EPA-R2-73-250
July 1973 Environmental Protection Technology 3?r:°~
Sludge Processing For
Combined Physical-Chemical-
Biological Sludges
Office Of Research And Development
U.S. Environmental Protection Agencv
Washington, O.C 20460
-------
RESEARCH REPORTING SERIES
Research reports of the Office of Research and
Monitoring, Environmental Protection Agency, have
been grouped into five series. These five broad
categories were established to facilitate further
development and application of environmental
technology. Elimination of traditional grouping
was consciously planned to foster technology
transfer and a maximum interface in related
fields. The five series are:
1. Environmental Health Effects Research
2. Environmental Protection Technology
3. Ecological Research
4. Environmental Monitoring
5. Socioeconomic Environmental Studies
This report has been assigned to the ENVIRONMENTAL
PROTECTION TECHNOLOGY series. This series
describes research performed to develop and
demonstrate instrumentation, equipment and
methodology to repair or prevent environmental
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.
-------
EPA-R2-73-250
July 1973
SLUDGE PROCESStNG FOR COMBtNED
PHYS tCAL-CHEMICAL-BIOLOG tCAL SLUDGES
by
D. S. Parker
P. J. Zadtck
K. E. Train
For the
Central Contra Costa Sanitary District
Walnut Creek, California 94596
Grant No. R80I445
Project No. 17080 FSF
Project Offfcer
Robert B. Dean
U.S. Environmental Protection Agency
National Environmental Research Center
Cincinnati, Ohio 45268
Prepared for
OFFICE OF RESEARCH AND MONITORING
U.S. ENVIRONMENTAL PROTECTfON AGENCY
WASHINGTON, D.C. 20460
-------
EPA Review Notice
This report has been reviewed by the Environmental
Protection Agency and approved for publication.
Approval does not signify that the contents neces-
sarily reflect the views and policies of the
Environmental Protection Agency, nor does mention
of trade names or commercial products constitute
endorsement or recommendation for use.
11
-------
ABSTRACT
Full scale combined sludge generation from a treatment sequence con-
sisting of lime clarification, nitrification and denitrification was studied.
Pilot scale studies were conducted to wet-classify and d$water the com-
bined sludges by means of two-stage solid bowl centrifugetion. Anaero-
bic digestion of first stage centrate was also studied on a pilot scale.
Predicted and measured sludge production agreed well using four coagu-
lation modes in the chemical primary.
First stage centrifugation (wet classification) , over a wide pH range,
achieved high capture of calcium carbonate in relatively dry (42 to 57
percent total solids)cakes as well as good rejection of magnesium, phos-
phorus, and iron compounds in the centrates.
At pH 11.0 or below, second stage centrifugation dewatered first stage cen-
trate to produce 18 percent total solids cakes, with 80 percent solids
recoveries. Dewatering deteriorated at a pH greater than 11.0.
Anaerobic stabilization of thickened and unthickened first stage centrates
showed volatile matter destructions of over 40 percent, high methane-
content gas, and substantial increases in soluble magnesium and alkali-
nity during digestion.
The State of California Water Resources Control Board supported this
project under Research Grant Project No. 1-RDGI.
This report was submitted by the Central Contra Costa Sanitary District in
fulfillment of Project No. R801445 under the partial sponsorship of the
Environmental Protection Agency.
iii
-------
CONTENTS
Section Page
I Conclusions 1
II Recommendations 3
III Introduction 5
IV Advanced Treatment Test Facility 11
V Summary of Liquid Processing Investigation 21
VI Sludge Production and Thickening 29
VII Wet Classification of Combined Sludges 43
VIII Centrate Processing by Centrifugation 59
DC Centrate Processing by Anaerobic Stabilization 83
X Special Studies 97
XI Discussion 105
XII Acknowledgements 117
XIII References 119
XIV Glossary 123
XV Appendix 125
-------
FIGURES
Page
3-1 ATTF System 6
3-2 ATTF Solids Processing System 9
3-3 Solids Processing - Lime Recovery Mode 10
4-1 ATTF Flow Diagram and Design Data 13
4-2 Centrifuge Processing Schematic 15
4-3 Bench Scale Digester Schematic 16
4-4 Pepcon Treatment Schematic 15
6-1 Comparisons of Denitrification Systems 35
6-2 Clarifier Solids Loading as a Function of Flow
and Flocculator Solids Level 38
6-3 Sludge Profile on October 24, 1972 39
7-1 Constituent Recovery During Classification 47
7-2 Summary of Constituent Recoveries During
Wet Classification 48
7-3 Solids Recovery atFlocculation pH of 11.5 51
7-4 Solids Recovery at Flocculation pH of 11.0 52
7-5 Solids Recovery at Flocculation pH of 10.2 53
7-6 Dewatering at Flocculation pH of 11.0 54
7-7 Comparison of Centrifuges for Solids Recovery 55
7-8 Comparison of Centrifuges for Dewatering 56
8-1 Effect of Polymer Type on Separation in Centrifuge 63
8-2 Effect of Auger Speed on Polymer Requirement 65
8-3 Effect of Pond Setting on Recovery of Solids 66
8-4 Effect of Centrifugal Force on Recovery 67
vi
-------
FIGURES (Con't.)
Page
Effect of Feed Rate on Recovery 68
Comparison of Two Sharpies Machines for Second
Stage Operation 70
8-7A,B Effect of Flocculation pH on Solids Recovery 71
8-8 Effect of Adjusting pH of Centrifuge Feed 74
8-9 Effect of pH on Solubility of Calcium 75
8-10 Effect of pH on Solubility of Magnesium 76
8-11 Effect of pH on Solubility of Phosphorus 77
8-12 Effect of Adding Sludge to Second Stage Feed 80
9-1 Overall Volatile Matter Destruction in Lime
Digesters as a Function of Solids Retention Time 94
11-1 Solids Processing Balances for Incineration/
Recalcination at pH 10.2 108
11-2 Solids Processing Balances for Incineration/
Recalcination at pH 11.0 109
11-3 Solids Processing Balances for Incineration/
Recalcination at pH 11.5 110
11-4 Solids Processing Balances for Digestion-Land
Disposal at pH 11.0 113
i
A-l Lime Digester Operation During Low Feed Solids
Period for SRT of 15 days 126
A-2 Lime Digester Operation During Low Feed Solids
Period for SRT of 20 Days 126
A-3 Lime Digester Operation During Low Feed Solids
Period for SRT of 25 Days 127
A-4 Control Digester Operation During Low Feed Solids
Period for SRT of 15 Days 128
-------
FIGURES (Con't.)
Page
A-5 Lime Digester Operation During High Feed Solids
Period for SRT of 10 Days 129
A-6 Lime Digester Operation During High Feed Solids
Period for SRT of 15 Days 130
A-7 Lime Digester Operation During High Feed Solids
Period for SRT of 20 Days 131
A-8 Lime Digester Operation During High Feed Solids
Period for SRT of 25 Days 132
A-9 Control Digester Operation During High Feed Solids
Period for SRT of 15 Days 133
A-10 Unit Gas Production for Lime Digester During Low
Feed Solids Period, SRT of 15 Days 134
A-11 Unit Gas Production for Lime Digester During Low
Feed Solids Period, SRT of 20 Days 135
A-12 Unit Gas Production for Lime Digester During Low
Feed Solids Period, SRT of 25 Days 136
A-13 Unit Gas Production for Control Digester During
Low Feed Solids Period, SRT of 15 Days 137
A-14 Unit Gas Production for Lime Digester During
High Feed Solids Period, SRT of 10 Days 138
A-15 Unit Gas Production for Lime Digester During High
Feed Solids Period, SRT of 15 Days 139
A-16 Unit Gas Production for Lime Digester During High
Feed Solids Period, SRT of 20 Days 140
A-17 Unit Gas Production' for Lime Digester During High
Feed Solids Period, SRT of 25 Days 140
A-18 Unit Gas Production for Control Digester During High
Feed Solids Period, SRT of 15 Days 141
-------
TABLES
No. Page
5-1 Summary of Primary Sedimentation Performance
at Design Flows 23
5-2 Summary of Limed Primary Sedimentation
Performance 23
5-3 Summary of Lime/Ferric Primary Sedimentation
Performance 24
5-4 ATTF Performance Summary, April 16-July 15, 1972 27
5-5 Process Parameters for Nitrification and
Denitrification 28
6-1 Summary of Sludge Production for Several Modes
of Chemical Primary Operation 31
6-2 Comparison of Calculated Sludge Composition to
Grab Sample Analysis 33
6-3 Nitrification Sludge Production and Yield 33
6-4 Comparison of Sludge Yields for Conventional
and Modified Denitrification Systems 36
6-5 Summary of Chemical Primary Operating Conditions
During Thickening Test Periods 36
7-1 Run Data for Wet Classification 46
i
8-1 Constituent Separation During Centrifuging
of First Stage Centrate 60
8-2 Laboratory Beaker Test of Polymers for Second Stage
Centrifuge Feed 62
8-3 Effect of Flocculation pH on Second Stage Cake
Dryness 73
8-4 Effect of Adjusting pH of Centrifuge Feed on Solids
Composition 73
1x
-------
TABLES (Con't)
No.
8-5 Comparison of Observed and Theoretical
Magnesium Solubilities 79
8-6 Hardness Release Due to Bacterial Action 79
9-1 Digestion Performance Parameters 87
9-2 Feed and Digested Sludge Characteristics 89
9-3 Digester Sludge Heavy Metals Analyses at End
of High Feed Solids Run " 89
10-1 Effect of Recycle of Second Stage Centrate 98
10-2 Vacuum Filter Leaf Studies on Primary Lime
Sludge Flocculated at pH 11.0 100
10-3 Vacuum Filtration Tests on First Stage Centrate
Generated at pH 11.0 102
10-4 Filtration Rates on Pepcon Processed and Unprocessed
Centrates 102
11-1 Material Balance Comparisons at Various pH Levels 111
11-2 Chemical Costs for pH Processing Modes in
Incineration/Recalcination Flow Scheme 111
11-3 Comparisons Between Recalcination and Digestion
Solids Processing Alternatives at pH 11.0 for
30 mgd Plant 115
11-4 Comparisons of Large Scale Digester Operations
Calculated for Three Types of Treatment Plants 115
-------
SECTION I
CONCLUSIONS
1. Combined sludge generated in a chemical primary sedimentation tank
where lime is the coagulant can be accurately estimated from changes
in quality constituents across the process and knowledge of chemical
doses.
2. Considerable sludge thickening takes place in the primary sedimentation
tank. Only the first half of the rectangular tank, however, is effective
for thickening. Thicker sludge is obtained at pH 11.0 and 11.5 than
at pH 10.2.
3. Combined sludge can be effectively classified with a solid bowl centrifuge.
For a flocculation pH of 11.0 and 11.5 approximately 9C percent of the
calcium carbonate fed to the machine was recovered in the cake. At
the same time, 50 to 75 percent of the other constituents were rejected
in the centrate. The calcium carbonate-rich cake was simultaneously
dewatered to a total solids content of greater than 50 percent. At pH
10.2, the calcium carbonate recovery was 75 percent while the- reject-
ion of the other materials was 60 percent . Cake dryness at pH
10.2 was 42 to 49 percent total solids.
4. Centrate from the first stage classification centrifuge can be dev/atered
in a second' stage solid bowl centrifuge. Anionic polymer addition is
required at a dose of at least two Ib per ton of dry solids. The pH of
operation affects the cake dryness obtained. Below pH 11.0, total
solids in the cake of 17 to 18 percent can be obtained. Above pH 11.0
dewatering deteriorates, and a median cake total solids of 12 percent
was produced.
5. Anaerobic digestion can stabilize first stage centrate, but volatile matter
reduction is less for centrate processing than for conventional raw
sludge-processing . Further, virtually all of the magnesium hydroxide
fed to the digester is solubilized due to the pH of digester operation
(pH 7.0 to 8.0) . The supernatant from digestion, if returned to the
process, would cause an increase in the lime dose required because
of recycled alkalinity and magnesium. Calcium carbonate and hydroxy-
apatite are not dissolved in the digestion process to a significant degree.
6. Combined sludge can be stabilized by chlorination treatment. However,
supernatant from a chlorination stabilization step cannot be returned to
the process, as the total dissolved solids of the reclaimed water vail
be increased.
-------
7. Two feasible solids disposal alternatives have been developed for
combined sludge processing. One involves incineration of the waste
solids with separate recalcination of the calcium carbonate. The
other solids disposal alternative involves disposal of the calcium
carbonate in landfill with digestion of the waste solids. The incinera-
tion/recalcination alternative is preferable, since chemical costs are
significantly less and far less solids are produced for ultimate
disposal.
-------
SECTION II
RECOMMENDATIONS
1. One of the most pressing needs of designers of advanced waste treat-
ment systems is for meaningful design information on solids disposal
systems. While this report presents information useful for some of
the advanced waste treatment systems, there is a large number of
systems for which there is no data at all. An intensive research
effort is required to fill the gaps that exist.
2. It is recommended that future investigations of processing combined
sludges include the following:
a. Material balance studies on heavy metals across each
step in the solids processing scheme.
b. Metal oxide material balances across each solids pro-
cessing step where incineration is employed in the
processing scheme.
c. Additional studies on dry classification of ash from
recalcining furnaces.
d. Effects of temperature on thickening of chemical com-
bined sludges.
e. Effect of recycled material (inerts, heavy metals, etc.)
on an integrated solids two-stage centrifugation,
recalcination and incineration.
f. Studies of chemical sludge stabilization by chlorination
with subsequent land disposal.
3. It is recommended that solid bowl centrifuges employed in wet
classification and dewateri.ng of combined physical-chemical-bio-
logical sludges should be designed to include the following features:
a. The capability of stepwise variation in the centrifuge
force from 1100 to 3100 times the force of gravity.
b. The capability of stepwise variation of the bowl pond
setting for liquid discharge levels.
c. The capability of continuous variation of auger speed
relative to bowl speed.
-------
d. The capability of internal polymer introduction to the bowl.
Anaerobic digestion of first stage centrate from centrifnging combined
sludges should be designed to operate with feed solids of at least
3 percent volatile solids. Also, solids retention time should be close
to 20 days to attain a 40 percent volatile matter reduction.
-------
SECTION III
INTRODUCTION
Increasingly, waste treatment systems are being designed to produce
effluents of substantially higher quality than car. be obtained by conventional
primary or secondary processes. Most commonly, these new "advanced
waste treatment" systems are oriented toward the removal of nutrients in
order to prevent over-fertilization of receiving waters. These advanced
treatment systems commonly remove either phosphorus or nitrogen, or
both. Other systems have been developed for enhanced heavy metals
removal.
The removal of phosphorus normally entails precipitation by a metal salt,
which may be iron (ferric or ferrous) , aluminum (alum or sodium aluminate) ,
or calcium (lime) These coagulants may be used in the process for other
purposes, such as heavy metals removal. The use of the metal salt sub-
stantially increases the mass of sludge which must be removed by either
primary or secondary treatment. Further, the dewatering and solids
processing characteristics of the resulting sludges are substantially different
from either primary-generated raw sludge or secondary-generated bio-
logical sludges. Since there are very few operational full-scale advanced
treatment systems (most results to date are from pilot plants) , little informa-
tion is available on which to base the design of advanced treatment solids
processing systems. Nevertheless, wastewater agencies and consultants
are faced with the immediate problem of designing full-scale solids pro-
cessing facilities for a host of advanced treatment systems.
This project involved the study of solids processing for an advanced treat-
ment system, termed the ATTF system. The ATTF system is a tailored
sequence of processes specifically developed for the purpose of reclaiming
municipal wastewater for industrial use and meeting discharge requirements L
The basic ATTF flow diagram is shown on Fig. 3-1. The process couples
lime clarification with a combined oxidation-nitrification step. The use of
lime in the primary treatment stage removes much of the organic carbon
load from the oxidation-nitrification stage, thus allowing stable oxidation
of ammonia to nitrate. In addition to the removal of organic matter, the
lime enhances the removal of phosphorus, heavy metals (toxic materials)
and viruses. Other advanced treatment systems have used lime in tertiary
treatment after conventional secondary treatment, but by bringing lime
addition forward to the initial stage of treatment, the load on the treatment
processes which follow is reduced, and operation is stabilized. Biological
denitrification follows nitrification, converting nitrate to nitrogen gas.
-------
RAW SEWAGE
LIME
LIME REACTOR
(PREAERATION)
__POLYMER OR
FERRIC CHLORIDE
PRIMARY
SEDIMENTATION
CHEMICAL
PRIMARY
EFFLUENT <
OXIDATION
NITRIFICATION
1
SECONDAR
SEDIMENTATION
TANK
C02
TANK
Y
TANK
1
^
SLUDGE TO
^ SOLIDS
PROCESSING
-AIR
RETURN
SLUDGE
WASTE SLUDGE
RAW SEWAGE
N2
METHANOL
DENITRIFICATION
TANK
•MIXING
AERATED
STABILIZATION TANK
I
FINAL
SEDIMENTATION TANK
I
RETURN
SLUDGE
WASTE SLUDGE
-»- TO
RAW SEWAGE
CHLORINE CONTACT
ADDITIONAL TREATMENT
FOR INDUSTRY
Figure 3-1
ATTF System
-------
Although not shown in the diagram, multi-media filtration will be providec
after denitrification to polish the flow stream before distribution to industry
or discharge.
The ATTF system produces a combined physical-chemical-biclogicai sluuge.
Raw sewage solids, calcium phosphate (hydroxyapatite) , magnesium
hydroxide, and calcium carbonate are precipitated in the primary process,
together with waste biological sludge from the nitrification system and
ferric hydroxide when iron salts are used. The primary underflow thus
contains all of the sludge produced in the system.
The specific purpose of this project was the development of useful desicn
and operating information lor the 30 mgd Ceruraj. Contra Costa Sanitary
District (CCCSD) Reclamation Plant, which will be under construction in
early 1973. As a result, avenues of investigation had a more practiccl
orientation than.a pure research project would. The results of this research.
however, will have application wherever lime is used the primary staoe <>\
treatment. Examples are:
1. Lime precipitation in the primary for phosphorus removal,
followed by 'biological treatment by activated sludge for
BOD removal2-3'4.
2. Lime precipitation in the primary for phosphorus removal
and organics reduction, followed by activated sludge for
nitrification .
3. Lime precipitation in the primary, followed by filtration
and activated carbon adsorption6-22i
4. Lime precipitation in the primary for improved organics ,
grease, and heavy metals removal, followed by ocean
disposal'.
The first two categories listed will produce sludges nearly identical to the
ATTF system sludge; the latter two categories will produce sludges \viih
similar characteristics. ,
Sludge processing investigations for the ATTF system were oriented towards
segregating recalcinable calcium carbonate from.the other sludge con-
stituents, which would build up as inerts in the system if not removed before
the lime sludge was recalcined and reused. Recycle of reusable lime is
economically desirable because it reduces the new lime requirement of tne
process and reduces the quantity of ash destined for ultimate disposal.
Further, much of the cost of recalcination would have to be spent in any
case as a sludge processing cost.
-------
The first stage (30 mgd) of the CCCSD Water Reclamation Plant will employ
the solids processing system shown in Fig. 3-2. A two-stage centrifuge
system will be used to classify the sludge into two major fractions. The
first fraction will contain most of the recyclable lime sludge. The second
fraction will contain most of the magnesium, phosphorus and organics.
The first stage solid bowl centrifuge will be run at high throughput rate
and without polymer (deliberate inefficiency) to classify the heavy calcium
carbonate into the cake and put most of the lighter organics and chemical
inerts into the centrato. The first stage cake will be recalcined and returned
to the process, while the first stage cent rate will be conveyed tc the second
stage centrifuge. The second stage centrifuge will be run at low throughput
and with polymer for maximum soiids capture. Second stage cake will be
incinerated prior to disposal, and second stage cent rate will be returnee tc
the raw sewage.
A major goal of this investigation was the definition of the effect of changes
in process variables on the operation of each element in the solids processing
system. Particular attention was paid to sludge thickening in the primary
and to the operation of both stages of centrifugation.
Since the initial construction constitutes only one-quarter of the ultimate
plant capacity (and the waste sludge furnace can be a recalcine furnace
in a future stage) alternatives for first-stage centrate disposal system
were investigated. A schematic listing the centrate processing alternatives
is shown on Fig. 3-3. Primary effort was directed to centrifugation (as
mentioned before) and anaerobic digestion as the chief processing alterna-
tives .
The digestion process was carried out on a bench scale with first-stage
centrate as feed, thickened to two different solids levels. The effect of
solids retention time on digestion performance was studied by simultaneously
running four centrate digesters at several solids retention times. Also
a control digester was run on conventional sludge at a specific solids
retention time. The digesters were well mixed so that hydraulic and
solids retention times were equal. The digested sludges were withdrawn
daily and portions were subjected to odor and drying tests.
Some limited studies on vacuum filtration were done on the first stage
centrate. Also, an electrolytic chemical oxidation processing scheme was
investigated briefly.
-------
MAKEUP r-WASTE BIOLOGICAL SOLIDS
LIME ,
-RECYCLED LIME
RAW
PRIMARY CLARIFIER
FIRST STAGE
THICKENER
PRIMARY
*•
EFFLUENT
WET
CLASSIFICATION
HIGH
CaC03
CAKE
SECOND
STAGE
THICKENER
MULTIPLE
HEARTH
RECALCINE
FURNACE
'GAS
SCRUBBER
RECALCINED
ASH
REJECTS
*•
CENTRATE TO
PRIMARY
DRY
CLASSIFICATION
RECYCLE
LIME
STORAGE
MULTIPLE
HEARTH
FURNACE
ASH TO DISPOSAL
MAKEUP
LIME
CENTRATE RETURN
LIME
REACTOR
(PRE AER-
ATION)
ATTF
PRIMARY CLARIFIER
PRIMARY
•*•
EFFLUENT
Figure 3-2 ATTF Solids Processing System
-------
LIME
RAW
PRIMARY
»•
Centrate Processing Alternatives
Incineration or
Recalcination
MULTIPLE
HEARTH
INCINERATOR
2ND STAGE
CENTRATE TO
PRIMARY
CLARIF.ER
SLUDGE
TO
LAND
DISPOSAL
RECALCINED PRODUCT
OR ASH
Figure 3-3
Solids Processing - Lime Recovery Mode
10
-------
SECTION IV
ADVANCED TREATMENT TEST FACILITY
The construction and operation of the Advanced Treatment Test Facility by
CCCSD is one of a series of steps leading to the design and construction
of a 30-mgd water reclamation plant to be completed in 1975. The first
significant step towards water reclamation was taken in December 1969,
when CCCSD and Contra Costa County Water District (CCCWD) signed a
memorandum of understanding which paved the way for a joint effort which
will culminate in the recycling of treated wastewater to industries in
Contra Costa County. The cooperative program was outlined which included
a water reclamation feasibility study, a solid waste disposal investigation,
a sampling and analysis program, and a pilot plant demonstration program.
In February 1970, a Class IV research and demonstration grant was issued
by the Federal Water Pollution Control Administration (now Environmental
Protection Agency) to carry out a two-year on-site demonstration that
renovated water could be produced of sufficient quality to replace industry's
present raw water supply for most uses. Concurrent with this research
work, CCCSD's engineering consultants (Brown and Caldwell) completed
a preliminary design report for wastewater treatment expansion and
modifications. The treatment scheme recommended for the ultimate plant
consisted of lime coagulation-sedimentation, nitrification and denitrification.
This process sequence later became known as the ATTF system.
In April 1972, CCCSD completed contract negotiations with CCCWD for sale
of the reclaimed water. The water contract set quality limits, among which
were 1 mg/1 total phosphorus and 5 mg/1 total nitrogen. A maximum value
for total dissolved solids was established at 375 mg/1 greater than industry's
present water source, the Contra Costa Canal. The phosphorus limit of
1.0 mg/1 as P was set to prevent calcium phosphate scaling of heat exchangers*
The nitrogen limit of 5 mg/1 was aimed at preventing algal growths on
distribution structures and open recirculating cooling towers.
The original CCCSD-CCCWD pilot plant demonstration program had not been
designed to evaluate the ATTF process sequence. In separate systems both
lime treatment of raw sewage and nitrification were evaluated, but not as
parts of the sequence proposed in the ATTF design. Denitrification was
attempted late in the pilot program after Brown and Caldwell had proposed
the ATTF sequence of processes. However, stable nitrification and
denitrification were never successfully demonstrated on a pilot scale. This
led regulatory agencies to question the reliability of the proposed ATTF
system. In order to estabish the credibility of the ATTF system and generate
design information, CCCSD undertook the construction of a full-scale
Advanced Treatment Test Facility. During the initial construction of the
11
-------
ATTF, CCCSD was not assisted by either state or federal research grants.
The test facility was placed in operation in late November 1971.
As an integral part of the ATTF system, solids processing had to be evaluated
to produce both a feasible processing 'scheme and design information for
the selected processing system. Studies were conducted on the centrifugation
of the primary lime sludge to determine the feasibility of classifying the
calcium carbonate from the sludge in order to recover lime solids. These
studies provided useful design information, but CCCSD sought to under-
take a further reasearch effort to obtain larger scale centrifuge performance
data as well as to evaluate anaerobic digestion of lime sludges. To finance
this effort, CCCSD sought research grants for funding the "Combined
Sludge Processing Project". Both the State of California and Environmental
Protection Agency provided funds for the project. Experimentation began
in the summer of 1972.
This report deals with the results of the entire research effort on solids pro-
cessing, including that conducted prior to commencement of grants. Solids
processing equipment used in the study is described in this section. Also
described are the analytical methods employed in the analysis of solids
streams generated in the ATTF process.
SOLIDS PROCESSING EQUIPMENT
Solids processing consists of sludge generation and handling as well as
sludge treatment. The treatment schemes tested in this project dealt with
both classification and stabilization of chemical sludges.
Primary Solids Generation and Handling
The chemical primary operation begins with the addition of lime slurry,
ferric chloride, and recirculated primary sludge to the raw sewage flow
stream. The next step, preaeration, serves as both a flocculation and
grit removal step. Flocculated solids are settled out as chemical primary
sludge in the primary sedimentation tank. The settled sludge is removed
from the primary tank by a variable speed pump through a timer-actuated
valve which regulates the relative proportion of wasted to recirculated
sludge. Wasted sludge is pumped to a digester together with the con-
ventional raw sludge from the parallel control primary tank. Recirculated
sludge is returned to the preaeration chamber to maintain the high ^olids
level necessary for efficient flocculation. A flow diagram and design dajta
for the chemical primary and other features of the ATTF system are shown
in Fig. 4-1.
