EPA-600/2-77-210
November 1977
Environmental Protection Technology Series
WASTEWATER CHARACTERIZATION AND
PROCESS RELIABILITY FOR
POTABLE WASTEWATER RECLAMATION
Municipal Environmental Research Laboratory
Office of Research and Development
U.S. Environmental Protection Agency
Cincinnati, Ohio 45268
-------
RESEARCH REPORTING SERIES
Research reports of the Office of Research and Development, U.S. Environmental
Protection Agency, have been grouped into nine series. These nine broad cate-
gories were established to facilitate further development and application of en-
vironmental technology. Elimination of traditional grouping was consciously
planned to foster technology transfer and a maximum interface in related fields.
The nine series are:
1. Environmental Health Effects Research
2. Environmental Protection Technology
3. Ecological Research
4. Environmental Monitoring
5. Socioeconomic Environmental Studies
6. Scientific and Technical Assessment Reports (STAR)
7. Interagency Energy-Environment Research and Development
8. "Special" Reports
9. Miscellaneous Reports
This report has been assigned to the ENVIRONMENTAL PROTECTION TECH-
NOLOGY series. This series describes research performed to develop and dem-
onstrate instrumentation, equipment, and methodology to repair or prevent en-
vironmental degradation from point and non-point sources of pollution. This work
provides the new or improved technology required for the control and treatment
of pollution sources to meet environmental quality standards.
This document is available to the public through the National Technical Informa-
tion Service, Springfield, Virginia 22161.
-------
EPA-600/2-77-210
November 1977
WASTEWATER CHARACTERIZATION AND PROCESS
RELIABILITY FOR POTABLE WASTEWATER RECLAMATION
by
Albert C. Petrasek, Jr.
Texas A&M University
for
Dallas Water Utilities
Dallas, Texas 75201
Project No. R-803292-01
Project Officer
John English
Wastewater Research Division
Municipal Environmental Research Laboratory
Cincinnati, Ohio 45268
MUNICIPAL ENVIRONMENTAL RESEARCH LABORATORY
OFFICE OF RESEARCH AND DEVELOPMENT
U.S. ENVIRONMENTAL PROTECTION AGENCY
CINCINNATI, OHIO 45268
-------
DISCLAIMER
This report has been reviewed by the Municipal Environmental Research
Laboratory, U.S. Environmental Protection Agency, and approved for publica-
tion. Approval does not signify that the contents necessarily reflect the
views and policies of the U.S. Environmental Protection Agency, nor does
mention of trade names or commercial products constitute endorsement or
recommendation for use.
-------
FOREWORD
The U.S. Environmental Protection Agency was created because of in-
creasing public and government concern about the dangers of pollution to the
health and welfare of the American people. Noxious air, foul water., and
spoiled land are tragic testimony to the deterioration of our natural envi-
ronment. The complexity of that environment and the interplay between its
components require a concentrated and integrated attack on the problem.
Research and development is that necessary first step in problem solu-
tion and it involves defining the problem, measuring its impact, and
searching for solutions. The Municipal Environmental Research Laboratory
develops new and improved technology and systems for the prevention, treat-
ment, and management of wastewater and solid and hazardous waste pollutant
discharges from municipal and community sources, for the preservation and
treatment of public drinking water supplies and to minimize the adverse
economic, social, health, and aesthetic effects of pollution. This publica-
tion is one of the products of that research; a most vital communications
link between the researcher and the user community.
The study described here was undertaken to evaluate the performance and
reliability of an advanced wastewater treatment system designed to produce
an effluent suitable for reuse for potable purposes. Sound management of
water resources must include consideration of all potential uses of properly
treated municipal wastewater as alternatives for meeting future water de-
mands .
Francis T. Mayo
Director
Municipal Environmental Research
Laboratory
-------
ABSTRACT
This research effort was initiated to quantify water quality criteria
of importance in evaluating the performance of a wastewater treatment facil-
ity producing a product water potentially available for potable reuse. Ad-
ditionally, the reliability of individual unit processes was evaluated and
the effects of process instability on product water-quality were investi-
gated.
The sequence of unit processes utilized in the study to treat municipal
wastewater consisted of screening, degritting, primary clarification, bio-
logical treatment with completely-mixed activated sludge, high-pH lime coag-
ulation, single-stage recarbonation with liquid carbon dioxide, gravity
filtration, and two-stage activated carbon adsorption. Flows through the
pilot plant ranged from 9.6 liters per second (152 gpm) for the activated
sludge influent to 1.1 liters per second (18 gpm) for the product water.
Twenty-four-hour composite samples were collected daily for routine analy-
ses; weekly composite samples were utilized for metals determinations.
The final product water complied with the quality criteria of the
National Interim Primary Drinking Water Regulations in all respects. Sig-
nificant process instabilities had little effect on product water quality
due to the redundant nature of the treatment system employed.
This report was submitted in partial fulfillment of Grant No.
R-803292-01 by the Water Utilities Department, City of Dallas, Texas under
the sponsorship of the U.S. Environmental Protection Agency. This report
covers the period July 1, 1974 to January 31, 1975.
-------
CONTENTS
Foreword iii
Abstract iv
Figures vi
Tables ix
Acknowledgment xi
1. Introduction 1
Background of the Dallas Program 4
Research Objectives 6
2. Conclusions 8
3. Recommendations 10
4. Description of Research Facilities 12
City of Dallas Collection System 12
Demonstration Plant 17
5. Sampling and Analytical Procedures 35
Sampling Procedures 35
Analytical Procedures 35
6. Plant Operation and Performance 41
Completely- Mixed Activated Sludge 44
Upflow Calrifier 57
Single-Stage Recarbonation Basin 61
No. 1 Mixed-Media Filter 66
Activated Carbon Contactors 71
7. Process Reliability 81
Organic Materials 81
Total Suspended Solids and Turbidity 85
Total Phosphorus 85
Nitrogen Compounds 91
Metals 91
Compliance with the National Interim
Primary Drinking Water Regulations 106
-------
FIGURES
Number Page
1 Existing City of Dallas water supply network 2
2 Potable water supplied and wastewater flows for 1975 5
3 Possible sources of influents for the Pilot Plant at
the White Rock Sewage Treatment Plant 19
4 Dallas Water Reclamation Research Center - Demo Plant 21
5 Elevation of the No. 1 aeration basin 24
6 Elevation of the No. 1 final clarifier 26
7 Elevation of the upflow clarifier 28
8 Typical elevation of the No. 1 and No. 2 gravity filters .... 31
9 Elevation of downflow granular activated carbon contactors ... 34
10 Demonstration Plant configuration for the potable reuse
study 42
11 Mixed Liquor Suspended Solids concentrations and seven-day,
moving-average sludge ages for the activated sludge pro-
cess 49
12 Aeration basin D.O. concentration and uptake for the
completely-mixed activated sludge system 50
13 Sludge Volume Index for the activated sludge process 52
14 Effluent TSS and NH3-N concentrations for the completely
mixed activated sludge system 53
15 Effluent TSS as a function of effluent NH3-N for the acti-
vated sludge system 54
16 Effluent total COD and soluble TOC concentrations for the
activated sludge system 55
-------
FIGURES (continued)
Number Page
17 Effluent COD as a function of effluent TSS for the
activated sludge system .................... 56
18 Fecal coliform for the activated sludge effluent ........ 58
19 Upflow clarifier effluent TSS and ferric chloride doses ..... 60
20 Total phosphorus concentrations observed in the upflow
clarifier effluent ...................... 62
21 Total phosphorus as a function of pH .............. 63
22 Upflow clarifier effluent pH values ............... 64
23 Effluent total plate counts and fecal col if onus for the
upflow clarifier ....................... 65
24 Effluent pH values for the single-stage recarbonation
basin ............................. 68
25 Effluent COD values for the mixed-media filter ......... 70
26 Effluent total suspended solids concentrations for the
mixed-media filter ...................... 72
27 Effluent turbidity values for the mixed-media filter ...... 73
28 Media expansion characteristics for the No. 1 mixed-media
filter ............................ 75
29 COD concentrations for the activated sludge effluent and
the final product water .................... 78
30 Product water COD as a function of influent COD to the
activated carbon system .................... 79
31 Turbidity, total phosphorus, and soluble TOC values for
the product water ....................... 80
32 Log-normal frequency distribution for observed COD values. . . . 82
33 Histogram of product water COD concentrations .......... 83
34 Normal probability distribution of COD values observed in
the effluent of the No. 3 carbon column ............ 84
35 Log-normal frequency distribution of observed TOC con-
centrations .......................... 86
vii
-------
FIGURES (continued)
Number Page
36 Log-normal frequency distribution of observed TSS con-
centrations 88
37 Log-normal probability distribution of the observed
turbidity values 89
38 Log-normal frequency distribution of observed total
phosphorus concentrations 90
39 Log-normal frequency distribution of NHg-N concentrations. ... 92
40 Log-normal frequency distributions of N02 & N03-N con-
centrations 93
41 Log-normal frequency distribution of organic nitrogen
concentrations 94
42 Observed variations in arsenic concentrations 96
43 Observed variations in boron concentrations 97
44 Observed variations in barium concentrations 99
45 Observed variations in cadmium concentrations . 100
46 Observed variations in chromium concentrations 101
47 Observed variations in copper concentrations 102
48 Observed variations in iron concentrations 103
49 Observed variations in mercury concentrations 104
50 Observed variations in manganese concentrations 105
51 Observed variations in lead concentrations 107
52 Observed variations in selenium concentrations 108
53 Observed variations in zinc concentrations 109
54 Comparison between the NIPDWR criteria and product
water quality 110
Plate
1 Overall view of the Central Wastewater Treatment Plant
looking south 18
2 Aerial photograph of Demonstration Plant 22
viii
-------
TABLES
Number Page
1 Summary of existing water supply for year 2020 3
2 Sizes and lengths of gravity mains in the sanitary
sewer collection system 13
3 Summary of industrial and commercial discharges to
the collection system 14
4 Influent metals concentrations for fiscal year 1975 15
5 Characteristics of raw wastewater at the Nhite Rock STP 16
6 Technical data for the No. 1 aeration basin 23
7 Technical data for the No. 1 final clarifier 27
3 Technical data for the upflow clarifier 29
9 Media specifications for the No. 1 mixed-media filter 32
10 Proposed schedule of analyses 40
11 Summary of Demonstration Plant performance 43
12 Performance summary of the completely-mixed activated
sludge system 45
13 Hydraulic process control parameters for the completely-mixed
activated sludge system 46
14 Process control parameters for the completely-mixed activated
sludge system 48
15 Water quality summary for the upflow clarifier 59
16 Water quality summary for the single-stage recarbona-
tion basin 67
17 Water quality summary for the No. 1 mixed-media filter 69
IX
-------
TABLES (continued)
Number Page
18 Operations summary for the No. 1 mixed-media filter 74
19 Water quality summary for the granular activated
carbon contactors 77
20 Comparison of mean and median COD and TOC concen-
trations for different process flows 87
21 Mean metals concentrations observed in weekly
composite samples 95
22 Results of biocide determinations on product water 112
-------
ACKNOWLEDGMENTS
This report is the product of cooperative effort by so many people
that individual mention becomes unfeasible. To the entire group of Dallas
Water Utilities personnel and the Texas A&M Research Foundation members we
want to convey our gratitude for their support and cooperation.
In particular we want to offer our thanks to Mr. Henry J. Graeser,
Director, Dallas Hater Utilities, who originated the plan and inspired
the effort; Dr. I. M. Rice, Assistant Director for Engineering and Planning
during the program and since the successor to Mr. Graeser as Director, for
grant preparation and financial planning in addition to general grant
administration; and to Dr. Harold W. Wolf, Texas A&M Professor of Civil
Engineering and Project Director, for his guidance, advice, and constant
effort in executing the program.
XI
-------
SECTION I
INTRODUCTION
As shown in Figure 1, the City of Dallas in north central Texas is
situated in the upper Trinity River Matershed. This region must be class-
ified as naturally water deficient; in fact, there is only one natural lake
in the whole State of Texas. In the upper Trinity River Basin, average
annual precipitation ranges from about 40 inches per year in the eastern
portions to 20 inches per year in the westerly sections. In general, aver-
age annual precipitation and evaporation rates are about equal, and drought
periods in excess of 60 days are not uncommon. These factors combine to
make water a very valuable resource in North Central Texas.
The City of Dallas derives its drinking water supply from the exten-
sive reservoir network shown in Figure 1, and estimates indicate that an
adequate supply of fresh water exists to meet demands anticipated to the
year 1995. The water supply is derived from six reservoirs on three water-
sheds, with estimated safe yields for the year 2020 indicated in Table 1.
During the middle 1950's North Central Texas experienced a protracted
drought. The City of Dallas investigated two alternate sources of water
to augment its dwindling supply. One was to pump water from the highly
mineralized Red River into the existing reservoir system, and the other was
to utilize the waters of the West Fork of the Trinity River which carry a
considerable amount of pollution from the wastewater effluents of the City
of Ft. Worth and of the mid-cities between Ft. Worth and Dallas. In 1955
the thought of "drinking someone else's sewage" did not meet with wide
acceptance from either the general public or the City Council and the deci-
sion was made to import water from the Red River. Consequently, the City
survived the drought by utilizing the Red River over a three-year period --
but not without effect (1). A most important point emerging from this
period is that as early as 1955 a major American city had seriously eval-
uated the possibility of an indirect wastewater reuse as an alternate supply
of drinking water.
With respect to total water resources management in the upper Trinity
River Basin indirect but intentional wastewater reuse is of considerable
importance to the City of Dallas. Both the Bachman and the Elm Fork Water
Purification Plants of the City of Dallas withdraw water from the Elm Fork
of the Trinity River. Estimated water supply for the year 2020 includes
328,573 cu m per day (86.8 MGD) from Garza-Little Elm Reservoir, 37,854 cu m
per day (10.0 MGD) from Grapevine Reservoir, and 156,337 cu m per day (41.3
MGD) in return flows. The return flows are composed exclusively of waste-
water effluents discharged approximately 33 kilometers (20 miles) upstream
from the water plant intake structures. The return flows constitute 30
1
-------
ro
Lewisville
Lake
Grapevine
Lake
SABINE RIVER
BASIN
^JECHES\RIVER
BASIN
TRINITY RIVER
BASIN
I
Figure 1. Existing City of Dallas
supply network.
-------
TABLE 1
SUMMARY OF EXISTING WATER SUPPLY FOR THE YEAR 2020
Reservoir and Basin
Trinity River Basin
Lewisville (Garza-Little Elm)
Grapevine
Ray Hubbard
La von
Return flows
Estimated
MGD
86.8
10.0
55.4
10.0
41.3
Safe Yield
M3/day
328,573
37,854
209,711
37,854
156,337
Sabine River Basin
Tawakoni 162.8 616,263
Neches River Basin
Palestine 102.0 386,111
TOTAL AVAILABLE SUPPLY 468.3 1,772,703
-------
percent of the 522,764 cu m per day (138.1 MGD) that will be used as a source
of drinking water, and represent a valuable water resource. However, travel
time from the point of discharge to the most distant intake structure is less
than one day, allowing little time for significant natural purification to
occur. Under these circumstances, the City of Dallas has considerable inter-
est in water and wastewater treatment technology since it is imperative that
the public health be safeguarded.
