
47 (2):461 Mar. 1963.
17. McDermott, G. N., Post, M. A. Jackson, B. N. and Ettinger,
M. B. Nickel in Relation to Activated-Sludge and Anaerobic
Digestion Processes. JWPCF. 37:163. Feb. 1965.
18. McDermott, G. N., Moore W. A., Post, M. A., and Ettinger,
M. B. Copper and Anaerobic Sludge Digestion. JWPCF. 35:655.
May 1963.
19. Mancy, K. H., Westgarth, W. C., andOkun, D. A. The Applications
of the Galvanic Cell Oxygen Analyzer to Waste Control Programs.
Proc. 17th Ind. Waste Conf., Lafayette, Ind. May 1-3, 1962. Eng.
Ext. Ser. 112. Eng. Bull., Purdue Univ. 47 (2):508. Mar. 1963.
20. Bozich, T. A. The Toxicity of Metals on the Activated-Sludge
Process. Master's Thesis. Case Inst. Technology. Cleveland,
Ohio. 1959.
21. Masselli, J. W., Masselli, N. W., andBurford, G. The Occurrence
of Copper in Water, Sewage, and Sludge and Its Effects on Sludge
Digestion. New England Interstate Water Pollution Control Com-
mission, 73 Tremont Street, Boston, Mass, June 1961.
22. Dawson, P. S., and Jenkins, S. H. The Oxygen Requirements of
Activated-Sludge Determined by Manometric Methods. Sewage
and Ind. Wastes. 22:490. 1950.
23. Jenkins. S. H. Trade Waste Treatment. J. Inst. Sewage Purif. Part
2. 193. 1957.
24. Tarvin, D. Metal Plating Wastes and Sewage Treatment. Sewage
and Ind. Wastes. 28:1371. 1956.
190 INTERACTION OF HEAVY METALS
-------
25. Stones, T. The Fate of Chromium During the Treatment of Sewage.
J. Inst. Sewage Purif, Part 4. 435. 1955.
26. Stones, T. The Fate of Copper During the Treatment of Sewage.
J. Inst. Sewage Purif. Part 1. 82. 1958.
27. Stones, T. The Fate of Nickel During the Treatment of Sewage.
J. Inst. Sewage'Purif. Part 2. 252. 1959.
28. Stones, T. The Fate of Zinc During the Treatment of Sewage.
J. Inst. Sewage Purif. Part 2. 254. 1959.
29. Pettet, A. Effect of Metal Finishing Wastes on Sewage Purification.
J. Inst. Sewage Purif. Part 1. 36. 1956.
30. Earth, E. F., Sal'otto, B. V., McDermott, G. N., et al. Effects of
a Mixture of Heavy Metals on Sewage Treatment Processes.
Proc. 18th Ind. Waste Conf. Lafayette, Ind. April 30-May 2, 1963.
Eng. Ext. Ser. 115. Eng. Bull., Purdue Univ. 48(3):616. May 1964.
31. Ettinger, M. B. Heavy Metals in Waste-Receiving Systems.
Presented at Interdepartmental Natural Resources Seminar, Ohio
State Univ. Columbus, Ohio, March 1963.
32. Dobbs, R. A., and Williams, R. T. Elimination of Chloride Inter-
ference in the Chemical Oxygen Demand Test. Anal. Chem.
35:1064. July 1963.
33. Earth, E. F., Ettinger, M. B., Salotto, B. V., and McDermott,
G. N. Summary Report on the Effects of Heavy Metals on the
Biological Treatment Processes. JWPCF. 37:86. Jan. 1965.
34. Ross, W. E., and Steeg, H. R. Richmond, Ind., Solves Its Gar-
bage-Sewage Problems. Am. City. 67:132. Sept. 1952.
35. English, J. N., Earth, E. F., Salotto, B. V., and Ettinger, M. B.
A Slug of Chromic Acid Passes Through A Municipal Treat-
ment Plant. Presented at 19th Ann. Purdue Ind. Waste Conf.
