TREATAB1LITY STUDIES
for the
INORGANIC CHEMICALS MANUFACTURING
POINT SOURCE CATEGORY
Prepared for
Effluent Guidelines Division
Office of Water and Waste Management
U.S. Environmental Protection Agency
Washington, D.C. 20460
Robert B. Schaffer, Director
Effluent Guidelines Division
G. Edward Stigall, Branch Chief
Inorganic Chemicals Branch
Elwood E. Martin
Project Officer
Contract No. 68-01-5767
July 1980
-------
NOTICE
This report has been reviewed by the Effluent Guidelines
Division, Office of Water and Waste Management, U.S.
Environmental Protection Agency, and approved for publication.
Approval does not signify that the contents necessarily reflect
the views and policy of the U.S. Environmental Protection
Agency, nor does mention of trade names or commercial products
constitute endorsement or recommendation for use.
-------
TABLE OP CONTENTS
LIST OF FIGURES
LIST OF TABLES
ACKNOWLEDGEMENTS
1.0 INTRODUCTION
1.1 OBJECTIVES
1.2 SCOPE OF TREATABILITY STUDIES
2.0 DESCRIPTION OF CONTROL TECHNOLOGIES
2.1 GENERAL CONSIDERATIONS
2.2 PRECIPITATION OF HYDROXIDES
2.3 PRECIPITATION OF SULFIDES
2.4 OXIDATION
2.5 CHROMATE REDUCTION
2.6 FLUORIDE PRECIPITATION
3.0
PROGRAM METHODOLOGY
3.1 DESCRIPTION OF TEST APPARATUS DESIGN
AND OPERATION
3.2 SAMPLING AND ANALYTICAL PROCEDURES
3.3 WASTE WATER CHARACTERIZATION
3.4 TREATMENT OPTIMIZATION
4.0 PROGRAM DEVELOPMENT
4.1 GENERAL CONSIDERATIONS
4.2 LOGISTICS CONSIDERATIONS
5.0 STATISTICAL ANALYSIS
5.1 OBJECTIVES OF STATISTICAL ANALYSIS OF
TREATABILITY DATA
5.2 ASSUMPTIONS CONCERNING MEASUREMENT OF
POLLUTANT CONCENTRATION LEVELS
5.3 STATISTICAL METHODOLOGY OF PERFORMANCE
ASSESSMENT
Page
vii
xi
xvii
1
1
2
5
5
7
8
8
9
11
11
12
15
15
19
19
20
21
21
21
22
iii
-------
TABLE OF CONTENTS - continued
6.0
5.4 ASSUMPTIONS CONCERNING 30-DAY AVERAGE
POLLUTANT LEVEL MEASUREMENTS
5.5 COMPUTATIONAL PROCEDURES
NICKEL SULFATE SUBCATEGORY
6.1 INTRODUCTION
6.1.1 General Considerations
6.1.2 Sample Point Location
6.2 TREATABILITY TEST MODEL OPERATION
6.2.1 Treatment Technology Tested
6.2.2 Waste Water Characterization
6.2.3 Details on Treatability Test
Operation
6.3 TEST RESULTS
6.3.1 Discussion of Results
6.3.2 Statistical Evaluation
6.3.3 Conclusions
7.0 HYDROFLUORIC ACID SUBCATEGORY
7.1 INTRODUCTION
7.1.1 General Considerations
7.1.2 Sample Point Location
7.2 TREATABILITY TEST MODEL OPERATION
7.2.1 Treatment Technology Tested
7.2.2 Waste Water Characterization
7.2.3 Details on Treatability Test
Operation
7.3 TEST RESULTS
7.3.1 Investigation of Anomalous Results
7.3.2 Discussion of Treatment Results
7.3.3 Statistical Evaluation
7.3.4 Conclusions
8.0 COPPER SULFATE SUBCATEGORY
8.1 INT RODUCTION
8.1.1 General Considerations
8.1.2 Sample Point Location
8.2 TREATABILITY TEST MODEL OPERATION
8.2.1 Treatment Technology Tested
8.2.2 Waste Water Characterization
8.2.3 Treatability Test Operation
8.3 TEST RESULTS
8.3.1 Discussion of Results
8.3.2 Statistical Evaluation
8.3.3 Conclusions
23
23
25
25
25
25
28
28
28
28
32
32
34
34
39
39
39
39
40
40
40
42
42
42
46
48
48
57
57
57
57
57
57
59
59
59
59
67
67
-------
TABLE OF CONTENTS - continued
9.0 CHLOR-ALKALI SUBCATEGORY (DIAPHRAGM CELL/
GRAPHITE ANODE)
9.1 INTRODUCTION
9.1.1 General Considerations
9.1.2 Sample Point Location
9.2 TREATABILITY TEST MODEL OPERATION
9.2.1 Treatment Technology Tested
9.2.2 Waste Water Characterization
9.2.3 Treatability Test Operation
9.3 TEST RESULTS
9.3.1 Discussion of Results
9.3.2 Statistical Evaluation
9.3.3 Conclusions
10.0 TITANIUM DIOXIDE SUBCATEGORY (CHLORIDE PROCESS)
10.1 INTRODUCTION
10.1.1 General Considerations
10.1.2 Sampling Point Locations
10.2 TREATABILITY TEST MODEL OPERATION
10.2.1 Treatment Technology Tested
10.2.2 Waste Water Characterization
10.2.3 Details on Treatability Test
Operation
10.3 TEST RESULTS
10.3.1 Discussion of Results
10.3.2 Statistical Evaluation
10.3.3 Conclusions
11.0 CHROME PIGMENTS SUBCATEGORY
11.1 INTRODUCTION
11.1.1 General Considerations
11.1.2 Sample Point Location
11.2 TREATABILITY TEST MODEL OPERATION
11.2.1 Treatment Technology Tested
11.2.2 Waste Water Characterization
11.2.3 Details on Treatability Test
Operation
11.3 TEST RESULTS
11.3.1 Discussion of Results
11.3.2 Statistical Evaluation
11.3.3 Conclusions
12.0 SODIUM DICHROMATE SUBCATEGORY
12.1 INTRODUCTION
Page
79
79
79
79
79
79
81
81
81
81
86
92
93
93
93
93
94
94
94
97
97
97
102
102
111
111
111
111
111
111
111
114
116
116
116
116
125
125
v
-------
TABLE OF CONTENTS - continued
12.1.1 General Considerations 125
12.1.2 Sample Point Location 125
12.2 TREATABILITY TEST MODEL OPERATION 125
12.2.1 Treatment Technology Tested 125
12.2.2 Waste Water Characterization 128
12.2.3 Details on Treatability Test
Operation 128
12.3 TEST RESULTS 132
12.3.1 Discussion of Results 132
12.3.2 Statistical Evaluation 134
12.3.3 Conclusions 134
13.0 SODIUM BISULFITE SUBCATEGORY 141
13.1 INTRODUCTION 141
13.1.1 General Considerations 141
13.1.2 Sample Point Location 141
13,2 TREATABILITY TEST MODEL OPERATION 141
13.2.1 Treatment Technology Tested 141
13.2.2 Waste Water Characterization 143
13.2.3 Details on Treatability Test
Operation 143
13.3 TEST RESULTS 143
13.3.1 Discussion of Results 143
13.3.2 Statistical Evaluation 155
13.3.3 Conclusions 155
14.0 SODIUM HYDROSULFITE SUBCATEGORY (FORMATE PROCESS) 161
14.1 INTRODUCTION . 161
14.1.1 General Considerations 161
14.1.2 Sample Point Location 161
14.2 TREATABILITY TEST MODEL OPERATION 161
14.2.1 Treatment Technology Tested 161
14.2.2 Waste Water Characterization 163
14.2.3 Details on Treatability Test
Operation 163
14.3 TEST RESULTS 163
14.3.1 Discussion of Results 163
14.3.2 Statistical Evaluation 178
14.3.3 Conclusions 185
APPENDIX A Statistical Summaries of Treatment Data A-l
APPENDIX B lodate Demand Curves - Sodium Bisulfite B-l
APPENDIX C lodate Demand Cruves - Sodium C-l
Hydrosulfite
VI
-------
LIST OP FIGURES
2-1
3-1
3-2
6-1
6-2
6-3
6-4
6-5
7-1
7-2
7-3
7-4
7-5
7-6
8-1
8-2
8-3
8-4
Comparative solubilities of metal hydroxides and
sulfide as a function of pH
Inorganic waste water-treatment test system
Aeration system
General waste water treatment process flow diagram
showing the sampling point at plant #369. (Nickel
sulfate manufacture.)
General waste water treatment process flow diagram
at plant #120 showing the sampling point. (Nickel
sulfate manufacture.)
Effluent nickel concentration as a function of pH
Estimated performance of proposed BAT treatment
Estimated performance of proposed BAT treatment
Paqe
6
13
14
26
27
33
36
37
Relationship of free and total fluoride concentration
in the raw and treated waste 47
Estimated performance of proposed BAT treatment 50
Estimated performance of proposed BAT treatment 51
Estimated performance of proposed BAT treatment 52
Estimated performance of proposed BAT treatment 53
Estimated performance of proposed BAT treatment 54
General process flow diagram at plant #034 showing
the sampling point. (Copper sulfate manufacture.) 58
Estimated performance of proposed BAT treatment 69
Estimated performance of proposed BAT treatment 70
Estimated performance of proposed BAT treatment 71
VII
-------
LIST OF FIGURES - continued
Page
8-5 Estimated performance of proposed BAT treatment 72
8-6 Estimated performance of proposed BAT treatment 74
8-7 Estimated performance of proposed BAT treatment 75
8-8 Estimated performance of proposed BAT treatment 76
8-9 Estimated performance of proposed BAT treatment 77
9-1 General process flow diagram at'plant #967 showing
the sampling point. (Chlorine/caustic, diaphragm
cell manufacture.) 80
9-2 Estimated performance of proposed BAT treatment 88
9-3 Estimated performance of proposed BAT treatment 89
9-4 Estimated performance of proposed BAT treatment 90
9-5 Estimated performance of proposed BAT treatment 91
10-1 Sources of waste samples for the titanium dioxide
subcategory (chloride-process) 95
10-2 Estimated performance of proposed BAT treatment 104
10-3 Estimated performance of proposed BAT treatment 105
10-4 Estimated performance of proposed BAT treatment 106
10-5 Estimated performance of proposed BAT treatment 107
10-6 Estimated performance of proposed BAT treatment 108
11-1 General waste water treatment process flow diagram
at plant #894 showing the sampling point. (Chrome
pigment manufacture.) 112
11-2 Estimated performance of proposed BAT treatment 120
11-3 Estimated performance of proposed BAT treatment 121
11-4 Estimated performance of proposed BAT treatment 122
11-5 Estimated performance of proposed BAT treatment 123
11-6 Estimated performance of proposed BAT treatment 124
viii
-------
LIST OF FIGURES - continued
Page
12-1
12-2
12-3
12-4
12-5
13-1
13-2
13-3
13-4
13-5
14-1
14-2
14-3
14-4
14-5
14-6
General waste water treatment process flow diagram
at plant |493 showing the sampling point. (Sodium
dichromate manufacture.)
136
137
138
139
Estimated performance of proposed BAT treatment
Estimated performance of proposed BAT treatment
Estimated performance of proposed BAT treatment
Estimated performance of proposed BAT treatment
General process flow diagram at plant #282 showing
the sampling points. (Sodium bisulfite manufacture.)142
Estimated performance of proposed BAT treatment 157
Estimated performance of proposed BAT treatment
Estimated performance of proposed BAT treatment
Estimated performance of proposed BAT treatment
General process flow diagram at plant #672 showing
the sampling points. (Sodium hydrosulfite
manufacture.)
Estimated performance of proposed BAT treatment
Estimated performance of proposed BAT treatment
Estimated performance of proposed BAT treatment
Estimated performance of proposed BAT treatment
Estimated performance of proposed BAT treatment
158
159
160
162
180
181
182
183
184
IX
-------
-------
LIST OF TABLES
Paqe
1-1 Inorganic Chemical Industry Subcategories
Evaluated by Treatability Tests
3-1 List of Approved Analytical Test Procedures Used
6-1 Waste Water Characterization for the Nickel Sulfate
Subcategory, Plant #369
6-2 Treatability Test Conditions and Analytical
Results
6-3 Treatability Test Conditions and Analytical
Results
6-4 Comparison Between Proposed BAT Limitations and
Estimated Treatability Performance for the Nickel
Sulfate Subcategory (Alkaline Treatment)
7-1 Waste Water Characterization for the Hydrofluoric
Acid Subcategory
7-2 Treatability Test Conditions
7-3 Analytical Results for the Plant Selected for
Study in the Hydrofluoric Acid Subcategory
7-4 Comparison Between Proposed BAT Limitations and
Estimated Treatability Performance for the
Hydrofluoric Acid Subcategory
8-1 Waste Water Characterization for the Plant Selected
for Study in the Copper Sulfate Subcategory
8-2 Effect of pH on Solubility of Pollutants
8-3 Treatability Test Conditions
8-4 Analytical Results for the Copper Sulfate
Subcategory
8-5 Treatability Test Conditions
. 3
16
29
30
31
35
41
43
44
49
60
60
61
62
64
XI
-------
LIST OP TABLES - continued
8-6
8-7
8-8
9-1
9-2
9-3
9-4
10-1
10-2
10-3
10-4
10-5
11-1
11-2
11-3
Analytical Results for the Copper Sulfate
Subcategory
Comparison Between Proposed BAT Limitations and
Estimated Treatability Performance for the
Copper Sulfate Subcategory (Lime Treatment)
Comparison Between Proposed BAT Limitations and
Estimated Treatability Performance for the
Copper Sulfate Subcategory (Caustic Treatment)
Waste Water Characterization for the Plant Selected
for Study in the Chlor-Alkali Subcategory, Diaphragm
Cell (Graphite Anode)
Treatability Test Conditions
Analytical Results for the Plant Selected for
Study in the Chlor-Alkali Subcategory
Comparison Between Proposed BAT Limitations and
Estimated Treatability Performance for the
Chlor-Alkali Subcategory
Waste Water Characterization for the Plant
Selected for Study in the Titanium Dioxide
Subcategory (Chloride Process)
Effect of pH on Toxic Metal Solubility
Treatability Test Conditions
Analytical Results for the Plant Selected for
Study in the Titanium Dioxide (Chloride Process)
Subcategory
Comparison Between Proposed BAT Limitations and
Estimated Treatability Performance for the
Titanium Dioxide (Chloride Process) Subcategory
Characterization of Raw Waste Water from the
Chrome Pigment Subcategory
Effects of Addition of Ferrous Sulfide to the
Chrome Pigments Waste Water
Treatability Test Conditions
Page
65
68
.73
82
83
84
87
96
98
99
100
103
113
113
115
Xll
-------
LIST OF TABLES - continued
Page
11-4 Analytical Results for the Plant Selected for
Study in the Chrome Pigments Subcategory
11-5 Comparison Between Proposed BAT Limitations and
Estimated Treatability Performance for the Chrome
Pigments Subcategory
12-1 Analyses of Treated Chromate Waste Water Solutions
after 10 days Reaction Time with Sodium Sulfide at
pH of Greater than 8.0
12-2 Characterization of Sodium Bichromate Waste Water
12-3 Characterization of Sodium Bichromate Batches
Used for the Test Runs
12-4 Characterization of Pickle Liquor
12-5 Treatability Test Conditions
12-6 Analytical Results for the Plant Selected for
Study in the Sodium.Bichromate Subcategory
12-7 Comparison Between Proposed BAT Limitations and
Estimated Treatability Performance for the
Sodium Bichromate Subcategory
13-1 Sodium Bisulfite Waste Water Characterization
13-2 Treatability Test Conditions
13-3 Analytical Results for the Plant Selected for
Study in the Sodium Bisulfite Subcategory
13-4 Sodium Bisulfite Subcategory Treatment by
Aeration. Batch 1
13-5 Sodium Bisulfite Subcategory Treatment by
Aeration. Batch 2
13-6 Sodium Bisulfite Subcategory Treatment by
Aeration. Batch 3
13-7 Sodium Bisulfite Subcategory Treatment by
Aeration. Batch 4
13-8 Sodium Bisulfite Subcategory Treatment by
Aeration. Batch 5
117
119
129
129
130
130
131
133
135
144
145
146
147
148
149
150
151
xxii
-------
LIST OF TABLES - continued
13-9
13-10
13-11
14-1
14-2
14-3
14-4
14-5
14-6
14-7
14-8
14-9
14-10
14-11
14-12
14-13
Sodium Bisulfite Subcategory Treatment by
Aeration. Batch 6
Sodium Bisulfite Subcategory Treatment by
Aeration. Batch 7
Comparison Between Proposed BAT Limitations and
Estimated Treatability Performance for the
Sodium Bisulfite Subcategory
Waste Water Characterization for the Sodium
Hydrosulfite Subcategory
Treatability Test Conditions
Analytical Results for the Plant Selected for
Study in the Sodium Hydrosulfite Subcategory
Sodium Hydrosulfite Subcategory Treatment by
Aeration. Batch 1
Sodium Hydrosulfite Subcategory Treatment by
Aeration. Batch 2
Sodium Hydrosulfite Subcategory Treatment by
Aeration. Batch 3
Sodium Hydrosulfite Subcategory Treatment by
Aeration. Batch 4
Sodium Hydrosulfite Subcategory Treatment by
Aeration. Batch 5
Sodium Hydrosulfite Subcategory Treatment by
Aeration. Batch 6
Sodium Hydrosulfite Subcategory Treatment by
Aeration. Batch 7
Sodium Hydrosulfite Subcategory Treatment by
Aeration. Batch 8
Sodium Hydrosulfite Subcategory Treatment by
Aeration. Batch 9
Sodium Hydrosulfite Subcategory Treatment by
Aeration. Batch 10
Page
152
153
156
164
165
166
168
169
170
171
172
173
174
175
176
177
xiv
-------
LIST OF TABLES - continued
14-14 Sodium Hydrosulfite Subcategory Treatment
by Aeration. Batch 11
14-15 Comparison Between Proposed BAT Limitations and
Estimated Treatability Performance for the
Sodium Hydrosulfite Subcategory (Formate Process)
Paqe
177
179
xv
-------
-------
ACKNOWLEDGEMENTS
The treatability study was conducted by Jacobs Environmental
Division of Jacobs Engineering Group Inc. of Pasadena,
California under the direction of Mr. Henry Cruse, Vice
President, and Mr. Michael Warner, Program Manager. Major
contributors were Mr. Santander Barros, Mr. Dale Newkirk, Mr.
Dev Srinivasan, Dr. David Ben Hur, Ms. Maureen Smith, and Dr.
Ben Edmondson. Dr. Richard Pomeroy of James M. Montgomery
Incorporated is gratefully acknowledged for his technical
assistance.
Field and analytical support was provided by Versar, Inc.
in Springfield, Virginia under the direction of Mr. Edwin
Abrams, Operations Manager, with the assistance of Mr. Edward
Rissmann and Mr. Ken Randolph.
The guidance and support provided by G. Edward Stigall,
Chief of the Inorganic Chemicals and Service Industries Branch
and Elwood E. Martin, Project Officer, is greatly appreciated.
xvi i
-------
-------
SECTION 1.0
INTRODUCTION
1.1 OBJECTIVES
The major purpose underlying this Treatability Study is to
evaluate the achievable performance of proposed Best Available
Technologies (BAT) for the treatment and control of pollutant
discharges, and to provide empirical treatment system performance
information applicable to selected inorganic chemical
subcategories. The study specifically concentrated on those
subcategories in the Inorganic Chemicals Industry for which
analytical data on raw waste waters and treated effluents either
do not exist or are deficient, and for which data are needed for
purposes of comparison with proposed effluent limitations
currently being proposed by the Effluent Guidelines Division of
the Environmental Protection Agency (1).
This study focuses on available treatment technologies which
have been selected as the basis for the proposed BAT regulations.
The majority of plants in the particular industries under study
practice the' BPT level of treatment in accordance with existing
or previous regulations. In most cases, the BAT level of
treatment can be achieved by adding onto the BPT systems a final
polishing step designed to remove additional toxic metals and
other pollutants of concern from the process waste streams.
Although BAT treatment is generally known and practiced to some
extent in the Inorganic Chemicals Industry, its application in
the particular product subcategories under study represents a
transfer of technology within the industry. For this reason, a
demonstration of applicability and an evaluation of performance
with regard to the treatment of these particular waste streams
are required in the regulation development process.
(1) U.S. Environmental Protection Agency. Development Document
for Proposed Effluent Limitations Guidelines, New Source
Performance Standards, and Pretreatment Standards for the
Inorganic Chemicals Manufacturing Point Source Category.
June 1980.
-------
1.2 SCOPE OF TREATABILITY STUDIES
Table 1-1 outlines, the inorganic chemical subcategories
studied, treatment technologies tested, and pollutants evaluated
in each subcategory.
-------
TABLE 1-1. INORGANIC CHEMICAL INDUSTRY SUBCATEGORIES
EVALUATED BY TREATABILITY TESTS
Subcategory Pollutants of Concern Treatments Tested
Hydrofluoric Acid/
Aluminum Fluoride
Chlor-Alkali
(Diaphragm Cell
with graphite
anodes)
Chrome Pigments
TSS, F, Ni, Zn, Cr(T) Lime plus filtration
TSS, Pbf Cr(T), Ni
Lime followed by iron
sulfide plus
filtration.
TSS, Cr(T), Pb, Zn, Cd Treatment by FeS to
precipitate residual
metals. (Cr had
already been removed
by reduction with S02
followed by precipi-
tation as Cr(OH)3.)
Sodium Dichromate
Nickel Sulfate
TSS, Cr(T), Ni
TSS, Ni
a)Simultaneous chro-
mate reduction and
precipitation by
Na2S followed by
alkaline precipita-
tion plus dual
media filtration.*
b)Chromate reduction
by ferrous chloride,
then precipitation
by lime; dual
media filtration.
a)Caustic soda plus
filtration.
b) Lime, plus
filtration.
*0riginally proposed treatment although modified to alternate b during
the treatability investigations.
-------
TABLE 1-1 - continued
Subcategory
Pollutants of Concern
Treatments Tested
Copper Sulfate
TSSf Cu, Nir Se
Sodium Bisulfite TSS, COD, Zn
Sodium Hydrosulfite TSSf COD, Cr(T), Zn
Titanium Dioxide TSS, CR(T), Fe, Ni,
(Chloride Process) Zn
a)Caustic soda plus
filtration.
b)Lime plus filtration,
Aeration.
Aeration
plus filtration.
Lime plus
filtration.
-------
SECTION 2.0
DESCRIPTION OF CONTROL TECHNOLOGIES
2.1 GENERAL CONSIDERATIONS
The treatment technologies applied in the treatability tests
consisted of well established unit processes. These processes
include removal of heavy metals by precipitation as hydroxides
and/or sulfides, reduction of chromate, precipitation of
fluorides with lime, separation of solids by settling and dual
media filtration, and oxidation by aeration. The primary
chemical unit processes are discussed below.
2.2 PRECIPITATION OF HYDROXIDES
The most widely used treatment technology for the removal of
most metals from chemical wastes is their precipitation as
hydroxides. Caustic soda (NaOH) and slaked lime (Ca(OH)2) are
the two precipitants most commonly used for this purpose.
The amount of a metal ion, as for example, Zn++, that can
remain in solution diminishes with increasing pH, but a point is
reached at which the metal can be in solution as a negative ion,
as for example, Zn02=. The amount of this ion that can be in
solution increases with increase of pH. Thus, there is, for any
metal showing this "amphoteric" character, a pH at which the
solubility is minimal. This is often called the isoelectric
point. Figure 2-1 shows this behavior of the hydroxides. The
figure is illustrative and relative, and must not be considered
as a representation of quantitative information.
The normal isoelectric pH of a metal, is not always the
optimal point for its removal, because other components,
especially complexing agents, may have major effects. For any
particular type of waste water, the optimal pH needs to be found
by trial.
-------
NOTE:
10
10
1 i 1 1 1 1 1
01 2 3 45 67 8 9 10 11 12 13 14
10
10
,-12
pH
1. Solubilities for metal hydroxides are taken from curves by Freedman and
Shannon, "Modern AUcaline Cooling Water Treatment," Industrial Water
Engineering, Page 31, (Jan./Feb. 1973).
2. Plotted data for metal sulfides based on experimental data listed in
Seidell's solubilities.
Figure 2-1. Comparative solubilities of metal Hydroxides and sulfide
as a function of pH.
-------
2.3 PRECIPITATION OF SULFIDES
As shown in Figure 2-1, the solubilities of the sulfides of
most of the common metals are orders of magnitude lower than the
solubilities of the hydroxides. Therefore, the residual
concentrations of one of these metals in a. waste water after
treatment by a soluble sulfide is expected to be lower than after
precipitation by a hydroxide. This would make it appear that
sulfide precipitation should be the method of choice. This is
not necessarily the case. Attention must be given to the
amenability of the precipitate to removal, economic factors, air
pollution with hydrogen sulfide (H2S), byproduct disposal, and
the actual environmental significance of the traces of metals
that may remain.
Sulfide precipitation is usually accomplished by addition of
Na2S or NaHS. Either compound, when dissolved in water, gives a
solution in which the sulfur is largely in the form of HS-. The
Na2S hydrolyses to give HS- and OH-, and thus produces a more
alkaline solution, but there is very little S= unless the pH is
above 12. The precipitation of a divalent metal is essentially
as follows:
M
+ HS
•» MS + H
Sometimes, precipitation is accomplished by adding freshly
prepared ferrous sulfide. An exchange reaction (metathesis)
occurs, as follows:
From the relative solubilities of the hydroxides and the
sulfides, it would appear that the residual metal concentration
in a waste water would be much lower if sodium sulfide is used as
the precipitant, but this is not necessarily so, because the
amenability of the precipi'tate to separation by settling and
filtration is an important factor. The sulfide precipitates have
less tendency to flocculate and a greater inclination to produce
colloidal precipitates than do the hydroxides.
There is a disinclination in industry to use the sulfide
method because of the possibility of the mixing of the sulfide
solution with an acid solution and the consequent release of H2S,
a highly toxic gas that has claimed many lives in industrial
-------
accidents. The hazard is greater when using NaHS or Na2S, but it
can also happen with FeS.
Regardless of how the metals are precipitated, they will be
removed as a sludge that must be disposed of in an
environmentally acceptable manner.
2.4 OXIDATION
Chemical oxidation is another technique commonly used for
the destruction of pollutants in a waste. The cheapest oxidant
is air. Oxygen may be dissolved in the waste water by bubbling
air through it, or by bringing the waste water and air into
contact in some other way. Sometimes pure oxygen is used, and
sometimes oxidation has been accomplished by the use of chlorine
or hydrogen peroxide. Often, biological growths are instrumental
in bringing about the reaction between oxygen and the substance
(reducing agent) with which it reacts. The biological oxidation
process is difficult to utilize on inorganic chemical industry
wastes without addition of substantial quantities of dilution
water and/or nutrient supplements. In addition, many of the
pollutants found are toxic to biological growths. In the
chemical subcategories studied herewith, only the sodium
hydrosulfite industry currently employs biochemical oxidation of
oxygen demand.
Oxidation by air was tested for treating waste waters from
the sodium bisulfite and sodium hydrosulfite subcategories. The
treatment was aimed at the reduction of the inorganic
constituents which contribute to the high COD. The rationale of
using air oxidation was that most of the COD present in the waste
was believed to be due to sulfite and hydrosulfite, which will
react with oxygen. It was found, however, that various other
forms of sulfur are present and that elemental sulfur may be
present or may be produced when the waste is acidified. Attempts
to measure COD in such samples give highly erratic results.
2.5 CHROMATE REDUCTION
The reduction of chromium from the hexavalent form to the
trivalent form is essential, since hexavalent chromium cannot be
removed by alkaline precipitation (unlike the trivalent form).
It should be noted that chromium sulfide does not exist in
aqueous systems as it is readily oxidized to the hydroxide.
Therefore, sulfide precipitation is not a viable technology for
chromium removal.
Chromate is reduced in the chrome pigments industry by
introduction of sulfur dioxide into the raw waste under acidic
-------
conditions. Chromate may also be reduced using ferrous chloride
as is done in the sodium dichromate industry.
2.6 FLUORIDE PRECIPITATION
The conventional method of treating fluoride bearing wastes
is to precipitate the fluoride as calcium fluoride by the
addition of lime. The reaction is:
Ca(OH)2 + 2F = CaF2 + 20H
Using this process alone, it is difficult to remove free
fluoride to concentrations less than 8 mg/1 under ideal
conditions due to the solubility of calcium fluoride.
-------
-------
SECTION 3.0
PROGRAM METHODOLOGY
3.1 DESCRIPTION OF TEST APPARATUS DESIGN AND OPERATION
In order to study the variety of BAT treatment concepts
under evaluation, pilot scale test equipment was designed to
incorporate the following operations:
2.
3.
4,
5,
Chemical reactions, including
a. pH adjustment
b. reduction
c. oxidation by air
d. precipitation
e. stirring
Sedimentation
Decantation
Filtration
Sludge removal
The various chemical treatment reactions were performed in a
30 gallon linear polyethylene tank with conical bottom and
equipped with a stirrer. The conical bottom was designed to
assist in sludge removal through a bottom valve. The amount of
waste water treated was usually between 20 to 25 gallons.
In cases where treatment required addition of caustic soda
or lime, the stirrer was started and the chemical was added until
a selected pH was reached. Most other chemicals were added in
predetermined amounts.
When aeration was required, air from a diaphragm compressor
was fed to six porous media diffusers placed in the bottom of the
11
-------
tank. The mixer was operated
the dispersion of the air.
during aeration runs, to improve
If the treatment included precipitation and clarification by
settling, the supernatant was decanted by the use of a hose
serving as a siphon, and the sludge was then withdrawn through
the bottom of the tank. After the supernatant was returned to
the tank, it was pumped to the filter.
The filter consisted of a 4-inch ID PVC column 40 inches
high, with removable flange plates at each end. A stainless
steel screen, 30 mesh with 0.01 inch diameter wire, was inserted
into the column about six inches from the bottom, for support of
the media. The bottom 9 inches of the media was silica sand with
an effective size between 0.4 and 0.66 mm, supporting 18 inches
of anthracite coal with an effective size between 0.7 and 1.7 mm
and a uniformity coefficient of 1.7. The filter media was
normally replaced before each test and it was washed with water
for about 20 minutes before being used; therefore, no
backwashing was practiced.
The flow to the filter was controlled by a needle valve and
monitored by use of a rotameter. The rate in all cases was kept
at 0.25 gpra, equivalent to a filtration rate of 3.1 gpm/ft2. By
use of a throttling valve at the bottom, the liquid level over
the filter media was usually kept between 1.0 and 1.5 inches.