Bearing in mind the occasional shortages of rainfall and the conse-
quences of a prolonged drought, and knowing also that new sources of supply
are disappearing rapidly, the City of Dallas has felt for some years that
the building of new reservoirs on streams to augment the potable water supply
would become virtually impossible. The fact that eastern cities in partic-
ular have been able to produce potable, palatable waters from rivers that
have been increasingly polluted has led to consideration that some day the
actual recycle of waste waters, properly treated, may actually be a means of
survival in the semiarid southwest and west.
Dallas has for several years pursued studies of wastewater and water
treatment methods that might lead to the production of potable water com-
pletely acceptable for all uses. Because there are technological, legal,
and esthetic considerations to be satisfied, Dallas has followed a cautious
policy that more than is known now must be learned about constituents of
any water proposed for reuse, the three large areas still relatively unsat-
isfied being viruses, heavy metals, and organics. Any or all of these may
be found to some degree in recycled waters, and enough work must be done to
satisfy all parties that no risk attaches to either health or comfort from
the prolonged use of such waters. Dallas will want to know that the most
stringent standards can be met before proposing recycle.
BACKGROUND OF THE DALLAS PROGRAM
As a result of the previously described circumstances, namely, the
consideration of using the heavily polluted West Fork of the Trinity in 1955,
the present indirect use of upstream effluents, however minor, and the eco-
nomics dictating that maximal amounts of wastewaters be salvaged if practical,
the Dallas Water Utilities have pursued an active, viable wastewater reuse
program since June 1970. Available wastewater flow is shown in Figure 2.
The Demonstration Plant of the Dallas Water Reclamation Research Center,
which is described in detail inasubsequent section of this report, was built
and brought on-line in late July 1969. Equipment check-out and/or modifi-
cation consumed at least nine months, and the facility was in service for
almost a year before the staff considered it to be truly operational. Addi-
tionally, a laboratory and administration building was constructed to pro-
vide laboratory capabilities for the research program and office space for
several different Water Utilities Department activities. The laboratory
facilities were occupied in the Spring of 1971.
The funds for construction of the laboratory and administration build-
ing were provided by Federal Grant No. WPC Tex 588, to the amount of
$571,093 of which the City provided 52 percent. The Demonstration Plant was
-------
en
o so
Drinking Water Demand
Total Wastewater Effluent Flow
5 15 25 5 15 25 5 IS 25 5 15 25 5 15 25 5 15 25 5 15 25 5 15 25 5 15 25 5 15 25 5 15 25 5 IS 25
JAN FEB MAR APR MAY JUN JULY AUG SEPT OCT NOV DEC
1975
Figure 2. Potable water supplied and wastewater flows for 1975.
-------
constructed under Grant No. 17080 EKG from the Environmental Protection
Agency (EPA) for a total cost of $689,015 of which the City of Dallas pro-
vided 45 percent.
From start-up until January 1971 the Dallas Water Reclamation Research
Center operated on additional funds provided in Grant No. 17080 EKG for the
study of various water reclamation unit processes. The initial phase of the
project was directed at evaluating different sequences of unit processes to
expand and up-grade the City's Central Wastewater Treatment Plant.
At the conclusion of these studies additional unit processes were in-
vestigated by utilizing a number of different wastewaters as influents to
the unit processes at the Demonstration Plant. For the most part the inves-
tigations were short term (2 to 4 weeks), and these data have been presented
in the Final Report for the project (2).
From June 1972 through February 1974 the Research Center evaluated the
removal of metals and viruses through three different advanced wastewater
treatment sequences. Concentrations for over twenty different metals were
evaluated in the influent and effluent of each unit process over the pro-
tracted study period. This project was funded through Grant No. S-801026
from EPA for a total cost of $200,287, of which the City shared 41 percent
(3,4).
RESEARCH PRESENTED IN THIS REPORT
The present project, Grant No. S-803292-01, was started in June 1974
and concluded during November 1975. This project was conducted in two
phases, both of which had distinctly different research objectives.
Characterization for Potable Reuse
In the first part of this research effort the selected treatment train
was operated to produce the best possible quality of effluent. This product
water was then examined in the light of present and proposed standards for
its potential reuse as a potable water. The research objectives were:
1. Production of a high-quality product water.
2. Monitoring of the effluent for parameters to indicate
whether potable quality according to available criteria
had been achieved.
3. Evaluating the reliability of the sequence of unit
processes utilized in the study as it relates to
wastewater reuse.
Ultraviolet Disinfection of Municipal Effluents
Phase two of this project was a study of the feasibility of utilizing
ultraviolet light as a disinfection process for municipal wastewater efflu-
ents . The objectives of the ultraviolet disinfection project were as
follows:
-------
1. Evaluate the feasibility of using ultraviolet light as a
disinfection process for municipal wastewater effluents.
2. Determine what, if any, additional treatment is required to
achieve fecal coliform counts less than 200 per 100 ml on a
consistent basis.
3. Evaluate the effect of photo-reactivation on ultraviolet
disinfection by performing light/dark studies.
This research effort was conducted for a total cost of $262,754, of
which the EPA provided 27 percent of the funds and the City of Dallas the
remainder.
Since the two portions of the study had such different objectives, this
report discusses only the research effort conducted during the first phase
of the project. The work performed on ultraviolet disinfection will be
presented in a separate report.
-------
SECTION II
CONCLUSIONS
The sequence of unit processes used in this research effort, i.e.
screening, degritting, primary clarification, biological treatment with
completely-mixed activated sludge, high-pH lime coagulation, single-stage
recarbonation, gravity filtration, and activated carbon adsorption through
two gravity-flow contactors operating in series, produced a consistently
high-quality product water. When evaluated in terms of mean observed con-
centrations, the product water easily complied with the criteria for maximum
contaminant levels (MCL) for potable water promulgated by the U.S. Environ-
mental Protection Agency in the National Interim Primary Drinking Water
Regulations (NIPDWR).
During this study the activated sludge system experienced one major
process upset that resulted in very high COD and TSS concentrations in the
effluent from the biological process. However, the 87 minute empty-bed
contact time in the activated carbon adsorption system was sufficient to
prevent any significant increase in the COD concentrations in the product
water, and for this reason extended empty-bed contact time should be con-
sidered in the design for potable wastewater reuse.
Although turbidity breakthrough did occur on the mixed-media filter as
a possible result of improper backwashing procedure, the downflow carbon
contactors provided important supplementary filtering capacity such that the
turbidity of the product water was not increased significantly. Difficulties
with filter performance should be minimal under normal operating conditions,
but the redundancy gained by using downflow carbon columns is a significant
factor to be considered when designing a facility for potable reuse.
The metals removals observed during this study were quite variable,
ranging from an increase for mercury to an 85.6 percent reduction for arsenic.
Every metal for which there is a standard in the NIPDWR was substantially
reduced in concentration by the activated sludge process, and for this reason
one may conclude that biological processes are important in limiting the
metals concentrations in the product water. Additionally, most of the metals
with well-defined adverse health effects were reduced in concentration by at
least 50 percent. Considering the concentrations and removals of the various
metals observed during this research, the best technique for controlling
metals concentrations in the product water is to limit metals discharges into
the collection system, and then employ an effective treatment sequence.
Operation of the activated sludge system to achieve complete nitrifi-
cation is important for a facility producing an effluent for potable reuse
and employing the same treatment sequence used in this study. Lower soluble
8
-------
COD concentrations will result from operation at the elevated sludge ages
necessary to develop an adequate population of nitrifying microorganisms.
If the activated sludge process is operated to nitrify, and long empty-bed
contact times are provided in the carbon adsorption system, the production
of a product water with a very low COD concentration (0-3 mg/1) is feasible.
Certain water quality parameters that were routinely monitored during
this study have great utility as process control parameters for wastewater
reclamation facilities that are producing a product water intended for pot-
able reuse. Since operation of the activated sludge process in a nitrify-
ing mode produces water with a lower COD, and since chlorination as a disin-
fection process is affected by the presence of ammonia, ammonia nitrogen
and organic nitrogen are extremely valuable process control parameters.
Monitoring and control of pH on the lime coagulation and recarbonation pro-
cesses is essential if process performance and cost effectiveness are to be
maximized. Alkalinity determinations can be used for process control on the
chemical treatment systems, and since alkalinity is destroyed during nitri-
fication it is a valuable process control tool for the biological process.
Control of product water turbidity is important for proper disinfection,
and the turbidity of various process flows is an important indication of how
well the individual unit processes are operating. With the singular excep-
tion of biochemical oxygen demand (BODs) -- because of the five-day period
all of the gross organic parameters have the capability of being important
process control parameters.
During this study no nitrogen removal process per se was incorporated
in the treatment sequence. Nitrate nitrogen concentrations were consistent-
ly below the 10 mg/1 standard promulgated in the NIPDWR, but higher nitrate
nitrogen concentrations can and should be anticipated. Under these circum-
stances nitrogen removal will probably be necessary, although the process
utilized may only be required to treat 25 to 50 percent of the total product
flow.
The product water produced during this study easily complied with the
criteria promulgated in the NIPDWR- however these water quality criteria
were not developed for potable water derived from wastewater and the com-
parison must be viewed with this in mind.
-------
SECTION III
RECOMMENDATIONS
The following recommendations are made as a result of the potable reuse
research program conducted at the Dallas Water Reclamation Research Center,
and the experience gained during operation of the Demonstration Plant.
1. Since the consistent production of a high-quality product water is
essential for potable reuse, every effort should be made to opti-
mize the reliability of the entire treatment sequence and each
individual unit process.
a. Demonstrated reliability of the mixing and oxygen transfer
equipment used in the activated sludge system should be
considered more important than mass transfer efficiency,
since mechanical failure will adversely effect all sub-
sequent unit processes and product water quality will be
impaired.
b. Chemical feed equipment should be sized and duplicated so
that both reserve capacity and standby capacity assure
continuing of operation.
c. All chemical feed lines, and especially the lime feed
lines, must be accessible for routine cleaning and scale
removal.
d. If high pH lime coagulation is employed, the effluent from
the chemical clarifier should flow by gravity to the recar-
bonation basin to avoid scale formation in pumps.
e. Two-stage recarbonation with intermediate clarification
should be used to improve calcium removal and reduce the
hardness of the product water.
f. Down-flow carbon contactors should be used so that the
facility will have a redundant filtering capability.
2. Significant removals of toxic metals were observed during this
project. However, the employment of adequate processes to remove
metals should also be accompanied by a source control program so
that metal discharges into the municipal wastewater system can be
carefully controlled and limited.
10
-------
3. High pH (> 11.2) lime coagulation is a very effective disinfection
process, and for this reason should be considered in the design of
any potable reuse treatment sequence.
4. The activated sludge process should be operated to nitrify such
that the COD loading to the activated carbon adsorption system
will be reduced, and performance of the carbon improved.
5. Additional research should be performed to evaluate the capabilities
of high-pH lime coagulation as a disinfection process. Particular
emphasis should be placed on the destruction of parasitic organisms
and enteric viruses.
11
-------
SECTION IV
DESCRIPTION OF RESEARCH FACILITIES
CITY OF DALLAS COLLECTION SYSTEM
The sanitary sewer collection system serving the City of Dallas consists
of 5074.005 kilometers (3,136.734 mi) of gravity mains. The lengths of mains
of different diameters are summarized in Table 2. During the fiscal year
covered in this report the City abandoned 2.536 kilometers (1.576 mi) of
sanitary sewer main, and added 35.094 kilometers (21.811 mi) of mains to the
collection system. The best available estimates indicate that the total
length of laterals is between one and two times the length of the gravity
mains, indicating a total collection system length between 10,000 and 15,000
kilometers (6215 and 9323 miles). The City has no combined sewers.
In addition to the normal domestic wastes discharged to the collection
system, the City has significant contributions from industrial and commercial
establishments. During fiscal year 1974 the industrial discharges repre-
sented 12.1 percent of the total flow received at the Central Plant, or
21.93xl06 cu.m. (5.794xl09 gal) per year. The 221 significant industries
monitored by the Water Utilities Department discharged a total of 1.70xl07
kg of BOD5 (3.7396xl07lbs) and 1.375xl07 kg (3.0263xl07 Ibs) of total sus-
pended solids TSS to the collection system. When expressed in terms of con-
centration, the BOD5 and TSS of the industrial discharges are 773 mg/1 and
626 mg/1, respectively. The BOD5 discharge represents 37.7 percent of the
total load entering the Central Plant, while the industrial TSS discharges
represent 30.7 percent of the total TSS load.
The activities of commercial establishments, including restaurants,
wholesale food preparation facilities, and service facilities (principally
car washes), have substantial impact on wastewater characteristics. General-
ized information relating to the industrial and commercial discharges are
summarized in Table 3. The predominant effect of commercial activities is
to increase the organic and solids loadings; however, certain of the service
activities (car washing) can have appreciable impact on metals concentrations.
The concentrations of certain metals in the influents of the Dallas and
White Rock plants are given in Table 4. The column headed "combined" is a
calculated, flow-weighted concentration for all wastewaters arriving at the
Central Plant which comprises both the Dallas and White Rock plants. Typi-
cal data for raw wastewater entering the White Rock STP are presented in
Table 5, and upon inspection the wastewater appears to be representative of
domestic wastewaters. Most of the industrial waste discharges enter the
Dallas STP, and for this reason the wastewaters entering the White Rock
plant are more suitable for wastewater reuse studies.
12
-------
TABLE 2
SIZES AND LENGTHS OF GRAVITY MAINS IN THE SANITARY SEWER COLLECTION SYSTEM
PIPE
(cm)
10.16
15.24
20.32
25.40
30.48
35.56
38.10
40.64
45.72
53.34
60.69
68.58
76.20
83.82
91.44
99.06
106.68
114.30
121.92
129.54
137.16
152.40
160.02
167.64
182.88
198.12
205.74
213.36
228.60
Total
DIAMETER
(in)
4
6
8
10
12
14
15
16
18
21
24
27
30
33
36
39
42
45
48
51
54
60
63
66
72
78
81
84
90
length of mains
(km)
2.513
1814.508
1934.113
450.758
244.624
0.759
176.397
0.290
123.037
59.057
47.435
21.334
40.954
8.367
22.632
7.435
12.790
7.416
16.939
0.454
10.095
26.182
0.732
3.560
8.333
1.936
0.714
1.263
2.248
5,047.005
LENGTH
(mi)
1.562
1127.724
1202.059
280.148
152.035
0.472
109.658
0.180
76.468
36.704
29.481
13.259
25.453
5.200
14.066
4.621
7.949
4.609
10.528
0.282
6.274
16.272
0.455
2.236
5.179
1.203
0.444
0.785
1.397
3,136.734
13
-------
TABLE 3
SUMMARY OF INDUSTRIAL AND COMMERCIAL DISCHARGES TO THE COLLECTION SYSTEM
T f
Business
Industrial
Commerical
Food Preparation
oerv i ce
Facilities
Total Commercial
Nnmhov r\f _ P"I ni.i ___ _ ROD,- T'X'N £VQ;JCO
Establishments (cu.m./mo.) (gal. /mo.) (mg/1) (mg/1) (mg/1)
221 1.824xl06 4.82xl09 773 626
9fin -. -- QD*^ fi7R 1 R^
onn w_ 9f\R 1 H7Q 90?
