PB88-131289
TECHNICAL RESOURCE DOCUMENT
TREATMENT TECHNOLOGIES FOR
CORROSIVE-CONTAINING.WASTES
VOLUME 2
Alliance Technologies Corporation
Bedfo.rd , .MA
Dec 87
U.S. DEPARTMENT OF COMMERCE
National Technical Information Service
-------
EPA/600/2-87/099
December 1987
PB88-1312d9
TECHNICAL RESOURCE DOCUMENT
TREATMENT TECHNOLOGIES FOR CORROSIVE-CONTAININ3 WASTES
Volume II
by
Lisa Wilk
Stephen Palmer
Marc Breton
Alliance Technologies Corporation
Bedford, MA .01730
EPA Contract Number 68-02-3997
Project Officer
Harry M. Freeman
Alternative Technologies Division
Hazardous Waste Engineering Research Laboratory
Cincinnati, OH 45268
HAZARDOUS WASTE ENGINEERING RESEARCH LABORATORY
OFFICE OF RESEARCH & DEVELOPMENT
U.S. ENVIRONMENTAL PROTECTION AGENCY
CINCINNATI, OHIO 45268
-------
TECHNICAL *E*O*T DATA
i am Mr n**ru trforr
MO.
EPA/600/2-87/099
TIT4.1 AMOSUBTITLi
pels*?? VaTSB 9/AS
M*OB1 DAT I , _ 0 _ •-
December 1987
Technical Resource Document: Treatment Technologies
for Corrosive-Containing Wastes Volume II
»o***i*c OM6AM iXAtio* coot
Wilk, Stephen Palmer, Marc Breton
PIIMOIIMIN6 OH6ANIZATIO* KAMI AMD AOO*CW
Alliance Technologies Corporation
213 Burlington Road
Bedford. Ma 01730
M6!'
68-02-3997
13. »»0*»0«INC AGIMCT KAMI AND
HAZARDOUS WASTE ENGINEERING RES. LAB.
OFFICE OF RESEARCH AND DEVELOPMENT
U.S. ENVIRONMENTAL PROTECTION AGENCY
CINCINNATI, OH A5263
13 TV»| Of MlPOftT AMD »t«>0a CSVIMIB
14
COOC
EPA/600/12
MOTES
It
This Technical Resource Document (THD) for wastes containing corrosives
is one in a series of five document* which evaluate waste management
alternatives to land disposal. In addition to this THD for corrosive wastes,
the other four TRDs in the series address land disposal* alternatives for the
following waste categories: dioxins; solvents; nonsolvent halogenated
organics; and metals/cyanides. The purpose of these documents is to provide a
comprehensive source of information that can be used by environmental
regulatory agencies and others in evaluating available waste management
options, which include waste minimization and recycling as well as treatment.
Emphasis has been placed on the collection and interpretation of performance
data for proven technologies. However, all potentially viable technologies
are identified and discussed. When possible, cost and available capacity data
are provided to assist the user of the T&Ds in assessing the applicability of
technologies to specific waste streams.
•IT WOHM AMD OeCUMf NT ANALYSIS
i Fvld Growf
. DlSTMiftl/TlON BTATiMIMT
RELEASE TO PUBLIC
It HCW«IT» C^ASt (T*u*rF*fti
UNCLASSIFIED
UNCLASSIFIED
-------
NOTICE
This document has been reviewed in accordance with
U.S. Environmental Protection Agency policy and
approved for publication. Mention of trade names
or commercial products does not constitute endorse-
ment or recommendation for use.
11
-------
FOREWORD
As hazardous waste continues to be one of the more prominent
environmental concerns to the people of the United States and other
countires throughout the world, there are continuous needs for research
to characterize problems and develop and evaluate alternatives to addressing
those problems. The program of the Hazardous Waste Engineering Research
Laboratory are designed to contribute to satisfying these research needs.
This Technical Resource Document for Treatment Technologies for
Corrosives-Containing Wastes compiles available information on those
technologies. It is intended to provide support for the land disposal
prohibition, currently being considered by the EPA, and to provide
technical information for those individuals and organizations concerned
with the subject waste streams. Those wishing additional information on
the various technologies should contact the Hazardous Waste Engineering
Research Laboratory.
Thomas R. Hauser
' ' . Director
Hazardous Waste Engineering Research Laboratory
-------
ABSTRACT
This Technical Resource Document (TRD) for wastes containing corrosives
is one in a series of five documents which evaluate waste management
alternatives to land disposal. In addition to this TRD for corrosive wastes,
the other four TRDs in the series address land disposal alternatives for the
following waste categories: dioxins; solvents; nonsolvent halogenated
organic*; and metals/cyanides. The purpose of these documents is to provide a
comprehensive source of information that can be used by environmental
regulatory agencies and others in evaluating available waste management
options, which include waste minimization and recycling as well as treatment.
Emphasis has been placed on the collection and interpretation of performance
data for proven technologies. However, all potentially viable technologies
are identified and discussed. When possible, cost and available capacity data
are provided to assist the user of the TRDs in assessing the applicability of
technologies to specific waste streams.
Neutralization processes are the most commonly applied methods for
managing corrosive wastes. Alkaline reagents commonly used to neutralize
strongly acidic wastes (>5,000 mg/L) are lime and caustic soda. Limestone
treatment is often used to neutralize more dilute acidic wastes. Acids, such
as sulfuric, hydrochloric, and carbonic acid, are generally used for the
neutralization of alkaline waste streams. The applicability of each of these
neutralization reagents to corrosive wastes is discussed in specific
sections, with attention given to restrictive waste characteristics which
impact both the ability of the technology to effectively treat the specific
waste under consideration and the cost of application.
The recovery/reuse technologies examined in detail include the following:
• Evaporation/Distillation • Electrodialysis
• Crystallization • Reverse Osmosis
• Ion Exchange • Donnan Dialysis
• Solvent Extraction • Thermal Decomposition
These technologies, like the neutralization technologies, are described in
terms of their effectiveness in treating corrosive wastes, their associated
environmental impact, and advantages/Iimitations arising from specific waste
characteristics. Although emphasis is placed on performance data, cost data
and the capacity/state o£ development of each technology is presented to
assist in evaluating ana selecting options. Approaches to the selection of
neutralization and recovery/reuse options are reviewed in the final section of
this TRD.
iv
-------
CONTENTS
Foreword ..... iii
Abstract iv
Figures vi
Tables xi
Acknowledgement xx •
Project Summary . . 1
1. Introduction 1-1
2. Identification of RCRA Corrosive Wastes 2-1
3. Corrosive Waste Sources, Generation, and Management 3-1
Corrosive Waste Sources 3-1
RCRA Corrosive Waste Generation and Management 3-24
4. Neutralization Treatment Technologies 4-1
General Considerations 4-2
Mixing of Acid and Alkali Wastes 4-40
Limestone Treatment 4-56
Lime Slurry Treatment 4-76
Caustic Soda Treatment . . . ". 4-99
Mineral Acid Treatment '. 4-117
Carbonic Acid Treatment . . . ' 4-137
5. Recovery/Reuse Technologies • • 5-1
Introduction » . . . 5-1
Evaporation/Distillation 5-3
Crystallization . 5-28
Ion Exchange ..'. •• 5-47
Electrodialysis 5-84
Reverse Osmosis 5-115
Donnan Dialysis and Coupled Transport 5-138
Solvent Extraction 5-153
Thermal Decomposition 5-177
Waste Exchange 5-195
6. Considerations for System Selection 6-1
General Considerations 6-1
Waste Management Process Selection ..... 6-3
-------
FIGURES
Number
Page
4.1.1 Neutralization of waste by calcium hydroxide 4-6
4.1.2 Alternative concepts for wastewater equalization 4-9
4.1.3 Treatment trains for corrosive wastes 4-15
4.1.4 Treatment of concentrated organics and oily wastewater
emulsions 4-16
4.1.5 Approximate ranges of applicability of treatment techniques
as a function of organic concentration in liquid waste
streams .........'. 4-19
4.1.6 Investment cost for flocculation/clarification units 4-25
4.1.7 Settling rate curves 4-27
0 -
4.1.8 Investment cost for sludge storage/thickening units 4-30
4.1.9 Hardware cost for recessed plate filter presses 4-30
4.1.10 Annual cost for disposal of industrial sludge 4-31
4.2.1 Conventional wastewater treatment system for electroplating . . 4-42
4.2.2 Construction costs for reinforced concrete reactor . 4-50
4.2.3 Investment cost for continuous neutralization unit 4-51
4.3.1 Upflow limestone bed neutralization process configurations . . 4-58
4.3.2 Single bed, upflow limestone treatment system schematic .... 4-60
4.3.3 Flow diagram of complete biochemical oxidation and
limestone neutralization process 4-67
4.3.4 Continuous limestone powder neutralization process ...... 4-70
VI
-------
FIGURES (continued)
Number Page
4.3.5 Construction cost for limestone powder feed system 4-73
4.4.1 Flowsheet of a lime slurry system 4-79
4.4.2 Small volume, high calcium, hydrated lime slurry
treatment system 4-80
4.4.3 Investment costs for hydrated lime feed systems 4-94
4.5.1 Neutralization of waste by sodium hydroxide 4-104
4.5.2 Titration curve for the neutralization of a 12 sulfuric
acid solution with sodium hydroxide and sodium carbonate . . 4-105
4.5.3 Process schematic of two-stage neutralization system 4-108
4.5.4 Process schematic of sodium hydroxide batch treament system . . 4-111
4.6.1 Coal-liquefaction facility wastewater treatment system .... 4-123
4.6.2 Process schematic showing plating/etching waste treatment
system 4-126
4.6.3 Flowsheet for proposed recovery of waste peeling sludge .... 4-130
4.7.1 Solubility of carbon dioxide in water . 4-140
4.7.2 Submerged combustion pilot unit 4-142
4.7.3 Compressed carbon dioxide treatment system 4-145
4.7.4 Construction cost for recessed plate filter press 4-148
5.1.1 Process flow diagram for commonly used evaporation systems . . 5-4
5.1.2 Process flow diagram for a two-stage vacuum evaporation/
distillation system to recover spent HN03/HF pickling
liquors . . 5-7
5.1.3 Process flow diagram for a single-stage vacuum evaporation/
distillation system to recover spent HN03/HF pickling
liquors 5-8
5.1.4 Nitric and hydrofluoric acid concentrations in cumulative
distillate fraction versus original volume distilled .... 5-10
VII
-------
FIGURES (continued)
Number Page
5.1.5 Flow diagram of evaporative recovery system installed at
Superior Plating, Inc 5-17
5.2.1 Flow diagram of crystallization system for recovery of
sulfuric acid pickling liquor 5-29
5.2.2 Flow diagram of two-stage recovery system for nitric-
hydrofluoric acid pickling liquor 5-31
5.2.3 Flow diagram of crystallization system for the recovery
of caustic soda aluminum etching solution 5-32