12
-------
SUlSUN
BAT
. t
PILOT PLANT DESIGN DATA
EXISTING TREATMENT PLANT
PLANT LOAOlNO
MINIMUM, HOD
4ve*A0r PHY WlATttt*, tiet>
/»«A* mtT WfATHfX, M9P
*eo9l«t/t
SVIHMHO toupt, loeo i§*/o*r
PILOT PLANT DESIGN DATA
PRCACRATIDH AND CHIT REMOVAL TANKS
KU*9t* . COMTHOL
MVtttllt - en*Mlt*.l.
WfOTX, ft
LtH4TH, fT
Avttitet **rtn otfttt. rr
afrenrion nut. Ml
OtMfOH FLO* ft* f***, *«»
LfMC FEED C9UIPUENT
irvr HUM fir tr0K*9t TMK
PBI«ARr SE&IMENTATtON TANKS
HUHtt* ' eonrnOL
HUMttlH - CNCWtCAI.
WIDTH, fT
tVtHAOl mArt It DtfTH, fT
oeriHftoM Tme, nut
avgnrLow mre, OAL /so f r
otsiOH now mi TANK, ueo
CAFtarr, OP*
PRIMARY EFFLUENT PUMP
CAPACITY, BP"
TOTtL Ml*S, fT
H0ltftro#e*, Mf
CARSON OfOXIOC fCEDCH
HUMtt*
fAFAtnr, Ltt/DAr
CHEMICAL PRIMARY TRCATM
•OOj HtUOVtL
ftnetHT
ioo-**o
PCRFOHMANCE
BOO,. I.OS/OA1
Sl/S«*OEI> SOtlOS, LBS/0*f
PfitMttY EfFLatHT TO SfCOHDAHY T*ttTUtN
foOf, t.*s/o**
SUtMHMO SOLIDS, L9S/W
DlStOH rt.OW, CPU
»oo lojwtta, i.ti/>ooo cu rr/o*r
tvfXAQt tr<*rrft OCPTH, rr
VISION ft,o», am
oveatLQ* ntT€, oti/se tr/OM
OtTtNTtC* T,ut, Hits
RETURN NITRIFIED SLUDOE AIR LIFT PUMP
ttuuati*
etpACtrr, efu
SECONOARV EFFLUENT PUMP
«U«»f*
MMC«rr, 8P«
TOTAL HC*O, 't
HOtsirawitt , HP
METMANOL FEED PUMP
sss;,.«^.
KNiril.nciir.ON TA»x
AERATED STAfl.L.ZAT.ON TANK
NVMOIK
w»0r«, T
LINSTH. rr
-------
Centrifuges and Related Equipment
Chemical primary sludge was pumped to anyone of four 2500-gal. tanks
where it was stored until fed to the centrifuge (Fig. 4-2) . Carbon dioxide
diffusers were installed in two holding tanks to enable pH adjustment of the
centrifuge feed. A bar screen removed the rags from the feed sludge, pre-
venting pump clogging or blockage of the centrifuge feed tube. Air diffusers
kept the sludge mixed during storage. A similar but much smaller feed
and storage system was used in the test work done prior to research grant
work.
The usual operating procedure began with operation of the centrifuge in a
first stage mode (centrifuging chemical primary sludge) and collection of
the centrate in one of the storage tanks. The necessary machine changes were
then made, and the machine was run in a second stage mode (centrifuging
first stage centrate). Three centrifuges were tested, two extensively.
The Sharpies P600 horizontal centrifuge used in the study had a bowl
diameter of 6 inches, a bowl length of 14 inches, and a feed rate range of
0 to 12 gpm. The centrifuge feed pump was a Moyno 12-gpm variable
speed pump.
The Sharpies P3000 horizontal centrifuge used in the study had a bowl diameter
of 14 inches, a bowl length of 30 inches, and a feed rate range of 10 to 50
gpm. The centrifuge feed pump was a Moyno 50-gpm variable speed pump.
A Bird solid bowl centrifuge with 6-inch bowl diameter and 12-inch bowl
length was employed for limited test work on lime sludges. The feed rate
range was 0 to 9 gpm. A Robbins-Meyers Model PS HE feed pump was used.
Polymer was mixed in an 800-gallon tank and pumped to the centrifuge feed
tube via a small variable speed chemical pump. Polymer normally was fed
internally into the feed zone.
.Pilot Digestion Equipment
The digestion work was carried out oh a bench scale, using 5-gal glass
bottles for both the reactors and the gas collection vessels. Sludge lines
and valves were constructed of polyvinyl chloride, while gas lines were
constructed of Tygon tubing. ,Fig. 4-3 shows a flow schematic of the
apparatus. The digesters were mixed by 10-gpm external circulation
pumps activated by a timer.
The digesters were placed in a water bath composed of a molded plastic
water bed frame kept at a constant temperature by the water bed's thermostati-
cally controlled heating element. Heat input was augmented by a 250-watt
-------
AIR
BAR 1
SCREEN*
T
I
i
I—CXH
CHEMICAL
SLUDGE
FEED—^
I
TO WASTE
OR PRIMARY
CENTRIFUGE
-[XI »• WASTE
RECYCLE
POLYMER
TANK
CENTRIFUGE
FEED PUMP
WATER
Figure 4-2
Centrifuge Processing Schematic
15
-------
1
1
DISPL
L1QU
COLLE
ACED
D
CTOR
T
-
_
-
••
s— GAS LINE
— \ K. ^
}
1 GAS
SAMPLE
VALVE
-
1
— - :
—
-
-
j
=_ — —
t
4x3—1 FEED
FUNNEL
1
-1X1—)
SAMPLE
TAP
GAS DIGESTER A~^
COLLECTOR (Volume Ca ibrated ) TIMER ACTUATED
BOTTLE CIRCULATION PUMP
CENTRATE
FEED
(Volume Calibrated)
Figure 4-3
Bench Scale Digester Schematic
ALTERNATE DISCHARGE FOR ONCE
THROUGH SYSTEM
CATHODE LINE
PEPCON
CELLS
DISCHARGE
Figure 4-4
Pepcon Treatment Schematic
16
-------
heating element. The entire water bath was enclosed in a large insulated
wooden box, hinged at the top. The heating water was kept well circulated
by an external pump. Both water circulation and digester sludge circulation
lines were insulated to reduce heat loss. The apparatus maintained sludge
temperature within 0.5 C of water bath temperature (normally 38 C) .
Electrolytic Test Unit
A Pepcon electrolytic test treatment facility was operated to treat first stage
centrate from lirne sludges. The unit, produced by Pacific Engineering
and Production Nevada, consisted of the following major components:
Rectifier 1000 amp
Cells, 4 each, lead 500 sq in . each
dioxide anode 2000 sq in. total
Cathode, 4 each,
copper
Recirculation pumps,
2 each 24 gpm
Recirculation tank 30 gal
Brine tank 30 gal
The unit was modified by taking two cells (anode-cathode assembly) out
of the flow circuit and connecting the remaining two cells in parallel,
both hydraulic-ally and electrically. A schematic of the Pepcon apparatus
is shown in Fig. 4-4.
ANALYTICAL METHODS
A number of analyses on the solids streams were required to determine
performance of the various processes as well as component recoveries in
the various streams. Most of the analyses were standard techniques as
described in Standard Methods for_ the Examination of Water a_nd Waste water,
13th Edition. Analyses employed in this project which are listed in
Standard Methods are: total and volatile solids; total hardness (EDTA method)
calcium hardness (EDTA method) , (magnesium hardness by difference of
total and calcium hardness); soluble phosphate by the stannous chloride
method; total calcium, magnesium, iron, and heavy metals by atomic
absorption spectrophotometry; total Kjeldahl nitrogen; grease by magnesium
sulfate monohydrate dehydration plus Soxhlet extraction with hexane;
ammonia nitrogen on centrate of laboratory-centrifuged sludge by the
Nesslerization method; and sulfides by the methylene blue visual color-
matching method.
Additional analyses performed on sludges were calcium hydroxyapatite,
magnesium hydroxide, ferric hydroxide, calcium carbonate, organic matter,
and acid insoluble matter. Procedures for these analyses were as follows:
17
-------
I. A weighed sample of sludge or centrate was dried at 105 C
to obtain total solids.
2. A portion of the dried sample was transferred to a crucible,
ignited to 1000 C for 2 hours, then collected and weighed to
obtain loss on ignition (LOI) .
3. A portion of the residue from ignition was dissolved in hydrochloric
acid, and the insoluble matter filtered out and weighed to give
acid insoluble matter.
4. The filtrate from the above was analyzed for Ca, Mg, Fe (when
required) by atomic absorption, and for PO4 by the stannous chloride m
method.
The calculations from the above steps were as follows:
1. From PO4, the calcium hydroxyapatite was calculated. (All PC>4
was assumed to be in the hydroxyapatite form) .
2. Calcium (net) was determined by deducting from the total the
calcium equivalent to calcium hydroxyapatite.
3. The net calcium was calculated to be present as calcium carbonate.
4. Carbon dioxide equivalent to the calcium carbonate was deducted
from the LOI matter.
5. Magnesium was calculated as magnesium hydroxide (assuming
all Mg present in this form) .
6. Water equivalent to magnesium hydroxide was also deducted from
the LOI.
7. Where iron was present to a significant degree, iron previously
determined was calculated to ferric hydroxide.
8. Water eauivalent to ferric hydroxide was also deducted from the
LOI.
9. Net LOI was called organic matter.
10. Acid insoluble matter was called the inerts (mostly SiO2) .
This analytical method typically yielded recoveries of around 99.9 percent.
IB
-------
Analyses on, digested sludge were performed in the following manner:
1. A total solids analysis was made as before.
2, A portion of the dried sample was "wet ashed" with nitric and
sulfuric acids, then finally diluted to 250 ml with deionized
water.
3. The solution was analyzed for Ca, Mg, and heavy metals by
the atomic absorption technique and for PC>4 by the stannous
chloride method.
4. Total Kjeldahl nitrogen was determined by starting with the
original wet sample, using the approach given in Standard
Methods.
5. Grease was determined on the original wet sample by the
method mentioned earlier.
19
-------
SECTION V
SUMMARY OF LIQUID PROCESSING INVESTIGATION
Interpretation and use of the results of the solids processing investigation
requires an understanding of the liquid processing system which generates
the solids. A specific concern is the possible operating modes of the liquid
processing system.
ATTF SYSTEM
As previously stated, the ATTF system consists of lime clarification in a
chemical primary, followed by nitrification and denitrification. Performance
of the primary clarifiers is the main point of concern, since all solids
wasted from the process are separated in the primary. Full details on the
liquid processing system are 'available elsewherel.
Chemical Primary
After the raw sewage has been screened, lime is added and flash mixed
with recycled primary sludge. When used, ferric chloride and anionic
polymer were added at the same point. The sewage is then conveyed to
the first element in the treatment system, preaeration. Heavy grit particles,
such as sand and unreacted lime, settle out in the preaeration tank. The
mild air agitation in the tank just after lime addition promotes the growth
of large flocculent particles which settle readily. Operation of the test
facility has demonstrated that air mixing in the preaeration tank can be
substituted for mechanical mixing and thus avoid fouling problems caused
by rags contained in the raw sewage. Rags would tend to rap around the
paddles of mechanical mixers, causing continuous maintenance problems.
Waste activated sludge is also returned to the preaeration tanks for cosettling
and concentration in the primary tanks. The ATTF system incorporates no
separate waste activated sludge thiqkening. All of the solids to be processed
pass through the primary sedimentation tanks.
Carbon dioxide in the preaeration air has caused a minor operational
difficulty; scaling occurs inside the orifices of the coarse bubble diffusers.
Initially, the tank had to be drained once every week or two to clean the
plugged orifices. The plant staff solved the problem by installing a high
pressure air blowoff line and periodically blasted loose the scale. This
procedure has extended the period between manual cleaning of the orifices
to two months. In the CCCSD Water Reclamation Plant preaeration air will
be introduced through swing arm diffusers, which will allow maintenance
without having to take the tanks out of service.
21
-------
After flocculation, the suspension is allowed to settle in rectangular primary
tanks. At conventional overflow rates for rectangular primaries (average
dry weather surface loading of 1300 gpd/sf and 1.5 hours residence time),
the chemical primary far out-performs a parallel conventional primary
settling tank (Table 5-1). Peak hourly flows caused the surface loading to
reach 1900 gpd/sf.
Primary clarifier operation is managed so that the contractual requirements
for industrial reuse on final effluent water quality can be met*. To
accomplish this, the pH of operation has been varied, and supplemental
coagulants utilized to modify effluent qualities. Data for the various modes
of operation are shown in Tables 5-2 and 5-3. Of particular interest is the
effect of process pH. which determines the required lime dosage. Operation
at pH 11.0 in the primary nearly equaled operation at pH 11.5, with one
important exception, phosphorus removal (Table 5-2) . The mean total
phosphorus of 2.3 mg/1 in the primary effluent at pH 11.0 caused the final
effluent total phosphorus to exceed the 1.0 mg/1 limit.
Ferric chloride was tested as a supplemental coagulant to flocculate colloidal
phosphorus. At pH 11.0 a 12 mg/1 dose of ferric chloride reduced the
primary effluent phosphorus to less than 1.0 mg/1 (Table 5-3) . An even
lower lime dose and pH (pH 10.2) is possible if it is compensated by a
higher ferric chloride dose (24 mg/1) . Under the latter mode of operation,
the primary effluent total phosphorus averaged only 0.7 mg/1.
Removals of conventional parameters (BOD, SS, TOG) were all similar for
each mode of operation, and therefore in themselves did not dictate a
particular method for raw sewage coagulation. However, a substantial
difference in the increase of hardness across the primary was observed.
For CCCSD sewage, operation at a pH of 11.0 with ferric chloride as a
supplemental coagulant yielded the lowest effluent hardness of the coagulation
modes tested. While the theorectical minimum solubility point for calcium
in clean water is close to pH 10, the level of calcium in the pH 10.2 effluent
was as high as at pH 11.0. However, operation at pH 11.0 with ferric chloride
produced the greatest removal of magnesium and yielded the least hardness
increase across the process.
If hardness is considered a water quality limit, why not use a two-stage
recarbonation system, with intermediate settling to enhance hardness
removal? Analysis of previous pilot-demonstration work^ made during the
ATTF investigation indicated that little, if any, benefit would be gained from
the additional settling and recarbonation stage. During the previous studies
the average hardness increase across the two-stage operation varied from
-8 to +20 mg/1 according to the pH mode. The average for all pH modes was
+4. This is very likely due to the effect observed in the present study,
namely, that the soluble calcium level is relatively constant with operating
22
-------
Table 5-1
Summary of ternary Sedimentation Performance at Design Flows
Constituent
(mean value)
BOD
ss
vss
Turbidity
TOC
Soluble organic carbon
Total phosphorus as P
Orthophosphate as P
Settleable solids
Calcium hardness
Magnesium hardness
Grease
"Raw
sewage
mg/1
222
2ie
-
-
111
20
9.7
8.5
9°
68
113
55
Control primary
mg/1
137
64
57
47b
77
22
-
_
0.8°
_
-
39
% removed
38
70
_
31
-10
91
29
Chemical primary
mg/1
74
53
31
32b
52
29
2.8
1.8
0.12C
153
66
16
% removed
67
75
53
-45
71
79
99
71
Operation from February 29 to March 9 with average flow of 2.66 mgd at pH 11.0 without ferric chloride.
JTU
Cml/l
measured, typical value shown.
"as CaCO,
Table 5-2
Summary of Limed Primary Sedimentation Performance
(mean value)
BOf>5
SS
VSS
Turbidity
TOC
Soluble organic carbon
Total phosphorus as P
Settleable solids
Calcium hardness6
Magnesium hardness6
Hardness increase6
Grease
pH 11.5 operation*
Ca (OH) : 500 mg/1
Raw
sewage
mg/1
190
199
-
107
16
9.4
8.2d
76
96
57
Control primary
mg/1
103
57
43
35°
59
16
-
0.3d
_
32
%
removed
46
71
-
45
0
95
-
44
Chemical primary
mg/1
50
41
20
16C
37
23.4
0.96
nil
168
40
36
12
%
removed
74
79
65
-46
90
100
-
-28
79
pH 11 . 0 operation*1
Ca (OH)2: 400 mgA
Raw
sewage
mgA
192
195
-
118
17
9.2
9.5d
76
103
53
Control primary
mg/1
121
57
46
40C
68
21
-
0.2d
-
42
%
removed
37
71
-
42
-24
98d
-
-
21
Chemical primary
mg/1
60
47
25
26°
48
28
2.3
<0.1d
156
59
36
19
%
removed
69
76
-
59
-65
75
>99
-
-20
64
December 22, 1971 to February 10, 1972, at average flow 1.30 mgd.
February 11 to February 28, 1972, at average flow 1.12 mgd.
°JTU
e
as CaCOn
23
-------
Table 5-3
Summary of Lime/Ferric Primary Sedimentation Performance
Constituent
(mean value)
BOD,
5
ss
vss
Turbidity
TOC
Soluble organic carbon
Total phosphorus as P
Settleable solids
Calcium hardness as
CaCO
Magnesium hardness
as CaCO
o
Hardness increase as
CaCO
O
Grease
pH 11.0 operation
Ca(OH) : 400; FeCl : 14
£ «$
Raw
sewage
mg/1
210
305
130
23
9.5
13. 2d
75.5
90
146
Control primary
mg/1
109
69
49
45C
72
18
0.13d
removed
48
78
45
21
99
-
-
Chemical primary
mg/1
53
27
14
14c
37
24
0.85
nil
139
32.5
6
9.5
removed
75
91
72
-4
91
100
-
-
-4
94
pH 10.2 operation
Ca(OH) : 289; FeCl : 24
2 3
Raw
sewage
mg/1
178
235
117
20
9.4
12. ld
72
90
66
Control primary
mg/1
106
59
48
41C
68
17
.1
removed
40
75
42
15
99
Chemical prin|
mg/1
59
31
19
15 c
43
25
0.68
<0.1d
148
70
56
8
remove'
66
87
63
-25
93
>99
-34
88
March 23, 1972 to April 5, 1972 at average flow 1.19 mgd.
bApril 7, 1972 to April 30, 1972 at average flow 1.20 mgd.
CJTU
dml/l
24
-------
pH at CCCSD . It also follows that little calcium carbonate would be
available for recalcination from a second stage precipitator at CCCSD .
All of the previously reported data were for periods when underflow solids
were recycled to the preaeration basin. The objective of recycling solids
is improvement of the rate of the precipitation reactions involving calcium,
magnesium and phosphorus . Lime reactor TSS was maintained in the range
of 1000 to 2500 mg/1. To test the effect of solids recycle on the primary over-
flow, solids recycle was discontinued from June 3 to June 22, 1972. The
major performance change observed was in effluent hardness. During this
period of operation at pH 11.0 with a ferric chloride dose of 10 to 16 mg/1,
the hardness increase across the primary averaged 32 mg/1. This compares
to the hardness increase of 6 mg/1 for March 23 through April 5, when under-
flow solids were recycled to hold flocculator solids in the range of 900 to
1500 mg/1 TSS (Table 5-3) . When flocculator solids were increased to 1700
to 3900 mg/1 for the period of April 25 to 30, the hardness increase across
the primary was 16 mg/1. On the basis of hardness data obtained to date,
it appears desirable at CCCSD to recycle underflow solids at a rate which will
maintain lime reactor TSS in the range of 900 to 1500 mg/1 when operating
at pH 11.0.
Nitrification
After primary clarification, the chemical primary effluent is passed directly
to the oxidation-nitrification tanks with no intervening recarbonation stage.
The chief source of carbon dioxide (CC>2) for recarbonation is the CO£
generated in the nitrification process . Carbon dioxide is derived from the
oxidation of both carbon and (through an intermediate step) nitrogen *.
This biologically-produced C02 is usually sufficient for recarbonation without
the need for addition of CC>2 from external sources. When needed, external
carbon dioxide is vaporized from liquid storage and fed directly to the first
bay of the oxidation-nitrification tank .
Primary clarifier performance is controlled to maximize organic removal in
the primary, thereby controlling sludge age in the oxidation-nitrification tank.
In this way, the longer sludge ages (greater than 10 days) required for year-
round nitrification can be attained without separating the carbonaceous oxi-
dation and nitrification stages as is done in the three-sludge
Lime clarification ahead of nitrification also provides a significant buffer
against heavy metals that would be toxic to the nitrifying organisms . High-
lime treatment is one of the most efficient processes for heavy metals removal8-20
Denitrification
A denitrification system was developed at the ATTF which couples an anoxic
residence period for nitrate reduction with an aerobic period for liquid and
25
-------
sludge stabilization1. This process proved to be so efficient for nitrogen
removal that the effluent contains on the average less than 2 mg/1 of total
nitrogen, compared to approximately 30 mg/1 in the raw sewage to the plant.
Results
Performance data for a representative three months of operation are shown in
Table 5-4. Of particular interest is the fact that the 90 percentile performance
level for the various constituents does not vary widely from the median per-
formance level. This provides statistical confirmation of the ATTF system's
operating stability.
Table 5-4 shows that the concentration of organics, as measured by BOD and
organic carbon, is low in both the nitrified and denitrified effluents. Opera-
tion of the activated sludge process for complete nitrification results in high
organic removals. Suspended matter in the nitrified and denitrified effluents
is also exceptionally low. This is attributed in part to the fact that the sludge
has had a high sludge volume index (Table 5-5) . Effluents from sludges
with a high sludge volume index (SVI) are quite commonly very clear. For
instance, Keeter, et al17 found that effluent solids decreased linearly with
SVI up to the point of clarifier failure. The mechanism for this enhanced
clarification is flocculation of dispersed particles in contact with the sludge
as it settles.
Nutrients are effectively removed in the ATTF system. Total nitrogen in the
denitrified effluent averages less than 2 mg/1. More than half of the effluent
nitrogen is organic nitrogen, and a major fraction of that will be removed
by effluent filtration. Total phosphorus is below the 1 mg/1 contractual limit.
Also of interest are the changes in alkalinity which occur through the process
(Table 5-4) . Lime treatment causes a net increase in alkalinity of the raw
sewage by adding hydroxide ion (even though removing phosphorus alkalinity)
Nitrification, which is an acid producing process, causes a decrease in
alkalinity. Denitrification is a base (hydroxide ion) producing process,
which causes another increase in alkalinity. The net effect of all these pro-
cesses is a decrease in the total alkalinity of the water through the process.
2.6
-------
Table 5-4
ATTF Performance Summary, April 16 - July 15, 1972
Constituent
BODg
TOC
SOC
SS
Turbidity
Settleable solids
Organic N
NH3 - N
NO3 - N
NC-2 - N
Total P
Ortho P
TDS
Conductivity
Alkalinity11'1
Qah
Mgh
Temp range
Raw sewage, mg/1
mean
203
122
22
214
13. 5 d
9.86
9.74
583
1230&
215
67.9
84.0
16. 5-23. 0J
median
199
120
23
212
13. Od
9.57
9.90
561
1210S
218
66.0
84.0
-
90%a
235
152
25
295
16. 5 d
11.19
11.24
750
1366^
237
78.5
95.6
-
Chemical primary,
mg/1
mean
57
42
27
26
median
54
41
27
23
12. 8C, 12. Oc.
.084'
24.0
.86
.61
254
155
38.6
-
1 Od
23.8
.59
.46
249
150
34.0
-
9o<;;a
79
55
31
45
23C
.37d
.
28.5
1.85
1.75
293
184
74.0
-
Nitrified effluent,
mg/1
mean
3.6
8.9
5.6
4.5
median
3.5
S.5
5.5
4.0
1.4° 1.3C
.26
.48
27
.015
1.04*
1.00
634
0
.30
27
.013
.72f
.71
636
1226g 1223g
105 106
159k 161k
57. 3k 61. Ok
-
-
90',a
G.S
11.5
6.7
7.S
2.2C
.67
.58
32
.027
2.111
1.80
724
1394g
127
187k
73. Ok
-
Denitrified effluent,
mg/1
mean
3.2
9.5
5.9
4.5
1.4°
1.1
.31
.48
.009
.50
.52
537
1160
183
1741
30.1
16-24J
median
3.0
n.o
Ci.O
4.0
1.3°
1.1
.30
0
.008
.30
.36
551
1147"
187
172
30. 01
-
nor*
4.S
11
U.(i
7.7
l.sc
2.5
.40
.79
.01S
.93
.90
016
1318^
217
199
48. 41
-
90% of observations are equal to or less than stated value.
b
pH 10.2 operation to June 1, pH 11.0 thereafter.
Jackson Turbidity Units
dml/l
June 27 to July 15 only.
April 16 to June 1 only.
Micro rnhos
has CaCOg
April 16 to May 31 only.
^degrees C.
kApril 16 to May 31 only.
June 1 to July 15 only.
27
-------
Table 5-5
Process Parameters for Nitrification and Denitrification
Parameter
Flow, mgd
Residence time, hr
MLVSS, mg/1
vss
SVI, ml/g
SRT, day
oxid or rem Ib/lb
MLVSS/day
TOC rem Ib/lb MLVSS/
day
BOD5 and NOD load lb/
1000 cu ft/day
M/N ratio
Nitrification
.47 ±' .04
6.5 (aeration)
2985 ± 388
83.8
294 ± 57
21. 3 ± 5.2
.038 ± .008
.047± .010
38
Denitrification
.47 ± .04
.8.2 (reactor)
.79 (stabilization)
3115 ± 449
83.6
250 ± 50
56. 5 ± 23. 6a
.13 ± .05
-
-
3,54± 1.07
jMay 1 to June 16 only, on April 16 to 30 and June 17 to July 15 wasting from activated
sludge into the denitrification system.
28
-------
SECTION VI
SLUDGE PRODUCTION AND THICKENING
There are two interfaces between the ATTF liquid processing and sludge
processing systems: one is at the point where waste solids stream enters
the sludge processing sequence, the other is at the point where residual
solids return to the main flow stream (such as in a centrate) . The problem
of residual solids is discussed in later sections. The present section deals
with the quantity and quality of sludge entering the processing sequence.