The flange closing the top of the filter permitted it to be
operated under pressure.
Usually the filtrate was recirculated to the reaction tank
for a period of generally 30 to 40 minutes to increase the
efficiency of the filtration operation. Then the entire batch
was filtered and the filtrate was sampled. Figures 3-1 and 3-2
illustrate the apparatus.
3.2 SAMPLING AND ANALYTICAL PROCEDURES
Samples were collected from the raw waste water, clarified
supernatant, and the' filtrate for each of the treatability tests.
All the samples were split into three for analysis. One of the
portions was unpreserved and utilized for the TSS determination.
The second portion was filtered through a Whatman Filter Paper
No. 40. All constituents of the filtered waste water were
described as "dissolved". The filtrate was preserved by adding
HN03 and was subsequently used for the determination of dissolved
metals. The third portion was not filtered, but was also
preserved with HN03. This sample was subjected to "total" metals
determinations. All samples were collected in one liter plastic
containers. When the treatability tests were carried out in the
field, the sample bottles were refrigerated with ice and sent by
12
-------
R
en
4J
w
n
4J
$
s
u
S;
rH
^
S!
13
-------
AIR. Jf
A»R POROUS
DIFFUSER
"
AIRFLOW
COMTROU
Figiice 3-2. Aeration sy-
14
-------
overnight delivery to the analytical laboratories. The samples
were always kept refrigerated following receipt until they were
analyzed. Refrigeration of the samples prior to the analytical
determinations was also practiced when the test runs were
performed in the Pasadena and Springfield laboratories. The
analytical determinations were done following methods approved by
EPA as indicated in Table 3-1.
3.3 WASTE WATER CHARACTERIZATION
Waste water samples, usually 5 gallons, were collected for
characterization from each subcategory prior to the test. runs.
Pollutant concentrations were determined, and in some cases, more
complete analyses were made so as to provide a better
understanding of the chemistry of the waste waters. The
analytical methods used were EPA approved methods.
3.4 TREATMENT OPTIMIZATION
Characterization of the waste provided the basis for
optimization of the treatability tests in each subcategory. In
each series of treatability tests involving precipitation of
heavy metals by alkaline treatment, the initial raw waste sample
was used to determine the pH for optimum metal removal. .This was
accomplished by running a series of beaker tests at approximately
pH intervals of 0.5 covering the theoretical optimum
precipitation range for the pollutants under consideration. _The
concentration of the metals remaining in solution was.determined
at each pH value and on the basis of the test results, an optimum
pH was selected.
15
-------
TABLE 3-1. LIST OF'APPROVED ANALYTICAL TEST PROCEDURES USED
:===========:
Parameter
:=======:=:====:=:=:====:
Method
Acidity as CaC03
Alkalinity as CaC03
Methyl Orange Acidity
as CaC03
Phenolphthalein
Alkalinity as CaC03
Carbonate
Total Suspended Solids
Total Dissolved Solids
Total Residue
Fixed Residue
Total Nitrogen
(Kjeldahl)
Sulfate
Nitrate
Phosphate
lodate demand
Chloride
Fluoride (free)
Phenolphthalein end point
Methyl orange end point
Methyl orange end point
Phenolphthalein end point
Calculation based on alkalinity
Gravimetric GFC filtration, 104 degrees C
Gravimetric GFC filtration, 180 degrees C
Gravimetric, 104 degrees C
Gravimetric, 550 degrees C
Digestion and distillation followed
by titration or nesslerization
Gravimetric, BaS04 precipitation
Brucine method
Vanadomolybdophosphoric acid
lodide-iodate titration*
Mercuric nitrate method
Ion electrode
*The iodide-iodate acid titration is not specific for any one
compound, but it reacts with any sulfite, hydrosulfide, thiosulfate
and possibly other substances. When one of these substances is
dominant, the titration is treated as a measure of that substance.
16
-------
TABLE 3-1 - continued
===:=== = = = ====: r= = = = = ==s:
Parameter
Method
Fluoride (total)
Total Hardness as CaCOS
Calcium as Ca
Magnesium as Mg
Chemical Oxygen Demand
PH
Nickel
Zinc
Chromium (Total)
Lead
Cadmium
Selenium
Iron
Sodium
Potassium
Copper
Aluminum
.Chromium (hexavalent)
Distillation followed by ion electrode
EDTA titration
Digestion followed by atomic absorption
Digestion followed by atomic absorption
Dichromate reflux
Ion electrode
Digestion followed by atomic absorption
Digestion followed by atomic absorption
Digestion followed by atomic absorption
Digestion followed by atomic absorption
Digestion followed by atomic absorption
Digestion followed by atomic absorption
Digestion followed by atomic absorption
Digestion followed by atomic absorption
Digestion followed by atomic absorption
Digestion followed by atomic absorption
Digestion followed by atomic absorption
Colorimetric, Diphenlycarbazide method
17
-------
-------
SECTION 4.0
PROGRAM DEVELOPMENT
4.1 GENERAL CONSIDERATIONS
Preparations to conduct the treatability studies started in
July 1979. Jacobs Engineering Group Inc. as part of Contract
No. 68-01-5767 undertook the task of performing treatability
tests on nine industrial subcategories in which the achievability
of the proposed effluent limitations using BAT needed to^be
demonstrated. Jacobs developed a comprehensive work plan which
included, for each subcategory, the following:
1. Selection of an industrial plant from each of the nine
chemical subcategories under consideration. Whenever possible,
this selection was based on considerations as to how
representative the plant to be studied would be for the entire
subcategory being considered.
2. Development of the treatability tests to be performed on
each subcategory.
3. Review of the process flow diagrams for each of the
plants selected in order to determine the most representative
point for collection of the waste water to be used in the
treatability tests.
4. Contacts with the appropriate plant personnel. This was
essential and necessary in order to explain to them the purpose
of the treatability studies and obtain their collaboration and
permission to collect the required waste water samples.
5. Concurrently with the implementation of steps 1 through
4, work was started on the design, selection, and procurement of
equipment, and assembly of treatability units.
6. Negotiations with Versar, Inc. for implementation of
portions of the study.
19
-------
4.2 LOGISTICS CONSIDERATIONS
Most of the plants selected for study were located on the
East Coast and in the South. Since Versar Inc. is centrally
located with respect to plants in the East Coast and in the
South, the treatability work was divided on a geographical basis
to take full advantage of Versar's laboratory and mobile units.
Jacobs concentrated on plants located on the West Coast.
20
-------
SECTION 5.0
STATISTICAL ANALYSIS
5.1 OBJECTIVES OF STATISTICAL ANALYSIS OF TREATABILITY DATA
As part of this treatability study, selected inorganic
chemicals manufacturing plants employing various available
treatment technologies, and thought to be representative of the
inorganic chemicals industry, were chosen for study. Effluent
samples were drawn from appropriate points in raw process waste
water and treated waste water streams. The concentration levels
(mg/1) of chemical pollutant parameters in these samples were
determined by laboratory measurements and the resultant data
presented for statistical analysis. The main objective of the
analysis was the demonstration of achievable performance levels
for BAT treatment technologies presently under consideration for
effluent limitations guidelines in the Inorganic Chemicals
Manufacturing Point Source Category.
A demonstration of this nature requires certain assumptions
regarding the applicable statistical or probabilistic models
used. These assumptions are outlined in the following sections.
5.2 ASSUMPTIONS
CONCENTRATION LEVELS
CONCERNING
MEASUREMENT
OF
POLLUTANT
In the formulation . and calculation of statistical
characteristics of treatability performance from laboratory
measurements, individual sample measurements of pollutant
concentration levels were assumed to follow the lognormal
distribution, a well known and generally accepted statistical
probability model used in pollution analysis, and which is
appropriate for measurements of levels taken on a daily basis.
This assumption is equivalent to the assertion that the
logarithms of individual measurements follow a normal probability
model. It was also assumed that sampling at a given plant was
conducted responsibly, and in such a way that the resulting
measurements can be considered statistically independent and,
21
-------
therefore, amenable to standard statistical
procedures.
estimation
5.3 STATISTICAL METHODOLOGY OF PERFORMANCE ASSESSMENT
To measure treatability level for a given process employing
a certain treatment technology, the "probability performance" was
calculated for each parameter in the treated effluent stream.
This quantity is the probabilty that a given stream will have a
pollutant level (mg/1) that is within a stipulated effluent
guideline limitation level. This amounts to the probability that
a plant employing the proposed BAT effluent treatment will
produce a 30-day average concentration less than the
corresponding limitation level. Where a pollutant discharge
level is measured by the average concentration in thirty
individual daily measurements (the 30-day average), the
probability performance represents the proportion or fraction of
the 30-day averages that will be less than or equal to the given
limitation level.
For this study, the maximum likelihood estimation procedure,
based upon a sequence of "runs" of treatments, was used to
estimate the probability performance. This method chooses as the
estimated probability performance that value which is most likely
to have produced the observed sample consistent with the model
chosen. Standard maximum likelihood procedures are available and
widely known, and are the methods on which the statistical
estimation in this study are based.
The data included in the statistical analysis were
successively screened for outliers through use of the t-statistic
and on the basis of technical considerations. For each
subcategory studied, the statistical results are presented in
terms of both the screened and unscreened data. The t-statistic
is defined by the equation:
max ((x - X)/S, (X - X )/S)
max min
Where: xmax is the datum corresponding to the greatest
parameter level in a particular run, and xmin
to the smallest.
X is the sample average over all repetitions of runs,
s is the sample standard deviation of x.
22
-------
For those cases where a treatability sample had a maximum
value or minimum value sufficiently large or small so as to
produce a t-statistic exceeding the 99 percent confidence limits
from the t-distribution, that datum was rejected as an outlier
and treatability statistics were recomputed on the screened
sample.
5.4 ASSUMPTIONS CONCERNING 30-DAY
MEASUREMENTS
AVERAGE POLLUTANT LEVEL
Even though individual pollution concentration measurements
are assumed lognormally distributed, that assumption does not
extend to the statistical behavior of averages, in this case
where 30-day averages are to be used. However, if averages are
taken over a "reasonably large" number of days, a statistical
principle, the Central Limit theorem, assures that probabilities
pertaining to such averages may be computed using _ the normal
probability distribution. A 30-day average contains enough
individual daily measurements to insure that the ?o™al
probability model is satisfactory for use in final calculation ot
probability performance.
5.5 COMPUTATIONAL PROCEDURES
To compute the maximum likelihood estimates (MLE's) of
probability performance, it is necessary to compute the MLE s of
the long term average pollution level and the long term_ standard
deviation of pollution level using logarithms of the individual
measurements of data for each run. These estimates, the maximum
likelihood estimates of the mean logarithm and standard deviation
of logarithms are done using standard statistical formulae.
These estimates are computed and used to obtain the MLE's of
the long term average, A, and long term standard deviation, S, of
the pollutant concentration. Dividing the estimated long term
standard deviation, S, by the square root of the number of days
(30) in the average, gives S*, the estimated standard error_of
the 30-day average, and which can be used in computing
probability performance. For any limitition value, L, tne
estimated probability performance, p, is computed as:
Probability(30-day average does not exceed L) = Pr(z)
where z = (L-A)/S*
and Pr(z) = probability that a standardized normal
value does not exceed z.
23
-------
Standard statistical measures of pollution level in treated
effluent for each subcategory are tabulated and recorded in
Appendix A. These include minimum (Min) arithmetic sample
average (Avg), maximum (Max) standard deviation (Stdv), and
coefficient of variation (C.Var).
Included with the technical analysis of each subcategory are
the probability performance curves and estimates of long term
average for each parameter. Note that the estimate of long term
pollution average is obtained by maximum likelihood methods from
the lognormal distribution and does not necessarily equal the
sample arithmetic average given in Appendix A.
Where a parameter curve is quite "steep", i.e., for those
parameters that show a sharp increase in probability (of a 30-day
average not exceeding a given value) for a small increase in
pollution concentration, this is primarily due to a relatively
small standard deviation. In other words, the steepness of the
curve relates to the degree of consistency in the sample results.
24
-------
SECTION 6.0
NICKEL SULFATE SUBCATEGORY
6.1 INTRODUCTION
6.1.1 General Considerations
To test the BAT concept as proposed for this subcategory, a
treatability model unit was set up at PJB Laboratories, a
division of Jacobs, in Pasadena, California. The tests were
carried out from September 4 to November 12, 1979, and a total of
24 runs were completed in this period.
Samples were collected from Plant #369 which operated on a
non-daily, batch type basis, so waste water was not always
available. In this case, the treatability team running the unit
was advised in advance by plant personnel when to be ready for
sample collection. It became evident, however, that it would not
be possible to obtain enough samples from this plant alone to be
statistically significant in the time span available to perform
the tests. Therefore, Plant #120 was also selected for study.
Arrangements were made to have a technician at that plant collect
and send samples for three consecutive days. Enough waste water
was collected during each sampling occurrence to run two tests,
using lime and caustic soda solution, for pH adjustment. This
was done in order to provide a basis for a direct comparison
between the results obtained by each treatment method.
6.1.2 Sample Point Location
Figures 6-1 and 6-2 are flow diagrams for the current waste
water handling facilities of the two plants selected for study.
Samples for use in performing treatability tests were collected
from the locations indicated on the diagrams.
25
-------
PROCESS
WASTES
i
NaOH
WASTE WATER
COLLECTIC»T
SUMP
TREATMENT
TANK
DECANT
BATCH OPERATION
FLOOR
SUMP
OVERFLOW
LEGEND
Sampling point
FLOOR
SUMP
Figure 6-1.
TREATED EFFLUENT
TO SEWER
General waste water treatment process flow diagram showing the
sampling point at plant #369. (Nickel sulfate manufacture.)
26
-------
OTOER NICKEL WASTES
NiSO. PROCESS
WASTE
MONITORING
SHED WEIR
BOX
e-**
-*.
SUMP
1
REACTION
TANK
SOLIDS TO NiSO,
PROCESS
SAND
FILTER
BACKWASH
BACKWASH
TANKS
LEGEND
Sampling point
DISCHARGE
Figure 6-2. General waste water treatment process flow diagram
at Plant #120 showing the sanpling point.
(Nickel sulfate manufacture.)
27
-------
6.2 TREATABILITY TEST MODEL OPERATION
6.2.1 Treatment Technology Tested
Treatment consisted of three separate steps. These steps
included removal of nickel as nickel hydroxide by chemical
precipitation, followed by a settling period for solids
separation and then dual media filtration of the clarified waste.
6.2.2 Waste Water Characterization
Table 6-1 shows the results of analysis of two waste water
samples from Plant #369. A major difference between samples is
that the first shows a large amount of phosphate, accompanied by
high concentrations of sodium and potassium. The second shows no
phosphate. The sample with phosphate is, of course, strongly
buffered. Nickel was not precipitated until the pH was 12.5.
Also, the nickel seemed to be complexed to a degree not accounted
for by the phosphate. The sample contained ammonia and/or
amines. However, the concentrations of these compounds were not
large enough to account for the extensive complexation of nickel.
Extraction of the solution showed the presence of a polyglycol
which may contribute to the chelation.
The second sample also appeared to be buffered, in this case
on the acid side. The buffering is provided by a bicarbonate
system. No phosphate was detected, nor was there any polyglycol.
Overall, this sample behaved in a manner predictable by
solubility data with essentially complete precipitation of nickel
at a pH of 10.
treatment
6.2.3 Details on Treatability Test Operation
Tables 6-2 and 6-3 show information related to
conditions for each of the treatability tests.
The treatability tests on waste water from Plant #369 were
performed by first trying a pH of 10. If nickel precipitation
and/or settling occurred after some observation period, the test
was conducted following the normal procedure, namely, settling
and ^ dual media filtration of the clarified waste. When
precipitation and/or settling did not occur, the pH was
r.eadjusted to 12.5 and the test was conducted as before.
particular situation occurred during runs 1, 8, and 9 of
6-2 and runs 1, 4, 6, and 7 of Table 6-3.
then
This
Table
The precipitations were carried out using a solution
containing 333 g NaOH per liter, or a calcium hydroxide
suspension made by adding 100 g of Ca(OH)2 to enough water to
total one liter.
28
-------
TABLE 6-1. WASTE WATER CHARACTERIZATION FOR THE NICKEL
SULPATE SUBCATEGORY, PLANT #369(1)
Parameter
Results
Sample I Sample II
pH
Temperature (degrees C)
Total Acidity
(as CaC03)
Total Alkalinity
(as CaC03)
Total Suspended Solids
Total Dissolved Solids
Total Residue
Fixed Residue
Nickel
Zinc
Sodium
Potassium
Calcium (as CaC03)
Magnesium (as CaCO3)
Chloride
Sulfate
Nitrate
Phosphate
Kjedahl N
7.2
24
780
3,500
1,090
14,100
18,100
16,950
470
1.50
1,250
4,170
12
29
335
1,395
536
6,570
65.5
6.4
25
1,020
140
725
5,300
4,350
2,440
890
OO*7
. it
435
290
/"O
68
•-5 £•
Jo
241
2,036
20
. 2.
None detected
35.6
EFFECT OF pH ON NICKEL SOLUBILITY^ PLANT
==========—-------- Sample II
NaOH
pH
Nickel
NaOH
pH
Nickel
518
1062
1214
1290
1366
1822
2278
3036
4404
7592
7.98
9.00
9.86
10.01
10.44
10.94
11.46
11.96
12.51
12.94
480
500
510
480
480
480
410
5.30
1.20
0.67
50
217
819
1139
1215
1291
1291
1291
1367
1746
7*01
7.94
8.49
8.97
9.48
9.71
10.02
10.30
10.57
11.60
970
810
130
34
11
9
2.
3.
3.
7.
5
1
7
5
(1) mg/1 unless specified otherwise
29
-------
I
0
3
1
P*
ff\
SB
3 ;} JO1 2 § 3 3 3 § S !8
H S m § S? "? 0 °. 0
Ij Zi ;0 .«CI* ° tr> CN o «o
Mt-ICN *0 ,-f i-J M ,_] rH CO VO
H n o
OrHlO • O O* lf)^f» ^trt
Hr-iCN r-in rH 33oO t^S
rH *-* O in VO IO
CT>r1cN Sr\l 3 H3 H ?! WO
wir-ltN ^3CN rH rHCNrH rH1^
^^^ •— V
>? ^"* m ^5 rH
^5 rH O
mt~ rHCN ^reo H iHrHon r^vo
1 Sin S. § 2 in§°i "*
|VOH
r-t
O IO
r^ sf
• .
rH O
O CN
r*- r*^
> *
O O
8
332
a
oo oo oo
^r CO VD '^ OJ O4
m CN * * • •
CN CN CN CN co ro
O O
O O
H H
in m
o o mo
CN ro • .
IT) ^ O O
P
O ro o
rH CN t^
o\
o o en
•* ro CN
CN
a
m
30
-------
SUBCHTEOOKf:
TABLE 6-3. TFEKEABILirY TEST CONDITIONS AND ANALYTICAL FESOLTS
Nickel Sulfate TREATMENT: Caustic Soda Plus Dual Media
Filtration
Parameter
Test Nuifcer
Data
Volume of Waste Water
Treated (gallons)
Kaw Waste Water pH
Volume of Precipitant
Solution Used (mis)
pH Achieved by Lime
Addition
Mixing Tine (mins)
Settling Time (rains)
Supernatant pH After
Reaction
Filtrate pH
Filtration Tine (mins)
Nickel, mg/1
Raw Waste —
Total Hi
Dissolved Ni
Supernatant —
Total Ni
Dissolved Ni
Filter Effluent-
Total Ni
Dissolved Ni
Total Suspended
Solids, TSS, mg/1
Raw Waste
Supernatant
Filter Effluent
Plant S369
TREATABUJTY TEST CONDITIONS
1
9/5
20
NDA(1)
1250
12.6
15
195
12.5
12.4
90(2)
2
9/6
20
6.51
300
10.0
15
200
9.6
7.7
35
3
9A4
20
5.40
500
10.0
15
4150
6.8
6.8
30
4 5
9/21 9/28
25 25
6.33 3.26
1000 1280
12.6 10.0
15 15
135 3825
12.5 9.6
12.5 9.3
45 40
6
10/17
25
2.00
1380
12.6
15
240
12.6
12.6
40
ANAIOTICAL
NDA
NDA
320
0.49
0.5
0.5
NDA
89
105
1000
950
0.50
0.14
0.06
0.12
920
34
12
1900
1650
1.50
1.50
1.50
NDA
1070
24
16
520 4850
300 4500
0.13 1.70
0.06 1.40
<0.01 <0.01
<0.01 <0.01
2620 800
44 46
36 55
490
460
1.50
0.20
0.55
0.18
132
19
30
7
10/22
25
4.95
250
10.0
15
255
8.5
5.3
45
RESULTS
840
700
8.2
8.2
7.2
7.2
mi
20
27
8
10/30
25
6.43
250
10.0
15
285
9.1
7.1
105(3)
710
630
2.7
7.6
5.2
5.2
1090
37
22
9
1V9
25
6.70
1430
12.6
15
4185
12.7
12.9
35
•1170
330
0.31
0.21
•0.11
0.08
7500
32
32
Plant
10
11/2
25
7.65
40
10.5
15
120
10.7
7.9
65
12
11
2.1
0.2
0.05
0.14
20
8
8
#120
11
1V5
25
7.66
30
10.7
15
105
10.3
10.6
65
11
10
2.0
0.05
0.14
0.04
30
2
2
12
11/6
25
7.92
30
10.7
15
120
10.8
11.6
50
19
17
2.7
0.07
0.20
0.16
55
3
3
(1) - NDA = No data available.
(2) - Total recirculation of ttie filtrate back to the reaction tank practiced for 50 minutes.
(3) - Total recirculation of the filtrate back to the reaction tank practiced for 1 hour.
31
-------
For treating waste water from Plant #120, tests were
conducted at pH values of 10 and 10.5 (see run numbers 10, 11,
and 12 of Tables 6-2 and 6-3). From conversations with plant
personnel, it was learned that they run their nickel treatment
system in that pH range. Hence, the same values were used in
.this study. Later bench scale experiments showed that a minimum
pH value of 10 would be optimum for nickel removal from the waste
from this particular plant.
Recalculation of the filtrate back to the settling tank was
practiced for runs 1 and 8 of Table 6-3. In the other cases,
recirculation was not practiced, but samples for analysis were
taken only at the end of the filtration of the batch.
6.3 TEST RESULTS
6.3.1 Discussion of Results
The analytical results for the treatability tests performed
are presented in Tables 6-2 and 6-3.
These tables and Figure 6-3 show the dependence of pollutant
removal on the final pH after the reaction has gone to
completion. This pH value is, in turn, a function of the nickel
concentration and buffering capacity of the waste being treated.
Figure^ 6-3 also indicates that caustic soda and lime are
essentially equivalent if compared at the same pH.
Poor nickel removal was observed in some of the tests.
Unsatisfactory levels of nickel were obtained during runs 3, 6,
and 7 when lime was the treating reagent added and during runs 3,
7, and 8 when caustic soda was the precipitant used. It appears
that in some cases the precipitation reaction was slow, with the
fm !Lthat ^he PH continued to fall during the mixing period.
When the available OH- was exhausted, the pH was too low to carry
the precipitation to completion. This is a condition that can be
remedied in practice. BAT should be defined as treatment of the
waste water with either lime or caustic soda to a final pH, after
the precipitation reaction, of 9.5 or above. There were fourteen
runs that came within this definition. These tests, therefore,
can serve as a demonstration of the capability of this
technology.
Finally, it should be pointed out that an amber, goldish
coloration was observed in the filtrate from some of the tests
when caustic soda was the precipitant used. This happened during
funS i--,4' 6' and 9 and indicates that the supernatant attacked
the filter media, picking up some solids during its passage
through the filter media. This may explain the poor TSS removal
32
-------
*•"*
,— 1
!
C
0 4J
s J
C3 in
O Ci3
c-o
H
-------
observed for those 4 runs. An examination of Table 6-3 reveals
that the filter media attack occurred at a pH value of 12.6.
6.3.2 Statistical Evaluation
A statistical analysis was performed for total suspended
solids and nickel in Figures 6-4 and 6-5. The statistical
parameters used to develop the analysis appear in Appendix A. A
comparison between the proposed maximum 30-day average
concentration and the estimated experimental 30-day average is
presented in Table 6-4. The proposed BAT limitations are
designed such that compliance can be achieved at least 95 percent
(95th percentile) of the time. The curves presented in Figures
6-4 and 6-5 show the estimated probability performance for a
range of maximum 30-day average values.
6.3.3 Conclusions
The treatability test results serve to indicate the general
applicability of the proposed BAT regulations to the treatment
technology applied. Results show that the pollutant
concentration basis for the proposed BAT maximum 30-day average
effluent limitations would not be achievable with the prescribed
treatment technology under the conditions of the tests. The
estimated long term average concentration for nickel is within
the proposed BAT maximum 30-day, average; however, the
statistical analysis indicates that the proposed limitation can
only be met 56 percent of the time. The test results indicate
that the proposed BAT limitation be based on 0.27 mg/1 for nickel
which is consistent with levels reported in the literature.
Toxic metal removal by alkaline precipitation under the
conditions of the test appears to be equally efficient when
either caustic soda or lime are used. However, application of
either treatment chemical may require evaluation on a case by
case basis to guarantee that no interferences exist which would
preclude the use of one chemical over the other. .
Test results also indicate that dual media filtration
appears to provide significant additional removal of nickel from
the clarifier effluent and is needed to ensure consistent
achievement of the desired effluent quality.
34
-------
TABLE 6-4. COMPARISON BETWEEN PROPOSED BAT LIMITATIONS AND
ESTIMATED TREATABILTIY PERFORMANCE
TOR THE NICKEL SULFATE SUBCATEGORY (ALKALINE
TREATMENT)
STREAM: FILTER EFFLUENT
Pollutant
Concentration Basis
(mg/1)
Proposed BAT
Maximum
30-Day Average
Est. Treat. Performance
30-Day Average
Nickel 0.20
Total Suspended Solids, TSS 47
0.32
57
35
-------
.SUBCATEGORY
Nickel Sulfate
POLLUTANT
Nickel
PRECIPITANT
Alkaline
Proposed Maximum 30-day Average
(mg/1): 0.20
95th Percentile (Z = 1.64) (mg/1) : 0.32
Long Term Average (mg/1): 0.19
Standard Deviation of 30-day Averages (mg/1): 0.078
Probability of Achieving Proposed
Maximum 30-day Average (%) : 56
Number of Observations : 14
Estimated Prcbability That Any 30-day
Average Does Not Exceed a Given Maximum
OOOOOOOOOH
• •••••••••
H to u> ** tn cr» -J co ^o o
n n
f*>
^
s
j*
\S
/
/
/
/
/
/
/
f
/
A
/
/
/
/
f
/
r
tf
/
s
s
fT
W*
eSS.
0.20 0.25
Maximun 30-day Average (mg/1)
Figure 6-4. Estimated Performance of Proposed
BAT Treatment
36
0.30
-------
SUBCATEGORY
Nickel Sulfate
POLLUTANT
Total Suspended Solids
PRECIPITANT
Alkaline
Proposed Maximoa 30-day Average
(mg/1): 47
95th Percentile (Z = 1.64) (mg/1) : 57
Long Term Average (mg/1): 34
Standard Deviation of 30-day Averages {mg/1): 14
Probability of Achieving Proposed
t^ximuftt 30-day Average (%): 82
Nunber of Observations: 14
1.0
0.9
0.8
0.7
0.6
0.5
I 0.4
I 0.3
0
I °'2
* 0.1
0.0
10.0 20.0 30.0 40.0 50.0
Maximum 30-day Average (mg/1)
Figure 6-5. Estimated Performance of Proposed
BAT Treatment
37
60.0
-------
-------
SECTION 7.0
HYDROFLUORIC ACID SUBCATEGORY
7.1 INTRODUCTION
7.1.1 General Considerations
Treatability studies were conducted in an experimental unit
set up in the vicinity of the plant where the waste was
generated. A fresh sample of the raw waste stream was collected
each day. Between 4 September and 10 October 1979, seventeen
treatability test runs were completed using the proposed BAT
level treatment (lime treatment plus filtration).
Determinations of pHr turbidity, and acidity were made at
the site. Other analyses were made after overnight
transportation of the sample to the laboratory in Pasadena.
Additional treatability studies were found necessary after
completion of the initial 17 treatability runs. The purpose of
these studies was to determine whether changes in the sample
characteristics had occurred due to logistic lag time (the time
between sample collection and laboratory analysis). These
studies were conducted between 30 October and 8 November 1979 at
the laboratory in Pasadena, California, and the time lag was
reduced from overnight (16-18 hours) to three hours.
7.1.2 Sample Point Location
Samples were collected directly at the point of the kiln
discharge. At this point, the treated recycle water mixes with
the expelled reacted ore (primarily gypsum solids) and is
conveyed by open channel to the waste water treatment system.
The process waste stream is currently neutralized with soda ash
(Na2C03) prior to clarification in retention ponds.
39
-------
7.2 TREATAL1LITY TEST MODEL OPERATION
7.2.1 rn.t!eit-....-.ac. Technology Tested
The proposed BAT level of treatment considered in the
treatability investigation includes lime treatment of the raw
waste water, clarification, . and dual media filtration. BAT
treatment proposes the removal of TSS, fluorides, and heavy
metals in addition to acid waste neutralization.
7.2.2 Waste Water Characterization
The characteristics of a typical waste water sample from the
HF plant selected for the tests is shown in Table 7-1. Portions
of this waste water having an original pH of 3.1 were treated
with NaOH to pH levels between 8 and 12. Additional tests were
run at pH values of 3.1, 10, and 10.5 to determine the levels of
complexed fluoride and aluminum. The upward adjustment of the pH
was made with NaOH. The results are shown below.
Fluoride, ion, mg/1
Fluoride, total, mg/1
Aluminum, mg/1
pH = 3.10 pH = 10 pH = 10.51
78
193
27
35
109
0.36
43
101
0.24
The non-free fluoride at the lowest pH may be due to un-
ionized HF, and perhaps in part to AlF6(-3). In any case, the
aluminum precipitates readily at the higher pH. There is the
possibility that fluoride in the alkaline solutions is partly in
the form of silicofluoride, SiF6(-2).
In order to assess possible interferences with the
analytical procedures and determine if other pollutants were
present, emission spectroscopic analysis of the raw sample total
dissolved solid was performed. The results showed the major
metal to be sodium; minor amounts of aluminum, calcium,
magnesium, silicon, silver, and lead were indicated, as well as
traces of boron, antimony, iron, and manganese. Quantitative
tests for lead by atomic absorption were negative. The lead
indicated on the emission spectrograph may have been a chance
contaminant, or the "amount was too small to be shown by atomic
absorption under the conditions of the test.