5.2.4 Solubility of ferrous sulfate in various sulfuric acid
concentrations 5-34
5.2.5 Solubility of iron in mixed acid containing 12.5 percent
nitric acid and different amounts of free hydrofluoric
acid at varying temperatures 5-36
5.2.6 Delay of formation of visible crystals in oversaturated
mixed acid versus initial concentration of iron 5-38
5.3.1 Cocurrent ion exchange cycle • 5-49
5.3.2 Effect of revising acid regenerant on chemical efficiency . . . 5-50
5.3.3 Schematic of a fixed bed reverse flow ion exchange system
for the recovery of chromic acid from a dilute solution . . . 5-52
5.3.4 Basic operation of the acid purification unit using
a continuous bed RFIE system 5-53
5.3.5 Schematic of inverse mode of operation 5-68
5.3.6 Schematic of normal mode of operation 5-69
5.4.1 Schematic of a coocentrating-diluting electrodialysis process . 5-85
5.4.2 Schematic of an ion transfer/ion substituting electrodialysis
process 5-86
5.4.3 Diagram of an electrolytic electrodialysis cell . 5-88
5.4.4 Flow diagram for electrodialysis treatment of chrome line
at(Seaboard Metal Finishing Company 5-92
vnx
-------
FIGURES (continued)
Number Page
5.4.5 Relationship between current and product concentration
during operation of ED unit for chromic acid recovery 5-95
5.4.6 Product concentrations achieved during operation of the
pilot scale ED unit at Seaboard Metal Finishing Company .... 5-96
5.5.1 Reverse osmosis membrane module configurations 5-117
5.5.2 Reverse osmosis module arrangements 5-119
5.5.3 Schematic of reverse osmosis/evaporation system used to
recover zinc cyanide plating solution at American
Electroplating Company 5-127
5.5.4 Annual.operating costs for various RO system conversions . . . 5-134
5.6.1 Membrane stack unit 5-139
5.6'.2 Co-transport process for chromate recovery 5-141
5.6.3 Counter-transport process for metal cation recovery 5-142
5.6.4 Membrane test system 5-146
5.7.1 Generalized flow diagram of solvent extraction process .... 5-154
5.7.2 Flow diagram of AX process 5-156
5.7.3 Flow diagram of Kawasaki process 5-158
5.7.4 Distribution of nitrate between 752 TBP in kerosene and
water as a function of concentration of total nitrate
in the aqueous phase •... 5-160
5.7.5 Distribution of sulfuric acid between 100Z TBP and water
as a function of total added hydrofluoric acid concentration . 5-161
5.7.6 Distribution of HF between 100Z TBP and water as a function
of added concentration of hydrofluoric acid and
sulfuric acid ••>. 5-163
5.7.7 Extraction of nitric acid and hydrofluoric acid by 75Z TBP
in kerosene from aqueous solution of metal sulfates as a
function of total aqueous nitrate and fluoride
concentration 5-164
IX
-------
FIGURES (continued)
Number Page
5.7.8 The distribution of nitric acid and hydrofluoric acid
between 752 TBP in kerosene and water as a function of
aqueous concentration of nitric and hydrofluoric acids . . . 5-165
5.7.9 Extraction of nitric acid with 752 TBP in kerosene as a
function of iron content 5-167
5.7.10 Extraction of hydrofluoric acid with 752 TBP in kerosene
as a function of iron content 5-168
5.8.1 General flow diagram of HC1 regeneration by thermal
decomposition 5-178
5.8.2 Flow diagram of a thermal decomposition process using a
rotating furnace 5-181
6.1.1 Corrosive waste management options 6-2
-------
TABLES
Number Page
1 Waste Management Alternatives to Land Disposal ........ 3
2 Corrosive Waste Quantity Handled by Industrial
Classification ....................... 7
3 Management Practice Summary for Corrosive Waste
National Estimates ..................... 9
4 Summary of Neutralization Technologies ............ 12
S Summary of Recovery/Reuse Technologies for Corrosive
Wastes ........................... 17
1.1 Scheduling for Promulgation of Regulations Banning
Land Disposal of Specified Hazardous Wastes ......... 1-2
2.1 Corrosive and Potentially Corrosive Waste Materials ...... 2-3
2.2 Dissociation Constants and pHs of Acids and Bases in
Aqueous Solutions at 25 °C .................. 2-7
3.1.1 Industrial Consumption of Primary Acid and Alkali Chemicals
By End Use ......................... 3-2
3.1.2 Phosphate Fertilizer Industry: Gypsum Pond Waste
Characteristics ....................... 3-7
3.1.3 Summary of Constituent Concentrations from Sampled
Iron and Steel Plants .................... 3-13
3.1.4 Corrosive Waste Characteristics in the Nonferrous Metal
Forming Industries ..... ' ...... . ......... 3-15
3.1.5 Surface Treatment Unit Operations ............... 3-17
3.1.6 Classification of Types of Discharge from Surface
Treatment Processes .................. ... 3-18
XI
-------
TABLES (continued)
Number Page
3.1.7 Acids Used in Metal Pickling Solutions 3-19
3.1.8 Raw Waste Characteristics - Variations Among Plants Involved
in the Manufacturing and Coating of Metal Products 3-21
3.2.1 Management Practice Summary for Corrosive Wastes
National Estimates 3-27
3.2.2 Corrosive Waste Quantity Handled and Recycled by
Industrial Classification 3-30
3.2.3 Number of Facilities Handling and Disposing of Corrosive
Waste by Industrial Classification 3-31
4.1.1 Acid/Alkaline Neutralization Agent Characterization 4-5
4.1.2 Summary of Organic Residual Treatment Process 4-20
4.1.3 Laboratory Analysis of the Neutralization of Spent
Acid Plating Waste 4-26
4.1.4 Summary of Sludge Dewatering Device Characteristics 4-29
4.1.5 Compatibility of Selected Waste Categories with
Different Waste Solidification/Stabilization Techniques . . . 4-33
4.1.6 Present and Projected Economic Considerations for Waste
Solidification/Stabilization Systems 4-34
4.1.7 Encapsulated Waste Evaluated by the U.S. Army Waterways
Experiment Station 4-37
4.1.8 Estimated Costs of Encapsulation 4-37
4.2.1 Characterization of Mutual Neutralization Waste Streams
From Coal-Fired Plant Applications 4-45
4.2.2 Summary of Application of Mutual Neutralization to Metal
Finishing Wastes . ; 4-47
4.2.3 Average Quality of Acid/Alkali Waste Streams 14-48
4.2.4 Mutual Neutralization Treatment Costs 4-53
4.3.1 Summary of High Calcium and Dolomitic Limestone Properties . . #-57
Xll
-------
TABLES (continued)
Number Page
4.3.2 Summary of Typical Operating Parameters 4-61
4.3.3 Summary of Limestone Bed Treatment Performance Data 4-63
4.3.4 Case Study 1: Bench Scale Limestone Treatment Data Summary . . 4-64
4.3.5 Summary of Limestone Treatment Pilot Plant Data 4-66
4.3.6 Summary of Limestone Neutralization Experiments 4-69
4.3.7 Continuous Limestone Powder Treatment Costs 4-71
4.4.1 Summary of High Calcium and Dolomite Quicklime Properties . . . 4-77
4.4.2 Summary of Typical Lime-Slurry Operating Parameters 4-82
4.4.3 Waste Carbide Lime Composition 4-84
4.4.4 Elemental and Anion Variation in U.S. Cement Kiln Dust .... 4-86
4.4.5 Summary of Lime-Slurry Neutralization Data 4-87
4.4.6 Full-Scale Automatic Lime Neutralization System
Characteristics 4-88
4.4.7 Nominal Design Criteria for Two-Stage Lime Neutralization
System . 4-91
4.4.8 Typical Operating Data for the Two-Stage Lime
Neutralization of Acidic Process Cooling Water 4-92
4.4.9 Continuous Hydrated Lime Neutralization Treatment Costs .... 4-93
4.4.10 Advantages and Disadvantages of Lime-Slurry Neutralization . . 4-96
4.5.1 Physical Constants of Pure Sodium Hydroxide 4-100
4.5.2 Sodium Hydroxide Neutralization: Summary of Typical
Operating Parameters 4-102
4.5.3 Summary of Sodium Hydroxide Neutralization Data 4-107
4.5.4 Continuous Caustic Soda Neutralization Treatment Costs .... 4-112
xiii
-------
TABLES (continued)
Number Page
4.5.5 Sodium Hydroxide Sludge Generation Factors 4-113
4.5.6 Advantages and Disadvantages of Caustic Soda Neutralization . . 4-115
4.6.1 Sulfuric Acid Physical Properties 4-118
4.6.2 Hydrochloric Acid Properties 4-120
4.6.3 Summary of Coal-Liquefaction Facility Wastewater
Characteristics 4-124
4.6.4 Summary of Neutralization/Precipitation Test Data 4-127
4.6.5 Summary Data on Caustic Tomato Peeling Operating and
Acid Usage 4-129
4.6.6 Continuous Mineral Acid Treatment Costs Using Sulfuric
and Hydrochloric Acids as the Neutralizing Agents 4-131
4.6.7 Advantages and Disadvantages of Sulfuric Acid Neutralization . 4-134
4.6.8 Advantages and Disadvantages of Hydrochloric Acid
Neutralization 4-135
4.7.1 Summary of Carbon Dioxide Physical-Property.Data 4-138
4.7.2 Summary of Automatic Carbon Dioxide Neutralization
Process Data 4-143
4.7.3 Compressed Carbon Dioxide Treatment Costs 4-146
4.7.4 Advantages and Disadvantages of Carbon Dioxide Neutralization . 4-150
5.1.1 Effects of Pressure on Distillate Losses 5-11
5.1.2 Summary of Operating Parameters and Results During Testing
of High Vacuum Vapor Compression Evaporation System at the
Charleston Navy Yard 5-15
5.1.3 Summary of Operating Parameters and Results Using
Evaporation to Recover a Cadmium Cyanide Plating Bath
at Superior Plating, Inc 5-18
5.1.4 Economic Evaluation of Vacuum Compressor Evaporator Unit
Employed at Superior Plating, Inc 5-19
xiv
-------
TABLES (continued)
Number Page
5.1.5 Summary of Results of Pilot-Scale Unit Installed at a
Chinese Steel Plant 5-21
5.1.6 Typical Operating Parameters During Testing of the Pilot-
Scale Evaporation/Distillation System at the NYBY Steel
Works in Sweden 5-22
5.1.7 Typical Capital Equipment Costs for Various Evaporation
System Capacities 5-20
5.1.8 Percentage Breakdown by Plating Type of Evaporation Units
Currently in Operation 5-25
5.2.1 Evaporation-Crystallization System for Recovery of
Nitric-Hydrofluoric Acid . 5-35
5.2.2. Typical Operating Parameters and Results for Sulfuric Acid
Recovery System Using Crystallization 5-40
5.2.3 Typical Performance of a Two-Stage Crystallization
System for the Recovery of Nitric-Hydrofluoric Acid 5-41
5.2.4 Economic Evaluation of Acid Recovery System Using
Crystallization Technique 5-43
5.2.5 Economic Evaluation of Two-Stage System for Recovery
of Nitric-Hydrofluoric Acid 5-44
5.3.1 Selectivities of Ion Exchange Resins in Order of
Decreasing Preferences 5-56
5.3.2 Typical Operating Parameters and Results for the APU
Installed at Continuous Colour Coat, Ltd. in
Rexdale, Ontario 5-60
5.3.3 Economic Evaluation of the APU Installed at Continous
Colour Coat, Ltd 5-61
5.3.4 Typical Operating Parameters and Results During Testing
of the APU for Recovery of Nitric Acid at Modine
Manufacturing Company in Racine, Wisconsin 5-63
5.3.5 Economic Evaluation of the APU Installed at Modine
Manufacturing Company in Racine, Wisconsin for the
Recovery of Nitric Acid ,, 5-64
xv
-------
TABLES (continued)
Number Page
5.3.6 Typical Operating Parameters and Results for the APU
Installed at Springfield Machine & Stamping, Inc.
