&EFK
United States
Environmental Protection
Agency
Industrial Environmental Research
Laboratory
Research Triangle Park NC 2771 1
EPA-600/7-79-248
November 1979
Treatability and
Assessment of Coal
Conversion Wastewaters:
Phase I
Interagency
Energy/Environment
R&D Program Report
-------
RESEARCH REPORTING SERIES
Research reports of the Office of Research and Development, U.S. Environmental
Protection Agency, have been grouped into nine series. These nine broad cate-
gories were established to facilitate further development and application of en-
vironmental technology. Elimination of traditional grouping was consciously
planned to foster technology transfer and a maximum interface in related fields.
The nine series are:
1. Environmental Health Effects Research
2. Environmental Protection Technology
3. Ecological Research
4. Environmental Monitoring
5. Socioeconomic Environmental Studies
6. Scientific and Technical Assessment Reports (STAR)
7. Interagency Energy-Environment Research and Development
8. "Special" Reports
9. Miscellaneous Reports
This report has been assigned to the INTERAGENCY ENERGY-ENVIRONMENT
RESEARCH AND DEVELOPMENT series. Reports in this series result from the
effort funded under the 17-agency Federal Energy/Environment Research and
Development Program. These studies relate to EPA's mission to protect the public
health and welfare from adverse effects of pollutants associated with energy sys-
tems. The goal of the Program is to assure the rapid development of domestic
energy supplies in an environmentally-compatible manner by providing the nec-
essary environmental data and control technology. Investigations include analy-
ses of the transport of energy-related pollutants and their health and ecological
effects; assessments of, and development of, control technologies for energy
systems; and integrated assessments of a wide range of energy-related environ-
mental issues.
EPA REVIEW NOTICE
This report has been reviewed by the participating Federal Agencies, and approved
for publication. Approval does not signify that the contents necessarily reflect
the views and policies of the Government, nor does mention of trade names or
commercial products constitute endorsement or recommendation for use.
This document is available to the public through the National Technical Informa-
tion Service, Springfield, Virginia 22161.
-------
EPA-600/7-79-248
November 1979
Treatability and Assessment
of Coal Conversion Wastewaters:
Phase I
by
P.C. Singer, J.C. Lamb III,
F.K. Pfaender, and R. Goodman
University of North Carolina - Chapel Hill
Department of Environmental Sciences and Engineering
Chapel Hill, North Carolina 27514
Grant No. R804917
Program Element No. EHE623A
EPA Project Officer: N. Dean Smith
Industrial Environmental Research Laboratory
Office of Environmental Engineering and Technology
Research Triangle Park, NC 27711
Prepared for
U.S. ENVIRONMENTAL PROTECTION AGENCY
Office of Research and Development
Washington, DC 20460
-------
ABSTRACT
The objectives of this project are to assess the environmental impact
of wastewaters originating from the production of synthetic fuels from coal
and to evaluate various technologies for the treatment of these
wastewaters. The major focus to date has been on aerobic biological
treatment which is projected to be the principal means of removing organic
impurities from these wastewaters and a cornerstone of any overall
wastewater treatment program.
A synthetic wastewater, designed to simulate a real conversion process
wastewater, was formulated and fed to a series of aerobic biological
reactors. Design and operation of the reactors is described, along with
performance data spanning two six-month periods of operation. In addition
to TOC, BOD, and COD data, the treated wastewaters were analyzed with
respect to their phenolic content and the presence of residual organics
using chromatographic techniques. Aquatic bioassays and mammalian
cytotoxicity tests were performed on the raw and treated wastewaters to
evaluate their potential environmental impact.
Experimental results from Phase I of the project, involving
coagulation, adsorption, and preliminary biological treatability studies,
are presented in this report. Model studies, using a simulated coal
conversion wastewater at 25% of full strength, suggest that coal conversion
wastewaters are biologically treatable via aerobic treatment processes;
ill
-------
phenol is completely removed by biological systems with sludge ages greater
than 5 days. Preliminary aquatic bioassay experiments and cytotoxicity
bioassays show that the toxicity of the wastewater is substantially reduced
following biological treatment and that the reduction in toxicity increases
with increasing sludge age. The adsorbability of the residual organics
following biological treatment on activated carbon was found to decrease
with increasing sludge age. Acidification and coagulation studies
indicated that tar can be effectively precipitated and removed from coal
conversion wastewaters by acidification to pH 4.5 to 5.0; coagulation using
alum appears to be an ineffective means of removing tar from these
wastewaters. Work on these tasks is continuing, and future reports
representing successive phases of the project will update these results.
This report was submitted in partial fulfillment of the requirements of
Grant No. R804917 by the Department of Environmental Sciences and
Engineering of the University of North Carolina at Chapel Hill under the
sponsorship of the U. S. Environmental Protection Agency. This report
covers the period June 1, 1978 to September 30, 1979.
iv
-------
CONTENTS
Page
Abstract iii
Figures vii
Tables xii
Acknowledgements xiv
1. Introduction 1
2. Conclusions 4
3. Pretreatment of Coal Conversion Wastewaters for the
Removal of Tars and Oils 6
Experimental Methods 10
Results and Discussion 17
Conclusions 36
4. Biological Treatment of Synthetic Coal Conversion
Wastewaters: Part 1 37
Formulation of Synthetic Coal Conversion Wastewater, 38
Description of Pilot Units 39
Operation of Pilot Units 43
Preliminary Results. 45
Summary of Preliminary Results 53
5. Biological Treatment of Synthetic Coal Conversion
Wastewaters: Part 2 56
Operating Procedures 56
Results 60
6. Kinetic Analysis of Biological Treatability Data 82
-------
7. Specific Organic Analysis and Environmental Assessment
of Treated Wastewaters: Preliminary Results 98
HPLC Analysis 98
Aquatic Bioassay 110
Health Effects Bioassay 122
8. Volatility and Air-Stripping of Organics During
Biological Treatment 127
Procedure 127
Results and Discussion 128
Conclusions 129
9. Biodegradability of Coal Conversion Wastewater
Constituents 133
Procedure 134
Calculations and Results 135
Discussion 143
10. Mixed Liquor Respiration Studies 145
11. Adsorption of Alkyl Phenols and Residual TOC Following
Biological Treatment 160
Procedures 161
Results 164
References 175
vi
-------
FIGURES
Number Page
1 Tar, Oil, and Grease Concentrations from Synthane Gasification
of Illinois No. 6 Coal After Various Stages of Treatment.... 9
2 Effect of Acidification on the Formation of Settleable Solids
in Coal Gasification Wastewater 19
3 Effect of Acidification on the Removal of Acetone-Soluble
Tar from Coal Gasification Wastewater 20
4 Effect of Acidification on TOC Removal from Coal
Gasification Wastewater 23
5 Effect of Acidification on COD Removal from Coal Gasification
Wastewater 24
6 Titration of Coal Gasification Wastewater with Sulfuric
Acid 25
7 Removal of Acetone-Soluble Tar by Coagulation with Alum
at pH 5.7 27
8 Removal of TOC by Coagulation with Alum 28
9 TOC Removal by Acidification in the Presence and Absence
of Alum 29
10 Production of Settleable Solids Resulting from Alum
Coagulation at pH 5.7 31
11 Effect of Polymer (DEAE-Dextran) Addition on the Removal
of Acetone-Soluble Tar at pH 6.0 32
12 Effect of Polymer (DEAE-Dextran) Addition on the Removal
of TOC at pH 6.0 33
13 Diagram of Experimental Biological Reactors 41
14 Performance Characteristics of the 5-Day Reactor 46
15 Performance Characteristics of the 10-Day Reactor 47
vli
-------
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
Performance Characteristics of the First 20-Day Reactor
Performance Characteristics of the Second 20-Day Reactor....
Effect of Residence Time on Reactor Performance and
Stability
Effluent TOC and Mixed Liquor Volatile Suspended Solids
Concentration in 5-Day Reactor
Effluent TOC and Mixed Liquor Volatile Suspended Solids
Effluent TOC and Mixed Liquor Volatile Suspended Solids
Concentration in 10-Day Reactor
Effluent TOC and Mixed Liquor Volatile Suspended Solids
Concentration in First 20-Day Reactor
Effluent TOC and Mixed Liquor Volatile Suspended Solids
Concentration in Second 20-Day Reactor
Effluent TOC and Mixed Liquor Volatile Suspended Solids
Concentration in 40-Day Reactor
Correlation Between COD and TOC in Biologically-Treated
Correlation Between BOD and TOC in Biologically-Treated
Correlation Between BOD and COD in Biologically-Treated
Effect of Residence Time on Phenols Removal
Relationship Between Solids Residence Time and TOC Loading..
Relationship Between Solids Residence Time and BOD Loading..
Relationship Between Solids Residence Time and COD Loading..
Kinetics of Substrate Utilization: BOD Basis
Kinetics of Substrate Utilization: TOC Basis
48
49
54
62
63
64
65
66
67
71
72
73
77
78
79
80
86
87
88
89
90
viii
-------
J/
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
HPLC Chromatographic Profiles of Raw Synthetic Wastewater. . .
HPT C rhmma fr>c»r;inhi c Prnfilpc nf S— Hflv Reactor Kf fluent .....
HPLC Chromatographic Profiles of 7.5-Day Reactor Effluent...
HPLC Chromatographic Profiles of 10-Day Reactor Effluent....
HPLC Chromatographic Profiles of 20-Day Reactor Effluent....
HPLC Chromatographic Profiles of 40-Day Reactor Effluent....
Identification and Quantitation of HPLC Chromatographic
Growth of Selenastrum capricornutum Exposed to Various
Growth of Selenastrum capricornutum Exposed to Various
Growth of Selenastrum capricornutum Exposed to Various
Growth of Selenastrum capricornutum Exposed to Various
Toxicity of Raw and Biologically-Treated Synthetic Wastewater
Toxicity of Raw and Biologically-Treated Synthetic Wastewater
Results of 20-Hour Clonal Toxicity Assay Using V-79 Chinese
HPLC Chromatographic Profile of Synthetic Wastewater for
Rate of Oxygen Utilization Resulting From Microbial
92
93
94
100
102
103
104
105
106
108
116
117
118
119
120
121
124
131
137
ix
-------
57 Rate of Oxygen Utilization Resulting from Microhial
Degradation of p-Cresol 137
58 Rate of Oxygen Utilization Resulting from Microbial
Degradation of o-Cresol 138
59 Rate of Oxygen Utilization Resulting from Microbial
Degradation of m-Cresol 138
60 Rate of Oxygen Utilization Resulting from Microbial
Degradation of 2 ,5-Dimethy 1 phenol 139
61 Rate of Oxygen Utilization Resulting from Microbial
Degradation of 2 ,3-Dimethylphenol 139
62 Rate of Oxygen Utilization Resulting from Microbial
Degradation of 2 ,6-Dimethy 1 phenol 140
63 Rate of Oxygen Utilization Resulting from Microbial
Degradation of 3 ,5-Dime thy Iphenol 140
64 Rate of Oxygen Utilization Resulting from Microbial
Degradation of 3 ,4-DietnthyIphenol 141
65 Experimental Set-Up for Mixed Liquor Respiration Studies.... 146
66 Common Types of Oxygen Utilization Curves 148
67 Respiration Curves Using Original Oxygen Utilization
Procedure 149
68 Respiration Rate of Aerated Sludge with 200 mg/1 Phenol 151
69 Effect of Time on Respiration Rate with 2000 mg/1 Phenol.... 154
70 Respiration Rate of "Fresh" Sludge from Reactor 155
71 Respiration Curves for Phenol Using Modified Oxygen
Utilization Procedure on Selected Substrates 157
72 Results Using Modified Oxygen Utilization Procedure on
Selected Substrates 158
73 Adsorption of Methylphenols by Activated Carbon 165
74 Adsorption of Dimethylphenols by Activated Carbon 166
75 Adsorption of Ethylphenols by Activated Carbon 167
76 Comparative Adsorption of Alkyl-Substituted Phenols by
Activated Carbon 168
-------
77 Adsorption of Phenol by Activated Carbon 170
78 Adsorption of Isopropyl phenol by Activated Carbon 171
79 Adsorption of Raw and Biologically-Treated Synthetic
Wastewater by Activated Carbon 173
xi
-------
TABLES
Number Page
1 Typical Oil, Grease, and Tar Concentrations in Synthane
By-Product Water 7
2 Typical Jar Test Results with an Illinois #6 Coal Wastewater
Produced During Synthane Gasification 8
3 Physical and Chemical Characteristics of Coal Gasification
Wastewater Sample 18
4 Results of Analysis of Residual Supernatant Layer After
Acidification of Wastewater Sample to pH 4.8 26
5 Phenol Removal by Acidification and Coagulation 35
6 Composition of Synthetic Coal Conversion Wastewater 40
7 Summary of Reactor Performance During Periods of Intensive
Analysis 51
8 Average Quality of Effluent from Biological Treatment Units. 55
9 Concentration of Inorganic Constituents in Quarter-Strength
Wastewater 57
10 Record of Equipment Malfunctions 61
11 Summary of Reactor Performance 59
12 Periods of Steady State Performance 75
13 Summary of Average Steady State Reactor Performance 76
14 Calculated Process Loading Factors for Biological Reactors.. 84
15 Summary of Kinetic Coefficients 95
16 Kinetic Coefficients from Biological Treatment of Synthane
and Synthoil Wastewater 97
17 Identification of HPLC Peaks 101
xii
-------
18 Concentrations of Major Phenolic Compounds in Reactor
Effluents 109
19 Effects of Raw and Biologically-Treated Wastewaters on
Fathead Minnows and Daphnia Pulex 113
20 Summary of Mammalian Cytotoxicity Data 125
21 Change in Total Organic Carbon Concentration Resulting
from Aeration 129
22 Results of HPLC Analysis of Aerated Wastewater 130
23 Biodegradation of Selected Coal Gasification Wastewater
Components 142
24 Effect of Aeration on "Activity" of Sludge 152
25 Characteristic Wavelengths for Maximum UV Absorbance of
Aqueous Phenolic Solutions 162
26 Langmuirian Coefficients for the Adsorption of Alkyl Phenols
on PX-21 Powdered Activated Carbon 172
xiil
-------
ACKNOWLEDGEMENTS
We would like to acknowledge the help provided by Michael Hughes,
Gerald Speitel, Dave Reckhow and Roger Rader in designing, constructing and
operating the biological reactors; Randall Williams for carrying out the
coagulation and acidification experiments; Chen Yen for performing the
adsorption studies; Teena Cochran, Liz Anderson, Marynoel Monson and Ann
Chan for their assistance in the analytical phases of the project; Anthony
Maciorowski, Jane Hughes, Cecily Beall, and Steve Shoaf for their aquatic
bioassay activities; Mark Sobsey, Randy Jones and Leslie McGeorge for
setting up and performing the mammalian cytotoxicity assays; and Jolene
Chinchilli and Dave Ruehle for carrying out the biodegradation studies.
The "assistance provided by Drs. Dean Smith and Thomas Petrie, our
Project Officers, and their colleagues at the Industrial Environmental
Research Laboratory of the U. S. Environmental Protection Agency at
Research Triangle Park, N.C., in guiding and facilitating the performance
of this research is also appreciated.
xiv
-------
SECTION I
INTRODUCTION
In a previous review (1) of the characteristics of wastewaters
generated as a result of the production of synthetic fuels from coal, it
was shown that these wastewaters contain substantial amounts of organic
substances, many of which can have an adverse impact on aquatic life and on
human health. CODs (chemical oxygen demands) on the order of 10,000-40,000
mg/1 and TOC (total organic carbon) concentrations of 5,000-10,000 mg/1
have been reported (2, 4) for these wastewaters. Specific organic analysis
of these wastes shows that 60-80% of the total organic carbon is phenolic
in nature, consisting of monohydric, dihydric, and polyphenols. The
remainder of the organic material identified consists of monocyclic and
polycyclic nitrogen-containing aromatics, oxygen- and sulfur-containing
heterocyclics, polynuclear aromatic hydrocarbons, and simple aliphatic
acids. The composition of the wastewaters from various coal gasification
and coal liquefaction processes appears to be relatively uniform,
especially with respect to the phenolic constituents, regardless of the
specific process technology or feed coal employed (1).
Aerobic biological processes appear to be the focal point of any
overall scheme for treating these wastewaters since a significant number of
the major constituents of the wastes are biodegradable. Accordingly,
suitable design and operating criteria for biological treatment of these
-------
wastewaters need to be developed. In order to establish such guidelines
and to evaluate the efficacy of biological treatment, treatability studies
need to be conducted.
Several research groups (3,4,5, and 6) have conducted and are currently
performing biological treatability studies on wastewaters produced from a
variety of pilot coal gasification and liquefaction facilities such as the
Hygas, Synthane, Lurgi, and H-Coal processes. Accordingly, these
treatability studies are closely bound to the conversion process being
investigated. However, since coal conversion processes are still in the
developmental stage, there is some question as to whether or not a
suitable, consistent and representative wastewater could be obtained for a
comprehensive analysis and assessment of overall coal conversion wastewater
treatability. Hence, the studies described in this report have been
conducted using a synthetic wastewater which has been formulated to be
representative in its organic composition, of actual coal conversion
wastewaters. The wastewater contains twenty-eight organic compounds,
inorganic nutrients, and pH-buffers.
The synthetic coal conversion wastewater is being used to feed several
bench-scale activated sludge reactors. In addition to generating
acclimatized organisms for separate biodegradability studies of model
compounds identified in actual coal conversion wastewaters, the pilot
reactors are being used to treat the synthetic wastewater under various
types of operating conditions. Effluents from the reactors are being
analyzed by gas chromatography, gas chromatography/mass spectrometry, and
high performance liquid chromatography to assess the degree of removal of
the various constituents in the raw feed, and to identify reaction products
-------
following biological treatment. Additionally, acute toxicity studies using
fish, Daphnia, and algae are being conducted to evaluate the biological
impact of the treated wastewaters on aquatic life. Acute mammalian
cytotoxicity and Ames mutagenicity analyses are also being performed on the
reactor effluents to assess their potential impact on human health. This
report presents some of the initial results of the biological treatability
evaluation, along with some additional studies directed at wastewater
pretreatment for acidification and coagulation, and adsorption of phenolic
substances and residual TOC from biologically-treated wastewater using
activated carbon.
-------
SECTION 2
CONCLUSIONS
Based upon preliminary model studies using a synthetic coal conversion
wastewater at 25% of full strength and aerobic biological processes without
solids recycle, coal conversion wastewaters appear to be biologically
treatable. TOC, COD, and BOD removal increase with increasing sludge age
(solids residence time) and reductions of 85 to 97%, 86 to 96%, and 99.8%,
respectively, have been obtained with sludge ages of 20 days. Phenol is
essentially completely removed with a sludge age of 5 days, while the
cresols and xylenols require 7.5 to 10 days and 20 days, respectively, for
removal to levels below 1 mg/1. Although some difficulties were
encountered in achieving stable reactor operation and steady state
performance, the TOC, COD, and BOD data appear to be in conformance with
commonly accepted kinetic models; the kinetic coefficients for microbial
growth and substrate utilization using the synthetic wastewater are of the
same order and essentially in agreement with the results of other
investigators performing biological treatability studies on samples of real
wastewaters generated from developing coal conversion technologies.
Preliminary aquatic bioassay experiments with fish, Daphnia and algae,
and a health effects acute cytotoxicity bioassay employing mammalian cells
exposed to raw and biologically treated synthetic coal conversion
wastewaters show that toxicity of the waste is substantially reduced
-------
following biological treatment, and that the reduction in toxicity
increases with increasing sludge age. These bioassays are continuing in an
attempt to develop more definitive data with respect to the biological
impact of the treated wastewaters.
Activated carbon adsorption studies show that alkyl-substituted phenols
are more strongly adsorbed than phenol and that the extent of adsorption
increases as the number of alkyl substituents and the length of the alkyl
chain increase. Position of the substituent alkyl group has no effect on
the extent of adsorption. With respect to the adsorbability of the
residual organic carbon following biological treatment, the extent of
adsorption appears to decrease with increasing sludge age of the biological
system. Interpretation of the results is based upon the polar nature
(aqueous solubility) of the residual organic compounds comprising the
effluent TOC from the reactors.
Acidification and coagulation studies show that tar can be effectively
precipitated and removed from real coal conversion wastewaters by
acidification of the waste to pH 4.5 to 5. Coagulation using alum appeared
to be an ineffective means of removing tar from these wastewaters.
These treatability studies are continuing, as are chemical and
biological assessment studies of the effluents following various degrees of
treatment. Particular attention will be directed at priority pollutants
present in these wastewaters. The results of these additional studies will
be available in the next report in this series. Ultimately, the
conclusions developed using the synthetic coal conversion wastewater will
be tested on real conversion wastewaters.
-------
SECTION 3
PRETREATMENT OF COAL CONVERSION WASTEWATERS FOR THE REMOVAL
OF TARS AND OILS
Wastewaters produced from coal gasification pilot plants have been
found to contain suspended material such as tar and oil as well as
dissolved organic compounds (2). These tars and oils are composed of an
organic matrix of compounds including some carcinogenic polycyclic aromatic
hydrocarbons and their derivatives. The presence of tar and oil in coal
gasification wastewaters can contribute to the fouling of equipment and
pipes and, in addition, has been shown to interfere with aerobic biological
treatment processes for coke plant wastewaters (7).