2200 3.9547xl05 l.OSxlO8 --
-------
TABLE 4
INFLUENT METALS CONCENTRATIONS FOR FISCAL YEAR 1975
Metal
Arsenic
Ban' urn
Boron
Cadmi urn
Chromium
Copper
Lead
Manganese
Mercury
Nickel
Selenium
Silver
Zinc
Q
Dallas STP
0.00474
0.660
0.507
0.035
0.345
0.234
0.443
0.096
0.00159
0.210
0.0170
0.580
oncentration (mg/1)
White Rock STP
0.02003
0.700
0.387
0.016
0.114
0.142
0.193
0.080
0.00125
0.088
0.0155
0.227
Combined
0.0164
0.691
0.416
0.0205
0.169
0.164
0.252
0.084
0.001331
0.117
0.0159
0.311
15
-------
TABLE 5
CHARACTERISTICS OF RAW MASTEWATER AT THE WHITE ROCK STP FOR
THE PERIOD OF OCTOBER 1, 1974 THROUGH SEPTEMBER 30, 1975
Flow = 408,823 cu.m./day
(108.0 MGD)
Grit = 0.0127 liters/cu.m.
(1.69 cu.ft./MG
Total Solids = 844 mg/1
Settleable Solids = 8.4 mg/1
TSS = 209 mg/1
COD = 417 mg/1
BOD5 = 191 mg/1
NH3-N = 16.4 mg/1
Org. N. = 13.5 mg/1
N02 & N03-N = 0.3 mg/1
pH = 7.2
16
-------
DEMONSTRATION PLANT
The Demonstration Plant of the Dallas Water Reclamation Research Center
is colocated with the White Rock Sewage Treatment Plant (STP) at the City of
Dallas' Central Wastewater Treatment Facility. The Central Plant complex is
situated on the south bank of the Trinity River approximately five kilometers
(3 miles) south of the City's central business district. The Central Plant
actually consists of the three treatment facilities described below, two of
which are trickling filter plants. The third facility is completely-mixed
activated sludge followed by tertiary mixed-media filtration.
Dallas Sewage Treatment Plant
The Dallas STP is the oldest wastewater treatment facility operated by
the City of Dallas. This single-stage, standard-rate, trickling filter facil-
ity consists of four bar screens and grit channels, twenty-four Imhoff tanks
which are operated as primary clarifiers, two rectangular primary clarifiers,
sixteen standard-rate trickling filters which are 53 meters (174 feet) in
diameter, and three final clarifiers.
White Rock Sewage Treatment Plant
The White Rock STP is a two-stage, high-rate trickling filter facility
without intermediate clarification. The plant consists of two bar screens
and grit channels, six rectangular primary clarifiers, four first-stage,
high-rate trickling filters, eight second-stage, high-rate trickling filters,
and four rectangular final clarifiers. All trickling filters are 53 meters
(174 feet) in diameter, and contain a maximum of 2.286 meters (7.5 feet) of
media.
Tertiary Treatment Complex
Under normal flow conditions the effluents from both the Dallas and
White Rock facilities will be discharged to the tertiary treatment complex
prior to discharge into the Trinity River. The tertiary complex consists
of twelve completely-mixed, activated sludge aeration basins and twelve
final clarifiers, followed by fourteen mixed-media gravity filters. This
facility is scheduled to come on-line during the fall of 1976, and pilot
plant research indicates that considerable improvement in water quality in
the Trinity River can be anticipated.
The physical relationship between the individual facilities present at
the Central Plant complex can be seen in Plate 1.
Demonstration Plant
All influents to the Demonstration Plant are pumped from the White Rock
STP and all effluents and sludges from the pilot plant are returned to the
headworks of the White Rock plant. As indicated in Figure 3, there are a
total of five possible influents which can be supplied at a maximum flow of
47.32 liter/sec (750 gpm), with the exception of the raw sewage pump that
is rated at 18.93 liter/sec (300 gpm). The discharges from all pumps are
17
-------
IIIIIIMIIIIMIJIIIIMIIIIIIIIIIIIIMIIIIIIIIIliin Illlllllllllllll
_ ji. *.. HWT^Mlli i ii 'lian i ' - * - .,-^~!- t
v '
=
TE
Plate 1. Overall view of the Central Hastewater Treatment Plant looking south.
-------
ILOT PLANT INFLUENT SOURCES
/T\ /T\ /O^
SCREENSCHANNELS PRIMARY
/// I CLARIFIERS I
/// « a
I STAGE I i --
TRICKLING | TRICKLING
| STAGE II
i TRICKLING
I FILTERS
1
FINAL
CLARIFIERS
\
ALL
PILOT PLANT
EFFLUENTS
AND
SLUDGES
Figure 3. Possible sources of influents for the Pilot Plant at the White Rock Sewaqe Treatment
Plant.
-------
routed to a valve station at the White Rock plant, from which the flow is
directed to the pilot plant through one of three main influent lines. Each
influent line services one of the major treatment modules (biological, chem-
ical, or physical) at the Demonstration Plant.
As indicated in Figure 4, which is a piping diagram of the major compo-
nents of the Demonstration Plant, the facility is relatively complex and very
flexible. The following unit processes are present at the pilot plant. They
are discussed in detail in the following sections of this report, and many
are identified in Plate 2 (aerial photograph of the Demonstration Plant.)
1. No. 1 completely-mixed activated sludge system
2. No. 2 completely-mixed activated sludge system
3. Upflow clarifier
4. Gravity-flow, mixed-media filter
5. Gravity-flow, dual-media filter
6. Gravity-flow activated carbon contactors (2 each)
7. Chlorine contact basins (2 each)
8. Ozone generator and contacting system
9. Reverse osmosis demoralization unit
10. Ultraviolet light disinfection units (2 each)
11. Chemical storage and feeding equipment
No. 1 Activated Sludge System --
The No. 1 activated sludge system consists of the No. 1 aeration basin
and the No. 1 final clarifier, and return sludge and effluent pumps. The
return sludge pump has a practical operating range of 6.3 to 47.3 liters/sec.
(100 to 750 gpm), while the effluent pump has an operational range of 3.2 to
20.5 liters/sec. (50 to 325 gpm).
No. 1 Aeration Basin --
The No. 1 aeration basin is a circular mild-steel tank erected above
ground; technical data for this unit are summarized in Table 6. Figure 5
indicates the major structural features of interest.
Several different types of mixing and oxygen transfer equipment have
been evaluated in this basin. Figure 5 shows the original equipment supplied
by Dorr-Oliver, which consisted of an 18.6 kw (25 HP) surface/submerged
turbine variable-speed mixer and an 11.2 kw (15 HP) variable-speed air com-
pressor. This system was removed and replaced with Fiscal in equipment.
20
-------
ro
Figure 4. Dallas Water Reclamation Research Center - Demo Plant
-------
ro
no
Plate 2. Aerial photograph
of Demonstration Plant.
-------
TABLE 6
TECHNICAL DATA FOR THE NO. 1 AERATION BASIN
Diameter
Depth
Volume
Theoretical Residence Time at
6.3 liters/sec (100 gpm)
Total Nameplate Power Input
Maximum Power Level
Air Flow
7.62 m
(25.0 ft.)
3.66 m
(12.0 ft.)
170,343 liters
(45,000 gal.)
560.2 m3
(6,030 ft.3)
7.5 hours
29.8 kw
(40 HP)
3
0.053 kw/m
(0.89 HP per 1000 gal)
(6.64 HP per 1000 ft.3)
118 or 236 liters/sec
(250 or 500 scfm)
23
-------
ro
O. a
ffeducer Of
Qt////
f/ex/6/e
Coujo/r'no
yff/'y/d F/ff. Coi//a/fny
_
. 70/7
Wafer Leve/
408.6O
Figure 5. Elevation of the No. 1 aeration basin.
-------
Aquarius, Inc. is the supplier of the Fiscalin system, which was oper-
ated throughout most of the reuse portion of this grant. During the UVphase
of this research effort the Aquarius equipment was replaced by a "jet aera-
tion" system manufactured by Penberthy-Houdaille.
No. 1 Final Clarifier
The No. 1 final clarifier is a circular, mild-steel basin erected above
ground. As initially provided by Rex Chainbelt, the unit had peripheral
feed and square effluent weirs in the center of the tank, as shown in Figure
6. The unit was not originally equipped with a surface skimmer, and the
center effluent weir configuration made addition of a skimmer quite complex.
Therefore, the basin was modified by removing the center effluent weirs and
bolting a new peripheral effluent weir to the inside of the existing influent
baffle skirt. A skimmer and scum collector were then fitted to the basin.
Sludge is removed by the head differential between the water surface in
the clarifier and the return sludge pump well, via a single armed header.
Technical data for this clarifier are presented in Table 7.
No. 2 Activated Sludge System
The No. 2 activated sludge system consists of the 28.4 m3 (7,500 gal)
completely-mixed aeration basin, and a three-hopper Smith and Lovelace final
clarifier. The aeration basin has a diameter of 3.14 m (10.3 ft.) and a
side water depth of 3.66 m (12.0 ft.). Mixing and oxygen transfer is effect-
ed by diffusers, and the maximum air flow is 53.8 liters/sec. (114 scfm).
Upflow Clarifier
An Infilco Densator, shown in Figure 7, is the chemical treatment unit
used at the Demonstration Plant. The unit consists of the main tank, 5.5 m
in diameter and 5.48 m deep, and an inner cylinder that serves as the rapid
mixing and flocculation zones. Influent enters the top of the inner cylinder
and the main tank serves as the upflow clarification compartment.
Energy input for mixing and flocculation is supplied via independent
turbine-type mixers, each of which is equipped with a U.S. Electric Vari-
drive that has a 10 to 1 turndown capability.
A 2.54 cm x 10.16 cm (1 in. x 4 in.) steel fluidizer bar is used to
prevent the sludge from over compacting. Table 8 summarizes the more signif-
icant technical information on the upflow clarifier.
Chemical Storage and Feed Equipment
Facilities are present at the Demonstration Plant to store and feed the
following chemicals.
1. Hydrated lime
2. Hydrated aluminum sulfate
25
-------
0-O. d/u/n. Tub/ng Handra/7-
PO
en
£/. 4O9.5O
Top Of Tank~J-L-
£/. 395.50
(/ns/de Tank)
Trough
$ Supports
Mox. Water Surface
\E/. 4O7.50
X
Wins/m/h
8MCYTW Oufpv/ Speed
.058R.PM. - Gen'/, f/ec
'/tH.f>@ /7SO/3PH/COcy.
93/}/jf/?t*l I/ ffis* 7~- /»_*'
Spaced Orr/ces
Figure 6. Elevation of the No. 1 final clarifier.
-------
TABLE 7
TECHNICAL DATA FOR THE NO. 1 FINAL CLARIFIER
Diameter
Depth
Volume
Surface Area
Weir Length
Surface Overflow Rate at
6.3 liters/sec (100 gpm)
Weir Loading at
6.3 liters/sec (100 gpm)
9.14 m
(30.0 ft.)
3.66 m
(12.0 ft.)
240,373 liters
(63,500 gal.)
240.4 m3
(8,480 ft.3)
65.68 m2
(707 ft.2)
27.13 m
(89 ft.)
8.50 m3/m2/day
(203.7 gpd/ft2)
20.09 m3/m/day
(1618 gpd/ft.)
27
-------
Figure 7. Elevation of the upflow clarifier.
28
-------
TABLE 8
TECHNICAL DATA FOR THE UPFLOW CLARIFIER
Rapid Mixing Zone
Diameter
Depth
Surface Area
Volume
Residence time at
6.31 liters/sec (100 gpm)
Mixing
2.59 meters (8.5 ft.)
0.64 meters (2.1 ft.)
5.26 sq.m. (56.6 sq.ft.)
3.35 cu.m. (886 gal.)
8.86 nrin.
3 to 30 rpm
Flocculation Zone
Diameter
Depth
Volume
Residence time at
6.31 liters/sec (100 gpm)
Mixing
2.59 meters (8.5 ft.)
3.29 meters (10.8 ft.)
17.3 cu.m. (4,570 gal.)
45.7 min.
1 to 10 rpm
Upflow Clarification Zone
Outside diameter
Inside diameter
Volume
Surface Area
Overflow rate at
6.31 liters/sec (100 gpm)
Weir loading at
6.31 liters/sec
Residence time at
6.31 liters/sec
5.54 meters (18.167 ft.)
2.59 meters (8.5 ft.)
96.8 cu.m. (25,580 gal.)
18.71 sq.m. (201.4 sq.ft.)
29.9 m3/m2/day
(717 gpd/sq.ft.)
33.5 m3/m/day (2,700 gpd/ft)
255.8 min.
Total theoretical residence time
at 6,31 liters/sec (100 gpm)
5.17 hours
29
-------
3. Ferric chloride
4. Dry polyelectrolytes
5. Liquid polyelectrolytes
6. Activated silica
7. Powdered activated carbon
8. Chlorine
All coagulants and coagulant aids, with the exception of lime, can be
fed to either aeration basin or final clarifier, to the upflow clarifier,
or in front of the filters or carbon contactors for use as filtration aids.
The lime slurry can be pumped to either of the activated sludge systems or
the upflow clarifier.
Ho. 1 Mixed-Media Filter --
The No. 1 filter is shown in Figure 8 as it was initially installed at
the Demonstration Plant. At that time the filtering media consisted of 0.91
meters (3.0 feet) of sand overlayed with 0.30 meters (12 inches) of anthra-
cite. The influent flow was split equally between the top and bottom of
the filter bed, and the effluent was withdrawn through a mid-bed collector
located 15 cm (6 inches) below the sand-anthracite interface. The filter
performance can be characterized as having been generally good; however,
structural deficiencies with the mid-bed collector resulted in frequent
maintenance and the unit was converted to a conventional gravity flow filter.
When the filter was rebuilt media supplied by Neptune Microfloc was
utilized and the media specifications are summarized in Table 9. The filter
has a nominal diameter of 1.2192 meters (4.0 feet) and a surface area of
1.17 sq.m. (12.57 sq.ft.). At a flow of 2.37 liter/sec. (37.5 gpm) the
filtration rate is 174.7 cu.m. per sq.m. per day (3 gpm/sq.ft.). Filter
backwashing is conventional and utilizes a surface wash in lieu of an air
scrub.
No. 2 Dual-Media Filter --
Structually, the No. 2 filter is almost identical to the No. 1 filter,
and this unit is also operated in the conventional gravity-flow mode. Media
consist of 30.48 cm (12 inches) of sand with a 60.96 cm (24 inches) anthra-
cite cap. The filter sand has an effective size of 0.57 mm and a uniform-
ity coefficient of 1.6. Air scour is normally used prior to backwash at a
rate of 20.2 liters per sec. per sq.m. (4 scfm per sq.ft.).
Activated Carbon Contactors
Both the No. 3 and the No. 4 columns at the Demonstration Plant serve
as granular activated carbon contactors, and construction details are shown
in Figure 9. Both units are 1.22 meters (4.0 feet) in diameter and use a
30
-------
Rim L l'/eHxlW, /<"
Backwash Drain
) Connections
To Have \
Reinforcing \
Plates Per \^
/ API Standards J
Figure 8. Typical elevation of the No. 1 and No. 2 gravity filters.
31
-------
co
ro
TABLE 9
MEDIA SPECIFICATIONS FOR THE NO. 1 MIXED-MEDIA FILTER
Material
Gravel
Garnet
Coarse, MS-11
Fine, MS- 7
Sand, MS-18
Anthracite, MS-19
Size
1.905 to 1.27 cm
(3/4 to 1/2 in.)