SLUDGE PRODUCTION BY PRECIPITATION
All sludge processed in the ATTF system is cycled through the primary and
appears in the primary underflow. The sources of sludge include the raw
sewage suspended solids, the solids that are wasted to the chemical primary
from the subsequent biological treatment stages, and, of course, the
inorganic chemical sludges that are precipitated due to lime and iron addition.
The latter are composed of calcium carbonate, magnesium hydroxide, ferric
hydroxide, calcium phosphate compounds, and a host of minor constituents such
as precipitates of heavy metals.
Measured Sludge Production
Sludge production was monitored in the ATTF system by setting the waste
sludge pumping rate at a known (and recorded) rate and by monitoring sludge
total solids by means of three grab samples per day. Sludge production is
tabulated in Table 6-1 for four different coagulation modes tested in the
chemical primary.
Calculated Sludge Production
Sludge production can also be calculated from analyses of influent and
effluent samples using the,data on suspended solids, phosphorus, magnesium,
and calcium, combined with chemical feed data. This was done as follows:
1. Suspended solids (SS) removed in the primary was taken as
the difference between the raw influent and effluent values
(sampled 5 days a week) .
2. Waste activated sludge (WAS) removed in the primary was
calculated from the metered quantities of sludge returned
to the chemical primary. During most of the test periods
shown, activated sludge was wasted to the denitrification
system (to build up biological solids) rather than to the primary.
29
-------
Phosphorus removed in the process was assumed to be
precipitated as hydroxyapatite (CAs (OH) (PO^ 3) .
4. Magnesium removed in the process was assumed to be in
the form of magnesium hydroxide (Mg (OH) 2) . Magnesium
was sampled five days a week.
5. Due to the high pH of coagulation, iron, when added to the
process as ferric chloride (FeC^) , was presum'ed to precipitate
as ferric hydroxide (Fe (OH) 3) rather than as a phosphate compound.
6. Calcium carbonate precipitation was calculated based on a mass
balance around the calcium ion. The equation used was:
CaCO3 precip = Cain + Cadose ' Caout - Cap;
where each component is expressed as calcium carbonate.
Data on calcium in the influent (Caj[n) and in the effluent (Caout)
were available from weekday analysis . Lime dose was measured
daily. The lime requirement for phosphorus was calculated
from the mass of phosphorus removed, assuming that the phos-
phorus was precipitated as hydroxyapatite.
The results of these balances are presented in Table 6-1 along with the ratios
of measured sludge production to calculated production . The maximum
difference between measured and calculated production rates was 15 percent
for the four modes of operation. Further, two modes had less measured pro-
duction than calculated, while in two modes measured production exceeded
calculated production . This would indicate that there is no systematic error
in the method of calculation of the mass of sludge production.
One significant source of error affecting these comparisons is that of measur-
ing actual sludge production. Only three grab samples were taken per day,
and accuracy in the rate of sludge pumping was based on the frequency of
operator attention , which was variable .
Comparisons between the periods are interesting; however, it should be
noted that raw suspended solids varied significantly through the test periods
and affected sludge quantities . Comparisons in sludge quantities for a common
set of of raw sewage characteristics are made in Section XI ,
Since the calculated and measured quantities are in substantial agreement
the engineer can construct with confidence a solids balance for a lime pre-
cipitation system using pilot plant data on the constituents . Carefully con-
ducted jar tests should also yield adequate data provided that the need for a
30
-------
Table 6-1
Summary of Sludge Production for Several Modes of
Chemical Primary Operation
"^~--^_ Operating pH,
— v^Chemical dosea
Parameter — ^~-~«^_
Sludge production
Ib per mil gal
mg/1
Calculated sludge production ,
mg/1
SS
WAS
Ca5(OH)(PO4)3
Mg(OH)2
Fe(OH)3
CaCCL
3
Total
Ratio measured production to
calculated production
pll 11. 5b
Ca(OH) :500
6930
832
182
0
42
34
0
541
799
1.04
pll 11.0°
Ca(OH) :400
£t
6190
743
171
7
37
31
0
419
665
1.12
pll 11.0
Ca(OH)2:400
FeCl314
6125
735
278
0
47
33
9
431
798
0.92
pH 10. 2e
Ca(OH)2:289
FeCl :24
3875
465
204
3
47
12
16
268
550
0.85
Dose in mgA.
bDec 23, 1971 to Feb 11, 1972.
CFeb 13 to March 21, 1972
March 23 to April 5, 1972.
6April 7 to April 30, 1972.
31
-------
reasonable scale-up in chemical dose is recognized. The authors hcve used
both pilot data and jar test results to construct the solids balance for a
number of proposed treatment plants.
Sludge Composition
The calculated distribution between sludge components shown in Table 6-1
can be compared to sludge analyses of grab samples collected during the
evaluation periods (Table 6-2) . The sludge analyses v/ere made in connection
with the centrifuge test work reported in later sections. The samples were
normally taken between 9 am and 5 pm, with most in the afternoon.
Before the comparisons could be made, the computed compositions in Table 6-1
had to be modified to take into account several factors not considered in the
computation. First, while very little activated sludge was wasted to the primary
during any of the periods shown in Table 6-1, activated sludge was wasted
to the primary on those individual days when centrifuge tests were made. This
wasting was in approximate proportion to the levels that will be maintained
in the full-scale plant. As a consequence, approximately 60 mg/1 of waste
activated sludge was in the sludge sampled and not in the constituent balances.
Second, in the constitutent balances it was assumed that all of the effluent SS
were organic in nature. While no specific tests were conducted on the effluent SS
other data (Table 5-4) indicates that about 20 mg/1 of CaCO- was lost over
the weirs. Lastly, in the sludge analyses, organics were taken as volatile
matter corrected for ignition losses related to calcium carbonate arid mag-
nesium hydroxide. To make comparisons to the sludge analyses. SS in the
constituent balance were assumed to be 80 percent volatile. Once these
factors were included in the constituent balance, the fractions could be
computed and compared to the grab sample sludge analyses.
Considering the fact that the sludge samples were grab samples while the
constituent balances were based on composites, there is reasonable agree-
ment between sludge analysis and computations. The grab samples would
reflect transient diurnal fluctuations and daytime conditions which would
normally be expected to differ somewhat.from a 24-hour composite.
Ferric hydroxide was consistently less in the analyses than in calculated
compositions. It is believed that is due to the method of ferric hydroxide
feed; the chemical was fed at a constant rate despite the fact that chemical
primary flows fluctuated diurnally. Grab samples in the afternoon, therefore,
would have lower iron-content than the calculated, composition, as the. latter
would reflect 24 hours of sludge generation.
Biological Sludge Generation
Biological sludge production was monitored in the ATTF^system to provide a
basis for projecting sludge generation in the full-scale "water reclamation plant.
-------
Table 6-2
Comparison of Calculated Sludge Composition to Grab Sample Analysis
Mode of primary operation
and date
p|i 11.5, no Fed,",
Fob 1, 1972
Feb 1
Feh 1
Fob 1
Fob 2
Feb 7
Mean
(Calculated)
pll 11.0, no Fed
Feb 14, 1972
(Calculated)
pll 11.0 with FeCls
April .">, 1072
(Calculated)
pli 10. 2 with Fet'ls
April 24, 1972
April 24
April 27
April 27
.Mean
(Calculated)
Constituent, percent
Organic
25 . 7
2(>.(i
29.2
2s. 4
34.0
30.0
2.S.9
(24.4)
33.0
(2S.O)
31.0
(33.3)
3<>.C>
35.1
32.S
32. C
34.3
(37.4)
Car(OH)(PO.)0
O ~i o
5.x
5.5
5.7
5.9
5.3
7.0
5.9
( 4.3)
*.Q
( 3.1)
6.4
( 5.4)
10.2
10.1
10. S
12.2
10.8
( 7.7)
MR(OH)9
(>.7
0.0
7.15
(i.l
7.9
C.O
0.7
(3.9)
4.0
(4.3)
4.5
(3.8)
1.4
1.6
l.S
1.7
1.7
(1.9)
FeSOfl)^
0
0
0
0
0
0
0
( 0)
0
( 0)
0.4
(1.0)
o.s
2.3
1.0
1.0
1.3
(2.6)
CaCO
«J
55.3
55 . 0
51.5
53.3
46. 5
51.0
52.1
(60.6)
49.0
(55.5)
47.2
(47.9)
40.0
45.4
48.7
49.7
46.0
(40.8)
Balance
6.s
0.9
5.9
li.4
6.3
6.0
C.4
( 6.0)
5.0
( 6.9)
10.5
( f.3)
11.0
5.5
4.9
2.8
6.0
( 9-3)
Table 6-3
Nitrification Sludge Production and Yield
Month,
1972
January
February
March
April
May
June
Mean
Ib BOD
removed
per day
188
158
266
211
243
163
205
j Ib SS
'produced
per daya
89
106
139
156
181
150
139
Yield,
Ibs SS produced
per Ib BOD removed
0.47
0.67
0.52
0.74
0.74
0.92
0.67
alncludes solids wasted, solids lost over weirs, and
changes in sludge inventory.
33
-------
Sludge production during the food processing season (July to October) is
not presented herein because it represented atypical conditions. The cannery
ceased operation in 1972 and no longer will be a contributor to the CCCSD system.
Nitrification Sludge. The net sludge yield in the oxidation-nitrification
stage of the system for six months of operation averaged 0.67 Ib SS per Ib BOD
removed (Table 6-3) . Waste activated sludge production for the full-scale
30-mgd plant is projected in Section XI.
Denitrification Sludge. As has been described in detail elsewhere^, the
denitrification system was initially operated in the conventional manner
(Fig. 6-1) , but operation in this mode did not produce an acceptable effluent.
The system was modified to incorporate an aerated stabilization tank between
the reactor and clarifier. As has been reported, this reduced sludge yields
to the point where the only wasting was a 4 mg/1 loss over the effluent weirs,
compared to 14 mg/1 obtained for the unmodified system or 16 mg/1 predicted
by sludge yield data for other denitrification systems^.
The data previously reported! was reanalyzed to develop sludge yield data
(Table 6-4) . The conventional system had a sludge yield of approximately
0.6 Ib SS per Ib nitrate-N. This is within the range of 0.2 to 0.9 obtained
by others 12 for suspended growth denitrification. After modification, the
sludge yield dropped to 0.07, well below that for the conventional system
or the value of 0.4 suggested by Christensen and Harremoesl2 in their com-
prehensive literature review. Thus, it can be seen that the modified denitri-
fication system produced only 12 to 17 percent of the sludge produced in
conventional denitrification systems. The result of this is that the denitrifi-
cation system has essentially no impact on solids processing in the ATTF
system.
SLUDGE THICKENING IN THE CHEMICAL PRIMARY
The chemical primary has two basic functions: the first is separation of the
solids from the liquid, and the second is thickening of these solids. While
the ATTF solids processing sequence incorporates separate thickening
stages (Fig. 3-2) , considerable thickening takes place in the primary tank
itself.
General Observations
Thickening depends primarily on solids loading on the clarifier bottom,which
can be expressed as Ibs per sq ft per day. The solids loading depends on the
rate of flow and solids content. In turn, solids content depends on the amount
of solids precipitated plus the amount purposely recycled from the underflow.
34
-------
A. CONVENTIONAL DEN ITRI F ICAT ION SYSTEM
METHANOL
NITRIFIED
EFFLUENT
ANOXIC MIXED
DENITRIFICATION REACTOR
T = 50 to 120 min.
Z
-AERATED
NITROGEN STRIPPING
CHANNEL T= 5 min.
DENITRIFIED
EFFLUENT
B. MODIFIED DENITRIFICATION SYSTEM
CO
en
METHANOL
NITRIFIED
EFFLUENT
ANOXIC MIXED
DENITRIFICATION
REACTOR
T = 50 min.
AERATED
STABILIZATION
TANK
T = 50 min.
DENITRIFICATIO
CLARIFIER
DENITRIFIED
EFFLUENT
MILDLY AERATED
PHYSICAL CONDITIONING
CHANNEL
Figure 6-1
Comparisons of Denitrification Systems
-------
Table 6-4
Comparison of Sludge Yields for Conventional and
Modified Devitrification Systems
Denitrification system
and test period
b
Conventional system
March 3 to 13, 1972
b
Modified system
May 2 to June 16, 1972
Nitrate-N
removed
Ib per day
106
104
SS wasted, Ib per day
Effluent
60
15.8
Change in
inventory
a
-8.3
Total
60
7.5
Yield,
Ib SS per Ib
Nitrate-N
.6
.07
Inventory record indicates an increase in inventory by as much as 20 Ibs/day. Record is somewhat erratic and
therefore inventory change is neglected.
bRefer to Fig. 6-1.
Table 6-5
Summary of Chemical Primary Operating Conditions
During Thickening Test Periods
Test period
Dec 1, 1971 to Feb 10, 1972
Feb 11 to Mar 23, 1972
March 25 to April 5, 1972
April 7 to April 26, 1972
April 26 to May 8, 1972
May 9 to May 21, 1972
May 25 to May 31, 1972
June 24 to July 15, 1972
July 15 to Oct 15, 1972
Primary
operating
pH
11.5
11.0
11.0
10.2
10.2
10.2
10.2
11.0
11.0
Waste
activated
sludge
no
yes
no
yes
yes
yes
yes
yes
yes
Supplemental
ferric chloride
addition
no
no
yes
no
yes
yes
yes
yes
yes
Supplemental
polymer
addition
no
no
no
no
no
yes
yes
no
no
36
-------
Solids content was routinely monitored by means of composited grab sample?
taken from the reactor (preaeration tank) . Solids loading cculd then be
calculated using the reactor solids level and flow data. Fig, 6-2 shows the
relationship between flow and reactor solids level. This figure proved useful
in interpreting the thickening data.
It was found that the inventory of sludge in the primary sedimentation tank
varied diurnally The amount of sludge in the tank was at a minimum early
in the morning, increasing to a maximum in the afternoon. Fig. 6-3 shows
two typical sludge profiles. The change in inventory is due to the fact that
sludge production varies diurnally, while the rate of sludge pumping was held
constant. On some occasions there would be no sludge at sampling station
No. 2 in the morning, resulting in very thin sludge in the underflow.
From the surface of the tank it could be seen that the sludge settled out very
rapidly. The sludge profiles,confirmed that the bulk of the sludge settled
out in the first three-eighths of the primary tank. Sludge was seldom found
past Station 5 unless thickening failure had occurred.
Whenever sludge intruded to Station 6 or beyond, under the effluent weirs,
large floes immediately began appearing in the effluent. As long as the sludge
layer was held in the first half of the tank.no large floes appeared. This
factor helps explain the observed stability of these clarifiers as a function of
flow rate. 1
Expansion of the'sludge layer into the effluent end of the primary was taken
as evidence of thickening failure, since no significant increase in underflow
solids concentration occurred at the time of the layer expansion. On this
basis, it is estimated that only 50 percent of the tank floor is effective in
thickening. Since the calculated solids loadings (Fig. 6-2) always included
the total floor area, actual solids loadings are at least twice those calculated.
Observation of sludge thickening in the primary was complicated by several
factors. These were:
I
1. Limiting solids loadings are difficult to assess. To establish actual
limiting loadings, a failure in the thickening must occur. Up
to the point of thickening failure, the TS attained in the underflow
is determined by the rate of pumping, i.e., by strictly a
dilution relationship. Therefore, unless thickening failure
occurs, daily measurements of underflow solids and associ-
ated computed solids loadings do not necessarily provide
useful information about thickening capability.
2. Because sludge was pumped at a constant rate, composite
underflow samples included thin night and early morning
samples which diluted the heavier afternoon and evening
samples.
37
-------
500O
4OOO
Oi
6
3000
-J
O
V)
2OOO
o
O
o
tooo
0
O
J L
f 2
FLOW, MOD
I I
500 tOOO
OVERFLOW RATE, GPD PER SO FT
15OO
Figure 6-2
Clarifier Solids Loading as a Function of Flow and Flocculator Solids Level
38
-------
EFFLUENT WEIRS
CHEMICAL PRIMARY TANK
SLUDGE INTERFACE AT 1700
SLUDGE HOPPER
110 FT
h-
LL.
o
SAMPLING
STATION
Figure 6-3
Sludge Profile on October 24, 1972
39
-------
3. Operation of the primary was dictated by an experimental
program which was not always compatible with the study
of sludge thickening.
Thickening data for the primary clarifier are summarized according to opera-
ting period in the following sections. Table 6-5 summarizes operating condi-
tions in the chemical primary for these periods.
December 1, 1971 to February 10, 1972, pH 11.5
During this period no thickening failures were observed, and limiting
solids loadings could therefore not be definitely established. Maximum
attained underflow concentrations are listed:
Solids loading Underflow TS,
Ib per sq ft per day percent
10 8.2
7 9
At about 7 percent TS and above, coning and bridging occurred in.the sludge
hoppers, and septic sludge zones developed. With steeper hoppers and better
agitation this should be less of a problem.
February 1 to March 23, 1972, pH 11.0
There was no thickening failure during this period. Maximum attained under-
flow concentrations are listed:
Solids loading Underflow TS,
Ib per sq ft per day percent
18 6.6
16 8.0
7 9.9
March 23 to April 5, 1972, pH 11.0
Sludge profile monitoring during this period indicated that the sludge was
pumped thin. No failure in thickening occurred. At 8 Ibs per sq ft per day
solids loading, up to 4.8 percent TS was attained.
April 7 to April 26, 1972, pH 10.2
Thickening failure occurred when the underflow waste rate was adjusted
to a rate inconsistent with the level to which the sludge could be thickened.
The limiting total solids level appeared to be 4.2 percent. This sludge could
be thickened to 4.2 percent up to at least a solids loading of 14 Ib per sq ft
per day.
40
-------
April 26 to May 8, 1972, pH 10.2
No thickening failure was observed during this period. At 12 Ib per sq ft per
day, the underflow solids could be thickened up to 3.9 percent TS.
May 9 to May 21, 1972, pH 10.2
Use of a polymer, ICI Atlasep 3A3 at a dose of 0.25 mg/1, yielded improve-
ments in sludge thickening over previous operation at pH 10.2. At a solids
loading of 16 Ib per sq ft per day, up to 6.0 percent TS was attained. No
thickening failure was observed.
May 25 to May 31, 1972, pH 10.2
Use of an alternate polymer. Dow A-23 at 0.29 mg/1, allowed solids loadings
to be increased to 29 Ib per sq ft per day at underflow TS of 4.3 percent. While
the Dow A-23 was not evaluated at the same solids loadings as the Atlasep 3A3,
it did not appear to be as effective.
June 1 to June 23, 1972
During this period underflow was purposely pumped thin and no significant
observations on thickening could be made.
June 24 to July 15, 1972, pH 11.0
Sludge during this period could not be thickened as well as in previous comparable
periods (e.g. , Feb 11 to Mar 23 or Mar 23 to April 15, 1972) . Based on
observations of sludge profiles at thickening failure, these limiting levels
are estimated:
Solids loadings Underflow TS,
Ib per sq ft per day percent
20 3
14 ' 4.3
The main difference between this period and the similar previous periods
was sewage temperature. During this period sewage temperature was between
20 and 24 C, while in the two previous periods the temperature was usually
between 15 and 18 C. It is possible that the chemical sludges do not thicken
as well at higher temperatures. This effect may be due to increased biological
activity in the sludge (see next section) .
July 15 to October 15, 1972, pH 11.0
During this period a cannery processing peaches and tomatos was contributing
to the system. Thickening in the primary during this period was quite erratic.
41
-------
Biological action formed gas bubbles in the solids very quickly, which
tended to make the solids less compact. Underflow solids were usually in
the range of 2.5 to 4.5 percent TS. Sewage temperatures ranged from 20
to 24 C.
Under these conditions sludge could be thickened to 4.0 percent at 20 Ib per
sq ft per day. There were a few days when sludge could be thickened to
5.0 percent at even higher loadings.
Summary
In general, sludges with a high pH (11.0 and 11.5) thickened better than the
sludge generated at pH 10.2. However, thickening of the latter sludge could
be significantly improved by the use of a small dose (0.25 ppm) of anionic
polymer. During the summer, under elevated temperatures, biological activity
caused "gassing" of the sludge and thickening deteriorated.
42
-------
SECTION VII
WET CLASSIFICATION OF COMBINED SLUDGES
Most of the lime used for raw sewage coagulation in the ATTF system is precipi-
tated as calcium carbonate. The amount of new lime required in the ATTF
system can be reduced by recovering, recalcining, and reusing the precipi-
tated calcium carbonate. Recalcining can be readily accomplished in a multiple
hearth furnace.
The problem in recovering lime from raw sewage sludge is that many consti-
tuents are precipitated with the sludge besides calcium carbonate. As
explained in Section VI, organics, phosphorus, magnesium, iron, and other
constituents are coprecipitated with calcium carbonate. These constituents,
if left with the calcium carbonate during recalcining, would be returned to
the process. Eventually, these "inerts" would build up to such a magnitude
that the solids processing facilities would be overwhelmed. This prob-
lem has been the major reason that lime recovery in the primary stage of
application has not been practiced ^. The problem of lime recovery has also
been a deterrent to the use of lime in raw sewage coagulation.
The system developed for lime recovery employs a centrifuge for separation of
the calcium carbonate from the other sludge constituents. This process is
termed "wet classification." The fundamental mechanism for classification
is based on the fact that the calcium carbonate settles more rapidly than any
of the other sludge constituents. By purposely operating the centrifuge
at incomplete solids capture, the calcium carbonate can be preferentially
recovered in the cake while the other solids are preferentially rejected to
the centrate. The investigation of wet classification was concerned with the
operating variables to optimize the separation of calcium carbonate from the
other solids,
A second aspect of the investigation which is integrally related to classification
was the dewatering of the separated calcium carbonate. Efforts were directed
toward reducing the water content of the calcium carbonate-rich sludge to
minimize the energy requirements for recalcination.
Solid bowl centrifuges manufactured by Sharpies division of Pennwalt Corpora-
tion were used for studying wet classification. Two machines were used, the
P600 and P3000. The P600 has a maximum capacity of approximately 10 gpm
of feed, while the P3000 can process a maximum of about 50 gpm of feed.
Runs were made under various operating conditions to obtain data for designing
the solids handling system for the CCCSD Water Reclamation plant. The experi-
43
-------
mental design was oriented tov/ards flexibility of operation so that several
variables couki be changed in the experiments at once. Thus the modes
tested were based on an ongoing evaluation of the test results and on the
investigator's broad experience in centrifuge operation.
Under each mode of operation the centrifuge was operated sufficiently long to
assure a steady state condition. Since the residence time in a centrifuge is
shore, a steady state condition when operating to classify the sludge is
attained in five to ten minutes. After reaching a steady state condition, the
pri.T.ary sludge (centrifuge feed) , the water-rich effluent (centrate) and the
solids concentrate (cake) were sampled. The analytical methods useci for
analyzing the samples are described in detail in Section IV.
The recovery of total solids was calculated by material balance using the
follow i n g eq u a tior. :
100
t ct - et
where:
Rt = recovery of total solids , percent
ct = total solids in cake, percent
et = total solids in centrate, percent
i"+ = total solids in feed, percent
The term _ " gives the fraction of total feed stream that is separated
H et '
as the calie stream. Multiplying the concentration of solids in the cake
by the fraction separated as cake gives the amount of solids recovered in
the cake stream for a unit of feed . Dividing the amount of solids in the cake
by the amount of solids in a unit of feed gives the fractional recovery of
solids. Recovery of total solids was reported as percent.
After calculating the total solids recovery, the recovery of an individual
constituent can be calculated by the following equation:
fc
.vnere:
Rp = recovery of a given constituent, percent
R{ = recovery of total solids, percent
cc = constituent in total solids of cake, percent
f,^ = constituent in total solids of feed, percent
44
-------
CONSTITUENT SEPARATION
The purpose of wet classification is to maximize recovery of calcium carbonate
while rejecting the other components to as great an extent as possible. The
factors affecting the classification of the solid components are discussed below .
A subsequent section of the report discusses dewatering of the calcium car-
bonate-rich cake.
Separation Efficiency
The calcium carbonate crystals settle more rapidly than the other solids. Conse-
quently , a partial recovery of the total solids with a centrifuge concentrates
the calcium carbonate in the cake. To get an indication of the efficiency of
component separation, the recovery of each component was compared with
the total solids recovery. The plots are shown in Fig. 7-1 for all of the runs
listed in Table 7-1.
There is a considerable scatter of data resulting from different degrees of separa-
tion and from experimental error. However, trend lines can be drawn through
the data points when it is recognized that two boundary conditions must be
satisfied. These are:
1. At 100 percent recovery of total solids, all constituents must
be completely removed.
2. At 0 percent recovery of total solids, no individual constituent
can be recovered.
These recovery curves are summarized in Fig. 7-2. From the figure, it can be
observed that there is a fairly broad range in which fairly high calcium
carbonate recovery is coupled with reasonable classification. For instance,
at 50 percent total solids recovery, 70 percent of the calcium carbonate is
recovered with less than 30 percent>of the other major constituents, whije at
70 percent total solids recovery, almost 90 percent of the calcium caibonate
is recovered with less than 50 percent of the other major constituents.
Centrifuge Variables
Besides demonstrating the feasibility of wet classification, the data prosc-nted in
Fig. 7-1 show the wide range in degree of separation attained. The variations
in separation are due primarily to differences in centrifuge operating condi-
tions. The principal variables are feed rate, centrifugal force, and pond set-
ting.
45
-------
Table 7-1
Run Data for Wet Classification
Process Parameters
Run
2-7
2-5
2-1
2-4
2-32
2-12°
8-1
8-4
8-12
8-9
8-8
8-5
,8-41
8-40
4-3-72
4-134
4-106
4-105
4-123b
Machine
Sharpies
P-600
Sharpies
P-3000
Sharpies
P-600
Centrifugal
force
per Ib
mass, g's
2100
1050
2100
2100
1500
1500
3050
2100
Auger
speed,
ARPM
50
35
1G
26
65
50
9
Auger
pitch,
in.