in Warren, Michigan for Sulfuric Acid Recovery 5-65
5.3.7 Average Constituent Concentrations in the New, Intermediate,
and Spent Bath Solutions 5-66
5.3.8 Summary of Results of Research Performed at
Electroplating Engineering, Inc. . 5-70
5.3.9 Recommended Minimum Concentrations (g/L) for Efficient
Metals Removal Using the Eco-Tech APU 5-73
5.3.10 Typical Capital Costs for Eco-Tech APU 5-75
5.3.11 Typical Operating Costs for Acid Purification Using
Continuous Countercurrent Ion Exchange (RFIE) 5-76
5.3.12 Economic Evaluation of the Acid Purification Process 5-77
5.3.13 Demonstrated Applications of Eco-Tech Acid Purification
Unit Using RFIE 5-79-
5.3.14 Comparison of Ion Exchange Operating Modes 5-81.
5.4.1 Laboratory-Scale Electrodialysis of Simulated Chromic
Acid Rinses Over a Range of Current Densities 5-91
5.4.2 Operating Parameters for Electrodialysis Demonstration Unit
for Recovery of Chromic Acid at Seaboard Metal
Finishing Company 5-93
5.4.3 Typical Operating Parameters and Results for the
Aquatech Electrodialysis Process 5-97
5.4.4 Typical Operating Parameters for Ion Transfer System
Developed by Innova Technologies, Inc 5-98
5.4.5 Summary of Results of Preliminary Testing of PRU 5-100
5.4.6 Summary of Operating Parameters and Results of Bureau of
Mines Tests Using Electrolytic ED to Treat Spent
Chromic-Sulfuric Acid Etjchants 5-101
5.4.7 Typical Operating Parameters for Scientific Control, Inc.
Electrodialytic ED Unit £or Recovery of Brass Etchants . . . 5-103
xvi
-------
TABLES (continued)
Number " Page
5.4.8 Economic Evaluation of Aquatech ED System to Regenerate
1.5 Million Gallons of HF/HN03 Pickling Liquor per Year . . . 5-105
5.4.9 Costs for Leasing Aquatech System 5-106
5.4.10 Estimated Fixed Capital Costs for Electrolytic Recovery Unit . 5-107
5.4.11 Estimated Annual Operatings Costs for Electolytic
Recovery Unit 5-109
5.4.12 Cost Comparison of Electrolytic Unit vs. Conventional
Treatment 5-110
5.5.1 Commercially Available RO Membranes Applicable to the
Treatment of Corrosive Wastes 5-116
5.5.2 Summary of the Characteristics of the Corrosive Wastes
Treated by Reverse Osmosis During the Preliminary Testing
Conducted by the Walden Division of Abcor, Inc 5-122
5.5.3 Analytical Test Methods Used During Preliminary
Investigations of Reverse Osmosis Applications Conducted
by Walden Division of Abcor, Inc 5-124
5.5.4 Analytical Results for Reverse Osmosis Treatment of
Spent Chromic Acid Plating Rinse 5-125
5.5.5 Analytical Results for Reverse Osmosis Treatment of
Spent Copper Cyanide Plating Rinse 5-126
5.5.6 Analytical Results for Reverse Osmosis Treatment of
Spent Cadmium Cyanide Plating Rinse . 5-126
5.5.7 Analytical Results for Reverse Osmosis Treatment of
Spent Zinc Cyanide Plating Rinse 5-126
5.5.8 Typical Operating Parameters During Testing of Reverse
Osmosis System at American Electroplating Company for
the Recovery of Zinc Cyanide Plating Rinse 5-129
5.5.9 Typical Composition of Zinc Cyanide Plating Bath at
American Electroplating Company 5-130
5.5.10 Capital Costs for Various RO System Conversions 5-132
5.5.11 Operating Costs for a Reverse Osijiosis System 5-133
-------
TABLES (continued)
Number Paze
5.6.1 Results of Preliminary Membrane Testing Performed by
the Southewest Research Institute 5-147
5.6.2 Results of Tests to Determine the Effect of Metal Ion
Concentration on Transport Rate 5-148
5.7.1 Operating and Design Parameters for the Multi-Stage
Pulsed Mode Column 5-170
5.7.2 Operating and Design Parameters for the Multi-Stage
Mixer-Settlers 5-171
5.7.3 Results of Tests Conducted at Stora in Sweden Using
the AX Process 5-172
5.7.4 Typical Composition of Spent Pickling Liquor at Kawasaki
Steel Chiba Works in Japan 5-173
5.7.5 Typical Operating Parameters and Results for Kawasaki
Process 5-174
5.8.1 Typical Operating Parameters for Thermal Decomposition
Acid Regeneration System 5-182
5.8.2 Typical Performance of a Thermal Decomposition System
for Regeneration of Hydrochloric Acid 5-184'
5.8.3 Typical Composition of Iron Oxide Byproduct Generated by
Thermal Decomposition Process 5-185
5.8.4 Composition of Wastes Fed to a Rotary Furnace During
Testing Conducted by the BMFT 5-187
5.8.5 Summary of Analytical Results of the Rotary Furnace
Pilot Tests Conducted by the BMFT 5-189
5.8.6 Economic Evaluation of Hydrochloric Acid Regeneration
Using Thermal Decomposition 5-190
5.9.1 List of Most Commonly Exchanged Corrosive Wastes 5-196
5.9.2 Industries Representing the Majority of Generators and
Potential Users Benefiting from Waste Exchange . ' 5-199
5.9.3 Examples of Economic Uses of Waste Exchanges for
Corrosive Wastes . .j 5-202
xvi 11
-------
TABLES (continued)
Number Page
5.9.4 Clearinghouse (Information) Waste Exchanges in
North America 5-203
5.9.5 Material Exchanges in North America 5-204
6.2.1 Summary of Recovery/Reuse Technologies for Corrosive Wastes . . 6-8
6.2.2 Guideline Considerations for the Investigation of Waste
Treatment, Recovery, and Disposal Technologies 6-10
'6.2.3 Summary of Neutralization Technologies 6-13
6.2.4 Major Cost Centers for Waste Management Alternatives 6-15
xix
-------
ACKNOWLEDGEMENT
The authors would like to thank Harry M. Freeman, the Hazardous Waste
Engineering Research Laboratory Work Assignment Manager, whose assistance and
support was utilized throughout the program. The authors also extend thanks
to other members of the HWERL staff for their assistance and to the many
industrial representatives who provided design, operating, and performance
data for the waste treatment technologies.
xx
-------
SECTION 1
INTRODUCTION
Section 3004 of the Resource Conservation and Recovery Act (RCRA), as
amended by the Hazardous and Solid Waste Amendments of 1984 (USWA), prohibits
the continued placement of RCRA-regulated hazardous wastes in or on the land,
including placement in landfills, land treatment areas, waste piles, and
surface impoundments (with certain exceptions for surface impoundments used
for the treatment of hazardous wastes). The amendments specify dates by which
these prohibitions are to take effect for specific hazardous wastes as shown
in Table 1.1. After the effective date of a prohibition, wastes may only be
land disposed if: (1) they comply with treatment standards promulgated by the
Agency that minimize short-term and long-term threats arising from land
disposal; or (2) the Agency .has approved a site-specific petition
demonstrating,-to a reasonable degree of certainty, that there will be no
migration from the disposal unit for as long as the waste remains hazardous.