When the results of the pH optimization tests are viewed, it
would appear that in the pH range 10-11 the pollutant levels are
minimized; hence, this is the preferred pH range for the
treatment. At lower pH levels, nickel and zinc are still present
in solution. At higher pH levels, it is anticipated that the
40
-------
TABLE 7-1. WASTE WATER CHARACTERIZATION FOR THE
HYDROFLUORIC ACID SUBCATEGORY(l)
Parameter
Result
Parameter
Result
pH 3.10
Temperature(deg.C) 23
Methyl Orange 514
Acidity (as CaC03)
Phenolpthalein 969
Acidity (as CaC03)
Total Suspended 63000
Solids
Fixed Residue 24900
Free Fluoride 78
Total Fluoride 193
Sodium 24100
Potassium
Calcium (as CaC03)
Magnesium
(as CaC03)
Chloride
Sulfate
Nitrate
Aluminum
Nickel
Zinc
Total Chromium
Hex. Chromium
91300
1650
260
632
45800
22
27
0.57
0.84
0.29
None detected
EFFECT OF pH ON SOLUBILITY OF POLLUTANTS
NaOH
Added
0
690
800
850
880
930
935
940
1290
pH
3.1
8.07
9.04
9.74
9.98
10.22
10.51
10.98
12.00
Free
Fluoride
78
41.4
32.7
32.7
35.3
38.5
42.6
50.7
111.2
Zinc
0.84
0.13
0.01
0.01
0.01
0.01
0.01
0.01
0.04
Nickel
0.57
0.39
0.24
0.17
0.01
0.01
0.01
0.01
0.01
Total
Chromium
0.09
0.08
0.08
0.08
0.08
0.08
0.08
0.08
0.08
(1) All values expressed in mg/1 unless otherwise specified
41
-------
solubility of zinc and aluminum would increase since they are
amphoteric.
7.2.3 Details on Treatability Test Operation
The conditions of the treatability tests
Table 7-2 for the initial seventeen test runs.
are presented in
In general, operation parameters including pH adjustment,
mixing time, settling time, and filtration time were varied
within the anticipated ideal range to optimize treatment process
efficiency in the test model. The need to vary these operation
parameters became apparent during the course of study, since the
removal of fluorides, TSS, and nickel did not achieve anticipated
levels. Additional test runs and laboratory experimentation
discussed in Section 7.3.3 were performed to verify these
anomalies.
Alterations in the test model reaction tank were found to be
necessary in runs 6 through 17 since the settled sludge blanket
level interfered with the tank outlet. An "L" tube was placed
inside the tank to position the outlet level approximately three
inches above the settled sludge blanket for mitigation of the
interference. The solids captured by the "L" tube after settling
were wasted before the beginning of each filter run.
Turbidity and pH in the filtered effluent was carefully
monitored during runs 6 through 17 to ensure that the dual media
filter performance was properly established before effluent
sample collection. Methyl orange and phenolphethaiein acidity
was monitored for the raw waste to determine the variability of
this waste characteristic, and also to check on measurements of
lime slurry dosage requirements.
7.3 TEST RESULTS
7.3.1 Investigation of Anomalous Results
The laboratory results for the 17 treatability test runs are
tabulated in Table 7-3. Note that the total fluoride
concentration was monitored only in the supernatant and filter
effluent waste streams.
Review of the experimental results during the treatability
test runs revealed the following observations:
1) The free fluoride concentration in the treated waste
water effluent varied widely between 27 and 65 mg/1,
and was not appreciably improved through the dual
media filtration process.
42
-------
§
•id
3
•tl
3
,_j
1
I
3
i
I
§
g
§
I
s
|
i
g
CN
^
1
1
tj
1
j
r-
rH
3
"sp
H
3
CM
rH
g
rH
01
CO
VO
in
CO
CM
rH
4J
m
i-t rg o in cj
O) Ol
\ in cn
o o vo o in
H M rH c3 eg
fn
o o o o ^*
5i
53" o 0 •» O
iH CM CM •
•s. H rH
CM CM C>4
CM P- Tl"
\ O co in .CM
Ot\ CM V O •
rH CO
CM •* 0 CO
cn CM o in CM
CM o co cn
p* CO
^0 § S ".
°* N S rH ^
CN CO 01 CM
m" 04" 5» S oj
iS in
o
o cn c*i
S a ' 1 1 2
CO
rH cn in co
eft CN cn in CM
^o | g «
oJ
» g g
** CM
^ o | | r
CM
3
CM
Sl8 o S 1
O
^
S
rH
O
O
CJ
I
o
cn
in
o
0
5
g
cn
o
CO
CM
rH
O
i
1
|
g
|
*
•8
i, g;
o
rH
tn
0
3
o
H
3
CM
O
rH
3
CO
o
rH
CO
O
rH
tn
O
«H
in
0
rH
CO
a
cn
3
«
O
m
3
10
• a
en
S
JS-8
«r
in
rH
a
o
in
rH
a
a
in
H
o
o
ve
8
o
10
8
9
O
cn
tn
rH
in
o
cn
j,
o
cv
a
o
a
rH
0
O
CM
rH
O
so
o
,cn
§
in
cn
rH
3
§
o
\o
o
p*
o
10
§
Settling Time
(mins)
CM
o
CM
O
CO
s
o
3
H
o
rH
iH
S
CM
S
rH
o
rH
rH
3-
CO
s
CM
g
3
H
S
in
3
S
vo
3
3
s!
3
o
3
3
en
en
o
o
rH
CO
en
o
a
Cn
cn
03
cn
cn
cn
ON
Oft
CO
Cn
o
CO
rH
g
*
• g
g
*
1
3
1^1 fll
!+) 215 ni3~.
j s s-| ^"§ 1
S M 6< •» g S —
u
a
1
S
i
CM
43
-------
a
o o o o
ss
dd
£5
O O
v •
3°
do
MM
• *
O O
en
o
• •
oo
S8
do
Sto
o
°°
do
SS
do
SS
do
co en
«r «K'
• •
00
a a
do
a a
do
gg
dd
ss
do
§\o
o
• •
o o
o o
do
SS
do
SS
• •
00
ss
do
O O OO OO
en en
d d
Sg
dd
oia
do
oo
do
SS
do
aa
O O
SS
gg
do
SS
do
O O
SS
• •
O O
ss
do
gg
do
•Ofi-t (MIO rHV
in in o o o o
• • • • • »
OO OO O O
gg
H d
SS
•-51-5
00
HH
as ss
• • • •
O O OO
OO O O
V
co in M en
oo oo
do do
I-H o
gg
dd
SS
rHH OO
SS
HH
ss
do
°°
do
10 iH O
SS
• •
o o
00
°.°.
Hr4
00
• •
o o
m oo
• •
o o
J-lrH
• •
o o
o
ss
do
en M
do
aa
aa
do
°,°j
H H
O O
M M
HH
O O
• •
HH
S8
HH
O O
in in
HH
g§
i-5 1-5
ss
HH
• •
00
ss
HH
gg
HH
gg
HH
O O
HO
HH
dd
o\ i*»
yjy,
do
cog
*9* ^J1
do
mo
cnH
d H
CO 9
CO CO
o o
oo
OH
• •
H H
tO 1C
• •
00
a a
do
as
d d
ss
do
ss
do
as
do
CO f-
in co
dd
to in
do
35
dd
en en
do
JSP
o o
Sto
CO
do
toin
• •
o o
ss
do
SS
do
SS
do
558
dd
SS
dd
pa enes
o in •*•
en too
to en en
•»*•*<
sss
rH a> m
u> m o
S^"*
o
• r- oo
§88
°>°i en
I"R
135
o O en
tp en in
o
c3
Smm
o oo
in rH en
ID m n
•sr
SP
oo en
rs. f*
U> rH
r*< oo
ag
V? Is*
Sen
N a\
rH O
o\ tn
enm
°,£
iH
co 55
m o
m m*
S0"
o in
o o o o o o
.00 into co co
HH do dd
SS "!5 SS
d d do HO
o
53
dd
• • H M
r> t> en o
M co en
es«» en
o oo
CO to H
N
O f> O
«* inro
in rH H
8*-
elSI
OH
enr- _
inenH
°»H
44
-------
2) The total suspended solids concentration varied between
11 and 363 mg/1, with an overall average of 153 mg/1 in
the treated waste water effluent. The dual media
filtration process showed very poor removal of suspended
solids in spite of both visual and instrumental
improvement of turbidity.
3) The removal efficiency for nickel was very poor, with
an average concentration of 0.49 mg/1 in the treated
waste water effluent.
In view of these anomalies, it was determined that
additional control experiments would be necessary to assess
whether the poor removal of suspended solids and other parameters
was a result of logistical lag times between sample collection
and analysis, or the existence of colloidal solids not amenable
to removal by the filtration processes. In the event that the
poor removal of suspended solids was a result of a colloidal
suspension, it was also a goal to determine what steps may be
taken to improve the removal of solids.
Raw hydrofluoric acid waste samples were transported from
the waste source to the treatability test model within a three
hour period to minimize logistic lag time for the experiments.
Control experiments to determine the effect of logistic lag
times indicated that the removal of free fluoride, nickel, and
TSS was consistent with observations made in the previous
seventeen runs. TSS levels remained high and unchanged through
the dual media filter, even with extended recycle, indicating the
existence of a fine colloidal suspension.
Since these results were not conclusive in eliminating the
existence of possible logistical problems, an additional
experiment on the relationship between time and suspended solids
was performed on the clarified supernatant and filtered effluent
samples for a period of one week. During this time, free
fluoride and TSS concentrations were monitored. The results of
the laboratory analyses revealed that the suspended solids
concentration almost doubled within a six day period for the
clarified supernatant. The result was particularly surprising in
view of the fact that the same trend was not observed for the
treated waste after passing through the dual media filter. The
increase in TSS indicates formation of a crystalline precipitate.
The concentration of free fluoride increased by approximately 5
mg/1 within six days, possibly because of a slow release of
fluoride from its complexes.
In summary, it can be concluded that both time lag and
existence of fine suspended solids contribute to observations of
poor and inconsistent removal of TSS in the treatability test
45
-------
runs and that the free fluoride concentration is also subject to
variations due to logistic lag time.
7.3.2 Discussion of Treatment Results
Results of the treatability investigations indicate that the
proposed BAT effluent limitations were not consistently achieved
for fluoride and nickel at the selected plant. All other toxic
pollutant parameters appear in good .agreement with anticipated
removal performance. The observations of unachievable
performance were given careful consideration throughout the
course of study, and a number of conclusions were found to partly
explain these results.
Free and Total Fluoride Removal
The solubility of CaF2 in pure water would yield about 8
mg/1 of F- at ordinary temperatures. It was anticipated that
this level could be approached in practice. However, the
treatability test results indicated fluoride concentrations
achieved in practice may be due, at least in part, to the
slowness of the attainment of equilibrium conditions between
solid and solution. In other words, super-saturation may play a
role. Support for this hypothesis is seen in the fact that there
was no precipitation until the fluoride ion concentration reached
65 mg/1, and then, up to a point, the residual fluoride decreased
as the initial concentration increased. Figure 7-1 shows that
residual fluoride decreased to about 35 mg/1 when the initial
fluoride was at concentrations above 300 mg/1.
The high ionic strength in the waste is another factor that
tends to increase the solubility of the free fluoride over and
above the effect of super-saturation. It is believed that
observations of free fluoride removal is plant specific in
nature. Lower fluoride values should be attainable where
treatment with soda ash (Na2C03) and plant recycle are not
practiced. It is likely that the fluoride residual could be
decreased by raising the level of calcium ion, but it cannot be
raised significantly until most of the sulfate and carbonate have
been precipitated. This might be accomplished if the initial
neutralization of the waste water were accomplished with lime
instead of Na2C03. In treatment systems where soda -ash is
employed to facilitate effluent reuse, fluoride concentrations
would tend to increase as a result of a deficiency of available
calcium in the presence of excess sulfate and
carbonate/bicarbonate ions. In the tests conducted, even the use
of lime for neutralization and alkaline precipitation did not
provide sufficient available calcium for efficient fluoride
removal because of the calcium demand exerted by the high levels
of sulfate and carbonate species already present in the reused
treatment system effluent.
46
-------
tr>
G
O
•H
4J
0)
u
G
8
8_-§
^
,8
H
fa
I
4->
•H
G
H
O
O
O!
0)
4->
to
I
tl
0)
4-)
0)
U
•O
(tf
>
G
H
G
O
•H
tt)
O
G
O
O
-8
•H
fc
s
1-1
fa
ni
4J
e
G
(0
•H
.G
W
G
O
•H
4J
(C
H
•S
•H
fa
47
-------
Total Suspended Solids Removal
The treatability results indicate that the dual media filter
will not effectively remove TSS concentrations below
approximately 150 to 200 mg/1. The high TSS concentration can be
attributed to colloidal or fine suspended solids not amenable to
physical separation in the filter. The composition of the
suspended solids is believed to be fine CaF2 and trace quantities
of CaS04. Examination of the suspended solids results indicates
evidence that the suspended CaS04 is effectively removed by dual
media filtration, whereas the CaF2 resists separation.
Total Nickel Removal
Removal of nickel in the form of a hydroxide precipitate is
dependent on a specific equilibrium solubility which may be
strongly influenced by high ionic strength and the presence of
fluoride. Thus, nickel removal would tend to be less efficient
and possibly erratic at plants where soda ash is used for
neutralization and effluent reuse. '
7.3.3 Statistical Evaluation
A comparison between the proposed BAT limitations and the
estimated treatability performance for each of the pollutants of
concern is presented in Table 7-4. A statistical analysis of the
treatability test runs is presented in Figures 7-2 through 7-6.
Appendix A summarizes key statistical parameters used in the
analysis.
7.3.4 Conclusions
The treatability test results do not provide an adequate
basis for assessing the general applicability of the proposed BAT
regulations for the Hydrofluoric Acid Subcategory. The
particular plant selected for study was 'an extreme case in the
sense that the kiln raw waste slurry incorporated a totally
reused treatment system effluent as the carrier. Thus, the kiln
solid residues were undoubtedly typical of the industry, but the
slurry transport medium was not because of the plant's practice
of 100 percent effluent reuse following soda ash treatment. The
high effluent concentrations of fluoride observed in the lime
treatment tests were not .unreasonable under these circumstances,
although the problem was not anticipated at the time the plant
was selected as the source of raw waste water for this study.
The results on nickel removal with lime treatment may also
be viewed as atypical of this technology possibly due to the
effects of ion pair or complex formation (resulting in a low
48
-------
TABLE 7-4. O3MPARISON BETWEEN PROPOSED BAT LIMITATIONS AND
ESTIMATED TREATABILITY PERFORMANCE FOR
THE HYDROFLUORIC ACID SUBCATEGORY
STREAM? Filter Effluent
Pollutant
Concentration Basis
(mg/D
Proposed BAT
Maximum
30-Day Average
Est. Treat. Performance
30-Day Average
Fluoride (T)
Nickel
Zinc
Chromium
Total Suspended Solids. TSS
33
0.15
0.52
0.04
68
94
0.81
0.11
0.096
230
49
-------
SUBCATEGORY
Hydrofluoric Acid
POLLUTANT
Fluoride
PRECIPITANT
Proposed Maximum 30-day Average
(mg/1) : 33
95th Percentile (Z = 1.64) (mg/1) : 94
Long Term Average (mg/1): 90
Standard Deviation of 30-day Averages (mg/1): 2.3
Probability of Achieving Proposed
Maximum 30-day Average (%): <0.01
Number of Observations: 13
Estimated Probability That Any 30-day
Average Does Not Exceed a Given Maximum
> o ooooo OOOH
• •••** t t • •
3 H to cj rf* tn cr» *j co u> o
-••
***
s
/
^
/
/
/
/
/
w
/
r
/
f
/
f
A
/
/
f
/
T
/
.^
r
**
V*
t~
\
^*
««•
t~*
Maximum 30-day Average (mg/1)
Figure 7-2. Estimated Performance of Proposed
Treatment
50
-------
SUBCATEGORY
Hydrofluoric Acid
POLLUTANT
Nickel
PRECIPITANT
Proposed Maximum 30-day Average
(mg/D
0.15
95th Percentile (Z = 1.64) (mg/1): 0.81
Long Term Average (mg/1): 0.57
Standard Deviation of 30-day Averages (mg/1): 0.15
\
Probability of Achieving Proposed
Maximum 30-day Average (%): g.24
Number of Observations: 17
1
a
1
4-
£
•r
r-
*r
X
i
0
w
1.0
,| 0.9
•3
| 0.8
1 °'7
3
5 o-6
M ^J
»
I °*5
1 0.4
a 0-3
1 02
2 0.2
§*
^ 0.1
i^*
*i*°
***
^
S
s
u-u 0.30 0.4
/
~
/
/
f
/
/
/
0 0.5{
Maximum 3
/
f
/
/
/
/
A
/
/
^
/
S
A
/
s
s
^^
^
) 0 .60 0 .70 0 .2
0-day Average (mg/1)
^"
^ «*
n*"*
0»
0 0 .90
Figure 7-3. Estijiated Performance of Proposed
BAT Treatment
51
-------
SUBCATEGORY
Hydrof luoric Acid
POLLUTANT
Zinc
PRECIPITANT
•
Proposed Maxiimm 30-day Average
(mg/1): 0.52
95th Percentile (Z = 1.64) (mg/1): o.ll
Long Term Average (rag/1): 0.074
Standard Deviation of 30-day Averages (mg/1): 0.024
Probability of Achieving Proposed
Maximum 30-day Average (%): >99
Nuttber of Observations: 17
1.0
0.9
0.2
0.0
o.d
•«•
.-••
2 o.d
w^
S
3 O.C
s
/
r
/
/
/
v
/
/
f
f4 0.05 0.06 0.0
Maximun 30-day
/
/
f
7 0.0
Aver
/
f
/
8 0.0
age (
/
/
9 0.1
mg/1)
/•
s
6 0.3
ir
^
1 0.1
•*•
2 O.J
3
Figure 7-4. Estimated Performance of Proposed
BAT Treatment
52
-------
SUBCATE60RY
Hydrof luoric Acid
POLLUTANT
Qiromium
PRECIPITANT
Proposed Maximum 30-day Average
(mg/1)
0.040
95th Peroentile (Z = 1.64) (mg/1) : 0.096
Long Term Average (mg/1): 0.076
Standard Deviation of 30-day Averages (mg/1): 0.012
Probability of Achieving Proposed
tteximum 30-day Average (%): 0.13
Number of Observations: 17
0.9
0.8
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0.0
o.c
iM*
•**
iS»*
^s
^r o.c
/
/
i
f
leT o.c
'
y
f
A
/
(
/
Y7 O.C
/
.
/
/
f
rf
^
k^
S
"7*
**
8 0.09 0.]
***•
NBttM
.0 0..
LI
Maximum 30-day Average (mg/1)
Figure 7-5. Estimated Performance of Proposed
BAT Treatment
53
-------
.SUBCATEGORY
Hydrofluoric Acid
POLLUTANT
Total Suspended Solids
PRECIPITANT
Proposed Maximun 30-day Average
(mg/1): 68
95th Percentile (Z = 1.64) (rag/1): 230
Long Term Average (rag/1): 170
Standard Deviation of 30-day Averages (mg/1): 36
Probability of Achieving Proposed
Maximm 30-day Average (%): 0.36
Nurrfoer of Observations: 17
Estimated Probability That Any 30-day
Average Does Not Exceed a Given Maximm
O OOOOO O O O H
• ••••• •• • •
H NJ CJ it* Ul ff> -JOOVOO
w*>
u.u 1(
**
^
s
to i;
s
/
/
/
/
JO 1^
/
/
-
/
/
/
F
to i*
/
A
/
/
f
0 1?
A
/
/
/
/
F
30 2C
/
/
F
s
0 2.
j*
•?*
<*e.
20 24
»*•
••-
*••
0 It
0
Maximum 30-day Average (mg/1)
Figure 7-6 . Estimated Performance of Proposed
BAT Treatment
54
-------
activity coefficient for nickel) at the high ionic strengths
encountered.
A manor conclusion that can be drawn from this study is that
the addition of dual media filtration after alkaline
precipitation and settling is not particularly effective for the
reduction of final TSS and total fluoride concentrations.
Further, dual media filtration does not appear to be justified on
the basis of additional toxic metal removal judging by the
results presented in Table 7-3.
55
-------
-------
SECTION 8.0
COPPER SULFATE SUBCATEGORY
8.1 INTRODUCTION
8.1.1 General Considerations
The treatability tests for the copper sulfate subcategory
were conducted at the in-house laboratory facilities of Versar,
Inc. located in Springfield, Virginia. Eight samples were
collected, at Plant #034 and expeditiously transported to the
laboratory facility where the test runs were conducted. A total
of 24 test runs were completed between October 19 and November
14, 1979, and all test runs were made within one week after raw
waste sample collection..
8.1.2 Sample Point Location
Figure 8-1 shows a general process schematic flow diagram
for Plant #034 showing the raw waste sample point location
selected for study. The stream sampled includes wastes from
leaks, spills, and washdown which collects at a sump in the
basement of the facility. About one quarter of this waste water
by volume is comprised of contaminated ground water from the
immediate area.
8.2 TREATABILITY TEST MODEL OPERATION
8.2.1 Treatment Technology Tested
The proposed BAT treatment includes alkaline precipitation
of toxic metal pollutants 'with either lime or caustic, followed
by clarification of suspended solids and final polish through a
dual media filter.
57
-------
65
g
1
1
ii
«
0
I*
S-
s
I
I
9i
S1
TT
B« a
3_»s
s
a
(D
.p
u
id
Q)
.p
w
H
cd
rd -rH
•H O
tr>
fi
ra g
tn id
OJ 01
o
o
-------
8.2.2 Waste Water Characterization
The major toxic pollutants studied for this subcategory
include copper, nickel, and selenium. Since optimum removal of
nickel and copper by alkaline precipitation occurs at slightly
different pH values, it was necessary to determine the pH for
optimum removal of pollutants. Therefore, a five gallon sample
of waste water was collected for study prior to the treatability
test runs. Results of the characterization are presented in
Tables 8-1 and 8-2. The copper concentration ^creases slightly
at higher pH values indicated in Table 8-2 as anticipated (Figure
2-1). Since the minimum solubility for all the pollutants
occurred at a pH of 10, it was decided to conduct the
treatability test runs at this pH value.
8.2.3 Treatability Test Operation
The BAT concept was tested comparing both lime and caustic
soda for alkaline precipitation of metals. Total recirculation
of the filter effluent back to the reaction/settling tank was
practiced for runs 1 through 4; however, the filter was operated
on a once-through basis for runs 5 ^through 12 Because
recirculation was found to have no appreciable effect on tne
efficiency of solids removal.
Chemical dosages for metal precipitation are shown in Tables
8-3 and 8-5 for each of the treatability test runs. Data
pertaining to the pH of raw and treated waste water, settling
time, time of filter effluent sampling after start of filtration,
and mixing time are also included in these tables.
8.3 TEST RESULTS
8.3.1 Discussion of Results
The analytical results tabulated in Tables 8-4 and 8-6 show
that both lime and caustic soda are effective in lowering the
concentrations of copper, nickel, and total suspended solids.
However, no significant removal of selenium was shown. Selenium
was present only in very low concentrations, probably as selenite
and selenate, both of which are relatively unaffected by pH.
Recirculation of the filter effluent back to the
reaction/settling tank did not seem to improve the filter_removal
efficiencies for any of the pollutants under consideration. It
is clear from an examination of Tables 8-4 and 8-6 that a 50
minute once-through filtration time was adequate in the test
model to provide a good solids separation.
59
-------
TABLE 8-1. WASTE WATER CHARACTERIZATION FOR THE PLANT
SELECTED FOR STUDY IN THE COPPER SULFATE SUBCATEGORY
Specific Waste Constituent (mg/1)
Copper Nickel . Selenium
Total 221 22 0.20
Dissolved 213 22 0.17
TABLE 8-2. EFFECT OF pH ON SOLUBILITY OF POLLUTANTS
ss===ss===Ss=s=ss==-=-=s===__=_____=___________s______=___=___
Precipitant
Sodium Hydroxide Lime
Amount Present (mg/1)
Final pH . Copper Nickel Copper Nickel
7-° 7.6 7.4 2.6 16~~2
8-° 9.7 6.7 2.5 10.0
9-° 9.6 2.2 3.5 2.2
10.0 13.4 1.6 4.7 0.7
60
-------
CCI,
CM
£j i — 1
•3
1 d
to O
Jj r~1
cn
r-H
s GO
i ^
1
*>
IT)
CO
CM
$
(5 rH
1
SUBCRTEGORY: Copper
Test Number
*< H cn
38 co* CM"
r—
^ °. ^
H CM CO CM
vo
\ o in
o o • *
H CM CO CM
\ rH CM
O O • *'
rH CM CO CM
\ ° °i
rH CM CO CM
\ 0 H
H CM CO CM
§ o H- *
rH CM CO rH
^ H rH
o ° -^ ^f
rH CM CO CM
OJ rH VO
Cn CM CO rH
CM co r-
cn' CM co i-H
^o M. ^
Cn" CM CO rH
rH co cn
\ o • *
Cn CM CO . rH
Q)x'*"
jU (1)
ft &Q
(o H3
o> H m O
us 1 2 »
g: nj |s -y
m u rH jg a 8J
S 3 H g -ri H
m >3 & id ^ O -d
jp rH jg o g 9> a
(rt OS — iy M ?7
a £ K cn
a o
. 0
o m in o
H H i-H rH
oo o ^o
. in CM
cn H rH cn
0 °.
S rl rH S
00 O f-
. in CM •
cn rH rH cn
H. 0 1-
S S rH 0^
cn o m
. in CM
Cn rH rH Cn
cn o m
.0 CM
in
m o
r-^ H
CO O
CM O
cn i-H
co o
r^ rH
CO
C!
Filtrate pH
Filtration Time (mi
61
-------
1
8
I
i
1 S
H A!
w -g
3 S
& V>
pi lj|
fj (ij
i i
65 H
*** 2
1 9
ts _9
t-J r~l
g ^
S 0}
H} .S
g M
H
1 1
a B
gj
>* §3
30
!
H
CM
H
3
o
rH
CT>
°°S
g1
-S
r- to
9j
rH
n
•p
Tj
a
m S
rH
I
3
CO
CM
rH
Test Number
O o
0 0
CO CM
O O
CT> O
CM CO
O O
•H rH
CM CM
in o
0 0
CM CM
o m
rH 0
CM CM
O O
CO CO
CM CM
o in
r- •<*
CM CM
m r?
rH
m in
in in
rH iH
Cn OO
^J* CO
CM CM
VO VO
CN CM
VO VD
rH rH
CO CO
Copper, Cu
Raw Waste
Total Cu
Dissolved Cu
Supernatant
00 O
m »*
o o
CM ^
t-~ CM
• •
0 0
vo in
CO CM
•H 0
Cn OO
O rH
• •
iH O
^J* r~J
• •
o o
CM O
CM CM
rH 0
rH VO
CO CO
rH 0
in rH
VO CM
O CD*
0 O
H vo
CM CM
O O
in "3*
H CM
o a\
f- CM
•H 0
CM r-
in rH
CD* O
-P
Total Cu
Dissolved Cu
Filter Eff luer
vo oo
0 O
0 0
CO CM
iH rH
O 0
o en
H 0
0 0
CM 0
rH rH
* •
0 0
in en
rH 0
0 O
l-H rH
O O
00 ^f
rH H
• *
0 0
en CM
0 rH
O O
00 VO
CN rH
0 0
in en
CM CM
• •
o o
VO rH
in co
o o
oo a\
o o
• •
O 0
Total Cu
Dissolved Cu
Nickel, Ni
o o
9 *
r- vo
CO CO
0 0
r-! [-^
CM CO
0 0
oo r~
co. co
0 O
en oo
co co
0 0
CO CO
o o
in in
co co
0 0
*^3* CO
co co
^1* rH
CO CO
to co
B r-
Q O
K co
co en
en oo
co co
CnrH
en o
CM CO
^j* CI3
00 OO
CM CM
Raw Waste
Total Ni
Dissolved Ni
Supernatant
rH cn
H 0
O O
in en
rH 0
0 O
co m
CM O
o o
V
on m
rH 0
o o
V
in in
0 0
o o
V
cn m
H 0
o o
V
o m
CM 0
0 0
•^ co
CM rH
O C3
vo r--
co ^
0 CD
CM rH
0 C>
vo oo
CM O
0 CO
cn r>
0 O
cJ o
•P
Total Ni
Dissolved Ni
Filter Eff luen
oo r~
o o
o o
vo r-~
o o
o o*
in in
o o
O C3
V V
in in
0 0
o o
y v
m m
o o
o o
V V
m in
o o
O 0
V V
r-~ oo
CD 0
0 CD*
33
CD* CD*
33
CD CD*
5- H
CD CD*
rH rH
CM CM
O O
in in
0 0
CD* CD*
Total Ni
Dissolved Ni
62
-------
rrt
TABLE 8-4 - continue
rH
rH
H
O
rH
rH
00 \.
.3
r» w
0
1
VO
I
in Q)
rH
^
"*
co
CM
rH
Test Number
CM CM
0 H
0 0
CO CM
rH rH
* •
0 0
CM CO
rH rH
• •
O O
CO CO
rH rH
• •
0 0
CM CM
rH rH
O O
-a- co
HrH
O O
CM CM
rH H
O 0
rH CM
rH rH
CD 0
CM rH
rH rH
CD O
•si1 rH
•-H rH
O O
CO CO
rH rH
O CD
00 00
t-H rH
0 0
Selenium, Se
Raw Waste
Total Se
Dissolved Se
m ^r
rH rH.