2
3
1
Pond
number
1
3
1
3
1
2
3
1
1
4
Feed
rate,
gpm
8
10
10
12
9
3
23
49
25
49
21
49
19
43
10
5
7
14
4
Floccula-
tion
pll
11.5
11.0
10.2
Recoveries in Cake, percent
Total
solids
67
G3
69
G5
Uli
86
(>4
5G
G9
47
72
64
61
55
62
57
48
49
97
Calcium
carbonate
90
8G
90
78
84
n, 100
96
89
88
57
9G
75
85
76
87
79
69
77
nr 100
Magnesium
hydroxide
30
27
43
35
GG
51
44
34
64
31
58
45
36
32
30
38
21
32
97
Hydroxy-
apatite
23
20
25
24
18
70
10
11
30
13
34
41
11
9
23
23
18
16
89
Ferric
hydroxide
31
23
6
12
87
Organic
matter
41
41
43
49
59
65
50
42
43
32
56
55
42
35
38
41
33
35
91
Other
materials
57
51
57
87
44
60
66
60
67
50
66
66
76
74
56
41
52
30
~100
0)
Three pounds of Allied Chemical colloid 726 polymer added per ton of dry solids in the feed.
Two and a half pounds of ICI America Atlasep 2A2 polymer added per ton of dry solids in feed.
-------
100
80-
Magnesium Hydroxide
Calcium Carbonate
Hydroxyapatite
Ferric Hydroxide
40 60 SO IOO
I I
Other Material •
Organic Matter
60 8O 100 0 2O
RECOVERY OF TOTAL SOLIDS, PERCENT
Figure 7-1 Constituent Recovery During Classification
47
too
-------
100
O 2O 40 60 8O tOO
RECOVERY OF TOTAL SOLIDS, PERCENT
Figure 7-2
Summary of Constituent Recoveries During Wet Classification
48
-------
F_ee_d_Rale. One of the most important parameters in the depi'1:- t: fiquipr.itv. lr:
the feed rate and the effect of variations in feed rate. For tunctcJy, vide
variation in feed rate cur: be tolerated for the purpose of ccr.siiluor.t t>-pr;rr--
tioiis . Examination ot the data in Table 7-] reveals that for runr: made ur,.->r
the.sarne conditions, except, for feed rate, there was only a slight ucto:ic:v-
tion in constituent separation with an increase in feed rate.
2£ni£yH.?lal.Z.9I££- 'rvnolner important variable in centrifuge op-,-ratior it- the
centrifugal force. Based on experience of Sharpies personnel, most of the runs
were made at a centrifugal force per unit mass of 2100 timer f!:e nccx.lei ^ir-i .-,-"
gravity (g) . However, a few runs were made at different y ]ovo!s. O\ c-r
the range of about 1000 to 3000 g, recovery of calcium carbonate increased only
slightly with an increase in centrifugal force. For ciewateriu? cf the high
calcium carbonate cake, as will be discussed in a subsequent ration ot tnir.
section, centrifugal force, is an important operating parameter.
Pond__^ettin_g . The depth of the liquid layer in the centrifuce bowl would be ex-
pected to be an important variable. Increasing the depth of the liquid (incro
pond number) definitely had an adverse effect on the separation of constituents.
The best constituent separation was always attained at the lov, esl pond set-
ting. For example, runs 8-4 and 8-5 (Table 7-1) held all process parr.meters
constant except pond setting. The pond setting of 1 as compared to 3 gave a
higher recovery of calcium carbonate and a greater rejection cf all other
components .
Lime Dosage, or jp_H
There is some indication that calcium carbonate is more easily separated from the
Other components at a high pH than at a low pH. However, other variables
were also changed simultaneously with pH, and the pH effect could therefore
not be firmly established. Runs made at pH of 10.2 and 11.5 were made with
different machine auger s, while runs made at pH 11.0 were made with the
larger of the two machines. Also, the concentration of the various consti-
tuents in the sludge was substantially different at the various pH levels
(refer to Section XI) . and that also could account for the differences in
degree of separation.
Polymer Addition
Two runs were made with polymer addition to increase the capture of solids .
Because of the limited amount of data, it is difficult to determine the effect
of polymer on classification. In run 2-12; the recovery of calcium carbonate
and rejection of the other components were better than would be expected,
based ori the relationship presented in Fig.. 7-2 between total solids recovery
and recovery of individual constituents. For the other run (4-123) a 97 percent
recovery of total solids was obtained, and consequently, the recovery of all
constituents was high , resulting in little classification .
49
-------
Based on results from run 2-12, in which an 86 percent recovery of total
solids was obtained, it appears that a polymer can be-used in wet classifi-
cation to increase both the total solids and calcium carbonate recoveries.
DEWATERING AND RECOVERING SOLIDS
Besides separating the individual constituents of the solids, the dewatering
of the recovered cake must also be considered. Factors which are important
are the centrifuge operating conditions and the lime dosage or pH level used
in the primary treatment from which the sludge is produced.
Centrifuge Variables
As with the separation of solid constituents, the extent of dewatering is affected
by several centrifuge operating variables. The effects of feed rate, centrifugal
force, and pond setting were studied to determine their relative importance.
Feed Rate. As was noted in the discussion of constituent separation, feed
rate can be varied over a wide range with only a minor effect on separation.
Feed rate was also found to have little effect on total solids recovery at all
pH levels (Figs. 7-3, 7-4, and 7-5) . For the smaller of the two machines
tested, the P600, a change in feed rate from 4 to 12 gpm, a threefold change,
resulted in less than a 10 percent decrease in total solids recovery. Approxi-
mately the same effect of feed rate was observed with the larger machine, the
P3000. The cake dryness seemed to improve slightly with an increase in
feed rate, but the improvement was too small to be considered significant.
Centrifugal Force. For constituent separations centrifugal force seemed to
have a slight influence. For dewatering and solids recovery, however, the
tests showed that centrifugal force is an important variable. Increasing the
centrifugal force improved dewatering and increased solids recovery (Figs. 7-3,
7-4, 7-6, and 7-7). A centrifugal force of 2100g appears to be adequate
for dewatering.
Pond Setting. The liquid depth (pond setting) had a minor effect on total
solids recovery but a substantial effect on cake dryness (Fig. 7-6) . For
dewatering, as was previously observed for constituent separation, the
best results were obtained with as shallow a liquid depth as possible (pond
setting No. 1) .
Comparison of Machines. A few runs were made with both the P600 and P3000
during the same time period. The results are presented in Figs. 7-7 and 7-8.
The observed difference in performance of the two machines is possibly due to
differences in augers and auger speeds rather than machine size. The auger
50
-------
too
90
I-
2 BO
o
ft:
Uj
0. 70
ft: 6O
ki
50
-------
en
Uj
i*
O
O
Uj
tt
Q
•J
too
90
BO
7O
60
2O
10
O
SYMBOL
©
0
&
•
A
POND
1
2
3
1
3
FORCE, g
2100
2100
2100
1500
1500
CAKE,%SOLIDS
55.7- 58.1
38.5 - 4 1.6
30.2-39.0
51.3 - 52.5
25.1 - 32.8
ARPM
26
26
26
20
20
Machine: Sharpies P3000
pH r II.0
10
15
20 25 3O
FEED RATE, gpm
35
4O
45
50
Figure 7-4
Solids Recovery at Flocculation pH of 11.0
-------
en
CO
IOO
90
h.
5: 80
Uj
o
I 70
«^
x
Q: 60
ki
ft:
8
50
30
220
O
to
Machine: Sharpies P600
Force, g : 2100
Pond r 1
Auger pitch, in. : I
®
ARPM : 50
Cake, % Solids : 41.2-48.6
J.
23456789
FEED RATE, gpm
Figure 7-5
Solids Recovery at Flocculation pH of 10.2
©
IO II 12 13 14
-------
60
50
Uj
tu
a.
40
Ui
2*:
O 30
to
Q
O 2O
10
13
Machine: Sharpies P 3000
Auger pitch, in.: 3
SYMBOL
©
B
A
•
A
POND
1
2
3
1
3
FORCE ,g
2. 100
2100
2(00
1500
1500
©
O
10
20 30
FEED RATE , gpm
4O
Figure 7-6
Dewatering at Floccuiation pH of 11.0
50
-------
en
en
O
(fe
lu
too
9O -
»
" 8O —
7O -
50
30
10
O
O
L
O
Sharpies Centrifuges
Force , g : 2IOO
pH : II.0
SYMBOL
©
A
•
A
MACHINE
P-3OOO
P-600
P-3OOO
P-600
POND
I
I
3
3
ARPM
26
50
26
5O
PITCH
3 in.
1 in.
3 in.
1 in.
8
P-6OO
1O
12
JL_
14
IO
2O
3O 4O SO
P-30OO
FEED RATE, gpm
60
7O
16
SO
Figure 7-7 Comparison of Centrifuges for Solids Recovery
-------
en
cr>
60
50
LU
o
ft 40
Uj
0.
Uj
30
co
Q
O 20
co
10
0
O
0
O-
<§>
JO.
Sharpies Centrifuges
Force, g -. 2100
pH : 11.0
o
SYMBOL
0
A
•
A
MACHINE
P-3000
P-600
P-3000
P-600
POND
1
1
3
3
ARPM
26
50
26
50
PITCH
3 in.
1 in.
3 in.
1 in.
8
P-600
10
12
14
10
20
30 40 5O
P-3OOO
FEED RATE , gpm
60
70
16
80
Figure 7-8 Comparison of Centrifuges for Dewatering
-------
removes solids from the centrifuge bowl. The studies on second stage opera-
tion (Section VIII) give an indication of the importance of proper auger speed.
A high auger speed relative to the bowl results in excess water being removed
v/itii the settled solids. If the auger speed is too low, there will be an insuffi-
cient rate of solids removal, but the solids that are removed will be relatively
dry.
For the two machines tested, the smaller machine with a higher auger speed pro-
duced a higher total solids recovery but a lower cake dryness than the larger
machine. If the auger speeds of the two machines were comparable, it is
anticipated that sclids recovery and cake dryness would also be comparable.
Lime Dose or pH
As discussed under "constituent separations", the effect of lime dose or pH
could not be established with absolute certainty because other variables were
changed simultaneously. As lime dose or pH in the primary tank increased,
it appeared that solids recovery also increased.
SUMMARY
The calcium carbonate in primary lime sludge can be preferentially separated
from the other constituents in the sludge by wet classification in a centrifuge.
For a flocculation pH of 11.0 and 11.5 approximately 90 percent of the calcium
carbonate was recovered while rejecting 50 to 75 percent of the other solid
constituents. At a flocculation pH of 10.2 the calcium carbonate recovery was
about 75 percent when rejecting 60 percent of the other materials. The
calcium carbonate-rich cake was simultaneously dewatered during wet classifi-
cation to a total solids content of 42 to 57 percent.
To attain a high recovery of calcium carbonate during wet classification and
simultaneously devvater the captured solids, the centrifuge must be designed
and operated to separate the constituents. The best results were obtained
with a low liquid depth in the bowl (pond setting of 13 . Increasing centrifugal
force- improved the dewatering substantially and the calcium carbonate
recovery slightly. A centrifugal force of 2100 g appears to be adequate for
v;et classification and dewatering.
Another variable in centrifuge operation is the auger which removes solids
from the centrifuge bowl. While a study of auger pitch and speed was not
included in this study,the study of second stage centrifuge operation indi-
cates that the auger shape and speed are important variables. For an
operating installation flexibility in adjusting auger speed is considered
desirable to attain the optimum recovery and concentration of calcium car-
bonate under any given set of conditions.
57
-------
The centrifuge can be operated over a wide range of feed rates with only a
small effect on the separation. An increase in feed rate caused a minor
decrease in calcium carbonate recovery; however, only a small change in
separation occurred for a twofold change in feed rate. Feed rate had only a
minor effect on solids recovery and no significant effect on dewatering over
a 3:1 range of feed rates.
In summary, the tests showed that a solid bowl centrifuge can adequately wet
classify and dewater lime sludge.
58
-------
SECTION VIII
CENTRATE PROCESSING BY CENTRIFUGING
In the wet classification of primary sludge, calcium carbonate is concentrated
for recalcining and reuse as lime. The other constituents in the sludge are
preferentially rejected to the centrate. A second centrifuge stage can be used
to recover and dewater the rejected solids from the wet classification. The
centrate from the second centrifuge stage can then be recycled to the primary
treatment section of the plant. To avoid a large buildup of solids in the primary
treatment section, the recycle stream must have a low solids content. Therefore,
a high recovery of solids in the second stage is necessary.
If the recovered solids are to be incinerated, the water content of the solids
must be as low as possible. Efficient dewatering will minimize the fuel require-
ment for incineration. Consequently, in the second centrifuge stage both the
recovery and the dewatering of the solids should be as nearly complete as
possible.
The solids to be captured in the second centrifuge stage are the slow settling
solids which were rejected in the first centrifuge stage. Different conditions
must therefore be used in the second stage than in the first stage. Efficient
capture requires a high centrifugal force and the addition of a polymer to
improve the settling characteristics. The factors affecting the capture and
dewatering of solids are discussed in this section.
The centrate from the wet classification of primary sludge was accumulated in
a storage tank and fed to a centrifuge to study the capture and dewatering of
solids in first stage centrate. Most of the study was made with Sharpies
centrifuges, either the P600 or the P3000. A few runs were made with a 6-inch
Bird unit. Samples of feed, centrate and cake were analyzed and recoveries
calculated as described in Section VII. Results of all the centrifuge runs on
first stage centrate are presented in Table 8-1.
CONSTITUENT CLASSIFICATION
In wet classification calcium carbonate is preferentially captured, and the
other solid constituents are preferentially rejected to the centrate. In the
second centrifuge stage a high capture of solids with little or no discrimination
between constituents is desirable. The test results (Table 8-1) showed that
some classification of constituents seems to be unavoidable, although the
degree of separation was much less than for the first stage. The highest
59
-------
Table 8-1
Constituent Separation During Centrifuging of First Stage Centrate
Process Parameters
Rim
2-21
2-54
2-4-1
2-52°
4-1
4-2
3-1
3-fi
3-12
3-1 li
4-11K
Average
of all runs
Machine
Sharpies
P-liOO
Bird
li in. bowl
Sharpies
P-GOO
Centrifugal
force per
Ib mass,
K's
3050
2100
Anger
speed,
ARPM
12
24
IK
: 2-1
3050
10
10
2090 nad
2100
30
Auger
pitch,
in.
1
nad
1
Pond
numlier
4
3-1/2
3
na"
3
Feed
rate,
Kpm
2.4
3.4
2.0
3.4
2.0
2.5
2.2
2.2
2.0
2.0
3.3
Floccula-
tion
pll
11.5
11,0
10.2
Polymer
type"
725
2A2
Polymer
dose
lb/tonfc
«,5
3.3
3.5
3.3
H.8
i;.i>
(i.2
7.5
9.5
7.0
5.0
Centrifuge Performance
Dewate r-
ing
Total
solids
in cake,
percent
15.4
14.5
14.4
13.7
14.8
15.9
12.1
11.5
14.6
17.7
18. li
14.8
Recoveries in Cake, weight percent
Total
solids
G9
87
80
81
80
78
89
93
87
81
85
83
Calcium
carbonate
82
99
93
~100
-100
~100
~100
/vino
•vlOO
~100
95
97
Magnesium
hydroxide
84
79
100
66
88
77
99
~100
~100
'VlOO
85
89
Hydroxy-
apatite
69
87
80
81
79
80
94
99
85
68
94
83
Ferric
hydroxide
80
78
96
Organic
matter
63
78
71
71
71
67
79
83
69
62
89
73
Other
materials
52
99
-vlOO
81
74
70
89
80
89
86
(24)
77
o*
o
Allied polymer colloid 725 or ICI America polymer Atlasep 2A2.
Pounds polymer per ton of dry solids in feed.
'"Feed consisted of 11 percent sludge and 89 percent first stage centrate.
Information not available.
-------
recoveries were obtained for calcium carbonate, wnile the; JO.VCS-L laccvrr^s
were for organic matter and the constituents designated as ''other fr.atf-nal"
However, the extent of classification between constituents v.as i'elativcly
small. Consequently, variations in operating conditions hod or.ly a minor
affect on the relative recovery of individual'constituents, and the rr.ajoi con-
cern was the dewatering and caoture of total solids.
SOLIDS RECOVERY AND DHWATERING
As mentioned previously a high efficiency in both capture and dewaterinc; of
solids is desirable. Efficient capture will minimize the recycle of solics to
the sewage treatment facilities, and efficient aewatering will minimize the
heat requirements for incineration. Since a polymer is required to coagulate
the small particles and improve their settling characteristics, one of the more
important considerations is the effect of centrifuge operating conditions or.
the amount of polymer required to attain a given separation. While a thorough
study to define quantitatively the effect of each variable is beyond the scope
of this study, sufficient data were obtained to establish the approximate
efficiency of solids capture and dewatering and the relative importance ot the
operating variables.
Polymer Type
Several polymers are available for flocculating solids in a centrifuge. Since
the type of polymer has a marked effect on the degree of separation, a study
of various polymers was made to identify one that could be used satisfactorily.
Several polymers were given cursory examination in the laboratory with a
beaker test. The test consisted of adding the polymer to first Stage centrate
in a beaker and gently stirring the mixture. The effectiveness of the polymer
was judged by visual observation of the floe.
The results of the beaker tests are shown in Table 8-2 for a centrate from
a primary flocculation pH of 10.2. A second series of tests was run for a
primary flocculation pH of 11.5 but deleting Diamond Shamrock 930 polymer
and adding American Cyanamid Magnafloc 835A and Calgon 2700. Similar
results were obtained from both tests. In both eases ICI Atlasep 2A2 and
Allied Colloid 725 produced the best floes.
The two polymers that appeared best by the beaker test were further evalu-
ated in centrifuge tests. The results are shown in Fig, 8-1s ICI Atiasep
2A2 gave the best results both in the beaker test and the centrifuge runs and
was therefore used for all subsequent studies of the second centrifuge stage.
61
-------
Table 8-2
Laboratory Beaker Test of Polymers
for Second Stage Centrifuge Feed
Polymer
Size of Floe
a
ICI Atlasep 2A2
Allied Colloid 725
Hercules 836.2
ICI Atlasep 3A3
Allied Colloid 726
Dow A-23
Diamond Shamrock 930
large
medium
small
very small
very small
very small
very small
Centrate used in the tests was produced
from a primary flocculation pH of 10.2.
Polymer dosage in each case was 5 Ib
per ton dry solids.
62
-------
O)
00
IOO
I-
2 SO
O
Q:
loj
Q.
X
a: 6O
Uj
^
o
o
Uj
to 4O
Q
-J
O
CO
20
1 1 r
Machine: Sharpies P 600
Centrifugal force, g : 3050
Auger pitch , in. : 1
Auger speed ARPM ; 20
Feed rate, gpm : 4.3
Feed, % Solids : 1.2
F locculation pH : 11.5
Polymers :
ICI Atlasep 2A2
Allied Colloid 725
SYMBOL
©
A
POLYMER
2A2
725
CAKE, % SOLIDS
11.6- 11.9
11.8- 12.0
J.
123456
POLYMER DOSAGE, LB/TON DRY SOLIDS
Figure 8-1
Effect of Polymer Type on Separation in Centrifuge
8
-------
Centrifuge Operating Variables
It was beyond the scope of the present study to establish the quantitative
effect of the many interdependent variables in centrifuge operaiion. The
study was restricted to obtaining data for selection of prototype equipment
sizes and for establishing the areas where operating flexibility is necessaiy
Auger Speed. Since the solids separated by centrifugal force are removed
from the bowl with an auger, the auger speed relative to the bowl (A RPiVi)
is one of the important operating variables. An example of the effect of auger
speed is shown in Fig. 8-2. The polymer requirement to attain an 80 percent
recovery of solids was reduced to about half by increasing the auger speed
from 8 to 12 A RPM. However, the extent of dewatering was reduced by
the increased auger speed. Apparently there is an optimum auger speed that
gives the best balance between polymer usage and extent of dewatering for
a given feed and set of operating conditions.
Pond Setting. Several factors are involved when the liquid depth is changed.
Increasing the depth increases the liquid residence time, which should improve
the separation. Offsetting to an extent the increased residence time is a reduced
centrifugal force at the liquid surface due to the fact that the surface is nearer
the hub of the revolving bowl. Consequently, the average centrifugal force
is reduced. Increasing pond depth also increases solids removal efficiency
As the depth is increased the solids are more easily "lifted over" the beach
of the bowl by the auger. For the second centrifuge stage Fig. 8-3 shows
that increasing the liquid depth reduced the required polymer dosage to attain
a given recovery, but also reduced the solids content of the cake.
On occasions difficulty was experienced in getting the solids removed from the
bowl at a pond setting of less than four, particularly at low polymer dosages.
The solids would tend to build up in the bowl and then "dump11, making an
unstable operating condition.
Centrifugal Force. Recovery would normally be expected to improve with an
increase in centrifugal force, with the only limitation being the equipment and
operating cost relative to the improvement in degree of recovery As shown
in Fig. 8-4, the solids recovery is improved as the centrifugal force is increased.
However, on the Sharpies P3000 machine no improvement in recovery was
obtained by increasing the centrifugal force above 2,100 g. The cake dryness
did continue to improve, but the improvement from 2,100 to 3,200 g wa£ small.
Apparently there is little advantage in applying a centrifugal force greater
than 2,100 g.
Feed Rate. In the centrifuge test runs shown in Fig. 8-5 the feed rate was
varied from 14 to 29 gpm with no significant change in polymer requirement
or percent of solids recovered. There was an apparent slight increase in cake
64
-------
too
CT)
en
SI
Ul
O
tfc
Uj
Q.
, 60
Uj
>
O
O
O)
Q
O
0)
20
0
0
Machine: Sharpies P 3000
Centrifugal force, g: 2100
Auger pitch , in. : 3
Polymer: JCI Atlasep 2A2
SYMBOL
©
A
pH
11.0
11.0
FEED RATE
gpm
16.1
16.8
POND
4
4
AUGER
ARPM
12
8
FEED,
% SOLIDS
1.7
2.5
CAKE ,
% SOLIDS
10.9- 13,5
15.2- 15.8
23456
POLYMER DOSAGE, LB/TON DRY SOLIDS
Figure 8-2
Effect of Auger Speed on Polymer Requirement
-------
IOO
O)
01
8O
Uj
o
Q:
X
Qc
Ul
>.
O
O
to
Q
O
CO
6O
20
-e-
Mochine: Sharpies P 600
Centrifugal force, g: 2100
Auger pitch, in. : 1
Polymer: ICJ Atlasep 2A2
SYMBOL
O
A
•
A
pH
10.2
10.2
1 1.5
\ 1.5
FEED RATE
gpm
2.2
2.3
4.5
4.5
AUGER
ARPM
50
60
30
30
POND
3
4
3'/2
4
FEED,
% SOLIDS
1.9
1.9
2.1
2.1
CAKE,
% SOLIDS
16. 3-19.1
13. 7-15.1
11.8
11.7- 12.8
JL
23456
POLYMER DOSAGE, LB/TON DRY SOLIDS
Figure 8-3
Effect of Pond Setting on Recovery of Solids
8
-------
100
cr>
BO
UJ
o
EC
lu
a.
ct
Uj
>
o
o
lu
CO
Q
O
CO
6O
4O
20
Auger pitch, in, : 3 for P-3000
I for P-600
Polymer: ICI Atlasep 2A2
SYM'
BOL
0
A
m
A
•
pH
11.0
11.0
11.O
11.2
11.2
MACHINE
P-3000
P-3000
P-30OO
P-600
P-6OO
FEED RATE,
gpm
30.7
29.0
30.7
3.5
3.5
POND
4
4
4
3'/2
3l/2
AUGER
ARPM
18
11
16
10
IO
FORCE,
g
320O
2100
I50O
3050
2IOO
CAKE,
% SOLIDS
16.0- 17.1
15.5- 17.4
1 1.4- 11.8
12. 8-13. 8
1 1.4-12.8
I
23456
POLYMER DOSAGE, LB/TON DRY SOLIDS
8
Figure 8-4 Effect of Centrifugal Force on Recovery
-------
IOO
en
oo
80
UJ
o
Uj
\eo
X
ft:
Uj
o
o
!y ^o
O
CO
20
Machine: Sharpies P 3000
Pond:4
Auger pitch, in. : 3
Centrifugal force, g: 2100
Polymer: Atlasep 2A2
SYMBOL
©
&
A
FEED RATE
gpm
29.0
14.0
1 1.0
AUGER
ARPM
It
6
4
FEED,
% SOL IDS
1.6
1.5
1.6
CAKE,
% SOLIDS
15.5-17.4
16.8- 19.0
16.4-18.0
_L
_L
J-
O
23456
POLYMER DOSAGE, LB / TON DRY SOLIDS
8
Figure 8-5 Effect of Feed Rate on Recovery
-------
water content at the higher feed rate. At the low feed rate of 11 gpm, about
a third of the maximum rate used, solids recovery was slighly better, and
the polymer requirement was reduced. Fig. 8-5 shows that a wide range
of feed rates can be used without changing the efficiency to any great extent.
For the runs at different feed rates the auger speed was also varied to maintain
a relatively constant ratio of feed rate to auger speed. Since the concentration
of solids in the feed was essentially constant, the amount of solids removed
per revolution of the auger was also maintained constant, thereby essentially
eliminating the effect of auger speed on cake dryness.
\
Comparison of Machines. Two different sizes of Sharpies centrifuges were
used for the experimental work. To estimate the effect of machine size on
solids recovery, all of the data for the two machines for similar feed rates
and at similar centrifugal forces were plotted in Fig. 8-6. The figure shows
that the larger machine, the P3000, gave better solids recovery with a lower
polymer dose than the smaller machine, the P600.
Effect of Feed pH on Centrifuge Performance
Since pH affects the composition of the solids in sludge, it was postulated
that the ease of separation of the solids from the liquid might also be affected.
A study was therefore made of the effect of varying the pH of the feed to the
centrifuge for both the first and second centrifuge stages. The study con-
sisted of two phases. The first phase evaluated the effect of changes in the
pH of flocculation, which over the course of the experimental work was varied
from 10.2 to 11.5. The second phase used sludge from a constant flocculation
pH and evaluated the effect of adjusting the pH of the feed to the first and
second centrifuge stages.
Flocculation pH. It is difficult to identify the effect of flocculation pH, be-
cause the study extended over a period of several months and encompassed
several changes in operating conditions. There is also the problem that the
centrifuge work was performed to obtain design data rather than to define
the effect of all operating variables such as pH. Consequently, changes were
made in centrifuge sizes and operating conditions without regard to flocculation
pH.