Liquid acidic wastes (i.e., pH less than or equal to 2.0) will be banned
from land disposal effective July 8, 1987 with an exception granted for wastes
which are disposed via underground injection. Restrictions for these wastes
will be enacted by August 8, 1988. Alkaline waste (i.e., pH greater than or
equal to 12.5 or wastes which are strongly corrosive to steel) and non-liquid
waste disposal restrictions will be promulgated by May 8, 1990. However, the
1984 RCRA Amendments authorize the Agency to extend the effective dates of
prohibitions for up to 2 years nationwide if it is determined that there is
insufficient alternative treatment, recovery or disposal capacity.
1-1
-------
TABLE 1.1. SCHEDULING FOR PROMULGATION OF REGULATIONS BANNING
LAND DISPOSAL OF SPECIFIED HAZARDOUS WASTES
Waste category
Effective datea
Dioxin-containing waste
Solvent-containing hazardous wastes
numbered F001, F002, F003, F004, F005
California list:
- Liquid hazardous wastes, including free liquids
associated with any solid or sludge containing:
- Free or complex cyanides at ^1,000 mg/L
- As 2,500 mg/L ~~
- Cd >100 mg/L
- Cr+-* >500 mg/L
- Pb 250"0 mg/L
- Hg >20 mg/L
- Ni ^134 mg/L
- Se ~100 mg/L
- Ti ^130 mg/L
- Liquid hazardous wastes with:
•- pH <,2.0
- PCBs ^50 ppm
-Hazardous wastes containing halogenated organic
compounds in total concentration _>1,000 mg/kg
Other listed hazardous wastes (§§261.31 and 32), for
which a determination of land disposal prohibition
must be made:
- One-third of wastes
- Two-thirds of wastes
- All wastes
Hazardous wastes identified on the basis of
characteristics under Section 3001
Hazardous wastes identified or listed after enactment
11/8/86
11/8/86
7/8/87
7/8/87
7/8/87
8/8/88
6/8/89
5/8/90
5/8/90
Within 6 months
aNot including underground injection for which land disposal restrictions
will be promulgated by 8/8/88.
1-2
-------
PURPOSE AND SCOPE
This Technical Resource Document (TRD) for corrosive RCRA wastes
identifies recovery and treatment alternatives to land disposal for these
wastes and provides performance data and other technical information needed to
assess potentially applicable alternatives. This document is one of a series
of documents designed to assist regulatory agency and industrial personnel in
meeting the land disposal restrictions promulgated by the 1984 RCRA
Amendments. To minimize redundancy, emphasis has been placed on treatment
technologies (i.e., neutralization) which specifically address the corrosive
nature of RCRA wastes. Similarly, discussion of recovery practices has been
restricted to methods which are capable of achieving adequate performance at
extreme conditions of pH. Although emphasis is placed on performance data for
these processes, cost data and technical factors affecting performance (e.g.,
restrictive waste characteristics) are discussed to assist in the evaluation
of alternative approaches to land disposal.
DOCUMENT ORGANIZATION AND CONTENT
The following section (Section 2) -will identify the hazardous wastes of
concern which meet the RCRA definitions of corrosive wastes. Available
information concerning waste stream characteristics, generation, and
management practices will be provided in Section 3. Following sections
(Sections 4 and 5) will discuss neutralization and recovery practices, which
are available as alternatives to land disposal. Each process will be reviewed
with regard to the following four factors:
1. Process description, including design and operating parameters,
applicable was'te types, pretreatment requirements, and
post-treatment and disposal of residuals;
2. Case study and performance data which identifies the range in
potential applications, processing equipment, and system
configurations;
3. Cost of treatment; and
4. Present status of the process.
1-3
-------
Virtually all corrosive wastes will have to undergo some form of
neutralization as part of the treatment/disposal process. Thus, handling of
non-corrosive waste constituents will be discussed as pre- or post-treatment
(neutralization), as appropriate. Treatment and disposal alternatives for
these non-corrosive constituents and treatment residuals will be identified.
However, the reader is referred to related Technical Resource Departments for
detailed performance data since this was beyond the scope of this document.
A final section (Section 6) provides approaches to identifying and
selecting appropriate technologies for corrosive waste streams. Although
emphasis is placed on technical approaches, economic and institutional
concerns are also discussed to assist in process selection.
1-4
-------
SECTION 2
IDENTIFICATION OF RCRA CORROSIVE WASTES
As specified in the EPA regulations for identifying hazardous waste, a
waste is defined as a RCRA corrosive waste if it meets eitner of the following
criteria:
• It is aqueous and has a pH less than or equal to 2.0 or greater than
or equal to 12.5; or
• It is a liquid and corrodes steel (SAE 1020) at a rate greater than
0.25 in./yr at a test temperature of 130°F.
Thus, wastes which are all solids are not subject to restrictions
applicable to RCRA corrosive wastes. However, wastes which contain both
liquids and solids may be classified as RCRA corrosive wastes depending upon
the characteristics of the liquid fraction. EPA has established a test
2
protocol called the paint filter test which is designed to separate free
liquids by gravity from the waste matrix. If the recovered liquid meets the
above criteria for corrosive waste, the entire waste is considered to be
corrosive. The term "solids" is used throughout this document. It is assumed
that corrosives which were characterized as "solids" in the literature,
actually contain residual quantities of liquid which meet the definition of
corrosive waste.
Corrosiveness was determined to be a hazardous characteristic because
improperly managed highly acidic or alkaline wastes can present a danger to
3
human health and the environment through the following mechanisms:
• Harm to transporters and other persons coming into contact with the
waste;
• Solubilization of toxic constituents of solid wastes thereby
enhancing their transport into ground and surface water;
2-1
-------
• Chemical reactions with co-disposed wastes which can result in
generation of heat or toxic fumes; and
•. Altered pH of surface waters to the detriment of aquatic organisms.
Although these effects occur below pH of 3 and above pH of 12, EPA
ultimately promulgated less restrictive standards (i.e., pH less than or equal
to 2.0 or greater than or equal to 12.5) in order to exclude certain materials
from regulation. Specifically, these include otherwise non-hazardous wastes
such as lime stabilized sludges (pH 12.0 to 12.5), which can be put to
agricultural or other beneficial uses, and substances such as cola drinks and
many industrial wastewaters (pH 2.0 to 3.0).3 EPA felt that these less
restrictive limits would still encompass those wastes which are most likely to
involve damage to the skin, solubilization of toxic substances or harmful
chemical reactions.
In addition to pH, EPA has elected to express corrosiveness in terms of
the metal corrosion rate since many RCRA wastes are stored, transported and
land disposed in steel containers. Therefore, the metal corrosion rate is
indicative of a compound's ability to escape from its container or corrode
other containers thereby increasing the potential for advers.e chemical
reactions or hazardous substance release. The EPA standard for metal
corrosion rate was adopted from the Department of Transportation
classification for compounds which exhibit severe rates of corrosion at
temperatures which may be encountered during handling of hazardous materials.
EPA has listed two wastes from specific sources (K062, Kill) and five
discarded commercial chemical products and associated off-specification
materials, containers, and spills (U006, U020, U023, U123, U134) as materials
which meet the criteria for corrosive hazardous wastes. As shown in
i
Table 2.1, certain other listed hazardous wastes h4ve also been reported which
were characterized by the generators as corrosive. : Finally, the waste code
i
D002 has been assigned to wastes which also meet the definition of corrosive
wastes but are not identified as hazardous in Part 261, Subpart D. These
i ;
non-toxic corrosive compounds and other compounds which can react with water
>\ 4
to form corrosive wastes are also listed in Table 2.1.
2-2
-------
TABLE 2.1. CORROSIVE AND POTENTIALLY CORROSIVE WASTE MATERIALS
A. RCRA Wastes Listed Due to Corrosivity
D002 - Any waste not listed in Subpart D which has either: 1) 12.5
-------
TABLE 2.1 (Continued)
BASES -
K060 - Ammonia still lime sludge from coking operations
U012 - Aniline
U092 - Dimethylamine
U098 - 1,1-dimethylhydrazine
P046 - Ethanamine, 1,1-dimethyl-
2-phenyl-
P053 - Ethylene diamine
U133 - Hydrazine
U167 - 1-Napthylamine
U174 - N-Nitrosodiethylamine
U194 - n-Propylamine
U196 - Pyridine
D221 - Toluenediamine
C. Potentially Corrosive Son-Listed Wastes
ACIDS -
Acetic acid
Acetimidic acid
Adipic acid
Butanoic acid
Carbamic acid
Chlorosulfonic acid
Chromic acid
Citric acid
Dithiocarbamic acid
Fulminic acid
Hydrochloric acid
BASES -
Aminoethanolamine
Ammonium hydroxide
Caustic potash Solution
Caustic soda Solution
Cyclohexylamine
Diethanolamine
Diethylenetriamine
Diisopropanolamine
Dimethylformamide
Hexamethylenediamine
Hexamethylenetetramine
Methylethylpyridine
Monoethanolamine
Monoisopropanolamine
Hydrogen chloride
Hydrogen peroxide
Isocyanic acid
Methanesulfonic acid
Nitric acid
Oxalic acid
Phosphoric acid
Phosphorofluoric acid
Propionic acid
Sulfurie acid •
Vanadic acid
Morphaline
Potassium hydroxide
Sodium carbonate
Sodium hydrosulfide
Sodium hydrosulfite
Sodium Hydroxide
Triethanolamine
Triethylamine
Triethylenetetramine
Trimethylamine
Urea
(continued)
2-4
-------
TABLE 2.1 (Continued)
D. Chemicals that React in Water to Give Acids
Compound
Acetic anhydride
Aluminum chloride '
Benzoyl chloride
Bromine
Chlorosulfonic acid
Maleic anhydride
Nitrogen tetroxide
Nitrosyl chloride
Oleum
Phosphorus oxychloride
Phosphorus pentasulfide
Phosphorus trichloride
Polyphosphoric acid
Sulfur monochloride
Sulfuryl chloride
Titanium tetrachloride
Resulting Acid
Acetic acid
Hydrogen chloride + aluminum hydroxide
Hydrogen chloride + benzole acid
Hypobromous acid
Hydrogen chloride + sulfuric acid
Maleic acid
Nitric acid + nitric oxide
Hydrogen chloride + nitrous acid
Sulfuric acid + sulfur trioxide
Hydrogen chloride + phosphoric acid
Hydrogen sulfide + phosphoric acid
Hydrogen chloride + phosphoric acid
phosphoric acid
Hydrogen chloride + sulfuric acid + others
Hydrogen chloride + sulfuric acid
Hydrogen chloride + tri-hydroxyhalides
E. Chemicals that React with Water to Give Bases
Compound
Anhydrous -ammonia
Ethylene imine
Lithium aluminum hydride
Sodium
Sodium amide
Sodium hydride
Resulting Base
Ammonium hydroxide
Monoethanolamine
Lithium hydroxide + hydrogen + aluminum
hydroxide
Sodium hydroxide + hydrogen
Ammonia + sodium hydroxide
Sodium hydroxide + hydrogen
Sources: References 1, 4, 5 and 6.