Johnson et al. (4) have published the results of experiments designed
to describe the removal of tar and oil from Synthane coal gasification
wastewaters using different coagulants. Table 1 gives typical
concentrations of oil and grease, and acetone-soluble tar found in Synthane
by-product water. These analyses apply to the decanted supernatant layer
from the water-cooled condensers. The raw condensate, containing- about
22,000 mg/1 of tar, oil, and grease, was allowed to settle 3 to 6 hours
before the supernatant was drawn off. The authors report that this layer
is typical of the wastewater that would be sent to a treatment plant fuom a
commercial Synthane facility.
-------
TABLE I. TYPICAL OIL, GREASE, AND TAR CONCENTRATIONS
IN SYNTHANE BY-PRODUCT WATER* (All values in mg/1)
Type of
Coal Gasified
Montana Rosebud
Montana Rosebud
Illinois #6
Illinois #6
Oil and Grease*
910
840
1,020
460
Acetone-Soluble
Tar*
1,180
1,150
1,970
550
*After Johnson et al. (4)
"""Freon-soluble materials measured according to Standard Methods (8).
*Does not include components measured as oil and grease.
Table 2 gives the results of jar tests performed with Synthane
supernatant. The coal used in this particular run was Illinois #6 coal.
Conditions (pH, alum dose, and polymer dose) of the experiments are given,
yet no quantitative data are presented for evaluation of the different
conditions applied. Figure 1 shows a tar reduction from 1,000 mg/1 in the
supernatant layer to 600 mg/1 after treatment consisting of pH depression
and alum addition followed by pH adjustment. The exact conditions of
coagulant dose and pH are not specified. The removal of tar and oil from
the supernatant was 47 percent for a total of 97 percent reduction from the
raw condensate.
It appears that acidification and alum addition are effective means of
removing tar and oil, but the appropriate conditions for achieving best
removal are not clear. Quantitative data describing the removal of tar and
-------
TA3LE 2. TYPICAL JAR TEST RESULTS WITH AN ILLINOIS #6 COAL WASTEWATER PRODUCED DURING SYNTHAXE GASIFICATION*
do
Jar #
1 +
2 +
3+
4*
5 +
6+
7*
8*
9*
10*
11*
12*
Adjusted
PH
1.9
2.0
3.5
2.4
2.5
1.5
6.9
6.8
6.7
7.1
6.9
6.9
pH After Alum
Neutralization (mg/1)
7.0 25
7.0 50
7.0 25
7.0 50
7.0 100
7.0 150
20
40
60
80
100
120
Observations
Some surface oil and tar
Some surface oil and tar
Surface oil with some noticeable floe formation
Considerable oil on surface, noticeable floe formation
Much surface oil, excellent floe formation
Much surface oil, opaque liquid, excellent floe formation
Much surface oil, no floe
Much surface oil, no floe
Much surface oil, floe visible
Much surface oil, floe visible
Much surface oil, significant floe
Much surface oil, less floe than in jar #11
*After Johnson et al. (4)
+Alum alone
*Alutn and polymer; polymer dosage = 2 mg/1
-------
Figure 1. Tar, oil, and grease concentrations from Synthane gasification of
Illinois No. 6 coal after various stages of treatment.
(After Johnson et al., 4)
25
20
CO
o
T"
x
_J
UJ
V)
<
UJ
oc
q
cc
o
z
o
oc
\-
ai
y
O
o
15
10
21.8
70.7
Illinois No. 6 condensate
1.1
3.7
0.6
1.9
\ X
80
70
60
50
Q
IU
C/3
o
o
u.
o
I
cc
UJ
o.
UJ
CO
UJ
OC
o
30 =f
O
»
cc
5
40
20
10
TOTAL
CONOENSATE
SUPERNATANT
TREATED
SUPERNATANT
V)
O
Z
i
-------
oil from coal gasification wastwaters are necessary for evaluating
treatment methods. This section describes the results of a study directed
at the removal of tar and oil from a representative wastewater produced
during coal gasification operations. The control methods investigated
include coagulation using alum and synthetic organic polymers and
acidification to precipitate dissolved tars and oils.
EXPERIMENTAL METHODS
Sample Handling and Storage
A five-gallon sample of a representative coal gasification wastewater
was obtained from the Industrial Environmental Research Laboratory (IERL)
of the U. S. Environmental Protection Agency (EPA) at Research Triangle
Park, North Carolina. (For the pretreatment studies directed at tar and
oil and grease removal, it was decided to work with real coal conversion
wastewaters.) The sample was received in September, 1977 and was
immediately stored in a 10 C refrigerator. In an effort to maintain
sample integrity, no attempt was made to preserve chemically the
wastewater. The only major changes observed in this wastewater over the
study period were a gradual darkening of the amber color and some settling
of suspended materials. The wastewater was always well mixed prior to
withdrawing samples in order to ensure that representative samples would be
collected. The last few samples taken from the container were somewhat
concentrated with respect to solids, however.
10
-------
Experimental Procedures
Initially, the sample was characterized with respect to pH, TOG, COD,
total phenols, tar, oil and grease, and suspended solids. Standard jar
test procedures were then employed to investigate the effects of
coagulation and precipitation on the sample. The procedure for each set of
jar test experiments is outlined below.
Procedure for Coagulation Experiments—
A series of 100 ml samples of well-mixed wastewater were removed from
the refrigerated storage container with a volumetric pipet. Each sample
was placed in a 150-tnl beaker and covered with Parafilm. The aliquots were
allowed to warm to room temperature, usually 18 to 22 C, before
experimentation began.
The pH of each sample was adjusted to the desired level by the dropwise
addition of H SO . The pH was measured using a Fisher combination
electrode and pH meter.
Various dosages of coagulant were added. During the course of
coagulant addition, the pH was maintained constant by adding NaOH, if
necessary.
Samples were flash-mixed for 1 min on a magnetic stirrer to disperse
the coagulant. Samples were removed from the stirrer and immediately
placed on a Phipps and Bird jar test machine for 20 min, at a mix speed of
20 RPM. No attempt was made to prevent the loss of volatile materials such
as NH , H S, CO , or volatile, low molecular weight organic species
during the stages of mixing.
After mixing, the samples were removed from the jar test machine and
allowed to settle under quiescent conditions. After 1 hour, aliquots of
11
-------
the supernatant were removed from each beaker by carefully lowering the tip
of the pipet beneath the surface of the sample. Fifty-mi were removed for
tar, and oil and grease analyses, and a 5 ml sample was collected for TOG,
COD, and total phenols analyses. The solid residue remaining in the beaker
after settling was filtered and analyzed for settleable solids.
Coagulants Used in Experiments—
Alum (A1,,(SO, )„ ' 18H 0)—Stock solutions of 5, 20, and 50
243 2
g/1 were prepared as needed from reagent grade alum. Use of a particular
stock solution depended on the dose of alum required in the jar test
experiment.
Organic Polyelectrolytes—DEAE-Dextran, a high molecular weight
(2,000,000) cationic polyelectrolyte made from Dextran 2000, and Dow C-31
Purifloc, a cationic polyelectrolyte with a molecular weight of 30,000 were
employed in these experiments. Stock solutions of 1 percent (10 g/1)
Purifloc were made by pipetting 5 ml of the 20 percent concentrated stock
solution and diluting to 100 ml. Stock solutions of 1 g/1 of the dry
DEAE-Dextran powder were made by slowly adding the weighed amount of
polymer to distilled water and mixing in a high-speed Waring blender. Once
dissolved, the solution was removed from the blender and diluted to 1 1. A
stock solution of 0.1 g/1 was made by dilution of the 1-g/l stock.
Procedure for Acidification Experiments—
Experiments designed to determine the effectiveness of acid addition on
tar removal were conducted using procedures similar to those of the
coagulation experiments. However, no coagulants were added after the pH
was adjusted to the desired level. Thereafter, the samples were mixed and
12
-------
allowed to settle for 1 hour, and aliquots of the resultant supernatant
layer were withdrawn for analysis as in the coagulation experiments.
Analytical Methods—
Total Organic Carbon (TOO)—TOC was used as an indicator of the degree
of removal of organic material from the wastewater samples by determining
values before and after treatment. TOC was measured using a Beckman 915A
TOC Analyzer coupled with a Beckman 215B Infrared Detector. Organic and
inorganic carbon standards were prepared with carbon-free water according
to methods outlined by Beckman Instruments; 1,000 mg/1 stock solutions of
the organic standard (reagent grade potassium acid phthalate) and the
inorganic standard (a solution of NaHCO and Na CO ) were prepared.
All other standards were subsequently made from dilutions of the 1,000 mg/1
stocks.
Duplicate 10- or 20-Vl injections of the organic and inorganic
standards were made into the total carbon and inorganic carbon channels,
respectively. Calibration curves were obtained by plotting percent
absorbance vs. mg/1 of total carbon or inorganic carbon.
Since there was concern that the wastewater may contain significant
concentrations of volatile organics, the samples were not acidified and
purged with nitrogen to remove inorganic carbon prior to analysis as
recommended in Standard Methods (8). Instead, separate determinations of
inorganic carbon were made as described in the following paragraph.
Duplicate 10- or 20-yl injections of each sample were made with an
automatic syringe to ensure reproducibility. In cases for which the
resulting peak heights were not consistent for the same sample, injections
13
-------
were made until reproducible peak heights were obtained. Because the TOG
of the wastewater was very high, dilutions of the treated and untreated
samples were made in order to obtain values that would be "on scale"
(between 0 and 1,000 mg/1). Typically, 5 ml of sample was diluted to 100
ml using distilled, deionized water. Absorbance readings on successive
injections were averaged, and total carbon and inorganic carbon values were
determined from the calibration curves. Inorganic carbon values were then
subtracted from the total carbon values, yielding TOC values.
Acetone-Soluble Tar; Oil and Grease—Tar is operationally defined as an
acid-insoluble organic residue. Its meaning is derived strictly from the
procedure employed in its measurement. This analytical procedure was
suggested by Johnson et al. (4) and modified for use during this study.
Oil and grease, operationally defined as acid-insoluble, freon-soluble
material, was determined in accordance with the partition-gravimetric
method outlined in Standard Methods , Section 502A (8).
Oil and grease, and tar determinations were usually made on the same
aliquot. For the raw wastewater, a 50 ml aliquot of a well-mixed sample
was removed from the storage container. For treated samples, including
controls, 50 ml was withdrawn from the supernatant layer after chemical
treatment and settling of solids. These aliquots were placed in 125 ml
separatory funnels. The pH in all cases was then depressed to 1.6 by the
dropwise addition of H SO to precipitate all acid-insoluble
2 4
materials. These acid-insoluble materials were then extracted with three
15-ml. washings of freon, 1,1,2-trichloro-l ,2 ,2-tri f luorethane (reagent
grade from Fisher Chemical Company). The three washings were combined in
another separatory funnel where any water extracted into the freon was
14
-------
allowed to separate from the freon. The freon layer was then poured into a
clean, dry distilling flask of known tare weight. The freon was driven off
under an applied vacuum through a rotary evaporator over a 70 c water
bath. The flask was immediately transferred to a dessicator for 30 min and
then weighed on an analytical balance. The difference in weight between
the tare weight and the weight of the flask plus the freon-extracted
material equaled the weight of oil and grease in the 50-ml aliquot of the
supernatant. This value was subsequently multiplied by 20 to determine the
concentration of oil and grease in mg/1.
The acid-insoluble residue remaining in the supernatant aliquot after
freon extraction was drained onto a #40 Whatman filter disk to which a
vacuum was applied. The filter disk was then washed with reagent grade
acetone until it was apparent from the color of the disk that all
acetone-soluble material had been extracted. The amount of acetone-soluble
residue was a very small fraction of the total residue.
These acetone-soluble washings were collected in a.250 ml beaker and
subsequently refiltered through another #40 Whatman filter disk to remove
any acetone-insoluble material that might have been rinsed off the original
filter disk. Following filtration, the acetone solution was poured into a
clean dry distilling flask of known tare weight and rotary evaporated at
70 C, until all the acetone was removed. The flasks were placed in a
103 C oven for 30 minutes to remove any water contamination. The flask
was removed from the oven, cooled to room temperature in a dessicator, and
weighed on an analytical balance. The difference between the weight of the
flask and the acetone-extracted residue and the tare weight of the flask
equaled the weight of acetone-soluble tar in the 50 ml aliquot of
15
-------
supernatant. As with oil and grease, this value was multiplied by 20 to
obtain the concentration of tar in mg/1.
On some supernatant samples, only acetone-soluble tar was determined.
The pH of these samples was depressed to 1.6 as before. After the solids
precipitated, they were filtered through a #40 Whatman filter disk, and the
freon extraction step was omitted. Thereafter, the procedure was identical
to that outlined above for tar determination.
Chemical Oxygen Demand (COD)—COD was determined for selected samples
according to the procedures outlined in Section 508 of Standard Methods
(8). Samples were diluted in the same way as with TOG.
Total Phenols—The concentration of total phenols was determined by the
4-aminoantipyrine coloritneteric procedure outlined in Standard Methods,
Section 510 A and C (8, 9). Phenols react with 4-aminoantipyrine at pH
10.0 +0.2 in the presence of potassium ferricyanide to form a colored
antipyrine dye. Distillation of the sample is required as a first step to
separate the phenolic compounds from nonvolatile impurities that interfere
with color formation. The phenolic compounds measurable by this technique
include phenol (C H,OH), ortho- and meta-substituted phenols, and under
6 5
certain pH conditions, para-substituted phenols in which the substituent is
a carboxyl, methoxyl, halogen, or sulfonic acid group. All other
para-substituted phenols cannot be detected by this procedure.
Samples had to be diluted so that no more than 0.5 mg phenol was
present in 100 ml. The standard curve was not linear for values above this
concentration. After distillation of the sample and addition of the
reagents, absorbance of the colored antipyrine dye was measured at 510 nm
with a Varian spectrophotometer. These readings were compared with a
16
-------
standard curve of absorbance vs. mg phenol/100 ml. Results are reported as
mg/1 C H OH.
6 j
Because the percentage of various phenolic compounds present in the
given coal gasification wastewater was not known, phenol itself was
selected as the standard. Any color produced by the reaction of other
phenolic compounds with the reagents is reported as phenol (8).
Substitution generally reduces color response, so the values reported for
total phenols in the wastewater are less than they would be if an
equivalent amount of phenol was present in the wastewater.
Settleable Solids—Settleable solids are operationally defined as the
amount of nonfilterable residue created after wastewater samples are
chemically treated and allowed to settle for 1 hour. After aliquots of the
supernatant layer were withdrawn for TOG, COD, tar, and oil and grease
analyses, the remaining layer of settled solids was filtered through a
Whatman glass fiber filter and analyzed for nonfilterable residue in accord
with Standard Methods, Section 208D (8).
RESULTS AND DISCUSSION
General Wastewater Characteristics
Physical and chemical characteristics of the coal gasification
wastewater sample are given in Table 3. The values of the parameters can
be compared with the corresponding values given for Synthane gasification
byproduct water in Tables 1 and 2. The wastewaters produced during the
Synthane gasification process and the IERL-EPA wastewater sample are
17
-------
reasonably similar with respect to the amounts of acetone-soluble tar and
oil and grease.
TABLE 3. PHYSICAL AND CHEMICAL CHARACTERISTICS OF COAL
GASIFICATION WASTEWATER SAMPLE.
(All values except pH in mg/1.)
pH 8.3
TOG 5,740
Oil and grease 920
COD 17,400
Acetone-soluble tar 1,950
Suspended Solids 750
Total Phenols 2,000
Treatment by Acidification
Figure 2 shows a plot of settleable solids created as a function of pH
adjustment. Upon acidification, solids precipitate from solution and
settle out under quiescent conditions. Solids production increases as acid
is added, until a pH of about 5.7 is attained. Thereafter, no additional
settleable solids are produced regardless of the amount of additional acid
introduced to the system. Approximately 1,600 mg/1 of settleable solids is
produced by depressing the pH to any value below 5.7.
Figure 3 is a plot of the residual acetone-soluble tar as a function of
pH. A gradual decrease in residual tar occurs as the pH is depressed from
18
-------
2000
1,750
1.500
-^ 1,?50
j
o
Z
Ǥ ipoo
8
CO
<
UJ
750 --
m^m
% 500
250
8.0
7.0
6.0
5.0
4.0
3.0
2.0
1.0
PH
Figure 2. Effect of acidification on the formation of settleable solids in coal gasification wastewater,
-------
5
g
ipoo -
o
a 500
K
8.0
7.0
6.0
3.0
2.0
1.0
5.0 4.0
pH
Figure 3. Effect of acidification on the removal of acetone-soluble tar from coal gasification vaatevaten.
-------
8.3 to about 6.0. From pH 6.0 to pH 5.3, a sharp decrease in the
concentration of residual tar is observed. Below pH 5.3, no significant
improvement in tar removal is noted.
Some degree of scatter occurs among the data points, especially along
the steep section of the curve around pH 6, but not enough to negate the
observation of a trend. The scatter probably results because of difficulty
in weighing heavy objects (distilling flasks) to find milligram differences
in weight.
The analytical procedure for the determination of oil and grease was
not effective in demonstrating removal trends of these impurities as a
function of pH due to contamination of the solvent extract by water. When
the freon extraction procedure was performed, it appeared that some of the
amber-colored aqueous layer was miscible with the normally clear freon
layer. The addition of an extra step to allow for separation of the water
and freon layers did not eliminate the problem. Once the freon was
evaporated, small amber-colored water droplets remained in the distilling
flask. The flask was weighed, and the values obtained were reported as
acid-insoluble, freon-soluble oil and grease. When freon was added to the
distilling flask in order to dissolve the oil and grease for cleaning
purposes, it was noted that all of the so-called oil and grease did not
dissolve. It was evident that complete separation of true oil and grease
from treated and untreated wastewater was not possible because of
contamination of the oil and grease extract by water. The oil and grease
analysis adopted for these experiments (8) was designed for the petroleum
and food processing industries. Its applicability to the analysis of this
coal gasification wastewater is questionable.
21
-------
Figures 4 and 5 show TOC and COD removal as a function of pH. The
greatest increase in removal of each is observed between pH 6.5 and 4.8.
The actual shapes of the curves correspond closely with each other and with
those describing settleable solids production and acetone-soluble tar
removal. The greatest increase in solids production and the greatest
decrease in residual tar, TOC, and COD occur between pH 6.5 and pH 4.8.
A summary of the results of acidification of the wastewater to pH 4.8
is shown in Table 4. While approximately 1,835 mg/1 of tar is removed from
the wastewater by acidifying to pH 4.8, 1,550 mg/1 of settleable solids is
concurrently produced. In addition, TOC and COD reductions of 1,220 mg/1
and 2,730 mg/1, respectively, are associated with the observed tar
removal. In view of the aromatic nature of the molecules comprising the
tar, these reductions in TOC and COD are consistent with each other and
with the observed removal of tar.
Figure 6 depicts an acid-base titration curve describing the
neutralization of the coal gasification wastewater by strong acid.
Titratable species such as bicarbonate, ammonia, and carbonate are at very
high concentrations in coal gasification wastewaters (2). Predictions
concerning the quantity of strong acid required to precipitate tars can be
made from the data described by the curve. It appears that approximately
40 meq/1 of acid is required to depress the pH of the wastewater to about
5.0. At this pH, 95 percent tar removal can be achieved (refer to Figure
3).
22
-------
7000
6000
5000
_j
o
~ 4000
cc
<
o
3000
cc
! 2000
tooo
8.0
7.0
6.0
5.0
4.0
PH
3.0
2.0
1.0
Figure 4. Effect of acidification on TOC removal from coal gasification wastewater.
-------
20000
17500
_ 15POO
O
Q 1^500
UJ
O
2 10000
UJ '
NJ X
*" O
500
iu
O 5.000
2500
8.0
7.0
6.0
5.0
4.0
pH
3.0
2.0
1.0
Figure 5. Effect of acidification on COD removal from coal gasification wastewater.
-------
8.0
7.0 -
6.0 -
5.0 -
4.0 -
PH
3.0
2.0
1.0
20
Figure 6.
40 60 80 100
ACID ADDED (MEQ/L )
Titration of coal gasification wastewater with sulfuric acid.
120
-------
TABLE 4. RESULTS OF ANALYSIS OF RESIDUAL SUPERNATANT
AFTER ACIDIFICATION OF WASTEWATER SAMPLE TO pH 4.8
Untreated
Treated
(acidified to pH 4.8)
Difference
% Reduction
Tar, mg/1
1,950*
115
1,835
94%
COD, mg/1
17,400
14,670
2,730
16%
TOC, mg/1
5,620
4,400
1,220
22%
*Average of 6 determinations
Effectiveness of Coagulants
Alum—Figure 7 shows a plot of residual supernatant tar as a function
of alum dose, at pH 5.7. As alum dosage is increased up to 1,000 mg/1,
there is no concommitant reduction in acetone-soluble tar. In fact, there
appears to be an increase in the tar concentration. The reason for this is
unclear but could be due to restabilization of tar by the added Al(lII).
Initially, 1,950 mg/1 of tar was present in the wastewater sample at pH
8.3. Tar is reduced to approximately 500 mg/1 simply by acidification to
5.7.