All of the data on centrifuge operation were reviewed, and specific runs were
selected which appeared to offer a valid basis for comparison. Since no two
runs at different flocculation pH levels were made under the same operating
conditions, a direct..comparison is impossible. However, it was established
that solids recovery varies with the solids removal rate, which is a function
of auger speed and solids feed rate. A comparison was therefore made between
runs at different flocculation pH levels but a constant ratio of solids feed rate
to auger speed. The results are shown in Fig. 8-7 for different levels of
centrifugal force. The figure does not show any relationship between ficccu-
69
-------
V]
O
IOO
8O
iu
o
ft:
UJ
Q.
**
X
ft:
Uj
o
o
ft:
co
o
co
6O
4O
2O
O
*•>
©
0
A
0
0 © A
0 r^
A
A
A
A
©
A
A
F locculation pH : 11.0
SYMBOL
A
©
•
MACHINE
P600
P 3000
P 3000
FORCE, g
2600
2100
32OO
FEED RATE
gpm
2.0-4.6
it-29
11-31
CAKE,
% SOL IDS
15. 3-18. 4
15. 5-19.0
15. 7-19.6
I
23456
POLYMER DOSAGE, LB/TON DRY SOLIDS
8
Figure 8-6 Comparison of Two Sharpies Machines for Second Stage Operation
-------
100
80
©
Ul
o
1C
X
it
UJ
>
o
o
O
Q
-J
O
Machine: Sharpies P-600
Polymer: ICI Atlasep 2A2
EJ
ZO
A
SYMBOL
©
A
E3
pH
10.2
1 1.0
11.5
FEED
RATE,
gpm
3.3
2.0
2.0
AUGER
ARPM
30
11
16
FEED,
%
SOL IDS
1.9
1.7
1.8
Lb Solids
in Feed per
Auger Revolution
O.OI8
O.O26
O.OI9
POND
3
3'/2
3'/2
FORCE,
g
2100
26OO
2IOO
I
I
23456
POLYMER DOSAGE, LB/TON DRY SOLIDS
100
1
1
Machine: Sharpies P-600
Polymer : ICI Atlasep 2A2
0
/2
FORCE,
g
3050
3050
26OO
260O
3050
3O50
FEED,
% SOL IDS
.8
.9
.7
.7
.2
.4
Lb Solids in Feed
per Auger Revolution
0.035
0.035
0.033
0.041
O.030
O.04I
O
23456
POLYMER DOSAGE, LB/TON DRY SOLIDS
Figure 8-7A, B
Effect of Flocculation pH on Solids Recovery
71
-------
lation pH and solids recovery. The observed differences in recovery could
well have been due to other differences in operating conditions.
Flocculation pH values greater than 11.0 had a detrimental effect on second
stage cake water content, as shown in Table 8-3.
Adjustment of Centrifuge Feed pH. Some lime sludge processing flow sheets
include pH adjustment of the sludge with carbon dioxide ("recarbonation")
prior to sludge processing. The purpose of recarbonation is to dissolve
the magnesium hydroxide fraction which may hinder dewatering if left in the
sludge 21.
Sludge generated at a flocculation pH of 11.0 was adjusted to various levels to
assess the effects on centrifuge performance. On some occasions a pH drop
was accomplished by bacterial action in the primary, on most occasions carbon
dioxide was used. When the pH was raised, lime was used.
Solids recovery in the first centrifuge stage varied between 55 and 58 percent.
Cake solids varied from 53 to 60 percent, with the wettest cake from a feed pH
of 10.5. The observed variations are probably all within the range of experi-
mental error, and it was concluded that the effect of pH on first stage separa-
tion is negligible.
The data for the second stage centrifuge are presented in Fig. 8-8. Several
combinations of first stage feed pH and second stage feed pH are indicated.
The following observations can be drawn from the data:
1. Adjusting the feed to the first stage centrifuge to a pH greater
than 11.0 makes dewatering in the second stage more difficult.
A much higher polymer dosage was required for solids capture
for first stage feeds adjusted to pH 11.5 than for lower pH
sludges.
2. Adjustment of the feed to the second stage centrifuge to a pH
greater than 11.0 seems to have no detrimental or beneficial
effect.
3. In general, there appears, to be little or no benefit derived from
recarbonaticn of the first or second stage feed sludge.
Effect of pH Adjustmenton Composition of Solids. Adjusting the1 pH of the
centrifuge feeds was expected to change the solubility of the constituents and,
consequently, change the composition of the solids. To estimate the extent of
change in solids composition the soluble calcium, magnesium, and phosphorus
in the feed and centrate from several first stage centrifuge runs were deter-
inined. The results are given in Figs. 8-9, 8-10 and 8-11. The change in
72
-------
Table 8-3 Effect of Flocculation pH on Second Stage Cake Dryness
r\TJ
pH
11.2 to 11.5
11.0
10.2
Total solids in cake, percent
Range
11.0 to 14.4
15.5 to 19.6
13.7 to 20.3
Median
11.8
17.2
18.4
Table 8-4 Effect of Adjusting pH of Centrifuge Feed on Solids Composition
pH
11. 5
11. 0
10. 5
10. 0
9.5
9.0
Supernatant Liquid Composition, mg/la
Calcium
as CaCO0
440
290
220
180
180
230
Magnesium
as CaCC>3
10
30
50
80
120
200
Phosphorous
as PC>4
0.04
0.07
0.10
0.17
0.38
1.16
Percent of Solids Dissolved
Sludgec
CaC03
0.7
-0.4
-0.6
-0.6
-0,3
Mg(OH)2
-0.4
0.1
1.1
1.9
3.7
Ca5OH(P04) 3
nil
nil
nil
0.01
0.04
Centra te
CaCOg
3.8
-1.5
-2.5
-2.5
-1.2
Mg(OH>2
-0.9
0.3
1.9
4.2
8.1
Ca5OH(P04) 3
nil
nil
nil
0.01
0.05
From trend lines shown in Figures 8-9, 8-10, and 8-11.
Compared to material at pH 11.0. A negative value indicates material precipitated rather
than dissolved.
°For a sludge containing 5 percent total solids, and solids consisting of 40% CaCOS, 5% Mg(OH>2
and 8% Ca5OH (PO4)3.
For a first stage centrate containing 2 percent total solids, and solids consisting of 20% CaCC>3,
6% Mg (OH)2 and 16% CasOH (PO4)3.
73
-------
IUO
PERCENT
do
O
U?
CT
N SECOND ST>
Cft
O
^ 40
Q:
LU
O
O
LU
tt: 20
Q
•J
0
n
Iliiii
* *°
1 I
Flocculation pH : 1 1.0
Machine: Sharpies P-3000
Fe
Au
_ A Poi
Cei
9 A So
d
ed Rate.gpm: 14.8-16.6
ger Speed, A RPM : 12 __
id : 4
itrifugal Force, g: 2100
ids in 2nd Stage Feed,
percent ; 1.5-1.8
~ A 3 ^ O _4 Polymer; ICI Atlasep 2A2 ~
° SYMBOLS
Q ° FIRST STAGE FEED pH
— 9.8 10.5 11.5
0 3 •
A A A
a a •
2nd STAGE FEED pH
9.0 - 9.5
9.8 - 10.6
1 1.2 - 1 1.9
i i . i i i i
1234567
POLYMER DOSAGE IN SECOND STAGE FEED, LB/TON DRY SOLIDS
Figure 8-8
Effect of Adjusting pH of Centrifuge Feed
8
-------
700
6OO -
i I i I
CD Soluble calcium in digester
supernatant
© Soluble calcium in First stage
centrate and feed
Soluble calcium in sludge
after bacterial action
11.5
I2.O
Figure 8-9
Effect of pH on Solubility of Calcium
75
-------
0 O \00
B Soluble magnesium
in digester supernatant
r © Soluble magnesium in First
stage centrate and feed
Soluble magnesium in sludge
after bacterial action
8.5
11.5
12.0
Figure 8-10
Effect of pH on Solubility of Magnesium
76
-------
1.6
1.2
1.O
0.8
0
CL
5 0.6
O.4
0.2 —
O
0
0
0
0
0 Soluble phosphorus in
digester supernatant
© Solubility of phosphorus
in First stage centrate
and feed
©
8.5
9.O
9.5
10.0
pH
IO.5 11.0
11,5 12.0
Figure 8-11
Effect of pH on Solubility of Phosphorus
77
-------
solids composition was estimated from the curves. As shown in Table 8-4,
the change in solids composition over the pH range of 9.0 to 11.5 was small.
The Largest change in centrate solids occurred in magnesium. An estimated
8 percent decrease in the magnesium content of the centrate solids occurred
when the pK was reduced from 11.0 to 9.0.
The observed change in solubility of magnesium with pH is much less than
would be expected for equilibrium conditions. The magnesium solubility can
be estimated from solubility equilibrium--^. The comparison between cal-
culated and observed solubilities is shown in Table 8-5.
The large discrepancy between the calculated equilibrium conditions and the
observed results is evidently due to the fact that equilibrium was not attained
when adjusting pH. Evidently, a longer reaction time is required than was
employed when making the pH adjustment. Reaction times varied, but usually
were several hours.
In another experiment a sample of sludge was gently stirred in a series of
beakers for differing time periods. During that time the pH decreased as a
result of bacterial action. The results are shown in Table 8-6 and plotted
in Figs. 8-9 and 8-10.
Even though 13 hours elapsed while the pH was reduced from 10.1 to 9.4
the amount of dissolved magnesium was far below the equilibrium value.
This again indicates a slow approach to equilibrium.
Also shown in Figs. 8-9, 8-10 and: 8-11 are the levels for soluble calcium,
magnesium and phosphorus in the supernatant from.the experimental digesters
described in Section IX, In this case, many days of reaction time was avail-
able and for the digester with liquid retention times of 15 days or greater
nearly complete dissolution of magnesium was indicated. This indicates that
the reaction time for solubilization is measured in days rather than hours.
In conclusion, perhaps the reason that sludge recarbonation (decreasing
the pH) had little effect on sludge solids composition is that major changes in
sludge constituents did not occur within the reaction times employed in the
study.
Effect of Adding Sludge to Second Stage Feed
In an effort to reduce the polymer requirement, chemical primary sludge1 was
added to the second stage feed. Jt was hoped that the fibrous solids-in the
sludge would assist the capture of second stage solids. The results are shown
ii\ Fig. 8-12. At a given polymer dosage, a higher solids recovery and a
slightly dryer cake were obtained v/hen sludge was added to the feed. However,
since no polymer is required to capture approximately 60 percent of the solids
78
-------
Table 8-5
Comparison of Observed and Theoretical Magnesium Solubilities
pH
9.5
9.0
Solubility of magnesium as calcium carbonate, mg/1
Calculated
1 3,610
36,100
Observed
120
200
Table 8-6
Hardness Release Due to Bacterial Action
pH
10.1
9.9
9.7
9.6
9.4
9.4
Time, hr
0
2
4.3
6
11
13
Solubility, mg/1
Calcium
as CaCO0
3
206
206
214
226
240
260
Magnesium
as CaCO0
3
66
92
122
140
192
216
79
-------
too
CO
o
8O
6O
ki
o
ct
Uj
X
tfc
kl
* 40
CO
Q
20
SLUDGE CORRECTED FOR
FIRST STAGE CAKE
Machine; Sharpies P 600
Pond : 3 '/2
Auger pitch, in : 1
Auger speed, ARPM '- 24
Centrifugal force , g : 2IOO
Feed rate, gpm : 3.4
Flocculation pH : 11.5
Polymer: 1CI Atlasep 2A2
SYMBOL
0
A
FEED COMPONENTS
1 1 % SLUDGE ,
89% 1st STAGE CENTRATE
100% 1st STAGE CENTRATE
FEED
% SOLIDS
2.3
1.8
CAKE
% SOLIDS
12,8-13.7
10.4-12,7
I
23456
POLYMER DOSAGE, LB/TON DRY SOLIDS
8
Figure 8-12
Effect of Adding Sludge to Second Stage Feed
-------
in the raw sludge which was added to the centrate, the results should be
corrected for the first stage solids. Correcting the recoveries and polymer
dosages for those solids in the combined sludge which required no polymer
gives the corrected curve in Fig. 8-12. After correction the data still show
a slight improvement in solids recovery by adding sludge to the second stage
feed. However, the benefit is marginal.
SUMMARY
The study of centrifuging first stage centrate demonstrates that a high re-
covery and satisfactory dewatering of the solids can be obtained. For a
flocculation pH of 11.0 a recovery of 80 percent or more of the solids can be
realized using a polymer dose of two pounds per ton of dry solids. Cake
solids content greater than 15 percent can be achieved. With the proper
operating conditions a polymer dose of two pounds per ton of solids should
be adequate for capturing 80 percent of the solids for any flocculation pH
between 10.2 and 11.5. However, dewatering of the solids was less efficient
at a flocculation pH of 11.5 than at lower pH values.
An evaluation was made of the effect of recarbonation or lime addition to adjust
the pH of either the chemical primary sludge or first stage centrate prior to
centrifuging. The pH adjustment had little or no beneficial effect on solids
recovery in the pH range tested. Raising the pH of the first stage feed above
11.0 increased the polymer requirement substantially, but at or below 11.0
adjustments in pH had at most a minor effect on the polymer requirement.
The polymer requirement for a given solids recovery was sensitive to certain
machine configurations and operating conditions. Considerably less polymer
was required with the Sharpies P3000 centrifuge than with the smaller Sharpies
P600. Whether the difference was due to differences in size or in certain config-
urations, such as the auger pitch and shape, is not known.
One of the most important centrifuge operating variables is the rate of solids
removal from the centrifuge bowl. For a given machine, the rate of solids
removal is mainly a function of auger speed. The auger speed must be carefully
adjusted to allow the rate of solids removal to match the rate of solids capture
in the centrifuge. Consequently, a high degree of flexibility in adjusting
auger speed is necessary.
The capacity of a centrifuge appears to be far more dependent on the rate of
solids removal than the liquid flow rate. Consequently, satisfactory solids
recovery can be obtained in a given unit over a relatively wide range of feed
rates provided that the rate of solids removal is adjusted for differences in
rate of solids intake.
81
-------
As would be expected, increasing the centrifugal force improved the sep-
aration and decreased polymer requirement. However, increasing centri-
fugal force above 2,100 g produced marginal benefits.
Increasing liquid depth in the centrifuge bowl improved solids recovery but
decreased the efficiency of dewatering. A pond setting of four was found to
be optimum for the beach height of the machines used.
In summary, satisfactory solids recovery and dewatering of first stage cen-
trate can be attained with a centrifuge. For an operating installation, flexibility
in the rate of polymer addition and rate of solids removal from the centrifuge
bowl will improve operating efficiency.
82
-------
SECTION IX
CENTRATE PROCESSING BY ANAEROBIC STABILIZATION
Although incineration of the waste solids is planned for the initial stages of
the CCCSD reclamation plant, incineration is not the only available alternative
for later stages. Anaerobic digestion may be attractive, provided that suit-
able disposal sites are available. Moreover, since sludge digestion is a net
energy producing process as opposed to an energy consuming process such
as incineration, availability of fossil fuels may in the future establish digestion
as a preferable solution to the sludge disposal problem.
Studies were conducted on a bench scale to evaluate anaerobic digestion as a
means of processing the primary lime sludge solids in the first stage centrate
from the solid bowl centrifuge. The centrate was thickened to two different
solids levels. The low solids lime sludge was fed to the digesters at a TS
concentration of 2.5 percent. The high solids tests were conducted at TS
concentrations of 3.3 and 7.1 percent. Digester performance was evaluated
at each level. Digesters were operated at different solids retention times (SRT)
for the lime sludge feed, while a control digester was operated using raw pri-
mary (non-chemical) sludge feed as a comparison. Also, limited data were
obtained from operation of a main plant digester which was fed both lime sludge
and conventional primary sludge.
OBJECTIVES OF STUDY
Because there is very little information available on the anaerobic stabilization
of chemical sludges, the pilot study was conducted to evaluate operational
parameters. Among these are:
1. Operational characteristics of digesters, such as pH, alkalinity,
volatile acids and gas composition.
2. Digester performance, including gas production, volatile matter
destruction, loading effects, and odor and drying characteristics
of digested sludge.
3. Concentration ranges of soluble calcium, magnesium, and orthophos-
hosphate in the digesters; total calcium, magnesium and phos-
phorus in the digested sludge; and heavy metals, ammonia
nitrogen and soluble sulfide levels in the digested sludge.
4. Corrosion and scaling evidence in circulation lines used in the
experimental system.
83
-------
DESCRIPTION OF STUDY
The bench scale digester work was conducted using five-gallon glass bottles
as the reactors. The reactors were kept at 38 C. The digesters were filled
to a three-gallon level and kept well-mixed by recirculating the sludge with
external pumps which were activated tor approximately five minutes every
half-hour by an electric timer control switch. Gas produced from the digesters
was collected by the hydraulic displacement principle in separate calibrated
bottles filled with a slightly acidic solution. Because the digesters were well
mixed, solids retention time equaled hydraulic retention time. Initially three
digesters were fed lime sludge at different retention times, and later a fourth
was added. This gave operating data at four different retention times from
10 to 25 days. A control digester was fed raw primary sludge at a 15-day SRT.
Analyses
The feed and digested sludges were periodically analyzed for pH, alkalinity,
total and volatile solids, and soluble calcium, magnesium, and orthophosphate.
Also a limited number of analyses were conducted for ammonia nitrogen, solu-
ble sulfides, heavy metals, hexane extractable material, and total calcium,
magnesium and phosphorus. The digester gas was analyzed for carbon
dioxide by an Orsat apparatus, and this method was checked by chromatograph
to verify the carbon dioxide and methane content of the gas.
Performance Parameters
The experimental data were used to calculate the following parameters:
1. Percent volatile matter reduction.
2. Pounds of volatile matter fed per cubic foot of digester capacity
per day.
3. Cubic feet of total gas and methane gas per pound of volatile
matter fed (or destroyed).
Chronological Description of Experiment
The first period of digester operations evaluated the performance when
digesting first stage centrate of relatively low total solids (2.5 to 3.0 percent)
containing 45 to 50 percent volatile solids. The digesters were operated for
47 days. Three digesters were fed lime sludge and operated at solids retention
times of 15, 20, and 25 days. A control digester was run for three weeks at an
SRT of 15 days, with a feed of about 2.5 percent total solids containing 68
percent volatile solids.
84
-------
The second phase of experimental work lasted 40 days for the lime sludge
digesters and 60 days for the control digester During this second period
the tv/o lime sludge digesters that had been operating with an SRT of 15
and 25 days were fed sludge of 6 to 7 percent total solids of which 45 to 50
percent were volatile. A new lime sluage digester with a 10-day retention time
was placed in operation on this same feed. The digester that had been
operating at a 25-day SRT was fed lime sludge with 3 percent total solids
in order to evaluate the effect of a slight increase in feed solids - possibly
enough to overcome any near-washout conditions which may have been
present during the previous runs with low feed solids. During the second
phase the control digester was fed conventional sludge of 4 to 6 percent total
solids of which 60 to 65 percent were volatile.
Additional analyses during this second period included ammonia nitrogen
(NH3~N); total calcium (Ca) , magnesium (Mg) and phosphorus (P); heavy
metals; and hexane extractable materials. At the completion of the study,
all circulating lines and pumps were examined for corrosion and deposits.
PERFORMANCE HISTORY AND OPERATIONAL SUMMARY
The operational results presented in this section will compare the stabili-
zation of lime sludge arid conventional sludge. Of particular importance is the
evaluation of the ability of the anaerobic digestion process to stabilize lime
sludge.
Low Feed Solids Operation
To initiate the digestion process, seed sludge from the treatment plant
digesters was placed in the reaction bottles, and a progressive feeding
schedule was commenced. The initial operation of the bench-scale lime
and control digesters was characterized by slow adaptation of the methane
fermenting bacteria, resulting in high volatile acids for the first two weeks.
After acclimation of the methane bacteria, volatile acids decreased from
around 2000 ppm to under 100 ppm (principally as acetic acid) This
reduction was accompanied by a simultaneous increase in gas production.
The initial high volatile acids levels were probably the result of shocks
both in type of feed and in temperature imposed upon the seed sludge.
Values of pH reached equilibrium at about 7.7 for the lime sludge digesters
and 7.0 for the control digester. The soluble alkalinity (as CaCC^) was
generally between 3500 and 4000 ppm for the lime sludge digesters, while
the control digester was usually between 2700 and 3000 ppm. Time plots of
pH, alkalinity and volatile acids are shown in the appendix.
It is interesting to note that all of the lime sludge digesters showed a
sharp drop in gas production simultaneously after about two weeks of
85
-------
operation. Gas production curves are shown in the appendix. The
drop could have been caused by a toxic contaminant in the common lime
sludge feed supply. A contributing cause could have been that the
concentration of organisms was so low due to the dilute feed that resis-
tance to toxic compounds was poor 13. The low solids content in the lime
sludge digesters -was possibly near the borderline of washing out at the
lower solids retention times.
The control digester performed well for about three weeks on the lower
solids feed. Thicker feed was then introduced, and the sudden shock
caused a drop in performance until the digester could equilibrate again.
Table 9-1 shows a summary of the important performance parameters for the
low feed solids period. The following observations can be drawn from
comparisons between the control and lime sludge digesters:
1. Better volatile matter reduction was observed in the control
digester than in the lime sludge digesters.
2. Higher methane content occurred in the lime sludge digester
gas than in the control digester gas. This was due primarily
to the pH difference between the two types of digesters; CC>2
was absorbed more readily in the lime sludge digesters.
3. Methane production per Ib of volatile matter destroyed was
about the same for control and lime sludge digesters.
4. Methane production per Ib volatile matter fed was greater
in the control digester than in the lime sludge, digesters.
It was also noted that there was no detectable hydrogen sulfide (t^S) odor
in the lime sludge digester gas, while the control digester gas did have a
slight H2S odor. The higher pH in the lime sludge digesters caused more
sulfide to be in the ionized form in solution, whereas the lower pH in the
control digester permitted more favorable equilibrium to hydrogen sulfide
gas.
It is possible to observe from comparisons among the lime sludge digesters
the following:
1. Greater volatile matter destruction and more methane producefd per
Ib of volatile matter destroyed (or fed) in the lower rate (20 and
25 day) lime digesters than in the unit having an SRT of 15 days.
2. Essentially no difference in the methane produced per Ib of
volatile matter fed in the 20 and 25 day digesters. Overall
performance of these two units was quite similar.
86
-------
Table 9-1
Digestion Performance Parameters
JFeed sludge
ipe and solids
etention time
ime, 15 day
me, 20 day
me, 25 day
iw (control) ,
15 day
me, 10 day
me, 15 day
me, 20 day
me, 25 day
iw (control),
5 day
Low feed solids operation
Loading; Ib
VM per cu
ft per day
.05*
.04a
.03"
.07a
Typical pH
range
7.6-7.8
7.6-7.8
7.6-7.8
G.8-7.2
Digester
TS,bwt
percent
2.2
2.0
2.0
1.5
Overall VM
destruction
percent
34.2
40.4
44.8
49.6
Stable VM
destruction?
percent
34.9
45.9
44.8
49.6
Cu ft gas
. per Ib VM
destroyed d
12.3
13.2
13.8
16.9
Cu ft CH4
per Ib VM
destroyed*1
11.3
12.1
12.6
12.3
Cu ft CH4
perlb
VM fed"
3.8S
4.84
4.66
6.02
Percent
COa in
gasd
8.4
8.4
8.9
27.1
High feed solids operation
.20e
.14e
.ioe
.08*
.14°
7.4-7,5
7.4-7.6
7.4-7.0
7. 4-7. 6
7.1-7.2
5.9
5.6
5.5
2.7
2.5
33.6
38.3
3S.8
41.8
54.4
35. 6
41.8
38.8
41.8
58.1
13.1
12.9
13.3
15.1
13.9
11.7
10.9
11.3
13.3
9.8
3,93
4.17
4.38
5.55
5.29
15.0
15.5
15.0
12.3
30.0
laaed on lime feed of 2.7 percent total and 45 percent volatile solids; and control feed of 2.5 percent total and 68 percent volatile solids.
'alues near end of runs.
table periods shown on gas production curves in Appendix,
Average values.
laaed on lime feed of 7.1 percent total and 47 percent volatile solids, for>10, 15, and 20 day SRT lime digesters; and lime feed of 3.3 percent
>tal and 45 percent volatile solids for 25 day SRT digester; and control feed of 8.3 percent total and 63 percent volatile solids.
87
-------
The soluble calcium, magnesium and orthophosphate analyses are listed for
both the digested and feed sludges in Table 9-2. From these data one
can see that:
1. Soluble orthophosphate is lower in the control digester effluent
than in the feed. This is postulated to be because the feed
sludge pH is lower than the digester pH, making possible the
precipitation of phosphate under anaerobic conditions.
2. Lime sludge digestion releases a slight amount of soluble ortho-
phosphate. Since anaerobic conditions reduce the pH of lime
sludge the solubility of phosphate is increased. The amount
released, however, is far less than the amount that could potentially
be released if it were possible to dissolve all of the precipitated
phosphorus. For instance, the sludge composition data in Section VI
would indicate a value of total phosphorus (as orthophosphate) of
about 750 mg/1 for feed solids at 2.5 percent TS. Thus, less than
1.5 percent of the available orthophosphate was dissolved in the
lime sludge digester.
3. Magnesium is released to a high concentration in the lime
digesters. The concentration of soluble magnesium in the lime
digesters was 2000 ppm as calcium carbonate (CaCC>3) . Literature
ranges for the effects of magnesium cation concentration on
digester organisms are: stimulatory, 75-150 ppm as Mg (300-600
asCaCOg); moderately inhibitory, 1000-1500 ppm as Mg (4150-6250
as CaCCU); and strongly inhibitory, over 3000 ppm as Mg (12,500
ppm as CaCC^) . The concentration of magnesium in the lime
sludge digester would there fore be between the stimulatory and
moderately inhibitory ranges. The presence of other elements
which could act either synergistically or antagonistically is
unknown. Again, the sludge data in Section VI would indicate
a total value of magnesium (as CaCOg) ot 1700 mg/1 for a 2.5
percent TS feed. Thus, essentially all of the magnesium pre-
cipitate is dissolved in the digester.
4. No significant increase in soluble calcium or magnesium occurs
in the control digester operation.
High Feed Solids Operation
The lime and control feed sludges were further thickened in an attempt
to build a higher concentration of organisms in the digesters and possibly
achieve more stable operation,. especially in the lime sludge digesters.