2-5
-------
Ultimately, the pH or metal corrosivity of a waste should be determined
experimentally. Test methods for these properties are uncomplicated,
inexpensive to perform and are accurate provided that representative waste
samples have been collected. Estimates of pH and corrosivity are provided in
789 10
the literature for pure solutions at various concentrations. ''
However, many wastes will be contaminated with interfering compounds which
will limit the usefulness of these approximations.
The presence of chloride and sulfate ions accelerates corrosion by
interfering with the development of protective films and contributing to the
breakdown of passive films already in existence. Dissolved oxygen and
elevated temperature also act to stimulate the corrosion process.
Alternatively, calcium and bicarbonate ions tend to limit attack. Solutions
with high pH cause limited corrosion but, in practice do not usually corrode
steel at a rate which would classify them as hazardous under the RCRA
definition. Thus, acidic solutions which are very low in pH, contain chloride
or sulfate ions and/or contain dissolved oxygen should be tested for
corrosivity. EPA suggests using a simple immersion test (NACE Standard
TM-01-69) to make this determination.
In most industrial applications of acids or alkalis, the presence of
contaminants will generally act to create a more neutral solution. The
magnitude of this effect will be dependent on the specific type and
concentration of the contaminants. If representative waste samples are not
available for testing, the pH for a pure component solution can be estimated
provided that the solute concentration and dissociation constant are known.
The latter can be obtained from standard engineering texts for most acids and
789 10
bases. ' ' ' Examples of pH values' for compounds with varying
dissociation constants and concentrations are provided in Table 2.2.
Calculational methods for pH are summarized below.
For monoprotic acids:
pH - -log X - -log f "I * (Ka2 * 4KaM)
2-6
-------
TABLE 2.2. DISSOCIATION CONSTANTS AND pHs OF ACIDS
AND BASES IN AQUEOUS SOLUTIONS AT 25°C
Compound
Formula
(1st step) Concentration
PH
Acids
Hydrochloric acid HC1
Sulfuric acid 1^504 •
Sulfurous acid H2S03
Oxalic acid 02^04
Orthophosphoric acid H3?04
Bases
Sodium hydroxide NaOH
Potassium hydroxide KOH
Lime CaOH
103 1 N
0.1 N
0.01 N
103 1 N
0.1 N
0.01 N
1.3 x 10~2 0.1 N
5.9 x 10"2 0.1 N
7.5 x 10~3 0.1 N
103 IN
0.1 N
0.01 N- .
103 1 N
0. 1 N
0.01 N
3.74 x 10~2 (Saturated)
0.1
1.1
2.0
0.3
1.2
2.1
1.5
1.5
1.5
14.0
• 13.0
12.0
14.0
13.0
12.0
12.4
Source: Reference 5.
2-7
-------
where X equals the hydronium ion concentration, K is the acid dissociation
constant and M is the molar concentration of the undissociated acid (i.e., the
solution formality). These equations can be applied to bases by substituting
pOH and K^ where appropriate in the above equation and using the following
relationship:
pH = 14 - pOH
For strong acids (bases) which are almost completely ionized, pH becomes the
negative logarithm of the product of the formal acid (base) concentration and
the number of moles of hydronium (hydroxide) ions formed per formula weight of
.acid (base).
2-8
-------
REFERENCES
1. Federal Register. 40 CFR Part 261. Environmental Protection Agency.
Regulations for Identifying Hazardous Wastes.
2. Federal Register. February 25, 1982. Volume 47, p. 8311.
3. USEPA. Identification and Listing of Hazardous Waste Under RCRA,
Subtitle C, Section 3001 - Corrosivity Characteristics (40 CFK 261.22).
U.S. Environmental Protection Agency. Washington, D.C. PBS1-184319.
May 1980.
4. Camp, Dresser & McKee Inc., Boston, MA. Technical Assessment of
Treatment Alternatives for Wastes Containing Corrosives. Prepared for
USEPA under Contract No. 68-01-6403. September 1984.
5. Sax, I. N. Dangerous Properties of Industrial Materials. Van Nostrand
Reinhold Company. New York, N.Y. 6th Edition. 1984.
6. Versar, Inc. Springfield, VA. National Profiles for Recycling - A
Preliminary Assessment. Draft Report prepared for USEPA under Contract
No 68-01-7053. July 8, 1985.
7. Weast, R. C. CRC Handbook of Chemistry and Physics. CRC Press, Inc.
Boca Raton, Florida. 1984.
8. Dean, J. A. Lange's Handbook of Chemistry. McGraw-Hill Book Company,
New York, N.Y. /12th Edition. 1979.
9. Lyman, W. J. and D. H. Rosenblatt. Handbook of Chemical Property
Estimation Methods. McGraw-Hill Book Co. New York, N.Y. 1982.
10. Green, D. W. Perry's Chemical Engineers' Handbook. McGraw Hill Book
Company, New York, N.Y. 6th Edition. 1984.
2-9
-------
SECTION 3
CORROSIVE WASTE SOURCES,
GENERATION AND MANAGEMENT
3.1 CORROSIVE WASTE SOURCES
The primary industrial applications for acids and bases which result in
generation of corrosive wastes are: (1) use as chemical intermediates in the
inorganic and organic chemical manufacturing industries; (2) use as a metal
cleaning agent in metal production and fabrication industries; and (3) use in
boiler blowdown and stack gas treatment, primarily in electricity generating
facilities. Other significant corrosive waste sources include refining
processes in the petroleum industry and pulping liquor in the paper industry.
Table 3.1.1 summarizes industrial consumption of primary acid and base
chemicals by end use. As shown, production of fertilizers consumes nearly
half (48.3 percent) of all domestic production of acids and bases, including
sulfuric acid, phosphoric acid, urea and nitric acid. The second largest
industrial use category is the production of primary chemicals (16.& percent)
followed by various uses as a chemical intermediate in the production of
glass, explosives, soaps and detergents, synthetic fibers, resins, coatings,
plastics and adhesives (7.0 percent).
Other industries which consume large quantities of corrosive materials
include primary metal, metal fabricating and equipment manufacturing
industries. Corrosive chemicals are primarily used as fluxing agents in the
production of steel and to clean or prepare metal surfaces prior to annealing
or coating operations. The latter uses are nonconsumptive and often involve
highly concentrated solutions. Therefore, a high quantity of waste is
generated from these industries relative to the amount of raw material
consumed.
3-1
-------
TABLE 3.1.1. INDUSTRIAL CONSUMPTION OF PRIMARY ACID AND ALKALI CHEMICALS BY END USE (103) TONS/YEAR)
M
Chemical ipccies
Sul (uric acid
Phosphoric acid
Sodium hydroxide
Sodium carbonate
Urea
Nitric acid
Muriatic acid
Acetic acid
Adiple acid
Acrylic acid
Methyl methacryUte
Hydrofluoric acid
Potassium hydroxide
Citric acid
Sodium hydroaulfite
Chloro.ulfonic acid
Chromic acid
Propionic acid
Ethylene diamine
Formic acid
Sodium hydrosulfide
TOTAL
PERCENT
ProHuc- Fertil-
tion itert
33,354 21.255
9,215 8,4)5
9,141
8, 468
7,610 4,870
7,340 4,698
2,580
1,469
570
375
362
330
220 20
101
60
45
36
33
29
24
15
81,377 39,298
100 48.3
Primary
chemical
manufac-
turing
2,616
5,484
1,524
76
1,174
650
1,264
68
330
-
116
99
10
-
42
0
19
17
12
2
13,503
16.6
Hetat
production, Soap.
Chemical _ Pulp Mater and food cli-anlne., nnd
inter- Petroleum and .tick «as mfg. or preparation, Explo- deter-
mediate* industry paper treatment product. 4 finishing Mining lives tile* Textile)
981 2,616 654 1,308
- - J70
366 1,828
3,387d - 254 593
533 ... 761
-
650 - - 370
....
492 ....
30C 10C
307 18 -' -
20
>0 - -
>0 - - 71
- 38 - >0
.
-
6 _ . . >0
>0 -
3 >0
5,739 3,670 2,774 1,911 1,772
7.0 4.5 3.4 2.4 2.2
400C 1,308 - 327
95 ....
366 - - 274
593
- - - -
0 - 1,101
650 -
29
5=
1
135 17 - -
55
>0 - - 13
>0 >0
-
18 - - - >0
'.
>0
>0 - - - 3
2 ....
1,666 1,330 1,101 98B 307
2.0 1.6 1.4 1.2 0.4
Otherb
1,889
95
823
2,117
1,370
O67
260
176
10
>0
36
42
<46
< 7
<22
3
<18C
< 8
<12
< 6
11
7,318
9.0
'Include, production of Rla.a, flberi, resins, coating., plastic, and adhesive..
^Include, net export..
ce»timated.
Source: Adapted from References 1 and 2.
"U.e in gla.a production.
'16,000 tons uaed a. a wood preservative.
-------
Other high volume industrial uses of strong acids and bases include use
as an alkylation catalyst in petroleum refining, chemical bleaching in the
paper industry, textile treating and finishing, food products manufacturing,
the recovery of nonferrous minerals, (e.g., copper and uranium ores in mining
applications), and oil and gas well acidizing for enhanced recovery. Some of
these uses are nonconsumptive and result in the generation of high
concentration, high volume corrosive waste; e.g., textile treatment baths, and
spent acid alkylation catalysts. Conversely, food industry and oil and gas
drilling generate little waste relative to quantities of acid/alkali
consumed. These uses are either consumptive or result in waste which is
exempt from RCRA regulation; e.g., acid mine drainage.