1.27 to 0.635 cm
(1/2 to 1/4 in.)
<0.635 cm
(
-------
3.048 meter (10.0 foot) charge of carbon, or 1998 kg (4396 Ibs.) at a bulk
density of 0.56 kg per liter (35 Ibs. per cu. ft.). The carbon contactors
use the same air scrubbing system for backwashing as the No. 2 filter.
Calgon Filtrasorb 400, 8x30 mesh, was used throughout the study.
Chlorine Contact Basins
The Demonstration Plant has two chlorine contact basins which may be
operated in parallel or in series. Each basin is 5.49 meters (18.0 feet)
long, 2.26 meters (7.41 feet) wide, and 0.48 meters (1.58 feet)deep. Each
basin has a volume of 5.95 cu.m. (1573 gal.), which results in a theoretical
residence time of 15.7 minutes at a flow of 6.31 liters per sec. (100 gpm).
Eleven fiberglass baffles were installed in each basin, such that plug flow
would be closely approximated. Dye studies have been used to quantify the
hydraulic characteristics of the end-around baffling system, and observed
residence time distribution functions closely approximate theoretical values.
The chlorination equipment is capable of a maximum feed of 22.7 kg (50 Ibs.)
of chlorine per day.
33
-------
Top Of Filter
Elev. 412.33
-6" Backwash Drain
EUv. 407.67
-Manhole El. 403.85
18'Lcnape Type Qol
-Manhole El. 391.08
? 16" Double Crab Type
Rtinforcinq Plate
Tvp. For All Pipe Conn
3/16
3'Overflow El. 411,45
Top of Wash
Trough El. 408.17
4" Influent El.403.67
GO) Loss of Head
Press Taps
I' St. Sfl. Cplq
w/ pluqs
Top of Filter Media
El. 401.83 (Typ.)
Th«»« Taps To Be
Connected To Loss of
H«od Metennq
Others Move St. St'l.
Pipe Pluqs.
60 Mesh Screen 18-8
St.ifl. Welded Over
Holes On Inside Of
Tank (Typ.)
Kudo Nozzles
Appro*. 6" on Ctrs.
Top Of L)nd«rdrom
El. 59I.6S
-3" Air Wash
El. 390.75
Effluent
And Air Wash
£' Backwash Dram
Elev. 407.67
r ,
Anchor Bolts (By Others)
24
(By Others)
" Grout
(By Others)
Figure 9. Elevation of downflow granular activated carbon contactors.
34
-------
SECTION V
SAMPLING AND ANALYTICAL PROCEDURES
SAMPLING PROCEDURES
The sampling procedures described below were utilized for the duration
of this research effort. Samples for routine wet chemistry and metals ana-
lyses were collected by the operators on duty at the Demonstration Plant,
while samples for microbiological analyses were collected by the staff micro-
biologist or microbiology laboratory assistants.
Routine Chemistry Samples
Samples for routine wet chemistry analyses were collected by the plant
operators seven days a week at 1 AM, 5 AM, 9 AM, 1 PM, 5 PM, and 9 PM. Wide-
mouth, half-pint plastic bottles were used for sample collection. These
sample bottles were placed in a refrigerator until transported to the labo-
ratory, at which time they were composited by the staff chemists. Since the
Demonstration Plant was operated at hydraulic steady-state, equal volumes
(400 ml) of each of the six grab samples were used for the 24-hour composite
sample.
Metals Samples
Samples for metals determinations were collected by the plant operators
at the same time samples for routine analyses were collected. Four hundred
milliliter fractions were composited in one-gallon amber bottles to which
redistilled nitric acid (10 ml per liter) had been previously added for
sample preservation.
Microbiological Samples
Either the staff microbiologist or the microbiology laboratory techni-
cians collected all samples for microbiological evaluation. The samples were
collected in 125 ml, wide-mouth glass bottles with glass stoppers that had
been previously dry sterilized at 350°F for one hour.
ANALYTICAL PROCEDURES
The analytical procedure used in this research effort followed the 13th
Edition of Standard Methods for the Examination of Hater and wastewater in
so far as practicable (5).
35
-------
Flow
The influent flows to all unit processes, with the exception of the_re-
carbonation basin, were measured by BIF/Brooks magnetic flow meters. This
combination of meters and recorders proved to be unreliable, and the City's
instrumentation technicians lacked the expertise necessary to keep the flow
meters properly calibrated and maintained. These problems resulted in the
installation of several physical flow measuring elements (orifices, weirs,
and venturi sections) such that accurate flow measurements could be obtained.
Biochemical Oxygen Demand
The procedure used for determining the five-day biochemical oxygen
demand was as given in Section 219 of Standard Methods, and the dilution
technique described in paragraph 4 (ii) was utilized.
Chemical Oxygen Demand
The following procedures were utilized to determine COD values on the
routine samples. The low level technique was employed for those samples
where the COD was expected to be less than 50 mg/1.
High Level Technique
The procedure used was as described in Section 220 of Standard Methods.
Low Level Technique -
The COD of low level samples was determined by using the procedure given
on page 19 of Methods for Chemical Analyses of_ Hater and Wastes 1971 (6).
Two modifications were made to the procedure. The amount of mercuric sulfate
was reduced from 1.0 to 0.4 grams, and the ferrous ammonium sulfate solution
was 0.01 N instead of 0.025 N.
Total Organic Carbon
All total organic carbon determinations were made using a Beckman Model
915 Total Organic Carbon Analyzer.
Total Residue
in
Total solids determinations were made in accordance with the procedure
Section 224A of Standard Methods.
Nonfiltrable Residue
ure
filters
Total suspended solids determinations were made by employing the proced-
in Section 224C of Standard Methods and using 2.4 cm diameter glass-fiber
and Gooch crucibles,
36
-------
Total Dissolved Solids
Total dissolved solids were computed by subtracting the nonfiltrable
residue from the total residue.
Total Phosphorus
The single reagent method given in Methods for Chemical Analyses of
Hater and Hastes 1971 was used for all total phosphorus determinations. The
amount of ammonium persulfate used was increased from 0.4 to 0.5 grams, and
the amount of combined reagent was increased from 8 ml to 10 ml.
Ammonia Nitrogen
Ammonia nitrogen determinations were made by using an ion-specific elect-
rode and the Known Addition Method (7). The electrode used was an Orion
Model 95-10.
Total Kjeldahl and Organic Nitrogen
Total Kjeldahl nitrogen was determined by using an ion-specific elect-
rode and the Known Addition Method after completing the digestion phase of
the procedure given in Section 216 of Standard Methods. Organic nitrogen
was determined by subtracting the ammonia nitrogen from the total Kjeldahl
nitrogen.
Nitrite Nitrogen
Nitrite nitrogen determinations were made using the procedure described
in Section 134 of Standard Methods.
Nitrate Nitrogen
The phenoldisulfonic acid method, Section 213D of Standard Methods, was
used to determine combined nitrite-nitrate nitrogen. Nitrate nitrogen was
computed by subtraction of the nitrite nitrogen.
Sulfate
Sulfate was determined by an indirect atomic absorption spectroscopy
method by adding a known concentration of barium chloride to form a barium
sulfate precipitate. The barium concentration in solution was then deter-
mined by atomic absorption, and the sulfate concentration determined by
subtraction.
Chloride
Chloride concentrations were determined by the mercuric nitrate method
described in Section 112B of Standard Methods.
37
-------
Silica
Silica determinations were made using the molybdate method presented in
Methods for Chemical Analyses of_ Hater and Wastes 1971.
Alkalinity
Total and phenolphthalein alkalinity was determined by using the pro-
cedures given in Section 102 of Standard Methods using methyl orange.
Turbidity
Turbidity was determined by the nephelometric method described in Sec-
tion 163A of Standard Methods with a Hach Model 2100A Turbidimeter. The
standard references were formazin polymer suspensions.
Color
Color determinations were made by plant operators using a Hellige Aqua
Tester and platinum-cobalt color disk.
Metals Determinations
Samples for metals analyses were filtered through a glass fiber filter
and then concentrated by a factor of ten. Concentration was accomplished by
heating (below the boiling point) a 500 ml sample until the volume was re-
duced to less than 50 ml, and then making up to volume in a 50 ml volumetric
flask.
Atomic Absorption -
Atomic absorption spectroscopy was utilized to determine the concentra-
tions of aluminum, barium cadmium, calcium,cobalt, copper, chromium, iron,
lead, magnesium, manganese, silver, strontium, and zinc. A Perkin-Elmer
Model 403 was used for these analyses, and standard procedures given in the
Perkin-Elmer Operator's Manual (8) and Standard Methods were employed.
Flame Emission -
Sodium and potassium concentrations were determined by flame emission
spectrophotometry by operating the P-E 403 in that mode, and using methods
given in Standard Methods.
Arsenic -
Arsenic concentrations were determined by using the silver diethyldithio-
carbamate method presented in Section 104A of Standard Methods.
Boron -
The curcumin method given in Section 107A of Standard Methods was used
to determine boron concentrations, with an ion-exchange modification to
38
-------
remove cationic interferences.
Beryl 1i urn -
The morin fluorometric method (9) was employed for beryllium determi-
nations.
Mercury -
Mercury concentrations were determined by the fTameless atomic absorp-
tion method with a Perkin-Elmer Model 209B atomic absorption spectrophoto-
meter.
Molybdenum -
The dithiol method of Brown, et al.(10) was used to determine molybdenum
concentrations.
Selenium -
Selenium concentrations were determined by employing the diaminobenzi-
dine method given in Section 150A of Standard Methods.
Silica -
Silica determinations were made by using the heteropoly blue method
given in Section 151C of Standard Methods.
Vanadium -
The catalytic oxidation method presented by Brown, et al. (U.S. Geolog-
ical Survey) was used to determine vanadium concentrations (10).
Sampling Frequency -
The sampling frequency used during the project period is shown in Table
1,0. Most of the more conventional water quality parameters that had process
control significance were evaluated daily and on 24-hour composite samples.
Those parameters necessary for general background information, such as chlo-
rides and sulfates, were evaluated on weekly composite samples. Due to time
and cost limitations, trace element determinations were made on weekly com-
posite samples.
39
-------
TABLE 10
PROPOSED SCHEDULE OF ANALYSES
Raw
Parameter Sewage
Chlorides
Sul fates
Alkalinity
Dissolved Solids
Nitrates
Nitrites
Ammonia
Hardness
Total P
Trace Elements
COD MPR
TOC
BOD MPR
Turbidity
Color
Suspended Solids MPR
Temperature
PH
Conductivity
Total Col i forms
Fecal Col i forms
Sample Frequency:
MPR = Municipal Plant Records
D = Daily Grab
DC = Daily 24 Hr. Comp.
2M = Twice Monthly
A.S.
influent
DC & 4P
DC
DC
DC
we
DC
we
DC
DC
DC
D
DC
DC
A.S.
Effluent
we
we
DC & 4P
we
DC
DC
DC
we
DC
we
DC
DC
DC
D
DC
DC
D
D
W
we
4M
4P
Upflow
Multi -Media
Clarifier Filter
Effluent
DC
DC
DC
DC
DC
DC
DC
DC
= Weekly
= Weekly
= 4 Times
= 4 Mrs.
Effluent
we
DC
DC
DC
we
DC
DC
DC
DC
DC
DC
Grab
Comp.
Per Month
@ Plant
Carbon
Column
Effluent
we
we
DC
DC
DC
DC
DC
Final
Chlorinated
Effluent
we
we
DC
we
DC
DC
DC
we
we
we
DC
DC
4M(DC)
DC
DC
DC
D
DC
DC
D
D
-------
SECTION VI
PLANT OPERATION AND PERFORMANCE
The Demonstration Plant of the Dallas Water Reclamation Research Center
was configured as shown in Figure 10 for the potable reuse research effort
discussed in this report. The basic process control philosophy was to pro-
duce as high a quality product water as possible by operating the completely-
mixed activated sludge system to nitrify, maintaining a pH of at least 11.3
in the upflow clarifier, and employing extended empty-bed contact time by
utilizing two granular activated carbon contactors in series operation.
The primary coagulant fed to the upflow clarifier was hydrated lime
with a minimal dose of ferric chloride used as a flocculation aid. Single-
stage recarbonation of the high pH effluent from the upflow clarifier was
accomplished by using liquid carbon dioxide. Two-stage recarbonation with
intermediate clarification would have reduced the calcium in the product
water and been much more desirable; however, these facilities were not avail-
able at the Demonstration Plant.
The recarbonated water was then pumped to the No. 1 mixed-media filter,
which is a conventional gravity-flow filter with Neptune Microfloc media and
a rate of flow controller on the influent line. Both granular activated
carbon contactors were operated in series, with the trailing (No.3) contactor
being loaded with virgin carbon. The final effluent was then disinfected
with a chlorine solution in the chlorine contact basin.
Table 11 presents a brief summary of the performance of the Demonstra-
tion Plant during this research effort. All values reported are arithmetic
means for the study period; the raw wastewater data were obtained from the
monthly reports of the Wastewater Treatment Division, Dallas Water Utilities
Department.
The plant performance was excellent during most of the project period,
as the effluent quality indicates. Nitrification was not complete, since an
average of 2.6 mg/1 of ammonia nitrogen was present in the final product
water; however, the total nitrogen removal approached seventy-two percent
with only 9.1 mg/1 of nitrogen being found in the product water. The Demon-
stration Plant has no nitrogen removal process per se, and this significant
reduction in total nitrogen was not anticipated.
The observed reductions in gross organic materials as measured by the
five-day BOD test and COD test were about as anticipated except that the
average product water COD concentration of 3.3 mg/1 is somewhat lower than
one might predict.
41
-------
Q =
Qr = 8.3 lit/sec
132 gpm
Ho. 1
Final
Cl arif ier
9.6 lit/se
252 gpm I No. ]
Aeration
Basin
No. 1 Activated Sludge System
Q = 6.9 lit/sec
109 gpm
Recarbonation
Basin
3.8 lit/sec,
61 gpm
Up-Flow Clarifier
Q = 2.4 lit/sec
38 gpm
Chlorine
Contact
Basin
Q = 1.1 lit/sec
18 gpm
Q = 2.0 lit/sec
32 gpm
No. 3 Column
No. 4 Column
No. 1 Mixed-Media Filter
Activated Carbon System
Figure 10. Demonstration Plant configuration for the potable reuse study.
-------
TABLE 11
SUMMARY OF DEMONSTRATION PLANT PERFORMANCE
Parameter
TSS
BOD
5
COD
UOD*
NH -N
3
Org.N
NO -N
2
NO ,&NO -N
2 3
Total N
Total P
pH
Raw Wastewater
(mg/1)
212
198
442
375
16.9
14.6
0.0
0.5
32.0
12.6
7.1
AWT Effluent
(mg/1)
2
1.3
3.3
14
2.6
0.6
0.16
5.9
9.1
0.4
6.9
Reduction
(percent)
99.1
99.3
99.3
96.3
84.6
95.9
71.6
96.8
"Ultimate Oxygen Demand = 1.5 (BOD5.) + 4.6 (NH3-N)
43
-------
Product water total suspended solids (TSS) averaged 2 mg/1, which is
well below the practical working limit of the procedure.