The data presented in Figures 8 and 9 show that even at very high alum
doses and at various pH values, little if any TOC removal is observed. The
curve from Figure 5, which demonstrates TOC removal as a function of the
26
-------
ui
_i
ffi
2,000
V50
1,500
1250
ui 1,000
750
Q
CO
UJ
ec 500
250
250
750
Figure 7.
500
ALUM DOSE (MG/L)
Removal of acetone-soluble tar by coagulation with alum at pH 5.7
1000
-------
7000
5000
5pOO
O
g 4J300
^
o
u
T ^
ho 35
03 <
3.000
I- 2000
1,000
pH 6.0
250
pH 5.3-
500
1000
750
ALUM DOSE (MG/L )
Figure 8. Removal of TOC by coagulation with alum.
1250
1500
-------
7,000
8.0
PH
Figure 9. TOG removal by acidification in the presence and absence of alum. (Alum Dose = 1,000 mg/1.)
-------
degree of acidification alone, without alum addition, is shown in Figure 9
for purposes of comparison.
From the data presented in Figure 10, it appears that the added
aluminum is being totally solubilized by the constituents of the wastewater
and that no precipitation of Al(OH) is occurring. Normally, at pH 5.7,
aluminum is insoluble and can be expected to hydrolyze to form abundant
amounts of Al(OH) , which should precipitate from solution. Figure 10
demonstrates that no additional solids were created despite the massive
addition of alum. This observation suggests that aluminum is being
complexed (solubilized) by various dissolved species present in the
wastewater and, as such, cannot be expected to act as a very effective
coagulant as demonstrated in Figures 7, 8, and 9.
The presence of ligands with the potential of forming complexes with
multivalent cations such as Al is well documented for coal gasification
wastewaters (2). Ligands capable of forming complexes with the Al
cation include CN , SCN , and the anions of organic acids, all of which
can compete with hydrolysis reactions. In view of the alum dosage applied
to the samples (up to 1,500 mg/1 as alum), the ability of the ligands in
the wastewater to complex aluminum must be considerable.
Polymers—In contrast to alum, the two cationic polyelectrolytes tested
in this study were shown to be capable of removing tar and TOC from the
wastewater. Figures 11 and 12 demonstrate the effect of increasing dosages
of DEAE-Dextran applied at pH 6.0. Tar is reduced from 1,260 mg/1 to 260
mg/1 at pH 6.0 upon addition of 100 mg/1 of DEAE-Dextran. TOC is
concommitantly reduced by 300 mg/1 at this dosage and by greater amounts at
higher polymer dosages.
30
-------
2000
1,750
X500
-ti-
~ 1250
o
u>
UJ
ui
UJ
\ooo
750
500
250
250
750
It-
Figure 10.
500
ALUM DOSE ( MG/L )
Production of settleable solids resulting from alum coagulation at pH 5.7
1000
-------
2000
KJ
DC
<
CD
D
I
UJ
1500
125°
750
500
250
50
125
150
175
75 100
POLYMER DOSE (MG/L )
Figure 11. Effect of polymer (DEAE-Dextran) addition on the removal of acetone-soluble tar at pH 6.0.
-------
GpOO
5000
4000
flC
u
I **»
o
oc
o
< 2pOO
1,000
100
200
300
400
500
Figure 12.
POLYMER DOSE (MG/L)
Effect of polymer (DEAE-Destran) addition on the removal of TOC at pH 6.0,
-------
Similarly, Dow C-31 Purifloc was also capable of removing TOC from the
wastewater. A TOC reduction of 600 mg/1 was observed when 200 mg/1 of
Purifloc was applied at pH 8.3. No determinations of tar removal were made
during these experiments, however.
DEAE-Dextran has a molecular weight of 2,000,000 (10) which is more
than sufficient for destabilization by interparticle bridging. Purifloc
does not have as high a molecular weight, but does have a higher charge
density and can act to destabilize by adsorption and charge neutralization.
Considerable dosages of DEAE-Dextran and Purifloc had to be applied
before significant reductions in tar and TOC were observed. Some polymers
can destabilize certain colloidal suspensions at very low dosages (10),
making them economically desirable to use. The dosages required to
destabilize tar in these wastewaters appear to exceed dosages that would be
economically feasible.
Further experimentation with Purifloc and DEAE-Dextran at different
dosages and pH's and experimentation with other cationic, anionic, and
nonionic polymers was precluded due to lack of a sufficient quantity of
wastewater.
Phenols
The concentration of total phenols was determined for samples that were
acidified as well as for samples that were acidified and treated with
various doses of alum. The concentration of total phenols was not
decreased by any type of treatment (Table 5). While the solubility of
acetone-soluble tar is dramatically reduced as the pH is depressed, no
change in the solubility of phenolic compounds is observed.
34
-------
TABLE 5. PHENOL REMOVAL BY ACIDIFICATION AND COAGULATION.
(All values in mg/1.)
PH
8.3
8.3
8.3
3.1
2.8
5.0
5.3
5.3
Alum Dose
mg/1)
1,000
1,000
250
1,000
Total Phenols
(mg/1 as phenol)
1,900
2,030
1,920
2,000
1,920
1,920
2,000
1,950
Summary
Results of acidification experiments on the coal gasification
wastewater supplied by the Industrial Environmental Research Laboratory
(IERL) showed that a 94 percent reduction in residual acetone-soluble tar
could be obtained when the pH of the sample was depressed to approximately
4.8 (Table 4). This reduction in tar corresponds to a 16 percent decrease
in COD and a 22 percent decrease in the TOC of the wastewater. These
values appear to be consistent with each other.
Johnson et al. (4) were able to achieve a 47 percent reduction in tar,
oil, and grease in Synthane by-product water after batch treatment
consisting of pH depression and alum addition followed by pH adjustment to
35
-------
7.0 (Figure 1). In addition, Johnson et al. observed "floe formation" at
various pH's and alum dosages up to 1,500 mg/1. However, for the
wastewater used in this study, alum was incapable of acting as an effective
coagulant, most likely because of the solubilization of aluminum by ligands
present in the wastewater.
CONCLUSIONS
1. Acidification of a representative sample of coal gasification
wastewater obtained from the EPA proved to be an effective means to remove
dissolved and suspended acetone-soluble tars. TOC and COD reductions were
observed concotnmitantly with tar reductions. In order to obtain about 95
percent tar removal from the wastewater, approximately 40 meq/1 of strong
acid was required.The addition of this amount of acid to the wastewater
depressed the pH of the waste to about 5.0.
2. Alum was ineffective in chemical treatment of the wastewater. It is
proposed that complexes are formed between aluminum and ligands present in
the wastewater, resulting in the solubilization of aluminum and the
inhibition of its effectiveness as a coagulant.
3. Two organic cationic polyelectrolytes were shown to be effective
coagulants, but only at high dosages. The cost of these polymers most
likely precludes their use on a large scale.
4. The analytical method for the determination of oil and grease was not
effective in demonstrating removal of these materials from the wastewater.
36
-------
SECTION 4
BIOLOGICAL TREATMENT OF SYNTHETIC COAL CONVERSION WASTEWATERS: PART 1
Aerobic biological processes will most likely be the principal means of
treating coal conversion wastewaters for the removal of phenols and the
other organic impurities. In order to evaluate the biological treatability
of a wastewater, biotreatability studies should be conducted using the
specific wastewater for which the treatment is being developed. In this
study, however, and at this time, it is not feasible to use actual
wastewaters from coal conversion operations since coal conversion processes
are still in the developmental stage and it is unlikely that a suitable,
consistent, and representative wastewater could be obtained. Accordingly,
a synthetic organic wastewater was formulated to provide a mixture of
organic compounds, at known and reproducible concentrations, to be used in
acclimatizing and maintaining microbial cultures for preliminary
biotreatability studies. The synthetic wastewater was used to feed several
bench-scale pilot reactors. In addition to generating acclimatized
organisms for biodegradability studies (see Section 9), analysis of
effluents from the reactors provides information on wastewater
characteristics before and after various degrees of biological treatment.
This section presents preliminary results of this biotreatability study.
37
-------
FORMULATION OF SYNTHETIC COAL CONVERSION WASTEWATER
Several criteria were employed in choosing specific compounds to be
included in the synthetic wastewater, and their concentrations. Because it
was desired to use this waste as a means of developing an acclimatized
culture of microorganisms, most of the compounds selected are known or
thought to be biodegradable. However, not all of the identified
constituents of coal conversion wastewfters can be utilized by
microorganisms. Accordingly, some compounds presumed to be slowly
degradable or non-degradable, as deduced from earlier biodegradation
experiments (1), were also included (e.g., 2-indanol, indene,
2-methylquinoline, and 3,5-xylenol).
In formulating the composition of the synthetic wastewater, it was
desired that concentrations of the various components should be similar to
those encountered in real wastewaters. Accordingly, reference was made to
a summary of the constituents identified in coal conversion wastewaters
(1), and the range, midrange, and median concentrations were determined for
each constituent and for each class of compounds (e.g., cresols, xylenols,
heterocyclic N-compounds, etc.). From each class, one or more compounds
were chosen based upon considerations of biodegradability and reported
concentration. The specific compounds chosen were usually the compounds
within each class which were reported at the highest concentrations in the
real wastewaters. Often, if a class contained many components, or if
differences in biodegradability among the components of a given class were
anticipated, more than one chemical from that class was chosen. The
concentration selected was the median value reported for that compound in
the real wastewater, or the median of the class if only one compound from
38
-------
that class was picked. When the concentration of a specific compound
selected was not known, it was included in the synthetic wastewater at the
median concentration for its class.
Table 6 presents the composition of the wastewater formulated in this
manner. Twenty-eight organic components are included, as well as inorganic
nutrients and pH-buffers. The synthetic wastewater represents all major
classes of organics present in real wastewaters for which data are
available, and virtually all specific organic compounds which have been
reported to be present at high concentration. The total organic carbon
(TOG) concentration of all the components is 4,636 mg/1.
DESCRIPTION OF PILOT UNITS
Four 25-liter biological reactors were constructed for use in the
initial phases of the pilot program. (The number of reactors has since
been increased to eight.) Each reactor consists of a 7-1/2 inch ID lucite
tube, four feet long, fitted at the bottom with a stainless steel cone with
a 45° slope (Figure 13). Each reactor has overflow and sampling
connections located at appropriate heights to retain the desired volume of
contents in the reactor and to permit withdrawal of samples from desired
elevations. The stainless steel cone is equipped with connections to
permit draining of the unit, if desired, and nipples for introducing air
and feed solution at the bottom of the cone.
A compressor, operating through a pressure regulator, supplies air to
each reactor at a rate adequate to insure thorough mixing and maintenance
of aerobic conditions in the mixed liquor at all times. The rate of air
supply is controlled through use of rotameters and needle valves.
39
-------
TABLE 6. COMPOSITION OF SYNTHETIC COAL CONVERSION WASTEWATER
Compound
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
27.
28.
Phenol
Resorcinol
Catechol
Acetic Acid
o-Cresol
p-Cresol
3,4-Xylenol
2,3-Xylenol
Pyridine
Benzoic Acid
4-Ethylpyridine
4-Methylcatechol
Acetophenone
2-Indanol
Indene
Indole
5-Me thy Ire sore inol
2-Naphthol
2,3,5-Trimethylphenol
2-Me thy Iqu incline
3,5-Xylenol
3-Ethylphenol
Aniline
Hexanoic Acid
1-Naphthol
Quinoline
Naphthalene
Anthracene
NH4C1 (1,000 mg/1 as N)
MgS04 ' 7H20
CaCl2
FeNaEDTA
Phosphate Buffer: KH2P04
K2HP04
Na7HP04
Concentration, mg/1
2,000
1,000
1,000
400
400
250
250
250
120
100
100
100
50
50
50
50
50
50
50
40
40
30
20
20
20
10
5
0.2
theoretical TOC = 4,636 mg/1
3,820
22.5
27.5
0.34
170
435
• 7H70 668
40
-------
SAMPLING PORTS-
OVERFLOW
REACTOR STAilD
EXHAUST
SYSTEM
FEED SOLUTION
VARIABLE SPEED
ti—rr* PERISTALIC PUMP
AIR SUPPLY
ROTAMETER
PLEXIGLASS REACTOR
GLASS FEED TUBE
STAINLESS STEEL CONE
Figure 13. Diagram of experimental biological reactors.
41
-------
The units were fed synthetic wastewater from a glass storage reservoir
mounted on a large magnetic mixer. The wastewater was introduced into each
reactor by a variable-speed peristaltic pump. During this phase of the
biotreatability study, the reactors were operated as continuous-flow
activated sludge systems with no recycle of solids (biomass). Hence,
solids residence time (sludge age) equalled hydraulic detention time.
Hydraulic detention times of 5, 10, and 20 days were investigated during
this first phase of study. The pump feeding the 5-day reactor was operated
continuously. Pumps supplying feed to the other reactors were actuated by
a clock which operated them for a predetermined period once every half
hour. Two parallel reactors were operated at a 20-day detention time: one
reactor was isolated for use as a chemostat to provide seed organisms for
parallel biodegradation investigations, while the other 20-day reactor was
used with the 5- and 10-day reactors to provide operating data to
characterize reactor performance as a function of solids residence time.
Overflow from each reactor was collected in a glass reservoir and the
amount of wastewater actually fed was determined daily by measuring the
amount of effluent collected in that container.
Because of the potential hazards associated with some of the chemicals
in the wastewater, and because it was desired to eliminate objectionable
odors in the working area, an exhaust system was installed to vent the
units continuously to the outside of the building. The exhaust system
consisted of a blower mounted at the outside wall, thereby maintaining the
air ducts under a slight vacuum to insure that gases from the reactors
always flow into the exhaust system and not into the room. The feed
reservoir was also vented to the exhaust system in order to prevent the
escape of gases into the room from that unit, as well.
42
-------
OPERATION OF PILOT UNITS
The synthetic wastewater was made up in 16-liter batches.
Carbon-filtered Chapel Hill tap water was used as dilution water to which
the 28 constituents, shown in Table 6, were added. This was accomplished
by adding appropriate quantities from concentrated stock solutions,
prepared periodically from reagent grade chemicals and stored under
refrigeration until use. It was found that in order to prepare some of the
concentrated solutions an organic solvent was required to maintain
solubility of the component organics. Accordingly, acetone was employed
for this purpose. While this introduced an extra constituent into the
wastewater, it was believed that much of the acetone would be removed
through air stripping in view of the long detention times in the reactors.
Hence, the TOC concentration of the raw wastewater was actually somewhat
higher than that shown in Table 6.
The reactors were started up using activated sludge from one of the
Durham, North Carolina municipal wastewater treatment plants. The feed of
synthetic wastewater was increased gradually over a period of several days
to allow time for acclimatization of the microorganisms to the wastewater.
However, during the first few weeks after startup, all of the units began
to fail as evidenced by increased TOC concentrations in the effluents and
decreased solids concentrations in the reactors. Failure occurred first in
the five-day reactor, then in the ten- and twenty-day reactors. The exact
reason for failure is unknown, but several possibilities have been
considered. Operating procedures during the early stages of the
investigation were uncertain and made it possible for the concentration of
dissolved oxygen in the reactors to drop occasionally to zero. Also, the
43
-------
pH decreased to low levels (approximately 4) and remained there for
extended periods. Further there is a possibility that some wastewater
constituents could have exerted a toxic effect on the microorganisms aa
concentrations of the constituents built up in the reactor during the
period following startup. The pattern of failure, i.e. in order of
increasing reactor detention time, is consistent with the latter hypothesis.
Because of the possibility of toxic effects, and a desire to stabilize
operations as quickly as possible, it was decided to reduce the strength of
the synthetic feed to one-quarter of that in Table 6 during these initial
investigations. Other investigators (3, 4) have had to resort to similar
dilution procedures in order to treat coal conversion wastewaters
biologically. The resulting diluted version, with a theoretical TOC of
1,159 mg/1 is not inconsistent when compared with biotreatability
experiments being conducted by others. (The concentration of TOC measured
in the feed averaged 1,600 mg/1 over the course of the runs due to the
addition of acetone to solubilize the organic constituents in the feed.)
At a later date, the question of treating the synthetic wastewater at
higher strengths will be addressed. Accordingly, the reactors were started
up again using a synthetic wastewater diluted to one-quarter of the
concentrations specified in Table 6.
It should be noted that there was a significant change in color of the
synthetic feed solution over the several days during which it was used to
feed the reactors. Attempts were made to determine possible changes in
wastewater composition during this time through periodic measurements of
TOC and chromatographic scans using high performance liquid chromatography
(HPLC). Chemical changes accompanying the change in color from clear to
brown appeared to be minimal.
44
-------
Routine sampling of each reactor was performed three times a week.
Parameters measured included temperature, pH, mixed liquor suspended solids
(MLSS), mixed liquor volatile suspended solids (MLVSS), sludge volume index
(SVI), and total organic carbon (TOG). pH was measured
potentiometrically. Mixed liquor suspended solids concentrations were
determined using glass fiber filters in a Buchner funnel, followed by
drying of the filter in an aluminum dish at 103 C for 24 hours.
Filtrates from MLSS analyses were collected for total organic carbon
determinations using a Beckman 915 Carbon Analyzer. Sludge volume index
(SVI) was determined by allowing mixed liquor from the reactors to settle
for thirty minutes in a one liter graduated cylinder and calculating the
settled volume occupied by the MLSS (8).
Other samples were collected as desired for the measurement of
biochemical oxygen demand (BOD), chemical oxygen demand (COD), and for more
detailed analyses, including specific organic compounds using HPLC and
GC/MS, aquatic bioassays, and assessment of health effects. BOD and COD
analyses were conducted on samples from which suspended materials had been
removed through glass fiber filtration. Samples for HPLC and GC/MS
analysis and for aquatic bioassay and health effects assessment were
centrifuged, filtered, and frozen. The results of these analyses are
presented in Section 7.
PRELIMINARY RESULTS
Figures 14 through 17 show performance characteristics for each reactor
over the period from May to October 1978. The reactors operated without
serious incident from the beginning of May to the middle of June. The
-------
JUNE JULY
1978
x
a
MAR
9
8
7
6
5
4
3
APRIL
MAY
JUNE JULY
1978
AUG
SEPT
OCT
MAR ' APRIL MAY JUNE JULY AUG
1978
SEPT
OCT
Figure 14. Performance characteristics of the 5-day reactor.
46
-------
350
~ 300
_j
O 250
5
~ 200
o
g 150
100
50
loss of
aeration
MAR APRIL
MAY
JUNE ' JULY
1978
AUG
SEPT
OCT
1400
1200
<3 80°
? 600
% 400
200
0
MAR
_L
« «
APRIL MAY JUNE ' JULY ' AUG
1978
SEPT
OCT
9
8
7
6
5
4
3
APRIL
MAY
JUNE JULY
1978
AUG
SEPT
OCT
Figure 15. Performance characteristics of the 10-day reactor.
47
-------
350
~ 300
_i
0 250
5
~ 200
o
g 150
100
50
overfeed
_L
-L
loss of
aeration
MAR APRIL
MAY JUNE JULY
1978
AUG
SEPT
OCT
1400
1200
^1000
O 800
-~ 600
^ 400
200
0
1 - 1
,i, J
• i
z
a
MAR APRIL
9
8
7
6
5
4
3
MAY
JUNE JULY AUG SEPT
1978
OCT
MAR
APRIL
MAY
JUNE JULY
1978
AUG
SEPT
OCT
Figure 16. Performance characteristics of the first 20-day reactor.
48
-------
overfeed
loss of
aeration
ii
MAR APRIL
1400
1200
1000
800
600
400
200
0
MAY JUNE JULY
1978
AUG
SEPT
OCT
MAR APRIL
MAY
JUNE JULY
1978
AUG
SEPT
OCT
9
8
7
6
Is
4
3
2
i i I i i i i i
MAY JUNE JULY
1978
MAR APRIL
AUG
SEPT
OCT
Figure 17. Performance characteristics of the second 20-day reactor.
49
-------
operational data suggested that they had reached approximate steady-state
performance, and intensive data collection for this pattern of operation
was initiated in early June. Five sets of filtered samples from the
reactors were analyzed for BOD, COD, nitrogen species, and phosphorus as
shown in Table 7.
It had been planned that the analyses would be continued at intervals
of two days over a period of at least two weeks. If the data then
indicated that steady-state had, in fact, been attained, intensive sampling
would have been discontinued and the operations modified to another set of
reactor conditions. During the intensive sampling period in June, however,
the data for TOC and mixed liquor suspended solids indicated clearly that
steady-state operation had not been attained. Effluent TOC in all of the
reactors rose sharply beginning about June 9. This led to a decision to
postpone the intensive analysis program until more consistent performance
could be achieved.
The exact cause for the substantial change in performance which
occurred in June is unknown. However, a short time earlier, the time clock
controlling the feed to the reactors malfunctioned, resulting in an
overfeed to the 10- and 20-day reactors. This malfunction was corrected
and the feed rate was readjusted for normal operation.