Lafayette, Ind. May 5-7, 1964.
36. Wahl, A. J. Larson, C. C., Neighbor, J.B., et al. 1963 Operators'
Forum, JWPCF. 36:401. Apr. 1964.
37. The Round Table. Discussion by Doris Voshel., Grand Rapids,
Mich. Wastes Eng. 34:362. July 1963.
38. Phillips, M. B. Activated-Sludge Response to Excess Chromium
Waste. Presented at 38th Ann. Ohio Water Pollution Control
Conf. Cincinnati, Ohio. 1964.
References 191
-------
39. Carlson, P. R. Cyanide Waste Disposal Survey. Sewage and
Ind. Wastes. 24:1541. Dec. 1952.
40. Salotto B. V., Earth, E.F., Tolliver, W.E., and Ettinger, M.B.
Organic Load and Toxicity of Copper to Activated-Sludge Process.
Presented at 19th Ann. Purdue Ind. Waste Conf. Lafayette, Ind.
May 5-7, 1964.
192 INTERACTION OF HEAVY METALS
-------
SUBJECT INDEX
Acclimation, to
chromium, 13, 26
copper, 31, 45, 48, 120, 141
cyanide, 32, 40, 48, 64, 70, 97, 120
nickel, 83, 91
zinc, 64, 65, 70
Accumulation, of metal
aerator solids, 9, 12, 15, 18, 41, 42, 69, 82, 119, 138, 149
digesting sludge, 8, 52, 93, 110
Acids, volatile, in digesting sludge, 51, 110, 158, 183
Activated-sludge pilot plant
description, 4, 6, 61, 79, 80
operation, 7, 28, 61, 79, 97, 139
Activated-sludge treatment, effects of
chromium, 4
copper, 27, 32, 143
metal mixture, 97, 117
nickel, 79, 83
zinc, 61, 65
Addition of metal solution to pilot plants 5, 80, 98
Alkalinity of digesting sludge, 51, 95, 110, 158, 183
Ammonia
drinking water standard, 129, 132
effect on breakpoint chlorination, 114
in final effluent, 107, 114, 163, 187
Anaerobic digester
municipal plant, 157, 167, 183
pilot plant
description, 7, 8, 49, 62, 80, 97
operation, 7, 8, 49, 62, 81
Anaerobic digestion, relation of metal
chromium, 17, 126
copper, 49, 52, 126
193
-------
cyanide, 52, 64, 76
metal mixture, 97, 110
nickel, 92, 126
zinc, 73, 76, 126
Analytical methods
BOD, 8, 30, 64, 82, 100, 170
chromium, 100
COD, 30, 65, 82, 100, 141
copper, 30, 51, 100, 154, 170
cyanide, 30, 65, 100
nickel, 82, 100, 154
suspended solids, 100
turbidity, 100
zinc, 65, 100, 154
Assay
antibiotic, 94
vitamin, 94
Biological reductor, 17, 26, 137
BOD, of final effluent
chromium study, 11, 159
copper study, 32, 35, 143
metal mixture study, 100, 119
municipal plant, 159, 181
nickel study, 84, 91
zinc study, 65, 72
Bulking, 26, 47, 71, 125
Cadmium, 25
Carbon dioxide, in digester gas, 18, 110
Chloride interference, in
chemical oxygen demand, 30, 65, 82
chromium determination, 9
Chromic acid, 154, 155, 184
Chromium, in
aeration solids, 9, 12, 13, 26, 109, 156, 162, 176
digesting sludge, 9, 17-22, 26, 112, 157, 162, 176
final effluent, 9, 15, 108, 157, 171
primary effluent, 9, 12, 15, 155, 171
primary sludge, 9, 15, 18, 26, 108, 125, 156, 176
secondary sludge, 9, 15, 18, 108, 156, 171, 176
sewage, 9, 15, 155, 160
194 INTERACTION OF HEAVY METALS
-------
COD, of final effluent
chromium study, 11, 159
copper study, 32, 35, 40, 145
metal mixture study, 100, 119
municipal plants, 159, 181
nickel study, 84, 91
zinc study, 65, 72
Color
aeration solids, 17
effluents, 136, 164, 184, 186
sewage, 155, 186
Copper, in