COMPLETELY-MIXED ACTIVATED SLUDGE
The completely-mixed activated sludge system was operated at an average
flow of 9.6 liters per second (152 gpm) during this study. The influent,
pumped from the White Rock STP, was effluent from the primary clarifiers
during most of the project, although alternate influent sources were used on
some occasions. The performance of the completely-mixed activated sludge
system is summarized in Table 12, which presents arithmetic means for the
various water quality parameters. Geometric means have been reported for
total plate counts (TPC/ml), total coliforms (TC/100 ml), and fecal coliforms
(FC/100 ml).
The treatment obtained by the biological process was about what one
would expect for a conventional activated sludge system treating a "typical
municipal wastewater". Effluent total suspended solids averaged 22 mg/1
for the study period, which represents an observed reduction of approximately
90 percent. An equivalent reduction in gross organic compounds, as measured
by COD and TOC,was observed.
Nitrification was the major process control criterion for the activated
sludge system during this study. The effluent ammonia nitrogen concentration
averaged 3.7 mg/1 for the study, representing an ammonia nitrogen reduction
of 78.1 percent, which exactly coincides with the observed total Kjeldahl
nitrogen reduction. The combined effluent nitrite-nitrate nitrogen concen-
tration was considerably lower than expected, averaging 6.5 mg/1. The low
concentration of oxidized forms of nitrogen resulted in a total nitrogen
reduction by biological treatment of 58.1 percent, which is higher than one
would normally predict.
An alkalinity decrease of 76 mg/1 (as CaC03) resulted from nitrification
in the activated sludge system, which represents an observed alkalinity de-
crease of 6.5 mg/1 for each 1.0 mg/1 of ammonia nitrogen oxidized. The fac-
tor of 6.5 compares favorably with the theoretical value of 7.2 mg/1 of alka-
linity destroyed per mg/1 of ammonia nitrogen oxidized. However, when the
observed alkalinity reduction is compared with nitrate production the ratio
is 12.3. This is much higher than theory predicts, and when evaluated in
terms of the high observed total nitrogen removal, the conclusion is that
very significant denitrification must have occurred in the secondary clari-
fier.
Table 13 summarizes the hydraulic process control parameters for the
completely-mixed activated sludge system, and the facility operation must be
considered conservative when considering these parameters. The theoretical
residence time in the secondary clarifier of 6.9 hours is longer than norm-
ally encountered, and does provide an opportunity for uncontrolled denitri-
fi cation to occur.
Process control parameters for the activated sludge system are summa-
44
-------
TABLE 12
PERFORMANCE SUMMARY OF THE COMPLETELY-MIXED ACTIVATED SLUDGE SYSTEM
PARAMETER RAW
WASTEWATER
(mg/1)
SS
COD
TOC
NH3-N
Org.N
TKN
N02 & N03-N
Total N
Total P
PH
Alkalinity
Hardness
TPC/ml
TC/100 ml
FC/100 ml
F"
NaCl
S04
TDS
*Soluble TOC
212
442
148
16.9
14.6
31.5
0.5
32.0
12.6
--
--
A.S.
INFLUENT
(mg/1)
124
281
*39.2
15.4
15.4
24.1
0.3
24.4
8.7
7.1
228
188
3.9xl05
1.4xl08
8.5xl06
1.26
--
--
A.S.
EFFLUENT
(mg/1)
22
44
*n.3
3.7
3.2
6.9
6.5
13.4
6.0
7.3
152
176
2.1xl05
9.1xl05
1.3xl05
1.27
100
94
496
REDUCTION
(percent)
89.6
90.0
92.4
78.1
78.1
78.1
58.1
52.4
33.3
94.2
99.3
98.5
--
--
--
45
-------
TABLE 13
HYDRAULIC PROCESS CONTROL PARAMETERS FOR THE COMPLETELY-MIXED
ACTIVATED SLUDGE SYSTEM
Q Wastes
T(Q+Qr)
Clarifier T
Clarifier T (Q+Qr)
Overflow Rate
Weir Loading
9.59 liters/sec
152 gpm
8.33 liters/sec
132 gpm
4478 liters/day
1183 gpd
4.9 hours
2.6 hours
6.9 hours
3.7 hours
12.6m3/day/m2
310 gpd/ft2
29.8 m3/day/m2
310 gpd/ft2
46
-------
rized in Table 14. Mixed liquor suspended solids averaged 2723 mg/1 during
the study period, and the average MLVSS to MLSS ratio was 0.83. The return
activated sludge (RAS) concentration averaged 5858 mg/1, which is much lower
than one would encounter under "normal" operation. The return sludge flow
meter at the pilot plant does not register at flows less than about 6.2
liters per second (100 gpm) and a relative high recycle ratio of 0.87 was
used in order to obtain a reading on the return sludge flow meter. The high
recycle ratio clearly reduced the return sludge concentration.
The average food to microbe ratio and sludge age for the study period
were 0.50 kg COD applied / day / kg MLSS and 10.4 days, respectively. Both
values are approximately sufficient for maintaining nitrification in moder-
ate temperatures. Monthly average wastewater temperatures ranged from a
high of 29°C during July 1974 to a low of 18.3°C during December 1974 and
January 1975. The average wastewater temperature for the project period
was 24.2°C.
Figure 11 presents time-series plots of the mixed liquor suspended
solids concentration and the seven-day, moving-average sludge age. The MLSS
concentration fluctuated about the mean of 2723 mg/1 only slightly through
November 1974. However, in early December a major process upset occurred
when one of the plant operators left the waste sludge valve open for his
entire shift. The result was a decrease in MLSS from 3300 mg/1 to about
400 mg/1.
The loss of solids from the system produced a commensurate decrease in
sludge age, and very poor treatment. In order to rebuild the MLSS as quick-
ly as possible, ferric chloride feed was switched from the upflow clarifier
to the aeration basin of the activated sludge system. During the period of
December 7 to 31, 1974 an average ferric chloride dose of 27 mg/1 was fed to
the aeration basin; additionally, no sludge was wasted during this time.
The aeration basin dissolved oxygen concentration and dissolved oxygen
uptake rates measured in the mixed liquor and return sludge are shown in
Figure 12. During most of the project period the oxygen transfer equipment
provided by Aquarius, Inc. functioned well, and sufficient residual dis-
solved oxygen was present in the aeration basin to prevent the nitrifying
system from becoming oxygen limited. The Aquarius equipment was removed
for maintenance from August 13 to August 25, 1974, and during that time
mixing and oxygen transfer was maintained by utilizing four temporary air
diffusers. The performance of the diffusers was less than optimum, and the
process was oxygen limited with respect to nitrification.
The Aquarius equipment suffered a major break-down on December 21, 1974
and from then until the termination of this project the activated sludge
system operated with the stand-by diffusers providing oxygen transfer.
The plant operators performed uptake rate determinations on both the
mixed liquor and return sludge every six hours. The average dissolved
oxygen uptake rate measured in the mixed liquor was 25.1 mg-hr/1 for this
study. Considerable variability in oxygen demand is evident in Figure 12.
Peak oxygen demands of 66 mg-hr/1 were measured, and this value represents
47
-------
TABLE 14
PROCESS CONTROL PARAMETERS FOR THE COMPLETELY-MIXED
ACTIVATED SLUDGE SYSTEM
MLSS = 2723 mg/1
MLVSS = 2271 mg/1
MLVSS/MLSS =0.83
RAS = 5858 mg/1
F/M(COD) = 0.50 kg COD applied/day/kg MLSS
F/M(TOC)s = 0.070 kg soluble TOC applied/day/kg MLSS
F/M(NH3-N) = 0.028 kg NH3-N applied/day/kg MLSS
F/M(TKN) = 0.43 kg NH3-N applied/day/kg MLSS
Sludge Age = 10.4 days
Mixed Liquor D.O. Uptake Rate = 25.1 mg-hr/1
Return Sludge D.O. Uptake Rate = 34.3 mg-hr/1
Aeration Basin Temp = 24.2°C
Aeration Basin D.O. = 3.3 mg/1
48
-------
10 20 30 10 20 30 10 20 30 10 20 30 10 20 30 10 20 30 10 20 30
JUL | AUG | SEP | OCT | MOV | DEC JAN |
1974 I 1975
Figure 11 Mixed Liquor Suspended Solids concentrations and seven-
day, moving-average sludge ages for the activated sludge
process.
49
-------
10 20 30 LO 20 30 10 20 30 10 20 30 10 20 30 10 20 30 10 20 30
JUL
AUG
SEP OCT
NOV
DEC
1974
IAN
1975
Figure 12. Aeration basin D.O. concentration and uptake for
the completely-mixed activated sludge system.
50
-------
an average of one day's operation. The return sludge uptake rates are higher
than those for the mixed liquor, but obvious correlation exists.
Sludge volume index (SVI) values for the mixed liquor are shown in
Figure 13. The activated sludge system at the Demonstaration Plant has al-
ways operated with SVI values that are higher than one would like; however,
no adverse operation can be attributed to the high SVI. No attempt was made
to relate SVI and sludge age or any other parameter since ferric chloride
was fed to the aeration basin for three weeks.
Effluent ammonia nitrogen and total suspended solids concentrations for
the project period are shown in Figure 14. Nitrification was generally good
until December 7, 1974, at which time the previously discussed process upset
resulted in a total loss of the nitrifying population. From mid-August to
mid-October, 1974 nitrification was erratic. Sludge ages were consistently
above ten days, and should have been adequate to support complete nitrifi-
cation. During this two-month interval two maintenance problems were exper-
ienced at the Demonstration Plant, both of which had adverse effects on
maintaining proper process control for a nitrifying biological system. First,
one of the two air compressors in the oxygen transfer systems had bearing
and overheating problems which resulted in the process becoming oxygen limi-
ted on occasion. Second, the influent magnetic flow meter on the activated
sludge system was in almost constant need of calibration, making accurate
flow measurement totally impossible.
Many of the peaks in effluent TSS shown in Figure 14 can be directly
related to hydraulic washout from the secondary clarifier when influent
flows increased, but were not indicated by the flow recorder.
Over the past several decades different investigators have reported
problems with "rising sludge" in the secondary clarifiers of nitrifying
activated sludge processes. The cause of the rising sludge has generally
been attributed to uncontrolled denitrification in the secondary clarifier.
Furthermore, uncontrolled denitrification appears to be prevalent when only
partial nitrification is occurring. The experience gained at the Demonstra-
tion Plant tends to support the above observations; however, quantification
of the findings has been difficult. Figure 15 presents effluent total
suspended solids plotted as a function of effluent ammonia nitrogen. Some
degree of correlation is obvious, but too many other variables are involved
to permit definitive data analysis.
Effluent soluble TOC and effluent total COD concentrations for the
activated sludge process are shown in Figure 16. The effluent soluble TOC
data illustrate the capability of completely-mixed systems for removing
soluble organic compounds. The major process upset of December 7, 1974 did
little to increase the soluble TOC in the effluent. The observed transients
in effluent COD concentrations are much more pronounced, as would be expected.
Figure 17 shows effluent COD as a function of effluent TSS. Consider-
able scatter exists; however, the indicated correlation is obvious. One can
approximate a soluble COD of about 15 mg/1 from this graph, and after coagu-
lation and filtration the COD of the activated sludge effluent averaged
51
-------
40
10 20 30 10 20 30 10 20 30 10 20 30 10 20 30 10 20 30 10 20 30
IUL
BUG
SEP
OCT
NOV
DEC
1974
IAN
1975
Figure 13. Sludge Volume Index for the activated sludge process.
52
-------
10 20 30 10 20 30 10 70 30 10 20 30 10 20 30 10 20 30 10 20 30
JUL
AUG
SEP
OCT
NOV
DEC
1974
JAN
1975
10 ?0 30 10 20 30 10 20 JO 10 20 30 10 20 30 10 20 30 10 20 30
JUL
AUG
SEP
OCT
NOV
DEC
1974
JAN
1975
Figure 14. Effluent TSS and NH3-N concentrations for the
completely mixed activated sludge system.
53
-------
tn
110
100
90
80
00
C/3
30
20
10
i.» v
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18
EFFLUENT NH3-N (mg/l)
Figure 15. Effluent TSS as a function of effluent NH3-N for the activated sludge system.
-------
10 20 30 10 20 30 10 20 30 10 20 30 10 20 30 10 20 30 10 20 30
Jul
Stp
Oct
No*
Dtt
1974
Jan
1975
Figure 16. Effluent total COD and soluble TOC concentrations for the
activated sludge system.
55
-------
isu-
i n ft
14U
1 1 ft
1 oU
1 O ft
l£U
11 n
1U
i n A
1UU
" ^ on
\ 90
OJ)
o on
o 80
o
.^ 7fl
UJ / U
=3
U.
LL.
'*' c n
ou
en
J U
/in
411
O ft
J U
on
i U
i n
I U
*
. .
.
*:
*
V-
:" .«s
|v
..
; y
*.
*:
"
-.
1
*
I
1
*
.
.
.
*
0 10 20 30 40 50 60 70 80 90 100
EFFLUENT TSS (mg/l)
Figure 17. Effluent COD as a function of effluent TSS for the
activated sludge system.
56
-------
17mg/l.
Between July and November 1974, the effluent COD concentration was
typically 50 mg/1 or less. On those days when the effluent COD was signif-
icantly higher, the effluent TSS were generally above 40 mg/1 as a result
of hydraulic overloading of the final clarifier.
Fecal coliform data for the activated sludge effluent are shown in
Figure 18.
UPFLOW CLARIFIER
The upflow clarifier (Infilco Densator) at the Demonstration Plant was
used as a high-pH lime coagulation, flocculation, and settling basin for the
activated sludge effluent during this project. Hydrated lime was fed as the
primary coagulant, and ferric chloride was fed as a flocculation aid during
the first five months of the project. The average flow treated by the upflow
clarifier was 6.88 liters per second (109 gpm), and the average hydrated
lime dose was 284 mg/1.
The water quality data for the upflow clarifier are summarized in
Table 15. During this project the total suspended solids in the upflow
clarifier effluent were higher than expected. Figure 19 is a time-series
plot of the effluent TSS, and also indicates monthly average ferric chloride
doses (as FeCl3). During the first four months of the study the average
ferric chloride dose was 16 mg/1 and the effluent TSS averaged about 30 mg/1.
In an effort to reduce operating costs the average ferric chloride dose was
reduced to 5.3 mg/1 in November, and the increase in effluent TSS that
resulted can be clearly seen in Figure 19. During December 1974 and January
1975, the ferric chloride feed was diverted to the activated sludge system,
and the effluent TSS reflect this change in operation.
Another factor which was contributory to the high TSS was the absence
of sludge recirculation. The Densator has the capability of solids contact
operation when utilizing lime as the primary coagulant. However, the sludge
recirculation pump was used to pump the upflow clarifier effluent to the
recarbonation basin, and not for sludge recirculation needed for solids
contact operation. Prior experience with solids contact operation indicates
that the sludge recirculation is effective in reducing effluent solids carry
over and effluent turbidity.
A fifty percent reduction in COD was observed and can be directly at-
tributed to the removal of volatile suspended solids from the activated
sludge effluent. Although the effluent TSS from the upflow clarifier were
high, the volatile suspended solids rarely exceeded 5 mg/1.