O O
•* CM
rH H
O 0
CM •=*
r-H rH
0 CD
rH "*
rH rH
* •
O O
CM CO
rH rH
0 •
O O
CO CM
rH rH
0 CD
00 rH
rH rH
* •
O O
CM CM
rH rH
CD CD
0 rH
rH rH
O CD
•sJ1 CM
rH rH
CD CD
33
CD CD
CO CO
rH rH
CD CD
Supernatant
. Total Se
Dissolved Se
sa
o o
CM CM
rH rH
0 CD
CO CO
rH rH
CD CD
rHCM
"~i "~!
o o
H CM
rH H
CD CD
VO CM
rH rH
CD CD
H rH
rH rH
CD CD
O 0
rH rH
• •
O O
rH 0
rH rH
0 CD
rHin
r~i "~i
0 0
CM rH
rH rH
« %
O O
in >*
rH rH
• •
0 0
Filter Effluent
Total Se
Dissolved Se
Total Suspended
r-» -1-£^3f^ TTIC'C'
r~ co CM
rH CO CO
COCA CM
rH ^ CM
00
vo vo •
rH rH O
cr« o
00 "Sf VO
G\ rH
CM CM
^^4 • • •
CT^ lO ^*P
• vo
CO <* •
•=* rH in
o o
r- oo ^t"
rH
co -m
«c^< O •
rH CM CO
CM
• in
o o •
CTl rH CO
CM O VO
vo • •
rH CM rH
in oo CM
o • •
rH rH 0
DUJLJLUE9 , 0-OiJ
Raw Waste
Supernatant
Filter Effluent
63
-------
1
§
ci
i
i«
.S
o *-P
Tl S
id -H
CJ p4
•*
i 1
1
&
i
n
a
1
in
i
CO
P 0}
"m
m
?i1
SUBCATEGORY: Copp
oa
rH
^
3
cn
00
r-
vo
in
-
ro
01
1
cn
\
o o
rH Ol
§0
H 01
VO
rH
CD CD
H 01
3
0 0
rH 01
So
rH Ol
cn
0 0
rH 01
§ S
01
CD ^D
rH CM
l«
01
c> S
Cj\
< o
cn oj
r.
rH
^s O
cn 01
Date
Volume of Waste
Water Treated
(Gallons)
f^ ^3*
CO 01
H t~-
ro CM
o o
ro oj
o in
CO CM
O ">}<
ro CM
o in
ro oi
rH in
CO 01
H m
ro CM
o oo
CO CM
rH CM
ro oa
01 sf
CO CM
^r ^t*
ro CM
r
Raw Waste Water pH
Precipitant Dosage
quired ( gm CaO
13 ^,
rH en
3 vo
rH Cn
£3 vo
rH en
§00
4
rH cn
<3
c3 f"*>
rH rH
rH cn
g °1
rH cn
in
§ 0
H H
m
Is o
to H
S O
^H
J2J CD
VO rH
.R
•^ OO
vo cn
Vgal of was
Precipitant Solutic
Strength
pH Achieved by Lome
Addition
o
in in
rH rH
O
in CM
rH rH
rH S
O
H H
a §
rH S
in vo
r-i cn
in o
rH cn
in o
H cn
o
o in
CM rH
o
O CM
01 rH
O
O 01
CM H
Mixing Time (mins)
Settling Time (minE
ro
•«*
cn
in
cn
5
in
cn
rH
o
O
H
in
r--
^T\
t^*
*
cn
o
00*
b
Supernatant pH Afte
Reaction
O
• o
oo in
00
• o
r- in
« o
r- in
• o
r- in
CM
• o
en in
• o
cn in
in
• 0
oo m
in
• o
cn in
o
• o
cn in
in o
P> rH
r-. o
r^ H
in o
t^ H
*•»
a
Filtrate pH
Filtration Time (mi
(1) N = Normalitv
64
-------
rg
H
O
rH
in
I—1
5!
co
O O
O O
CO CO
o o
o en
co CM
in in
in in
CM CM
o o
O rH
CM CM
o in
CM CM
O CO
«* CO
CNJ CNI
O O
CO CO
CM CM
O J
• •
O O
r~ in
05 <*
• •
o o
00 CO
O H
• a
o o
rH
H O
« •
O O
en o
r- H
* •
en o
vo en
vo CM
• •
o o
co H
e •
o o
«M
O
••*
o o
r- oo
CO CO
o o
VO VD
CO CO
o o
• e
CO CO
o o
oo r*
CO CO
o o
e •
en oo
co co
o o
I**"* ijQ
CO CO
o o
in in
CO CO
o o
•«* CO
CO CO
co r~
• t
CM CO
CO CO
O O
O O
CO CO
H en
["** ^*
CO CO
^" CM
• t
CO 00
CM CM
.H
3
Jgl
1 *r-!
i P P
in co
H 0
• 9
0 0
vo vo
H o
o o
oo in
CO O
o o
V
in in
rH 0
• •
o o
V
•^ in
rH 0
C3 O
' V
o in
CO O
* *
o o
V
co in
HO
* •
0 0
V
H in
H 0
* *
0 0
V
H m
H 0
• *
O 0
H 00
vo o
• •
0 0
r>. p-
CM O
• •
o o
co en
in o
• •
o o
•H
jj "^
Isl
ill
oo in
0 0
* *
0 0
VO VD
O O
» •
0 0
CM VO
H H
0 0
VO VO
o o
• •
o o
in in
0 O
• *
0 0
V V
00 00
0 0
• e
0 0
in in
o o
* •
o o
V V
in m
0 0
• *
o o
V V
r-i in
r^ o
• •
0 O
0 0
OJ (%1
• •
o o
O H
H H
0 •
O O
^ s
rH Q
• C_]
o £3
.p
§ -H
3 IS
r- r-t
*H -d
.asl
1 1
3 ro w
•P -P w
H 6 -H
•H EH P
-------
*
1
1
I
vo
00
CN
rH
rH
rH
O
_
oo v
|
c
•r
p~ o
a
^
u
t
J
r—
s
ro
CN
u
fi
•p
da
o o
in H
rH rH
0 0
CN >*
rH H
0 O
CM «*
rH iH
O* 0
iH ^
rH rH
• •
O 0
•* CM
H H
CD O
CM CM
H rH
• •
O O
rH rH
O O
rH 0
rH rH
O O
CM CO
O O
*3< co
H rH
O 0
in •<#
rH H
0 0
Q)
rn
$ •?
•• -P W c>
•H JS jd W
oo co
rH rH
O O
CN rH
rH H
O 0
VO VO
rH rH
O O
in in
rH rH
0 0*
•«* co
rH rH
• %
O 0
CM rH
i-H iH
CD 0
Si<
3 H
33
O O
CM CM
rH rH
O C3
rHCO
CD CD
CN CO
rH rH
O 0
33
0 0
0)
-p OT
•P rS 5>
fO rH
CO CN
rH rH
0 0
H rH
rH H
CD CD*
VO **
H H
CD CD
VO VO
rH rH
O CD
CN CO
rH i-H
CD CD
CO ^
•H rH
CD CD
33
• •
0 O
in -^
rH rH
O O
33
CD* CD
CO
-------
8.3.2 Statistical Evaluation
Tables 8-7 and 8-8 compare the proposed BAT maximum 30-day
average concentration with the estimated treatability 30-day
average for lime and caustic treatment, respectively. The
results are based on the statistical analysis presented in
Figures 8-2 through 8-9 and information in Appendix A.
It should be pointed out that the proposed BAT limitations
must be met at least 95 percent of the time by any industry
having to comply with the limitations.
8.3.3 Conclusions
The proposed BAT treatment is effective for the removal of
pollutants whether lime or caustic is used as the treatment
chemical. Removal of selenium by alkaline precipitation is
ineffective at the low raw waste concentrations observed. The
initial selenium concentrations were near the lower limit of
treatability which is the basis of the proposed maximum 30-day
average limitation. Therefore, conclusions may not be made
regarding use of proposed BAT treatment in situations where
selenium may conceivably exist at higher concentration levels.
67
-------
TABLE 8-7. COMPARISON BETWEEN PROPOSED BAT LIMITATIONS AND
ESTIMATED TREATABILTIY PERFORMANCE FOR THE
COPPER SULFATE SUBCATEGORY (LIME TREATMENT)
STREAM: FILTER EFFLUENT
Pollutant
Concentration Basis
Proposed BAT
Maximum
30-Day Average
Est. Treat. Performance
30-Day Average
Copper
Nickel
Selenium
Total Suspended Solids, TSS
0.40
0.10
0.10
25
0.21
0.13
0,13
6.0
68
-------
.SUBCATEGORY
Copper Sulf ate
POLLUTANT
Copper
PRECIPITANT
Lime
Proposed Maximun 30-day Average
(nig/1) : 0.40
95th Percentile (Z = 1.64) (mg/1) :
Long Term Average (mg/1) :
Standard Deviation of 30-day Averages (mg/1) :
Probability of Achieving Proposed
Maximum 30-day Average
Nurrfoer of Observations :
(%) :
0.21
0.17
0.020
>99
12
1.0
o 0.8
n)
lity
P
Es
0.6
0.5
0.4
fi 0.3
0
9 0.2
0.1
0.0
I
0.13 0.15 0.17 0.19 • 0.21
Maximum 30-day Average (mg/1)
Figure 8-2. Estimated Performance of Proposed
BAT Treatment
69
0.23
-------
SUBCATEGORY
Copper Sulf ate
POLLUTANT
Nickel
PRECIPITANT
Lime
Proposed Maximum 30-day Average
(mg/1):
0.10
95th Percentile (Z = 1.64) (mg/1): 0.13
Long Term Average (mg/1): O.ii
Standard Deviation of 30-day Averages (mg/1): 0.014
Probability of Achieving Proposed
Maximum 30-day Average (%): 36
Nunfoer of Observations: 12
1
i
4-
1
\
I
1.0
(| 0.8
ri 0.7
6
'•a °-6
rfli
ffi
1 0.4
fl 0.3
Q)
flj
* 0.1
>-.•
1^.*
— ••
^*
^
^
/
j
r
j
'
j
f
/
/
j
/
^
^
y
/
^
/
^
^
^
^
j
^
X1
X
X*
.^
*«*
•*•
••i
•MH
°«° 0.08 0.09 0.10 0.11 0.12 0.13 0.
Maximum 30-day Average (mg/1)
14
Figxire 8-3. Estimated Performance of Proposed
BAT Treatment
70
-------
SUBCATEGORY
Copper Sulf ate
POLLUTANT
Selenium
PRECIPITANT
Lime
Proposed Maximum 30-day Average (mg/1) : 0.10
95th Percentile (Z = 1.64) (mg/1):
Long Term Average (itg/1) :
Standard Deviation of 30-day Averages (ing/1) :
Probability of Achieving Proposed
Maximum 30-day Average
Number of Observations:
(%) :
0.13
0.12
0.0031
<0.01
12
f
o
n
fei
-H
4J O
a
I
P
ge
Es
1.0
0.8
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0.0
0.116 0.113
0.120 0.122 0.124 0.126
Maximum 30-day Average (mg/1)
Figure 8-4. Estimated Performance of Proposed
BAT Treatment
71
0.128
0.130
-------
SUBCATEGORY
Copper Sulfate
POLLUTANT
Total Suspended Solids
PRECIPITANT
Lime
Proposed Maximum 30-day Average
(mg/1): 25
95th Percentile (Z = 1.64) (mg/1): g.o
Long Term Average (mg/1): 4.7
Standard Deviation of 30-day Averages (mg/1): 0.82
Probability of Achieving Proposed
Maximum 30-day Average (%): >gg
Number of Observations: 12
1.0
0.8
0.7
0.6
0.5
0.4
0.3
00
.2
.1
**
***
fS
S
S
j
r
s
R -4.
/
/
A
/
A
s
''n 4
/
*
/
'
f
/
^ 5
/
/
/
/
0 5.
S
^
S*
•?*
-
.R fi
-------
TABLE 8-8. COMPARISON BETWEEN PROPOSED BAT LIMITATIONS AND
ESTIMATED TREATABILITY PERFORMANCE FOR THE
COPPER SULFATE SUBCATEQORY (CAUSTIC TREATMENT)
STREAM: FILTER EFFLUENT
Pollutant
Concentration Basis
fag/1)
Proposed BAT
Maximum
30-Day Average
Est. Treat. Performance
30-Day Average
Copper
Nickel
Selenium
Total Suspended Solids, TSS
0.40
0.10
0.10
25
0.30
0.10
0.1.4
7.0
73
-------
SUBCATEGORY
Copper Sulfate
POLLUTANT
Copper
PRECIPITANT
Caustic Soda
Proposed Maximum 30-day Average
(mg/1): 0.40
95th Percentile (Z = 1.64) (mg/1) : 0.30
Long Term Average (mg/1) : 0.25
Standard Deviation of 30-day Averages (mg/1) : 0.034
Probability of Achieving Proposed
Maximum 30-day Average (%) : >99
Nunber of Observations: 11
Estimated Probability That Any 30-day
Average Does Not Exceed a Given Maximum
oo ooooo OOOH
• • ••••• •• ••
o H to cj *» 01 m -^4 oo vo o
<•>"
0*
3^
^
4
/
/
t
/
/
/
/
/
i
/
/
f
/
/
/
f
,J,
s
/
/
s
*r
**
***
'
— .*
0.18
0.22 0.26 0.30
Maximum 30-day Average (mg/1)
Figure 8-6. Estimated Performance of Proposed
BAT Treatment
74
.34
-------
SUBCATEGORY
Copper Sulfate
POLLUTANT
Nickel
PRECIPITANT
Caustic Soda
Proposed Maximum 30-day Average
(mg/1): 0.10
95th Percentile (Z = 1.64) (mg/1) : 0.10
Long Term Average (mg/1): 0.089
Standard Deviation of 30-day Averages (mg/1): 0.0076
Probability of Achieving Proposed
Maximum 30-day Average (%): 92
Number of Observations: 11
1.0
:*§ °-9
«
Estimated Probability That Any 30
Average Does Not Exceed a Given Ma
ooooooooo
• •••«••••
O I-1 to to *» Ui -j co
***
^
*'
s
s
r~
0.080
/
r
f
/
/
,
/
'
f
(
/
...jt
0.085 0.090 0.
Maximum 30-day Averag
t
s
s
/
*
^
,*— '
B*-1
095 0.100' 0.
e (mg/1)
^^*
105
Figure 8-7. Estimated Performance of Proposed
BAT Treatment
75
-------
.SUBCATEGORY
Copper Sulfate
POLLUTANT
Selenium
PRECIPITANT
Caustic Soda
Proposed Maximum 30-day Average
(mg/1): 0.10
95th Percentile (Z = 1.64) (mg/1): 0.14
Long Term Average (mg/1): 0.13
Standard Deviation of 30-day Averages (mg/1): 0.0026
Probability of Achieving Proposed
Maximum 30-day Average (%): <0.01
Nuntoer of Observations: 11
Estimated Probability That Any 30-day
Average Does Not Exceed a Given Maximum
3OOOOOO OOOH
• •§••• ••••
3 H N> OJ ife Ul 01 -JOOU5O
•M^
_••
**
/
/
*
/
r
,
f
/
*
J
f
)
f
J
f
/
f
A
/
J
?
AT
'
s
4*
^*
•MMM
MWMW
0.132 0.134 0.136 0.138
Maximum 30-day Average (mg/1)
Figure 8-8. Estimated Performance of Proposed
BAT Treatment
76
0.140
-------
.SUBCATEGORY
Copper Sulfate
POLLUTANT
Total Suspended Solids
PRECIPITANT
Caustic Soda
Proposed Maximum 30-day Average
(mg/1): 25
95th Percentile (Z == 1.64) (mg/1):
Long Term Average (mg/1):
Standard Deviation of 30-day Averages (mg/1) :
Probability of Achieving Proposed
Maximum 30-day Average
Number of Observations :
(%) :
7.0
5.6
0.86
>99
11
Estimated Probability That Any 30-day
Average Does Not Exceed a Given Maximum
O OOOOO OOOH
• » « • « • . . •-.
H tO U> rf* Ul CT* >J 00 VQ O
n n
L*"*1
'
s
J
^
f
*
^
i
/
t
I
/
f
/
f
/
f
,
/
/
t
s
r
,
*
^
^-«^
**•
mm*
**~
sss
£22
MBBI
aaaw
MMM
•Ml
4.0 ' 5.0 6.0 7.0
Maximum 30-day Average (mg/1)
Figure 8-9. Estiirated Performance of Proposed
BAT Treatment
77
8.0
-------
-------
SECTION 9.0
CHLOR-ALKALI SUBCATEGORY (DIAPHRAGM CELL/GRAPHITE ANODE)
9.1 INTRODUCTION
9.1.1 General Considerations
The on-site treatability studies were conducted at Plant
#967 which was selected for study. A total of 15 runs were
performed using raw waste water collected on a daily basis in the
period between October. 2 and October 23, 1979. Raw waste water
samples were split with plant personnel at their request during
the time the waste collection took place.
9.1.2 Sample Point Location
Figure 9-1 is a process flow schematic indicating the waste
water sampling point location for the treatability study. The
sampling point largely contains cell room wastes which are
normally laden with lead and asbestos.
9.2 TREATABILITY TEST MODEL OPERATION
9.2.1 Treatment Technology Tested
The treatability experiments for the waste water in this
subcategory involved an assessment of lead, chromium, nickel, and
total suspended solids (TSS) removal.
BAT treatment consists of pH adjustment to accomplish
alkaline precipitation of the heavy metals from the waste water,
followed by removal of the settled hydroxide metal sludge by
gravity separation. This is followed by treatment of the
clarified waste with a ferrous sulfide suspension for additional
metal precipitation. Dual media filtration of the clarified,
sulfide-treated waste completes the overall treatment for this
type of waste.
79
-------
a
CO
I
S1
Ojr
•p
id
11
m u
M 4J
CQ CQ
CTl
80
-------
9.2.2 Waste Water Characterization
Table 9-1 presents the results of the initial
characterization tests on the raw waste water. Beaker scale
treatability tests were performed to determine the pH for optimum
removal of pollutants. The results of those tests are presented
in Table 9-1. It can be seen from an inspection of the data that
treatment of the waste water to a pH of around 10 produces the
lowest concentrations of heavy metals in solution. Therefore,
this pH value was utilized to carry out the alkaline
precipitation process in the treatability test runs.
Beaker tests were not used to determine sulfide dosage for
optimum removal of the metals under consideration from the
supernatant after removal of the hydroxide sludge. The dosage
and concentration of the ferrous sulfide suspension used in the
treatability tests is given in Table 9-3.
9.2.3 Treatability Test Operation
As pointed out above, a pH of 10 was found to be optimum for
removal of lead and nickel. However, measurement of the raw
waste pH always showed a value above 11 when the treatability
tests were performed. Therefore, it was always necessary to
adjust the pH of the raw waste below 10 by addition of acid if
the effectiveness of lime in removing those metals was to be
demonstrated.
Adjustment of the pH to 10 was accomplished by adding
sulfuric acid and then making the final adjustment by adding
lime.
Table 9-2 summarizes
treatability test runs.
the operation parameters for the 15
9.3 TEST RESULTS
9.3.1 Discussion of Results
Analytical results are tabulated in Table 9-3 for the toxic
pollutant parameters of concern; namely, lead, nickel, and
chromium.
Review of the results indicates a relatively poor and
inconsistent removal of metals. Total lead concentrations vary
from 3.3 to <0.05 mg/1 in the filter effluent, nickel from 1.21
to <0.05 mg/1 and chromium from 0.138 to 0.04 mg/1.
81
-------
TABLE 9-1. WASTE WATER CHARACTERIZATION FOR THE PLANT
SELECTED FOR STUDY IN THE CHLOR-ALKALI SUBCATEGORY,
DIAPHRAGM CELL (GRAPHITE ANODE)(1)
Parameter
Result
Parameter
Result
Methyl Orange 23,000
Alkalinity (as
CaC03)
Hydroxide Alka- 17,800
linity (as CaCOS)
Total Suspended 166
Solids
Total Dissolved 180,000
Solids
Total Residue 179,000
Fixed Residue 171,000
Calcium 14
(as CaCOS)
Magnesium (as CaCOS) 1.3
Chloride 92,471
Sulfate 260
Nitrate 0.40
Sodium 65,900
Potassium 30
Chromium 0.14
Lead 1,160
Nickel 0.80
(1) All values in mg/1 unless specified.
PH
EFFECT OF pH ON SOLUBILITY OF POLLUTANTS(1)
Chromium Lead Nickel
8.0
8.5
9.0
9.5
10.0 .
11.0
0.18
0.19
0.06
0.21
0.06
0.17
10.9
5.5
4.5
4.3
2.4
26.5
4.13
3.74
3.26
2.48
2.58
4.13
(1) All values in mg/1 unless specified.
82
-------
fai
o «•? *d
||l
•til
IP
(3 i3J (0
J2 fl)
sal.
i
1
p
^
1
f
Q *fl)
•H C
||
IU
1
1
1
in
H
S
rH
a
1:1
s
cn
CO
r-
u>
in
-
n
CN
1
4J
CM VO H
O O rH P^ in CO
H CM H tH
^r vo en vo
o o rH in o co
H CM rH CM
\ • • r- •
O O rH rH CO Cn
H CM rH CM
CM f> VO ^
o o rH vo tn co
rH CM rH rH
0 H
4J rrt m ^| fj S^
?IsS i|| li
§ 1 ^ * 1 3 m I 1
S &&3 Is- a's 1* a;
CM ^^J
am o &\ in <*3 3 JQ
rH r*» rH »™i r~t Jg r— I
§r>
• m
fHrHr- rH rHr>inrH
o in
ass 3 aa«a
VO
aas a asosa
CO
aas 3 aSt-;a
o cn
ass. a aa^a
aas a as 2s
„
moo co'inm »in
o
o o o m tn y- -in
f»1CSr^ •"! t-tr^COrH
atno *O''nc-i«m
rH C- i-l «H H W H
!-4
inino -« "5 2 -! S
rH ^ I— rH iHCnCncM
SrH
a> ol
fr, r- >H rH r-4 CO rH
cn
inino n "!!D~r2
rr IH r» H H r* r* n
000 « in PI • o
ncnr* f-t i-HrHr-m
in
moo m min«o
^3*inr» rH i-H^r^d
Jd) -H OOtarHrH 01 >H S
HS EH ffi433*-'CJ £ EH ftfi
3 1".! f.| l§i«li |1§ -II si S si
3 i* s^ I'^sSI |a- |~ I1 g g^
•8
§
CM
1
•1
i
.5
J.
«
I
•s
0}
&
o .
ro
1
I1
•g
cn
o
•8
3
8
1
1.
5
1 »
II
rt M
3 *
(d u
II
-------
JJ
i
I
& i
8 *
ft T3
1 §
5y £
il ^
*| J>
PI E
§g *
g| •?
|| |
ie °l
|| ||
5 ''J ^C*
o.e)
cK S ,
M §1
i I g
S rH
***
I
g
i
1
a
3
rH
m '
rH
a
d
•3
i
i ^!
•5
col
f
rJ
^1
VO^
in
^t
m
H
1
i
4J
1
O
mO rH VO
iH O CO rH
CO
0
CN ** 3 CO*
O O rH
vo r— H •
V CO rH CM
go ^r-
vo in m vo
1*1 1** i*» •
in** CM vo
§8 S5!
v eg in r»
CO CM
co r> i-» •
eg N mm
8 in ""JcM
•0- rH «
^r CM in in
o
cgm
o o in
in ^r CM ^«
o o •
vo en TJ« r*
VO "» CM
00 H
VO in rH i Sa pi
| — «5
eg co SS
VO rH OO
V
SiS ON
do d d
mrH S3
n CM o o
V
^m S3
en *3* o o
V
vo eg
CM Or-]
rH IT! OO
Svo SS
inm d d
r-o men
oat "2"1.
rH O O
in ** m rH
oV ^
rH O O
tn o
en o rH
M CO O O
V
« ss
rH VO O O
"2H. S.3.
vo m o o
V
1 vo co vo
rH r- CO (O
I-H- 3d
• • • •
ro H en in
M (*1 CM PI
oo o o
CO 0 CM
00 • • *
rH CO CM CM
1
i « S s
I *Cf **H 'Tj
?S 5 H S >
JnH 3! | gl
! § S a -^ S a
fa
o
VO
m
o
Pi
o
en
CM
o
vo
vo
o
o
CO
S
m
5
r-
o
3
g
VO
1
O
CM
S
o
S
o
""•
I
f
i-H
(jj g;
*=J !s
w 3
o o
rjg
O O
ss
dd
H 0
m in
r-5 i-i
S3
do
SCO
in
do
^•rH
r* r*
dd
ss
do
S3
0 0
V V
ss
0 O
rH in
rH.O
"o o
V
vo'in
CJO
o o
v
vo in
00
' oo
V
as
, d d
vo in
CMrH
do
1
JS1
•* rt 2
(
1" rH
in vo
d d
SS
d d
m o
in in
d d
as
H r-5
•w in
O CO
rH 0
in in
d d
CM in
r- vo
d d
m vo
r* r*
d d
S3
?9
ss
O O
V V
§8
O O
V
ss
o o
V
in m
00
o o
V V
3d
d d
3d
do
! -g
g-|^. -^|
l$* 11
3*S •a § a
a-«j i
. ^
d d
SS
o o
™"
o o
sa
i-Id
d d
CMH
in in
dd
rH rH
r- r»
d d
FjP
d d
ss
o o"
V V
m in
O 0
o o
V V
tn in
o o
0 0
V V
in m
00
o o
V V
ss
o o
V V
S3
o o
o en
rH O
O 0
'•! ^
3lf al
J|I jl
J1 § fi S a
o-e
Sin ^
. . ^
odd
SS S
odd
or- gj
o o o
rtS S
H H H
Sr>. o
en o
O* O* H
CM CM CM
in in co
C3 0 CD
CM O ^
r» r> co
CD O O
gg a
O O H
tn in o
o o tn
o o o
V V
in tn en
o o ro
0* 0 O
V V
in m o
o o
-------
1!
.s
§
I
n
A
1
1
m
rH
3
i~i
CO
fH
C»J
r-l
H
O
rH
CT\ tp
•S
Si
oo 23
i
p- "5
rH
vo l ^ * -3
B J3 O ^ .jj) H
6 l^S ||
"A §82 ffS
8 W
ss
0 O
^j* in
O 0
o o
SS
o e>
gg
O 0
tn vo
°, °
o o
r-- r-
0 0
o o
in t^»
o o
o o
gg
o o
•* r*
o o
O C5
gg
0 0
as
O 0
r- iH
C5 rH
op
SB
o o'
m us
rH rH
0 O
H 01
H rH
O O
B
4
c el
Q i™ 1
1 ^ "
'
to v
0 O
o o
in -^
°. °.
o o
us in
o c>
.-ICO
r~ o
cs o
§§
o o
GO t^
00
O O
Q3 CO
O O
• o c>
S^f
o
0 0
ss
o o"
<4* 00
o o
o o
ss
o o
ss
o o
p^ VD
O 0
o o
in p*
rH rH
0 O
as
o o
i . S
1 -a ><
%'%- £%
3lti til
ill 3,i
=r)S I 8 3
S8 3
o o o
qi n 10
0 O CM
0 O 0
ss s
C3 O O
co o> r-
0 C3 01
o o o
\O *£> U3
O O (-(
o o o
tiO r** M
o o n
o o o
0 OJ CNJ
i-i O i-H
o o o
art s
o o o
-* irt •*
O O r-l
O O O
SS S
000
t* CT\ rH
p o ro
o o o
00 CO O%
o c» o
o o o
gg g
C5 O 0
•* «j- m
rH rH n
o d o
iH ^* M
rH rH (N
odd
! ,
4J E?
I4_j ^— * rrj ^ c
Sag™ E^
rH 0) W &H
JrJ H Q |U 3
d 8 a "§ ^ S
S S
00
s
r~
t~
co
a
in
•
3
a
33
°.§
^r e>)
s
85
-------
Review of Table 9-3 also indicates that pH adjustment
achieves a relatively high removal of the initial dissolved lead
concentration. However, the lead hydroxide precipitate formed by
the lime treatment process did not settle well in some cases.
Treatability tests 6, 7, 8, 9, 10, 12, and 15 indicated
relatively high concentrations of suspended lead hydroxide after
lime treatment. In view of these high concentrations of lead
hydroxide before sulfide treatment, it has been determined that
the experimental dose of ferrous sulfide proved insufficient to
meet the sulfide demand of suspended lead hydroxide for runs 8,
9, and 10 and of nickel hydroxide for those tests in which the
total nickel concentration in the supernatant after lime addition
is greater than 0.53 mg/1. Since lead is preferentially
precipitated as lead sulfide, there was not an adequate
concentration of free sulfide in most cases to appreciably remove
the nickel. Therefore, better metal removal efficiencies can be
obtained by either improving the metal hydroxide removal prior to
sulfide treatment or by increasing the sulfide dosage to satisfy
the suspended metal hydroxide demand. In addition to this
observation, evidence appears in Table 9-3 to support the
conclusion that insufficient contact time was provided in the
sulfide reaction prior to discharge. Review of the dissolved
lead _ concentrations before and after fil'tration indicates a
consistent improvement in the lead removal. Since this
improvement cannot be associated with a physical separation
process, it may be concluded that reaction of dissolved lead with
unreacted ferrous sulfide continued during the filtration step.
Dual media filtration was observed to achieve a high removal
of the lead sulfide indicating the poor settling characteristics
of the metal sulfides and the need for filtration to achieve a
physical removal.
Review of the suspended solids results indicates that the 15
to 20 minute settling time after lime treatment failed to provide
a good removal of TSS. Percent removal efficiencies after lime
treatment and settling periods up to 15 hours were attempted
without improvement in the clarified supernatant characteristics.
Filtration substantially improved TSS concentrations but failed
to provide low levels on a consistent basis.
9.3.2 Statistical Evaluation
Table 9-4 shows a comparison between the proposed BAT
maximum 30-day average and the estimated treatability 30-day
average. Figures 9-2 through 9-5 and Appendix A present the
statistical analysis.
86
-------
TABLE 9-4. COMPARISON BETWEEN PROPOSED BAT LIMITATIONS AND
ESTIMATED TREATABILITY PERFORMANCE FOR THE
CHLOR-ALKALI SUBCATEGORY
STREAM: Filter Effluent
Pollutant
Concentration Basis
(rag/1)
Proposed BAT
Maximum
Maximum 30-Day Average
Est. Treat. Performance
30-Day Average
Lead
Chromium
Nickel
Total Suspended Solids, TSS
0.22
0.05
0.10
12
0.085
0.087
0.66
51
87
-------
SUBCATEGORY
Chlor-AUcali
POLLUTANT
Lead
PRECIPITANT
Proposed Maximum 30-day Average
(mg/1): 0.22
95th Percentile (Z = 1.64) (mg/1):
long Term Average (mg/1):
Standard Deviation of 30-day Averages (irg/1):
Prdbability of Achieving Proposed
Maximum 30-day Average (%) •
Nurfoer of Observations:
0.085
0.073
0.0075
>;99
10
1.0
0.9
4J
-------
SUBCATEGORY
Chlor-ALkali
POLLUTANT
Chromium
Proposed Maxiimm 30-day Average
(mg/1)
95th Peroentile (Z - 1.64) (mg/D
Dong Term Average (mg/1)
Standard Deviation of 30-day Averages (mg/1)
Probability of Achieving Proposed
Maximum 30-day Average (%)
Number of Observations:
PRECIPITANT
0.05
••
0.087
0.077
0.0059
<0.01
15
1.0'
0.0
0 065 0.070 0.075 0.080 • 0.085
Maximum 30-day Average (mg/1)
Figure 9-3. Estimated Performance of Proposed
BKT Treatment
89
0.090
-------
SUBCATEGORY
POLLUTANT
PRECIPITANT
Chlor-Mkali
Proposed Maximum 30-day Average
Nickel
(mg/1): 0.10
95th Percent-He (Z = 1.64) (mg/i):
Long Term Average (mg/1):
Standard Deviation of 30-day Averages (mg/1):
Probability of Achieving Proposed
Maximum 30-day Average (%):
Nxzrfoer of Observations:
0.66
0.43
0.14
0;99
15
1.0
>,! °-9
/r* cl
R| 0.8
II 0.7
ti$
*« 0.6
&$
3j§ 0.5
w
3 ^ 04
R fl 07
i S1 0.2
3 jj)
^01
n n
*•*•
**
•^
^
^
/
/
/
—j.