The following is a brief summary, by industry, of acid/alkali
consumption, corrosive waste streams generated, and their general waste
characteristics.
3.1.1 Chemicals and Allied Products
The Chemicals and Allied Products Industries (SIC 28) represent the bulk
of acid/alkali production (90 to 95 percent) , consumption (80 percent or • •
12 3
more) ' , and waste generation (72 percent). These industries include
the inorganic chemical, fertilizer, and organic chemical manufacturing
industries as discussed below.
Inorganic Chemicals Industry—
The inorganic chemicals industry is responsible for most of the
production of sulfuric acid, phosphoric acid, sodium hydroxide, nitric acid,
some hydrochloric acid, hydrofluoric acid and other acids/alkalis. Many of
these processes generate wastes in the form of off-specification products,
wastewaters, and sludges (e.g., filtration residues) which meet EPA criteria
for corrosive wastes. However, due to a high degree of waste recycle, these
industries only generate a modest fraction of overall chemical industry
corrosive waste.
Hydrofluoric acid is generally produced by the action of sulfuric acid on
fluorspar which generates a calcium sulfate precipitate. Other substances
such as silicon tetrafluoride, fluosilicic acid, hydrogen sulfide, carbon
dioxide and sulfur dioxide can be generated due to impurities in the
fluorspar. The kiln discharge stream is typically below pH of 2.0 and is
3-3
-------
treated by neutralization with soda ash and clarified. Combined
wastewaters have a total suspended solids (TSS) range from 220 to 16,400 mg/L
and are typically between pH of 2.0 to 3.0. However, values in the corrosive
acid range have been reported. Heavy metal concentrations (Cr, Ni, Zn)
tend to be below levels which require treatment as a hazardous waste. Thus,
treatment of these wastes via lime neutralization and dual media filtration
provides a stream suitable for discharge.
Phosphoric acid is primarily produced by sulfuric acid acidulation of
phosphate rock generating a CaSO, precipitate which is removed by
filtration. Gypsum and contaminants such as fluosilicates, and aluminum and
iron phosphates are precipitated from the acid stream as the solution is
concentrated through evaporation. The precipitates are generally recycled and
used in dry fertilizer manufacturing. Wastewater sources include cooling
tower blowdown, containing ammonia and chromates, and boiler blowdown.
Nitric acid is primarily produced by ammonia oxidation in a converter
containing a platinum-rhodium catalyst which generates water as a by-product.
This is followed by further oxidation of nitric oxide to nitrogen dioxide
which is absorbed in a portion of the by-product -water to form nitric acid.
Major contaminants in the waste by-product water include unreacted ammoni'a,
nitric acid, and nitrogen oxides (e.g., NO, NO., NO,).
Sodium hydroxide is produced as a co-product with chlorine during the
electrolysis of sodium chloride in diaphragm or mercury cells. The reaction
product is evaporated to concentrate the caustic soda solution producing
sodium chloride brine muds. However, due to the buffering action of the salt,
these wastes typically have pH levels below 12.5 and thus may not qualify as
RCRA corrosive wastes.
Most sulfuric acid is produced by the contact process in which sulfur is
burned to sulfur dioxide which is then catalytically (vanadium pentoxide
impregnated on diatoraaceous earth) oxidized to sulfur trioxide. This is
absorbed in concentrated sulfuric acid and diluted to the desired
concentration. The principal waste stream generated is acid plant blowdown
(i.e., off-gas scrubber liquor) which has a typical pH of 1.8 to 2.5 and
g
contains high levels of Cu, Pb, Zn, and Fe. When by-product sulfur dioxide
from smelter gases is used as the source of sulfur, impurities such as arsenic
and fluorine are also commonly present.
3-4
-------
Other documented sources of corrosive wastes from production of primary
inorganic chemicals include combined raw wastewaters from the production of
chloride (pH ranges from 1.5 to 9.0, TSS averages 32 mg/L, Al averages
44 mg/L) and production of sodium fluoride (pH in excess of 12, IS averages
167,500 mg/L, F averages 16,000 mg/L).
Fertilizer Industry—
Potentially corrosive fertilizer industry wastes originate from
production of their intermediates (primarily sulfuric acid, phosphoric acid,
and urea), production of fertilizer products and boiler blowdown. Ammonia
effluents will not reach a high enough pH to be RCRA corrosives. However,
urea process condensates may exceed pH 12.5. Condensate from flashed gases
from the urea concentration step result in a solution containing urea,
ammonium carbonate, ammonia and carbon dioxide. The waste quantity has been
reported to range from 417 to 935 1/kkg of product with ammonia concentration
of 9,000 kg/1,000 kkg and urea concentration of 33,500 kg/1,000 kkg of
product.
In the phosphate fertilizer industry, significant discharges result from
water treatment and blowdown streams associated with the sulfuric acid
process, and gypsum pond water. Gypsum pond water is typically recycled in a
closed loop, requiring treatment and discharge only when pond holding capacity
is exceeded. The gypsum slurry, resulting from the filtration step in
phosphoric acid production, has a pH of less than 2.0 with soluble phosphates
(e.g., phosphoric acid) and fluoride (e.g., sodium silicofluoride) in the 8 to
15 g/L concentration range. Gypsum pond waste characteristics are
9
summarized in Table 3.1.2 for a phosphate fertilizer plant.
Organic Chemicals Industry—
The organic chemicals industry generates corrosive wastes in the
production of petrochemicals, polymers, organometallics, detergents,
pesticides, explosives, dyes, and pigments. Corrosive raw materials used in
high volumes include sodium hydroxide, sulfuric acid, sodium carbonate,
hydrochloric acid and acetic acid. In addition, hydrochloric acid is produced
in large quantities (90 percent of domestic production) as a by-product from
manufacture of vinyl chloride monomers, isocyanates, chlorofluorocarbons and
other chemical products.
3-5
-------
The production of petrochemicals generates effluents which are usually
the result of by-product water and water used for chemical transportation.
These are typically heavily contaminated with organic substances. For
example, acetic acid manufacturing wastewater contains high levels of any or
all of the following: formic acids (175 to 89,400 ppm), acetic acid (30 to
177,100 ppm), propionic acid (0 to 1,200 ppm), alcohols (260 to 30,000 ppm),
ketones (0 to 2,000 ppm) and aldehydes (0 to 92,000 ppm).10
The EPA has recently added product washwater from the production of
dinitrotoluene via nitration of toluene (Kill), 'to the list of hazardous
wastes from specific sources (40 CFR Part 261.32). These washwaters exhibit a
pH less than 2.0 and contain toxic concentrations of 2,4*-dinitrotoluene
(800 ppm) and 2,6-dinitrotoluene (200 ppm).
Organic chemical wastes from the polymer industry (including plastics,
resins, rubbers, and fibers) can be acidic or alkaline, and contain a variety
of contaminants. Effluents from reclaimed rubber manufacture have a very
high pH as well as large amounts of suspended solids (usually escaped bits of
rubber) and chlorides. Synthetic rubber production also yields a high
concentration of solids, but they are suspended in an acidic, rather than
alkaline, saline solution. . Cellulosics can be produced by the xanthate
process which generates a waste stream containing H-SO,, NaOH, heavy
3
metals, cellulosic materials, and sulfates. Polyester and alky resins are
formed by condensation polymerization of a dibasic acid and a polyfunctional
alcohol. The main wastes yielded by this process are caustic wash solutions
3
contaminated with unreacted volatile fractions of raw materials.
The pesticide industry routinely recycles a large fraction of its spent
12
acid wastes. However, potential sources for corrosives include:
• Waste products from acid recovery units;
• Nitric acid and other acidic wastewater from fractionation columns
in the production of halogenated aliphatic pesticides;
• High pH caustic scrubber water which typically contains dissolved
solids and low organic levels;
• Spent acid from settling tanks containing moderate organic content;
and
• Wet scrubbers for acidic solutions containing low organic
concentrations.
3-6
-------
TABLE 3.1.2. PHOSPHATE FERTILIZER INDUSTRY: GYPSUM
POND WASTE CHARACTERISTICS
Parameter3
pH, standard units
Total acidity, as CaCO^
Fluoride, as F
Phosphorus, as P
Silicon, as Si
Total solids
Total suspended solids
Chlorides, as Cl
Sulfates, as SO,
Sodium, as Na
Calcium, as Ca
Magnesium, as Mg
Aluminum, as Al
Chromium, as Cr
Zinc, as Zn
Iron, as Fe
Manganese, as Mn
NH. - N, as N
Total organic N, as N
Color, APHA units
Range
1.6 - 2.1
20,000 - 60.OOO
4,000 - 12,000
4,000 - 9,000
1,000 - 3,000
20,000 - 50,000
50 - 250
50 - 500
2,000 - 12,000
50 - 3,000
50 - 1,500
50 - 400
50 - 1,000
0.2 - 5.0
1.0 - 5.0 -
100 - 250
5 - 30
0 - 1,200
3 - 30
20 - 4,000
aAll values expressed as mg/L unless otherwise noted.
Source: Reference 9.
3-7
-------
An example of an acidic process waste from the production of organo-phosphorus
pesticides was provided by EPA as follows: pft ranging from 1.0 to 2.5, TSS of
36 mg/L, phenol concentration of 0.005 mg/L, total chlorine 5,730 mg/L, total
1 9
phosphorus 35.1 mg/L and total pesticides of 0.014 mg/L.
Phosphate compounds are a principal ingredient in detergents as are
alkalies (e.g., sodium carbonate), silicates, neutral soluble salts, acids
(e.g., sulfuric acid), insoluble inorganic builders (e.g., clays such as
kaolin) and miscellaneous organic compounds; e.g., ethylenediamine,
3
tetra-acetic acid and sodium carboxymethyIcellulose. Effluents from
detergent manufacture will contain a fraction of these and other contaminants
in the acid and alkali wastes.