The only significant change in the nitrogen forms was the 1.2 mg/1
decrease in organic nitrogen that resulted from the removal of volatile
suspended solids. The effluent suspended solids from an activated sludge
process are about eight percent nitrogen by weight, therefore, the 1.2 mg/1
organic nitrogen decrease is approximately equal to a 15 mg/1 decrease in
57
-------
OL
oo
10 20 30 10 20 3d 10 20 30 10 20 30 10 20 30 10 20 30 10 20 30 10 20 30
JUL
AUG
SEP
OCT
NOV
DEC
1974
JAN
1975
FEB
Figure 18. Fecal coliform for the activated sludge effluent.
58
-------
TABLE 15
WATER QUALITY SUMMARY FOR THE UPFLOW CLARIFIER
Parameter
TSS
COD
Soluble TOC
NH3-N
Org.N
TKN
N02&N03-N
N02-N
Total N
Total P
pH
TPC/ml
Total Co li/ 100ml
Fecal Col i/1 00ml
Activated
Sludge
Effluent
(mg/1)
22
44
11.3
3.7
3.2
6.9
6.5
0.55
13.4
6.0
7.3
2.1xl05
9.1xl05
1.3xl05
Upflow
Clarifier
Effluent
(mg/1)
72
22
8.8
2.8
2.0
4.8
6.7
0.74
11.5
1.02
10.95
<4.7xl01
<3.7
<3.2
Reduction
(percent)
--
50.0
22.1
24.3
37.5
30.4
30.4
14.2
83.0
83.0
>99.98
>99.9996
>99.997
59
-------
cy>
CD
10 20 30 10 20 30 10 20 30 10 20 30 10 20 30 10 20 30 10 20 30
JUL
A DC
16.8
SEP
14-5
OCT
16.0
NOV
5.3
FeCU DOSE (mg/l)
DEC
1974
JAN
1975
Figure 19. Upflow clarifier effluent TSS and ferric chloride doses.
-------
the suspended solids present in the activated sludge effluent.
The reduction in total phosphorus (83 percent) was good, but not excel-
lent. Figure 20 presents the total phosphorus concentrations in the effluent
of the upflow clarifier during this study, and a strong correlation with
effluent TSS is evident. The pH in the upflow clarifier and the total phos-
phorus concentrations in the effluent from the mixed-media filter are
shown in Figure 21. These data behave as expected, and predict a total phos-
phorus concentration of lessthanO.5 mg/1 when the pH is above 11.0.
The control of pH proved to be a more difficult problem than initially
anticipated. Figure 22 presents the pH values observed in the effluent of
the upflow clarifier. The principal process control criterion for the upflow
clarifier was to maintain a pH of 11.3. Clearly this condition was seldom
satisfied. Lime feed was difficult to control due to several factors, the
most serious of which was the line carrying the lime slurry to the upflow
clarifier. The line is a 1.9 cm (3/4 inch) PVC pipe about 15 meters (50 feet)
in length, and scale build-up is quite rapid. Approximately 11 liters of
1:1 HC1 were pumped through the lime feed line once a week to control scale,
and this procedure was relatively successful in preventing blockages. Addi-
tionally the calcium hydroxide content of the commercial lime varied from a
high of 98 percent to a low of 60 percent, which necessitated changes in
lime feed rates with each load of lime.
Although the pH in the upflow clarifier was slightly lower than the
intended value of 11.3, excellent bacterial kills were observed in this unit
process. Total plate counts were reduced almost four logs to a geometric
mean of less than 47 per ml for the project period. Greater than four log
reductions were observed for both total and fecal coliforms. The geometric
mean for total coliforms was less than 3.7 per 100 ml, and less than 3.2 per
100 ml for fecal coliforms.
Effluent total plate count and fecal coliform data for the upflow clari-
fier effluent are shown in Figure 23. The total count data have been conver-
ted to an organisms per 100 ml basis, such that the values are consistent
with the fecal coliform data. Positive tubes were found on only seven of the
forty-nine samples set up for fecal coliform determinations. The total plate
counts exhibited considerable variability, and were about three logs higher
than the fecal coliform data.
SINGLE STAGE RECARBONATION BASIN
A galvanized steel tank, 1.83 meters (6 feet) in diameter and 3.66 meters
(12 feet) in depth, was used as the single-stage recarbonation basin during
this study. The basin had an effective liquid depth of 3.40 meters (11 feet-
2 inches). Effluent from the upflow clarifier was pumped through a 5.08 cm
(2 in ) PVC line, and entered the recarbonation basin 36 cm (14 in.) above
the tank bottom. The effluent from the recarbonation basin flowed over a
90° V-notch weir and into a trough piped to the suction-side of a centrifugal
pump.
61
-------
cr>
ro
10 29 30 10 20 30 10 20 30 10 20 30 10 20 30 10 20 30 10 20 30
JUL
AUG
SEP
OCT
NOV
DEC
1974
JAN
197S
Figure 20. Total phosphorus concentrations observed in the upflow clarifier effluent.
-------
CTl
OO
12 0
11.0
as 10 o
o
u.
O.
-V
- + - «
. ^_.._
^ - _, 1,
1-- t 1 1
I f
0 5
10 1.5 2
TOTflL P IN FILTER EFFLUENT (mg/\)
2.5
Figure 21. Total phosphorus as a function of pH.
-------
10 20 30 10 20 30 10 20 30 10 20 30 10 20 30 10 20 30
JUL
»UG
SEP
OCT
NOV
DEC
1974
10 20 30
JIN
1975
Figure 22. Upflow clarifier effluent pH values.
-------
o
10 2B 30 » 20 3» 10 20 30 10 20
10°
Figure 23. Effluent total plate counts and fecal coliforms for
the upflow clarifier.
65
-------
Liquid carbon dioxide was stored in a 5.4X104 kg (6 ton) receiver. The
liquid carbon dioxide was passed through a pressure regulator and flow meter,
and into the 1.27 cm (1/2 in.) PVC diffusers mounted 0.3 meters (1 foot)
above the bottom of the recarbonation basin. The average carbon dioxide dose
for this project period was 420 mg/1 (3489 lbs./106gal.).
Water quality data for the recarbonation basin are summarized in Table
16. As utilized in this study, recarbonation is principally a pH control
process and the water quality data indicate few significant, changes, except
for pH, which was reduced from a mean value of 10.95 to 6.9.
Effluent pH values for the recarbonated effluent are shown in Figure 24.
The pH value desired during the study was 7.0, and these data indicate the
pH generally ranged from 6.5 to 7.0. The lime-coagulated activated sludge
effluent had very low buffering capacity at neutral pH values, and the plant
operators found it possible to maintain better process control at slightly
lower pH values.
The recarbonation basin was operated at an average flow of 3.8 liters/
sec. (61 gpm), which resulted in a theoretical residence time of 39 minutes.
Under these conditions the basin had a surface overflow rate of 126.4m3m2/day
(3104 gpd/sq.ft.), which is sufficiently low to permit some settling. Total
suspended solids removal averaged 59.7 percent, and the effluent TSS concen-
tration averaged 29 mg/1. The fifty percent reduction in total phosphorus
is in excellent agreement with the TSS removal, and should be expected.
NO. 1 MIXED-MEDIA FILTER
The No. 1 mixed-media filter is a relatively conventional gravity filter
equipped with Neptune Microfloc media (anthracite, sand, and garnet). During
this research effort, this filter was used to process the activated sludge
effluent that had been chemically coagulated with lime and recarbonated. The
water quality data for this unit process are summarized in Table 17. The
only water quality parameters that were significantly affected by filtration
were total suspended solids, turbidity, and COD.
A 29.2 percent decrease in COD was observed across the filter, and this
reduction can be attributed to the removal of volatile suspended material
from the product water. The arithmetic mean of the effluent COD values was
17 mg/1, which indicates that good treatment usually preceded filtration.
Figure 25 is a time-series plot of the effluent COD concentrations observed
for the filter, and for practical purposes these values are indicative of
the soluble COD in the activated sludge effluent. During the months of July
and August 1974, the COD values averaged approximately 20 mg/1 and a constant
decrease occurred in COD concentration until early December 1974, when the
average value was 9 mg/1. The major process upset of early December is
clearly evident, since the mean COD concentration increased over 300 percent.
It is interesting to note that with only 400 mg/1 of MLSS in the aeration
basin the COD in the filter effluent never exceeded 35 mg/1. A gradual
decrease in the COD concentration in the filter effluent occurred as the
microbial population in the aeration basin increased, and at the termination
66
-------
TABLE 16
WATER QUALITY SUMMARY FOR THE SINGLE-STAGE RECARBONATION BASIN
Parameter
TSS
COD
Soluble TOC
NH3-N
Org. N
TKN
N02&N03-N
N02-N
Total N
Total P
PH
TPC/ml
Total Col i/1 00ml
Fecal Col i/1 00ml
Upflow
Clarifier
Effluent
mg/1
72
22
8.8
2.8
2.0
4.8
6.7
0.74
11.5
1.02
10.95
<4-7xl01
<3.7
<3.2
Recarbonation
Basin
Effluent
mg/1
29
24
11.1
3.5
1.9
5.4
6.1
0.55
11.5
0.51
6.9
3.7x10^
750
44
Reduction
(percent)
59.7
--
5.0
--
9.0
25.7
--
--
50.0
--
--
67
-------
CT>
oo
10 20 30 10 20 30 10 20 30 10 20 30 10 20 30 10 20
JUL I »UC I SEP I OCT I NOV I DEC
10 20 30
Figure 24. Effluent pH values for the single-stage recarbonation basin.
-------
TABLE 17
WATER QUALITY SUMMARY FOR THE NO. 1 MIXED-MEDIA FILTER
PARAMETER
TSS
COD
Soluble TOC
NH3-N
Org. N.
TKN
N02 & N03-N
N02-N
Total N
Total P
PH
Color (Pt-Co)
Hardness
Turbidity (FTU)
F"
S(K
TPC/ml
TC/100 ml
FC/100 ml
RECARBONATION
BASIN
EFFLUENT
(mg/1)
29
24
11.1
3.5
1.9
5.4
6.1
0.55
11.5
0.51
6.9
--
--
--
3.7x10^
7.5xl02
4.4X101
NO. 1 MIXED-MEDIA
FILTER
EFFLUENT
(mg/1)
17
17
10.1
3.3
1.8
5.1
6.0
0.52
11.1
0.45
6.8
11
155
4.8
1.15
90
8.7xl03
1.5xl03
4.3xl02
REDUCTION
(PERCENT)
41.4
29.2
9.0
5.7
5.3
5.6
1.6
5.4
3.5
11.8
--
76.5
69
-------
o
10 20 30 10 20 30 10 20 30 10 20 30 10 20 30 10 20 30 10 20 30
JUL
AUG
SEP
OCT
NOV
DEC
1974
JAN
1975
Figure 25. Effluent COD values for the mixed-media filter.
-------
of the project the COD concentration in the filter effluent had returned to
very low levels.
The average effluent values for both TSS and turbidity are much higher
than one would expect for a filtered water. Figures 26 and 27 are plots of
the effluent TSS concentrations and turbidity values, respectively. Quite
obviously two major breakthroughs occurred, one in July the other in Septem-
ber 1974. With the exception of those two instances, effluent TSS and tur-
bidity are good.
As directed by their supervisor, the plant operators were backwashing
the filter at a rate of 8.9 liters/sec./meter2 (13.2 gpm/ft.2) until late
September. During that time the filter was inadequately cleaned during the
backwashing procedure, and the result was cracking of the filter-bed on two
occasions. Figure 28 presents bed expansion data for the filter as a function
of backwash rate. Under normal operating conditions the desired bed expan-
sion would be in the range of 30 to 50 percent, and an expansion of only 12
percent is realized at a backwash rate of 8.9 liters/sec./m2(13.2 gpm 1 ft.2).
After September 1974, the backwash rate was increased to 12.2 liters/sec./m2
(18 gpm/ft.2) and filter performance was substantially improved.
Sporadic turbidity breakthrough is evident after the backwash rate was
increased, and this problem can be attributed to one major difficulty. The
surface wash system on the filter was inadequate when the filter was modified
for the Microfloc media. As a result of inadequate service water pressure
and a poor bearing assembly, the surface wash worked only infrequently and
this had an adverse effect on filter performance, and periodic turbidity
breakthrough resulted.
Table 18 presents a month by month summary of the operation of the mixed-
media filter, and a summary for the complete project period. When considered
in terms of hours of service between backwash cycles, the filter operation
was good, averaging 51 hours. The wash-water consumption represented only
2.85 percent of the total filtered product water, and the average filtration
rate of 1.90 liters/seo/meter2 (2.8 gpm/ft.2) was selected to be represent-
ative of typical filter operations.
ACTIVATED CARBON CONTACTORS
The two granular activated carbon contactors at the Demonstration Plant
were operated in series for this research project. Each contactor was loaded
with 1700 kq (3750 Ibs.) of Calgon Filtrasorb 400, with a mesh size of 12 x
40 The carbon beds were 1.22 meters (4.0 feet) in diameter and 3.05 meters
(10 0 feet) deep The carbon in the No. 4 Column had been previously used,
but'had been very lightly loaded, while the carbon in the No. 3 Column was
virgin.
The No. 4 Column always preceded the No. 3 Column in the treatment
si^dW^^^
The theoretical empty-bed contact time (EBCT) was 30.7 minutes for the No. 4
71
-------
10 20 30 10 20 30 10 20 30 10 20 30 10 20 30 10 20 30 10 20 30
JUL
fllC
SEP
OCT
NOV
DEC
1974
JAN
1975
Figure 26. Effluent total suspended solids concentrations for the
mixed-media filter.
72
-------
10 20 30 10 20 30 10 20 30 10 20 30 10 20 30 10 20 30 10 20 30
JUL
DUG
SEP
OCT
NOV
DEC
1974
IAN
1975
Figure 27. Effluent turbidity values for the mixed-media filter.
73
-------
TABLE 18
OPERATIONS SUMMARY FOR THE NO. 1 MIXED-MEDIA FILTER
Month
July,
Aug. ,
Sept.
Oct. ,
Nov. ,
Dec. ,
Jan. ,
Product Water
(liters x 106)
(gals, x 106)
1974
1974
, 1974
1974
1974
1974
1975
Total for
Project
3.872
(1.023)
3.077
(0.813)
6.143
(1.623)
6.658
(1.759)
6.082
(1.607)
4.868
(1.286)
5.443
(1.438)
36.143
(9.549)
Backwash Water
(liters x 104)
(gallon x 104)
6.47
1.71
4.542
(1.20)
13.32
(3.52)
37.32
(9.86)
16.18
(4.27)
14.19
(3.75)
10.83
(2.86)
102.88
(27.18)
Number
of
Washes
10
6
15
24
13
12
10
90
Avg. Hours
Between
Washes
53
72
48
31
55
56
74
51
Filtration Rate
(liters/sec. /m2)
(gpm/ft.2)
1.76
(2.6)
1.70
(2.5)
2.04
(3.0)
2.17
(3.2)
2.04
(3.0)
1.76
(2.6)
1.76
(2.6)
1.90
(2.8)
Washwater Used
i (percent of
product)
1.7
1.5
2.2
5.6
2.7
2.9
2.0
2.85
-------
on
45-
40-
35-
30
E25- -
:20- -
'15- -
10- -
0
0
18.0-
16.5-
15.0-
13.5-
12.0-
J10.5-
U
e
z 9 °"
o
X
£ 7.5-
UJ
UJ
" 6.0-
4.5
3.0
1.5
0.0
s BACKWASH RATE 10 (liters/sec/meter2) ,5 18
1 1
5 1() BACKWASH RATE 15 (gpm/ft2) 20 25
,
/ -
/
/*
'/
/
/
/*
./
*/^ WATER TEMP. = 29.5°C
-
.