/
/
^
0.51
/
,/
/*
0
S
.6
0
X
0
-*
-.7
0.
Maxinttin 30-day Average (mg/1)
Figure 9-4. Estimated Performance of Proposed
BflT Treatment
90
-------
SUBCATEGORY
POLLUTANT
PRECIPITANT
Chlor-Alkali
Total Suspended Solids
Proposed MaxLmun 30-day Average
(mg/1): 12
95th Percentile (Z = 1.64) (mg/1):
Long Term Average (rag/1):
Standard Deviation of 30-day Averages (mg/1):
Probability of Achieving Proposed
Maximum 30-day Average (%)'
Number of Observations:
51
35
9.9
1.1
15
1.0
>, I °-9
"f'l
° 1 °-8
t§
1 °-7
+> C5
1*. 0.6
Estimated Probability
Average Does Not Excee
0 O O O O O
..«•••
o H to co *» tn
•H
•Hi
]
•••
!5
••a
***
••t
^H
X
D
4
S
/
25
/
/
^
•^
/
^
/
,/
^
>
/
,/
V 5 0
_/
X
X*
*^«
Maximum 30-day Average (mg/1)
Figure 9-5. Estimated Performance of Proposed
BAT Treatment
91
-------
Table 9-4 and Figure 9-2 show the results that are obtained
after screening of the analytical data for lead data outliers.
Test numbers 8f 9, and 10 have been rejected on a technical basis
and runs 1 and 4 on statistical grounds.
9.3.3 Conclusions
The treatability study conducted for the diaphragm cell
segment of the Chlor-Alkali Subcategory did not provide an
adequate basis for evaluating the performance of the proposed BAT
treatment. The extremely poor settling characteristics of the
metal hydroxide sludge formed during lime treatment precluded any
meaningful testing of the subsequent treatment steps involving
sulfide precipitation and filtration. The hydroxide settling
problem was not identified in time to attempt corrective measures
that ^would bring the results more in line with the known
practical capabilities of the treatment technology. The use of
coagulating agents in conjunction with adequate mixing and longer
settling periods may provide considerably improved separation.
The technique of recycling a portion of the sludge to "seed" each
batch may also prove to be helpful. Although lime was selected
as the source of alkalinity, largely because of its low cost, a
serious disadvantage in its use may have been incurred in this
case due to the presence of sulfate that was introduced when the
PH of the raw waste was lowered with sulfuric acid. A mixed
precipitate of negatively charged particles of CaS04.2H20 and
metal hydroxides may well have interfered with the
coagulation/settling process. Substituting soda ash or caustic
soda for lime would circumvent this potential source of
interference with the sludge settling process.
Thus, the experimental results presented for this
subcategory represent the outcome of the particular set of
experiments that were conducted during the relatively short time
frame available. They do not represent the actual performance
capabilities of the proposed BAT treatment. This work should be
taken as a starting point for the design of a more comprehensive
series of tests on this technology.
92
-------
SECTION 10.0
TITANIUM DIOXIDE SUBCATEGORY (CHLORIDE PROCESS)
on-site at the
1979. A total of
10.1 INTRODUCTION
10.1.1 General Considerations
The treatability tests were performed
selected plant from October 16 to October 27,
16 test runs were completed in this period.
Due to time limitations, waste water was always collected in
large enough composite samples to run two tests at the same time.
All samples generated by both treatability units were split with
the plant personnel in addition to the sample collected for
initial waste characterization.
10.1.2 Sampling Point Locations
The manufacture of titanium dioxide from rutile ore by the
chloride process produces several waste streams. At the selected
facility, the four main waste water streams go to sumps where
they commingle and are pumped to the treatment facility. These
waste sources and flows are tabulated below:
TREATABILITY WASTE SAMPLE COMPOSITION
Waste
Source
Plow
(gal/min)
Percent of
Total Flow
Estimated Treatability
Sample Composition By
. Sources (Percent)
1) Chemical sump, 200
oxidation step
and area runoff
2) Filter sump, 375
filtration and
acid wash plus
spills
18
33
20
35
93
-------
3) Boiler and cool- 275
ing tower sump
4) First reactor,
chlorination
step and scrub-
ber blowdown
280
24
25
25
20
TOTAL
1130
100
100
rector" is, in effect, a sump where the
water mixes with the chlorine recovery scrubber
(caustic) blowdown. This reactor is located next to the liming
mixer or reactor which is the reactor where the aqueous wastes
are collected for treatment.
CmT,tv,Qi« is n° °ne place to sa"»Ple the combined waste, a
composite sample was put together from the four principal waste
*!^h «= PS aS Sh°Wn in Figure 1°-1- The amount of water from
each source composited for the test sample on the first day was
im,r™ Y eXJCt Pr°P°rtion to the flows from the different
sources, as shown in the table above. However, difficulties
a*ne! *n-,.!eP*ng solids uniformly suspended (especially the very
dense solids from the first reactor) while transferring less than
SamSe c°ntainer. Consequently, volumes of waste water
!er aPP°rtioned so that full containers were used for
trftability units. This resulted in the proportions
mixrew J C°1Umn °f the above table- As n°te<3, the
mixture was a good approximation of the composite raw waste from
Q >X •!• o \j \JiL w S O •
10.2 TREATABILITY TEST MODEL OPERATION
10-2.1 Treatment Technology Tested
10.2.2 Waste Water Characterization
TableR10-ltS 2«
Table 10-1. An
?he Se2lirSS
the metals present
Water Characterization are shown in
series of laboratory tests were carried
t °f increasin9 P« °n the solubility oi
the raw waste. The results of these tests
94
-------
CHLORINATION
OXIDATION
AREA RUNOFF
FINISHING
UTILITIES
AREA AREA
RUNOFF RUNOFF
1. CHEMICAL SUMP
2. BOILER AND COOLING TOWER SLOWDOWN
3. FILTRATES
4. FIRST REACTOR
Figure 10-1. Sources of waste samples for the
titanium dioxide subcategory
(chloride-process)
95
-------
TABLE 10-1. WASTE WATER CHARACTERIZATION FDR THE PLANT
SELECTED FOR STUDY IN THE TITANIUM DIOXIDE
SUBCATEGORY (CHLORIDE PROCESS)
Constituent
Concentration (mg/1)
pH*
Calcium (as CaC03):
Chromium (Hexavalent):
Chromium:
Copper:
Iron:
Potassium:
Magnesium (as CaC03):
Sodium:
Nickel:
Zinc:
Total Suspended Solids:
Total Residual:
Fixed Residue:
Total Dissolved Solids:
Methyl Orange Acidity (as CaC03)
Total Hardness (as CaC03):
Chloride:
Nitrate:
4.0-6.5
4,250
0.025
50;. 0
0.11
136
11.6
34.8
292
1.29
0.69
381
10,930
7,620
7,134
600
4,300
390
0.41
*As indicated in Table 10-3, sixteen samples of the raw waste water
showed pH values ranging between 4.0 and 6.5.
96
-------
are presented in Table 10-2. Note that the metallic species in
the waste show the lowest overall solubility at pH values of 9.8
and above, and so 9.8 was chosen for the treatability test.
10.2.3 Details on Treatability Test Operation
The precipitant used in all the tests was agricultural grade
hydrated lime. This material was always fed to the reaction tank
as a powder.
After the initial addition of lime, the pH of the waste
water dropped to 8.5 and 9.5 for runs 1 and 2 respectively. In
order to bring the pH back up to more favorable levels for metal
removal, additional amounts of lime were added. These amounts
were 1.25 and 2.55 grams, respectively. The final pH\values
obtained were 9.05 and 10 .as indicated in Table 10-3.
Recirculation of the filtrate back to the reaction tank was
practiced for 15 minutes for all of the runs. Values reported
for the filtration time do not include this 15 minute
recirculation period.
Table 10-3 is a tabulation of operation parameters and
observations for the sixteen treatability test runs performed for
the Titanium Dioxide Subcategory.
10.3 TEST RESULTS
10.3.1 Discussion of Results
The pollutants studied in this subcategory included iron,
chromium (T), zinc, nickel and total suspended solids (TSS). The
analytical results for these parameters are presented in Table
10-4.
A review of the analytical results presented in this table
indicates that it is possible to meet the proposed limitations
for the pollutants under consideration by application of the BAT
level of treatment proposed for the Titanium Dioxide Subcategory.
As noted in Table 10-4, the concentrations of zinc in the
filtrate stream increased as the supernatant passed through the
filter. Contamination of the samples seems to be the most likely
explanation because the increase in concentration occurred for
both the dissolved and undissolved portions of that metal. In
other words, the increase in the dissolved zinc did not occur at
the expense of the total zinc. Sample contamination from the
filter media may have been favored at the low pH values reported
for the final effluent as indicated in Table 10-3.
97
-------
TABLE 10-2. EFFECT OF pH ON TOXIC METAL SULUBILITY
(All values in mg/1 except pH)
pH
7.0
7.5
8.0
8.5
9.0
9.5
9.8
10.0
11.0
12.0
————————__
Chromium
0.02
0.01
0.03
0.02
0.03
0.02
0.02
0.02
0.05
0.03
Copper
0.04
0.04
0.04
0.09
0.03
0.04
0.03
0.04
0.03
0.03
=======
Iron
0.08
0.09
0.08
0.07
0.07
0.09
0.08
0.07
0.07
0.07
Nickel
0.85
0.65
0.47
0.25
0.19
0.13
0.14
0.1.2
0.12
0.10
Zinc
0.028
0.011
0.021
0.011
0.038
0.030
0.021
0.014
0.014
0.011
Hexavalent
Chromium
<0.004
<0.004
0.004
0.004
0.004
0.004
<0.004
0.008
0.004
<0.004
98
-------
W
z
o
6-1
H
O
u
a
X
EH
H
H
1
9
2
m
I
o
rH
TABLE
nf
•rl
0)
rH
(fl
3
Q fi
CO O
3 -H
rH-P
&4
rn" S * R
0) )H
•ggf | of
nj IS'ffi gg. B"S
^ O 3- "^ ra CU £5 C3
Q t> CM i5
CO
en
CO
en
en
en
VO
en
en
oC
CO
en
CO
CO
CO
en
CO
en
CO
en
o
s
CO
on
CO
ot
o
s
0
ol
*j
I-
S j
53 *•
&
in
rH
a
m
rH
a
m
rH
in
rH
m
rH
a
in
rH
in
rH
m
rH
in
rH
m
rH
in
rH
m
CM
s •
1 »a
1 'i'i
3 •£{ *" *
2!
3
o
o
CM
rH
S
H
8
H
O
CM
rH
O
CM
rH
rH
CM
1-1
S
H
8
rH
S
c\i
H
rH
rH
1
EH
»
•H w
B'i
fli ^w-
UJ
^
CO
VO
09
CM
en
rH
en
_,
en
a\
CO
CT\
CO
en
CO
o
en
rH
en
H
en
o
en
rH
en
rH
en
in
CO
in
CO
I
1
s
S
,_ 'a
s
99
-------
ss
• •
oo
as
HH
S3
d d
as
00
0§
d0'
rM in
Ho
do
o in
do
in
• •
oo
S3
oo
si
do
t-CM
in a\
oo co
• •
00
^1 \Q
do
SS
• •
o o
o in
3$
gs
r*> co
CM CM
O O
!3
) O
°°
o d
o o
Pfl
\Q
d d
is
O O
en PI
ss
do
§8
d d
in m
H H
SS
do
CO (N
00
SS
o o
• •
o o
si
do
ss
do
S\o
o
d d
SS
00
m m
S3
CM \O
00
°
ss
do
is
o o
CM O
So1
o o
d d
00
do
o o
s°'
o o
en o
10 ro
n o
CM CM
So
do
CM CM
o o
do
in \o
do
r>) in
So
o
o o
Sef
dd
g
> o
!S
oo
d d
ID o
m ID
o o
oStn
d d
S3
d d
g§
do
So
o o
SS
dd
SCO
o
do
H o
• •
oo
00
do
o o
dd
> n
iS
en o
en to
d d
o in
vo co
ocn
HO
en CM Tf CM
§i 33
ss
o d
Sen
o
• •
o o
en in
in in
in r-4
co ro
en en
o o
do
§g
do
o o
d d
'P.
do
83
O O
o o
o d
SS
• •
O O
!§ Si
o o
d d
ss
d d
ss
do
100
-------
3
a.
3
a
rH
a
s
en *
5
-1
-I
VO
in
•w
m
CM
rH
a
1
r* o
53
PS
rH rH
8VO
M
m m
33
n in
m r-
in m
C4 H
23
S3
CM
rH O
VO *3*
rH CO
0 0
00 00
r- cn
i§
a
Sd
voR
3d
VO
S3
o in
cn rH
| £ Jj
•13 S
M
si
do
o cn
cn in
m o
d d
CM in
cn o •
d d
rH in
CM O
d d
§§
do
d d
p* cn
d d
58
O O
in o
o o
o o
S3
d d
coin
CO rH
d d
£3
d d
rH «H
d d
rH VO
CM 0
d d
O CM
CM rH
d d
i ?>"$
rH
ill
m vo
38
d d
o cn
jn-H
m rH
o o
Ct CO
vo m
0 O
3§
C3 O
O rH
-------
Excellent removal of total chromium was observed for most of
the tests. Since Cr+6 does not precipitate by pH adjustment, the
removal observed can be explained only if the chromium was
present in a reduced form, that is, as Cr+3.
The analytical results for total chromium show poor removal
for runs 3 and 4 as indicated in Table 10-4. The high values
reported may be the result of operational problems in the
titanium dioxide plant which led to the production of an atypical
waste. The chlorine recovery refrigeration compressor broke down
and the excess chlorine was pumped into the waste system. Also,
oil from the broken compressor was drained off into the chemical
waste sump during sampling. It will be noted that the
concentrations of chromium was exceptionally high in the raw
waste on that day. Furthermore, at pH levels at which trivalent
chromium would be quite insoluble, the analyses of supernatant
and filtrate show both total and dissolved chromium at similar
high concentrations. It is concluded that the chromium was in
the hexavalent form. The treatment applied was not designed to
remove chromate, and the capability of the method to yield low-
chromium concentrations should not be judged on the basis of the
abnormal waste water containing chromate. The chromate probably
came from the cooling tower, since as far as is known, it does
not arise in the processes of titanium dioxide production.
Therefore, chromium results for test runs 3 and 4 were rejected
from the statistical analysis.
Even though the dual media filter was effective in reducing
TSS from the clarified waste, its benefits are only marginal as
can be noted in Table 10-4.
10.3.2 Statistical Evaluation
Table 10-5 and Figures 10-2 through 10-6 show the results
obtained from a statistical analysis of the treatment data. It
should be pointed out that Figure 10-3 presents the probability
performance for total chromium after screening the data and
rejecting runs 3 and 4 for the reasons stated above.
10.3.3 Conclusions
The treatability test results provide a good basis for
assessing the general applicability of the proposed BAT
regulations for the chloride process segment of the Titanium
Dioxide Subcategory. Results show that the pollutant
concentration basis for the proposed BAT maximum 30-day average
effluent limitations are achievable with the prescribed treatment
technology under the conditions of these tests.
102
-------
TABLE 10-5. COMPARISON BETWEEN PEC-POSED BAT LIMITATIONS AND
ESTIMATED TREATABILI'IY PERFORMANCE FOR THE TITANIUM
DIOXIDE (CHLORIDE PROCESS) SUBCATEGORY
STREAM: Filter Effluent
Pollutant
Concentration Basis
(itg/1)
Proposed BAT
Maximum
30-Day Average
Est. Treat. Performance
30-Day Average
Iron
Chromium
Nickel
Zinc
Total Suspen
2.5
0.14
0.20
0.50
.ded Solids, TSS 64
0.21
0.051
0.093
0.047
12.0
103
-------
.SUBCATEGORY
Titanium Dioxide
POLLUTANT
Iron
PRECIPITANT.
Proposed Maximum 30-day Average
(mg/1)
2.5
95th Percentile (Z » 1.64) (mg/i)
Long Term Average (ng/1)
Standard Deviation of 30-day Averages (ng/1)
Probability of Achieving Proposed
Maximum 30-day Average
Nurber of Observations:
0.21
0.17
0.023
>99
16
1.0
0.4
0.0
0712
0.14
0.1
D.I
0.20.
0.22
Maxiimm 30-day Average (rag/1)
FJ.gure 10-2. Estiiated Performance of Proposed
BKT Treatment
104
-------
.SUBCATEGORY
Titanium Dioxide
POLLUTANT
Chromium
PRECIPITANT.
Proposed Maximum 30-day Average
(mg/1) : 0.14
95th Peroentile (Z = 1.64) (mg/1) j
Long Term Average (mg/1) :
Standard Deviation of 30-day Averages (mg/1) '
Probability of Achieving Proposed
Maximum 30-day Average
Number of Observations;
(%) :
0.051
0 . 046
0.0030
>gg
14
%
•?
3
f
<
«s
^
£*
r"
•f
^
f
]
I
t!
H
1.01
g*t .
- 0.9
g
1 n o
1 0.8
^ °*7
0) _ _
•5
G
SB O-4
8
Q O*3
g 0.2
•^ n i
0.1
0.0
*•-*-
-^"
rt
X
/\
^
4 n
,/
X1
/
f\
/
r\A A
J
/
J
s
/
'
A 1
/
•\AC
/
/
/
1
/
r\ n
f
dO
/
/
1
MB*
^n»
•» n
s>
rrr>
___„
n
nc'
j
Maximum 30-day Average (mg/1)
Figure 10-3. Estimated Performance of Proposed
BAT Treatment
105
-------
SUBCATEGORY
Titanium Dioxide
POLLUTANT
Nickel
PRECIPITANT
Proposed Maximum 30-day Average (mg/1) : o . 20
95th Percentile (Z = 1.64) (mg/1) : 0.093
Long Term Average (mg/1): 0.088
Standard Deviation of 30-day Averages (mg/1): 0.0028
Probability of Achieving Proposed
Maximum 30-day Average (%): >99
Nunber of Observations: 16
1.0
>ii °*9
^ S 0.8
ts
•S ^*^
^rrt °-6
ted Prcbabilitj
Does Not Excee
000
• • •
co ifc» in
Jj S* 0.2
03 ft)
"« 0.1
I**1*
***
^f
s
/
/
/
'
/
'
/
"
/
A
/
^•l* 0.082 0.084 0.086 0.0
Maximum 30-day AT
J
A
/
j
f
/
/
/
/
j
r
S
^ *•
•«*•
•••
38 0.090 0.092 0.094
/erage (mg/1)
Figiare 10-4. Estimated Performance of Proposed
BAT Treatment
106
-------
.SUBCATEGORY
Titanium Dioxide
POLLUTANT
Zinc
PRECIPITANT.
Proposed Maximun 30-day Average
(mg/1)
0.50
95th Percentile (Z = 1.64) (mg/1): 0.047
Long Term Average (mg/1): 0.041
Standard Deviation of 30-day Averages (mg/1): 0.0035
Probability of Achieving Proposed
Maximum 30-day Average (%): j>99
Number of Observations: 16
1.0
0.9
0.8
0.7
0.6
0.5
0.4
.3
00
.£
.1
/
/
jj
S'
/
/
/
/
j
/
/
/
/
f
/
f
/
F
/
/
S
/
S
'
^~
^*~
0**
•*~
-—
•^•""
°*
-------
.SUBCATEGORY
Titanium Dioxide
POLLUTANT
Total Suspended Solids
PRECIPITANT.
Proposed Maximum 30-day Average
(mg/1): 64
95th Percentile (Z = 1.64) (mg/1) i 12
Long Term Average (mg/1): 10
Standard Deviation of 30-day Averages (mg/1): 1.2
Probability of Achieving Proposed
Maximum 30-day Average (%): >99
Number of Observations: 16
1.0
I^t 3
fa §
i *R
°I o-8
4i O
f? * 0.6
gj 0.5
I*
75 «2 0.4
JOfl 0.3
43*-1
^J w O.^J
W i*
0.1
n n
JM«^
»-•
^-"
•^
^
•X"
X
^/
it
/
f
t
/
/
f
/
/
/
f
/
/
/
f
/
f
s*
f*^
-*•
^
1 0
8.0 9.0 . 10.0 11.0
Maximum 30-day Average (mg/1)
12.0
13.0
Figure 10-6. Estimated Performance of Proposed
BAT Treatment
108
-------
The results further indicate that depending on specific
plant raw waste characteristics, compliance with the proposed BAT
regulations may be achievable without the filtration process.
However, elimination ofjthe filtration process would require that
the alkaline precipitation of metals is carefully controlled at
the optimum pH and that sufficient settling time is given. In
addition, the proper operating conditions would have to be
established on a plant by plant basis giving careful
consideration to variability in raw waste characteristics.
109
-------
-------
SECTION 11.0
CHROME PIGMENTS SUBCATEGORY
11.1 INTRODUCTION
11.1.1 General Considerations
The treatability tests were carried out on-site at Plant
#894. A total of fourteen runs were completed from September 5
to September 24, 1979. Waste water samples were collected daily
and used for each run. Both raw and treated waste samples were
split with the plant personnel.
11.1.2 Sampie Point Location
The selected plant is currently practicing BPT treatment as
shown in Figure 11-1. Waste samples used in the treatability
study for the proposed BAT level treatment were collected at the
point of clarified effluent discharge. The sample point location
is indicated in Figure 11-1 for the selected plant.
11.2 TREATABILITY TEST MODEL OPERATION
11.2.1 Treatment Technology Tested
The treatment technology evaluated in the treatability tests
on chrome pigments consisted of additional metal precipitation as
metal sulfides. After settling and removal of the sulfide
sludge, the clarified waste was polished by means of a dual media
filter.
11.2.2 Waste Water Characterization
Results of the initial raw waste characterization are shown
in Table 11-1. Following the waste characterization,
treatability experiments were performed by adding 1.0 to 3.0
111
-------
n i
§!
112
-------
TABLE 11-1.
CHARACTERIZATION OP RAW WASTE WATER FROM THE
CHROME PIGMENT SUBCATEGORY
========
Specific Waste Constituent (mg/1)
Cadmium Chromium Iron Lead Zinc
Total
Dissolved
0.04
0.04
0.23
0.10
0.32
0.18
0.27
0.27
0.006
0.006
TABLE 11-2.
EFFECTS OF ADDITION OF FERROUS SULFIDE TO THE
CHROME PIGMENTS WASTE WATER
Amount Sulfide Added
(Percent Stoichio-
metric Requirement)*
Remaining Metal Concentration in
Solution (mg/1)
Cadmium Chromium
Lead
Zinc
0
50
75
100
125
150
200
400
0.04
0.05
0.04
0.04
0.04
0.04
0.04
0.04
0.10
0.07
0.07
0.08
0.08
0.08
0.08
0.07
0.27
0.20
0.20
0.20
0.20
0.20
0.20
0.20
0.006
0.110
0.040
0.030
0.030
0.020
0.050
0.020
* Fourteen treatability tests performed later showed that
the pH of the raw waste varied from 7.93 to 8.60.
113
-------
times the stoichiometric amounts of ferrous sulfide required for
precipitation of the metals present in the waste. This was
necessary to determine the ferrous sulfide dosage for optimum
removal of the metals that would have to be used during the test
runs. However, all of the added material was found to dissolve
and no precipitate formation could be observed. Chemical
analyses of the resulting treated waste water solutions showed
negligible reductions in the cadmium, chromium, and lead levels
as seen in Table 11-2. To overcome this problem, the procedure
was then modified so that the amount of ferrous sulfide added in
each treatability test was twice that required to form a
saturated solution (2) of this material. This resulted in a
solid phase still being present after initial dissolution of some
of the added ferrous sulfide. This condition was assumed to be
optimum for metal removal and was utilized throughout the
experiments.
11.2.3 Details on Treatability Test Operation
As shown in Figure 11-1, the waste water used in the tests
was not collected from any of the raw waste streams being
generated by the plant. Instead, the waste utilized came from
the clarified effluent. It was reasoned that the plant selected
for study was already practicing BPT and that there was no point
in repeating observations on this level of treatment.
A constant amount of ferrous sulfide reagent (260 ml) was
added to the waste during each of the test runs. The reagent
contained 2,800 mg/1 of ferrous sulfide.
Total recirculation of the filtrate back to the
reaction/settling tank was practiced for all the runs with the
exception of run 14. Recycle time varied from 45 to 90 minutes.
This provided seasoning of the filter bed prior to collection of
the waste sample for analysis and ensured efficient removal of
TSS from the supernatant.
Table 11-3 is a tabulation of the operational parameters and
observations made during the fourteen treatability test runs.
Review of the table reveals a relatively consistent operation of
the treatability test apparatus providing a very uniform effluent
quality.
(2) Solubility for FeS from the Handbook of Chemistry and
Physics, 51st Edition, Robert C. Weast, Ed. The
Chemical Rubber Company, 1970-1971, pg. B99
114
-------
d
0
•H
4J
•P
rH
•H
fa
.2
I
§
rH
«
•H
ra
„
1
1
&5
en
|
Q
EH
W
w
t<
B
J
1
1
g
tn
i
rH
rH
g
1
4J
I
•H
.a
o
S
S
W
rH
rH
CM
rH
rH
S
CO
*"•
vo
in
<»
.-
CM
H
H H
H
\ o • PI in
Cn CN CO rH rH
CN VO
\ o • m in
cn CM co H H
cn
X o • cn in
cn M co H H
CO
X 0 * «n in
Cn CM CO rH rH
^
\ O • ' M m
Cn CM CO rH rH
*r
3 *H
X o • m in
Cn CM CO rH H
rH CM
\ o •
S S *7 * ^ ~ S
H W 3 C S
U-l frt C 4) WO -H
O O -M -H &*
i< «H 01 m -P
-------
11.3 TEST RESULTS
11.3.1 Discussion of Results
Analytical results for the treatability test runs are
tabulated in Table 11-4 for the major pollutant parameters. This
table summarizes results for cadmium, chromium, lead, zinc, and
TSS. Review of Table 11-4 indicates very low initial
concentrations of the pollutants in the waste water samples. The
results indicate the importance of dual media filtration to
achieve further reduction of metal concentrations when initial
concentrations are too low for effective settling. Observation
of the experiments showed that formation of a sulfide sludge
blanket did not occur. This may be correlated with observations
of poor metal removal efficiencies in the clarification step.
The initial metal concentrations were consistent throughout
the period of the treatability investigations. Toxic metal
levels in the treated effluent also appear at consistent levels.
11.3.2 Statistical Evaluation
A comparison between the proposed BAT limitations, which
were developed on the basis of a 95 percent likelihood compliance
by any industry, and the estimated treatability performance is
presented in Table 11-5. The statistical analysis for each of
the pollutants is presented in Figures 11-2 through 11-6 and
Appendix A.
11.3.3 Conclusions
The treatability tests conducted to evaluate the proposed
BAT treatment for the Chrome Pigments Subcategory are
inconclusive as to the practical viability of sulfide
precipitation for the removal of additional toxic metals
following the BPT alkaline precipitation step. Due to a highly
efficient BPT system at the plant studied, the initial toxic
metal concentrations (BPT clarified effluent) were already near
the lower limit of removal obtainable with sulfide precipitation.
The results indicate that the major application of sulfide
precipitation and/or dual media filtration would be in specific
plant situations where less efficient BPT systems require
additional treatment to meet the BAT regulations.
116
-------
§
533
Ba
M
^ «
§3
1*
*§
•H Q
>* *O
W id rH
rH rH
PS 3 g
Q "
i is
J3 K, p
™ i i
II **
«l
«1
W H
H 0.
I 1
[J (J
H EC
EH EH
IS
.