The explosives industry generates wastes from three major processes: the
manufacture of TNT, the manufacture of smokeless powders, and the manufacture
of small-arms ammunition. TNT production results in nitric and sulfuric
acid wastes, contaminated with ammonia and nitrogen oxides, and alkaline wash
water containing sulfur impurities. Smokeless-powder wastes are similarly
concentrated in nitric and sulfuric acids, and generally contain a large
amount of waste sulfates and nitrogen oxides. Wastes from the manufacture of
small-arms munitions include, waste pickling acids contaminated with copper,,
zinc and grease from cutting oils and detergents, and wastewater from -
equipment washout. An example of a caustic waste (pH j> 12.5) is a
desensitized washout solution of RDXUCH.N.O.)-) and lead styphnate
(PbC,H02(NO.).) which has been treated with caustic. This typically
contains high lead levels (225 mg/L), high TDS (17,000 mg/L), and low TSS
(35 mg/L).13
The most comprehensive source of information regarding hazardous waste
characteristics and handling methods in the organic chemicals industry is the
14
Industries Studies Data Base (ISDB). This data was compiled by EPA
through actual chemical analysis of residual streams or from organic chemical
manufacturer responses to RCRA 3007 questionnaires. Wastes meeting the
definition of RCRA corrosives were classified by residual type,
physical/chemical properties, waste quantity and constituents, constituent
concentrations, and management practice. Nearly 37 million metric tons of
waste was characterized which is representative of approximately 72 percent of
all corrosive wastes generated.
3-8
-------
14
A summary of data analyzed in August 1985 is presented below. This
discussion is generally qualitative since little data were available to
establish aggregate waste volumes which exceeded various constituent
concentration levels.
Chemical industry corrosives tend to be very low in metal and oil
concentrations. The small percentage of waste streams which may require
specialized treatment for metals (less than 3 percent) tended to be small
volume streams. However, wastes are frequently contaminated with organics at
concentrations which require treatment, in addition to neutralization, prior
to discharge. The most frequently reported noncorrosive organics which may
require treatment are methanol, toluene, and chlorinated solvents.
The most significant corrosive constituent on a waste stream volume basis
is sodium hydroxide followed by sodium chloride, glutaric/adipic acid (low
concentration, high volume aqueous waste), sulfuric acid, and sodium
carbonate. The most highly concentrated of these are sulfuric and
hydrochloric acids. On average, sulfuric acid and sodium hydroxide streams
are generated in large volumes whereas sodium carbonate and the more expensive
hydrochloric acid wastes are generated in smaller volumes.
The vast majority of organic chemical industry corrosive wastes are
aqueous (93.3 percent) with roughly an even split between acidic and alkaline
wastes. Sludges constituted 4.2 percent of the total waste volume, but
nearly all of this was fluid enough to be disposed via deep well injection.
Organic liquids accounted for only 2.2 percent, collected gas streams
(predominately HC1) contributed 0.1 percent, and solids were only 0.01 percent
of the total waste generated. Solid corrosives include materials such as
spent adsorbents, filters, dried sludges, waste containers, and catalysts. An
example of these is spent phosphoric acid catalyst used in the production of
cumene from the alkylation of benzene with propylene.
3.1.2 Petroleum Refining Industry
The Petroleum Refining Industry (SIC 29) is a large consumer of sulfuric
acid (most of which is produced onsite), hydrochloric acid and, to a lesser
extent, sodium hydroxide and hydrofluoric acid. Concentrated sulfuric acid
(98 percent solution) or hydrofluoric acid are used to catalytically react
3-9
-------
light hydrocarbon molecules in alkylation units to yield higher octane
products. Spent acid sludge, which is contaminated with water (about
3 percent) and hydrocarbons (about 10 percent), is typically recovered on or
offsite through thermal and other treatment technologies. Sludges which
result from neutralization of alkylation waste will also contain heavy metals
such as nickel, copper, lead, selenium and arsenic. Other potentially
corrosive wastes from petroleum refining include acid/alkali washes, acid
sludges from sulfonation, sulfation and acid treatments, spent acid catalysts
(e.g., phosphoric acid catalyst used in polymerization processes), boiler
blowdowns, and caustic solutions from chemical sweetening and
desulfurization.
Re-refining of used oil by the acid clay process generates acid sludge
and dehydration wastewater which have pH less than 2.0. Sulfuric acid
treatment of used oil dissolves metal salts, aromatic and asphaltic compounds,
organic acids, water and other polar compounds. In addition to these
compounds, the resulting sludge will contain particularly high levels of lead
(2,000 to 10,000 ppm) and lesser, but significant levels of aluminum, barium,
magnesium, and calcium. As much as 10 percent of used oil can remain in
9
the spent acid sludge which contributes to its reported heating value of
9,000 to 11,000 Btu/lb. A sample of acid clay dehydration process water
showed a pH of 1.6 with high iron (197 mg/L), phenol (99 mg/L), boron (54
mg/L), silicon (52 mg/L) and lead (40 mg/L) content.
Caustic sludge is generated in small quantities from emulsion breaking of
used oils which results in a waste containing caustic, sodium silicate, oil,
lead and other metals. However, as a result of waste disposal problems,
the use of both acid and caustic clay processes have declined in favor of
various forms of vacuum distillation. Thus, these processes are no longer as
significant a source of corrosive wastes as they were prior to the
implementation of RCRA regulations.
3.1.3 Primary and Secondary Metals Industry
The Primary and Secondary Metals Industries (SIC 33) generate large
volumes of corrosive wastes from spent metal cleaning, preparation and
3-10
-------
pickling solutions. Other sources include certain rinses following acid
baths, fume scrubbers and small quantity residuals generated from treatment of
bath solutions.
The iron and steel industry accounts for approximately one-fourth of all
domestic hydrochloric acid demand for use in acid pickling baths. This
acid is used most frequently by large primary steel producers while sulfuric
acid is used in roughly one-half the amount by smaller, secondary steel
finishers. Smaller quantities of nitric and hydrofluoric acid are used by
18
specialty steel finishers.
The choice of acid is dependent on the base metal and desired surface
characteristics. Sulfuric acid penetrates oxide scale and reacts with the
base metal to form hydrogen which aids in removing the scale. Scale
eventually dissolves in the acid solution forming ferrous sulfate.
Hydrochloric acid reacts vigorously and directly with the oxide scale forming
soluble ferrous and ferric chloride. Combination acid pickling (hydrochloric,
sulfuric, nitric, and hydrofluoric acids) is reserved for specialty steels.
Wastewaters from these operations contain sulfates, fluorides and nitrates in
addition to iron salts. They also contain higher levels of toxic metals
(e.g., Cr, Ni) since specialty steels contain a wider range and higher levels
19
of alloying elements than do carbon steels. Pickling baths are highly
concentrated with acid (e.g., 5 to 15 percent H9SO, by weight) and, as a
20
result, are frequently recycled.
Other sources of acidic corrosives from the iron and steel industry
include pickling bath rinse waters, pickling bath scrubber systems and
scrubber effluent from cleaning absorber vent gases emitted from
21
(hydrochloric) acid regeneration systems. Rinse waters can be below pH of
2.0 if counterflow or high pressure spray systems are used to minimize water
consumption. Out of 43 rinse discharges sampled by the EPA, 22 had recorded
pH values which were less than or equal to 2.0. Combination acid rinses had
21
the lowest tendency to qualify as corrosive wastes (27 percent).
Wet fume scrubbers generate wastes with similar constituents as the
pickling baths but much lower concentrations of solids, oil and
22 23
grease. ' Much of this waste can be recycled or eliminated altogether
by replacing wet systems with dry systems such as acid demisters. Similarly,
3-11
-------
wet scrubber wastes from spent pickling bath regeneration fumes are commonly
recycled along with regenerated acid. A summary of constituent types and
concentrations for wastes associated with acid pickling is provided in
Table 3.1.3.
Alkali cleaning is practiced at over 60 carbon and specialty steel
facilities to remove dirt, oil, and grease prior to other finishing steps.
Typical-baths are composed of caustic and have pH of 12 to 13, oil and grease
concentration of 1,500 mg/L, iron concentration of 100 mg/L and high solids
23
content (TDS of 25,000 mg/L, TSS of 1,000 mg/L). Since alkali solutions
are' not as aggressive as acid pickling baths, the spent solution will contain
smaller quantities of most toxic metal pollutants. These baths are very
amenable to recycling through processes such as ultrafiltration since the
major contaminants are easily removed solids and oils. Low molecular weight
alkali and builders will readily pass through membranes. Rinse waters
generally do not have high enough pH to be considered RCRA corrosive
23
wastes. They are typically treated through oil skimming or flotation,
blended with acidic wastes and then treated for solids/metals removal.
The nonferrous metals industries, including primary and secondary copper,
lead and zinc, also generate significant quantities of corrosive wastes.
Primary copper smelters generate sulfuric acid from treatment of sulfur
dioxide off-gas in contact sulfuric acid plants. Dilute sulfuric acid waste
is generated from blowdown of conditioning towers which are designed to remove
dust from the metallurgical operations off-gas. Sampling data at three
facilities showed pH ranges of 1.8 to 2.5 with the following concentration
averages: TDS - 244,000 ppm, TSS - 1920 ppm, Zn - 218 ppm, Pb - 89.8 ppm,
Fe - 38.2 ppm, As— 59 ppm, Cd - 9.7 ppm and smaller amounts of Cu, Ni, and
g
Se . Several firms recycle this waste as a coolant to hot ESP units
following neutralization, thereby returning the metals content to
g
metallurgical processing.
Copper smelters also generate sulfuric acid waste from the
dimethylaniline absorption process which involves cleaning particulate matter
from SO. gas streams. This waste is similar to the sulfuric acid blowdown
waste described above. Other acidic wastes include spent sulfuric acid from
electrolyte solutions or sludges from recovery (e.g., dialysis, evaporators)
of copper from these solutions. These wastes contain high concentrations of
3-12
-------
TABLE 3.1.3. SUMMARY OF CONSTITUENT CONCENTRATIONS FROM SAMPLED IRON AND STEEL PLANTS (mg/L)
u>
I
Pickling baths
pH (units)
Dissolved iron
Oil and grease
Suspended solids
Fluoride
Nitrates
Toxic organics
Chlorinated solvents
Phthalates
Phenols
Other
Antimony
Arsenic
Cadmium
Chromium
Copper
Lead
Nickel
Silver
Zinc
Cyanide
Sulfaric
acid
-------
nickel sulfate along with copper (2 percent), lead (1 percent), iron, zinc and
nearly all the arsenic, antimony, and bismuth that comes in with the anode
8,18
copper. '
Primary lead smelters generate sulfur dioxide off-gases from sintering
machines which are treated by sulfuric acid production, as described
previously. Combined with copper and zinc smelters, these plants generated
2.8 million tons of sulfuric acid in 1982. Secondary lead smelters
generate sulfuric acid wastes from battery breaking and leaching operations.