/*
50 100 BACKWASH 150 fLOW 200 (gpm) 250 300 35
1 1 1 1
0 5 BACKWASH 10 FLOW (lil«rs/s«) '5 20
.9
r50
-45
-40
35
o
2.
a.
r-}
UJ
CO
-202
CJ
ly
UJ
Q.
15
10
5
0
0
Figure 28. Media expansion characteristics for the No. 1 mixed-media
filter.
-------
Column and 56.3 minutes for the No. 3 Column, with a resultant total empty-
bed contact time of 87 minutes for the complete activated carbon system.
Since the influent to the No. 4 Carbon Column was a filtered, chemically-
coagulated, activated sludge effluent, loss of head increased slowly and only
infrequently was backwashing required. The lead contactor was backwashed
daily, while the No. 3 Column was backwashed weekly. The backwash frequency
was principally a matter of convenience for the plant operators, since the
loss of head on either contactor rarely exceeded 0.6 meters (2 feet). Ad-
ditionally, no problems with hydrogen sulfide formation in the activated
carbon contactors were experienced.
Table 19 presents a water quality summary for the activated carbon
contactors. A very significant reduction in total suspended solids, 88.2
percent, was observed; however, if the filter had been properly operated
for the entire project, the solids loading to the carbon columns would have
been greatly reduced. Turbidity was reduced from 4.8 FTU to 0.7 FTU by the
carbon contactors, for an observed reduction of 85.4 percent.
Biologically refractory organic materials were reduced to very low
levels by the carbon contactors. The product water COD averaged 3.3 mg/1,
which represents an 80.6 percent reduction. Figure 29 is a time-series plot
of the COD concentrations in the activated sludge effluent and the final
product water from the No. 3 Carbon Column. The effect of the process upset
in early December 1974 on the quality of the activated sludge effluent is
quite obvious. However, only a very slight increase in the product water
COD resulted from what one would normally consider a major perturbation in
process control.
The effluent COD concentrations for the No. 3 Carbon Column are shown
as a function of the influent COD concentrations for the No. 4 Carbon Column
in Figure 30. As expected, these data exhibit relatively good correlation
and predict a COD removal of approximately 75 percent.
With the exception of organic nitrogen, no significant changes in the
various forms of nitrogen occurred as a result of carbon adsorption. Orga-
nic nitrogen concentrations were reduced by 66.7 percent, from 1.8 mg/1 to
0.6/mg 1.
Turbidity, total phosphorus, and soluble TOC data are shown for the
activated carbon effluent in Figure 31. The peaks in total phosphorus
coincide with increases in turbidity, which is to be expected.
The activated carbon contactors were very effective in reducing color
from 11 to 0.3 units on the platinum-cobalt scale.
76
-------
TABLE 19
WATER QUALITY SUMMARY FOR THE GRANULAR
ACTIVATED CARBON CONTACTORS
Parameter
TSS
BQD5
COD
TOC
NHs-N
Org.N
TKN
N02 & NOs-N
NOz-N
Total N
Total P
PH
Color (Pt-Co)
Alkalinity
Hardness
Turbidity (FTU)
F"
Sp. Cond.
NaCl
S0i+
TDS
TPC/ml
TC/100 ml
FC/100 ml
No. 1
Multimedia
Fi 1 ter
Effluent
(mg/1)
17
17
10.1
3.3
1.8
5.1
6.0
0.52
11.1
0.45
6.8
11
155
4.8
1.15
--
--
90
__
8.7xl03
1.5xl03
4.3xl02
No. 4
Carbon
Column
Effluent
(mg/1)
2
--
6.2
6.8*
2.1
1.1
3.2
6.1
0.20
9.3
--
6.8
0.7
227
-'-
897
130
--
630
330
440
37
No. 3
Carbon
Column
Effluent
(mg/1)
2
1.3
3.3
5.8*
2.6
0.6
3.2
5.9
0.16
9.1
0.44
6.9
0.3
--
139
0.7
1.18
128
--
611
710
320
23
Reducti on
(percent)
88.2
80.6
42.6
21.2
66.7
37.2
1.7
69.2
18.0
2.2
97.3
--
85.4
--
--
91.8
78.7
94.6
*See comment on p 85
77
-------
CO
160
140
120
100
^ 90
SO
60
40
20
Figure 29. COD concentrations for the activated sludge effluent and the final product water.
-------
Aft
4£
A J
«_
ji *
38 -
\
j
x 32
X
j 3t -
e
o
z 78-
o
co
S 2H
w
* 24-
d
, "-
0 <.*
o 20-
o
z 18~
LU
5 ..,
i 1«-
14-
12-
10-
s-
«
4-
0-
*
. «
0
k
4
«
. S.
»
1
«
2
EFFLU
*
E *
»
l
3
ENT
»
4
COD
5
- N
t
D. 3
7
CARB
S
ON C
9
:OLU»
10 1
IN
LI 12 13 14 15 1C 17 11
(mg/l)
10
>>
I/)
o
-o
01
o
rO
01
O
CJ
d)
Z5
o
E
O
O
c
to
It!
O
O
S-
O)
4->
rd
o
D-
O
OO
S-
3
CD
79
-------
00
o
1974 1975
Figure 31. Turbidity, total phosphorus, and soluble TOC values for the product water.
-------
SECTION VII
PROCESS RELIABILITY
When considering wastewater reclamation for potable reuse, process reli-
ability must be of paramount concern. During this research effort the three
operational problems described below perturbed the routine process control
of the Demonstration Plant:
1. Failure of an operator to close a valve on schedule while
wasting activated sludge caused a major loss of sludge and
created a major and prolonged process upset in December 1974.
2. Improper backwashing of the mixed-media filter during the
first months of the project resulted in two significant
turbidity breakthroughs.
3. Mechanical problems with the hydrated lime feed system made
pH control in the upflow clarifier very difficult.
Although these problems were totally undesirable when considered in
terms of experimental design, the fact that they occurred makes the research
effort more useful, since it permits evaluation of the stability of the over-
all treatment sequence.
ORGANIC MATERIALS
Logarithmic-normal probability distributions for the observed COD con-
centrations in the various process flows are shown in Figure 32. Log-normal
probability graphs will generally plot data which are skewed to the right as
reasonably straight lines, and for this reason are very useful in presenting
water quality data obtained from wastewater effluents.
Figure 33 is a histogram of the COD values observed for the effluent
from the No. 3 activated carbon contactor. These data are obviously skewed
to the right, and a normal probability distribution for these data is shown
in Figure 34.
Although the activated sludge process upset that occurred in early
December 1974 had some effect on the quality of the product water, the
impact was much less than one might have expected. In fact the three highest
COD values measured on the product water (29, 15, and 15 mg/1) were observed
during the first week of operation. Previous experience at the Demonstra-
tion Plant indicates that about two weeks are required before a new charge
of granular activated carbon will consistently produce a high quality
81
-------
1,000
UPFLOW
CLARIFIER
EFFLUENT
NO. 3
CARBON
COLUMN
EFFLUENT
10 20 30 40 50 60 70 80 90 95 98
PERCENT OF SAMPLES WITH COD CONCENTRATIONS LESS THAN OR EQUAL TO THE
INDICATED VALUES
Figure 32. Log-normal frequency distribution for observed COD values.
82
-------
45 ,
35
30
z 25
20
u
u.
O
15
10 -
5
fj
'*
5 10 15
RANGE OF COD VALUES (mg/l)
I
20
25
30
Figure 33, Histogram of product water COO concentrations.
83
-------
oo
18
16
14
12
\ 10
^_E
o
S 8
0.01
7
0.1
2 10 20 30 40 50 60 70 80 90 98 99
PERCENT OF TIME LESS THAN OR EQUAL TO PLOTTED VALUE
99.9
99.99
Figure 34. Normal probability distribution of COD values observed in the effluent of the No. 3
carbon column.
-------
effluent. It is interesting to note that the six highest COD values were
observed during the first two weeks following the start-up of the No 3
contactor.
Soluble TOC data for the activated sludge influent and effluent, and
total TOC data for the effluent from the No. 3 Carbon Column are shown in a
log-normal probability distribution in Figure 35.
Mean and median COD and TOC values are compared for the different sample
locations in Table 20. In most cases the median values are somewhat lower
than the mean value, and since these data are positively skewed this condi-
tion is to be expected.
The TOC data obtained on the effluent from the No. 3 Carbon Column seem
to be questionable. The TOC analyzer used for these determinations was cali-
brated such that full deflection was equivalent to a TOC concentration of
200 mg/1. Evaluation of BOD5 : COD : TOC ratios obtained on previous
research indicates that the COD values reported are correct, and that the
TOC data are incorrect. The average COD and TOC ratio observed during this
project was 2.08; hence, this ratio was used to calculate the mean and median
TOC concentrations shown on the table of 2.8 and 1.9.
TOTAL SUSPENDED SOLIDS AND TURBIDITY
Log-normal probability distributions for total suspended solids and
turbidity are shown in Figures 36 and 37, respectively. The effect of the
suspended solids and turbidity breakthrough that occurred on the No. 1 Filter
is obvious. However, since the gravity-flow carbon contactors function as
relatively good filters the quality of the product water remained relatively
good.
The turbidity in the effluent from the No. 3 Carbon Column exceeded 1.0
FTU in twenty percent of the samples, but only exceeded 3.0 FTU in three
percent of the samples. Filtration aids, such as polyelectroytes or alum,
were not used during this project, and their use would have improved product
water quality when difficulties with the filter were being experienced.
TOTAL PHOSPHORUS
The observed total phosphorus concentrations are presented in Figure 38
as log-normal probability distributions. The median total phosphorus con-
centration in the activated sludge influent of 8.4 mg/1 was reduced to 6.2
mg/1 by biological treatment.
The effluent from the upflow clarifier had a total phosphorus concen-
tration that exceeded 2.6 mg/1 in ten percent of the samples. The extreme
variability in the observed phosphorus concentrations can be directly
related to problems in controlling the lime feed to the upflow clarifier
and switching the ferric chloride feed from the upflow clarifier to the
activated sludge process.
85
-------
* SOLUBLE TO C
NO. 3
CARBON
COLUMN
EFFLUENT
0.1
10 20 30 40 50 60 70 80 90 95
98
PERCENT OF SAMPLES WITH TO C CONCENTRATIONS LESS THAN OR EQUAL TO THE
INDICATED VALUES
Figure 35. Log-normal frequency distribution of observed TOC
concentrations.
86
-------
TABLE 20
COMPARISON OF MEAN AND MEDIAN COD AND TOC CONCENTRATIONS FOR DIFFERENT PROCESS FLOWS
SAMPLE LOCATION
Activated Sludge Influent
Activated Sludge Effluent
Upflow Clarifier Effluent
Recarbonation Basin Effluent
No. 1 Mixed Media Filter Effluent
No. 3 Carbon Column Effluent
COD (mg/1)
Mean
281
44
22
24
17
3.3
Median
276
39
20
17
16
2.2
TOC tmg/1 )
Mean
39*
11*
9
11
10
5.8
(2.8)
Median
39*
11*
8
9
9
4
(1.9)
00
*Soluble TOC
( ) - Calculated TOC values
-------
NO. 3 CARBON COLUMN EFF
0.0
5 10 20 30 40 50 60 70 80 90
PERCENT OF SAMPLES £ INDICATED VALUE
95
98
Figure 36. Log-normal frequency distribution of observed TSS
concentrations.
88
-------
NO.l FILTER EFFLUENT
NO. 3
COLUMN
CARBON
EFFLUENT
2 5 10 20 30 40 50 60 70 80 90 95 98
PERCENT OF SAMPLES WITH TURBIDITY VALUES LESS THAN OR EQUAL TO THE
INDICATED VALUES
Figure 37. Log-normal probability distribution of the observed turbidity
values.
89
-------
UPFLOW
CLARIFIER
EFFLUENT
NO.3
CARBON
COLUMN
EFFLUENT
0.1-
10 20 30 40 50 60 70 SO 90 95 98
PERCENT OF SAMPLES WITH TOTAL PHOSPHOROUS CONCENTRATIONS LESS THAN OR
EQUAL TO THE INDICATED VALUES
Figure 38. Log-normal frequency distribution of observed total phosphorus
concentrations.
90
-------
The effluent from the No. 3 Carbon Column had very low phosphorus con-
centrations. Two samples had total phosphorus concentrations exceeding 2 0
mg/1, and only five samples exceeded 1.0 mg/1.
NITROGEN COMPOUNDS
Log-normal probability distributions for ammonia nitrogen, nitrite-
nitrate nitrogen, and organic nitrogen are shown in Figures 39, 40, and 41,
respectively. Nitrification was not achieved consistently during this pro-
ject as a result of difficulties encountered with the oxygen transfer equip-
ment and operator error. The NH3-N concentration in the activated sludge
effluent exceeded 2.0 mg/1 in forty percent of the samples, although the
median concentration was 0.8 mg/1. No significant changes in ammonia
nitrogen concentrations were observed in the processes following the acti-
vated sludge system.
The median nitrite-nitrate concentration observed in the product water
was 6.4 mg/1, and the nitrite-nitrate concentration exceeded 10 mg/1 in only
eight percent of the samples.
The median organic nitrogen concentration in the activated sludge efflu-
ent was 2.8 mg/1, and the AWT processes (principally that of activated carbon)
very significantly reduced the median organic nitrogen concentration to 0.45
mg/1.
METALS
Fifteen special weekly composite samples were analyzed for a total of
thirteen different metals at four different points in the treatment sequence.
These data are summarized in Table 21.
SILVER
Silver was present in such low concentrations that analysis for this
metal was terminated after the fifth weekly composite sample. At the low
silver concentrations observed in the City of Dallas wastewater, no signif-
icant change in the concentration of this metal occurred.
ARSENIC
Observed arsenic concentrations for the study period are shown in
Figure 42 Total removal of arsenic was excellent, averaging almost 86 per-
cent. High pH lime coagulation and filtration were the most important
processes effecting the removal of arsenic.
BORON
Boron concentrations are shown in Figure 43 and obviously no signif-
icant removal of boron was observed during this study.
91
-------
NO. 3
CARBON
COLUMN
EFFLUENT
0.1
10
20 30 40 50 60 70 80
90 95
98
PERCENT OF SAMPLES WITH NHj-N CONCENTRATIONS LESS THAN OR EQUAL TO THE
INDICATED VALUES
Figure 39. Log-normal frequency distribution of NH3-N concentrations,
92
-------
N0.3
CARBON
COLUMN
EFFLUENT
0.1
2 5 10 20 30 40 50 60 70 80 90 95 98
PERCENT OF SAMPLES WITH N02 AND N03 CONCENTRATIONS LESS THAN OR
EQUAL TO THE INDICATED VALUES
Figure 40. Log-normal frequency distribution of N02 & N03-N concentrations,
93
-------
100 -T
A.S.INFLUENT
UPFLOW
CLARiFIER
EFFLUENT
A. S. EFFLUENT
NO. 1 FILTER EFFLUENT
NO.3
CARBON
COLUMN
EFFLUENT
0.2
0.1
2 5 10 20 30 40 50 60 70 80 90 95 98
PERCENT OF SAMPLES WITH ORGAN 1C NITROGEN CONCENTRATIONS LESS THAN OR
EQUAL TO THE INDICATED VALUES
Figure 41. Log-normal frequency distribution of organic nitrogen
concentrations.