4,
rH
§
^,
rH
en
H
CN
"*
H
0
rH
rH
cn g
c
•H
S
CO S
'ec
•s
r» s.
r
-
* *
in
^r
CN
rH
st Number
e
CO rH
O O
CO rH
^T rH
d d
in \o
d d
O rH
fO CN
d d
CN rH
O 0
in TT
d d
in CN
o o
in m
o o
§ Jn
o o
as
r-l rH
O O
t» en
d d
to r*
tr> co
en CN
d d
la
d d
CM CM
en I-H
d d
" S
1 ! g"
4J 01
rH S O -H
ffl « frl O
g
r» rH
en rH
d d
to tn
en rt
d d
rH rH
d d
0 0
PI rH
O O
03 O
CM i-H
d d
m m
•H O
d d
sa
d d
in rH
d d
rH Q
in CN
d d
85
rH O
ss
d d
r* rH
CO rH
CM O
o o
vo tn
CO CO
en o
d d
r* en
tn co
H O
d d
S
i?"
pernatant
Total Cr
Dissolvec
en
CN CM
oo
oo
•am
oo
o o
en eN
O o
<3 0
CM ^*
d d
Sen
0
d d
00
d d
o o
o o
m tn
O 0
a d
in in
o o
CO CN
rH CM
d d
rH C*
OO
O O
O rH
CM CM
O 0
d d
O eg
en rH
o o
do
rH rH
0 0
do
g
gg^
iH rl CD
CQ
fl> O *H •*
rH «
b 3
in m
rH 0
d d
co en
in CM
d d
83
d d
tn rH
d d
(N H
d d
tn tn
CM O
o o
CO 00
rH rH
d d
co tn
d d
r- r*-
rH H
d d
ss
o o
§8
d p
rH O
** en
CM rH
O O
as
d d
as'
rH rH
O O
S
.w Waste
Total Pb
Dissolved
s.
rH O
d d
r- to
«* CM
o o
en o
CM CM
d d
in CN
d d
O CM
m H
d d
as
d d
CM •«•
CN CM
d d
O CM
fn rH
d d
3S
d d
en CM
d d
o
qs
d d
CM O
d d
en tn
d d
-
0 O
d d
•P
§ s
rH tJ 13
rl Ot Ul §
0) -P W -H
*J O -tH S
H EH Q |
fi 0
H H
O O
d d
m ^r
o o
d d
m **
o o
d d
o o
d d
vo vo
o o
d d
CM (N
0 O
d d
o o
d d
tn --a1
O O
d d
vo m
o o
d d
vo in
rH rH
d d
S3
d d
en ^«
0 0
rH rH
O O
0 0
rH rH
O O
en en
o o
o o
s
iw Waste
Total Cd
Dissolved
&
rH H
O O
O O
0 O
d d
S3
O O
r* 10
' O O
d d
in T
0 0
o o
CN rH
0 O
d d
tn •*
o o
d d
in T
o o
d d
0 0
o o
m CM
rH H
d d
0 0
o o
O CM .
rH 0
d d
en vo
o o
d d
en ^
m m
o o
d d
s
ipernatant
Total Cd
Dissolved
s
CQ
rH H
0 0
°'
-------
•o
§
5
§
i
rH
I
f*H
ro
N
r-l
r-l
O
r-|
cn \
1
c
-H
10
CO 0)
5
*o
a
*• ti
a
s
r-i
10 <
tn
CO
r-f
14
Z
.u
DI
to in
o o
0 0
°9
o to
ro H
O O
d d
tn ^j1
r-l H
O in
o o
o o
CM CM
o o
d d
CM CM
0 0
d d
("1 rH
0 0
d d
01 cn
CM r-l
0 0
o o
CO CO
CM CM
o o
o o
in to
CO CM
o o
o o
£
•o
4J i-t
B rJ O
c a a in
N S JJ 01
o ni
CM 00
r-l O
o o
o d
t^ to
to to
0 0
d d
r~ CM
0 0
d d
Sen
to
0 0
d d
1- <0
to in
o o
d d
CM r-<
in I-H
in o
o o
CO CO
0 0
d d
53- CO
O O
d d
CM CM
O O
d d
in CM
0 0
d d
i-l CM
0 0
d d
cn to
in r~
o o
d d
in in
ro ro
o o
o o
r-l i^
to to
o o
d d
£
4J *rt
Supernatan
Total Zn
Dissolve
r- in
o o
o o
^p cn
o o
d d
•* 0
in to
o o
CM cn
t-- to
o o
o o
co en
in to
0 0
d d
r^ CM
0 0
0 0
in in
0 0
d d
in r»
0 0
d d
ro CO
o o
d d
to to
o o
o o
ro co
0 0
o d
co r~
in in
0 0
d d
o cn
CO ^
o o
d d
cn to
CM CO
0 0
d d
4J
i— 1 *O *O
Filter Eff
Total Zn
Dissolve
Total Suspen
co ^f to
CM 00 H
r-l r-l
0 CM 0
o t- in
r-l r-l
f^ CO 00
• • •
O ^ ro
to
to to o
• I •
^r o H
H r-l
o * r-
r-i cn IH
•H
en *d" en
in co CM
t- en r-
•<# H *»
co ro in
TP co in
to
o o H
CM en en
in
t- o CM
• • •
r-l M1
0 0 CO
in TT •**
•» r-
o o to
CM o en
•* t-
in r^ ro
o 10 «»
r-l CO
•» to i1
to o ro
in r-i
4J
g
3
4J r-l
OOJ.1QS , TOO
Raw Waste
Supernatan
Filter Eff
118
-------
TABLE 11-5.
COMPARISON BETWEEN PROPOSED BAT LIMITATIONS AND
ESTIMATED TREATABILIIY PERFORMANCE
FOR THE CHROME PIGMENTS SUBCATEGORY
STREAM: Filter Effluent
Pollutant
Concentration Basis
Cmg/1)
Proposed BAT Est. Treat. Performance
Maximum 30-Day Average
30-Day Average
Chromium
Lead
Zinc
Cadmium
Tn-t-al Suspended Solids. TSS
1.1
1.4
1.1
0.19
1 37
0.044
0.13
0.050
0.041
5.5
119
-------
SUBCATEGORY
Chrone Pigments
POLLUTANT
Chromium
PRECIPITANT
Proposed MaxLmun 30-day Average
(mg/1) :
1.1
95th Percentile (Z = 1.64)
Long Term Average (mg/1):
Standard Deviation of 30-day Averages (mg/1):
Probability of Achieving Proposed
Maadunum 30-day Average (%).
Nurber of Observations:
.036
.0050
>99
14
CQ
W
1.0
>,i 0-9
*n
J "1 n s
j gj U.O
f| 0.7
> o
'« 0.6
II 0.5
pj
1 0.4
p n 3
-------
SUBCATEGORY
POLLUTANT
Lead
(rag/1): 1.4
Chrome Pigments
T~' '"••••.'" "*" ' '~" " 77-!
Proposed Maximum 30-day Average
a^5^—
95th Peroentile (Z = 1.64) (rog/1):
Long Term Average (mg/1):
Standard Deviation of 30-day Averages (mg/1):
Probability of Achieving Proposed
Maximum 30-day Average
Number of Observations:
PRECIPITANT
0.13
0.10
0.013
>99
14
4J CJ
J! «J
$
f
»~
£.
a #
•§§ 0.
0.9
0.8
0.7
0.6
0.5
4
ft
ca
«
0.3
0.2
0.1
0.0
0.08
0. 9 O.lO 0.11 0. 2
Maximum 30-day Average (mg/1)
Figure 11- 3. Estimated Performance of Proposed
BAT Treatment
121
-------
SUBCATEGORY
Chrone Pigments
POLLUTANT
Zinc
PRECIPITANT
Proposed Maximum 30-day Average
(mg/1):
1.1
95th Percentile (Z • 1.64) (mg/1):
Long Term Average (mg/1):
Standard Deviation of 30-day Averages (mg/1):
Probability of Achieving Proposed
Maximum 30-day Average
Number of Observations:
0.050
0.045
0.003
>99
14
K
1C
1
1
•r
r~
•r
a
«
1.0
M| 0.9
>| 0.8
fl"
'TJ °-6
1 w
I °*5
& 0.4
Q n "3
M v». J
ft\
§> „ „
KJ 02
^01
u.x
n n
**
x*
8 O.Ol
^
_>
^
^
/
^
/
0 0.042 0.04
-J.
/
^
/
^
!/*
4 0.046 0.0^
/
/•
x^
8 0.0-
^
M»«
•"•"
W.H
•••
0 0.052 0.054
Maximum 30-day Average (mg/1)
Figure 11-4. Estimated Performance of Proposed
BAT Treatment
122
-------
Proposed MaxLrmm 30-day Average
(rag/1): 0'.19
95th Percentile (Z = 1.64) (mg/1):
Long Term Average (mg/1):
Standard Deviation of 30-day Averages (mg/1):
Probability of Achieving Proposed
Maximum 30-day Average
Nunfoer of Observations:
0.041
0.034
0.0040
>99
13
2 0.28 .030 0.032 0.034- 0.036 0.038 0.040 0 .0'42 Q.044
0.0
Maximum 30-day Average (mg/1)
Figure 11-5. Estimated Performance of Proposed
BAT Treatment
123
-------
SUBCATEGORY
Chrome' Pigments
POLLUTANT
Total Suspended Solids
PRECIPITANT.
Proposed Maxinow 30-day Average
(mg/1) s 37
95th Percentile (Z = 1.64) (mg/1):
long Terra Average (mg/1):
Standard Deviation of 30-day Averages (mg/1) :
Probability of Achieving Proposed
Maximum 30-day Average
(%) :
Nurber of Observations:
5.5
4.5
0<56
>99
14
1.0
0.6
0.0
3 0
3-5 4'.0 4;5 5 0
Maximum 30-day Average (mg/1)
5
Figure 11-6. Estimated Performance of Proposed
BAT Treatment
124
6.0
-------
SECTION 12.0
SODIUM DICHROMATE SUBCATEGORY
12.1 INTRODUCTION
12.1.1 General Considerations
the treatability test runs for this subcategory were
performed at the in-house laboratory facilities of Versar, Inc.
located in Springfield, Virginia, between November 8 and December
5 1979 for Plant #493. During this period, a total of 18 test
runs were completed utilizing the proposed BAT treatment concept.
Two treatability test units were operated at the same^time
to expedite the study and ensure completion of the tests within
the required time period. Since the selected plant was at a
considerable distance from the laboratory test unit location, the
waste water samples were collected in two large batches of 120
gallons each and transported by truck from the plant site. All
test runs were completed within three weeks from receipt _of the
raw waste to prevent the possibility of sample deterioration to
the maximum extent practical.
12.1.2 Sample Point Location
Figure 12-1 shows the general waste water treatment process
flow diagram for Plant #493 and indicates the appropriate_ sample
point location used in the study. The sampling location includes
waste water from three primary sources including boiler and
cooling tower blowdown, scrubber water from a by-product sodium
sulfate operation, and spent ore residue.
12.2 TREATABILITY TEST MODEL OPERATION
12.2.1 Treatment Technology Tested
The proposed BAT treatment concept includes dual media
filtration added to BPT treatment to achieve a higher level of
125
-------
RAW WASTE WATER
IMPORTED AGIO
INDUSTRIAL WASTE
REACTORS
HOLDING TANKS
WATER
CLARIFIERS
TREATED
EFFLUENT
SLUDGE TO
LAND DISPOSAL
Saitpling point.
Figure 12-1.
General waste water treatment process flow diagram at plant #493
showing the sampling point. (Sodium dichromate manufacture.)
126
-------
suspended solids removal including metal hydroxides and sulfides.
Originally, it was proposed to apply BAT treatment utilizing
sodium bisulfide to reduce chromate to trivalent chromium and
precipitate other heavy metals followed by alkaline treatment and
clarification. However, this BPT treatment aproach had to be
abandoned because of a number of operational difficulties which
could not be conveniently and/or expeditiously mitigated. These
difficulties are outlined below:
1 it was discovered, during the initial waste
characterization, that addition of sodium sulfide at recommended
pH levels above 8 to avoid H2S evolution required excessive
reaction times to reduce chromate to the trivalent form, ^ring
ihe characterization, a series of tests were performed to
determine the sodium sulfide dosage for effective °h£om^e
reduction. These tests were performed at 1.3, 1.5, 2.0, 5.0, ana
10.0 times the calculated stoichiometnc sulfide demand which was
based on the analysis of heavy metals. Results of these tests
indicated that the reaction required one day or more to reduce
most of the chromate present. Specific results of these tests
are summarized in Table 12-1 which presents residual chromate
concentration after 10-day reaction times at the varioussodium
sulfide doses. Review of the table indicates that chromate
concentrations remained at significant levels between 0.9 and 2.4
mg/1 after excessive reaction times. Evidence of the slow
reaction could be observed visually as the waste solutions slowly
turned from a bright orange to green followed by the formation of
a precipitate. During the slow reaction, the pH was observed to
shift further to the basic side making circumstances ideal for
alkaline precipitation. The pH shift is well known to occur and
can be described by the following reaction:
8Cr04= + 3HS- + 17H20
8Cr(OH)3 + 3804= + 13 OH-
When Na2S dissolves in water, HS- becomes the prevalent_species
which reacts with the sodium dichromate. Since hydroxyl ions are
one of the reaction products, the final pH °or"sP°n^n^Y
increases. Formation of the hydroxyl ions would not, however,
preclude the use of lime for final pH adjustment in_providing a
control measure to consistently achieve the optimum pH for
alkaline precipitation.
2. In view of difficulties encountered with slow reaction
times, other tests were performed at pH values below 8. Results
of these tests indicated a more rapid reaction rate f"er
addition of the sodium sulfide, but^ was c°mPllcatedby evolution
of H2S gas. Since H2S gas emissions are a potential safety
hazard to personnel conducting the tests, the method was
abandoned. It is believed that this treatment approach may have
a potential application to best available technology ^J^^
in a closed treatment system. However, time constraints
127
-------
prevented a complete evaluation and redesign of a test model
treatment system that could be used to fully investigate the
technique without the safety hazard.
In view of the difficulties encountered with the use of
sodium sulfide for chromate reduction, this BPT treatment
alternative was abandoned and substituted by a second viable
alternative using ferrous chloride and hydrochloric acid (pickle
liquor). The treatment concept studied includes chromate
reduction with pickle liquor followed by alkaline precipitation
with lime, clarification* and dual mdia filtration for final
polish of the clarifier effluent.
12.2.2 Wastfe Water Characterization
Tables 12-1 through 12-4 present the results of the sodium
dichromate raw waste characterization. An initial sample was
collected to perform the tests. Results of the analyses are
shown in Tables 12-1 and 12-2. Table 12-3 presents additional
analyses performed on the two 120 gallon batches used for the
test runs. Review of the data revealed a wide range of
varibility in raw waste metal and suspended solids
concentrations. The wide variability was most likely due to the
heterogeneous nature of the waste. Suspended solids were
observed to settle readily and required a considerable amount of
agitation when taking samples for the test runs.
Table 12-4 presents the analyses performed on the pickle
liquor. It should be noted that the pickle liquor used for the
test runs was from a different source than that used by the
selected plant. Review of the data shows that the hydrochloric
acid (HC1) concentration was approximately eight percent;
whereas, the pickle liquor currently used by the plant has an HC1
content of 15 percent.
12.2.3 Details on Treatability Test Operation
A complete presentation of the operational parameters for
the test runs is shown in Table 12-5. in general, operational
parameters were selected to the degree possible on the basis of
plant experience. The primary differences between plant
practices and the test treatment conditions were, a)
concentrated HC1 was required as a supplement to the pickle
liquor used for the test runs and, b) the raw waste temperature
was 25 degrees C instead of 50 to 70 degrees C encountered at the
plant. In actual practice, the plant uses pickle liquor with a
higher HC1 content precluding the need of additional acid for pH
adjustment.
128
-------
TABLE 12-1. ANALYSES OF TREATED CHROMATE WASTE WATER
SOLUTIONS AFTER 10 DAYS REACTION TIME WITH SODIUM
SULFIDE AT pH OF GREATER THAN 8.0
Ratio of Sulfide Used to
Stoichiometric Sulfide
Requirement
Residual Chromate
Concentration (mg/1)
1.3
1.5
2.0
5.0
10.0
0.9
1.0
0.9
2.4
1.2
TABLE 12-2. CHARACTERIZATION OF SODIUM DICHROMATE WASTE WATER
Parameter
Total Suspended Solids
Total Residue Solids
Fixed Residue
Total Dissolved Solids
PH
Total Hardness (as CaC03)
Chlorides
Sulfate
Nitrate
Carbonate (as CaC03)
Bicarbonate (as CaC03)
Amount Present (mg/1)
75,800
76,900
70,800
7,000
9.5
4,000
275
3,100
0.80
3,000
40,000
129
-------
TABLE 12-3. CHARACTERIZATION OF SODIUM BICHROMATE BATCHES
USED FOR THE TEST RUNS
Parameter Concentration (mg/1)
Batch A Batch B
Chromium (Total) 1,500 800
Chromium (Hexavalent) 1,490 560
Nickel 7.82 0.60
Total Suspended Solids 5,500 7,100
Total Dissolved Solids 15,710 8,800
TABLE 12-4. CHARACTERIZATION OF PICKLE LIQUOR
Parameter Concentration (mg/1)
Ferrous Chloride (as Fe) 150,000
Chromium (Trivalent) 78 to 85
Nickel 250
Hydrochloric Acid 80,000
130
-------
31
$1:
b
i tti
a
;: s a
O Cl O
in * •
\o ts T o
O O CO
§ ro **
I
I
I
in
". =;
P- rH
U=
m. o
C~ rH
in
^S
r-
2s
sS
53
". o
r- rH
CO
o
o co o
ir> • •
o o o
^ ro ro
o CN rn
ON 0
5
3
o cs m
r-
d
r-
d
o
in
O C4 O iH
S ri en o CM
SOS O rH
• • ••
a\ r* n w CN
O PJ O rH
3 ri m ON
• • in
m c4 I-H CN
o o
S A' n
.S
I
in
O ^O CO
-------
The basic treatment concept used involved addition of pickle
liquor in an amount equivalent to more than two times the
stoichiometric amount required for chromate reduction. The
mixture was then adjusted between pH 2 and 3 with concentrated
hydrochloric acid and stirred for approximately three hours.
Dilution of the waste with additional water was necessary, as was
also practiced at the plant, due to the high solids content.
After stirring, the waste was adjusted to a pH of 9 with lime for
optimum heavy metal removal by alkaline precipitation.
Aeration of the waste water during alkaline pH adjustment
was practiced for all the test runs. Aeration was applied to
oxidize the remaining ferrous iron to its ferric form since
ferric iron has a lower solubility in alkaline conditions and the
possibility of coprecipitation with other metals may improve
settling characteristics of the sludge.
12.3 TEST RESULTS
12.3.1 Discussion of Results
The analytical results for the 18 treatability test runs are
tabulated in Table 12-6". Treatment results for the chromate ion
are also included in the table to ascertain the proportion of
total chromium contributed by the presence of chromate in the
final treated effluent.
The results presented in Table 12-6 show high total chromium
concentrations in the filter effluent for test runs 1, 5, 9, 15,
and 16. Test run number 1 is questionable since the chromium
concentration in the filtered effluent exceeds the concentration
in the treated supernatant by an amount greater than the expected
experimental error.
The poor results for runs 5, 9, 15, and 16 may all be
related to insufficient pickle liquor dosage to completely reduce
all the available chromate to trivalent chromium. This result is
not unlikely since the raw waste was very heterogeneous and
therefore highly varible in chromate content. In addition,
oxidation of some of the ferrous iron in the pickle liquor may
have occurred during storage, thereby decreasing its
effectiveness. It should be noted that the high effluent
chromium concentrations occurred more frequently in the later
runs which would tend to support this conclusion.
There is an apparent correlation between the pH of the
treated waste after the addition of pickle liquor and the poor
removal of total chromium for runs 9, 15, and 16. As shown in
Table 12-5, the pH's for these runs were 6.80,6.40, and 6.40
which are much higher than any of the pH values reported for the
132
-------
3
o o
CM P-
CO *3"
S3
O O
o in
in in
rHH
CM «*
rH H
o o
vo
.. p-
rH rH
a
^r in
rHrH
0
VO
O O
S3
0 0
O 0
co en
vo in
0 O
co P-
vo in
§
o o
VO P-
cn vo
co co
rH rH
r- co
rH rH
o P-
r- vo
H VO
(— VO
cn o
•a* CM
9 a
d d
S3
d d
CM i-l
1-1 I-l
in vo
03 CM
0 H
vo vo
co v
vo
r- in
CO CO
o o
r-
Sco'
co §
a a
in
•-I
voco
rHrH
do
sa
d o
vo o
vo r-
^r co
vo vo
o i1
in CM
rH rH
d d
a a
d o
CO 00
rH rH
co in
CM CM
vo in
in in
•« i-H
CM CM
d d
CM CM
d d
co vo
co co
co vo
CM H
d d
Sg
d d
0 O
en «* .
CO *3-
CO CO
rH 03
o o
•3" •*
TJI in
O
^*
^1*
co in
in CM
33
in in
(H O
«3" tO
s
vo o
in vo
a
S3
in in
13
rf
S*3*
0
d d
o o
V V
CO CM
in in
o~\ o
in vo
ss
??
p- in
o o
o o
O C3
d d
O rH
0 O
S
i
o o
d d
in in
in vo
< •«
0
o o
v v
in vo
o o
o o
d d
in
CO O
• o
V
in
co o
in
vo o
in
in o
in
in o
in in
co o
in
CO O
t-- in
^ o
in in
co o
•v in
>* o
"3< in
•er O
CM in
CO O
co in
vo o
r* in
CO O
*& in
CO O
r- in
CO O
88
o o
v v
in in
O 0
in in
0 0
in vo
O 0
in in
O 0
o o
v v
in in
o o
d d
v v
in ro
in in
O 0
in in
0 0
in in
o o
in in
0 O
o o
v v
0
d d
in in
o o
P- 0
0 rH
in in
0 0
in in
0 0
CO CO
o o
in in
o o
t- in
0 0
r~ in
0 0
rH in
i-l 0
in in
0 0
VO «3>
CM CM
•-I i-l
in p-
0 0
s
vo P-
o o
in in
O 0
in in
o o
vo en
o o
in in
o o
rH CO
rH 0
O O
in in
0 0
in in
0 0
vo o CM
co cn co
VO r-i CO
°-.rH
I H O
i CO »*
r-) O
-------
other test runs. Even though this fact may have some bearing
upon the reduction of chromate, it should not be a limiting
factor in the reduction process because the pH of the pickle
liquor treated waste was always adjusted to around 2.8 by
addition of HC1 which would accelerate the reduction of chromate.
Test runs 1, 5, 9, 15, and 16 were rejected from consideration in
the statistical analysis for chromium due to the aforementioned
difficulties.
Good total suspended solids (TSS) removal efficiencies were
obtained during lime precipitation where low TSS values may be
correlated with good sludge blanket settling characteristics.
However, the filtration step is very unpredictable and
inconsistent in its overall performance.
12.3.2 Statistical Evaluation
Table 12-7 presents a comparison of the proposed BAT maximum
30-day average concentration and the estimated treatability
performance developed in the statistical analysis. The
statistical analysis is presented in Figures 12-2 through 12-5 in
addition to Appendix A for the pollutants studied. Test run
numbers 1, 5, 9, 15, and 16 for total chromium have been rejected
on technical grounds as explained in the previous section.
Application of the t-statistic resulted in elimination of test
run 12 for nickel.
12.3.3 Conclusions
The treatability test results can be viewed as strongly
indicative, but not entirely conclusive, of what the proposed BAT
treatment concept for the Sodium Bichromate Subcategory is able
to achieve. Although the proposed maximum 30-day average
concentration for chromium and nickel was achieved after
screening out the questionable results, more comprehensive
results could be obtained by evaluating the kinetic aspects of
the^ treatment process variables and utilizing appropriate
equipment to improve the mixing of reactants and the measurement
of chemical dosages.
A major conclusion to be drawn from this study is that dual
media filtration does not appear to improve significantly the
clarifier "effluent quality on a consistent basis. Sludge blanket
settling characteristics appear very effective for the reduction
of suspended solids including metal hydroxide precipitates from
the liquid phase. It is anticipated that careful design and
operation of a clarifier unit should preclude the need for dual
media filtration to' reduce toxic metal pollutants as indicated by
the results presented in Table 12-6.
134
-------
TABLE 12-7. COMPARISON BETWEEN PROPOSED BAT LIMITATIONS AND
ESTIMATED TREATABILITY PERFORMANCE FOR THE
SODIUM DICHROMATE SUBCATEQORY
STREAM: Filter Effluent
Pollutant
Concentration Basis
dng/D
Proposed BAT
Maximum
30-Day Average
Est. Treat. Performance
30-Day Average
Chromium 0.32
Hexavalent Chromium 0.050
Nickel 0.17
Total Suspended Solids, TSS 26
0.29
0.20
0.095
460
135
-------
SUBCATEGORY
Sodium Bichromate
POLLUTANT
Chromium
PRECIPITANT
Proposed Maximon 30-day Average
(mg/1)
0.32
95th Percentile (Z = 1.64) (mg/1)
Long Term Average (mg/1)
Standard Deviation of 30-day Averages (mg/1)
Probability of Achieving Proposed
Maximum 30-day Average
Number of Observations:
0.29
0.25
0.029
>99
13
Estimated Probability That Any 30-day ,i
1.0
| 0.9
i
| 0.8
1 °-7
3 °-6
$
w
i °*5
1 0.4
i
0^
-~ ••*
i* o 2
* n i
U.J.
f^^
^*
*«-*
,X
^
X
>
/*
/
/
f
f
s
J
A
f
•
/
'
f
s
s
^
•**
•«•
u'u 0.19 0.21 0.23 0.25 . 0.27 0.29 0.31
Maximum 30-day Average (mg/1)
Figure 12-2. Estimated Performance of Proposed
BAT Treatment
136
-------
.SUBCATEGORY
Sodium Dichrcmate
POLLUTANT
Hexavalent Chromium
PRECIPITANT.
Proposed Maximum 30-day Average
(mg/l)j 0.050
95th Percentile (Z = 1.64) (mg/1) :
Long Term Average (mg/1):
Standard Deviation of 30-day Averages (rag/1):
Probability of Achieving Proposed
Maximum 30-day Average (%) :
Nunber of Observations:
0.20
0.096
0.062
23
14
&
f
8
S1
fi
4J
*
.?
^
•r
£
t
S
?i
i
a
W
l.O1
i 0.9
I
1 °-8
g
i u.;
C5
•a 0.6
i
J3 °*5
4J
S °-4
i
fl 0.3
§>
S 0.2
*! - -
0.1
0.0
^
o.c
•^
x
/
/
0.
/
05
/
/
t
/
/
0
/
10
(
/
,
'
{
t
'
0
X
.15
S
^
x-
**^
0
-fiS
.21
*^-
D
«**"
0
.25
Maximum 30-day Average (mg/1)
Figure 12-3. Estimated Performance of Proposed
BAT Treatment
137
-------
SUBCATEGORY
Sodium Bichromate
POLLUTANT
Nickel
PRECIPITANT
Proposed Maximum 30-day Average
(mg/1): 0.17
95th Percentile (Z = 1.64) (mg/1):
Long Term Average (mg/1):
Standard Deviation of 30-day Averages (mg/1):
Probability of Achieving Proposed
Maximum 30-day Average (%):
Number of Observations:
0.095
0.081
0.0088
>99
17
Estimated Probability That Any 30-day
Averaae Does Not Exceed a Given Maximum
1.0
0.9
0.8
0.7
0.6
0.5
0.4
.3
» n •?
u./
.1
•***
*f*
.**
*s
•° o.<
\S
/
^
^
^
s
•
.
/
.
t
/
t
/
/
/
/
/
f
/
^
s
s
ft
t^
**
570 0.075 0.080 0.085 . 0.090 0.(
Maximum 30-day Average (mg/1)
*r
**
**
•—•
95 0.100
Figure 12-4'. Estimated Performance of Proposed
BAT Treatment After Screening of Data
138
-------
SUBCATEGORY
Sodium Bichromate
POLLUTANT
Total Suspended Solids
PRECIPITANT
Proposed MaxLmun 30-day Average
(mg/1) : 26
95th Percentile (Z = 1.64) (mg/1): 460
Long Term Average (mg/1): 200
Standard Deviation of 30-day Averages (mg/1): 160
Probability of Achieving Proposed
Maximum 30-day Average (%): 13
Number of Observations: 18
1.0 :
>ii °-9
T "i
°i °-s
I*S
< £ 0.7
. . TS
4-1 C5
1 * 0.6
Estimated Probability
Average Does Not Excee
o o o o o o
• !••••
O H to u> >t» cn
—
5
s
0
^
•^~
i
/
0
^"
^
1
/
30
/
/
2
/
/
^
>
X
30 2
(
/
x'
X
30 3
-X
r^
X
^
30 3
y
X
x*
30 4(
X"
^*«
i— «
•^
,»»
JO 450
Maximuna 30-day Average (mg/1)
Figure 12-5. Estimated Performance of Proposed
BAT Treatment
139
-------
-------
SECTION 13
SODIUM BISULFITE SUBCATEGORY
13.1 INTRODUCTION
13.1.1 General Considerations
The treatability studies were carried out at PJB
Laboratories, a division of Jacobs, in Pasadena, California. A
total of 17 test runs were made between September 24 and November
5, 1979, using seven batches of waste water obtained from Sodium
Bisulfite Plant f282.
13.1.2 Sample Point Location
Figure 13-1 is a schematic representation of Plant #282
which was selected for study. The treatability tests were made
using waste water collected from the w§ste stream to the effluent
holding tanks as indicated in the figure.
The plant operated on a non-continuous program. Therefore,
an agreement was made to provide the sampling containers and the
plant personnel would collect the waste water and immediately
inform Jacobs' personnel that the samples were ready for pick up.
Enough waste water was always collected in air tight containers
to run two or more tests from each sample batch. All test runs
were performed expeditiously after receipt of the samples.
13.2 TREATABILITY TEST MODEL OPERATION
13.2.1 Treatment Technology Tested
The waste water from this industry has a high capability
react with elemental oxygen. Therefore, the main objective
the treatment is to reduce this oxygen-consuming capacity.
to
of
A
simple aeration process was used. Determinations of TSS and zinc
were made, but this test was not designed to provide optimum
141
-------
a
S1
£3
o «J
fi m
to s
co g
CN
* 0)
-
S
-------
conditions for the precipitation of zinc
physical separation of the solids.
and other metals or for
13.2.2 Waste Water Characterization
An analysis of a sample of the waste water yielded the
results shown in Table 13-1. The possible presence of sulfur
compounds other than sulfite and sulfate was not excluded. It is
highly probable that substances such as thiosulfate are also
present, but a complete determination of the many possible forms
of sulfur in the water was not undertaken.
13.2.3 Details on Treatability Test Operation
Table 13-2 shows the operational conditions for
treatability tests made in the sodium bisulfite subcategory'.
the
Five different air flow rates were used during the course of
study. These were: 11.5, 23, 35, 46, and 57 SCFH (standard
cubic feet per hour).
Following a procedure similar to that used by a plant
practicing this technology, the initial pH of the waste water was
always adjusted to around 9.5 After pH adjustment, the air flow
rate was then adjusted to the desired value. The progress of the
reaction was followed by continuous monitoring of iodate demand
by titrations in acid solution in the presence of iodide. This
is the EPA approved method for determining sulfite) see 41 FR
52780, 12/1/76). The iodate demand was then expressed in terms
of the equivalent oxygen demand exerted by the sulfite/bisulfite
in solution. COD determinations were also made, but as explained
later, the nature of the samples makes it impossible to obtain
reproducible results in that test.
13.3 TEST RESULTS
13.3.1 Discussion of Results
The analytical results are presented in Tables 13-3 to
13-10.
.It was anticipated that the principal impurity in the waste
water from this subcategory would be bisulfite, either as HS03-
or S205= which could be eliminated by oxidizing it to sulfate
with aeration. The waste is, in fact, more complex.
The simplest way to determine the presence of bisulfite if
no other reducing agents are present, is by titration with iodate
in the presence of iodide and acid. The amount of sulfite
143
-------
TABLE 13-1. SODIUM BISULFITE WASTE WATER CHARACTERIZATION
Parameter
Value
PH
Temperature,
Degrees C
4.8
25
Total Acidity (as CaC03)
Total Suspended Solids
Total Dissolved Solids
Total Residue Solids
Fixed Residue Solids
Chemical Oxygen Demand
Zinc
lodate Demand, as S03=
Sodium
Potassium
Calcium (as CaC03)
Magnesium (as CaC03)
Chloride
Sulfate
Nitrate
Ammonia
Thiourea
780
310
4,700
5,400
4,800
1,400
1.3
1,500
1,540
9.5
150
66
270
2,170
<0.1
260
22*
* Calculated
144
-------
j
o
g
a
.