aeration solids, 41-43, 109, 146, 162, 176
digesting sludge, 57, 112, 162, 176
final effluent,. 32, 35, 39, 42, 46, 47, 108, 147, 171
primary effluent, 35, 39, 147, 171
primary sludge, 35, 39, 42, 57, 108, 148, 176
secondary sludge, 35, 39, 42, 57, 108, 148, 171, 176
sewage, 27, 52, 139, 160
Cyanide, 25, 52
acclimation to, 32, 40, 45, 48
copper complex, 27, 28, 35, 45, 47, 52, 97, 120, 160, 185
in river samples, 186
zinc complex, 63-65, 70, 97, 120
Denitrification, 108
Design data
municipal plant
Bryan, Ohio, 151-153
Grand Rapids, Michigan, 167
Richmond, Indiana, 167
Rockford, Illinois, 167
pilot plant, 4, 6
Diffuser tube, pilot plant, 4, 7
Digester failure
chromium, 22, 127
copper, 52-56, 127
zinc, 76, 127
Digester gas
analysis, 100
carbon dioxide, 18
Subject Index 195
-------
Digester supernatant
chromium, 162, 171, 176
copper, 162, 171, 176
nickel, 162, 171, 176
zinc, 162, 171, 176
Dissolved oxygen
pilot plant, 5, 107
municipal plant, 163, 187
Dog food, 8, 62, 80
Dose, response curve
aeration phase, 120, 135
anaerobic phase, 127, 136
Drinking water standards, 129, 132
Electroplating bath
chromium, 25, 154
copper, 27
nickel, 81
zinc, 63
Extended aeration, 134, 146
Filtration, membrane, 9, 52, 83, 99, 109, 141, 155
Fish meal, 80, 139
Fish toxicity
ammonia, 129
cyanide, 129, 186
metal, 129, 186
Frequency distribution, data
copper, 31, 32, 118, 142
metal mixture, 100, 119
nickel, 84
zinc, 66
Garbage, 132, 176
Gas production, effect of
chromium, 18, 20, 157
copper, 52, 56
metal mixture, 110
nickel, 92
zinc, 73
municipal digesters, 157
196 INTERACTION OF HEAVY METALS
-------
Imhoff cone, 112
Loading factors
anaerobic digester, 8, 62, 99
municipal plant, 167
organic loading
activated sludge, 4, 62
anaerobic digester, 8, 51
relation to copper toxicity, 39, 139, 141
trickling filter, 167
pilot activated-sludge plant, 5, 29, 39, 61, 80, 98, 141
Longitudinal mixing
pilot plant, 6
municipal plant, 155, 164
Lick Creek, 151, 157
Metal balances
chromium, 10, 123, 158
copper
cyanide, 39
sulfate, 35, 41, 123, 148
metal mixture, 108
municipal plant, 178
nickel, 88, 123
zinc
cyanide, 70
sulfate, 69, 72, 123
Metal retention, patterns, 135
Microscopic examination, mixed liquor, 40
National Technical Task Committee on Industrial Wastes, 1, 117
Nickel, in
aeration solids, 82, 109, 162, 176
digesting sludge, 92, 112, 162, 176
final effluent, 88, 89, 108, 171
primary effluent, 89, 171
primary sludge, 88, 108, 176
secondary sludge, 88, 108, 171, 176
sewage, 81, 88, 160
Nitrification
chromium, 15
metal mixture, 105, 123
Subject Index 197
-------
municipal plant, 163, 187
oxygen use, 136
Objectives of study, 3, 61, 79, 132
Oil, 25
pH
digester, 18, 110, 158, 183
sewage, 31, 66, 155
Phosphorus, 139
Pilot plant
activated sludge
description, 4, 6, 61, 79, 80
operation, 7, 28, 61, 79, 97, 139
addition of metal solution, 5, 80, 98
anaerobic digester
description, 7, 8, 49, 62, 80, 97
operation, 7, 8, 49, 62, 81
design data
new, 6
old, 4
Plating wastes, source, 25, 31, 63
Polarographic analysis, of
copper, 141, 154, 171
nickel, 154, 171
zinc, 65, 154, 171