50
-------
TABLE 7. SUMMARY OF REACTOR PERFORMANCE DURING PERIODS OF INTENSIVE ANALYSIS
Date
(1978)
5/30
6/5
9/12
5/30
6/1
6/3
6/5
6/7
5/30
6/1
6/3
6/5
6/7
9/8
9/12
9/14
9/16
9/18
5/30
6/1
6/3
6/5
6/7
9/8
9/12
9/14
9/16
9/18
5/30
6/1
6/3
6/5
6/7
9/8
9/12
9/14
9/16
9/18
TOC
Sample "g/1
Raw Watte
W 1*
M II
5-day Reactor 430
399
463
469
" 521
10-day Reactor 95
93
98
130
143
90
" 112
112
" 116
" 119
20-day Reactor *l 47
64
65
70
70
34
51
47
" 53
57
20-day Reactor »2 57
59
57
99
123
39
53
" 51
54
56
BOD
•g/1
3,520
2,880
4,140
1,115
870
960
1,055
1,100
179
140
171
245
240
43
26
25
33
47
30
45
80
52
5
7
8
7
73
18
38
170
183
5
5
4
7
N02 N03
COD .g/1 «g/l
•g/1 a* N at N
5.880 0.03 11.0
5,800
5,450
1,600 0.005 3.3
1,648
1,728
1,744
2,112 0.12 6.8
400 0.064 2.0
360
488
532
616 0.05 5.6
275
315
330
320
340 0.07 5.5
348
352
400
368 0.07 5.6
190
180
210
190
292 0.07 3.2
280
356
496
552 0.06 4.4
200
195
220
240
NH3
•g/1
a* N
243
228
228
209
234
222
222
217
231
225
247
249
240
TKN
•g/1
a« N
239
243
273
370
231
243
330
t
242
246
330
244
254
290
Total Ortho-
Phosphate Phosphate
•g/1 »g/l
423
68 46
42 50
106 99
35 38
369 J33
42 41
435 400
50 51
-------
During July, August, and September, mixed liquor suspended solids and
TOC data indicated reasonably steady performance, with the possible
exception of the 5-day reactor, which had performed irregularly since
startup. In all units there was a pronounced tendency for pH to drift
downward during this period, although the change in pH did not appear to
affect the stability of the MLSS and effluent TOC. Accordingly, additional
samples were taken during September for detailed chemical analysis, as
shown in Table 7. Because of its erratic performance, the 5-day reactor
was not sampled intensively during this period. The 10- and 20-day
reactors produced very low effluent BOD's, indicating that almost all of
the biodegradable material had been removed. The COD reductions are
consistent with the reduction in TOC exhibited in Figures 14 through 17.
The nitrogen and phosphorus measurements indicate that there are sufficient
nutrients for biological activity and that microbial growth was not
inhibited by a lack of nutrients. The distribution among the nitrogen
species shows that no nitrification took place.
Although the performance of the reactors appeared to be reasonably
consistent during the September sampling period, the pH was unstable and
continued to drift downward as shown in Figures 14 through 17, indicating
clearly that steady-state operation had not really been attained. During
October, the pH in the reactors reached levels lower than 4.0, causing
concern about reactor stability due to the depressed pH. This concern was
compounded by sharp rises in effluent TOC following loss of aeration for
several hours because of compressor failure. Accordingly, in late October
this series of experiments was terminated.
52
-------
SUMMARY OF PRELIMINARY RESULTS
Overall performance of the units from March through October may be
summarized with a few pertinent observations. All of the reactors showed
excellent TOC removals from the feed level of approximately 1,600 mg/1.
Figure 18 summarizes TOC removal data for the months of July, August, and
September before major excursions in pH were experienced. The 5-day
reactor was capable of producing an average effluent TOC of about 200 mg/1,
with a range extending from about 80 to 300 mg/1. The 10-day reactor
produced an average effluent BOD of about 80 mg/1, with more consistent
performance as shown by the narrower range of approximately 60 to 120
mg/l. The two 20-day reactors performed in substantially identical
fashion, with effluent TOC's averaging 45 mg/L and a rather narrow
operating range of approximately 40-60 mg/1. Table 8 summarizes the
average performance of the reactors for the months of July, August, and
September taken from the data in Figures 14 through 17 and Table 7.
Due to continued difficulties with pH variations, changes were made in
the character of the synthetic wastewater to provide additional buffer
capacity and to eliminate acetone in preparing the synthetic feed. The
results of this modification are described in the following section.
53
-------
300 p
^
200
o
— 100
APPROXIMATE
RANGE
0 5 10 15
DETENTION TIME, 0C (DAYS)
Figure 18. Effect of residence time on reactor performance and stability.
20
-------
TABLE 8. AVERAGE QUALITY OF EFFLUENT FROM BIOLOGICAL TREATMENT UNITS
(All values in mg/1.)
BOD
COD
TOC
MLSS
Raw
Wastewater
3,510
5,710
1,600
Reactor
5
1,020
1,770
200
700
Detention
10
32
310
80
900
Time
20
7
192
45
950
(Days)
20
5
214
45
900
55
-------
SECTION 5
BIOLOGICAL TREATMENT OF SYNTHETIC COAL CONVERSION WASTEWATER'.PART 2
OPERATING PROCEDURES
In order to overcome the pH variability and resulting instability of
the biological reactors discussed in the previous section, positive pH
control was established through the use of a stronger phosphate buffer
system. Several other inorganic components were modified as well; a
complete listing of the inorganic constituents, including the phosphate
buffer, is provided in Table 9. It is unlikely that pH control will be a
problem in treating real conversion wastewaters due to the presence of
abundant amounts of bicarbonate/carbonate alkalinity in such wastewaters
(2).
In addition to providing a stronger buffer system for pH control, the
procedure for preparing the organic constituents was modified to eliminate
the need for large concentrations of acetone which had been employed in the
first phase to solubilize several of the slightly soluble components.
Stock solutions of the following form were prepared, depending upon the
aqueous solubility of the constituents:
a) organics dissolved in distilled water;
b) organics dissolved in other organics;
c) organics dissolved in methanol.
56
-------
TABLE 9. CONCENTRATION OF INORGANIC CONSTITUENTS IN
QUARTER-STRENGTH SYNTHETIC WASTEWATER
Concentration, mg/1
NH Cl 955
4
MgSO^ ' 7H20 5.63
CaCl 6.88
FeNaEDTA 0.085
NaHCO 75.0
Phosphate Buffer:
KH P04 213
544
* 7H.O 835
The remaining organic constituents were stored as dry powders and weighed
out as needed. Only three constituents were dissolved in methanol,
resulting in a substantial reduction in the need for extraneous organic
solvents. The TOC attributable to the methanol in the quarter-strength
wastewater was approximately 35 to 50 mg/1. The concentrations of the
origin8^ 28 organic constituents in the synthetic wastewater remained
unchanged.
The new organic stock solutions necessitated the make-up of larger feed
batches to maintain both wastewater consistency and a reasonable ease of
chemical handling. Initially, the feed was prepared in 100-liter batches.
57
-------
The synthetic wastewater was mixed in a stainless steel tank which was
connected to a 5-gallon glass vessel. The feed pumps for the reactors drew
wastewater from this glass vessel. Each 100-liter batch lasted for about
10 days. In February 1979 three new reactors were put on line, making a
total of seven reactors in operation. In conjunction with the start-up of
the new reactors, the volume of the synthetic wastewater batches was
increased to 200 liters, which represents the maximum capacity of the
system. The 200-liter batches provided continuous feed for approximately 9
days.
The duration of the feed batches led to concerns over possible chemical
changes in the character of the wastewater. Chromatogaphic analyses
established that some changes do occur but, as reported earlier, these
changes appear to be minimal.
During this second period of study, operation of the reactors and the
sampling procedures remained essentially unchanged. The only exceptions
were in evaluating sludge settleability and in the rates of air flow to the
reactors. The sludge volume index (SVI) test (see Section 4) became
difficult to conduct because the contents of the reactor became quite dark
when the pH was increased to the neutral range. Thus, the solid/liquid
interface was difficult to discern. Also, it did not appear that zone
settling was occurring. A new operational procedure was substituted for
the SVI test to provide an indication of sludge settleability. A one-liter
sample of mixed liquor from each reactor was placed in a 1,000 ml graduated
cylinder and allowed to settle for 30 minutes. After settling, the top 500
ml was decanted through a glass port constructed in the cylinder. This
supernatant liquid was mixed on a magnetic stirrer and a suspended solids
58
-------
analysis was performed in accordance with the MLSS procedure. A comparison
between the MLSS of the supernatant and the MLSS in the reactor provided an
operational measure of sludge settleability.
The other modification involved a reduction in the air flow rate to
each reactor. Between December 1978 and February 1979, the air flow rate
was reduced in several steps from 10 liters per minute to 1.5 or 2 liters
per minute, depending upon the reactor. Reactors with hydraulic detention
times of 5 days or less received 2 liters of air per minute, while the
others received 1.5 liters per minute. The change was undertaken mainly to
reduce foaming problems which had developed. The lower air flow rates
still provided intimate mixing and a generous supply of dissolved oxygen to
meet the metabolic needs of the organisms.
The three new reactors which were started up in February were operated
in the same manner as the other four reactors, i.e. completely mixed
activated sludge systems with no recycle of biomass. The new reactors were
operated at detention times of 3, 7.5, and 40 days. The 7.5- and 40-day
reactors were fed intermittently, at 30-minute intervals, while the 3-day
reactor was fed continuously. The three new reactors were started up using
the effluent from the first four reactors. (Since no settling is provided
in the experimental set-up, the effluent contains roetabolically-active
organisms as well as residual TOC.) The effluent was added to the reactors
three times per week over a 10-day period. The reactor contents were
aerated until the desired operating volume was developed, after which
feeding of the synthetic wastewater commenced at the appropriate rate.
Some mechanical difficulties were encountered at various times during
this second study period. These operational malfunctions probably account
59
-------
for a number of the transient responses in reactor performance.
Operational problems are listed in Table 10, along with the reactor(s)
affected.
RESULTS
Figures 19 through 24 show the operating characteristics of the 5-,
7.5-, 10-, the two 20-, and the 40-day reactors. The effluent TOG, in
general, decreases with increasing retention time, reflecting improved
treatment efficiency. (The influent TOC during this period of operation
was measured to be 1,040 +120 mg/1.) It should be noted that the scales
for each of the figures are not the same, so that caution must be exercised
in comparing the results. No difficulties were encountered in controlling
pH due to the increased buffer capacity of the raw feed; the pH held steady
at 6.9 to 7.4.
Attempts to treat the wastewater with a 3-day residence time met with
failure. Immediately after feeding of the 3-day reactor commenced, the
effluent TOC began to rise and within a few days approached the influent
TOC. This pattern was observed a second time, implying that the wastewater
cannot be treated with such a low sludge age system.
60
-------
TABLE 10. RECORD OF EQUIPMENT MALFUNCTIONS
DAY
46
60
62
76
79
80
82
105
113
115
119
143
146
146
148
218
220
DESCRIPTION
loss of air - 24 hrs maximum
pump malfunction - no feed for 2 days
reactor overfed
loss of air - 12 hrs maximum
loss of air - 24 hrs maximum
loss of air - 8 to 10 hrs
feed off due to flooding
reactor overfed
reactor underfed
reactor underfed; reactor leak
reactor . underfed
reactor overfed
reactor underfed
loss of air
reactor overfed
punctured feed line, reactor leak
punctured feed line, reactor leak
REACTOR(S) AFFECTED
all
5-day
5-day
all
all
all
all
20-day
5-day
3-day
3-, 5-, 7.5-day
5-day
5-day
all
5-day
3-day
40 -day
61
-------
600
O
-ZL
o
O
400
O
° 200
0
TOC
750
500
0 30 60 90 120 150 180
DflYS
Figure 19. Effluent TOC and mixed liquor volatile suspended solids
concentration in 5-day reactor.
62
-------
O
O
C-5
z
O
300
200
100
0
TOC
600
300 -
110 125 HO 155 170 185 200
DflYS
900
155 170
DRYS
Figure 20. Effluent TOC and mixed liquor volatile suspended solids
concentration in 7.5-day reactor.
200
63
-------
o
z-
O
o
300
o 200
lOO
TOC
i20Q
80 120 160 200
QRYS
Figure 21.
80 120 160 200
DRYS
Effluent TOC and mixed liquor volatile suspended solids
concentration in 10-day reactor.
64
-------
CJ
z
o
300
200
100
TOC
0
1200
800
400 -
0
50
100 150
DRYS
200
150
200
DRYS
250
250
pigure 22. Effluent TOC and mixed liquor volatile suspended solids
concentration in first 20-day reactor.
65
-------
o
z
o
0
Z
O
o
300
200
100
TOC
0
1200
800
400
40
MLVSS
0
80 120 160
DflYS
80
120
DRYS
200 240
160 200 240
Figure 23. Effluent TOC and mixed liquor volatile suspended solids
concentration in second 20-day reactor.
66
-------
CD
300
200
100
TOC
0
no iso
190 230
QRYS
270 310
g
o
o
250 -
0
110
150
190 230
DRYS
270 310
Figure 24. Effluent TOC and mixed liquor volatile suspended solids
concentration in 40-day reactor.
67
-------
A closer look at the TOC data in Figures 19 to 24 shows that, in
general, reasonably steady performance was maintained for about 140 to 170
days after which the effluent TOC increased somewhat. In fact, there
appears to be a slight upward trend in the TOC data, and a decrease in
MLVSS over the period of observation. Accordingly, it may be inappropriate
to speak of steady-state behavior, despite the rather consistent
performance of the reactors over this observation period of up to 8
months. Some of the observed fluctuations in the TOC and MLVSS data may be
attributed to the operational malfunctions as itemized in Table 10.
Additionally, significant fluctuations in the ambient temperature began at
about the 160th day of operation, marking the change from winter to
spring. The external perturbations may have influenced microbial kinetics
and altered the biomass population and distribution in the reactors,
thereby eliciting the unsteady responses observed in reactor performance.
The reason for this shift in performance characteristics is being
investigated.
Analyses of the effluent BOD and COD and the concentration of phenols
were also made at selected times during this period of investigation and
the results are shown in Table 11. These numbers, when compared against
the measured influent concentrations of 1,780, 2,830, and 575 mg/1 of BOD
COD, and phenols, respectively, reflect the excellent degrees of treatment
which were achieved. (Phenols were determined using the 4-aminoantipyrine
procedure (8, 9) which responds only to certain of the phenolic
constituents.)
Figures 25 to 27 present correlations among effluent BOD, COD, and TOC
illustrating the strong inter-dependence of these 3 parameters for this
68
-------
TABLE 11. SUMMARY OF REACTOR PERFORMANCE (All values in mg/1.)
A. 5-Day Reactor
DAY TOG*
126
131
133
140
147
154
161
168
169
175
261
235
232
271
340
348
362
363
361
362
BOD
112
126
235
485
430
360
150
COD
670
670
850
1,160
1,080
825
1,025
PHENOLS
54
94
33
186
940
B. 7.5-Day Reactor
DAY
164
168
175
185
192
194
C. UHDay Reactor
DAY
TOC
175
176
117
168
195
176
TOC
126
133
140
147
154
161
168
175
185
192
198
140
128
133
180*
116
112
148
182
180
158
175
BOD
10
3
6
10
BOD
COD
570
435
445
465
COD
5
5
5
57+
8
9
6
6
8
6
11
480
430
460
700+
460
470
410
460
380
465
400
PHENOLS
0.70
1.16
PHENOLS
71 +
0.62
3.3
(continued)
69
-------
TABLE 11 (continued)
D. 20-Day Reactor
DAY
TOG
BOD
COD
PHENOLS
126
133
136
140
147
150
154
157
161
168
175
185
192
196
198
203
204
210
217
218
224
226
231
233
E. 40-Day Reactor
DAY
193
198
205
210
212
219
224
226
231
240
252
254
259
273
282
114
106
136
114
129
111
111
92
108
88
105
155
167
176
176
175
175
164
199
201
182
188
194
193
TOC*
147
156
164
161
156
172
151
157
152
154
146
149
148
142
166
3
2
-
4
2
-
2
-
3
2
3
2
1
-
3
-
-
3
-
-
-
3
4
-
BOD
_
1
2
-
-
-
1
2
-
1
-
1
3
310
370
-
355
320
-
360
-
350
400
420
415
385
-
420
-
-
450
-
-
460
-
465
' —
COD
340
345
-
420
-
-
430
-
400
-
375
-
-
-
_
0.43
0.35
-
-
0.35
-
0.29
-
-
-
-
-
0.19
-
-
0.18
-
-
0.22
-
-
-
0.25
PHENOLS
_
-
0.11
-
0.18
0.12
-
0.15
-
0.10
-
0.11
—
—
0.09
"""Data questionable; all values are high for Lhis date, reflecting
probable equipment malfunction.
*Many of the TOC measurements reported in th« Table were made on sample]
taken the day before or the day after the date indicated.
70
-------
0 100200300400500600700800900 1000 1100 1200
TOC (MG/L)
Figure 25. Correlation between COD and TOC in biologically-treated wastewater.
-------
0 100 200 300 400 500 600 700 800 900 1000 1100 1200
TOC (MG/L)
Figure 26. Correlation between BOD and TOC in biologically-treated wastewater.
-------
200 400 600 800 1000 1200 1400 1600 1800 2000 2200 2400
BOD (MG/L)
Figure 27. Correlation between BOD and COD in biologically-treated wastewater.
-------
wastewater. The following relationships were developed from these plots:
TOC = 190 + 0.47 (BOD)
COD =2.57 (TOC) + 60
COD =642+1.16 (BOD)
The cluster of data in the low BOD range (<25 mg/1) were omitted in
developing these correlations. Consequently, the linear correlations
cannot be applied to effluent BOD concentrations less than 25 mg/1.
In attempting to summarize reactor performance over this observation
period, it was necessary to divide some of the data into separate periods
during which reasonably steady-state performance was observed. The TOC
data shown in Figures 19 to 24 were used for this purpose. For example
Figure 19 shows consistent performance of the 5-day reactor from the 20th
day to the 115th day, with an average TOC of 199 +30 mg/1. After the 139th
day, the average TOC was reasonably constant at 332 +40 mg/1. As indicated
above, the reason for this shift is not clear at this time but is believed
to be related to conditions external to the reactors, such as temperature
changes. Accordingly, Table 12 lists the periods of apparent steady-state
operation, and Table 13 summarizes the effluent data during these
steady-state periods. The summary results are plotted in Figures 28 to
31. It is clear that the removals of TOC, COD, BOD and phenols improve
with increasing sludge age or residence time in the reactors. BOD and
phenols are completely removed after 10 to 15 days, while the extent of TOC
and COD removal is essentially complete after the same time period, leaving
residual non-biodegradable TOC and COD concentrations of approximately 1QO
to 160 and 350 to 400 mg/1, respectively. These residuals cannot be
compared directly with the correlations above, since the correlations were
74
-------
TABLE 12. PERIODS OF STEADY STATE PERFORMANCE
REACTOR PERIOD 1 PERIOD 2
5-day
7.5-day
10-day
20 -day,
20-day,
40-day
#1
#2
day
day
day
day
day
day
20 -
111
1 -
1 -
1 -
111
day 115
- day 195
day 143
day 174
day 171
- day 174
day
day
day
day
day
139
146
176
174
176
- day
- day
- day
- day
- day
180
199
248
239
307
75
-------
TABLE 13. SUMMARY OF AVERAGE STEADY STATE REACTOR PERFORMANCE
(all values in mg/1)
REACTOR
5-day
7.5-day
10-day
20-day, #2
40-day
REACTOR
5-day
10-day
20-day #1
20-day #2
40-day
MLVSS
470
495
650
725
710
500
MLVSS
305
410
535
445
470
First
TOC
199
173
125
106
99
Second
TOC
332
158
179
191
157
Steady State Period
BOD COD PHENOLS
7.3 480 0.5
5 455
2.6 ISO
Steady State Period
BOD COD PHENOLS
310 980 65
7.7 435 2.0
2.7 430 0.30
1.6 386 0.12
Raw Feed Characteristics
MLVSS
Feed
TOC
1,040
BOD COD PHENOT.R
1,780 2,830 575
76
-------
500
400
300
(D
S
8 200
I-
ui
u. 100
10
X 1st steady state
O 2nd steady state
15
20
25
30
DETENTION TIME, 9 ( DAYS )
c
35
_TL
40
Figure 28. Effect of residence time on TOG removal,
-------
1000
800
600
00
Q
8
ill
D
400
X 1st steady state
O 2nd steady state
200
10
15
20
25
DETENTION TIME, 0 (DAYS)
C
Figure 29. Effect of residence time on COD removal.
30
35
40
-------
300
1st steady state
O 2nd steady state
UJ
10
15 20 25
DETENTION TIME, 9 ( DAYS )
30
35
Figure 30. Effect of residence time on BOD removal.
-------
50
o
5
40
oo
o
o
ui
a.
ai
D
30
20
10
X 1st steady state
O 2nd steady state
10
15
20
25
30
DETENTION TIME, 9 (DAYS)
c
Figure 31, Effect of residence time on phenols removal.
35
40
-------
not developed for low effluent BOD concentrations. (See Section 7 for
further discussion of these residuals.)
Comparison of the residual TOC and COD for this second period of
operation with the residuals discussed in Section 4 (e.g. Figure 18) for
the first period of reactor operation show a better degree of treatment
(lower TOC residuals) for the first period. It is believed, in view of the
results reported in Section 3, that some of the organics from the first
period of study were precipitated at the low pH at which the reactors were
operated (approximately pH 4). Hence, removal of organics during the first
period of study may not have been entirely through a biological degradation
mechanism.