-1
3
IM
O '
g
to
3
8
in
91
m
at
VO
to
S pi
8
0
in
in
-------
!
i
i
g
1
i
i
I
u
icS
CJ
s
I
5
a
a
••S1
iH
i-l
a
•A
O
iH
i
en M
1
"s
J
in
*
m
N
""*
1
A)
o^ oj CM tn r* o **
S § S *° ".1 ".«» S £
CM ** §§ SS o o
ovo 03 •••• ?l?g
* i-l r» rtO HO rH H
| S § " ".S. S.S. 2 g
03 a OOOO 33
I i s " 23 o-l 33
§0^ C3«> 3° So 00
S S esdoo f3S
vo o o o no
00 TTC4 OOCnS C3Q
§ » S rtH- 00 S S
S g S S ^.°. 2o c^ o
ncn a\ oooo iHo*
« o M o ^j«
oo SP ro^ono oo
s 5 s M oo od a a
O O CJ CJ
in in .H r-| d o'o m m
§o o vo in ^y en ^P| irt in
o 3 rt . . .T in S
•* •"! OOOO i-HrH
*na en oo'aa tsa
o 3
H s fn i-I o oo J M 03 ^
1
3
1
I
a
i
146
-------
TABIE 13-4. SODIUM BISULFITE SUBCKEBGOBY
TREATMENT BY AERATICN. Batch 1
RUN 1
Aeration Time, hr.
0.0
0.5
1.0
1.5
3.75
4.75
6.25
7.75
9.75
11.75
13.75
22.0
24.0
25.25
25.75
26.0
27.0
RUN 2
0.0
1.0
1.5
2.0
4.0
6.0
8.0
10.0
12.0
14.0
24.0
9/12/79
E«
9.6
9.5
9.5
9.5
9.7
9.7
9.9
9.8
9.9
9.9
9,9
9.4
9.2
9.4
9.3
9.3
9.2
9/12/79
9.5
9.6
9.65
9.6
10.0
9.9
9.9
9.9
10.0
10.0
9.5
AIR FDCW RATE:
lodate Demand
as Oxygen, rog/1
360
236
230
218
140
108
96
85
75
68
60
44
41
40
40
37
37
AIR FIOW RATE:
360
256
236
236
152
106
84
64
52
50
32
11.5 SCTH
COD
nxf/1
1850
1690
1100
1020
1090
790
880
930
930
1030
860
930
960
1020
J
46 SCFH
1050
1050
980
920
880
850
860
880
820
860
147
-------
TABLE 13-5. SODIUM BISULFITE SUBCATEGORY
TREATMENT BY AERATION. Batch 2
RUN 1
Aeration Time, hr.
0.0
1.0
1.5
2.0
4.25
5.25
6.25
23.0
24.0
RUN 2
0.01
0.5
1.0
1.5
3.25
4.25
23.0
24.0
RUN 3
0.0
1.0
1.5
2.75
4.75
6.25
9.25
22.25
23.25
9/26/79
EH
9.5
9.4
9.4
9.4
9.2
9.0
8.9
8.5
8.4
9/27/79
9.8
9.8
9.8
9.8
9.8
9.7
.9.6
8.4
9/28/79
6.5
9.5
9.5
9.4
9.4
9.3
. 9.0
8.6
8.6
AIR FLOW RATE:
lodate Demand
as Oxygen, mg/1
140
122
110
102
60
54
46
16
!6
AIR FLOW RATE:
174
160
148
126
100
88
22
22
AIR FLOW RATE:
184
160
146
100
82
68
50
20
20
35 SCFH
COD
mg/1
800
560
640
640
880
720
880
480
400
23 SCFH
560
320
640
640
640
560
800
560
23 SCEH
640
400
2400
480
560
400
480
480
400
148
-------
TABLE 13-6. SODIUM BISULFITE SUBCATEGORY
TREATMENT BY AERATION. Batch 3
RUN 1
Aeration Time/ hr.
0.0
1.0
1.5
3.0
4.5
6.0
9.5
22.5
23.5
24.5
29.0
RUN 2
0.0
0.5
1.0
1.5
3.0
4.5
6.0
9.75
22.25
10/1/79
pH
10.2
10.4
10.6
10.5
10.4
10.2
10.0
9.1
9.0
9.0
8.8
10/3/79
5.0
9.4
9.9
10.8
10.9
10.7
10.5
10.1
9.2
.AIR FLOW RATE:
lodate Demand
as Oxygen/ mg/1
942
810
608
348
246
210
98
40
36
34
26
AIR FLOW RATE:
1046
1004
740
568
198
152
116
68
20
35 SCFH
COD
rag/i
320
960
800
640
640
720
560
400
560
480
960
46 SCFH
880
800
880
560
560
640
640
560
560
149
-------
TABLE 13-7. SODIUM BISULFITE SUBCATEGORY
TREATMENT BY AERATICN. Batch 4
RUN 1
Aeration Time, hr.
0.0
0.5
1.0
1.5
2.0
3.5
4.5
5.0
6.0
22.25
RUN 2
0.0
0.5
1.0
1.5
2-0
4.0
5.0
6.5
9.75
22.75
RUN 3
0.0
0.5
1.0
1.5
2.0
4.0
5.5
7.0
... 10.0
23.0
10/12/79
EH
7.0
9.5
9.6
9.6
9.7
9.7
9.7
9.7
9.6
9.3
10/15/79
7.0
9.6
9.5
9.5
9.6
9.7
9.7
9.8
9.7
9.6
10/16/79
9.6
9.5
9.5
9.6
9.6
9.6
9.6
8.6
9.7
9.6
AIR FLCW RATE:
lodate Dsnand
as Oxygen, mg/1
808
568
348
172
120
58
45
39
31
6
AIR FLOW RATE:
856
684
568
398
316
84
58
38
22
7
AIR FLOW RATE:
.800
476
460
268
166
68
46
34
20
8
57 SCFH
003
ragr/i
1120
560
480
480
480
560
640
640
640
560
11,5 SCFH
800
480
400
260
400
400
35 SCEH
880
480
400
320
320
320
320
240
320
80
150
-------
TABLE 13-8. SODIUM BISULFITE SUBCATEGORY
TREATMENT BY AERATION. Batch 5
RUN 1
Aeration Time, hrs.
0.0
0.25
0.75
1.25
1.75
2.25
4.25
5.25
6.25
RUN 2
0.0
0.5
1.0
1.5
2.0
4.0
6.0
8.0
23.75
RUN 3
0.0
0.5
1.0
1.5
2.0
2.5
4.5
6.5
23.75
10/22/79
pH
6.6
9.5
9.9
10.0
10.2
10.2
10.2
10.1
10.1
10/23/79
9.5
10.6
10.9
10.95
11.0
10.85
10.5
10.5
9.8
10/24/79
9.5
10.35
10.2
10.4
10.4
10.5
10.4
10.4
9.2
AIR FLOW RATE:
lodate Demand
as Oxygen, mg/1
388
276
172
90
50
40
14
12
7
AIR FLOW RATE:
382
242
160
102
80
30
14
8
3
AIR FLOW RATE:
398
250
118
60
38
28
12
6
2
23 SCFH
COD
mg/1
400
400
400
240
240
160
240
160
160
46 SCFH
400
240
240
160
160
160
160
160
80
57 SCFH
240
160
160
160
160
320
320
320
240
151
-------
TABLE 13-9. SODIUM BISULFITE SUBC&TEGORY
TREATMENT BY AERATION. Batch 6
RUN 1
Aeration Time, hr
0.0
0.5
1.0
1.5
2.0
3.5
5.0
6.5
10.0
22.75
RUN 2
0.0
0.5
1.0
1.5
3.0
4.5
6.0
9.5
22.5
10/29/79
pH
9.7
10.2
10.4
10.4
10.4
10.6
10.6
10.6
10.5
9.9
10/31/79
9.4
9.6
9.7
9.8
10.0
10.0
10.0
9.8
9.3
AIR FLOW RATE:
lodate Demand
as Oxygen, mg/1
1160
874
774
648
530
208
178
140
112
50
AIR ELOW RATE:
1064
890
716
508
148
92
70
42
16
11.5 SCEH
COD
mg/1
1520
1120
1120
1040
1520
640
720
720
640
720
46 SCEH
1280
1120
960
960
800
880
960
880
880
152
-------
TABLE 13-10. SODIUM BISULFITE SUBCATEGORY
TREATMENT BY AERATION. Batch 7.
RUN 1
Aeration Tims, hr.
0.0
0.5
1.0
1.5
2.0
4.0
6.0
22.75
25.25
RUN 2
0.0
0.5
1.0
1.5
2.0
4,5
6.5
9.75
22.50
11/2/79
pH
9.5
9.5
9.6
9.6
9.7
10.2
11.0
10.1
10.0
11/5/79
9.5
9.6
9.6
9.7
9.8
10.7
11.0
10.6
9.8
MR .FLOW RATE:
lodate Demand
as Oxygen, mg/1
2328
2164
2052
1830
1450
876
338
84
84
AIR FLOW RATE:
2272
2158
1948
1734
1538
450
232
152
62
23 SCFH
COD
mg/i
2400
2160
2000
1360
1360
880
880
480
480
57 SCFH
2160
2160
1840
1680
1600
800
640
480
400
153
-------
present cannot be greater than indicated by this titration.
However, other compounds including sulfide, thiosulfate,
polysulfide, etc., will also react with iodate, so there may be
less sulfite than indicated by the titration. Despite this lack
of specificity, the test was a useful one as an.indicator of the
course of oxidation when the sample was aerated.
In cases where the only reducing agent is tetravalent sulfur
ru oxide' sulfite' bisulfite, or metabisulfite) , the COD
snouid be the same as the oxygen demand from the iodate. The
actual COD determinations nearly always gave results that were
much higher than could be accounted for by the iodate titration
(Table 13-4 through 13-10). This is no doubt due to compounds
that are oxidized to sulfate in the COD test, but that are not
oxidized by iodate or that are oxidized only to sulfur or
tetrathionate or possibly other intermediate products.
Of greatest concern is the fact that sulfur, thiosulfate,
and any other compounds that yield sulfur as an intermediate are
not likely to be fully oxidized in the COD test because elemental
sulfur is difficult to oxidize by wet reagents. The amount that
will be oxidized will vary, depending upon seemingly trivial
variations in the conditions of the test. Since the samples were
relatively high in COD, small aliquot sizes were used in the
analysis which may also cause variation due to the difficulty
involved in making small liquid measurements. The reduction of
COD in the aeration treatments generally appeared to be less than
50 percent of the original COD.
Whatever the explanation, the-conventional COD tests were
notably erratic, with variations much greater than are expected
when oxidizing organic matter.
«.», The proposed treatment would be expected to oxidized slowly
the sulfide and sulfite, but other forms of sulfur would be
oxidized only partially or not at all. The iodate titration
results showed good consistency, and quite satisfactorily
indicated the progress of aeration. The total reduction of
iodate demand was generally 90 to 99 percent of the initial
values.
„ „ -Tie fate,of. Deduction of iodate demand does not bear any
consistent relationship to the rate of air application over the
range of 11.5 to 57 SCPM. Evidently, the rate was determined by
the kinetics of the reaction rather than the supply of oxygen.
The rate always declines as the reaction proceeds, yet there is
not a consistent relationship between the iodate demand and its
rate of decline when comparing different runs. In the first two
batches, iodate demand (expressed as oxygen) seemed to level off
at about 3 to 5 mg/1 after 24 hours, but in the others it was
i««*M IE J°J?.few *enths of a mg/1 in some cases. It is
possible that this residual demand is due to thiosulfate
154
-------
the
BAT
13.3.2 Statistical Evaluation
A statistical analysis was performed for .z
demand (COD), iodate demand, total suspended solids, and zinc
Figures 13-2 through 13-5 and Appendix A. Results of
analysis are summarized in Table 13-11 where the proposed
maximum 30-day average is compared to the estimated performance
SO-dTaveragVvalues! The proposed BAT limitations are designed
such that compliance can be achieved at least 95 percent of the
time.
Tho statistical analysis for COD is based on values obtained
at the ^termfnaUon of "Separation test run. This approach for
data selection was used since it incorporates variability due to
Sample collection with variability in the i?borato^^na^n?
which should relate well to actual practice. Data point
selection for iodate demand was similar although not "critical
since the analytical results were very uniform throughout the
test runs. Appendix B presents the iodate demand curves based on
data in Tables 13-3 through 13-10.
13.3.3 Conclusions
The treatability test results serve as a good indication of
the general applicability of the treatment technology considered
to the proposed BAT regulations. Results show that the pollutant
concentration basis for the proposed BAT maximum 30-day average
effluent limitations is achievable for COD with the Prescribed
treatment technology. However, in view of the wide variability
observed for the conventional COD test, it is recommended that
the iodate/iodide test be considered for possible us e as the
basis for an effluent limitation on sulfite/ bisulfite or the
equivalent oxygen demand.
The zinc concentration was determined during the course of
study before and after pH adjustment with caustic soda, ^sults
indicate that there is a significant reduction in the dissolve^
zinc concentration due to alkaline precipitation. In actual
practice, clarification and possibly the use of chemical
coagulating or floculating agents may be required to assist in
separating the metal hydroxide precipitates.
155
-------
TABLE 13-11. COMPARISON BETWEEN PROPOSED BAT LTMITATIONS AND
ESTIMATED TREATABILITY PERFORMANCE
FOR THE SODIUM BISULFITE SUBCATEGORY
STREAM: Effluent
Pollutant
Concentration Basis
(mg/1)
Proposed BAT
Maximum
30-Day Average
Est. Treat. Performance
30-Day Average
Chemical Oxygen Demand
COD
Total Suspended Solids, TSS
lodate Demand (as Oxygen)
680
22
•d)
600
274
37
(1) Recommended in place of conventional COD for the proposed
limitations.
156
-------
SUBCATEGORY
POLLUTANT
PRECIPITANT
Sodium Bisulfite
Chemical bxygen Demand
(mg/1): 680
Proposed Maximum 30-day Average
—
95th Percentile (Z » 1.64) (mg/1): 600
Long Terra Average (mg/1): 480
Standard Deviation of 30-day Averages (mg/1): 75
Probability of Achieving Proposed
Maximum 30-day Average (%): >99
Number of Observations: 15
3
1.0
0.9
0.8
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0.0
mm
on
' 3
mm
00
MB
MM
«*""
MBi
EBB
*
MM
/
4
EMI
/
60
MM
1
r
MM
/•
MOB
OHM
^
/
j
^
^
^^
/
X
^
00 6
^
«^
«*°
=*«
9W
BOBK
Maximun 30-day Average (mg/1)
Figure 13-2. Estimated Performance of Proposed
BAT Treatment
157
-------
SUBCATEGORY
POLLUTANT
PRECIPITANT
Sodium Bisulfite
Total Suspended Solids
Proposed Maximum 30-day Average
(mg/1) : 22
95th Percentile (Z = 1.64) (mg/1) : 270
Long Term Average (mg/1): 240
Standard Devia-hion of 30-day Averages (mg/1): 23
Probability of Achieving Proposed
Maximum 30-day Average
(%)
Number of Cbservations :
-------
SUBCATEGORY
Sodium Bisulf ite
POLLUTANT
lodate Demand
PRECIPITANT
Proposed Maximum 30-day Average (nig/1) : None
95th Percentile (Z = 1.64) (mg/1): 37
Long Term Average (mg/1): 27
Standard Deviation of 30-day Averages (mg/1): 6.5
Probability of Achieving Proposed
Maximum 30-day Average (%):
Number of Observations: 14
Not Applicable
I
4J
H
1
1
1
1
t
«
1.0 '
i| °-9
•|
S °'8
,e
.1 °'7
(d
1 w
gj
S 0.5
Ki
§ 0.4
I 0-3
1 02
2
n)
< 0.1
0.0
1
•***
^X
?^
x'
^
V
f
f
7
f
r
/
<,
f
s
/
_,<
x^
x
*-•
0 20.0 25.0 30.0 . 35.0 40.0
Maximum 30-day Average (mg/1)
Figure 13-4. Estimated Performance of Proposed
BAT Treatment
159
-------
SUBCATEGORY
Sodium Bisulfite
POLLUTANT
Zinc
PRECIPITANT
Proposed Maximum 30-day Average
(mg/1): Q.50
95th Percentile (Z = 1.64) (mg/1): 1.2
Long Term Average (mg/1): 0.85
Standard Deviation of 30-day Averages (mg/1): 0.20
Probability of Achieving Proposed
Maximum 30-day Average
Number of Observations: 16
Not Applicable
1.0
>ii °-9
jj §
°| 0.8
f| 0.7
4J C5
^rrt 0.6
#1
3$ °*5
•H W
1§ 0.4
3D n "3
Q U. J
JD
a? n ?
M *
Q gj
ttl 5
« n i
U..L
IMM.
— •
x*
3 0.
X
•
^
6 0.
j
j
f
/
/
/
—j
/
/
f
0.8 0 .9 i.
*
/
r"
0
^
r
i
i
X
i
^
-*•"
i.
2
•—•
*—
1
«^«
3
MMB
Maximum 30-day Average (mg/1)
Figure 13-5. Estimated Performance of Proposed
BAT Treatment
160
-------
SECTION 14.0
SODIUM HYDROSULPITE SUBCATEGORY
(FORMATE PROCESS)
14.1 INTRODUCTION
14.1.1 General Considerations
The treatability tests were carried out at Plant #672 from
September 10 to October 11, 1979. During this period, a total of
18 test runs were completed. Enough waste water was collected
most of the time to run two tests simultaneously.
14.1.2 Sample Point Location
Samples were collected from the sodium hydrosulfite waste
stream as indicated in the process waste flow schematic in Figure
14-1 for the selected plant. Sample collection was made at the
end of the storage pond pipe inlet.
14.2 TREATABILITY TEST MODEL OPERATION
14.2.1 Treatment Technology Tested
The waste treatment processes tested for this subcategory
consisted of physical (mechanical) aeration to treat readily
oxidized chemical oxygen demand (COD) such as sulfite, and to
also test the use of dual media filtration. In view of the
difficulty involved in establishing a properly seeded and
representative biochemical oxidation system on a small test
scale, only chemical oxidation (i.e., physical aeration) was
studied. However, it should be understood that this technology
may be coupled with a well established biochemical oxidation
process to further oxidize biodegradable forms of COD and fully
represent the performance of the best available technology.
161
-------
a
8
A
8
18—1
A H
B
j
T
i
!•
I
-p
td
3
51
H
3
01
O
4
• v rn
H "d
a o
w
4J ^_x
(0
m
-P
-H O
•c a
o s
H -H
M-l rH
w (d
0) 10
ll
id c
M -H
0) S
CJ O
(1) rj
C5 to
162
-------
14.2.2 Waste Water Characterization
The results of the waste water characterization for the
sodium hydrosulfite subcategory are presented in Table 14-1
Review of the results indicates a high concentration of organics
and heat unstable inorganics at levels of approximately 1200
mg/1. Organics and heat unstable inorganics such as elemental
sulfur are determined as the difference between the total residue
solids and fixed residue solids.
14.2.3 Details on Treatability Test Operation
Table 14-2 presents the operational parameters -for the
treatability tests made in the sodium hydrosulfite subcategory.
The removal of COD in the waste water was accomplished by
diffusing air through the waste. Five different air flow rates
were used in the study including 11.5, 23, 35, 46, and 57 SCFH
(standard cubic feet per hour). The pH of the raw waste water
was always adjusted to around 1.2 by the use of lime or H2S04 as
required. Following this, and after adjustment of the air flow
rate to the desired value, the progress of the COD reduction was
followed by repeated iodate titrations.
14.3 TEST RESULTS
14.3.1 Discussion of Results
The analytical results and removal efficiencies for COD,
iodate demand, zinc, chromium, and TSS are shown in Table !4-f»
The variations in COD and oxygen demand from iodate (in acid
iodide solution) during aeration are shown in Tables 14-4 through
14-14 for each of the test runs made. Chemical oxygen demand
values obtained during test runs 1 through 6 in Tables 14-4
through 14-6 are from Table 14-3 which presents results obtained
at the Springfield, Virginia laboratory. All other COD values
were determined at the test site in a mobile laboratory. Even
though results are presented for chromium and zinc, the main
objective of this study was the removal of COD and TSS. Any
removal of chromium and zinc can only be considered as incidental
to the aeration tested.
Organic matter as well as sulfur compounds are present _in
this waste water. The COD values in the raw waste show a wide
range, from about 2,000 mg/1 to more than 20,000 mg/1. This
demand declines during aeration, but the results are erratic.
The COD test does not appear to be a very good parameter for
monitoring degree of treatment when physical aeration is applied.
This is perhaps due to the difficulty of oxidizing the sulfur
163
-------
TABLE 14-1.
WASTE WATER CHARACTERIZATION FOR THE SODIUM
HYDROSULPITE SUBCATEGORY
Parameter
Amount Present (mg/1)
Calcium
Chromium (total)
Chromium (hexavalent)
Potassium
Magnesium
Sodium
Zinc
Total Suspended Solids
Total Residual Solids
Fixed Residue Solids
Total Dissolved Solids
Methyl Orange Alkalinity
(as CaCOS)
Chloride
Sulfate
Nitrate
Carbonate (as CaCOS)
Bicarbonate (as CaCOS)
6.9
0.035
0.004
15
8
9,000
0.29
264
26,856
25,780
26,000
10,220
155
4,500
0.33
2,160
8,060
164
-------
o
3
a
5
.2
i
i
S
U|
!
I
S
5
•H
in
*-4
m
3
S
S
0,
00
r»
>»
in
,
^ S
vo
VD ^. S
in x» cs
^*
cs \o
" 5t M
- s»
- 1 "
5 »
" 3r M
«»4
- S S
i-4
" S S
- §»
o
rt of M
Batch Number
Date
Volune of Wdste
Hater Aerated
(gallons)
in in
. t in
>-i in • >a M o
r-l Jj -H «
""I S S R M S
«r in p> 9 V VO
0 O CM O
">. ". P. » M 0
p* p» m m \5
^i ^ r* ^ M o
r» [^ m t»i \o
p* p» m . «H 03
S " 2 S S M S
""" S «H *w a
s5* f .al li 1 ? 21^3
2»j Q-P -us *j -a „ s ij 3
3 K tn T3 fO E fl JJ gj»H
-------
8
1
l
1
|
]
!'
i
'§
|B
5
j§
* w
<
j
S
55
H
a
3
a
CM
S
o
H
-1
J
J
.
in
»
<*>
CM
"
1
o o H r-i H r»
d tn o o o r-i
^» ^? cn cn cn cn
00 £g SS
TP P^
o o o o o o
!§ ^§ §S
O 0 OO 0 O
pH in tn po H H
CM CM r-* CM CM CM
O O O O O O
g cn S 2! 2 ^
o o o o o o
•v CM vo in vo vo
ro vo CM fi o CM
o o o o o o
•Q< CM ^S* ^P CO CM
rt vo H r-i co vo
t^ c> ^* "V ro P*
i— I in in H ro vo
^r vo ro CM o CM
ro f*1 co PO ro co
O o o o o o
r-i m r> cn CM jg
CO CO CM CM CM CM
go co vo vo o
ro in cn co cn
^r CM cn cn in vo
SO CO O CO <£
fO CM VO CO U
O 0 00 00
CM vo cn r-i in o
in co vo •«• *a« cn
H cn ^" ^ -^ ro
H H H H H
O O 00 0 O
CO t*+ VO O ^" VO
S CO CM H" H H
o o o o o o
o in rH cn HO
^T CO iH H H
O 0 O O O O
2* ro vo co p* in
H CM rH rH H H
o o o o o o
<• po in in CM vo
rH CM H H H H
o o o o o o
r- rH vo o *r o
CM P^ ^1* CO H CM
CO CM CM CM CM CM
o o o o o o
cn co rH tn vo o
PH vo r- m *3i ro
CO CM CM CM CM (M
|3 | 8 § Jo§
O 1}HQ (3 H Q 'MHQ O £
rd T'nQ 4J fli 05 W ra tn o)
Kj^t/l-Ptu (d4JUl -MO? £n
Is sea eea |e« ||
CD
s
d
*3*
H
crv
o
^*
o
CO
d
o
•
o
H
T
m
**
o
CM
CM
CO
o
•<*
d
d
in
•
VO
o
H
^r
d
CN
d
vo
CO
H
vo
H
&
b
1 I
Aearated '
Total Chn
d d
CO O
o d
in vo
H O
d d
in o\
HO
do
So
d d
o?^
d o
in co
d d
38
do
o in
do
°°
d d
S3
o'o
CO O
CO H
• *
0 IN
S3.
HO
H CO
. H H
H O
ss
do
VO (M
CM O
d d
r*.
ss
r-i d
H CM
O O
HO
g
g*
ij M
O >
$31
SeS
I
10
H O
oo r-
HH
°°
od
O CO
X* CO
do
Sin
x>
HH
as
0 o'
CO O
HH
do
o
HCO
in CM
o o
do
a. s
do
SS
o o
o\ o
oi~
cn oa
CM VO
in o
oo
CMO
COH
do
00
oo
ss
do
Sg
d d
X* CO
H O
d d
g
CM *""* 1fl
|!l
H
CM ^r
^r t—
rH rH
HCM
o o
r-» r->
CM CM '
0 0
rO VO
m tn
HH'
ss
Oo
ss
do
in co
co a!
HH
O. O
oo
ss
do
o'o1
oo
§00
CO
crv cn
So
oo
o r»
HO
do
gg
do
00
O 0
H °
o 3
do
o o
g
•P.-.&
§ Q irt
HSl
OH
H3I
I8"
fo
i
CO
4J
1
1
i
a
s
8
s
t
^
s
53
1
*
5
a
•§'
d)
i
I
&
1
1
g
g
§
ra
•a
1
30D values rej
c aeration.
a §
3 S
166
-------
CO
r-t
5
a
a
a
s
^
rH
" 1
01 g
1
00 $
-a
»,
"
' *
CO
IS
1-1
is
a CM f* A
*** s P
CM 0* S «
m«H ,H m T~) m
rHO n n n n
Sm co r* MO
in fsj iH CM CM
r-i O O*O O O
*ya\ m r* *n H
oin ?*. \o r» m
r-I O i-| r-i rH f-i
t»>««3« in m rH CM
(SCM •# -*f ^^n
on ^« m in n
CM o \o ro o o
o co ^ r-. Tj» CM co CM ri
vorn in ^j« ^» ^ _
oo" o* o* oo
mo o o o o
oo* 0*0* o o
m<5 n iH o o
o o* o o o o
33 SS 58
•*fr-{ co r- r^ co
in o «H vo CM vo
co in in ro in n
in o* CM o o o
in o o ro CM o
co m co in in ^f
in o* «H" o" o o
o o* 0*0* o o
^ ro n tn 3; «>
en r-i in n on
o o o o o o
o ^
co ,-J nr-i m 10
• . n fi o o.
CM ° do .do*
CM xo p** CM m co
no nr-i o o
in" d do o o
|- J |J
cT x ^
11 s 2
ro
cs
\o
CO
CO *& CO
O,CMO.
O CM CO
^
3as
3S3
CM
23»
ScS".
1-1 CM r~
oeoi-i
in co CM
in <-< H
0 O r-l
crt CM in
CMfM
°S 3
CM CM
§rr co
co in
ro i— )
§38
n
o
o in vo
en • •
O CM CO
n m •
4J
L fliS'
Us
EH 05
167
-------
TABLE 14-4. SODIUM EKDROSULFITE SUBCATEGORY
TREATMENT BY AERATION. Batch 1
BUN 1
Aeration Time, hr.
0
2
4
6
8
10
12
16
RUN 2
0
2
4
6
8
10
12
16
9/10/79
PH
7.3
7.8
7.9
8.0
8.24
8.27
8.24
8.25
9/10/79
7.3
7.8
8.14
8.25
8.3
8.3
8.35
8.3
AIR FLOW RATE:
lodate Demand
as Oxygen, mg/1
25.3
27.6
27.4
27.6
22.8
21.4
19
16.6
AIR FLOW RATE:
25.7
25.0
24.7
21.8
18.6
18.0
15.2
13.6
57 SCEH
GOD (1)
mg/1
3190
. .-
-
-
—
—
-
2710
57 SCFH
3270
-
—
-
—
—
—
2460
(1) From Table 14-3
168
-------
TABLE 14-5. SODIUM HXDROSULFITE SUBCATEGORY
TREATMENT BY AERATION. Batch 2
HUN 3
Aeration Time, hr.
0
2
4
6
8
10
14
18
22
26
30
34
RUN 4
0
2
4
6
8
10
14
18
22
26
30
34
9/13/79
PH
7.3
7.5
7.93
8.11
8.15
8.16
8.2
8.16
7.0
7.2
6.24
6.71
9A3/79
7.2
7.27
7.26
7.4
7.3
7.28
7.25
6.74
6.05
6.14
6.17
6.02
AIR FLOW .RATE:
lodate Demand
as Oxygen, mg/1
26.5
26.3
23.0
20.9
18.6
16.7
14.0
12.4
8.7
2.9
0.2
0.2
AIR FLOW RATE:
26.5
27.7
25.1
22.9
21.8
21.2
16.0
9.8
4.2 .
1.4
0.4
0.4
57 SCFH
COD (1)
mg/1
1940
-
-
-
-
-
-
-
-
-
-
1750
57 SCFH
1940
-.
-
- •
-
-
-
-
-
-
-
1860
(1) From Table 14-3
169
-------
TABIE 14-6. SODItM HXDSOSULFITE SUBCATEGORY
TREATMENT BY AERATION. Batch 3
BUN 5
Aeration Time, hr.
0
2
6
12
18
20
24
28
32
36
44
48
52
56
60
RUN 6
0
2
6
12
18
20
24
28
32
36.5
44
48
52
9/19/79
E«
7.01
7.51
7.79
8.15
8.60
8.62
8.53
8.38
8.14
8.05
8,09
8.14
8.09
8.12
8.16
9/19/79
7.01
7.16
7.22
7.04
7.65
7.64
7.75
7.86
7.93
7.85
8.02
8.14
8.28
AIR ELCW RATE: 4'6 -SCFH
lodate Demand COD (1)
as Oxygen, rag/1 mg/1
236.7 4080
234
223.9
111.8
27.3
21
11.8
10
8.8
7.4
5.2
4.4
2.9
1.0
1.03 1190
AIR FLOW RATE: 46 SCTH
236.7 4080
283.5
272.2
155.4
10.1
9.6
9.2
8.6
9.0
8.6
7.6
6.8
6.5 2060
(1) From Table 14-3
170
-------
TABLE 14-7. SODIUM HH>ROSULFITE SUBCATEGORY
TREATMENT BY MIRATION. Batch 4
RUN 7
Aeration Time hr.