Primary treatment, removal of
chromium, 15, 159, 178
copper, 32, 39, 148, 178
metal mixture, 108
nickel, 89, 178
zinc, 70, 178
Probability, data
copper, 31, 38, 142
metal mixture, 100, 119
nickel, 84
zinc, 66
Pumps, pilot plant, 4
Rates, return sludge, 5, 61, 80, 139, 153
Rock River, 186
198 INTERACTION OF HEAVY METALS
-------
Sampling
device, 10, 29, 61, 80, 140
municipal plant, 154, 168
pilot plant, 10, 29, 64, 81, 140
Secondary treatment, removal of
chromium, 15, 159, 178
copper, 32, 39, 148, 178
metal mixture, 108
nickel, 89, 178
zinc, 70, 178
Seeding of digester, 51, 73
Sewage
flow
municipal plant, 151, 176
pilot plant, 8, 62, 80, 98, 140
fortified, 8, 62, 80, 98, 139
source of, 8, 62, 79, 98, 139
Sewage characteristics
Bryan, Ohio, 163
chromium study, 15
copper study, 32, 35, 143
Grand Rapids, Michigan, 181
metal mixture study, 99
nickel study, 83
organic load study, 143
Richmond, Indiana, 181
Rockford, Illinois, 181
zinc study, 66
Sludge, digested
alkalinity, 51, 95, 110, 158, 183
drainability, 183
elutriation, 24
filterability, 23
Sludge density, index, 40, 105, 125, 146
Slug dose, criterion, 122, 138
Slug dose of metal, to
activated sludge
chromium, 13, 123
copper, 41, 123
copper cyanide complex, 45, 160
municipal plant, 13, 151, 184
nickel, 90, 123
zinc, 71, 123
anaerobic digester
Subject Index 199
-------
chromium, 21, 128
copper, 58, 128
nickel, 93, 128
Soluble chromium, in
aeration solids, 12
digesting sludge, 9, 127, 162, 176
final effluent, 9, 109, 126, 171
primary effluent, 9, 109, 117, 155, 171
sewage, 155
Soluble copper, in
digesting sludge, 57, 58, 127, 162, 176
final effluent, 35, 39, 109, 126, 147, 171
primary effluent, 35, 39, 109, 117, 147, 171
sewage, 52
Soluble nickel, .in
digesting sludge, 92, 127, 162, 176
final effluent, 89, 91, 94, 109, 126, 171
primary effluent, 89, 91, 94, 109, 117, 171
sewage, 117
Soluble zinc, in
digesting sludge, 77, 127, 162, 176
final effluent, 71, 72, 109, 126, 171
primary effluent, 71, 73, 109, 117, 171
sewage, 73
Statistical analysis, data
copper, 31, 38
zinc, 67, 69
Sulfide
anaerobic digester, 95, 127, 136
effect on nickel removal, 90
sewage, 90, 133
Suspended solids, of final effluent
chromium study, 11, 159
copper study, 32, 35, 146
metal mixture study, 100, 119
municipal plant, 159, 181
nickel study, 84, 91
zinc study, 65, 72
Temperature, of
aeration phase, 7
anaerobic digestion, 7, 49, 62, 80
Tiffin River, 151
200 INTERACTION OF HEAVY METALS
-------
Trickling filter, effects of heavy metals, 167
Turbidity, of final effluents
chromium study, 12, 15, 159
copper study, 32, 35, 47, 143
metal mixture study, 100, 119
municipal plant, 159, 181
nickel study, 84, 91
zinc study, 65, 70, 72
Urine, 80
Waste treatment, joint municipal and industrial, 1, 3
Zinc, in
aeration solids, 69, 109, 162, 176
digesting sludge, 76, 112, 162, 176
final effluent, 69, 71, 108, 171
primary effluent, 71, 117, 171
primary sludge, 69, 76, 108, 176
secondary sludge, 69, 76, 108, 171, 176
sewage, 69, 160
Subject Index 201
GPO 82O—663—15
-------
BIBLIOGRAPHIC: Robert A. Taft Sanitary Engineering Center.