For reasons which cannot yet be explained, after 6 to 8 months of
relatively consistent performance, all of the reactors with the exception
of the 40-day reactor began to fail. The effluent TOC, COD, BOD, and
phenols began to increase, approaching the influent concentrations, and
oxygen-uptake measurements showed a significant reduction in metabolic
activity. The cause of this sudden failure is not known at present, but in
view of the long period of relatively consistent behavior, it is believed
that the failure is attributable to problems in the mode of operation of
the reactors and not to any fundamental problems in the biochemistry of the
system, i.e. it is believed that the synthetic wastewater is biologically
treatable. This point is being investigated further and will be addressed
in future reports.
81
-------
SECTION 6
KINETIC ANALYSIS OF BIOLOGICAL TREATABILITY DATA
In order to design an activated sludge process for treatment of coal
conversion wastewater, the parameters describing the kinetics of microbial
growth and substrate utilization for the given wastewater must be
determined. The data collected to date can be used to make a preliminary
determination of these requisite microbial growth coefficients as follows:
The kinetics of microbial growth can be described by the equation (11)
dX/dt - y dS/dt - k X (1)
where
X = concentration of microorganisms (biomass), in rag of mixed liquor
volatile suspended solids (MLVSS) per liter;
S = substrate concentration, in mg/1, on a BOD, COD, or TOC basis;
t = time, in days;
y = microbial yield coefficient, in rag of biomass (MLVSS) produced
per mg of substrate (on a BOD, COD, or TOC basis) consumed;
kj = microbial die-away coefficient, in days ~1.
Taking finite differences in Equation 1 and dividing through by X, the mean
biomass concentration over the time period At, one gets
(AX/At) / X = y (AS/At) / X - kd (2)
For the continuous-flow, completely-mixed reactors used in this
investigation, X is the steady-state biomass concentration in each reactor,
and At is the detention time of the reactor. Equation 2 can be re-written
82
-------
as
Here © can ^e defined as the mean cell residence time, solids retention
time, or sludge age, and is equal to the steady-state quantity of biomass
in the reactor, divided by the rate of biomass production. 6 has units
of time, and for reactor operation with no recycle of biomass, the mean
cell residence time is equal to the hydraulic retention time. The quantity
U in Equation 3 is defined as the process loading factor, or food to
microorganism ratio, and is equal to the quantity of substrate consumed (AS)
during the given reactor detention period (At) divided by the
steady-state biomass concentration (compare Equations 2 and 3). The
process loading factor can be computed on a BOD, COD, or TOC basis. Hence,
if the reciprocal of the sludge age is plotted against the process loading
factor in accordance with Equation 3, a straight line should result and the
microbial kinetic coefficients y and k can be determined.
The kinetics of substrate utilization can be described by the equation
(U)
dS/dt • (kSX)/(K + S) (4)
s
where
= specific substrate utilization rate, in mg of substrate (on a
BOD, COD, or TOC basis) per mg of biomass (MLVSS) per day;
= Michaelis-Menten coefficient or half-velocity constant, in mg/1,
on a BOD, COD, or TOC basis.
83
-------
TABLE 14. CALCULATED PROCESS LOADING FACTORS FOR BIOLOGICAL REACTORS
A. FIRST PHASE OF REACTOR OPERATION
0 c, DAYS
Process Loading Factor 5 10 20 20
UB, mg BOD/mg MLVSS-day OTTT OTST 0.18 0.19
Uc, mg COD/mg MLVSS-day 1.13 0.60 0.29 0.31
UT, mg TOC/mg MLVSS-day 0.40 0.17 0.082 0.080
B. SECOND PHASE OF REACTOR OPERATION
i) First steady-state period
0C, DAYS
Process Loading Factor _5 7.5 10 20 20 40
UB, mg BOD/mg MLVSS-day 0.48 OT27 0.12 — ~^
Uc, mg COD/mg MLVSS-day 0.63 0.36 0.17
UT, mg TOC/mg MLVSS-day 0.36 0.23 0.14 0.064 0.066 0.047
ii) Second steady-state period
Process Loading Factor
UB,
uc>
UT,
mg BOD/mg MLVSS-day
mg COD/mg MLVSS-day
mg TOC/mg MLVSS-day
5
rnr
1.22
0.47
10
0.43"
0.58
0.22
Q c , DAYS
20 20
0.17
OOQ
0.081 0.096
40
0.094
0.13
0.047
The other terms are as defined for Equation 1. Again, taking finite
differences and dividing through by X, the mean biomass concentration over
the time period At, one gets
(AS/At)/X = U = (kS)/(K + S) (5)
s
At low substrate concentrations, where S « K , a plot of U, the process
S
loading factor or food to microorganism ratio, vs S should be linear, with
84
-------
a slope of k/K . Further, Equation 5 can be manipulated to yield
s
1/U = [(K /k). 1/S] + 1/k (6)
3
Hence, if the reciprocal of the process loading factor is plotted against
the reciprocal of the steady-state effluent concentration, a straight line
should result and the kinetic coefficients k and K can be determined.
8
Using the "steady state" data summarized in Tables 8 and 13 for the two
aeries of biological treatability studies described in Sections 4 and 5,
the process loading factors for each of the reactors were calculated and
are listed in Table 14. The linear plots of the combined data from all
phases of operation, in accordance with Equation 3, are shown in Figures 32
to 34* The yield coefficients computed from the slopes of the straight
lines are 0.45, 0.18, and 0.16 based upon TOC, BOD, and COD utilization,
respectively. The die-away coefficient, determined from the intercept at
zero-loading, is a negative number (-0.015 to -0.03 days ) which casts
some doubt as to the accuracy of the "steady state" values or the validity
of the model.
figures 35 to 37 are plots of the process loading factor vs residual
substrate concentration in accordance with Equation 5 for the data from the
second-phase of reactor operation. The plots appear to be linear at the
lower concentrations, as would be the case if S « K . If the
S
non-biodegradable portion of the TOC and COD, determined from the
x-intercepts in Figures 36 and 37, are subtracted from the TOC and COD
effl°ent values, and the reduced data plotted in accordance with Equation
6 the linear relationships illustrated in Figures 38 to 40 result.
Alternative linear modifications of Equation 5 have also been tested, for
example:
85
-------
0.20
0.15
00 *-
> 0.10
<
Q
0.05
X
0.10
TOC LOADING
0.20
TOC
0.30
/MG TOC \
\MG MLVSS -DAY/
MG MLVSS
Figure 32. Relationship between solids residence time and TOC loading.
0.40
-------
0.20 |-
00
o.18 MG MLVSS
MG BOD
)
MG MLVSS-DAY
Figure 33. Relationship between solids residence time and BOD loading.
-------
0.20
0.15
00
oo
V)
< 0.10
Q
0.05
0.5
1.0
Y = 0.16
MG MLVSS
MG COO
1.5
COD LOADING
/MG COD N
I MG MLVSS - DAY/
MG MLVSS
Figure 34. Relationship between aollda residence tine and COD loading.
-------
Q
ai
oo
vo
2
z
Q
<
g
o
§
1.0 r
0.8 -
0.6 -
0.4
0.2
O 1st steady state
2nd steady state
100
200
300
400
EFFLUENT BOD ( MG/L )
Figure 35. Kinetics of substrate utilization: BOD-basis.
-------
0.50
0.40
O
ui
5
C/J
O
(D
8 0.30
O
C3
Q
<
O
8
0.20
0.10
non-biodegradable TOC
1st steady state
non-biodegradable TOC
2nd steady state
100
0 1st steady state
Q 2nd steady state
200
300
400
EFFLUENT TOC ( MG/L )
Figure 36. Kinetics of substrate utilization: TOC-basis.
-------
1.0 r
o
ULJ
s
CO
8
O
O
o
0.8
o
0.6
[1st steady state
5
<
3
8
0.4
0.2
non-biodegradable COD
2nd steady state
non-biodegradable COD
JO-
O 1st steady state
O 2nd steady state
200
400
600
800
1000
1200
1400
EFFLUENT COD (MG/L )
Figure 37. Kinetics of substrate utilization: COD-basis,
-------
Figure 38. Linearized substrate utilization plot: BOD-basis.
10.0 r
EFFLUENT BOO
( MO/L K
-------
OJ
<
O
i
g
o
z
§
g
20 t-
16 -
12 -
8 -
4 .
o
o
o
I
.01
1
.02
1 t
.03 .04
1
i
.05
i i
.06 .0
Figure 39.
EFFLUENT BIODEGRADABLE TOC ( MG/L )'1
Linearized substrate utilization plot: TOC-basis.
-------
o
8
Q
O
O
O
o
z
Q
O
Q
O
O
8.0
6.0
4.0
2.0
•3D
0
.01
.02
.03
.04
.05
.06
.07
EFFLUENT BIODEGRADABLE COD
(MG/LK
Figure 40. Linearized substrate utilization plot: COD-basis.
-------
(S/U) = 1/k (S) + K /k
s
(7)
and
(U) = -K (U/S) + k (8)
s
Table 15 lists the range of values for the specific substrate utilization
rates and the Michaelis-Menten coefficients, determined from the slopes and
intercepts of these different forms of linear plots. Also shown are the
calculated yield coefficients as determined earlier from Figures 32 to 34.
It should be noted that the values reported here for the specific
substrate utilization rate, are relatively low when compared to domestic
wastewater and many other industrial wastes, reflecting the slow kinetics
of the biochemical oxidation for phenolic wastewaters. The
Michaelis-Menten coefficients, however, are in the same range as those
reported for many other wastewaters (9).
TABLE 15. SUMMARY OF KINETIC COEFFICIENTS
Substrate
Basis
TOC
BOD
COD
Specific Substrate
Utilization Rate, k
mg substrate/mg MLVSS-day
0.33 - 1.4
0.67 - 0.74
0.64 - 1.3
Michaelis-Menten
Coefficient, Ks
mg/1
84 - 460
5.6 - 10.0
60 - 240
Microbial
Y
mg MLVSS/mg
0.45
0.18
0.16
Yield,
substrate
95
-------
Some of these values in Table 15 can be compared with those of other
investigators for other types of coal conversion wastewaters. The COD
yield coefficient is in the same range as those reported by Luthy and
Tallon (3) for full-strength, ammonia-stripped and for diluted Hygas
wastewater (y = 0.11 and 0.22, respectively). Their reported microbial
die-away coefficient was 0.02 day . On a BOD basis, Reap et al. (5)
reported a yield of 0.48 tng VSS/mg BOD and a die-away coefficient of 0.03
day for air-stripped H-Coal process wastewater. Their yield is
significantly higher than that reported here.
Table 16 lists the substrate utilization and microbial growth kinetic
coefficients determined by Drummond et al. (12) for the biological
treatment of Synthane and Synthoil wastewater. The specific substrate
utilization rates and the Michaelis Menten coefficients for the Synthane
wastewater are quite comparable to ours, but the microbial yields are
appreciably higher. The kinetic coefficients for the Synthoil wastewater
are stated to be preliminary , and are presented here for illustrative
purposes only.
The summary results reported in Table 15 should be considered only as
preliminary results at this time and should not be used for design
purposes. Biological treatability investigations are continuing and it is
anticipated that these further investigations will lead to firmer criteria
for designing biological treatment processes for coal conversion
wastewaters.
96
-------
TABLE 16. KINETIC COEFFICIENTS FROM BIOLOGICAL TREATMENT
OF SYNTHANE AND SYNTHOIL WASTEWATER*
VO
Substrate k Ks Y Kd
Basis mg substrate/mg MLVSS-day mg/1 rag MLVSS/mg substrate day"
A.
TOC
BOD
COD
Phenol
B.
TOC
BOD
COD
Phenol
Synthane Coal Gasification
0.49
0.46
0.96
0.63
Synthoil Coal Liquefaction
0.14
0.39
0.43
—
Wastewater
100
5
120
30
Wastewater*
3
14
10
—
1.1 0.04
0.53 0.02
0.47 0.05
—
1.2 0.004
0.61 0.02
0.36 0.004
—
*After Drummond et al. (12)
"""Preliminary data only
-------
SECTION 7
SPECIFIC ORGANIC ANALYSIS AND ENVIRONMENTAL ASSESSMENT
OF TREATED WASTEWATERS: PRELIMINARY RESULTS
Raw wastewater and treated effluent from the biological reactors were
collected at various times during the first two periods of study and
subjected to specific organic analysis by high performance liquid
chromatography (HPLC) and gas chromatography/mass spectrometry (GC/MS).
Only the results for HPLC analysis are available at this time. Aquatic
bioassays involving algae, Daphnia, and fish, and mammalian cytotoxicity
analyses were also conducted on the samples as a means of assessing the
aquatic and health impacts, respectively, of the biologically treated
wastewater. This section presents some of the preliminary results from
these specific organic and environmental assessment analyses.
HPLC ANALYSIS
Fresh samples of the reactor effluent were collected, filtered through
0.7 ym glass fiber filters, and injected into the HPLC. Care was taken to
insure that the samples were taken during stable periods of reactor
operation. Separation of the wastewater components in the samples was
achieved on the HPLC using a 60-minute water/acetonitrile solvent gradient
on a Waters pBondapak Clg analytical column. The eluted compounds were
98
-------
detected by both UV absorbance at 280 nm and fluorescence at 275 nm
excitation and 310 nm emission wavelengths.
Figure 41 shows chromatograms of the raw feed using both UV absorbance
and fluorescence detectors. These chromatograms were used for purposes of
compound identification. The numbered peaks correspond to the compounds
listed in Table 17; elution volumes of the constituents are also shown. It
should be noted that polarity decreases with increasing elution volume so
that the elution volumes corresponding to the chromatographic peaks can be
correlated to the polarity of the various organic compounds in the
mixture. It should also be noted that each peak may represent more than
one compound.
Figures 42 to 46 present chromatograms of the effluent from the 5-,
7.5-, 10-, 20- and 40-day reactors collected during the second series of
reactor operation (see Section 5). The UV absorbance chromatograms of the
reactor effluents were all prepared at the same sensitivity (xlO) and can
be compared directly; the chromatogram for the raw feed is at a lower
getisitivity (x5). The fluorescence chromatograms should be observed
carefully as they were each prepared at different detector sensitivities.
The UV chromatograms of the reactor effluents reflect the production of
highly polar compounds (e.g. aliphatic and aromatic acids, etc.) as a
result of biological treatment. The concentration of these polar
compounds, which appear at low and intermediate elution volumes, does not
appear to change with increasing degree of treatment, i.e. detention time.
Some of these peaks may be attributed to the production of aliphatic acids
which can be expected from the bacterial degradation of phenolic compounds.
99
-------
3,4
E
LLJ
O
If
a. *
If
9,10,11 1415,16,17.18
x tS
CM 5
LU
LU g
CO £
LU LU
cc c
§ 1
3°
U. CO
10
20
n
27
nr-v>T^i^rAf
30
50
6,7 \ 8
10 20 30 40
HPLC ELUTION VOLUME ( ml)
60
Figure 41. HPLC chromatographic profiles of raw synthetic wastewater,
(Numbers correspond to compounds in Table 17.)
100
-------
TABLE 17. IDENTIFICATION OF HPLC PEAKS
Peak
Number
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
Elution
Volume, ml
3
4
13
14
18
21
22
23
33
33
34
35
28
40
40
40
41
41
42
42
43
43
45
45
50
51
58
Compound Name
acetic, hexanoic, benzole acids
solvent
resorcinol
catechol
aniline
phenol
5-me thy 1 resorcinol
4-methylcatechol
p-cresol
o-cresol
2-indanol
acetophenone
4-ethylpyridine
pyridine
quinoline
3, 5-xylenol
3,4-xylenol
2,3-xylenol
indole
3-e thy 1 phenol
2-methylquinoline
1-naphthol
2-naphthol
2 , 3, 5-trime thy 1 phenol
indene
naphthalene
anthracene
101
-------
E
o
UJ O
a
in
<
>
O
o «»
X ^
UJ X
lei
(N
UJ c
2-2
S a
81
u. ro
10
20
30
50
60
10
20
30 40 50
HPLC ELUTION VOLUME ( ml)
Figure 42. HPLC chromatographic profiles of 5-day reactor effluent.
60
102
-------
«
III
50
60
i-
Ul
g
*i
ll
u. «
ijv
10
l\
20 30 40 50
E HPLC ELUTION VOLUME (ml)
Figure ^' HPLC chromatographic profiles of 7,5-day reactor effluent.
60
103
-------
I
o
O X
3£
00 .>
o
"
UJ
CC e
§ =
°
HPLC ELUTION VOLUME ( ml)
Figure 44. HPLC chromatographic profiles of 10-day reactor effluent,
104
-------
LU X
ii
cc S
10
20
30
40
50
60
10
20
30
HPLC ELUTION VOLUME ( ml)
Figure 45. HPLC chromatographic profiles of 20-day reactor effluent.
105
-------
I
o
S2
z x
00 £•
c >
^J ^^
CO vJ
QD c
< g
D
20
30
40
50
60
o « .
x
in 5
-
tu c
^ o
§ £
W LU
UJ
£ E
O c
U. CO
10
60
20 30 40
HPLC ELUTION VOLUME (ml)
Figure 46. HPLC chromatographlc profiles of 40-day reactor effluent.
60
106
-------
On the other end of the UV chromatograms (at the high elution volumes),
a greater reduction of the more non-polar compounds (e.g., pyridine,
quinoline, xylenols, ethylphenols, trimethylphenol and naphthols) can be
seen with increased reactor detention time. This is especially true for
the reactors with detention times greater than 5 days, where the
chromatographic peaks for these compounds were too small to detect among
the peaks for the cellular metabolites produced during biological
treatment. (It should be noted that direct probe mass spectral analysis
indicates that the two significant UV peaks at elution volumes of 47 and 51
mis are phthalic acid esters, possibly arising from the tygon tubing or
plexiglass used in construction and operation of the biological reactors.)
HPLC traces of the reactor effluents were used to obtain approximate
concentrations for several of the major constituents in the raw feed to the
reactors. Tn view of the selectivity of the fluorescence wavelengths used
for phenolic compounds and the complexity of the UV chromatograras due to
the presence of large quantities of UV-absorbing cellular by-products in
the reactor effluents, all quantitation of phenolics was accomplished using
fluorescence detection. It should be noted that the elution volume at
which a given peak occurs may vary slightly from one chromatogram to the
next depending upon column age, sample volume, and sample concentration.
Hence, quantitation of the compounds of interest was performed by preparing
several chromatograms with standard additions of these compounds. An
illustrative chromatogram of this "spiking" technique is shown in Figure
47. The concentrations of the major phenolic compounds in each of the
reactor effluents, as determined in this manner, are listed in Table 18.
concentrations in some cases are reported as less than a certain value
107
-------
O
8
UJ
o
•—
n
N
o
oo
tt
§
g J
ui
UJ
oc
10
20 30 40
HPLC ELUTION VOLUME (ml)
50
60
Figure 47. Identification and quantitation of HPLC chromatographic peaks from 10-day reactor. (A.) Chromatogram
of 10-day reactor effluent. (B.) Chromatogram of 10-day reactor effluent spiked with 20 mg/1
resorcinol, 20 mg/1 phenol, 10 mg/1 p-cresol, 10 mg/1 3,4-xylenol, and 10 mg/1 2,3,5-trimethylphenol.
-------
TABLE 18. CONCENTRATIONS OF MAJOR PHENOLIC COMPOUNDS IN REACTOR EFFLUENTS (mg/1)
Compound
catechol
resorcinol
phenol
cresols
o-cresol
p-cresol
xylenols
3,4-xylenol
2,3-xylenol
3,5-xylenol
Raw
Feed (1/4-strength)
250
250
500
162.5
100
62.5
135
62.5
62.5
10.0
5-day
reactor
4/12/79
<0.5
<0.5
0.9
22.2
33.6
5-day
reactor
4/24/79
<0.5
<0.5
0.6
30.2
31.4
7.5-day
reactor
5/7/79
<0.2
<0.2
<0.2
0.2
1.0
10-day
reactor
4/25/79
<0.5
<0.5
<0.4
0.8
2.5
20-day 20-day
reactor reactor
4/25/79 5/4/79
<0.2 <0.1
<0.2 <0.1
<0.2 <0.1
<0.005 <0.02
1.4 <0.01
40-day
reactor
8/30/79
<0.02
<0.02
<0.13
0.036
0.007
2,3,5-trimethyl-
phenol
12.5
9.0
7.0
0.6
1.3
<0.08
<0.02
<0.004
-------
where that value represents the detection limit of the HPLC fluorescence
detector for that compound at the sensitivity used for the analysis. The
table shows that removal of the phenolics increases with increased
detention time and that phenol, resorcinol and catechol are almost
completely removed by the 5-day reactor. The cresols are completely
removed within 7.5 to 10 days (to concentrations less than 1 mg/1) while a
20 day sludge age is required to reduce the concentrations of the xylenols
and trimethylphenol below 1 mg/1. These observations are in accordance
with the phenol data using the aminoantipyrine wet chemical procedure
reported earlier in Section 5, and with the biodegradability results in
Section 9. These results are significant from the standpoint of reactor
performance since a large portion of the organic carbon in the influent
feed is comprised of phenolic compounds.
Analysis of the reactor effluents is continuing using HPLC and other
fluorescence wavelengths, gas chromatography/ mass spectrometry, and direct
ion probe mass spectrometry on HPLC-separated fractions. These results
should be available in the next report in this series.