88
-------
Table 9-2
Feed and Digested Sludge Characteristics
Sludge type
and SET
Lime feed
Control feed
Lime, 15
day
Lime, 20
day
Lime, 25
day
Control, 15
day
Lime, In
day
Lime, )5
day
Lime, 20
day
Lime, 25
day
Control, 15
day
Total
solids, wt
percent
2.5-8.0
2.5-li.l
Volatile
solids, wt
percent
•15-55
(iO-70
pll
range
s.H-9,4
5. 3-0,5
Alkalinity,^
mg/1, as
CaCOj
050-1000
600-1200
Volatile
acids,1*
mc/1
514- 050
S<0-1300
Soluble
Ca, mg/1
as CaCO3
140-5 GO
300-400
Soluble
Mff mg/1
as CaCQs
3SO-I9GO
420- 480
Soluble P,C
mg/1 as
0-PO4
3.4-9.9
117-124
Total Ca?
mg/1 as
CaCOs
29.200
4,000
Total Mgf
mg/1 as
CaC03
8503
1498
Digested sludges, low Teed solids period
2.2
2.0
2.0
1.5
41
42
42
02
7.6-7.S
7.6-7, H
7.6-7.S
G.8-7.2
5.3
5.11
5.5
2.7
'2.5
36
35
35
3V
47
7.4-7.5
7.4-7.5
7.4-7.0
7.4-7.11
7.1-7.2
3400-4200
3500-4500
3500-4500
2750-3100
100- 750
150- 400
liO- 500
100-1000
350
330
370
340
2000
2000
2200
300
11.6
11.4
14.6
50.8
-
-
-
-
-
-
-
-
Total P,
rag/1 as
P04
624
489
-
-
-
-
Grease,
mg/1
9344
7616
Ammonia
N, mg/1
120
190
-
-
-
-
Digested sludges, high feed solids period
4200-5500
4000-5500
4500-5400
3700-4350
3000-3(100
S50-1200
350- 700
350- 550
52- 250
50- 150
6*0
Ii(iO
580
U20
390
3450
3500
3500
2200
330
31.4
29.7
29.7
23.1
48.2
19,800
30,450
29,400
13,650
3,052
6314
3860
32(>1
1946
1510
1291
4791
4798
1829
474
3300
3074
2184
1352
2211
-
-
-
-
Soluble,
sulfide,
mg/1
.1
.1
-
-
-
-
550
600
520
300
580
.1
,1
.1
1
.1
CD
CD
values near ends of runs.
Typical range during stable operation.
Average value near end of run.
Spot analyses only near end of high feed solids run, insufficient analyses run to be considered typical.
Table 9-3
Digester Sludge Heavy Metals Analyses at End of
High Feed Solids Run
Digester type
and SRT
Lime, 15 day
Control, 15
day
nig/Kg of dry total solids
Cr
70
7H
Cu
351;
3itH
Pb
239
3-13
Zn
522
I;K.|
rd
12
15
Ni
G7
liC.3
mo/Kg of dry fixed solids
Cr
109
inn
Cu
55(5
812
Pb
373
700
Zn
815
l.'l 05
Cd
19
,11
Ni
104
135:5
mg/1
Cr
•1
3
Cu
21
13
Pb
1-1
11
'/n
30
23
Cd
1
1
Ni
4
•M>
-------
The pH decreased from around 7.7 for the previous low lime feed solids
period to around 7.5 for the high solids period in all the lime sludge
digesters. The control digester pH showed no significant change from
the previous low solids period. Alkalinities increased in the lime sludge
digesters to around 5500 ppm as CaCOg in the 10, 15 and 20-day digesters,
and to around 4400 ppm in the 25-day digester (still on "low" solids feed).
The control digester alkalinity rose slightly to around 3400 ppm. Operating
characteristics for all digesters during the high feed solids runs are shown
in the appendix.
Performance parameters are shown in Table 9-1 for the high feed solids
period. The following observations can be drawn from comparisons between
the lime sludge and control digesters:
1. Better reduction of volatile matter was obtained in the control
digester than in the lime sludge digester.
2. Again, higher methane content was found in the lime sludge
digester gas than in the control digester gas.
3. Lime sludge digesters produced more methane per Ib of volatile
matter destroyed than did the control digester.
4. The control digester produced more methane per Ib of volatile
matter fed than did the 10, 15, and 20-day SET lime sludge digesters,
but produced slightly less than did the 25-day lime sludge digester.
From comparisons among the lime sludge digesters it is possible to conclude
the following:
1. Percentage volatile matter destruction became progressively better as
solids retention time increased.
2. The lime sludge digester with a SRT of 25 days produced more
total gas and more methane gas per Ib of volatile matter destroyed
and fed than did any of the other lime sludge digesters.
Once again, the control digester exhibited a decrease in orthophosphate, while
the lime sludge digesters showed a small release of orthophosphate. As before,
the lime sludge digesters indicated a nearly complete release of magnesium
into solution.
The range of operating characteristics for the digesters in the high solids run
are listed in Table 9-2 and are plotted on graphs in the appendix. It is
apparent that the volatile acids residual in the lime sludge digesters was
unexpectedly high, especially in the higher rate (lower residence time) digesters,
90
-------
This appears to be an equilibrium situation in which the methane bacteria
population was limited by hydraulic washout. The concentration of soluble
metals may have inhibited the growth rate of the methane bacteria. Despite
the fact that the volatile acids levels were often quite high, buffering in the
system prevented a significant drop in pH. For the high solids feed runs,
operation was quite stable in terms of volatile matter destruction and gas
production. In conventional digesters, such high levels of volatile acids
would indicate incipient failure of the digestion process14. This is not the
case for the lime sludge digesters run during high solids feed.
Table 9-2 lists some analyses run on the digested sludge. The analyses
for soluble magnesium, calcium and phosphorus are averaged values near
the end of each run, while the analyses for total magnesium, calcium and
phosphorus are spot analyses only, done on dried sludge and converted to
mg/1 values. These numbers are not to be considered typical concentrations
but only approximations, because not enough analyses were done to establish
reliable values. This is why there are inconsistencies in comparing total
calcium or magnesium in the feed sludges with those values in the digested
sludges.
Looking at the values for soluble magnesium in the lime sludge digesters
and comparing them to the rough values of total magnesium, one can see
that most of the magnesium in the lime sludge digesters was in the soluble
state. The data also show an increase in ammonia nitrogen from 120-190
mg/1 in the feed sludges to 500-600 mg/1 in the digester sludges for both
lime and control digestion. These levels of ammonia are not considered
to be in the toxic range 131
The level of soluble sultides in all digesters was very low. Probably most
of the sulfides were precipitated as metal sulfides. The levels of grease in
both the feed sludges and digested sludges can be considered only approxi-
mate values, because the number of analyses was not statistically signi-
ficant .
Table 9-3 shoxvs sludge heavy metal concentrations found in both the
control digester and the lime sludge digester operated at the same 15-day
SRT. Heavy metals levels except for nickel are about the same for both
types of digested sludge. There is no explanation for the large difference
in nickel concentrations.
The 25-day lime sludge digester which was op-rated at over 3 percent
feed solids in the high feed solids period showed better performance than in
the previous low feed solids period. Although the previous feed was only
slightly lower in solids, the increase may have been enough to prevent
significant washout of organisms by dilution.
91
-------
Oth e r Obseryatlons
Odor tests on the digested sludges revealed no offensive putrescible type
odor from either lime sludge or control digesters in either high or low feed
solids runs. In the high feed solids run, the 10-day digester sludge did
have a slightly sharp odor due to the high level of volatile acids, but it
was not considered offensive.
Drying tests were conducted on the digested sludges. Tests were con-
ducted by simultaneously drying equal volumes of sludge in flat-bottomed
pans both outside under direct sunlight and in the lab using a controlled
temperature oven at both 80 F and 120 F. Results showed essentially the
same drying rate for lime sludge and control sludge, and there was no
clear effect of total solids levels on sludge drying. Dried control sludge
was much darker than the dried lime sludge.
Lines and pump internals were examined at the end of the run for deposits
and corrosion. The piping material was plastic (PVC) but some metal (cast
iron) sections had been inserted in lines of the control and three lime
sludge digesters. Results showed all lines and pump internals to be very
clean. No evidence of corrosion was detected. Of course, the length of
the study may have been too short to allow much deposition or corrosion
to occur.
During the course of the low and high feed solids runs a main plant
digester was fed a mixture of lime sludge and control primary sludge,
both fed directly from primary sedimentation tanks. The mass of volatile
solids fed to the digester from the control primary sedimentation tank was
roughly the same as the mass of volatile solids fed from the lime sludge.
During the time of monitoring, the hydraulic residence time in the digester
was about 6 days, which is quite a high rate of feed. The gas composition
of this digester averaged 29 percent CC^, compared with an average of
37 percent for the other main plant digesters operating on primary non-
chemical sludge only. The average operating characteristics were:
alkalinity of 2700 mg/1; volatile acids of 840 mg/i; typical pH 7.1. The
volatile matter destruction could not be precisely calculated, but was approxi-
mately 50 percent while loading in Ib volatile matter per cu ft per day was
0.18. Gas production was 7.4 cu ft gas per Ib volatile matter fed.
It was found that the digester was not well mixed, even though the gas
recirculation system was in operation. A solids profile taken on the: digester
showed the following concentrations at levels below the liquid-gas interface:
at approximately one-third the depth down from the interface the solids
content was .62 percent TS; at approximately half way down the digester
the solids content was .71 percent TS; while at the bottom draw-off the
solids content was 9.6 percent TS. This shows the need for effective gas
recirculation systems in digesters. Because this digester was not well
92
-------
mixed, the hydraulic and solids retention times were not equal, and the
6-day hydraulic residence time was less than the solids retention time.
Other studies on digesting mixtures of lime and conventional sludges
report lower CO2 content in the gas, stratification in the digester, and slightly-
lower gas production than in conventional digestion processes15 A recent
study of digestion of lime primary sludge with a small amount of waste
activated sludge reported excellent gas quality and stable operation at bcth
high (8-12 percent) feed TS and low {3-4 percent) feed TS*6
High and Low Feed Solids Operations Comparisons
A comparison of the high and low feed solids runs yields the following
observations:
1. Higher feed solids operation was characterized by higher carbon
dioxide levels in the digester gas.
2. Higher levels of soluble calcium, magnesium and orthophosphate
occurred during higher feed solids operation.
3. Lower digester pH was observed during high feed solids operation.
4. Slightly better efficiency in volatile matter destruction occurred
during the high feed solids operations for SRT values up to 20
days (Fig. 9-1) . Below an SRT of 20 days, the low feed solids
digestion requires longer retention times than the high feed solids
operation to achieve the same reduction in volatile matter.
5. Generally higher methane production per Ib volatile matter
destroyed (and fed) was found during the higher feed solids
run.
6. Higher alkalinity and volatile acids occurred during the higher
feed solids run.
7. Operation was considerably more stable when the lime digesters
were fed thickened feed.
The control digester exhibited better volatile matter reduction efficiency
during the higher feed solids period, but gas production (total gas as
well as methane gas) per Ib of volatile matter destroyed (and fed) decreased
during the higher feed solids run. The methane content in the gas was
slightly lower during this period as well.
93
-------
70
60
O
O
rs
ct
^
CO
Uj
Q
ft:
LU
5
LU
§
LU
O
ft:
LU
a.
50
4O
3O
HIGH FEED
SOLIDS I
-f
CONTROL DIGESTER,
HIGH FEED SOLIDS RUN
CONTROL DIGESTER,
LOW FEED SOLIDS RUN
4F
v
SOL|DS
w
I
/O
/5
20
25
Figure 9-1
Overall Volatile Matter Destruction in Lime Digesters
as a Function of Solids Retention Time
30
94
-------
Comparison Between Pilot Control Digester and CCCSD Main Plant Digesters
In addition to the experimental data from the pilot digester work, data from the
operation of the CCCSD main plant digesters was available for the computation
of digester performance parameters. Records of CCCSD digester operation
{on conventional sludge) during 1971 were examined and analyzed to compare
the performance of the CCCSD main plant digesters with that of the pilot control
digester. The CCCSD employs three primary digesters and one secondary
digester.
The pilot digester, under high (5.3 percent) feed solids operation, showed
slightly less volatile matter destruction (58 percent vs. 61 percent) and unit
methane production (9.8 vs. 10.7 cu ft CH4 per Ib VM destroyed) than did
the CCCSD digesters (with 5.8 percent feed) . The solids retention time,
however, was significantly greater for the CCCSD digester operation (35 days)
than for the pilot control digester operation (15 days) . Loading rates for the
CCCSD and pilot digesters were. 12 and .14 (pounds of volatile matter per
cubic feet of digester capacity per day) respectively. The pilot control digester
gas showed lower carbon dioxide content (30 percent) than did the CCCSD
digester gas (37 percent). The pilot control digester under low (2.5 percent)
feed solids and a loading rate ot .07 (Ib VM per cu ft per day) showed less
volatile matter destruction (50 percent vs. 61 percent) but higher unit methane
production (12.3 vs. 10.7 cu ft CH4 per Ib VM destroyed) than did the CCCSD
digesters.
In general, however, the performance of the pilot control digester at a 5.3
percent feed and a 15-day SRT was virtually the same as that of the CCCSD
digesters under 5.8 percent feed and 35-day SRT.
The pilot control and CCCSD digester performance comparisons show that:
first, digestion performance on a pilot scale digester can be considered repre-
sentative of large-scale operation; and second, that the CCCSD main plant
digesters at an SRT of 35 days have substantial reserve capacity.
95
-------
SECTION X
SPECIAL STUDIES
A number of solids processing studies of an exploratory nature were made in
conjunction with the main efforts. The exploratory studies included evaluation
of centrate recycle to the liquid processing system, laboratory vacuum filtra-
tion, and sludge stabilization by chlorination. The results of these studies
provide useful information for evaluating different processes and processinc
schemes.
SIMULATED CENTRATE RECYCLE
The envisioned sludge processing scheme in the first stage of the treatment
plant will include a two-stage centrifugal separation step. The calcium car-
bonate in the solids from the first stage will be calcined and recycled as lime
to the primary treatment. Solids from the second stage, high in phosphorus,
magnesium, and organic content, will be incinerated. The second stage cen-
trate will be recycled to the head oi the treatment plant. To determine if
recycling the centrate will cause any serious treatment problems,a short study
of centrate recycle from a simulated two-stage centrifuge system was made.
For producing a recycle centrate, a single-stage centrifuge separation was
made under conditions that would remove essentially the same solids as will
be removed in the two-step scheme. In earlier work, it was observed that
a single-stage centrate similar in composition to a second-stage centrate can
be pi'oduced by operating under conditions of high solids capture. This
requires the use of a polymer, which is also in the second stage of a two-stage
centrifugation process.
The total sludge production for each day was stored and centrifuged during
the day shift. The centrate was pumped directly to the chemical primary
preaeration tank. Consequently, the concentration of recycle centrate in the
influent to the treatment plant was about three times as high during the day
shift as it would be in a continuously operated process. The high concentration
of recycled centrate during the day shift would be expected to accentuate
any deleterious effect which might occur during continuous operation.
Due to mechanical failure of the polymer feed system the study was limited to
3| days of satisfactory operation. For the next U days the polymer was added
to the centrifuge feed inlet line instead of the polymer inlet tube in an effort to
continue the study after the mechanical failure. As a result of the change in
location of polymer addition the amount of solids in the centrate increased
substantially.
97
-------
Table 10-1
Effect of Recycle of Second Stage Centrate
Process parameters
Period of Centrifuge Operation
Start
Finish
Centrifuge Operating Conditions
(P3000)
Feed rate, gpm
Polymer, Ib per ton
Centrifuge Performance
Feed, total solids, percent
Centrate, total solids, percent
Cake, total solids, percent
Solids removed, percent
Primary Treatment Conditions
Influent flow, mgd
Operating pll
a
Primary Treatment Performance
Influent SS, mg/I
Effluent SS, mg/1
SS removed, percent
Run date, September 1972
18
0930
1740
22
1.9
2.G
0.3
23.5
90
1.2
11.0
22S
27
88
19
1115
1730
23
1.4
3.3
0.7
20
0915
1015
20
1.4
3.7
0.7
24.0 22.6
S2 i S4
1.0 0.8
11.0 . 11.0
216
28
87
264
33
88
21
0815
1130
17
1.5
2.7
0.0
24.7
so
1200
1750
25
3.1b
3.1
1.5
26.9
57
1.0
11.0
272
27
90
22
0300
1300
25,
4.2b
2.9
1.0
25.3
G9
0.9
23
0.8
-
24
1.4
11.0
288
27
91
25
1.4
11.0
276
25
91
26
1.5
11.0
204
34
83
27
1.6
11.0
260
32
88
28
2.9
11.0
276
47
83
ICI Atlasep 3A3 polymer, Ib per ton of dry solids in feed.
Polymer inlet tube, plugged; subsequently polymer was added to feed inlet line.
p
Based on average of ii to $ grab samples per day.
Results from 24-hour composite samples.
98
-------
There was no apparent deterioration of the primary effluent, as shown in Table
10-1, during the 5-day period when the centrate was recycled. Visual observ-
ation of samples obtained from various depths in the primary clarifier with a
Kemmerer sampler indicated no change in settling rate for the first three days.
After changing the location of polymer addition there appeared to be a slight
deterioration of settling rate as a result of the excess amount of solids in the
centrate recycle stream.
The recovery in the centrifuge was better on the first day of operation than on
the subsequent days. There was no clear indication that the difference resulted
from recycled centrate. Other factors such as variations in polymer dosage
and amount of cannery wastes in the sewage could have been responsible. For
2k days of operation after the first day the total solids in the centrate and the
percent solids removed were essentially constant, indicating no buildup of
recycled "fines" in the sludge. The poor separation for the last U days was
due to the change in the location of polymer addition.
While the study was of too short duration to establish the effect of recycle with
a high degree of confidence, the results did not show any deleterious effect.
VACUUM FILTER LEAF TESTS
Filter leaf tests were conducted using primary lime sludge generated at pH
11.0, to which either anionic polymer or activated (nitrification) sludge waste
solids were added. Some first stage centrate was also filtered using anionic
polymer as a conditioner. All filter runs were made using an Eimco filter leaf
of 0.1 sq ft area, 15 in. mercury vacuum, and Eimco filter media No. Dy-453
(yarn-spun staple) . Sludge pH at time of filtration was usually below 11.0,
as the pH of the sludge was observed to drop on standing, perhaps due to
bacterial action.
In the first experimental series shown in Table 10-2, primary sludge was
filtered at pH values of 9.4-9.5 with activated sludge solids added to give
a concentration range from 0 to 3 percent of the total sludge solids. In the
second experimental series, the primary sludge was filtered at pH values
of 10.4-10.8 with an activated sludge solids range from 0 to 1.9 percent of
the total solids. Considerably higher cake resistance values and lower filtra-
tion rates were observed for the lower pH series than for the higher pH series.
The values of specific resistance within each pH range were fairly close, with
no apparent relationship to the amount of activated sludge solids added.
In the next experimental series, primary sludge at pH values of 10.45-10.83
was filtered with varying amounts of polymer. With the exception of the single
data point at pH 10,83 (no polymer added), there was a trend of decreasing
specific resistance with increasing polymer dose up to 25 mg/1. In view of
99
-------
Table 10-2
Vacuum Filter Leaf Studies on Primary Lime Sludge Flocculated at pH 11.0
Sludge type and
variable involved
Combined chemical
sludge with acti-
vated sludge solids
added at pH range
of 9.39-9.55
Combined chemical
sludge with acti-
vated sludge solids
added at pH range
of 10.40-10.83
Combined chemical
sludge with polymer
at pH range of
10.45-10.83
First stage centrate
at constant polymer
dose with varying
pH
Total solidsf
percent
3.74
2.50
3.96
2.40
1.97
3.58
3.59
3.38
3.37
3.12
3.58
3.20
3.24
2.74
2.83
1.49
2.22
2.20
2.16
Filtration
PH
9.39
9.42
9.55
9.54
9.42
10.38
10.60
10.55
10.50
10.40
10.83
10.45
10.50
10.50
10.50
11.48
10.49
10.19
9.59
Polymer dose
mg/1
0
0
0
0
0
0
0
0
0
0
0
5
10
15
25
28
28
28
28
Percent WAS
solids0
0
O.G
1.1
1.2
3.0
0
0.4
0.9
1.3
1.9
0
0
0
0
0
0
0
0
0
Cnke resistance
x 107
9.0
10.1
13.7
10.5
10.3
2.0
1.8
2.4
2.0
2.7
2.0
2.8
2.0
1.5
1.2
7.3
2.3
4.3
0.4
Filtration
rate6
0.7
0.4
0.8
0.4
0.4
1.5
1.9
2.1
1.8
1.7
1.5
1.4
1.4
1.4
1.3
0.3
0.3
0.2
0.2
Cake TS,
percent
28.9
28.5
28.7
27. e
27.3
23.7
21.5
21.5
21.4
20.9
23.7
21.8
21. G
23.4
21.8
18.9
20.1
18. G
18.3
Total solids of starting sludge before any additions of polymer or waste activated sludge (WAS).
ICI Atlasep 2A2 anionic polymer.
c
Waste activated sludge solids, wt percent of total solids.
Specific resistance, sec /gram, on Eimco filter leaf media No. Dy -453.
6
Lb solids per sq ft per hr.
100
-------
the results presented on the eftect of sludge pH, it is probable that pH alone
is the reason that the single data point at pH 10, 83 without polymer has a
lower specific resistance than the run at pH 10.45 with 5 ppm polymer. Despite
the fairly wide range of specific resistance values, the filtration rates were
all fairly close throughout the polymer dose range.
The last filtration series was conducted using first stage centrate samples at
a dosage of 28 ppm polymer and varying the filtration pH from 9.6 to 11.5.
Cake specific resistance values decreased as pH increased from 9.6 to 10.5,
but then increased again at pH 11.5. Filtration rates were very low for all
of these runs, perhaps due to the low feed solids concentration.
Using first stage centrate generated at pH 11.0, a separate experimental series
consisting of three runs was made to determine the percent recovery of solids
obtainable with the vacuum filter leaf apparatus. Three separate filter media
were tested at a vacuum of 10 in. mercury. The results listed in Table 10-3
indicate that good solids recoveries can be achieved with vacuum filtration.
It is interesting to note that other reported values for the filterability of lime
sludge (with no polymer or activated sludge added) have been much higher
than the yields observed in this study. Sludge produced from a solids contact
type reactor-clarifier at underflow TS of 10 to 12 percent allowed vacuum filter
yields of 10 to 12 Ibs per sq ft per hr° These thicker sludges have much
better vacuum filtration characteristics than the thinner sludges observed in
this studv.
GST TESTS
A device developed by Baskerville and Gale^, which measures capillary
suction time (CST) , was used to investigate the dewatering properties of
combined chemical sludges. When using the CST apparatus, a sludge sample
is put in a small circular tube placed vertically on filter paper. Capillary
suction pressure causes filtrate to travel through the tilter paper at a rate
postulated to be dependent largely on the filterability of the sludge, and
practically independent of the hydrostatic pressure of the sample in the
vertical tube. Probes in contact with the filter paper cause the clock to start
as the liquid interface reaches the tirst probe, and stop as the interface reaches
the second probe. The time, interval for the liquid to pass between the two
probe reference points is related to the sludge filterability. The unit (Model
92) is manufactured by Triton Electronics, Ltd., England.
Results on a variety of chemical sludge and first stags centrate samples showed
that there w-is no reliable correlation between capillary suction time (CST) and
sludqe specific resistance as measured by the filter leaf apparatus. There
were also no trends in CST values with varying polymer concentrations in the
sludge.
101
-------
Table 10-3
Vacuum Filtration Tests on 1st Stage Centrate Generated at pH 11.0
Run no.
1
2
3
Filter media
used
Eimco Dy-453
spun staple
Eimco NY
332F
spun fill
Eimco NY
319F
multiple fill
Contrate TS
percent
2.3S
2.3R
2.3S
Effluent TS
percent
0.2(>
0.42
0.2(>
Cnke TS
percent
19.3
20.2
21.0
Filtration rate,
Ih per sq ft per hr
.45
.45
.3(5
Solids recovery
percent wAv
94
.S3
94
Table 10-4
Filtration Rates on Pepcon Processed
and Unprocessed Centrates
Centrate
batch no .
A
B
Filtration rate, Ib per sq ft per hr
Pepcon processed
centrate
.08
.12
Unprocessed centrate
.08
.11
102
-------
CENTRATE TREATMENT AND STABILIZATION BY CHLORINATION
A Pepcon electrolytic test facility was operated to treat centrate from the first
stage lime sludge centrifuge. Basically, this unit consists of electrolytic cells,
with concentric electrodes connected to a rectifier. When chlorides are
present in the process stream, chlorine is produced at the electrodes and can
be used for chemical oxidation of organic material. Centrate was pumped and
recirculted through the annular space between the electrodes, with chlorine
production and chemical oxidation occurring during this time. All tests were
run at current density of 4 Kamp-hr per 1000 gal with a salt dose of 5 g/1 to
supply the necessary chlorides. The unit was manufactured by Pacific Engi-
neering and Production Co. of Henderson, Nevada.
Results of processing centrate through the Pepcon unit showed that electrolytic
treatment did substantially reduce biological activity and bleached the color
from brownish-black to milky tan. There was also evidence of improved
settling in the electrolytically processed centrate. Equal amounts of the Pepcon
processed centrate and unprocessed centrate were poured into holes in the
ground at opposite ends of the ATTF mobile laboratory as a qualitative test
of lagooning or land disposal. The Pepcon treated centrate remained odor-free
and attracted very few flies, while the untreated centrate had a pronounced
odor and attracted many flies. This test was performed in the warm summer
months, during which the most demanding conditions are placed on the disposal
lagoons.
Pepcon-processed centrate did not show significantly improved filterability
over unprocessed centrate. Filtration rates were extremely low for both pro-
cessed and unprocessed centrate (Table 10-4) .
Tests with the Pepcon unit were of limited duration and did not allow study of
power optimization. The power cost is a significant item in considering this
type of treatment. Also, salt must be added to the stream to be processed in
order to facilitate chlorine production at the electrodes. If the treated centrate
were concentrated by thickening prior to land disposal and the decanted fluid
returned to the chemical primary, the added salt would increase total dissolved
solids in the main flow stream by 60 mg/1. In light of the water reclamation
objectives for this project, which limit total dissolved solids of the product
water1, this decant return could not be tolerated in the CCCSD water reclama-
tion plant.