Interaction of Heavy Metals and Biological Sewage Treatment
Processes. PHS Publ. No. 999-WP-22. 1965. 201 pp.
This volume, a collection of 10 research papers originating
at the Robert A. Taft Sanitary Engineering Center, describes
the effects of chromium, copper, nickel, and zinc on sewage
treatment processes. Results of pilot plant studies and full-
scale municipal plants are given.
For each of the metals and combinations of metals studied,
the effects on the aerobic and anaerobic treatment processes,
under continuous dosage, are given. The data presented allow
a reasonable estimate to be made of the amount of metallic
wat fes that a treatment plant can receive and accomplish
the desired efficiency of treatment. The effects of slug dis-
charges of the metals on the aerobic and anaerobic processes
under pilot plant conditions and at municipal plants are
presented.
The concentrations of the metals in the various sludges
and effluents produced by a treatment plant are given. Metal
balances conducted for each of the studies show the amount of
metal removed by primary and secondary treatment.
ACCESSION NO.
KEY WORDS:
Activated sludge
Anaerobic digestion
Cyanide
Metals
Municipal treatment
Pilot plant
Waste water
BIBLIOGRAPHIC: Robert A. Taft Sanitary Engineering Center.
Interaction of Heavy Metals and Biological Sewage Treatment
Processes. PHS Publ. No. 999-WP-22. 1965. 201 pp.
This volume, a collection of 10 research papers originating
at the Robert A. Taft Sanitary Engineering Center, describes
the effects of chromium, copper, nickel, and zinc on sewage
treatment processes. Results of pilot plant studies and full-
scale municipal plants are given.
For each of the metals and combinations of metals studied,
the effects on the aerobic and anaerobic treatment processes,
under continuous dosage, are given. The data presented allow
a reasonable estimate to be made of the amount of metallic
wastes that a treatment plant can receive and accomplish
the desired efficiency of treatment. The effects of slug dis-
charges of the metals on the aerobic and anaerobic processes
under pilot plant conditions and at municipal plants are
presented.
The concentrations of the metals in the various sludges
and effluents produced by a treatment plant are given. Metal
balances conducted for each of the studies show the amount of
metal removed by primary and secondary treatment.
ACCESSION NO.
KEY WORDS:
Activated sludge
Anaerobic digestion
Cyanide
Metals
Municipal treatment
Pilot plant
Waste water
BIBLIOGRAPHIC: Robert A. Taft Sanitary Engineering Center.
Interaction of Heavy Metals and Biological Sewage Treatment
Processes. PHS Publ. No. 999-WP-22. 1965. 201 pp.
This volume, a collection of 10 research papers originating
at the Robert A. Taft Sanitary Engineering Center, describes
the effects of chromium, copper, nickel, and zinc on sewage
treatment processes. Results of pilot plant studies and full-
scale municipal plants are given.
For each of the metals and combinations of metals studied.
the effects on the aerobic and anaerobic treatment processes,
under continuous dosage, are given. The data presented ^'.'ow
a reasonable estimate to be made of the amount of metallic
wastes that a treatment plant can receive and accomplish
the desired efficiency of treatment. The effects of slug dis-
charges of the metals on the aerobic and anaerobic processes
under pilot plant conditions and at municipal plants are
presented.
The concentrations of the metals in the various sludges
and effluents produced by a treatment plant are given. Metal
balances conducted for each of the studies show the amount of
metal removed by primary and secondary treatment.
ACCESSION NO.
KEY WORDS:
Activated sludge
Anaerobic digestion
Cyanide
Metals
Municipal treatment
Pilot plant
Waste water
------- |