AQUATIC BIOASSAY
Traditionally, short-term lethality tests with fish have received
greatest emphasis in developing and evaluating pollution abatement
programs. More recently, the realization that elimination of lower
organisms may have serious environmental consequences has led to increased
reliance on algae and invertebrate assays as well. Several aquatic
organisms, including the fathead minnow (Pitnephales promelas), a cladoceran
(Paphnia pulex), and an alga (Selenastrum capricornutum) have become widely
110
-------
accepted bioassay organisms, and routine bioassay procedures are available
(8, 13)-
In this section, results of some preliminary bioassay experiments
performed with fathead minnows, D. pulex, and S. cajpricornutum exposed to
raw and treated synthetic coal conversion wastewaters are presented.
Although the preliminary findings suggest that toxicity of the raw
wastewater is substantially reduced following biological treatment, the
oresent data are not definitive. However, they provide basic information
regarding: a) sample collection and handling of synthetic effluents for
conducting bioassays; b) procedural modifications of bioassay methods
necessary to accommodate toxicity data interpretation; and c) formulation
of appropriate experimental protocols for more detailed analyses of acute
toxicity problems associated with coal conversion wastewaters.
Tn the preliminary studies presented herein, no attempt was made
analytically to verify actual effluent concentrations from the
biotreatability reactors nor to monitor possible loss of TOG through
volatilization or biodegradation during the course of the bioassays.
Rather, nominal concentrations of the effluents were prepared by serial
dilution procedures using volumetric glassware.
The raw synthetic wastewater for toxicity tests was drawn from the
reservoir feeding the bioreactor pilot units. All samples were drawn from
freshly-made synthetic wastewater, from 2 to 10 hours after preparation.
If the raw feed was not to be used immediately, aliquots were frozen and
stored until needed. Effluents from the 5-, 10-, and 20-day biological
reactors were collected from glass reservoirs which contained the overflow
from the pilot units. Prior to sample collection each day, the reservoirs
111
-------
were emptied to ensure that the respective effluents were less than 24
hours old. The reactor effluents were centrifuged, and the resultant
supernatant was filtered through glass wool to remove suspended solids.
Filtrates were frozen in 1 to 3 liter aliquots for the Daphnia and
Selenastrum toxicity tests. However, due to the large sample volumes
required for tests with the fathead minnows, daily collections were
centrifuged, filtered, and frozen in 5-gallon soft glass jars, until
composite effluent samples of 35-40 liters were accumulated. All samples
subjected to the aquatic bioassay tests were collected during the second
set of biotreatability studies described in Section 5.
General bioassay procedures for fathead minnows and Daphnia pulex were
conducted in accordance with Standard Methods(S). Test concentrations of
the effluents were prepared on a volume percent basis by dilution with
carbon-filtered dechlorinated tap water. The dilution water had an average
pH of 7.2, total alkalinity of 36.7 mg/1 (as CaCO-j), total hardness of
30.8 mg/1 (as CaCO-,), and total residual chlorine of 0.04 mg/1. All
tests were conducted in a constant temperature laboratory at 18 to 20 C
as static non-renewal bioassays. Two replicates, each containing 15 test
organisms, were used for each effluent concentration. Sample volumes for
each replicate were 10 liters and 100 ml, respectively, for fathead minnows
and Daphnia. For the former, the duration of the test was 96 hours and
death was used as the bioassay endpoint. For the latter, toxicity tests
were conducted for 48 hours with immobilization as the endpoint. The
effluent concentrations used in these preliminary experiments, as well as
their respective effects on both test organisms, are summarized in Table 19
112
-------
TABLE 19. EFFECTS OF RAW AND BIOLOGICALLY-TREATED WASTEWATERS
ON FATHEAD MINNOWS AND DAPHNIA PULEX
FATHEAD
MINNOWS
RAW WASTEWATER
Concentration,
t by volume
3.2
1.8
1.0
0.56
0.32
0.18
0.10
Percent
Mortality
100
100
16.6
0.0
3.3
0.0
6.7
5 -DAY REACTOR
Concentration
% by volume
30
25
20
15
05
2.5
10 -DAY
Concentration
•L by volume
28
21
18
13.5
10.0
7.5
5.6
3.2
Percent
Mortality
LOO
100
100
100
13.3
3.3
REACTOR
Percent
Mortality
40.0
20.0
20.0
0.0
0.0
0.0
0.0
0.0
DAPHNIA
PULEX
RAW WASTEWATER
Concentration,
% by volume
3.2
1.8
1.0
0.56
0.32
5-DAY
Concentration
% by volume
10
5.6
3.2
1.8
1.0
10-DAY
Concentration
% by volume
32
18
10
5.6
3.2
20 -DAY
Concentration
% by Volume
32
18
10
5.6
3.2
Percent
Immobilization
83.0
83.0
86.0
60.0
50.0
REACTOR
Percent
Immobilization
73.3
50.0
33.3
16.7
0
REACTOR
Percent
Immobilization
100
93.3
63.0
60.0
23.3
REACTOR
Percent
Immobilization
40
10
13.3
0.0
0.0
113
-------
Although several algal assays have been initiated to examine the growth
response of Selenastrum capricornutum exposed to both raw and treated
synthetic coal conversion wastewater, the current discussion is limited to
one range-finding test. This particular test was conducted to delineate
effluent concentrations for more definitive toxicity analysis, and only one
test flask was used for each concentration. However, the data are of
interest in that they provide some basis of comparison between the raw feed
and all three bioreactor effluents. In order to remove bacteria from the
effluent samples prior to preparing the dilutions, all samples were
filtered through 0.2 Mm membrane filters. Raw feed concentrations of
0.001, 0.005, 0.01, 0.05, 0.1, 0.5, 1.0, and 5.0%, as well as reactor
effluent concentrations of 0.01, 0.05. 0.1, 0.5, 1.0. 5.0, 10.0, and 50.0%
were inoculated with 1x10 algal cells per ml, and incubated for 12
days. Growth curves for Selenasjrum capricornutum cultures exposed to
selected raw and treated wastewater concentrations of interest are shown in
Figures 48 through 51.
The preliminary toxicity data for fish and Daphnia summarized in Table
19 do not meet appropriate assumptions for reliable estimation of LCc^'s
over the concentration ranges tested. However, the dose-response curves
constructed from these data for both fathead minnows (Figure 52) and
Daphnia pulex (Figure 53) reflect an obvious trend of marked toxicity
reduction of the synthetic wastewater following biological treatment.
Further, the extent of toxicity reduction appears to be related to the
degree of biological treatment as reflected by the solids residence times
of the biological reactors. Considering the number of readily
biodegradable compounds present in the raw feed, these results are not
unexpected.
114
-------
While growth data for Selenastrum capricornutum are admittedly limited,
it is of interest to note that relatively low concentrations, of the raw
and treated synthetic wastewaters either stimulated growth or were not
significantly different from the controls. On the other hand, if the 1, 5,
and 10% dilutions are compared for all the effluents tested (see Figures 48
to 51)» toxicity again appears to be reduced to a greater extent with
increasing reactor detention time. It should be pointed out, however, that
several problems were encountered with both fungal and bacterial
contamination of the algal cultures. In subsequent tests with the raw
vastewater, it was found that after several days, bacterial populations in
the test flasks began to proliferate. Therefore, the observed reduction in
cell yield may have been due to inter-specific competition, as well as
toxicity.
As the results are only of a preliminary nature, the effluents from
each of the biological reactors will be studied more fully in the future,
in order to provide definitive answers concerning acute toxicity of the
wastewaters to the three aquatic organisms described above. Specific
objectives to be considered include detailed characterization of the
toxicity curve associated with the raw wastewater and reactor effluents,
Definitive assessment of reduced toxicity resulting from biological waste
treatment, and potential loss of wastewater constituents through
volatilization and biodegradation during the course of the bioassay.
Further, studies are being initiated to ascertain whether continuous-flow
toxicity tests will provide a more realistic impact assessment than static
bioassays with fish.
115
-------
10' -
0.1%
106
E
I 10E
O
u
UJ
U
10-=
0123456789 10
Figure 48. Growth of Selenastrum capricornutum exposed to various dilutions
of raw synthetic wastewater.
116
-------
10' +•
4567
TIME (DAYS)
8 9 10 11 12
Figure
49.
Growth of Selenastrum gapricornutum exposed to various dilutions
of 5-day reactor effluent..
117
-------
1234
6 7 8 9 10 11
103
Figure 50. Growth of Selenastrum capricornutum exposed to various dilutions
10-day reactor effluent.
118
of
-------
10
CONTROL
t io5--
01234567 89 10 11 12
TIME (DAYS)
Figure -*!• Growth of Selenastrum capricornutum exposed to various dilutions
of 20-day reactor effluent.
119
-------
s
oc
o
I
ill
O
98
95
90
80
70
60
50
40
30
DC
£ 20
10
•K raw feed
O Reactor 1 ( 5 day solids
detention )
A Reactor 2 ( 10 day solids
detention )
RAW FEED
0.18
50
0.32 0.50 1.0 1.8 2.5 3.2 10 15 21 28
PER CENT EFFLUENT ( BY VOLUME )
Figure 52. Toxicity of raw and biologically-treated synthetic wastewater to fathead minnows (Pimephales promelas).
-------
99.99
<
N
00
O
ai
O
DC
UJ
Q.
Date of grab sample
3-13 - 79
4 4 - 79
4-4-79
4-4-79
= raw feed
= Reactor 1
= Reactor 2
= Reactor 3
REACTOR 2
RAW FEED
REACTOR 1
REACTOR 3
Figure 53.
0.01
PER CENT EFFLUENT ( BY VOLUME )
Toxicity of raw and biologically-treated synthetic wastewater to Daphnia pulex.
-------
Although the sample collection and handling protocols discussed above
appear to be appropriate for the fish and daphnia bioassays, the nature of
the raw wastewater and the biologically treated effluents has hindered
effective study of algal toxicity. In order to overcome the bacterial
contamination problems associated with the algal assay procedure, an axenic
(bacteria-free) culture of Selenastrum capricornutum has been isolated and
will be used for future tests.
HEALTH EFFECTS BIOASSAY
A clonal toxicity assay, employing the Chinese hamster V79 cell line
(13), was used to compare the relative acute toxicities of the effluents
from the biological reactors and the raw synthetic wastewater. This assay
measures the colony-forming ability of mammalian cells exposed to
toxicants. The purpose of this test was to evaluate the effectiveness of
biological treatment in alleviating potential human health effects
associated with coal conversion wastewaters. While direct extrapolation of
in vitro test results to in vivo conditions is difficult, cell culture
methods are considered valuable for assessing the relative toxicities of
environmental pollutants.
Effluent samples from the 10- and two 20-day reactors during the first
series of biological treatability studies described in Section 4 were
collected on September 17, 1978 and from the 5-day reactor on October 28
1978. The samples were centrifuged and then filtered through a series of
Nuclepore polycarbonate filters consisting of a 1.0 pm prefilter and a 0.2
Urn ultimate filter. The filtrates were collected and aliquoted in small
glass prescription bottles, which were then frozen and stored at -80°C.
122
-------
A. sample of the raw synthetic wastewater which had been aged for two days
was collected, treated, and stored in a similar manner. Individual
aliquots of frozen reactor effluents and raw feed were thawed immediately
prior to their use and the remainder of that aliquot was discarded at the
end of the day.
A series of dilutions of each wastewater was made in
distilled-deionized water. The addition of 2x or 4x nutrient medium to the
dilution tubes maintained physiological conditions at final test
concentrations ranging from 0.25 to 75% of the wastewater sample being
te8ted. Two hundred cells were seeded per 60 mm tissue culture dish and
allowed to incubate and attach for 3 hours in 3 ml of normal cell growth
medium. Duplicate iJ >he; were then treated with appropriate dilutions of a
i- wastewater. Each pair of dishes received a single concentration of
test naterials. After an exposure period of 20 hours, growth medium
containing the test materials was removed. The cells were washed once in a
phosphate buffered saline solution and re incubated in 3 ml of fresh growth
medium. Exposed single cells were allowed to grow into colonies and were
then fixed and stained after 7 days. The number of colonies for each
exposure condition was calculated as a percent of the number of colonies in
untreated control plates, and expressed as the relative plating efficiency.
The results of the clonal toxicity assay are shown in Figure 54, where
concentration-dependent survival curves havt- been plotted using the average
Of the data points from duplicate clonal toxicity experiments. The
concentrations indicated represent dilutions of the samples being tested.
Concentrations producing 50% lethality (LC^Q'S) are shown in Table 20
along with the corresponding TOC concentrations of the wastewater sample
123
-------
§
E
3
1
*
U
UJ
U
u.
LL
UJ
C3
z
<
0.
UJ
H
_)
UJ
cc
IUU
90
80
70
60
50
40
30
20
10
n
\ q
-
-
«
"
-
-
-
\
\
,
KEY:
Synthetic feed
5 - day reactor effluent
10 - day reactor effluent
1st 20 - day reactor effluent
2nd 20 - day reactor effluent
Figure 54.
0.5 - 1 2 5 10 20 30 40 50 60 70 80 90 95
WASTEWATER CONCENTRATION (%)
Results of 20-hour clonal toxicity assay using V-79 Chinese hamster cells.
98
-------
TABLE 20. SUMMARY OF MAMMALIAN CYTOTOXICITY DATA
Sample
Haw wastewater
5— day reactor effluent
IQ-day reactor effluent
First 20-day reactor effluent
Second 20-day reactor effluent
TOC
mg/1
1600
200
80
45
45
LC5Q
%
1.0
3.0
23.5
80 *
80 *
* The two 20-day reactors did not produce 50% lethality at the highest
concentrations tested (75%). The LCjjQ values shown are extrapolated from
the plots in Figure 54.
125
-------
under examination. It is apparent from Figure 54 and Table 20 that V79
cytotoxicity decreases with increasing degree of wastewater treatment as
measured by solids residence time.
It is interesting to note in Table 20 that while the 5-day reactor
provided an 87.5% (8-fold) reduction in TOC compared to the raw wastewater,
the LCcQ was reduced only three-fold. This suggests that most of the
easily-degradable TOC may not be very cytotoxic. On the other hand, the
95% (20-fold) TOC reduction produced by the 10-day reactor corresponded to
a 23-fold reduction in cytotoxicity, while the 97% (33-fold) reduction in
TOC produced by the 20-day reactors corresponded to an 80-fold reduction in
cytotoxicity compared to the raw wastewater. This suggests that TOC
removal, in itself, may not be a very useful indicator of toxicity
reduction.
A similar series of acute cytotoxicity tests is currently being
performed on the effluents from the second set of biodegradability studies
described in Section 5. Additionally, the raw and treated wastewaters are
being assayed for their mutagenicity potential, using the Ames test. The
results of these further, more comprehensive health effects assays will be
presented in the next report in this series.
126
-------
SECTION 8
VOLATILITY AND AIR-STRIPPING OF ORGANICS DURING BIOLOGICAL TREATMENT
An experiment was performed to provide a preliminary estimate of the
Quantity and nature of the organics lost by air-stripping in the biological
reactors. To simplify the data analysis, microorganisms were excluded by
using fresh unseeded synthetic wastewater. The combined effect of
biological activity and aeration on the production and potential loss of
volatile organics will be addressed at a later date.
PROCEDURE
One of the reactors was filled with 26 liters of freshly-prepared
1/4-strength synthetic coal conversion wastewater. In addition, a BOD
bottle was filled and stoppered to serve as a control. At no time was
there any indication that air had been trapped in the BOD bottle. The
control was stored alongside the reactor for the duration of the experiment
(26 hours).
Initially, at time zero, a sample was taken from the bottom port of the
y-eactor. Part of the sample was stored under refrigeration for future TOC
measurement. The remainder was taken immediately to the analytical
laboratory for HPLC analysis. Before the raw wastewater was one hour old,
aeration was started at an air flow rate of 1 liter/min, corresponding to
the air flow rate in the biological treatability studies. (The gas
127
-------
transfer properties of the reactor are characterized by an overall mass
_i
transfer coefficient for oxygen, k a, of 1.8 hr .) Samples were taken
]_»
from the lower port at 1, 7.5, and 26 hours of aeration. These were
handled in the same manner as the initial sample. In addition, the control
was also sampled at 26 hours. Due to minor operational problems with the
HPLC, analysis of all but the initial sample was delayed from several hours
to two days. A final sample was collected after 97 hours for TOC
analysis.
Total organic carbon was measured using the total carbon channel on a
Beckman 915 Carbon Analyzer. A 10 ml aliquot was acidified with 2 drops of
50% HC1 to decrease the pH below 4, and the sample was purged with nitrogen
for 4 minutes, to strip out any inorganic carbon prior to injection into
the TOC analyzer. The HPLC procedure was identical to that described in
Section 7.
RESULTS AND DISCUSSION
The results of the TOC analyses are listed in Table 21. It is clear
that no loss of volatile organic carbon resulted from aeration of the
wastewater under conditions parallel to those in the biological
treatability studies.
Table 22 lists the results of the HPLC analyses after various times of
aeration. The reference HPLC chromatogram used for this purpose is shown
in Figure 55. The chromatogram is similar to that presented in Section 7
for the raw wastewater, although some minor differences are apparent due to
ageing of the column. It is recommended that the HPLC results be
interpreted with some degree of caution since delays of a day or more
between sample collection and analysis might have significantly affected
128
-------
TABLE 21. CHANGE IN TOTAL ORGANIC CARBON CONCENTRATION
RESULTING FROM AERATION (mg/l)
Time
(hours)
0
1
26
97
TOC in Reactor TOC in Control
1,083
1,081
1,068 1,076
1,110
the integrity of the sample. It should be noted that the wastewater
darkened almost immediately after aeration was initiated, whereas the
sealed control remained relatively colorless. When the control was opened
after 26 hours for sampling, it also began to darken due to exposure to
air- Despite this change in the appearance of the sample, changes in the
traces are relatively minor. There is some variability associated
peak #3 (resorcinol and catechol) which cannot be explained, and
-everal of the minor peaks, primarily those of the relatively non-polar
compounds, disappear after 26 hrs. In general, however, it can be
concluded that there is little difference between the chemical composition
o{ the aerated sample and that of the control.
CONCLUSIONS
The volatility of the synthetic wastewater constituents is sufficiently
low that no significant loss in TOC can be attributed to aeration when only
the constituents of the raw wastewater are considered and the wastewater is
aerated under conditions likely to be encountered during biological
treatment. If any volatile organics are produced during biological
129
-------
TABLE 22. RESULTS OF KPLC ANALYSIS OF AERATED WASTEWATER
U>
O
Peak Ho.*
Constituent
HPLC
Seneitivity*
Concentration
in Raw
Waatevater
(«g/D
PEI
LK HEIGH
IT*
Aerated Sanptei
0 hr* .
1 hr** 1 hr**
7.5
hr.
26 hra
Control
26 hr.
1
3
4
5
6
7
g
9
10
11
12
13
U
1$
16
Benzole Acid
Reaorcinol
Catechol
Ana 1 ine
Phenol
5-Methylreaorcinol
4-Methylcetechol
o-Cre»ol
p-Cr*sol
2-Indanol
3.4-Xylenol
Pyridine
Quinoline
Indole
2,3-Xylenol
3,5-Xylenol
2-Naphthol
2-Methylquinoline
3-Ethylphenol
1-Naphthol
2.3,5-Trinethylphenol
Indene
Naphthalene
8.1
4.4
28
1.6
6.4
42
7.4
3.0
20
1.0
2.2
3.4
1.3
0.87
6
5
1.4
1.2
3.1
0.58
21
8.3
0.9
25
250
250
5
500
12.5
25
100
62.5
12.5
19 c
62.5
30
2.5
12.5
62.5
10
12.5
10
7.5
5
12.5
12.5
1.25
3.0
77.0
0
71.6
2.7
43.8
43.0
18.5
0
16.4
4.6
' 4.5
1.1
1.4
3.1
69.2***
0
70.0 69.2
2.6 2.9
42.5 42.4
42.2 42.0
22.0 19.5
0 0
14.2 14.8
3.3 3.2
4.2 4.0
0.9 0.8
1.1 1.1
0.2
114
0
69.2
2.1
42.8
38.0
•
17.5
0
5-0
4.0
2.0
1.2
0.9
0
91.3
0
67.2
1.9
42.0
32.0
12.5
0
2.0 •
4.0
0.6
*
0
0
0
94.0
0
68.9
2.4
42.8
36.0
15.5
0
8.8
5.2
0
1.2
1.4
10 •icroliter aanple, UV-abaorbanee at 280 n», valuea in 0.0005 abaorbance unita.
* aee Figure 55
£ HPLC aanaitivitjr in mg/l per 0.0005 abaorbance unita.
** A duplicate of the 1 hour ample waa «de aa peak* I, 3 and 4 vere inadvertently loat on the firit injection.
*** Peak #3 appeared with • (boulder here. Vile could indicate that two eloeeljr overlapping peak* created the erratic
reaalta.
-------
Ul
o
<
00
cc
<
D
HPLC ELUTION VOLUME
Figure 55.
HPLC chromatographic profile of synthetic wastewater for stripping evaluation.
correspond to compounds in Table 22.)
(Numbers
-------
treatment, however, it is conceivable that they may be stripped during
aeration. However, this remains to be proven under normal conditions of
aeration.