94
-------
TABLE 21
MEAN METALS CONCENTRATIONS OBSERVED IN WEEKLY COMPOSITE SAMPLES
Metal
Ag
As*
B 1
Ba
Cd
Cr
Cu
Fe
Hg*
Mn
Pb
Se*
Zn
A. S.
Influent
(mg/1)
0.0024
20.9
0.29
0.18
0.012
0.077
0.141
0.65
0.297
0.071
0.072
4.4
0.113
A. S.
Effluent
(mg/1)
0.0026
17.0
0.30
0.12
0.005
0.027
0.029
0.59
0.164
0.041
0.034
2.9
0.063
Reduction "by
Biological
Treatment
(percent)
Inc.
18.7
Inc.
33.3
58.3
64.9
74.4
9.2
44.8
42.3
52.8
34.1
44.2
No. 1
Filter
Effluent
(mg/1 )
0.004
4.1
0.29
0.14
0.003
0.012
0.091
0.12
0.139
0.011
0.034
0.5
0.068
Reduction by
Chemical
Treatment &
Filtration
(percent)
Inc.
75.9
3.3
Inc.
40.0
55.5
Inc.
79. 7
15.2
73.2
0
82.8
Inc.
No. 3
Carbon
Column
Effluent
(rag/1)
0.0024
3.0
0.27
0.14
0.002
0.020
0.046
0.10
0.512
0.011
0.035
0.9
0.063
Reduction by
Carbon
Absorption
(percent)
40
26.8
6.9
0
33.3
Inc.
49.5
16.7
Inc.
0
Inc.
Inc.
7.3
Total
Observed
Reduction
(percent)
0
85.6
6.9
& t-l LJ
83. 3
74.0
67.4
84.6
Inc.
84.5
51.4
79.5
44.2
VD
CJ1
* - micrograms per liter
Inc. - Increase
-------
A. S. Inf.
A. C. Eff.
Filter Eff
n A. S. Eff.
3 10 17 24
Jan
1975
Figure 42. Observed variations in arsenic concentrations,
96
-------
A.S. INF.
A.C. EFF.
A A FILTER EFF
a a A.S. EFF.
7 13 20 27
Dec
1974
3 10 17 24
Jan
1975
8 15 22
Nov
0.0
Figure 43. Observed variations in boron concentrations.
97
-------
BARIUM
Barium is a relatively refractory metal, and only a 22.2 percent total
reduction was observed. All of the observed reduction occurred in the acti-
vated sludge process. Barium concentrations are shown in Figure 44.
CADMIUM
Total cadmium removal was good, averaging 83.3 percent. The activated
sludge process reduced the mean cadmium concentration from 0.012 mg/1 to
0.005 mg/1, and was the most significant process with respect to the removal
of this metal, as indicated in Figure 45.
CHROMIUM
The observed chromium concentrations are shown in Figure 46, and appear
well behaved with the exception of the samples collected during the week of
December 27, 1974 through January 2, 1975. The samples on the filter efflu-
ent, and most obviously the effluent from the No. 3 Carbon Column, exhibit
peaks that appear to indicate sample contamination with acid-dichromate.
Chromium removal was excellent, averaging 74 percent. Both the biological
and the physical-chemical processes were responsible for significant chro-
mium removal.
COPPER
As indicated in Figure 47, copper reductions were quite significant
during this study. The observed total reduction was 67.4 percent, and the
activated sludge process was responsible for the greatest removal, reducing
the mean concentration from 0.141 mg/1 to 0.029 mg/1.
IRON
Iron reductions were quite significant, averaging 84.6 percent. The
two peaks in the activated sludge effluent, shown in Figure 48, that occurred
during the ninth and tenth weeks were the result of ferric chloride feed to
the activated sludge process.
MERCURY
Mercury concentrations are shown in Figure 49, and these data are very
erratic. No significant removal of mercury was observed during this research
effort;in fact, an increase of about 0.2 micrograms per liter was observed,
although the activated sludge process did reduce the mercury concentration
by about 45 percent.
MANGANESE
Manganese removal was very good, averaging 84.5 percent for the project
period. The concentrations observed in the weekly composite samples are
shown in Figure 50. The data for the effluents from the No. 1 filter and
the No. 3 Carbon Column are very well behaved and average 0.011 mg/1. The
98
-------
0.24
0.22
A. S. INF.
A. C. EFF.
A FILTER EFF
1974 j 1975
Figure 44. Observed variations in barium concentrations.
99
-------
0.035
0.030
0.025
0.020
0.015
0.010
0.005
0.000
Figure 45. Observed variations in cadmium concentrations,
100
-------
A. S. Inf.
A. C. Eff.
Filter Eff
A. S. Eff.
0.012
0.000
Figure 46. Observed variations in chromium concentrations.
101
-------
0.288
0.264
0.240
0.216
0.192
0.168
0.144
Q_
o
0.120
0.096
0.072
0.048
0.024
0.000
A.S.Inf.
A. C. Eff.
Filter Eff
A. S. Eff.
1974 1975
Figure 47. Observed variations in copper concentrations,
102
-------
1.04
1.00
0.90
A.S. Inf.
A C. Eff.
Filter Eff
A S. Eff
0.10
0.04
Figure 48. Observed variations in iron concentrations.
103
-------
5.001
0.50-r
0.45
A.S. Inf.
A.C.Eff.
A A Filter Eff.
1974 1975
Figure 49. Observed variations in mercury concentrations,
104
-------
0.08
0.07
0.06
0.05
oc
E
0.04
0.03
0.02
0.01
0.00
A A Filter Eff.
D D A. S. Eff.
3 10 17 24
Jan
1975
7 13 20 27
Dec
1974
Figure 50. Observed variations in manganese concentrations.
105
-------
activated sludge is quite variable, and a significant peak develops at
precisely the same time the major process upset occurred. The manganese
concentration in the activated sludge effluent then decreases as the biomass
in the activated sludge system becomes re-established.
LEAD
The lead concentrations observed during this study were somewhat erratic,
as the data in Figure 51 indicate. The only significant lead removal observed
was the result of biological processes; the physical-chemical processes were
rather ineffectual in altering lead concentrations.
SELENIUM
As the data in Figure 52 indicate, selenium is present in the City of
Dallas wastewater at very low concentrations. This metal's concentration
was reduced by an average of 79.5 percent, and the physical-chemical processes
were particularly effective.
ZINC
The changes in zinc concentrations observed during this research are
difficult to quantify because the recarbonation basin was a. galvanized
steel tank. The data shown in Figure 53 indicate that a relatively constant
44 percent zinc reduction could be anticipated from the activated sludge
process. As a result of zinc pick-up from the recarbonation basin it is
impossible to make any definitive statement with respect to the physical-
chemical processes.
COMPLIANCE WITH THE NATIONAL INTERIM PRIMARY DRINKING WATER REGULATIONS
The objective of this research effort was the production of a product
water that was of potable quality, and for the purposes of defining potable
water quality, those water quality criteria presented in the National Interim
Primary Drinking Water Regulations will be utilized.
Figure 54 presents a graphical comparison between the water quality
criteria established in the NIPDWR (excepting coliforms) and the arithmetic
means for the various water quality parameters monitored in the product
water from the No. 3 granular carbon contactor.
Nitrate Nitrogen
During this study the combined nitrite-nitrate nitrogen concentration
averaged 5.9 mg/1, and the nitrite nitrogen concentration averaged 0.16 mg/1.
Nitrate nitrogen concentrations were equal to, or exceeded, the 10 mg/1
standard in approximately eight percent of the samples.
Due to the instability of the activated sludge process during this
study, it is quite likely that higher nitrate nitrogen concentrations would
106
-------
0.11
0.10
0.09
0.08
0.07
£0.06
0.05
0.04
0.03
0.02
0.01
A.S.INF.
A. C. EFF.
FILTER EFF
D A. S. EFF.
i
Figure 51. Observed variations in lead concentrations.
107
-------
A. S. INF.
A. C. EFF.
A A FILTER EFF
0 D A. S. EFF.
3 10 17 24
Jan
1975
Figure 52. Observed variations in selenium concentrations,
108
-------
0.157
0.145
0.133
0.121
0.109
0.097
0.085
0.073
0.061
0.049
0.037
0.025
A.S.INF.
A. C. EFF.
Filter EFF
Figure 53. Observed variations in zinc concentrations.
109
-------
OBSERVED
CONCENTRATIONS
0.0001
CONTAMINANTS
Figure 54. Comparison between the NIPDWR criteria and product
water quality.
no
-------
occur if this research was repeated. Additionally, the total Kjeldahl nitro-
gen concentrations observed in the influent to the activated sludge process
(average - 24.1 mg/1) are lower than the 30 to 40 mg/1 which might normally
be anticipated in a municipal wastewater. For these reasons, nitrate nitro-
gen concentrations should still be considered a potential problem to the
potable reuse of wastewaters.
Turbidity
Although some difficulties were experienced in-the operation of the
filter, the turbidity values observed for the product water meet the NIPDWR
standards in all respects.
Metals
The means of the metals concentrations observed in the product water
were considerably lower than the standards promulgated in the National
Interim Primary Drinking Water Regulations (NIPDWR). The metals concentra-
tions observed in the activated sludge effluent were also lower than the
criteria defined in the NIPDWR. The mean concentrations of only three metals
(cadmium, chromium, and lead) in the effluent from the primary clarifiers
at the White Rock STP exceeded the EPA standards.
Three individual weekly composite samples exceeded the NIPDWR criteria
for one individual metal. Chromium was found at 0.145 mg/1 in the tenth
week, lead was present at 0.059 mg/1 during the twelfth week, and a mercury
concentration of 5 micrograms per liter was found during the fifteenth week.
Biocides
Determinations for chlorinated hydrocarbons and chlorophenoxys were
made on only one twenty-four hour composite sample collected from 12 noon
October 24, 1974 to 12 noon October 25, 1974. The results are summarized in
Table 22, and all compounds were present in concentrations much lower than
the NIPDWR standards.
Ill
-------
TABLE 22
RESULTS OF BIOCIDE DETERMINATIONS ON PRODUCT WATER
Observed
COMPOUND EPA Standard Concentration
(mg/1)(ng/l) (ng/1)
Chlorinated Hydrocarbons:
Endrin 0.0002 200 < 2
Lindane 0.004 4,000 < 0.3
Methoxychlor 0.1 100,000 <15
Toxaphene 0.005 5,000 N.A.
Chlorophenoxys:
2,4-D
2,4,5-TP
0.1
0.01
100,000
10,000
1,750
<50
112
-------
REFERENCES
1. Wolf, H.W. & Esmond, S.E. Water quality for potable reuse of wastewater.
Watex and_ Sewaje_ Works_ 121, 48 (Feb. 1974).
2. Wastewater Reclamation Studies^ in_ DaJJas^. Texas. Final Report on Project
No. 17080EKG, Dallas Water Utilities Department and Texas A&M Research
Foundation, March 1973.
3. Research and Development Division Projects Receiving Federal Assistance
1969-1976, Report to the Director, Dallas Water Utilities Department,1976.
4. Wolf, H.W., Safferman, R.S., Mixson, A.R., and Stringer, C.E. Virus
inactivation during tertiary treatment. Virus^ Survival in Water and
Hastewater Systems, Ed. Malina, J.F., Jr., and Sagik, B.P. Center for
Research in Water Resources, The University of Texas at Austin (1974).
5. Standard Methods for the Examination of Water and Wastewater, 13 Ed.,
APHA, Washington, D.C.,1971.
6. Methods for Chemical Analysis of Water and Wastes. U.S. Environmental
Protection Agency, Office of Technology Transfer, Washington, D.C., 1974.
7. Instruction Manual Ammonia Electrode Model 95-10, Orion Research Incor-
porated, Cambridge, Mass., 1972.
8. Analytical Methods for Atomic Absorption Spectrophotometry, Perkin-
Elmer Corp., Norwalk, Conn., 1968.
9. Sill, C.W. and Willis, C.P. "Fluorometric Determination of Submicrogram
Quantities of Beryllium", J. Anal. Chem., 31_, 4 598 (April 1959).
10. Brown, Eugene, Skougstad, M.W. and Fishman, M.J., Techniques of Water-
Resources Investigations of the United States Geological Survey, Chapter
Al_, Methods for Collection and Analysis of Water Samples for Dissolved
Minerals and Gases, GPO, Washington, D.C., 1970.
113
-------
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before completing)
1 REPORT NO
EPA-600/2-77-210
3. RECIPIENT'S ACCESSION-NO.
. TITLE AND SUBTITLE
WASTEWATER CHARACTERIZATION AND PROCESS RELIABILITY
FOR POTABLE WASTEWATER RECLAMATION
5. REPORT DATE
November 1977
(Issuing Date)
6. PERFORMING ORGANIZATION CODE
7 AUTHOR(S)
Albert C. Petrasek, Jr.
8. PERFORMING ORGANIZATION REPORT NO
9. PERFORMING ORGANIZATION NAME AND ADDRESS
10. PROGRAM ELEMENT NO.
Dallas Water Utilities
Dallas, Texas 75201
1 BC 611
11. CONTRACT/GRANT NO.
R - 803292
12. SPONSORING AGENCY NAME AND ADDRESS
Municipal Environmental Research Laboratory--Cin.,OH
Office of Research and Development
U.S. Environmental Protection Agency
Cincinnati, Ohio 45268
13. TYPE OF REPORT AND PERIOD COVERED
Final/1974-1975
14. SPONSORING AGENCY CODE
EPA/600/14
15. SUPPLEMENTARY NOTES
Project Officer: John English
Phone: 513/684-7613
This research effort was initiated to quantify water quality criteria of
importance in evaluating the performance of a wastewater treatment facility
producing a product water potentially available for potable reuse. Additionally,
the reliability of individual unit processes was evaluated and the effects of
process instability on product water-quality were investigated.
The sequence of unit processes utilized in the study to treat municipal
wastewater consisted of screening, degritting, primary clarification, biological
treatment with completely-mixed activated sludge, high-pH lime coagulation, single-
stage recarbonation with liquid carbon dioxide, gravity filtration, and two-stage
activated carbon adsorption. Flows through the pilot plant ranged from 9.6 liters
per second (152 gpm) for the activated sludge influent to 1.1 liters per second
(18 gpm) for the product water. Twenty-four-hour composite samples were collected
daily for routing analyses; weekly composite samples were utilized for metals
determinations.
The final product water complied with the quality criteria of the National
Interim Primary Drinking Water Regulations in all respects. Significant process
instabilities had little effect on product water quality due to the redundant
nature of the treatment system employed.
17.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b.lDENTIFIERS/OPEN ENDED TERMS
c. COSATI Field/Group
Waste Treatment
*Water Reclamation^ Nutrients
Nutrients
Re Iiabi1ity
*Wastewater Reuse
Heavy Metal
*Tertiary Treatment
Dallas, Texas
13 B
B. DISTRIBUTION STATEMENT
Release to public
19. SECURITY CLASS (This Report)
[JnclassifipH
21. NO. OF PAGES
126
20. SECURITY CLASS (Thispage)
Unclassified
22. PRICE
EPA Form 2220-1 (9-73)
114
A-UA60VKNMENTPRINTING OFFICE 1977 757-140/6593
------- |