0
1
2
3
4
8
13
' 17
23
29
9/20/79
,pE
6.5
6.75
6.46
2.6
2.6
2.6
2.6
AIR FLOW RATE:
lodate. Demand
as Oxygen, mg/1
929.1
425.3
340.8
5.2
0.5
0.24
0.3
46 SCFH
ODD
mg/i
11520
22800
22420
21840
21460
20120
20310
18010
17240
14690 (1)
(1) From Table 14-3
171
-------
TftBEE 14-8. SODIUM HYDHQSULFITE SUBCATEGORY
TREATMENT BY AERATION. Batch 5
HUN 8
Aeration Time, hr.
0
1
2
3
4
8
12
20
24
RUN9
0
1
2
3
4
8
12
20
24
28
32
36
40
44
9/24/79
EH
9.2
7.75
7.86
8.07
8.17
8.16
8.08
7.8
8.2
9/24/79
8.80
7.78
7.98
8.04
8.16
8.15
8.16
8.20
8.17
8.12
8.10
8.25
8.05
8.01
AIR FLOW RATE:
lodate Demand
as Oxygen, mg/1
39.1
34.2
37.9
32.2
33.4
29.6
25.3
0.5
0.4
AIR FLOW RATE:
41.8
36.2
38.4
32.6
38.3
34.0
30.3
23.5
19.9
13.3
7.0
1.2
0.4
0.3
46' SCFH
ODD
mg/l
1460
843
728
613
1130
-
1090
854
892
57 SCPH
1460
1050
1160
1160
1160
1090
1010
1010
854
854
776
854
698
776
172
-------
TREATMENT BY AERATION. Batch 6
RUN 10
Aeration Time, hr.
0
1
2
3
4
8
12
16
20
24
28
32
36
40
46
50
56
RUN 11
0
1
2
3
4
8
12
16
20
24
28
32
36
9/26/79
EH
6.89
-
-
-
7.54
7.57
7.54
7.61
7.58
7.48
7.40
7.40
7.46
7.48
7.22
7.00
6.90
9/26/79
7.1
-
-
-
3.2
2.75
2.81
2.82
2.82
2.82
2.90
2.93
2.97
AIR FI£W RATE:
lodate Demand
as Qxygesn, rag/1
439.0
-
-
-
285.3
279.0
227.7
188.6
162.1
111.5
72.1
43.0
23.8
15.8
12.5
5.5
2.2
AIR ELCW ROTE:
232
-
-
-
0.66
3.0
3.1
2.8
2.0
2.0
0.4
0.3
0.4
35 SCTH
COD
rag/1
3410
5620
5300
5300
5220
5220
4740
4620
4340
4400
4300
4220
4100
3980
3670
3390
3310
35 SCEH
3410
5380
5540r
5500
5260
4790
4640
4170
4100
3710
3720
3720
3400
173
-------
TBBEE 14-10. SODIUM HXDBOSULFITE SUBCAIEGORY
TREA3MENT BY AERflUCN. Batch 7
RUN 12
Aeratim Time, hr.
0
2
3
5
6
10
14
18
22
26
30
38
42
46
50
BUN 13
0
2
3
5
6
10
14
j.8
22
26
30
38
42
46
50
54
63
10/1/79
EH
11.8
-
7.46
-
7.4
-
5.81
5.31
5.19
5.1
5.0
4.9
4.9
4.88
4.9
10/1/79
8.4
-
8.36
-
8.44
-
8.93
8.99
8.97
8.9
8.8
8.6
8.7
8.5
8.5
8.6
8.67
AIR Et£W RHCE:
lodate Demand
as Oxygen, rog/1
947
-
947
-
869.2
-
83
110.2
5.4
5.6
5.0
4.1
4.3
5.4
5.4
AIR EICW RKTE:
800
-
800
-
614.8
-
181
HI. 4
62.2
42.6
27.3
17.8
17.3
14.7
12.9
11.1
10.0
35 SCEH
COD
rcg/1
7340
8750
7950
8030
-
17600
18560
9200
5760
6320
5600
4560
4650
4400
4240
35 SCEH
7340
-
6610
6570
-
5290
5850
5170
4970
4770
4730
4090
4330
4330
4180
3700
3630
174
-------
TABLE 14-11. SODIUM HYDROSULFITE SUBCAIEGORY
TREATMENT BY ..AERATION. - Batch 8
RUN 14
10/4/79
AIR FK3W RATE:
35 scm
Aeration Time, hr.
0
1
2
3
4
8
20
pH
11.46
3.5
3.07
3.12
lodate. Demand
as Oxygen, mg/1
273
0.4
0.4
0.3
COD
mg/1
6700
5320
5130
4730
4580
4650
4500
175
-------
TABLE 14-12. SODIUM .EKDROSULFOTE SUBCATEGORY
TREATMENT -BY AERATION. . .Batch .9
,RUN 15
Aeration Tame, hr.
0
1
2
3
4
9
17
21
25
29
31
33
RUN 16
0
1
2
3
4
9
17
21
25
29
33
41
45
49
10/8/79
PH
9.35
5.97
5.88
5.78
5.68
4.13
4.18
4.18
4.18
4.17
4.18
4.17
10/8/79
10.4
1.48 .
7.55
7.62
7.65
7.75
7.80
8.13
8.13
7.79
7.66
7.58
7.58
7.30
.AIR .FLOW .RATEi
lodate Demand
as Oxygen, rag/1
88.6
81.1
78.6
83.1
75.7
8.2
2.6
2.2
' 2.1
0.7
0.6
0.4
AIR FLOW RATE:
47.2
98.6
100.8
106.7
90.7
79.9
58.7
42.8
37.8
26.8
19.6
10.0
4.6
0.9
23 .SCEH
COD
itg/1
2410
2740
3020
2940
2780
2460
2600
2600
2520
2520
2450
2370
23 SCPH
2410
1320
1350
1320
1400
1290
1260
1220
1180
1180
1140
1080
1010
1010
176
-------
TABLE 14-13. SODIUI4 HYDKDSULFITE SUBCATBGCHY
TREATMENT BY AERATIDN. Batch 10
.RUN 17
Aeration Time, hr.
0
1
4
13
18
24
28
37
10/10/79
PR
7.27
7.29
7.44
7.52
7.57
7.60
7.66
7.62
AIR .ELOW .RATE:
lodate Demand
as Oxygen, mg/1
223.4
200.2
173
212.2
188.2
176.6
177
.164.4
23 .SCFH
COD
rag/1
4340
3600
3720
3720
3440
3320
3080
3000
TABLE 14-14. SODIUM HYDROSUIPITE SUBCATEQCRY
TREATMENT BY AERATION. Batch 11
RUN 18
Aeration Time, hr.
0
1
2
6
16
10/11/79
pa
11.5
4.85
3.89
2.34
4.10
AIR FLOW .RATE:
lodate Demand
as Oxygen, ntg/1
484.8
347.2
337.6
0.8
0.6
23 SCFH
GOD
rog/1
4900
4430
4030
3640
3400
177
-------
from some of its intermediate oxidation states, such as elemental
sulfur. Sulfide was sometimes present, as noted by the odor, and
it is quite likely that elemental sulfur would be produced to
some extent in the oxidation processes. The amount of this
sulfur that would be oxidized by chromic acid in the COD test
would be variable and not reproducible in the standard procedure.
The extent of oxidation of COD by aeration was generally in the
range of 20% to 60%. , •.
The iodate demand, as determined by titration in the
presence of acid and iodate, was monitored during the aeration
test runs. The results are shown in terms of the equivalent
oxygen demand. The iodine produced from the iodate reacts with
sulfide, polyfides, sulfite, thiosulfate, and possibly with some
other sulfur species. The iodate demand (as oxygen) was always
much less than the COD. The ratio was highly erratic, but was
usually between 1% and 10%.
During the aeration runs, the iodate demand declined. The
results are somewhat erratic, possibly indicating poor
reproducibility with this particular waste. However, the iodate
demand does approach zero after a sufficient period, usually
within 24 hours. The air supply rate, ranging from 11.5 SCFH to
57 SCFH, did not appear to affect the rate of decline of the
demand. The rate evidently is controlled by the kinetics of the
reaction rather than the air supply over the range of conditions
of the runs.
The pH generally declined during aeration, and sometimes it
declined sharply to levels between 2.5 and 5.0. In these cases,
the disappearance of iodate demand was very rapid.
The decline of iodate demand shows that sulfide, sulfite,
and hydrosulfide were approaching zero. These are .substances
that react with oxygen fairly readily although the rate limiting
steps in the reaction mechanisms are not known. The iodate
demand test is probably an adequate indication of the tendency of
the waste water to deplete oxygen from receiving waters by
chemical reactions. There remains the possibility that
biochemical processes can cause oxygen depletion, and use of a
properly established biochemical oxidation treatment system
should further reduce the COD, as is currently practiced.
14.3.2 Statistical Evaluation
Table 14-15 and Figures 14-2 through 14-6 show the results
obtained from a statistical analysis of the treatment data. A
statistical analysis is included for zinc, chromium, and iodate
demand although the treatment technology tested was specifically
designed to evaluate chemical oxygen demand and total suspended
solids removal only.
178
-------
TABLE 14-15. COMPARISON Bh'lWEtN PROPOSED BAT LIMITATIONS AND
ESTIMATED TREATABHiITY PERFORMANCE FOR
THE SODIUM HYDKOSULFITE SUBCATEQORY (FORMATE
STREAM: Effluent
Concentration Basis
(rcg/1)
Pollutant
Proposed BAT
Maximum
30-Day Average
Est. Treat. Performance
30-Day Average
Chemical Oxygen Demand,
COD
Total Suspended Solids,
TSS
2600
25
3000
110
179
-------
.SUBCATEGORY
Sodium Hydrosulf ite
POLLUTANT
Chemical Oxygen. Demand
PRECIPITANT
Proposed Maxiimin 30-day Average
(mg/1): 2600
95th Percentile (Z = 1.64) (mg/1): 3000
Long Term Average (mg/1): 2600
Standard Deviation of 30-day Averages (mg/1): 270
Probability of Achieving Proposed
Maximum 30-day Average (%): 58
Number of Observations: 17
1.0
0.8
0.7
0.6
0.5
0.4
.3
0.2
.1
0.0 20
--
!•**
^
s
0 22i
/
/
/
/
0 24(
/
f
/
/
A
/
0 26i
/
'
A
/
/
*
/
^
0 28
^
*
^
s
00 30
^*
tf**1
•«**
BS
00 32
30
Maximum 30-day Average (mg/1)
Figure 14-2. Estimated Performance of Proposed
BAT Treatment
180
-------
SUBCATEGORY
Sodium Hydrosulf ite
POLLUTANT
Total Suspended Solids
PRECIPITANT
Proposed Maximum 30-day Average
(mg/1): 25
95th Percentile (Z = 1.64) (mg/1): no
Long Term Average (mg/1): 50
Standard Deviation of 30-day Averages (mg/1): 34
Probability of Achieving Proposed
Maximum 30-day Average (%): 24
Nurrber of Cbservations: 17
Estimated Probability That Any 30-day
Aver-ama lYies Not- FIxra¥»d a fiiwen Maximum
1.0
— _
0.9
0.8
• u.;
0.6
0.5
0.4
0.3
j 0.2
t
' 0.1
0.0
/
/
f
/
^
'
/
'
f
'
f
/
A
/
f
/
s
f
s
'
^
s
s
s
^
*!*
*ts.
**»
•••••
^w«
•Mn
10 20 30 40 50 60 70 80 50 100 llo 120 130
Maximum 30-day Average (mg/1)
Figure 14-3. Estimated Performance of Proposed
BAT Treatment
181
-------
SUBCATEGORY
Sodium Hydrosulf ide
POLLUTANT
Zinc
PRECIPITANT
Proposed. Maxintun 30-day Average
Cmg/D
0.50
95th Percentile (Z = 1.64) (mg/1)
Long Term Average (mg/1)
Standard Deviation of 30-day Averages (mg/1)
Probability of Achieving Proposed
Maximum 30-day Average (%);
Number of Observations:
7.5
2.9
2.8
Not Applicable
17
fci
0.6
0.0
.0.0 2.0 4.0 6 0
Maximum 30-day Average (mg/1)
Figure 14-4. Estimated Performance of Proposed
BAT Treatment
182
8 0
-------
SUBCATEGORY
PRECIPITANT.
Sodium Hydrosulfite
(mg/1)
Proposed Maximum 30-day Average
95th Percentile (Z » 1.64)
Long Term Average (mg/1)
Standard Deviation of 30-day Averages (mg/1)
Probability of Achieving Proposed
Maximum 30-day Average
Nvnfoer of Observations:
0.10
6.5
1.3
3.2
Not Applicable
17
0.0
.Q.Q
2 Q 4;0. 6,0
Maximum 30-day Average (mg/1)
Figure 14-5. Estimated Performance of Proposed
BAT Treatment
183
-------
SUBCATEGORY
Sodium Hydrosulfite
POLLUTANT
lodate Demand
PRECIPITANT
Proposed Maximun 30-day Average
(mg/1):
None
95th Percentile (Z = 1.64)
Long Terra Average (mg/1) :
Standard Deviation of 30-day Averages (ng/1) :
Probability of Achieving Proposed
Maximum 30-day Average
6.4
3.5
1.8
(%) :
Not Applicable
Nurfoer of Cbservations :
17
d Probability That Any 30-day
oes Not Exceed a Given Maximum
3OOOOOOM
• • • • « . .
•J & in a\ -*i CQ \o o
jlj Q «.-.
JO 02
n 1
u.x
0.0
M»«<»
,„
X^
x*
X
/*
/^
^/
^
/•
/
/
/
jf
s
/
/
4+
^X
4*
+*
Maxiimm 30-day Average (rtq/1)
Figxire 14-6. Estimated Performance of Proposed
BAT Treatment
184
6 0
-------
14.3.3 Conclusibns
Physical
chemical
The treatability test results indicate that
aeration does not significantly reduce the overall
oxyqen demand in the sodium hydrosulfite waste water
14-14 shows the estimated performance 30-day average which
exceeds the proposed BAT maximum 30-day average concentration
(achieved with biochemical oxidation) by 400 mg/1. It can be
concluded from these tests that biochemical oxidation is an
essential waste treatment process for the reduction of COD for
this subcategory.
Review of the results also indicates that dual media
filtration removes substantial quantities of suspended solids
although a greater removal appears achievable when preceded by
biochemical treatment.
The experimental results presented herewith represent the
outcome of the particular set of experiments conducted during the
available time frame. Hence, the results do not represent the
actual performance capabilities of the proposed BAT treatment.
185
-------
-------
APPENDIX A
STATISTICAL SUMMARIES OF
TREATMENT DATA
A-l
-------
Table Mo: A-01
Statistical Summary of Treatment Data for
Chlor-Alkali (Diaphragm Cell)
Treatment Effluent
Parameter (mg/1)
No
Win
Avg
Max
Stdv C.Var
Total Suspended Solids 15
Nickel 15
Total Chromium 15
Lead 10
1.00 30.44 92.00 29.20 0.96
0.05 0.38 1.21 0.37 0.97
0.04 0.08 0.14 0.03 0.41
0.05 0.08 0.29 0.08 0.96
A-2
-------
Table No: A-02
*
Statistical Summary of Treatment Data for
•Hydrofluoric Acid
Treatment Effluent
Parameter (mg/1)
No
Win
Avg
Max Stdv C.Var
Total Suspended Solids 17 11.00 149.74 363.00 106.98
Nickel 17 0.03 0.49 1.10 0.36
Zinc 17 0.01 0.07 0.32 0.09
Total Chromium 17 0.01 0.07 0.15 0.04
Fluoride 13 67.00 89.69 109.00 13.10
0.71
0.72
1.26
0.53
0.15
A-3
-------
Table No: A-04
Statistical Summary of Treatment Data for
Titanium Dioxide (Chloride Process)
Treatment Effluent
Parameter (mg/1)
No
Min
Avg
Max
Stdv C.Var
Total Suspended Solids 16
Nickel 16
Zinc 16
Total Chromium 14
Copper 16
Iron 16
2.00 9.85 20.00 5.69 0.58
0.06 0.09 0.13 0.02 0.19
0.02 0*04 0.08 0.02 0.48
0.03 0.05 0.12 0.02 0.49
0.03 0.04 0.05 0.01 0.21
0.03 0.18 0.69 0.15 0.87
A-4
-------
Table No: A-06
Statistical Summary of Treatment Data for
Chrome Pigments
Treatment Effluent
Parameter (mg/1)
No
Min
Avg
Max
Stdv C.Var
Total Suspended Solids
Zinc
Total Chromium
Lead
Cadmium
14
14
14
14
13
1.
0.
0.
0.
0.
00
02
01
03
01
4
0
0
0
0
.44
.04
.'04
.10
.03
9.
0.
0.
0.
0.
60
07
18
24
10
2.
0.
0.
0.
0.
51
02
04
06
02
0.
0.
1.
0.
0.
57
39
09
59
69
A-5
-------
Table No: A-08
Statistical Summary of Treatment Data for
Sodium Bichromate
Parameter (mg/1)
Treatment Effluent
No
= =: =sss==r
Min
Avg
Max Stdv C.Var
Total Suspended Solids 18
Nickel 17
Total Chromium 13
Hexavalent Chromium 14
3.00 175.29 832.40 277.13 1.58
0.05 0.09 0.50 0.11 1.20
0.09 0.25 0.79 0.19 0.77
0.00 0.12 0.90 0.26 2.16
A-6
-------
Table No: A-11C
Statistical Summary of Treatment Data for
Copper- Sulfate
Caustic Treatment
Treatment Effluent
Parameter (mg/1)
No
Min
Avg
Max
Stdv G.Var
Total Suspended Solids 11 1.10
Nickel 11 0.05
Copper 10 0.07
Selenium 10 0.11
5.44 12.60 3.69 0.68
0.09 0.20 0.05 0.53
0.25 0.66 0.18 0.72
0.13 0.16 0.01 0.11
A-7
-------
Table No: A-11L
Statistical Summary of Treatment Data for
Copper Sulfate
Lime Treatment
Treatment Effluent
Parameter (mg/1)
No
Min
Avg
Max
Stdv C.Var
Total Suspended Solids 12 0.90
Nickel 12 0.05
Copper 12 0.06
Selenium 12 0.10
4.55 14.00 3.78 0.83
0.11 0.41 0.11 0.96
0.18 0.56 0.14 0.77
0.12 0.16 0.02 0.15
A-8
-------
Table No: A-12A
Statistical Summary of Treatment Data for
Nickel Sulfate
Alkaline Treatment
Treatment Effluent
Parameter (mg/1)
No
Min
Avg
Max
Stdv C.Var
Total Suspended Solids 14 2.00 27.79 105.00 28.55 1.03
Nickel 14 0.01 0.16 0.55 0.18 1.12
A-9
-------
Table Nos A-14
Statistical Summary of Treatment Data for
Sodium Bisulfite
Unfiltered Supernatant
Parameter (mg/1)
No
Min Avg Max Stdv C.Var
Total Suspended Solids 16 120.00 242.81 690.00 169.82 0.70
Zinc 16 0.12 0.92 4.00 ' 1.23 1.34
A-10
-------
Table No: A-14
Statistical Summary of Treatment Data for
Sodium Bisulfite
Unfiltered Supernatant
Maximum Aeration Time
Parameter (mg/1)
No
Min
Avg
Max Stdv C.Var
Chemical Oxygen Demand 15 80.00 453.33 960.00 264.76 0.58
Sulfite(lodate) 17 2.00 24.59 84.00 22.64 0.92
A-ll
-------
Table No: A-15
Statistical Summary of Treatment Data for
Sodium Hydrosulfite
Treatment Effluent
Parameter (mg/1)
No
Min
Avg
Max Stdv C.Var
Total Suspended Solids 17 0.30 29.32 126.00 32.89 1.12
Zinc 17 ; 0.03 2.11 9.93 3.18 1.50
Total Chromium 17 0.01 1.25 9.09 2.88 2.30
A-12
-------
Table No: A-15
Statistical Summary of Treatment Data for
Sodium Hydrosulfite
Unfiltered Supernatant
Maximum Aeration Time
Parameter (mg/1)
No Min Avg Max Stdv C.Var
Chemical Oxygen Demand 17 776.0 2503.4 4500.0 1160.6 0.46
Sulfite(lodate) 17 0.20 3.50 16.60 5.19 1.48
A-13
-------
-------
APPENDIX B
IODATE DEMAND CURVES FOR
SODIUM BISULFITE
B-l
-------
12
15
18
21
24
Time (Hrs)
1800
1700
1600
1500
1400
1300
1200
1100 g
1000
900
800
700
600
500
400
300
200
100
\ •
8
27
Figure B-l . Effect of aeration 'on sulfite concentration,
(airflow rate: 11.5 SCFH)
B-2
-------
10
Sodium Bisulfite Subcategory
Batch 1; Run 2
1800
1700
9 12 15
Time (Hrs)
18
21
24
Figure B-2
Effect of aeration on sulfite cxancentration
(airflow rate: 46.0 SCFH)
1
8
n
B-3
-------
8
OP
100 -
90 -
80
70
60
50
40 -
30 -
20 -
10 -
Sodium Bisulfite Subcategory
'Batch 2; Run 1
SO3 Removed
SO., Concentration
- 200
•a. 100
024
Figure B-3
8 10 12 14 16 18 20 22 24
Time (Hrs)
Effect of aeration on sulfite concentration
(airflow rate: 35.0 SCFH)
H
I
•S
B-4
-------
100
Sodium Bisulf ite Subcategory
Batch 2; Run 2
I •
8
40 -
30 -
20 -
10 -,
9 12 15
Time (Hrs)
900
SO3 Concentration
- 300
- 200
100
18 21 24
Figure B-4
Effect of aeration on sulfite concentration
(airflow rate: 11.5 SCFH)
B-5
-------
dP
100
90
80
70
60 -
50 -
40 -
30 -
20 -
10
Sodium Bisulfite Subcategory
Batch 2; Run 3
SO, Concentration
- 1000
900
- 800
- 700
600 g
- 500
- 400
- 300
- 200
100
1
12 15 18
Time (Hrs)
21 24
Figure B-5 .
Effect of aeration on sulfite concentration
(airflow rate: 23.0 SCFH)
-------
100
90
80
Sodium Bisulfite Subcategory
Batch 3; Run 1
6
12 15 18
Time (Hrs)
21 24 27
Figure B-6
Effect of aeration on sulfite concentration
(airflow rate: 35.0 SCFH)
30
B-7
-------
100 -
Sodium Bisulfite Subcategory
Batch 3; Run 2
SO3 Removed
SO3 Concentration
20
10
12 15
Time (Hrs)
Figure B-7 . Effect of aeration on sulfite concentration
(airflow rate: 46.0 SCFH)
B-8
-------
100
8
*>
Sodium Bisulfite Subcategory
Batch 4; Run 1
SO3 Concentration
Figure B-8
12 15
Time (Hrs)
Effect of aeration on sulfite concentration
(airflow rate: 57.0 SCFH)
B-9
-------
100 -
df>
Sodium Bisulfite Subcategory
Batch 4; Run 2
SO., Removed
S03 Concentration
I
i
Figure B-9
12 15 18
Time (Hrs)
Effect of aeration on sulfite concentration
(airflow rate: 11.5 SCFH),
B-10
-------
8
dP
100
Sodium Bisulfite Subcategory
Batch 4; Run 3
SO., Concentration
I
1
8
12 15
Time (Hrs)
Figure B-10. Effect of aeration on sulfite concentration.
(airflow rate: 35.0 SCFH)
-------
100 -
Sodium Bisulfite Subcategory
Batch 5; Run 1
.SO, Removed
I
2000
1800 ]
1600 .
S03 Concentration
15
18
21
24
.400
200
Time (Hrs)
Figure B-ll. Effect of aeration on sulfite concentration
(airflow rate: 23 SCFH)
B-12
-------
100 -
Sodium Bjsulfite Subcategory
Batch 5; Run 2
12 15.
Time (Hrs)
SO- Concentration
21
24
Figure B-12. Effect of aeration on sulfite concentration
(airflow rate: 46 SCRH)
1
B-13
-------
100
80
60
50
30
20
10
Sodium Bisulfite Subcategory
Batch 5; Run 3
""^ SO3 Removed
-
-
:
-
-
2000
1800
1600
1400
1200
1000
800
600
400
icentration . 2QO
centration (m
§
Sf
9 12 15
Time (Hrs)
Figure B-13. Effect of aeration on sulfite concentration
(airflow rate: 57 SCFH)
-------
dp
100
90
Sodium Bisulfite Subcategory
Batch 6; Run 1
SO., Concentration
Time (Hrs)
Figure B-14. Effect of aeration on sulfite concentration.
(airflow rate: 11.5 SCFH)
-------
100 -
<#>
5600
5200
4800
4400
4000
3600
3200
2800
2400
2000
1600
1200
800
400
Time (Hrs)
Figure B-15. Effect of aeration on sulfite concentration
(airflow rate: 46 SCFH)
B-16
-------
Sodium Bisulfite Subcategory
Batch 7; Run 1
SO., Concentration
I
8
12. 15 18
Tine (Hrs)
Figure B-16. Effect of aeration on sulfite concentration
(airflow rate: 23 SCFH)
B-17
-------
100 -
Sodium Bisulfite Subcategory
Batch 7; Run 2
SO., Concentration
I
9 12 15
Time (Hrs)
Figure B-17. Effect of aeration on sulfite concentration
(airflow rate: 57.0 SCFH)
B-18
-------
APPENDIX C
IODATE DEMAND CURVES FOR
SODIUM HYDROSULFITE
C-l
-------
100
90
80
70
60
50
40
30
20
10 -
Sodium Hydrosulfite Subcategory
Batch 1; Run 1
S03 Concentration
130
120
110 P
100 ~
90 '$
80
70
60
50
40
30
20
10
Time (Hrs)
Figure C-l . Effect of aeration on sulfite concentration
(airflow rate: 57.0 SCFH)
C-2
-------
100 -
Sodium Ifydrosulf ite Subcategory
Batch 1; Run 2
SO3 Concentration
Timei (Hrs)
Figure C-2 . Effect of aeration on sulfite concentration.
(airflow rate: 57.0 SCFH)
4 -3
-------
100 -
90 -
80
70
60
8 50
40
30
20
10
Sodium Hydrosulfite Subcategory
Batch 2; Run 1
SO., Removed
SO., Concentration
140
130
120
110
100
90
80
70
60
50
40
30
20
10
12 16 20 24 28
Time (Hrs)
32 36 40
1
ro
Figure C-3 . Effect of aeration on sulfite concentration
(airflow rate: 57.0 SCEH)
0-4
-------
100
Sodium Hydrosulfite Subcategory
Batch 2; Run 2
S03 Removed
SO., Concentration
I 1
0 4 8 12 16 20 24 28 32 36
Time (Hrs)
20 -
10 -
Figure C-4 . Effect of aeration on sulfite concentration
(airflow rate: 57.0 SCFH)
C-5
-------
Sodium Hydrosulf ite Subcategory
Batch 3; Run 1
SO_ Concentration
10 -
0 4 8 12 16 20 24 28 32 36 40 44 48 52 56
Time (Hrs)
Figure c-5 . Effect of aeration on sulfite concentration
(airflow rate: 46.0 SCFH)
1500
1400
1300
1200
1100
1000 &
900 •!
800
700
600
500
400
300
200
100
60
c-e
-------
Sodium Hydrosulfite Subcategory
Batch 3; Run 2
100 -
S0_ Concentration
1200
1140
1080
1020
960
900
840
780
720
660
600
540
480
420
360
300
240
180
120
60
1—I
>
1
0 4 8 12 16 20 24 28 32 36 40 44 48 52
Time (Hrs)
Figure C-6 . Effect of aeration on sulfite concentration
(airflow rate: 46.0 SCFH)
-------
ff
20 -
10 -
Sodium Hydrosulf its Subcategory
Batch 4; Run 1
S03 Removed
S0_ Concentration
4800
4400
4000
3600 :
3200 :
2800
i
2400
2000
1600
1200
800
400
H1
0 2 4 6 8 10 12 14 16 18 20 22 24 26 28
Time (Hrs)
Figure C-7 . Effect of aeration on sulfite concentration
(airflow rate: 46.0 SCFH)
30
08
-------
1
o\o
Sodium Hydrosulfite Subcategory
Batch 5; Run 1
•I
I
8 10 12 14 16 18 20 22 24
Time (Hrs)
Figure c-8 '•.. Effect of aeration on sulfite concentration
(airflow rate: 46.0 SCFH)
C-9
-------
210
200
190
180
170
160
30
20
10
Time (Hrs)
Figure C-9 . Effect of aeration on sulfite concentration
(airflow rate: 46.0 SCFH)
C-10
-------
8
df>
Sodium Hydrosulfite Subcategory
Batch 6; Run 2
S03 Removed
SO3 Concentration
Time (Hrs)
10 -
Figure C-ll. Effect of aeration on sulfite concentration
(airflow rate: 35.0 SCEH)
001
-------
100
Sodium Hydrosulfite Subcategory
Batch 7; Run 1
SO3 Concentration
I
20 24 28
Time (Hrs)
Figure C-12. Effect of aeration on sulfite concentration
(airflow rate: 35.0 SCFH)
C-12
-------
Sodium Hydrosulf ite Subcategory
Batch 7; Run 2
SO3 Concentration
4000 J.
3600 jj
3200 jj
2800
2400
2000
1600
1200
800
400
CO
8
0 4 8 12 16 20 24 28 32 36 40 44 48 52 56 60 64
Time (Hrs)
Figure C-13. Effect of aeration on sulfite concentration
(airflow rate: 35.0 SCFH)
c-ia
-------
PI
o
to
Sodium Hydrosulfite Subcategory
Batch 8; Run 1
SO_ Concentration
1400
1300
1200
1100
1000
900
800
700
600
500
400
300
200
100
8 10 12 14 16 18 20
Time (Hrs)
Figure C-14. Effect of aeration on sulfite concentration
(airflow rate: 35.0 SCFH)
C-14
-------
00
100
90
80
70 -
60 -
50 -
40 -
30 -
20 -
10 -
Sodium Hydrosulfite Subcategory
Batch 9; Run 1
A
SO., Concentration
520
480
440 p
400 ~
360 +J
I
320 g
280 O
r'
240 S
200
160
120
80
40
3 6 9 12 15 18 21 24 27 30 33 36
Time (Hrs)
Figure C-15. Effect of aeration on sulfite concentration
(airflow rate: 23.0 SCFH)
C-15
-------
Sodium Hydrosulfite Subcategory
SO_ Concentration
3 4
Tims (Hrs)
Figure C-18. Effect of aeration on sulfite concentration
(airflow rate: 11.5 SCFH)
------- |