132
-------
SECTION 9
BIODEGRADABILITY OF COAL CONVERSION WASTEWATER CONSTITUENTS
The reactors used to evaluate the biological treatability of coal
conversion wastewaters are essentially microbial oxidative degradation
systems. A variety of microorganisms utilizing many different pathways are
involved in the biodegradation of the aromatic and aliphatic molecules in
the raw wastewater being fed to the reactors. Interpretation of the
results of reactor operation can be greatly aided by knowledge of the
biodegradation pathways used by the microorganisms for the various
individual components of the wastewater. Knowledge of biodegradation rates
and pathways can also provide insight as to metabolic products that might
accumulate and be present in the reactor effluent and to identify whether
or not certain substrates might act as metabolic inhibitors.
An initial screening analysis with respect to the biodegradability of
coal conversion wastewater constituents using biochemical oxygen demand
(BOD) techniques and an unacclimatized microbial seed was described in an
earlier report (1). That analysis revealed preliminary information about
the potential biodegradability of fifty-one constituents of coal conversion
vastewaters. In this section, the results of a more complete
biodegradability study, utilizing standard tnanometric techniques (14) and
an acclimated seed from the biotreatability reactors, are described.
133
-------
PROCEDURE
Two hundred ml of mixed liquor from one of the 20-day reactors was
centrifuged for about 15 minutes. The supernatant was decanted and the
solids were resuspended in 200 ml of 0.01M phosphate buffer. The
suspension of seed organisms was aerated for 15 to 30 minutes to allow
metabolism of any remaining wastewater constituents. A subsample of
approximately 50 ml was removed from the flask and autoclaved, to be used
ultimately as a sterile control.
Volume-calibrated manometers and Warburg flasks were used. Eleven ml
of aerated seed was pipetted into the reaction compartment of each Warburg
flask. Eleven ml of autoclaved sterilized seed was also added to one flask
of each compound as a control. Three ml of 10% KOH and one piece of
accordion-folded filter paper was added to the center well of each flask.
All substrate solutions were prepared in 0.01M phosphate buffer. This
concentration allowed 0.6 mg in the 1.0 ml added to the 11 ml seed, giving
a final concentration of 50 mg/1 in each flask. Each substrate was run in
duplicate or triplicate and each had a sterile seed control. For each
experiment, two flasks received no chemical addition, receiving instead 1.0
ml of phosphate buffer. These flasks comprised the controls for endogenous
0 uptake. A thermobarometer containing approximately 15 ml of buffer
was included in each run to correct for changes in ambient temperature and
pressure. The manometric apparatus was assembled and each flask was placed
so that it would be immersed in the 30 C water bath. Flasks vere allowed
to equilibrate for 20 minutes with the stopcock open. After equilibration
at time zero, the stopcock was closed, fluid in the manometer was adjusted
to the 250 mm level in the closed arm and the reading was recorded from the
134
-------
open arm. This procedure was repeated every ten minutes, thereby keeping
volume constant. After 10 minutes of incubation to allow pressures to
stabilize, the chemical contents (substrate) in each side arm were
carefully tipped into the reaction flask. The duration of the 0 uptake
measurements lasted from 3 to 5 hours. At the end of each experiment,
gubsamples of the reaction flasks were centrifuged to remove solids. These
samples were frozen and analyzed within several days by gas chromatography
to determine the concentration of substrate remaining. Initial samples of
each substrate solution were also analyzed.
CALCULATIONS AND RESULTS
Changes in the fluid level of the thermobarometer (TB) were substracted
fro.u net changes in each manometer to obtain TB-corrected net changes.
Cumulative sums of oxygen uptake were computed and multiplied by the
appropriate flask constant to obtain the amount of gas exchanged in
microliters, for the appropriate time intervals. Flask constants were
previously calculated from known parameters of temperature, pressure, fluid
volume, gas volume, and solubility (14). Net oxygen uptake was determined
by subtracting endogenous oxygen uptake from the gross uptake measured by
the respirometer. Replicates for each compound were averaged and converted
to Wmoles 0 uptake. From the analytical data, the amount of substrate
originally added to each flask was determined and converted to umoles.
•jtiis value was multiplied by a factor f, determined from the balanced
equation:
135
-------
substrate + f 0 -> CO + HO (g)
to obtain the umoles of 0 required for complete oxidation of the known
amount of substrate. The percent of the theoretical oxygen demand (THOD)
was computed by dividing the 0 utilized during the experiment by the
0 required for complete oxidation and multiplying by 100. On the graph
for each substrate (see Figures 56 through 64), gross umoles 0 uptake
due to substrate utilization, and net uptake (endogenous uptake subtracted
from gross uptake) were plotted.
In an effort to obtain information on how readily the substrates were
degraded, an initial rate of oxygen uptake was determined for each
compound. From the graphs, the total umoles of 0 utilized per hour
during the linear phase of oxygen uptake was divided by grains of mixed
liquor suspended solids in 11 ml of reactor contents to obtain these
initial rates shown in Table 23. Data on mixed liquor suspended solids
(MLSS) was obtained from reactor operating data. These values cannot be
directly compared as they depend to an extent on the initial amount of
substrate added which varied somewhat among different compounds. .Note that
Table 23 also includes sterile substrate concentrations or the amount of
substrate remaining in the reaction flask which contained autoclaved seed.
Theoretically, no biological oxidation should have occurred. Therefore,
any decrease from S initial to S sterile may be due to chemical oxidation
or adsorption of the chemical to solids. In a few cases, S sterile is
larger than S initial. The most likely explanation involves the difficulty
136
-------
Figure 56.
40 •
30 -
20 •
10 •
0
30 -
20 -
10 -I
pigure
Rate of oxygen utilization resulting from microbial degradation of
phenol. (Net oxygen uptake equals total oxygen uptake less endogenous
respiration.)
57.
Total 02 Uptake
Net 02 Uptake
50
100 150
TIME (Minutes)
200
250
300
50
100 150
TIME (Minutes)
200
Rate of oxygen utilization resulting from microbial degradation of
p-cresol.
137
-------
30
Total 02 Uptake
Net 02 Uptake
20
CM
o
o
10
150
200
Figure 58.
TIME (Minutes)
Rate of oxygen utilization resulting from microbial degradation of
o-cresol.
30
I 20 1
CM
O
o
£ 10
50
100
150
200
Figure 59.
TIME (Minutes)
Rate of oxygen utilization resulting from microbial degradation of
m-cresol.
138
-------
30 -
20 -
10 -
Total 02 Uptake
o e Net O2 Uptake
Figure 60.
30 -
20 •
O
*
a- 10 •
Figure 61,
250
300
TIME (Minutes)
Rate of oxygen utilization resulting from microbial degradation of
2,5-dimethylphenol.
50
100 150
TIME (Minutes)
200
250
300
Rate of oxygen utilization resulting from microbial degradation of
2,3-dimethyIphenol.
139
-------
Figure 62. Rate of oxygen utilization resulting from microbial degradation of
2,6-dimethylphenol.
30-
o>
.£
I
CM
O
o
20 -
30 -
w 20
O
J
o
10
-0 Total O2 Uptake
-e Net O2 Uptake
50
100 150
TIME (Minutes)
200
250
300
0 50 100 150
TIME (Minutes)
Figure 63. Rate of oxygen utilization resulting from microbial degradation of
3,5-dimethylphenol.
140
-------
30 •{
20 i
-------
TABLE 23. BIODEGRADATION OF SELECTED COAL GASIFICATION WASTEWATER COMPONENTS
Substrate
Phenol
p-Cresol
o-Cresol
n-Cresol
2 ,6-Dimethylphenol
2 ,5-Dimethylphenol
K) 2, 3-Dime thy 1 phenol
5 ,5-Diaiethylphenol
3 ,4-Dimethylphenol
Date MLSS
(1979) (rag/1)
3/28
3/7
2/23
3/7
3/28
3/17
3/17
4/6
4/27
617
766
794
766
617
615*
615*
643
701
Si
(mg/1)
41.7
29.1
38.2
50
50
50
50
46.7
50
sf
(mg/1)
0
1
0
3
25
29
1
21
0
.0
.5
.3
.5
.7
.7
.7
ssterile ™OD
(mg/1) (mg/1)
27
24
46
38
29
50
40
33
64
99.2
73.3
96.3
126.0
131.0
131.0
131.0
122.4
131.0
Initial Net Rate of
THOD 02 Uptake
(umoles) % THOD (umole 02/hr~8 MLSS)
37.2
27.5
36.1
47.3
49.1
49.1
49.1
45.9
49.1
52.4
56.4
48.5
43.3
1.0
11.2
42.8
5.4
45.8
1.
0.
0.
o.
0.
0.
0.
0.
0.
238
570
893
807
015
163
606
057
973
*Estimated
-------
of tipping the entire side arm contents into the reaction flask at the
beginning of the experiment. Some variability results.
DISCUSSION
In this phase of the investigation, essentially three groups of
molecules were examined. Phenol constitutes one group, the cresols, with
three different isomers, the second group, and five isomers of the xylenols
the third. The overall pattern of metabolism observed was consistent with
expectations based upon laboratory studies published in the literature
(15). Phenol was degraded extensively and at the highest rate, the cresols
as a 8r0uP being less biodegradable than phenol, but more degradable than
the xylenols. This follows the general pattern that has been observed in
laboratory studies of rates of biodegradation that total amounts of
biodegradation are less when increasing numbers of substituents are added
to aromatic ring systems.
The xylenols were metabolized to a lesser extent than either of the
other groups, but still showed significant amounts being degraded over a
five hour period. The pattern of metabolism for the different isomers of
xylenol reflects the usefulness of laboratory studies with pure cultures in
predicting the biodegradability of components in mixed systems with mixed
cultures. The two isomers of xylenol showing the largest amount of
metabolism were the 2,3 and 3,4 isomers. In both cases, the two methyl
groups are ortho to one another. The three other isomers in which the
methyl groups were either raeta or para to one another showed lesser amounts
o{ metabolism. This is most likely the result of steric interference with
143
-------
the second hydroxylation of the ring, which is necessary before ring
opening and further metabolisn can proceed.
Further studies of the microbiological degradation of specific
individual components in the coal gasification wastewaters are currently
underway using seed organisms from the reactors. These studies will
concentrate initially on the degradation of other components suspected to
be present in real coal gasification wastes. Subsequent studies will focus
on the degradation of mixtures of these compounds.
144
-------
SECTION 10
MIXED LIQUOR RESPIRATION STUDIES
Early in the project it was recognized that there were several
compounds in coal conversion wastewaters with potential for exerting toxic
effects on biological treatment systems. This raised significant questions
about practical treatability, and the wastewater concentrations at which
treatment systems might perform satisfactorily, as well as other
environmental factors influencing their operation. Accordingly, it was
deemed advisable to initiate screening studies suitable for short-term
evaluation of wastewaters to determine:
1. biodegradability of wastewater constituents under conditions as
close as possible to those in operating reactors, and
2. toxicity of the constituents to mixed cultures present in the
reactors, including approximate concentration levels at which such
toxic effects might occur.
Several techniques have been employed by past investigators (16-22) to
evaluate effects of wastewater constituents. The procedure selected for
use here was developed several years ago in the laboratories of the
University of North Carolina and subsequently has been applied by many
investigators under widely varying conditions to assess acute effects of
chemicals and wastewaters on wastewater treatment systems (16, 23, 24).
rather simple experimental setup is shown in Figure 65. Details of the
145
-------
BOD
BOTTLE
MAGNET
MAGNETIC MIXER
OXYGEN
ANALYZER
I
115v
115 v
RECORDER
115v
Figure 65. Experimental set-up for mixed liquor respiration studies,
146
-------
concepts and techniques have been described earlier (16), and may be
summarized here as follows:
(a) Samples of mixed liquor from a biological reactor are aerated for
several hours to ensure that the mixed culture approaches
"endogenous" respiration.
(b) A BOD bottle is filled with the mixed liquor, agitated
continuously to maintain homogeneous conditions, and the
endogenous respiration rate is measured using a dissolved oxygen
(DO) electrode.
(c) A known quantity of the wastewater or chemical being evaluated is
added to the sample and the new oxygen uptake rate is
determined. Increases in oxygen uptake rates indicate higher
levels of metabolic activity, usually because of biological
utilization of the added constituents. Decreases in respiration
rates indicate inhibitory effects on some of the organisms by
toxic constituents.
(d) This procedure is repeated with several aliquots of mixed liquor
at different concentrations of wastewater or constituents,
usually extending from concentrations well below those in the
actual wastewater to substantially higher levels.
Figure 66 shows the types of curves which have been observed commonly
with this technique, with respiration rates plotted as a function of added
substrate concentration. Curve 1 illustrates the situation which might be
observed when a chemical which is neither used by the culture nor
inhibitory to it is added to the respirometer. As the substrate
concentration is increased, the respiration rate does not depart
significantly from that of the endogenous sample, indicating no significant
^
effect on metabolism of organics in the system. Curves 2, 3, and 4
illustrate the types of relationships which have been observed for
chemicals having various characteristics with respect to utilization by the
organisms and toxic impact on their metabolic activities.
Figure 67 summarizes data obtained in this fashion for several
constituents of the synthetic wastewater. The level of each constituent is
147
-------
Figure 66. Common types of oxygen utilization curves.
oo
O
UJ
UJ
u.
O
Ul
I-
o:
z
O
Ul
oc
USED, NOT TOXIC
NOT USED, TOXIC
(2)
NOT USED, NOT TOXIC
SUBSTRATE CONCENTRATION
-------
1300
3,4 XYLENOL
\ACETIC ACID
^ rRESORCINOL
0-CRESOL-^ -~^
PYRIDINE
ENDOGENOUS
50 100 150 200
PERCENT OF CONCENTRATION IN WASTEWATER
Figure 67. Respiration curves using original oxygen utilization procedure.
-------
expressed as a percentage of its concentration in the full-strength
synthetic wastewater. Respiration rates are expressed as percent of the
"endogenous" rate for the sample of mixed liquor to which the constituent
was added.
Although the techniques used in these preliminary studies have proven
valuable in many earlier applications, some of the results raised concern
because effects of certain chemicals were not consistent with data reported
by other investigators (16, 23). For example, Figure 67 shows that the
respiration rate for phenol increased sharply with increased phenol
addition at low concentrations and continued to rise at phenol
concentrations up to 100% of that in the synthetic wastewater (2,000
mg/1). Even at 200% (4,000 mg/1), phenol did not inhibit biological
activity enough to depress respiration below the original endogenous rate.
This is inconsistent with observations by previous investigators who have
reported inhibitory action by phenol at concentrations at or above about
500 mg/1 (25). Another problem encountered was development of respiration
rate curves as a function of substrate concentration which were
inconsistent with past experiences in these types of analyses. Difficulty
also was frequently encountered with reproducibility of results, in general,
A series of tests was executed in which several aliquots of endogenous
mixed liquor were dosed with different concentrations of phenol and samples
removed from each periodically to determine respiration rate. At low
concentrations, the respiration rate for sludge which had been aged for
about 24 hours before adding phenol increased with time, as shown in Figure
68. Table 24 shows that "fresh" samples of mixed liquor were capable of
using phenol at substantially higher rates than the aerated sludge
150
-------
700
600
500
CO
O
UJ
O 400
O
U!
300
1
ff 200
100
200 mg/l Phenol
* 2/3/79
+ 2/5/79
ENDOGENOUS RATE
4 6
TIME (HOURS)
8
10
pigure 68. Respiration rate of aerated sludge with 200 mg/l phenol.
151
-------
TABLE 24. EFFECT OF AERATION ON "ACTIVITY" OF SLUDGE
RESPIRATION RATES, ug 02/l-min-g MLSS
Sludge Sample Initial Rate With 500 mg/1 Phenols
Aerated
Aerated
Aerated
"Fresh"
24
24
24
hours
hours
hours
(2/3/79)
(2/5/79)
(2/28/79
sludge from Reactor (4/11/79)
0.
0.
0.
0.
21
24
17
25
0.
0.
0.
0.
31
37
25
54
Ul
-------
immediately after phenol addition. These data suggest that organisms in
the system might have lost some of their ability to utilize the chemical
during the period of aeration. At elevated phenol concentrations, the
respiration rates were observed to decrease steadily over a period of
several hours, as illustrated in Figure 69, showing progressive inhibition
of biological activity by the phenol, at an initial phenol concentration of
2000 mg/1.
All of the difficulties led to the conclusion that experimental
procedures should be modified to yield results which are more amenable to
accurate interpretation, and to allow better reproducibility. The new
procedure consists of removing mixed liquor from the reactor and
determining its "initial" respiration rate immediately. Several checks
indicated that in reactors producing high levels of TOG removal,
respiration rates do not drop significantly over a period of 2 to 3 hours,
as illustrated by Figure 70. This indicates that relatively little excess
food exists in these reactors and that the organisms are in a metabolic
state approaching endogenous respiration. Whether truly endogenous or not,
the respiration rates are stable for extended periods, and can be
demonstrated to be reproducible. After determining the "initial"
respiration rate, aliquots of mixed liquor are dosed at different levels
with the chemical in question and oxygen uptake rates are determined
immediately and at several times thereafter. This allows assessment of the
effects of different chemical concentrations on sludge actually present in
the reactor, and evaluation of extended exposure to the chemical in
question, as contrasted with determining only the acute impact measured by
the original procedure.
153
-------
31.0
Ut
5
e»
CM
O
o
»—
X
LU
b
<
OC
O
uj 6.8
oc
ENDOGENOUS RATE
6
Figure 69.
TIME (HOURS)
Effect of time on respiration rate with 2000 mg/1 phenol.
-------
500
400
O
UJ
OC
200
UJ
O
OC
UJ
a.
100
50
100
150
TIME (MINUTES)
Figure 70. Respiration rate of "fresh" sludge from reactor,
-------
Figure 71 shows results obtained using the new procedure, with phenol
as the substrate. The curve labeled 4/7/79 shows the effect of phenol on
the respiration rate of "fresh" sludge immediately after phenol addition.
The upper curve for runs made on 4/11/79 agrees reasonably well with data
for corresponding concentrations on the earlier curve, illustrating
reproducibility.
The lower curves in Figure 71 on 4/11/79 indicate respiration rates of
the samples after 3, 5 and 30 hours of continuous aeration without feed.
The curve for time zero shows no toxicity even at 100% of the feed
concentration (2,000 mg/1). Data after several hours of aeration show
significant toxicity at 2,000 mg/1 phenol, the extent of inhibition
increasing with time. The points for 400 mg/1 show decreases in
respiration rates with time of aeration, which could indicate slight
inhibition or, perhaps, exhaustion of the phenol supply in the sample.
Clearly, the conclusion which one would reach about the effect of phenol is
radically different after several hours of aeration from that observed
immediately after its addition to the sample. Apparently, the inhibitory
action of phenol is exerted over a period of several hours.
Figure 72 shows results obtained using the modified procedure with
3,5-xylenol, 2-methylquinoline, acetic acid and the synthetic wastewater.
All of them indicate utilization by the mixed culture in the reactor and
none show immediate inhibition within concentration ranges employed in the
tests. It is especially significant that the composite synthetic
wastewater does not show any inhibitory effects, even at concentrations
twice as high as those of the synthetic formulation. However, it must be
recognized that inhibition may become evident subsequently, as shown in
156
-------
PHENOL CONCENTRATION ( mg/l)
1000 2000 3000
4000
c 300
uj 200
oc
Ln
S 100
UJ
U
OC
UJ
Q.
4/11/79
A 0 HOURS
HOURS
"A 5 HOURS
30 HOURS
4/7/79
INITIAL RATE
50 100 150 200
PERCENT OF CONCENTRATION IN WASTEWATER
Figure 71. Respiration curves for phenol using modified oxygen utilization procedure.
-------
0
500 f.
ui
K 400
<
2E 300
ft
ui
oc
S -i
00 < 200
H-
LU
u
oc
01
0.
100
FEED CONCENTRATION (% OF SYNTHETIC WASTEWATER )
50 100 150
200
I
3,5 XYLENOL
2 METHYLQUINOLINE
ACETIC ACID
"A
SYNTHETIC WASTEWATER
20 40 60
3.5 XYLENOL AND 2-METHYLQUINOLINE CONCENTRATIONS (mg/l)
ACETIC ACID CONCENTRATION (multiply by 10)
80
Figure 72. Results of using modified oxygen utilization procedure on selected substrates.
-------
Figure 71 for the curves developed on 4/11. This situation indicates a
need for checking the effect of substrate concentration at two or more
times, instead of only immediately after addition to the culture.
Evaluation of all chemicals in the synthetic wastewater are now being
repeated, using the modified procedure. Within the next few months, the
studies will be extended to evaluate the effects of other wastewater
constituents not presently included in the synthetic mixture, including:
cyanide, thiocyanate, ammonia, and selected priority pollutants. Also,
respiration measurements are being used periodically to monitor metabolic
activity in the experimental activated sludge reactors.