On the basis of these limited studies it appears that tirst-stage centrate can be
stabilized either by chlorination or by anaerobic digestion as described in
Section IX. Further engineering studies will be required to establish the
practicality and economic feasibility of stabilization by chlorination.
103
-------
SECTION X!
DISCUSSION
The studies conducted on chemical sludge production, thickening, classifica-
tion and dewatering, along with the studies on digestion, provided sufficient
design data to construct integrated solids processing flow sheets. Two
alternative solids processing schemes have been constructed for the disposal
of waste solids—one employing recalcining and incineration and the other
employing digestion. In both alternative processing schemes, two stage centri-
fugation was used to classify and dewater (or thicken) the sludge.
CHEMICAL PRIMARY OPERATION
Optimum primary operating pH is a function of the performances and costs of
both the liquid and solids processing phases. It was shown in Section V that
operation at pH 11.0 resulted in the least hardness increase across the process;
this factor dictates operation at pH 11.0 for the CCCSD water reclamation
plant, since minimizing hardness of the product water is desired for the purpose
of industrial use. Except for this, liquid processing performance was nearly
equivalent at all three operating pH levels (10.2, 11 and 11,5) .
Sludge generation increased as the lime dose was elevated between pH 10.2
and 11.5. Sludge balances tor four operating periods agreed exceptionally
well with the solids balances constructed on the basis of changes in sewage
constituents across the process. Further, sludge composition by chemical
analysis was in reasonable agreement with the calculated compositions, con-
sidering the possible sources of error.
On the basis of the close agreement obtained, solids balances can be constructed
for design purposes which can be expected to project realistic solids loadings.
WET CLASSIFICATION
The extensive testing on centrifugal wet classification of combined sludge has
demonstrated that the bulk of the calcium carbonate can be recovered in the
cake while the undesired compounds can be rejected in the centrate. Further,
this classification can be effectively accomplished with the solid bowl centri-
fuge over a wide range in sludge pH (pH 10.2 to 11.5). The calcium carbonate-
rich cake that is produced is unusually dry, with a total solids content usually
greater than 50 percent.
105
-------
CENTRATE PROCESSING ALTERNATIVES
Wet classification by centrifuge produces a centrate that contains most of the
organics, magnesium hydroxide, hydroxyapatite, ferric hydroxide, and other
inerts. This stream has to undergo further treatment prior to disposal. Two
alternatives were considered: ( 1) dewatering by thickening and centrifugation
followed by incineration and (2) anaerobic digestion of the solids followed by
land disposal.
Centrate Processing by a Second Stage Centrifuge
It was found that a solid bowl centrituge can dewater centrate from a classi-
fication centrifuge, provided an anionic polymer is fed to the machine in a
dose of at least two Ib per ton dry solids. The pH of operation in the primary
affects the cake dryness obtained; below 11.0 a median cake total solids of
17 to 18 percent was obtained. An operating pH of above 11.0 appears to cause
a deterioration in cake dryness; the median cake total solids was 12 percent.
Solids recoveries of 80 percent were possible in the second stage centrifuge.
Centrate Processing by Anaerobic Stabilization
Bench-scale studies of anaerobic digestion showed that both unthickened and
thickened first stage centrate can be effectively stabilized. Volatile matter re-
duction is lower for centrate processing than for conventional raw sludge pro-
cessing, but gas yields per unit volatile matter were the same. The sludge had
an unoffensive odor in both the wet and dry states. The supernatant, however,
contained virtually all of the magnesium which had been fed to the digester,
as apparently all had been dissolved at the digester operating pH. This super-
natant if returned to the process will cause an increase in the lime dose caused
by the recycled alkalinity and magnesium.
SOLIDS PROCESSING ALTERNATIVES
The ATTF solids processing sequence as originally proposed consisted of two
stage centrifugation of the thickened primary sludge, with recalcination of
the first stage cake and incineration of the second stage cake. This sequence
was the basis for material balance calculations at each of the three operating
pH modes.
As a second alternative, another solids processing scheme was proposed, em-'
ploying land disposal of the first stage cake and digestion (with land disposal)
of the second stage cake, in lieu of recalcination and incineration. A material
balance flowsheet was constructed for this processing alternative, assuming
a primary operating pH of 11.0 (with 14 mg/1 FeClg).
106
-------
Recalclnation and Incineration Alternative
Three material balance flowsheets for the recalcination-incineration processing
sequence were constructed (see Figures 11-1, 11-2 and 11-3) for pH modes
of 10.2, 11.0 and 11.5 respectively.
Assumptions. The material balances were calculated for the first stage of
the water reclamation plant at a raw sewage average dry weather flow rate
of 30 mgd. The balances were based on a common set of raw wastewater
qualities as listed: BOD, 216 mg/1; suspended solids (SS) , 240 mg/1;
phosphorus, 10 mg/1; calcium as calcium carbonate, 75 mg/1; and magnesium
as calcium carbonate, 92 mg/1.
Performance of the primary at each pH level is based on the liquid processing
results in Section V. Performance of the first and second stage centrifuges
is based on the results presented in Sections VII and VIII.
Waste activated sludge production is based on the experimental yields deter-
mined in Section VI. All of the waste activated sludge is assumed to preci-
pitate in the primary. Overall sludge production is estimated based on the
constituent balance procedure presented in Section VI.
Incineration and recalcination performance was not tested during the project
but was estimated. Dry classification, a proprietary process marketed by the
Envirotech Corp., is shown in the flow sheet, but was not tested during
the project. As a result no credit was given to dry classification in the
balances.
It should be noted that the "organic matter" in the material balance flow sheets
consists of 20 percent inert material, whereas the organic matter defined in
Section IV and discussed in Sections VII and VIII consists of net "loss on
ignition" material, corrected for carbon dioxide and hydrated water lost
during ignition. Inerts are defined as the acid insoluble material described
in the analysis procedures of Section IV.
i
Mass Balance Comparison. Table 11-1 summarizes the calculated results
that are presented in~Figures 11-1 to 11-3. At pH 11.0 less total new lime is
required than-at pH 11.5 or 10.2. However, the least solids are generated
in the primary at pH 10.2. Thirty percent more solids are generated at pH
11.0, while 39 percent more are generated at pH 11.5 than pH 10.2. Approxi-
mately the same amount of waste solids are processed in the incinerator at
each pH level. However, the amount of solids to be recalcined varies with
operational pH.
A basis for judging the effectiveness of wet classification is to use the data
in Fig. 11-1 to 11-3 to compute recycled solids. At pH 11.0, the centrate
stream from the second stage centrifuge carries 22,059 Ib/day of recycled
solids. At the same time, 31,355 Ib of solids are returned to the primary
with the recalcined lime. This totals 53,194 Ib of unwanted solids returned
107
-------
FLOCCULATION CONDITIONS
289 mg/l Ca(OH)2, (pH10.2), 24 mg/i FeCl3, 0.25 mg/i Polymer
c
oc
— *
New time (CaO)
28,308 lbs/day
I First pass pr
i Waste AS , CaC03
12,995 lbs/day fl Ca5(OH)(pOU)3
1 ' Mg (OH)2
Fe(OH)3
1— ./_
/
1 Settleable i ' ' Pre-
1 Primary Solids f— — •* Qeratj0n Pl">mc
Ysiudf
% Ti
_£297 C
Organics 79,38
CaC03 66,03
Ca5(OH)(P01*)o 17,52
Chemical Mg (on)2 3,56
Recycle Mg o 1,1*1*
Fe (OH)3 1*,36
re2^3 6U
Inerts 10,89"
' •' ' i
ecipitation |
63,225 lbs/day 1
12,620 lbs/day j
3,199 lbs/day
3,998 lbs/day
iry Clorifier »-To Aerot
Nit rificc
__
?e
5
5PM
1 li8/^ 1 "1 Thickener
L lbs/day ' 1
3 lbs/day 1 ^\_^ ^^
•> lbs/day ^^1
5 lbs/day. |
j lbs/day 1
5 lbs/day |
i lbs/day | | ' ~
i Total 183,852 i 1
1 •" ' I i OrffAnlon
_, r
i
1 CaO 26,31*6 lbs/day 1
' CaCOo 2,1*76 lbs/day /
\ Ca5(OH)(P01t)3 3,506 lbs/day /
1 Mg (OH)2 0 lbs/day Mult
1 Mg 0 1,111 lbs/day ^ec
1 Fe (OH)? 0 Ibs/day porn
i Fe20, 4 586 lbs/day R.!Ca
Inerts 9,9H* lbs/day l"ur
Total 1*3,939
1
/7S
1 (Class
\
1 Cake 1 Ca5(OH)(POU
1 J^/yg 1 Mg(OH)2
^vt i 71 n™ 1 Mg 0
-vj 1.73 CFM | Fe (OH)3
'Pie I 1 iner^s
rth I „ , .
Icine |_ Total
nace
— — — —
L! CaC03
ICa5(OH)(P01l
x K °
h \ 1 *" 2°3
if far 1 '
/ Total
Centr
I Organics 51,
1 CaC03 16,
1 Ca5(OH)(POl*)o lU,
| Mg^(OH)2 J 2,
I Mg 0 1,
Fe (OH)3 3,
1 ' Fe20o
| Inerts 6,
ion- |_Total 96'
ition
c
:
£
| — i 1st Stage
1 — 1 Centrifuge
2l*5 QPM
C
lbs/day Recovery, % [^ .
27,783 35
>*9,523 75
)3 3,506 20
1,070 30
l*3l* 30
655 15
96 15
1*,357 1*0
87,1*21* 1*7
16,778 lbs/day
)3 12,620 lbs/day 1
2,323 Ibs/day i
2.987 lbs/day i
598 lbs/day I _, .
508 lbs/day H Tluckenei 1
022 lbs/day ^>^ ^^
1*96 Ibs/day 7"^
Oil lbs/day 3.5% TS
711* lbs/day 21(0 Ib per day 122° GPM
51*1* lbs/day of polymer - ' •
535 lbs/day at 3<0 lb " ^na stage
Ij28 per ton TS Centrifuge
J
Cake
entrate _,
.yk TS r3* TS
28 GPM ^'37 CFM, .
r i
| lbs/day Recovery, %
i Organics 38,699 75
CaC03 16,178 78
1 Ca5(OH)(POl*)3 12,620 90
| Mg (OH)2 2,122 85
i Mg 0 860 85
Fe(OH)3 3.3U3 90
1 Fe203 It89 90
ake | Inerts 5,555 88
| Total 78,866 8l
1 I
n
-n
2nd
Stage
Centrate
0.7% TS
196 GPM
J U^ , organics 12,899 lbs/day
/ \ ' CaC03 330 lbs/day
] 1 Ca5(OH)(POl*)3 1,1*02 lbs/day '
Multiple ] Kg (OH)2 371* lbs/day |
Hearth i Mg o 151 its/day i
incinerator | £g>3 371 lbs/day ]
| Inerts 980 lbs/day |
L Total l6 "562
i (
J
13,291* lbs/day p~ Ash to
1»7,1»02 j Disposal
ate Recycle
Figure 11-1 Solids Processing Balances for Incineration/Recalcination at pH 10.2
-------
FLOCCULATION CONDITIONS
New
23,'
Waste AS T
9,?U6 Ibs/day h"
[• Settleable ,
! Primary Solids
«• 53.U78 Ibs/day!
Chemical
Recycle
C80
CaC03
ca5 ton) (POL
Mg (OH)2
Mg 0
Fe(OH)3
Inerts
lime (CaO)
291 Ibs/day
| First pass precipitation !
CaC03 Id*, 208 Ibs/day
-j Ca5 (OH) (POU)3 12,370 Ibs/day j
Mg (OH)2 8,621* Ibs/day j
. Fe (OH)3 2,299 Ibs/day _|
/
\ <> Pre-
i i aeration
"\
Organios
CaCO,
Caj TOH)(PO|j)3
Mg (OH)2
Mg 0
Fe (OH)3
Inerts
Total
1
52,355 Ibs/day 1
1*,920 Iba/day 1
, 2,729 Ibs/day i
3 0 Ibs/day
l»,005 Ibs/day 1
0 Iba/day |
922 Ibs/day ,
18,559 Ibs/day
Cn hon '
Primary Clarifier *-To Aerat
1 Nitrificc
^f
Sludge
5$ TS
f391 GPM
75,238 Ibs/day ~~] »j Thickener
109,3117 ibs/day I I Thickener
18,190 Ibs/day | \^^ ^^
9,913 Ibs/day i T^
1»,605 Ibs/day | |
2,6vU Ibs/day 1 ^~~~
1,078 Ibs/day |
20,900 Ibs/day | |
2l*l,9>t5 1 1 „
P | Organics
n 1 CaC03
| 1 Cake ; Cas(OH)(POi,).
J u ** TS ^rja
f \ l'**™ | Ff(OH)3
Multiple 1 SSI.
Hearth 1
Recalcine L^*1 _
Furnace
L, CaCO,
I CajOI
x | MgO
-(Classifier 1 ,
\ J Total
^—•S
Centr
Organics 1*5,1
CaC03 10,9
.Ca5(OH)(POl*), 15,1)
Mg(OH)2 J 6,1)
Mg 0 2,9
Fe(OH)3 1,8
Fe203 7
Inerts 8,3
1*3 Ibs/day 1 T|
35 Ibs/day *l Tl
61 Ibs/day 1 ^^
•1*3 Ibs/day 1
193 Ibs/day |
72 Ibs/day , 1W Ib per day
60 Ibs/day 1 at 2.<5 Ib tier*
Total 91,962 i ton TS
ion- (_ i L.
tion i
i — i 1st Stage
1 — 1 Centrifuge
ffj, fS 239 GPM
L
Ibs/day Recovery, % i
30,095 i»o
98,1*12 90
, 2,729 15
5 3,!*70 35
1,612 35
802 30
323 30
12,51*0 60
11*9,983 62
10,716 I
(pofc), 12,369
5,9W i
s 12,21*8 l~~ Ash 1
ii3,oolt J DlsP°
ate Recycle
Cake
lentrate .a*
> M TS 16? TS
S.H'J) It> „„
>22 OPM 3-0° CFM
I Ibs/day Rei
1 Organics 31,11*9
1 CaCOo 10,716
1 Ca5(OH)(PO!*), 12,369
i MgtOH)2 a 5.151*
I MgO 2.391*
Fe(OH)3 1,1*98
' F«20-= 601*
Oke | Inerts 6,019
Total 69,903
|_
/* ^o I Organics
/ \ ' CaCOi
/ ] I Ca5(OH)(P01(}3
Multiple | Mg(oH)2
Hearth i M^u,
Incinerator ] ^°H)3
i Inerts
1 frttAl
1
0
sal
ickener
[HTS
I 188 GPM
2nd Stage I 1
Centrifuge ' — '
1
:overy, %
69
98
80
80
80 2na
80 Stage
80 Centrate
72 1.1H TS
161GPM
76
J
13,99!* Ibs/day
219 Ibs/day
3,092 Ibs/day
1,289 Ibs/day |
599 Ibs/day i
371* Ibs/day '
151 Ibs/day
2,3l»l Ibs/day |
22 059 1
Figure 11-2 Solids Processing Balances for Incineration/Recalcination at pH 11.0
-------
FLOCCULATION CONDITIONS
500 mg/l Ca(OH)2>(pH11.5)
Hew
29,
j~Waste AS 1
1 9»7'i6 Ibs/day I"
l__ 1
f" Settleable "j
^51,230 Ibs/dayJ
Chemical
Recycle
, Cao
1 CaC03
| Ca5(OH)(P01t)3
, Mg (OH)2
1 Mg 0
I Fe (OH)3
I Fe20o
: Inerts
Total
L_
lime (Cao)
61*6 tos/day
1 1 |
1 First pass precipitation | Organics 1*3, 1|
CaC03 129,198 Ibs/day i CaCO, 13,;
-| Ca5 {OH)(P01*)3 12,220 Ibs/day , Ca_ (0H)(PO|,), 15,£
Mg (OH)2 7,523 Ubs/day 1 Mg (OH)2 5,5
Fe (OH)3 0 Ibs/day | Mg 0 2,]
1_ J Fe (OH), C
' < Pre-
aeration
\
Organics
CaCOB
Ca.c (OH)(PO|()3
Mg (OH)2
Mg 0
Fe (OH) 3
Inerts
Total
1
61*,911 Ibs/day
6,101 Ubs/day
1*,563 Ibs/day
0 Ibs/day
2,685 Ibs/day
0 Ibs/day
0 Ibs/day
lit, 221* Ibs/day
92,1*6% j
t
Primary Clarifier
Vsiudge
7*TS
1291 GPM
73,563 Ibs/day ~~l
135,570 Its/day 1
19,858 Ibs/day |
8,533 Ibs/day ,
3,01*9 Ibs/day
0 Ubs/day 1
0 Ibs/day 1
ll*,89>* Iba/day ,
J
HI Cake
I Ll 52* TS
f >«1 2.5 CFM
Multiple
Hearth
Recalcine
Furnace
/ Ash \
-(Classifier \
\ /
^—^
Inerts 6,r
»-To Aeration- L^10**1 ^
Nitrification ~
3
2
*| Thickener I r~l 'St St°9e
l^ J 1 — 1 Centrifuge
"^v^' ^TS
1 252 GPM
l.
| j
i Ibs/day Recovery,* L
Organics 30,161 Ul
CaCC3. 122,013 90
1 Ca5 (OH)(POlt)3 1*,563 23
1 Mg 0 915 30
| Fe (OH)3 0
1 Inerts 8,192 55
J Total 168, UoU 66
CaCOj 13,286 Ibs/day
1 Cac toH)(POl^)3 12,220 Ibs/day 1
Mg 0 5,198 Ibs/day |
FepO^ 0 Ibs/day ,
~l i
57 Ibs/day 'M "
75 Ibs/day ^^»
73 Ibs/day
s31* 2$S l67 ^^ of
! IBs/day polMier at .-
Iba/dav po^ioer at ^
'02 Ibs/day 3,8 Ib/ton TS
U3 j
entrate Calce
.1* TS 13* TS
33 GPM 5.1*2 CFM
r
| Ibs/day Be
l Organics 30,815
CaC03 13,286
1 Caj (OH)(POI,)3 12,220
1 Mg (OH)2 U.958
| Mg 0 1,771
| Fe (OH) 3 0
'oke | Inerts 6,032
1 Total 69,082
L
n
ickener
3.1* TS
| 2OO GPM
2nd Stage i — i
Centrifuge ' — '
1
covery,*
71
98
80
83
83
90
80
J
2nd
Stage
Centrate
0,9* TS
163 OPM
/} L^ Organics 12,587 Ibs/day i
/ \ CaCO, 271 Ibs/day
] Ca5 tOH)(POl*)3 3,055 Ibs/day 1
Multiple "8 (OH)2 1,015 lis/day |
Hearth "« ° 363 Ibs/day ,
Hearth Fe (OH) Q lbs/d 1
Incinerator Fe2o 3 0 ^/^ \
Inerts 670 Ibs/day i
| Total
i
17,961 j
| Inerts 12,195 Ibs/day ^" ^sti to
1 Total 1*2,899 Disposal
Centrate Recycle
Figure 11-3 Solids Processing Balances for Incineration/Recalcination at pH 11.5
-------
Table 11-1
Material Balance Comparisons at Various pH Levels
pll
,0,2
.1.0
1,5
Primary
s Judge
production
11) per day
1*4,000
242,000
255,000
1st Stage
cake to
recalcining
Ib per day
S7.000
150,000
1GS.OOO
percent
TS
42
5S
52
Recycled
solids
from
recalcining
Ib per day
44,000
83,000
92,000
2nd Stage
cake to
incinerator
Ib per day
79,000
70,000
09,000
percent
TS
18
IS
13
Ash to
disposal8
Ib per day
47,000
43,000
43,000
Makeup lime
Ib per day
30, 000b
25,000b
32,000b
Percent
of total
CaO dose
52
31
31
From incine'-ation of 2nd stage cake.
Basis 94<;; CaO.
Table 11-2
Chemical Costs for pH Processing Modes in
Incineration/Recalcination Flow Scheme
2
0
3
FeCl3a
Ib per day
o,oood
3,500g
none
cost
$270
$160
Primary
Polymer"
Ib per day
036
none
none
cost
$95
-
-
Centrifuge
Polymer"5
Ib per day
240f
115h
1671
cost
$300
$170
$250
New ltmec
Ib per day
30,000
25,000
32,000
cost
430
360
450
Total
daily
cost
$1155
$ 690
$ 700
Total
annual
chemical
cost
$422,000
$252,000
$256,000
«ed on delivered cost of $90 per dry ton.
ised on delivered cost of $1.50 per Ib.
sed on delivered cost of $28.69 per ton of CaO (941).
se 24 mg/1.
se 0.25 mg/1
Dose 5.0 Ib/ton DS.
gDose 14 mg/1.
Dose 2.5 Ib/ton DS.
Dose 3.8 Ib/ton DS.
Ill
-------
to the primary clarifier, and amounts to 22 percent of the solids appearing
in the underflow. Without wet classification, recycled inerts would build up
to levels constituting several times the solids load that would be obtained if
the solids were just incinerated and hauled away.
Chemical Costs. While a full economic comparison was beyond the scope of
this study, it was possible to calculate new chemical costs for each pH level
(Table 11-2) .
Since pH 10.2 operation is characterized by high requirements of ferric
chloride and polymer, and a higher makeup lime requirement than that for
pH 11.0 operation, chemical costs are significantly higher for pH 10.2 opera-
tion ($1,155 per day) than for either pH 11.0 ($690 per day) or pH 11.5
($700 per day) operations.
Chemical costs for pH 11.0 operation are lower than those at pH 11.5 because
of both the lower makeup lime and lower centrifuge polymer requirements.
However, if centrifuge polymer requirements for pH 11.5 can be reduced
from the value used in the flow sheet balance (3.8 Ib per ton dry solids) ,
then pH 11.5 operation would be the least expensive of the three pH modes
in terms of chemical purchase costs.
The polymer costs were scaled up from the centrifuge test work where the
Sharpies P600 machine was used at pH 10.2 and 11.5. The polymer dose at
pH 11.0 was based on P3000 experimental data. There is reason to believe that
the polymer doses required for pH 10.2 and 11.5 operations may be excessive,
since comparisons of polymer requirements at pH 11.0 for both the P600 and
P3000 machine showed that more polymer was required for the smaller P600
machine.
Digestion and Land Disposal Alternative
In lieu of including recalcining and incineration furnaces in the solids pro-
cessing sequence, another sequence using digestion with land disposal of
sludge was evaluated {Fig. 11-4) .
This process includes wet classification for separation of the calcium carbonate
from the rest of the sludge. This heavy calcium carbonate would pose a
mixing problem in the digester if left in the sludge. The calcium carbonate-rich
first stage cake can be safely disposed of on land due to its high pH. The
high pH both kills pathogens and renders the sludge stable during drying^.
The first stage centrate is then thickened in a second stage centrifuge produc-
ing a 15 percent total solids cake. A small portion of the centrate is used to
dilute the cake to 8 percent for digestion, while the remaining second stage
centrate is recycled to the primary. The digested solids would be disposed
of in lagoons and drying beds. Supernatant from the lagoon would be pumped
to the primary tank.
112
-------
FLOCCULATION CONDITIONS
494 mg/l Ca(OH>2, (pHll.O), 14 mg/l f<£\$, 0.25 mgfl Polymer
aita AS 1
|9,7k6 Iba/dayJ
!u»a (CaO) 1
|93,kOO lb«/dayj
j"
* '«• pre-
30 MOD Bav Se«ag«^ „„,„2 " 11,515 Ib./dny I
• (OH)3 2,187 Ib./day i
^ ' '"^a*
1
1 j_
17,150 lb,/day 1 1 organic. 1,715 Ib./day ]
271 Ib./day 1 I caCOl 27 Ib./day 1
3,029 Iba/day . Ca5(OH)(PO,)i, 303 Ib./day I
2,303 Ib./day 1 1 i|g (OH)J 230 Ib./day i
k3U Iba/day j j Fe (OHH k3 Ib./day 1
23.187 | j Total 2,316
387 «•» 308'5*T
O.HTS
oae 1 — 1 t Organic.
— LJ 1 c«e°3
u'e 1 Ca^fOBXPOli)!
en. kl.Ugp.crt. 15* TS 1 f< «*3
5.5 cfm | | 5^5,!
day 1 1 Organic. 38,
day I CaC03 13,
day ' Ca5(OH)(POi1)3 12,
day 1 Kg (OH); 9,
day ] Ta (OH)3 I,
I Total 168,735 I i Total 7k,
Supernatant
Recycle
Organic. 3,736 Ib./day
| CaC03 l,lkk Ib./day
i Ca^fOH (Poij)} 1,091 Ibc/day 61.9 «
1 W !« J 155 Sl'/aS "* **•"»"
i Total 6,126 |
Kg** 3017 Ib/day
SB, 0.83* 3S, l.S* TS 1 o,oon
y 20,380 Bg/1 a. CaCO^
XXXXXX-is
.15
S«ttlad 3
1
173 Ib./day I . .X^^^x,
289 Ib./day 1 f <*
118 Ib./day I Digester
753 tb./Jay 8* TS ""' " " ""*
5,,5 ' w.i no 5.3* TS
1 Organic. 8,790 tb./day 1
.J oaco3 5,909 Ib./day I
TS Ca5(dH)(PO|,)j 5,633 Iba/day 1
cfla | Kg (OH)2 o Iba/day |
Luin* r* (OKh 816 ib./day
Total 21,lk8
_J
39,888 Ib./day 1 1 Organic. 23,1
13,316 Ib./day 1 1 CaCOj 13,0
12,k4l Ib./day I C.5(OH)(PO|,)3 U,k
9,9k2 Ib./day 1 1 Kg (OH)2 0
1,796 Ib./day | Pe (OH)3 1,7
- 1
3k Ib./day 1
127 Ib./day I
21 Iba/day 1
Ib./day 1
96 Ui./day I
76,863 J | Total 50,378 j
_,S Kg** 3906 Ib/day
^.^-•^ oa** 288 Ib/day
63gp» 3.63* TS
m'ra Organic. 10,608 Ib./day
• «» CaCOj 5,97k lbi/d«y
Sl»dg. c.5(OII )(POU), 5,697 Ib./toy
to dry- 1 Kg (OH}2 0 Ib./day
Ing bed | p. (on), 825 Ib./Jay
| Tatal 23,10k
Kg** 833 Ib/day
i
Organic. Ut,526 Ua/day 1
CaCOj 7,053 lb./*ay
Ca5(«ll)(rX)|,)3 6,7Jk Iba/day 1
Hg (CH)a 0 U»/«ay I
1 P. (OR)* ^^ lba/4ay
_| total 27,17k
Kg** 3O72 Ib/day
Ib/day
Figure 11-4
Solids Processing Balances for Digestion-Land Disposal at pH 11.0
-------
Assumptions. Primary solids production and centrifuge operation has the
same bases as the previous alternative presented.