159
-------
SECTION 11
ADSORPTION OF ALKYL PHENOLS AND RESIDUAL TOC FOLLOWING
BIOLOGICAL TREATMENT
Adsorption by activated carbon or spent char from the coal conversion
process itself might be considered to be an integral part in the overall
scheme for treatment of coal conversion wastewaters. Adsorption processes
might be applied to the adsorbable organic constituents in the raw
waetewater, thereby lessening the organic load applied to the biological
processes. Alternatively, adsorption processes might be utilized in a
post-treatment mode, following biological treatment, in order to remove
residual organic substances which are non-biodegradable, or which are not
removed as a consequence of the conditions under which the biological
treatment units are operated. While there is a significant body of
knowledge concerning the adsorption of phenol by activated carbon, little
information is available concerning the relative adsorbability of the alkyl
phenols and of the other organic constituents of coal conversion
wastewaters. Accordingly, the objectives of this portion of the overall
investigation of coal conversion wastewater treatability and assessment
were :
1) to evaluate the effect of alkyl substitution on the adsorption of
phenols by activated carbon, with particular attention directed at the
position, length, and number of the alkyl substituents; and
160
-------
2) to evaluate the adsorbability of the residual TOC following
biological treatment.
PROCEDURES
A series of batch adsorption experiments was conducted to evaluate the
equilibrium characteristics for the adsorption of phenol and several alkyl
phenols on powdered activated carbon. The experiments were carried out on
a single component basis, using phenol, 2-, 3-, and 4-methylphenol
(cresols), several isomers of dimethyl- (xylenols) and ethylphenols, and
2-isopropylphenol. All chemicals were reagent-grade, from Aldrich Chemical
Company, and were used as received. PX-21, a powdered activated carbon
from Amoco Research Corporation, Naperville, Illinois, with a specific
2
surface area of 2,800 to 3,500 nv /gm, was used as the adsorbent.
Various dilutions, ranging from 20 to 60 mg/1, of stock phenolic
adsorbate solutions were addded to different amounts of activated carbon,
ranging from 540 to 1,000 mg/1. The suspensions were mixed for four hours,
at room temperature, on a Phipps and Bird six-place gang stirrer.
(preliminary experimentation assured that equilibrium was attained within
four hours.) Aliquots of the suspension were removed at the end of this
contact period and centrifuged to remove the powdered carbon. The UV
absorbance of the resulting concentrate was measured using a Varian
•jechtron Model 635 spectrophotometer, employing the UV absorbance maxima
listed in Table 25. Comparison of the absorbance measurements to a
characteristic calibration curve for the substrate in question allowed for
calculation of the equilibrium adsorbate concentration.
161
-------
TABLE 25. CHARACTERISTIC WAVELENGTHS FOR MAXIMUM UV ABSORBANCE OF
AQUEOUS PHENOLIC SOLUTIONS
. - - •• - — • - • - - —
Compound Wavelength, (nm)
Phenol 269
2-Methylphenol 270
3-Methylphenol 271
4-Methylphenol 276.5
2-Ethylphenol 270
3-Ethylphenol 270.2
4-Ethylphenol 274
2, 3-Dimethylphenol 271
3, 4-Dimethylphenol 274
2, 6-Dimethylphenol 268
2-Isopropylphenol 269.5
For the studies directed at the adsorption of the residual TOG from the
biological treatment units, Nuchar WV-G, a granular activated carbon
provided by the Westvaco Chemical Co., Covington, W. Va., was used as the
adsorbent. The carbon was obtained from the manufacturer in crushed form
and was washed twice with distilled water, dried at 200 C in an oven for
24 hours, and allowed to cool in a dessicator where it was subsequently
stored until it was used.
162
-------
Effluent samples from the biological reactors during the second series
of steady-state studies were collected, centrifuged, and filtered through
Gelman glass fiber filters (type A/E) with a pore size range of 0.2 to 10
jan to remove suspended material and microorganisms. Batch equilibrium
adsorption studies were conducted on the fresh filtrate, without any
dilution of the samples. A parallel sample of the fresh raw synthetic
wastewater w'as also included in the adsorption studies. Various amounts of
activated carbon, ranging from 100 to 5,000 mg/1 were added to the filtered
effluent samples, and the suspensions were mixed overnight, at room
temperature, on a Phipps and Bird six-place gang stirrer to assure that
equilibrium had been reached.
At the end of t,ir contact period, the suspensions were allowed to
settle for 24 hours while the beakers were covered with Parafilm sheets to
changes in concentration due to evaporation and volatilization.
of the supernatant liquid were filtered through a combination of
Whatman GF/D and GF/F glass fiber filter papers, with a maximum pore size
of 0.7 ym, in order to remove all fine carbon particles. The filtered
samples were stored under refrigeration and ultimately analyzed on a
Becktnan 915 Carbon Analyzer. Before injection into the TOC analyzer,
approximately 3 ml of each sample was acidified with 1 drop of 6N HCl to
convert inorganic carbon to CO which was removed by purging with
helium. Precipitation of a small amount of material was noted as a result
of the acidification step. A 30 ul aliquot of each of the acidified,
purged samples^was injected into the TOC analyzer and the residual TOC
concentration was determined.
163
-------
RESULTS
Phenol Adsorption
Figure 73A is a plot of the data for the adsorption of 2-, 3-, and
4-methylphenol on PX-21. The data conform to the Langmuir adsorption
isotherm
qe = (bQ°C)/(l+bC) (10)
where
q = quantity of substrate adsorbed per unit weight of adsorbent, in
gm/m
C = equilibrium concentration of substrate in solution, in mg/1
b = constant characterizing strength of substrate-adsorbent bond, in
(tug/I)'1
o
Q = adsorptive capacity in gm/gm
Figure 73B is a plot of the same data in accordance with the linearized
form of the Langmuir isotherm
C/qg - (1/Q°)C + l/bQ° (11)
yielding the coefficients b = 0.172 (mg/1) and Q = 0.383 gm/gm.
Both plots indicate that position of the substituted methyl group appears
to have no effect on the extent of methylphenol adsorption. Figures 74A
and 74B, and 75A and 75B, which show similar data for several isomers of
dimethylphenol and ethylphenol, respectively, also support this argument.
Composite plots of the adsorption isotherms for phenol, the methyl,
dimethyl, and ethylphenols, and 2-isopropylphenol are shown in Figures 76A
and 76B. The curves for the methyl, dimethyl , and ethylphenols are
developed from Figures 73, 74, and 75, respectively, while the curves for
phenol and 2-isopropylphenol are derived from the data illustrated in
164
-------
I
8
I
1
uj 0.6
O
o
UJ
to
tc
0.4
O
u
a.
t
0.2
10 20 30
METHYLPHENOL CONCENTRATION, C ( mg/l)
= 2-METHYLPHENOL
= 3-METHYLPHENOL
= 4-METHYLPHENOL
100
uj 50
O
O
0 10 20 30
METHYLPHENOL CONCENTRATION. C ( mg/l)
Figure 73. Adsorption of methylphenols by activated carbon. (A.) Uptake as
a function of methylphenol concentration. (B.) Linearized
Langmuirian relationship.
165
-------
0.6
0.4
0.2
1 °
2, 6 - DIMETHYLPHENOL
0 = 2, 3 - DIMETHYLPHENOL
A = 3, 4 - DIMETHYLPHENOL
10 20 30
DIMETHYLPHENOL CONCENTRATION, C (mg/l)
60
P 40
20
UJ
Figure 74.
0 10 20 30
DIMETHYLPHENOL CONCENTRATION, C (mg/l)
Adsorption of dimethylphenols by activated carbon. (A.) Uptake
as a function of dimethyIphenol concentration. (B.) Linearized
Langmuirian relationship.
166
-------
i
I
-ETHYLPHENOL
-ETHYLPHENOL
* = 4-ETHYLPHENOL
10 20 30
ETHYLPHENOL CONCENTRATION, C ( mg/l) /
U
0 10 20 30
ETHYLPHENOL CONCENTRATION, C ( mg/l)
Figure 75. Adsorption of ethylphenols by activated carbon. (A.) Uptake as a
function of ethylphenol concentration. (B.) Linearized Langmuirian
relationship.
167
-------
I
<3
O)
i
UJ
O
Q
UJ
m
a
O
0.6
0.4
0.2
10 20
CONCENTRATION, C ( mg/l)
30
5)
111
O
200
150
100
50
(1)- PHENOL
(2) - METHYLPHENOL
(3) - ETHYLPHENOL
(4) - DIMETHYLPHENOL
(5) - 2- ISOPROPYLPHENOL
10 20
CONCENTRATION, C ( mg/l)
30
Figure 76. Comparative adsorption of alkyl-substituted phenols by activated
carbon. (A.) Uptake as a function of alkyl phenol concentration
(B.) Linearized Langmuirian relationship.
168
-------
Figures 77 and 78, respectively. The alkyl-substituted phenols are more
strongly adsorbed than phenol itself, and adsorption increases as the
length of the alkyl chain increases. While position of the substituted
alkyl groups has no effect on the extent of adsorption, adsorption is
enhanced when the number of substituents on the phenol molecule is
increased. Table 26 summarizes the Langmuirian coefficients for the alkyl
phenols investigated, both on a weight and molar basis.
>
The experimental results are in accord with adsorption theory and with
the results of previous investigators who have reported that the extent of
adsorption increases as the solubility of the adsorbate decreases. The
substitution of alkyl groups on the phenol molecule would be expected to
make the resulting alkyl phenol less polar and therefore less soluble in
water.
Adscubabitity of Biologically-Treated Effluent
For a mixture of organic compounds, adsorption equilibria can usually
be described by the Freundlich isotherm
qe - *C1/n (12)
/
where q an^ C are, respectively, the quantity of organic material (e.g.
TOC) adsorbed per unit weight of adsorbent (activated carbon), and the
residual equilibrium concentration of organic material (TOC) in solution.
•Me" and "n" are constants. The Freundlich relationship can be rearranged
to give
log q = log k + 1/n log C (13)
which yields a linear plot when log q is plotted against log C.
Figure 79 presents a linearized Freundlich plot showing the
adsorbability of the TOC comprising the raw synthetic wastewater and the
169
-------
0.2
(0
O
i
0.1
UJ
Q.
10 20
PHENOL CONCENTRATION, C ( mg/l)
30
200
150
1
~ 100
UJ
a
50
10
20
30
Figure 77,
PHENOL CONCENTRATION, C (mg/l)
Adsorption of phenol by activated carbon. (A.) Uptake as a
function of phenol concentration. (B.) Linearized Langmuirian
relationship.
170
-------
I
<3
i
01
O
Q
LLl
00
DC
O
1
a.
a.
2
0.
0.6
0.4
0.2
10 20 30
2 - ISOPROPYLPHENOL CONCENTRATION, C { mg/l)
2 - ISOPROPYLPHENOL CONCENTRATION, C ( mg/l)
Figure 78. Adsorption of isopropylphenol by activated carbon. (A.) Uptake
as a function of isopropylphenol concentration. (B.) Linearized
Langmuirian relationship.
171
-------
TABLE 26. LANGMUIRIAN COEFFICIENTS FOR THE ADSORPTION OF ALKYL
PHENOLS ON PX-21 POWDERED ACTIVATED CARBON
b Q° b Q°
Substrate (rag/I)"1 gm/gm (ramole/l)"1 (mmole/gm carbon)
Phenol
2-Methylphenol
3 -Methyl phenol
0.0878
0.172
0.234
0.383
8.26
18.6
2.49
3.54
4-Methylphenol
2-Ethylphenol
3-Ethylphenol 0.222 0.532 27.1 4.35
4-Ethylphenol
2,3-Dimethylphenol
2,6-Dimethylphenol 0.248 0.552 30.3 4.52
3,4-Dimethylphenol
2-Isopropylphenol 0.245 0.671 33.6 4.89
biologically-treated effluent from the 5-, 20-, and 40-day reactors. At
/
this time, these results can only be considered to be preliminary but they
do suggest that the adsorbability of the residual organic constituents
following biological treatment decreases with increasing extent of
biological treatment. This does not imply that the residual organics
cannot be removed to a further degree by adsorption on activated carbon,
but it does indicate that the solid/solution equilibrium is shifted such
that a greater amount of adsorbent would be required to achieve a given
level of TOC removal.
These results are not surprising in view of the previous discussion in
Section 7 relating to the relative polarity of the residual organics
comprising the effluent TOC from the biological reactors. High performance
172
-------
U)
-0.4 r.
1
o -0.6
I
8
en
O
5
OC
UJ
O
1
8
-0.8
UJ
2 -1.0
o
UJ
CD
CC
Q -1.2
-1.4
-1.8
-1.8
1.4
X
a
X
A—
Raw Feed
* 5-Day Reactor Effluent
-~A 20-Day Reactor Effluent
~~° 40 - Day Reactor Effluent
I
1.6
2.6
1.8 2.0 2.2 2.4
LOG CONCENTRATION IN SOLUTION ( C in mg TOC/I)
Figure 79. Adsorption of raw and biologically-treated synthetic wastewater by activated carbon.
2.8
-------
Liquid chromatographic analyses demonstrated that biological treatment,
while substantially reducing the overall concentration of TOC, resulted in
the production of a sizeable amount of polar material, presumed to be
biological metabolites of the biochemical reactions taking place. The
increased polarity of the residual TOC would tend to make the organics more
soluble in water and therefore less adsorbable by a hydrophobic substance
like activated carbon.
These adsorbability studies are continuing in order to define better
the response of the residual TOC in the biologically-treated effluent to
adsorption by activated carbon and spent char.
174
-------
REFERENCES
1. Singer, P. C., F. K. Pfaender, J. Chinchilli, A. F. Maciorowski, J. C.
Lamb III, and R. Goodman. Assessment of Coal Conversion Wastewaters:
Characterization and Preliminary Biotreatability. Report No.
EPA-600/7-78-181, U. S. Environmental Protection Agency, Washington, DC
(September 1978).
2. Forney, A. J., W. P. Haynes, S. J. Gasior, G. E. Johnson, and J. P.
Strakey. Analysis of Tars, Chars, Gases and Water in Effluents from
the Synthane Process. U. S. Bureau of Mines Technical Progress Report
76. Pittsburgh Energy Research Center, Pittsburgh, PA (1974).
3. Luthy, R. G. and J. T. Tallon. Biological Treatment of Hygas Coal
Gasification Wastewater. FE-2496-43, U. S. Department of Energy,
Washington, DC (December 1978).
4. Johnson, G. E., R. D. Neufeld, C. J. Drummond, J. p. Strakey, W. P.
Haynes, J. D. Mack, and T. J. Valiknac. Treatability Studies of
Condensate Water from Synthane Coal Gasification. Report No.
FERC/RI-77/13, U. S. Department of Energy, Pittsburgh Energy Research
Center, Pittsburgh, PA (1977).
5. Reap, E. J., G. M. Davis, J. H. Koon, and M. J. Duffy. Wastewater
Characteristics and Treatment Technology for the Liquification of Coal
Using the H-Coal Process. Proceedings of 32nd Purdue industrial Waste
Conference, Ann Arbor Science Publishers, Ann Arbor, MI (1977), pp.
929-943.
6. Scott, C. D., C. W. Rancher, D. W. Holladay, and G. B. Dinsmore. A
Tapered Fluidized-Bed Bioreactor for Treatment of Aqueous Effluents
from Coal Conversion Processes. Proceedings of Second Symposium on
Environmental Aspects of Fuel Conversion Technology, EPA-600/2-76-149
(June 1976).
7. Kostenbader, P. D. and J. W. Flecksteiner. Biological Oxidation of
Coke Plant Weak Ammonia Liquor. J. Water Pollution Control Federation,
41 (2)-.199-207 (February 1969).
8. American Public Health Association. Standard Methods for the
Examination of Water and Wastewater, 14th ed. Washington, DC (1975).
9. Luthy, R. G. Manual of Methods: Preservation and Analysis of Coal
Gasification Wastewaters. U. S. Department of Energy, DOE FE-2496-8
(July 1977).
175
-------
10. O'Melia, C. R. Coagulation and Flocculation. Chapter 2 in
Physiochemical Processes for Water Quality Control (W. J. Weber, Jr.,
ed,). Wiley and Sons, New York (1972), pp. 61-109.
11. Metcalf and Eddy. Wastewater Engineering. McGraw Hill Book Co. (1972),
12. Drutnmond, C. J. , G. E. Johnson, R. D. Neufeld and W. P. Haynes.
Biochemical Oxidation of Coal Conversion Wastewaters. Presented at
87th National Meeting of Amrican Institute of Chemical Engineers,
Boston, MA (August 1979).
13. Duke, K. M., M. E. Davis, and A. J. Dennis. IERL-RTP Procedures
Manual: Level I Environmental Assessment. Biological Tests for Pilot
Studies, EPA-600/7-77-043, U. S. Environmental Protection Agency (April
1977).
14. Umbreit, W. W., R. H. Burris, and J. F. Stauffer. Manometric
Techniques. Burgess Publishing Co., Minneapolis, MN (1964).
15. Alexander, M. and B. K. Lustigman. Effect of Chemical Structure on
Microbial Degradation of Substituted Benzenes. J. Agric. Food Chem.,
14:410-413 (1966).
16. Lamb, J. C., W. C. Westgarth, J. L. Rogers, and A. P. Vernimmen. A
Technique for Evaluating the Biological Treatability of Industrial
Wastes. J. Water Pollution Control Federation, 36:1263-1284 (1964).
17. Jenkins, D. The Use of Manometric Methods in the Study of Sewage and
Trade Wastes. Page 99 in Waste Treatment. Pergamon Press, New York
(1960).
18. Gellman, I. and H. Heukelekian. Biological Oxidation of Formaldehyde.
Sewage and Industrial Wastes, 22(10):1321 (October 1950).
19. Skrinde, R. T. and C. N. Sawyer. Application of the Warburg
Respirometer to Industrial Waste Analysis - Special Studies on
Radioactive Wastes. Proceedings of the 7th Industrial Wastes
Conference, Purdue University Extension Service (1952), p. 481.
20. Dickerson, B. W., C. J. Campbell, and M. Stankard. Further Operating
Experiences on Biological Purification of Formaldehyde Wastes.
Proceedings of the 9th Industrial Wastes Conference, Purdue University
Extension Service (1954), p. 331.
21. Stack, V. T. and R. A. Conway. Design Data for Completely Mixed
Activated Sludge Treatment. Sewage and Industrial Wastes, 31(10):1181
(October 1959).
22. Wilson, I. S. The Monsanto Plant for the Treatment of Chemical
Wastes. J. and Proceedures, Inst. Sew. Purif. (Brit.), 86 (1954).
176
-------
23. Lamb, J. C. Reports on Industrial Waste Investigations. Unpublished
reports prepared for a chemical industry. (1962, 1963).
24. Matthews, L. A. Application of the Oxygen Electrode to Problems of
Waste Treatability. Proc. 13th S. Mun. and Ind. Waste Conf., Duke
University (in press).
25. Heukelekian, H. and M. C. Rand. Biochemical Oxygen Demand of Pure
Organic Compounds. Sewage and Industrial Wastes, 27:1040-1050 (1955)
177
-------
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before completing)
1. REPORT NO.
EPA-600/7-79-248
3. RECIPIENT'S ACCESSION NO.
4. TITLE AND SUBTITLE
Treatability and Assessment of Coal Conversion
Wastewaters: Phase I
5. REPORT DATE
November 1979
6. PERFORMING ORGANIZATION CODE
7 AUTHOR(S)
P.C.Singer, J.C.Lamb m, F.K. Pfaender, and
R. Goodman
8. PERFORMING ORGANIZATION REPORT NO~
9. PERFORMING ORGANIZATION NAME AND ADDRESS
University of North Carolina--Chapel Hill
Department of Environmental Sciences and
Engineering
Chapel Hill, North Carolina 27514
10. PROGRAM ELEMENT NO.
EHE623A
11. CONTRACT/GRANT NO.
Grant No. R804917
12. SPONSORING AGENCY NAME AND ADORES
EPA, Office of Research and Development
Industrial Environmental Research Laboratory
Research Triangle Park, NC 27711
14. SPONSORING AGENCY CODE
EPA/600/13
15. SUPPLEMENTARY NOT
541-2708.
ES IERL-RTP project officer is N. Dean Smith, Mail Drop 61 9197"
' '
16. ABS1
:T The report gives Phase I results of (1) an assessment of the environmental"
impact of wastewaters originating from the production of synthetic fuels from coal
and (2) an evaluation of alternative technologies for treating these wastewaters.
Work on coagulation, adsorption, and preliminary biological treatment studies is
continuing. Future reports, representing successive phases, will update these
results. The major focus is on aerobic biological treatment which is projected to be
the principal means of removing organic impurities from these wastewaters and a
cornerstone of any overall wastewater treatment program. A synthetic wastewater
designed to simulate a real conversion process wastewater, was fed to a series of '
aerobic biological reactors. Design and operation of the reactors is described,
along with performance data spanning two 6-month operating periods. In addition to
TOC, BOD, and COD data, the treated wastewaters were analyzed for phenolic con-
tent and residual organics, using chromatographic techniques. Aquatic bioassays
and mammalian cytotoxicity tests were performed on the raw and treated wastewa-
ters to evaluate their potential environmental impact.
7.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b.lDENTIFIERS/OPEN ENDED TERMS
CQSATt Field/Group
Pollution
Coal
Coal Gasification
Waste Water
Assessments
Water Treatment
Aerobic Processes
Organic Compounds
Bioassay
Toxicity
Cytology
Pollution Control
Stationary Sources
Coal Conversion
Synthetic Fuels
13B
08G,21D
131
14B
06C
06A
06T
Release to Public
Unclassified
PAGES
192
Unclassified
i (This page)
RICE
EPA farm 2220-1 (9-73)
178
------- |