Digester operation was based on the experimental test results presented in
Section IX. These studies indicated that essentially all of the magnesium
hydroxide is solubilized during the digestion process, while only a small
portion of the calcium carbonate and hydroxyapatite fed to the digester is
dissolved. The small incremental effect of calcium and phosphorous recycle
to the primary was neglected. The main effect of returning the digester
supernatant stream to the primary is the increase in lime required to react
with the additional bicarbonate alkalinity and magnesium.
Digester design is based on a solids retention time of 25 days with a volatile
matter destruction of 42 percent. Lagoon performance is based on the plant
staff's experience with their present lagoons.
Mass Balance Comparisons. Table 11-3 summarizes the calculated results
for the digestion/land disposal alternative and compares them to the recal-
cination/incineration alternative for pH 11.0.
The digestion alternative produces only slightly more primary sludge than
does recalcination, but produces significantly greater quantities of solids
for disposal. The makeup lime requirements are substantially greater without
recalcination of the first stage cake. Total lime dose for the digestion alternative
is greater than for the recalcination case because of the additional soluble
magnesium and alkalinity being returned to the primary via the digester
supernatant stream.
Chemical Costs. New chemical costs are estimated to be $635,000 per year
instead of $252,000 per year for the incineration alternative. The greater
cost is primarily due to the fact that no lime is recycled in the process and
that the total lime requirement is greater for the digestion alternative.
Comparison to Other Digester Operations. To place the digestion alterna-
tive into perspective a comparison was made to the digester operations in a
conventional plant, and an activated sludge plant as shown in Table 11-4.
For a 30 mgd plant capacity, activated sludge treatment produces more sludge
(and volatile matter) to be digested than either a primary plant or an ATTF
type plant using digestion to process waste solids. Based on the same raw-
sewage composition and flow rate, an ATTF type plant produces slightly
more waste solids to digest and requires slightly more digester capacity than
does a primary treatment plant. Also, because volatile matter destruction
is greater for a conventional digester than for a lime sludge digester, more
methane production can be expected from the conventional primary digester.
114
-------
Table 11-3
Comparisons Between Recalcination and Digestion Solids Processing
Alternatives at pH 11.0 for 30 MGD Plant
Alternative
Recalcination
Digestion
Primary sludge production
Ib per day
242,000
266,000
Lime additions4
Ib per day
25,000
99,000
Total limeb '
dose
Ib per day
76,000
93,000
Total solids to
land fill
Ib per day
43,000C
291 , 000d
"As
CaO.
d
As 100% CaO.
Ash from incineration.
Calcium carbonate rich cake; excludes digested solids removed by top
soil company; wet basis.
Table 11-4
Comparisons of Large Scale Digester Operations
Calculated for Three Types of Treatment Plants
Plant
type
Primary
plant a
Activated
sludge
plant13
ATTF
type
plant0
CCCSD
present
plantd
Capacity
MGD
30
30
30
23
Total sludge to digesters
Ib per day
38,950
60,650
76,880
35,496
gal per day
94,000
146,000
115,000
73,000
TS, f
5.0
5.0
a.o
5.8
VM to
digesters
Ib per day
29,000
45,000
32,000
I
27,000
Total digester
capacity
cu ft.
312,000e
487,000e
386,000e
344,000
Cu ft.
CH4 per
Ib VM
destroyed
10.0
10.0
11.0
10.7
Cu ft.
CH4
per day
169,000
264,000
147,000
174,000
SRT,
days
25
25
25
35
aBased on influent SS of 240 mg/1 with 65% SS removal; 75$ VM in sludge solids; 58$ VM
destruction.
ABased on same data as (a) above, with additional assumption of 401 removal of influent 216
mg/1 BOD across primary, and .67 Ib SS produced per Ib BOD fed to activated sludge (AS).
CBased on digestion of 1st stage centrate solids, (42$ VM); pH 11.0 mode, with FeClg and
polymer addition; 425 VM destruction, with supernatant recycle to primary.
^Actual CCCSD operation for 1971; using three primary and one secondary digester; digester
gas 37S CO2; 76
-------
SECTION XII
ACKNOWLEDGMENTS
This investigation was a joint effort of the Central Contra Costa Sanitary
District (CCCSD) and its consultant, Brown and Caldwell. Studies conducted
on solids processing were part of a larger research effort conducted at
CCCSD1 s Advanced Treatment Test Facility (ATTF) . The ATTF was authorized
by the CCCSD Board of Directors, who saw the need to encourage the develop-
ment of tailored treatment systems suited to the purposes of reclaiming sewage
for industrial use and effluent discharge. Board members are: Don L. Allan,
President (1972) , Parke L. Boneysteele, Director, Charles J. Gibbs, Director,
Richard J, Mitchell, Director, and George A. Rustigian, Director. G. A.
Horstkotte, Manager-Engineer and David G. Niles, Water Reclamation Project
Manager, supervised the district's water reclamation plant expansion and
research efforts.
CCCSD STAFF
Under the direction of David G. Miles, plant superintendent, the CCCSD
treatment plant staff was involved in the research program on a daily basis.
Routine operation of the liquid processing system was performed entirely by
the main plant's operators. Virtually all equipment installation and plant
modifications were performed by the plant's maintenance staff, directed by
CCCSD's mechanical engineer, Fred S. Markey Chemical analyses of the
liquid processing system were performed by the plant chemical section under
the direction of Walter Fox, plant chemist. Keith Packwood had specific
responsibility for scheduling the analytical program.
BROWN AND CALDWELL STAFF
Within Brown and Caldwell, the overall plant expansion effort was super-
vised by Dr. David H. Caldwell, President of Brown and Caldwell. David
L. Eisenhauer served as Project Manager for the design of the plant expansion
and also supervised the design of the Advanced Treatment Test Facility.
Dr. Denny S. Parker directed process research at the ATTF. On the
Research staff at the ATTF were Fred Zadick, environmental/chemical engi-
neer, James Tyler, supervising chemist, and Kenneth Train, chemist/
chemical engineer.
OTHER CONTRIBUTORS
Environmental Quality Analysts, a division of Brown and Caldwell, performed
the detailed sludge analyses.
117
-------
The Sharpies Division of the Pennwalt Corporation participated in the
research on lime sludge centrifugation, Ken Kyte of the Warminster office
was particularly helpful, as he was able to utilize his experience on centrifu-
gation of lime sludges obtained at several other plants. Clem O'Donnell and
Jim McCormick, from the San Francisco office of Pennwalt Corporation,
assisted in many of the experimental runs.
A number of equipment manufacturers loaned equipment which significantly
expanded the scope and capability of the research program. The Sharpies and
Wallace and Tiernan Divisions of the Pennwalt Corporation loaned centrifuges,
sludge pumps, and a variable speed chemical feed pump. The Dow Chemical
Company loaned chemical injectors, a variable speed chemical feed pump,
and a Lightnin mixer for mixing polymer solutions. The Komline-Sanderson
Corporation loaned a dissolved air flotation unit equipped with influent and
chemical feed pumps. The Bird Machinery Corporation loaned a centrifuge
and participated in several days of testing. The Pacific Engineering and Pro-
duction Company of Nevada loaned the Pepcon chlorine oxidation system and
operated it for a week.
This project was funded from several sources. The California State Water
Resources Control Board and U.S. Environmental Protection Agency each
contributed a grant to the district. The district contributed an equal amount.
Project officers for the State Board and EPA were KurtWassermann and Robert
Dean, respectively.
REPORT PREPARATION
This report was written by Brown and Caldwell. Principal authors were
Denny S. Parker, Fred Zadick and Kenneth Train.
118
-------
SECTION XIII
REFERENCES
1. Horstkotte, G.A., Niles, D.G., Parker, D.S., and Caldwell, D.H, "Full
Scale Testing of a Water Reclamation System," presented at 45th Annual
Conference of the Water Pollution Control Federation, Atlanta, Georgia,
October 12, 1972.
2. Weller, L.W., "Design of Sewage Treatment Plant Additions at Rochester,
New York," presented at 45th Annual Conference of the Water Pollution
Control Federation, Atlanta, Georgia, October 12, 1972.
3. Black, S.A., "Lime Treatment for Phosphorus Removal at the New-
market/East Gwillionbury WPCP", Ontario Ministry of the Environment,
Research Branch Paper No. W3032, May 1972.
4. Schmid, L.A. andMcKinney, R.E., "Phosphate Removal by a Lime-
Biological Treatment Scheme", JWPCF, 41, No. 7, pp 1259-1284 (1969) .
5. Gaines, F.R. and Hartley, C.E., "Advanced Waste Treatment in Hatfield
Township," Water Pollution Control Association of Pennsylvania
Magazine, p. 8., May-June 1972.
6. Burns, D.E., and Shell, G.L., "Chemical-Powdered Carbon Treatment
of a Municipal Wastewater,',' Journal of the Water Pollution Control Federa-
tion, in press, 1972.
7. California State Water Resources Control Board, "Water Quality Control
Plan for Ocean Waters of California", Resolution No. 72-45, adopted
July6, 1972.
8. Central Contra Costa Sanitary District and Contra Costa Water District,
"Municipal Wastewater Renovation Pilot/Demonstration Porject", draft
report submitted to the Environmental Protection Agency, April 1972.
9. Barth, E.F., Brenner, R.C., and Lewis, R.F., "Chemical-Biological
Control of Nitrogen and Phosphorus in Wastewater Effluent", Journal
Water Pollution Control Federation, 40, p. 2040 (1968) .
10. Mulbarger, M.C., "The Three Sludge System for Nitrogen and Phos-
phorus Removal", paper presented at the 44th Annual Conference of the
Water'Pollution Control Federation, San Francisco, California,
October 1971.
119
-------
11. Keeler, C.E . , "Relationship of Sludge Density Index to the Activated
Sludge Process", Journal Water Pollution Control Federation, 35, 1166
(1963).
12. Christensen, M.H. and Harrernoes, P., Biological Denitrification in
Water Treatment, A Literature Survey, Report 2-72 Department of
Sanitary Engineering Technical University of Denmark, 1972.
13. McCarty. P.L., "Anaerobic Waste Treatment Fundamentals, Part Three,
Toxic Materials and Their Control" , Public Works, 91-94, Nov. 1964.
14. Graef, S.P., and Andrews, J.F., "Process Stability and Control
Strategies for the Anaerobic Digester," presented at the 45th Annual
Conference, Water Pollution Control Federation, Atlanta, Georgia,
October 8-13, 1972.
15. Welch, P.R., "Digestion Experiments at Flint, Michigan, on Sewage
Sludge, Garbage, and Lime Sludge Mixtures," Sewage Works Journal,
10, No. 2, pp 247-260, March 1938.
16. Van Fleet, G.L., Barr, J.R. and Harris, A.J. , "Treatment and Disposal
of Chemical Phosphate Sludges in Ontario," presented at the 45th Annual
Conference, Water Pollution Control Federation, Atlanta, Georgia,
October 8-13, 1972.
17. Baskerville, R.C., and Gale, R.S., "A Simple Instrument for Determin-
ing the Filterability of Sewage Sludges," Journal of the Institute of
Water Pollution Control, No. 2, p. 3, 1968.
18. Gulp, G.L., "Physical-Chemical Treatment Plant Design," EPA
Technology Transfer Design Seminar, Pittsburgh, Pennsylvania,
August 1972.
19. Caldwell, D.H. and Lawrence, W.B., "Water Softening and Conditioning
Problems," Industrial and Engineering Chemistry. Vol. 45_, No. 3,
pp 535-540, March 1953.
20. Smith, J.E., Hathaway, S.W., Farrell, J.B. and Dean, R.B., "Lime
Stabilization of Chemical-Primary Sludges at 1.5 mgd," presented at
the 45th Annual Conference, Water Pollution Control Federation, Atlanta,
Georgia, October 8-13, 1972.
21. Mulbarger, M.C., Grossman, E., Dean, R.B. and Grant, O.L., "Lime
Clarification, Recovery, Reuse, and Sludge Dewatering Characteristics,"
11' 12- PP 2070-2085,
December .969.
120
-------
22. Cohen, J.M. and Kugelman, I.J., "Physical-Chemical Treatment for
Wastewater," Water Research. Vol. 6, pp 487-492, 1972.
23. Nilsson, Rolf, "Removal of Metals by Chemical Treatment of Municipal
Waste Water," Water Research, Vol. 5, pp. 51-60, May 1971.
121
-------
SECTION XIV
GLOSSARY
arnp ampere (s)
AS activated sludge
C degrees Centigrade
Cc constituent in total solids of centrifuge cake, weight percent
Ct total solids in centrifuge cake, weight percent
cfm cubic feet per minute
cu ft cubic feet
DS dry solids
e^ total solids in centrituge centrate, weight percent
F degrees Farenheit
fe constituent in total solids of centrifuge feed, weight percent
f* total solids in centrifuge feed, weight percent
L rt
g gravitational force per pound mass, feet per (second)
gal gallons
gpd/sf gallons per day per square foot
gpm gallons per minute
in. inches
Ib pound
LOI loss on ignition
mg/1 milligrams per liter
MLVSS mixed liquor volatile suspended solids
M/N methanol to nitrogen (mass ratio)
NHg-N ammonia nitrogen
ppm parts per million
RC weight percent recovery of a given constituent
R-j- weight percent recovery of total solids
A RPM difference in revolutions per minute. For centrifuge,
A RPM is bowl speed minus auger speed.
sq ft square feet
sqin. square inches
SRT solids retention time
SS suspended solids
SVI Sludge Volume Index
TOC total organic carbon
TS total solids
TSb total suspended solids
VM volatile matter, interchangeable with volatile solids in this
report
VS volatile solids (or volatile matter)
VSS volatile suspended solids
WAS waste activated sludge
wt weight
123
-------
SECTION XV. APPENDIX
125
-------
5OOO ~\ 250O
4500
4000
2
•^
-J
3500
3000
25OO
2OOO
6
.
/5OO
§
1000
500
IO 2O 30 40
TIME SINCE STARTUP, DAYS
Figure A-l
Lime Digester Operation During Low Feed Solids Period for SRT of 15 Days
VOLATILE ACIDS
IO 2O 30 40
TIME SINCE STARTUP, DAYS
5O
5750 "I
5500
5000
45OO .
x
k
-•.
%
^fc
~j
4000 5
3500
3000 -I
2750
2500
2000
1500
Uj
1000
-j
o
5OO
Figure A-2
Lime Digester Operation During Low Feed Solids Period for SRT of 20 Days
126
-------
/ v , /£>—<* \ ALKALINITY
y
5750 -12750
6.9
IO 20 30 4O
TIME SINCE STARTUP, DAYS
50
55OO
5000
4500
4OOO
350O
25OO
2000
6
Q.
Q.
K-
s
/500
Q
o
^
Ui
1000
-J
o
5OO
-1 3OOO -1 0
Figure A-3
Lime Digester Operation During Low Feed Solids Period
forSRTof 25 Days
127
-------
7.9
7.8
7.7
7.6
7.5
7.4
7.3
7.2
7.1
\
6.7
i
\ /\
V \
/\ x ALKALINITY
I \ . o
/ \ \ '* *"
' \ \ / \ xx
' \ \ / \ /
( \ V \ X
\
VOLATILE
ACIDS
\
>,
\
\
o
5 IO 15 20
TIME SINCE STARTUP. DAYS
25
33OO
32OO
3IOO
3000
29OO
28OO a
27OO ^
26OO -J
25OO ^
24OO
23OO
22OO
2IOO
300O
2500
2OOO 6
£
0)
Q
/50O O
Uj
IOOO
500
O
Figure A-4
Control Digester Operation During Low Feed Solids Period
forSRTof 15 Days
128
-------
7.9
7.8 —
5600
55OO
5400
5300
52OO
5IOO
5OOO
49OO
4 BOO
47OO
46OO
45OO
44OO
43OO
420O
—\2200
2IOO
2OOO
I9OO
I80O
1700
e
Q.
I60O Q'
/5OO 2
I40O ^
-j
/3OO £
o
I20O ^
1100
IOOO
9OO
BOO
12
22 32
TIME SINCE STARTUP, DAYS
figure A-5
lime Digester Operation During High Feed Solids Period
for SRT of 10 Days
123
-------
6.9
6OOO
55OO
500O
x
k.
450O $ —
4OOO
35OO —
-13000
I5OO
I4OO
I3OO
I2OO
1100
Q.
lOOO^-
900 5
BOO iy
•j
7OO
6OO
5OO
4OO
•j
O
58 68 78 88
TIME SINCE STARTUP, DAYS
Figure A-6
Lime Digester Operation During High Feed Solids Period
forSRTof 15 Days
130
-------
3;
Q.
8.4
8.3
8.2
8.1
8.O
7.9
7.8
7.7,
7.6
7.5
7.4
A
P
ALKALINITY
-I >
/ f \ - .
-/ / \VOLATILE ACIDS
i I \ A
- / V ^—\v^
1
1
48 58 68 78 88
TIME SINCE STARTUP, DAYS
55OO
5OOO
450O
4OOO
35OO
e
Q.
CL
—J O
IOOO
9OO
8OO
7OO
6OO
500
40O
3OO
2OO
IOO
98
Figure A-7
Lime Digester Operation During High Feed Solids Period
for SRT of 20 Days
131
-------
a:
es.
8.O
7.9
7.8
7.7
7.6
7.5
7 4
' > 1 1
Q
7\
1 » Q
, \ *
1 \ n
_ / \ 1 1
•'•MU
fk ' \ 1 a 6
- fl N>J \4 \ALKALINITY
VI \
/
-
—
7 i ^
M
xx^
1 -
1
I 1
A » ^ t _|^
Uyk J\ >viuAK
^ VyOLATILE ACIDS
i 1 1 1
—
—
—
—
_
—
-
—
-
t*HJU
4300
42OO -
4 IOO -
4OOO g
390O - -
X
38OO ^
37 OO 2
-j
360O -
35OO
3400
3300
•* nnf\
IdUV
IIOO
IOOO
9OO
BOO
7OO
6OO
5OO
400
300
20O
IOO
n
48 58 68 78 88 98
TIME SINCE STARTUP, DAYS
Figure A-8
Lime Digester Operation During High Feed Solids Period
forSRTof 25 Days
132
-------
24
4OOO
3900
380O
3700
36OO
35OO
e
Q.
x
340O -
33OO
3200
3/00
3OOO
290O
280O
I2OO
I IOO
1000
9OO
BOO
700
6OO
500
4OO
3OO
2OO
IOO
O
§:
34
44 54 64 74
TIME SINCE STARTUP. DAYS
Figure A-9
Control Digester Operation During High Feed Solids Period
forSRTof 15 Days
133
-------
I
o
Q:
K
«0
Ul
Q
fc
Uj
Ul
-J
c
*t
o
It
Ul
0.
0
K
u.
o
25
20
15
10
I
I
AVERAGE VOLATILE MATTER
DESTRUCTION 34.2%
STABLE VOLATILE
MATTER DESTRUCTION
PERIOD
STARTED
HEAVY FEED
10 20 30 4O
TIME SINCE STARTUP, DAYS
50
Figure A-10
Unit Gas Production for Lime Digester During Low Feed
Solids Period, SRTof 15 Days
134
-------
30
Q
SK
o
Q
Ct
K
K.
Uj
-J
-J
O
Q
Ct
Ul
Q.
K.
u.
15
10
1 1 1
AVERAGE VOLATILE MATTER
DESTRUCTION 40.4%
STABLE VOLATILE
MATTER DESTRUCTION
PERIOD
STARTED-
HEAVY FEED
10 20 30 40
TIME SINCE STARTUP, DAYS
50
Figure A-ll
Unit Gas Production for Lime Digester During
Low Feed Solids Period, SRT of 2Q Days
135
-------
30
Q
Ui
O
Ui 25
(fc
Uj
K
K.
1
5 20
Q
-J
,5
10
K-
U.
o
I 1 I
AVERAGE VOLATILE MATTER
DESTRUCTION 37.2%
STABLE VOLATILE
MATTER DESTRUCTION
, PERIOD ,
HEAVY FEED
I
10 20 30 40
TIME SINCE STARTUP, DAYS
50
Figure A-12
Unit Gas Production for Lime Digester During Low Feed
Solids Period, SRTof25Days
136
-------
Q
Ul
O
Ul
Q
19
I-
U.
35
30
25
20
Ui
-j
O
It)
Q: 15
Ul
a.
10
-I— 1 1
AVERAGE VOLATILE MATTER
DESTRUCTION 49.6%
STARTED
HEAVY FEED
STABLE VOLATILE MATTER
DESTRUCTION PERIOD
I
10 15 20 25
TIME SINCE STARTUP, DAYS
30
Figure A-13
Unit Gas Production for Qontrol Digester During Low
Feed Solids Period, SRT of 15 Days
137
-------
18
Q ,7
Ul If
O
i is
Ul
Ul
Ul
•J
c
1
-J
o
Ul
Q.
to
o
15
14
13
12
II
10
9
8
7
6
AVERAGE VOLATILE MATTER
DESTRUCTION 33.6%
STABLE VOLATILE MATTER DESTRUCTION PERIOD
i
I
10 15 20 25
TIME SINCE STARTUP - DAYS
30
35
Figure A-14
Unit Gas Production for Lime Digester During High Feed
Sol ids Period, SRTof 10 Days
138
-------
Ul
x
o
17
16
15
to
Uj
Q 14
«t
U4
13
1 „
Ul
•4
r II
§ 10
Ul
STABLE VOLATILE MATTER DESTRUCTION PERIOD
AVERAGE VOLATILE MATTER DESTRUCTION 38.3%
48
53
58 63 68 73 78
TIME SINCE STARTUP, DAYS
83
88
Figure A-15
Unit Gas Production for Lime Digester During High Feed
Solids Period, SRTof 15 Days
139
-------
25
Q
Ul
O
f~
(Q
-J UJ
Q
Ul Q.
0-2
U. Ul
o
is
10
AVERAGE VOLATILE MATTER DESTRUCTION 38.8%
STABLE VOLATILE MATTER
. DESTRUCTION PERIOD .
l
48
53
58 63 68 73
TIME SINCE STARTUP, DAYS
78
83 85
Figure A-16
Unit Gas Production for Lime Digester During High Feed
Solids Period, SRT of 20 Days
30
25
20
15
10
Si!
CO
bj
i r i i r i
AVERAGE VOLATILE MATTER DESTRUCTION 41.8%
STABLE VOLATILE MATTER DESTRUCTION PERIOD
lu
-J
O
Uj
0.
CO
(a
o
49 54 59 64 69 74
TIME SINCE STARTUP , DAYS
79
84
89
Figure A-17
Unit Gas Production for Lime Digester During High Feed
Solids Period, SRTof25Days
140
-------
25
1 1 1 1 r
AVERAGE VOLATILE MATTER DESTRUCTION 54.4%
STABLE VOLATILE MATTER DESTRUCTION PERIOD
I
35 45 55 65 75
TIME SINCE STARTUP, DAYS
85
95
Figure A-18
Unit Gas Production for Control Digester During High
Feed Solids Period, SRT of 15 Days
141
(.U.S. GOVERNMENT PRINTING OFFICEU973 546-308/17 1.5
-------
SELECTED WATER
RESOURCES ABSTRACTS
INPUT TRANSACTION FORM
;. Report No.
2.
4. Tith SLUDGE PROCESSING FOR COMBINED
PHYSICAL-CHEMICAL-BIOLOGICAL SLUDGES
7. Author(s)
Parker, D.S., Zadick, F.J., and Train, K. E.
9. Organization
BROWN AND CALDWELL CONSULTING ENGINEERS ,
San Francisco, California
3. Accession No.
W
5. Report Date
6.
8. Performing Organization
Report No.
10. Project No.
17080 FSF
!2. Sponsoring Organization Central Coatra Costa Sanitary District,
is. supplementary Notes Walnut Creek, California
Environmental Protection Agency report number,
EPA-R2-73-250, July 1973.
//. Contract/Grant No.
R801445
13. Type of Report and
Period Covered
16. Abstract
Full scale combined sludge generation from a treatment sequence consisting of lime
clarification, nitrification, and denitrification was studied. Pilot scale studies were
conducted to wet-classify^and dewater the combined sludges by means of two-stage
solid bowl centrifugation. Anaerobic digestion of first stage centrate was also studied
en"8 pilot scale. Predicted and measured sludge production agreed well using four
coagulation modes in the primary. First stage centrif ugation (wet classification),
over a wide pH range, achieved high capture of calcium carbonate in relatively dry
(42 to 57 percent total solids) cakes as well as good rejection of magnesium,
phosphorus, and iron compounds in the centrates. At pH 11.0 or below, second stage
centrifugation dewatered first stage centrate to produce 18 percent total solids cakes,
with 80 percent solids recoveries. Dewatering deteriorated at a pH greater than 11.0.
Anaerobic stabilization of thickened and unthickened first stage centrates showed
volatile matter destructions of over 40 percent, high methane-content gas, and
substantial increases in soluble magnesium and alkalinity during digestion.
17a, Descriptors
*Sludge Disposal, *Sludge Treatment, *Ultimate Disposal, *Dewatering,
*Chemical Precipitation, *Sludge Digestion, Centrifugation, Solid Wastes
17b. Identifiers
*Chemical Sludge Processing, *Centrifugal Classification, *Centrifugal Dewatering,
*Recalcination ,Chemical Treatment.
17c. COWRR Field & Group OS E , OS D
18. Availability
19. Security Class.
(Report)
20. Security Class.
(P*ge) -
21. Wo. of
Pages
22. Price
Send To:
WATER RESOURCES SCIENTIFIC INFORMATION CENTER
U.S. DEPARTMENT OF THE INTERIOR
WASHINGTON. D. C. 20240
Abstractor D. S. Parker
^institution Brown and Caldwell Consulting Engineers
WRSIC 102 (REV. JUNE 1971)
GP 0 9 tS.261
------- |