One source estimated that the industry generates over 1 million gallons of
8
1 percent sulfuric acid waste annually from battery breaking. These
effluents contain lead, antimony, lead compounds and metal alloys. Zinc
leaching wastewater consists of the spent leaching liquor, dilute sulfuric
o
acid, zinc, antimony, lead, copper, metal sulfides and metal chlorides.
Another nonferrous metals industry which has been identified as producing
corrosive acidic wastes is the secondary aluminum industry. Scrubber
wastewater samples from chlorine demagging at three plants showed the
following range in waste characteristics: pli - 1.65 to 3.70, TSS - 44 to
934 mg/L, Al - 474 to 16,600 mg/L, Zn - 2.4 to 38.7 mg/L and smaller
quantities of Cd, Cr, Pb, Cu, and Ni. Waste characteristics from other
nonferrous metal forming industries are summarized in Table 3.1.4. These data
demonstrate the considerable variation in raw waste characteristics which can
be found in a given industry.
3.1.4 Fabricated Metals, Machinery, Electrical Supplies and Transportation
Equipment
Metal cleaning, coating, fabrication and plating operations (SIC 34
through 37) are another significant source of corrosive wastes. Metal
stripping, cleaning and plating result in both acidic and alkaline spent baths
which can be heavily contaminated with heavy metal salts, oil and grease,
phosphates, silicates, carbonates, inhibitors, surfactants, organic
emulsifiers, solvents and suspended solids. Acidic rinsewaters may also
have low enough pH to be characterized as RCRA corrosive wastes, particularly
if water conservation techniques such as counterflow rinsing are used.
Sludges and solid wastes can result from bath treatment and recovery
3-14
-------
TABLE 3.1.4. CORROSIVE WASTE CHARACTERISTICS IN THE NONFERROUS METAL FORMING INDUSTRIES
Industry
subcategory
Titanium
Titanium
Beryllium
Precious
metals
Refractory
metals
Refractory
metals
1 Zirconium-
(j, Hafnium
Magnesium
Surface
treatment
waste
stream
Spent
bath
Rinae
water
Spent
bath
Rinse
water
Spent
bath
Rinse
water
Spent
bath
Spent
bath
Total
Oil and suspended
grease solids
(mg/L) (mg/L)
NA 1-66
NA 480-3.360
1 240
1-8 1-3.000
1 140
1-6 8.0-140
1.87-83.7 8.70-12.6
1- 100.000 70-270
Cyanide Lead
pH (mg/L) (mg/L)
0.53-6.90 0.02 0.050-5.90
1.80-2.20 NA 65.0-214
,
0.32 0.04 NA
1.30-2.50 0.02 NA-
0.80 NA NA
1.50-2.10 NA NA
1-3.9 0.118-0.356 NA
0.80-12.60 ' NA NA
Other
Zinc (ntg/L) (ing/L) (mg/L) (mg/L) Type
0.020-0.660 1.7-52 1.1-215 NA Titanium
2.0-166 NA 74,000-98,000 NA Titanium
NA NA 79,000 0.18 Beryllium
NA NA NA 1.80-60.6 Cadmium
Silver
NA NA 0.27 6.30 Nicltel
NA NA 1.1-3,000 0.035-0.400 Nickel
0.17-7.5 104-681 6,500-17,100 NA Chromium
8.00-138 0.3-97 1.6-126 NA Chromium
Magnesium
metal(a)
(mg/L)
3.55-186
27,900-60,300
15,000
0.02-11.1
0.01-6.70
12.4
0.016-10.2
12-24
0.350-83,600
<1. 0-12, 700
NA - Data not available.
Source: Adapted from Reference No. 19.
-------
operations such as continuous filtering in spray applications. Table 3.1.5
lists surface treatment unit operations which may generate corrosive wastes.
Table 3.1*6 provides examples for general types of discharges.
Alkali cleaning is associated with lead, nickel, precious metals,
titanium, refractory metals, and zirconium forming operations and cleaning
19
lead/tin/bismuth surfaces. Alkaline cleaning is performed prior to acid
pickling to loosen scale (potassium permanganate and sodium hydroxide or soda
ash) and for removing oil, grease, waxes, soap, or dirt from metal surfaces
(sodium hydroxide, sodium carbonate, sodium metasilicate, sodium phosphates,
sodium silicate and wetting agents). These wastes have cleaner concentrations
25
of 30 to 90 g/L . They have lower concentrations of basis metal relative .
to acid baths since alkaline wastes are not generally corrosive. However,
they will tend to be higher in oil and grease as well as chlorinated organic
solvents since they are used after degreasing operations to remove hydrophobia
residue which interferes with subsequent surface treatment.
Sodium hydroxide is one of the most widely used paint strippers for
cleaning painting equipment and salvaging ferrous metal parts with defective
25
finishes. It is used in solutions of approximately 0.4 kg/L with
additives such as -surfactants, sequestering agents and/or solvent blends to
25
increase stripping effectiveness. Alkalis are also used in
electrocleaning (60 to 105 g/L of cleaner) and in the chemical milling of
aluminum (32 to 60 g/L NaOH). The latter are discarded when sodium aluminate
25
concentration reaches 111 to 146 g/L.
Other milling baths, etchants and pickling solutions are generally
comprised of acids. Traditional acid pickling solutions are based upon
inhibited suIfuric, hydrochloric or phosphoric acids with certain applications
for nitric, hydrofluoric and chromic acids (Table 3.1.7). Almost all metals
other than aluminum are etched in acid baths which contain chlorides such as
hydrochloric acid or ferric chloride. Highly concentrated solutions of
sulfuric acid and/or phosphoric and/or chromic acids are frequently used for
electropolishing. Conversely, brightening involves the use of more dilute
solutions of oxidants such as chromic acid, nitric acid or hydrogen
•j 25
peroxide.
3-16
-------
TABLE 3.1.5. SURFACE TREATMENT UNIT OPERATIONS
Preparative
Acid or alkaline pickling and etching (of metals)
Etching (of glass)
De-rusting (of metals)
Oil and grease removal (from metals, glass and plastics)
Machining and grinding (of metals and glass)
Bonding preparation (of metals)
Chemical and electrochemical machining (of metals)
Protective and Decorative
Electroplating (on metals and plastics)
Anodizing (of aluminum and magnesium)
Galvanizing
Rustproofing
Enameling
Immersion coating (of metals)
Metallization (of glass)
Chemical and electrochemical polishing
Passivation (of treated metal surfaces)
Painting and lacquering
Surface hardening (of metals)
Electroforming (on metals)
Coloring and bronzing (of metals)
Source: Reference 24.
3-17
-------
TABLE 3.1.6. CLASSIFICATION OF TYPES OF DISCHARGE
FROM SURFACE TREATMENT PROCESSES
Type Examples
Acids HC1, HoS04, H3PO4 (and acid phosphates), HF,
113603 (often discharged with dissolved heavy
metals present).
Alkalis NaOH, Na£C03 (frequently with phosphates,
silicates and detergents, and often containing oils
and oil emulsions).
Heavy metals Cd, Co, Cr, Cu, Fe, Mo, Mn, Ni, Pb, Sb, Sn, Zn
in solution
Complex-forming CN~, amines, Nlfy, EDTA, NTA, citrate, tartrate,
agents oxalate, gluconate.
Organic Aldehydes, ketones, alcohols, fatty and aromatic
additives carboxylic acids, carbohydrates, sulphonic acids,
dyes, phenols.
Solvents Trichloroethylene, toluene, xylene, alcohols.
Oils, waxes Sometimes discharged with detergents in emulsified
and greases form.
Inert solids Grinding materials (oxides, carborundum, etc.).
Source: Adapted from Reference 24.
3-18
-------
TABLE 3.1.7. ACIDS USED IN METAL PICKLING SOLUTIONS
Basis metal
Application
Acid type (% by volume)
Ferrous metals
excluding
stainless steel
Stainless steel and
high nickel alloys
Removing heavy scale
Rapid removal of light
scale or rust
Removal of light scale or
rust, prior to- painting
Heavy scale removal
Cuprous metals
Aluminum and
its alloys
Magnesium
Zinc
Galvanized steel
Light scale
Rapid scale removal
Slow scale removal
Brightening
Deoxidation
Scale removal
Scale removal when alloy
contains silica
Deoxidation
Deoxidation
Sulfuric (5 to 10%)
Hydrochloric (25 to 50%)
Phosphoric (15% or more)
Hot sulfuric (10% with sodium
thiosulfate or hydrosulfite
(1 to 2%)
Hydrofluoric (2%) with ferric
chloride (6 to 8%)
Nitric (20%) with hydrofluoric
(2 to 4%)
Hydrochloric (25 to 100%)
Sulfuric (5 to 10%)
Nitric acid modified with
chromic acid
Sulfuric inhibited with
chromic acid
Phosphoric acid
Chromic acid
Sulfuric or nitric acid with
chromic acid
Hydrofluoric acid
Sulfuric, hydrochloric or
phosphoric acid
Phosphoric acid
Source: Adapted from Reference 25.
3-19
-------
Other metal containing acidic corrosives are generated from the
manufacturing and coating (e.g., plating, electroplating, chromating,
phosphating, metal coloring) of metal products. Acidic paint strippers
include inhibited solutions of nitric, sulfuric and chromic acid which can be
combined with alcohols and glycol ethers. Phosphoric acid is used in hot
strippers for aluminum and zinc substrates. Nitric acids find use in removing
very tough hard coatings while sodium hydroxide has been applied for removing
25
polyurethane coatings.
As with other highly contaminated baths, these can be recycled to recover
metal, acid or alkaline constituents. The resulting treatment residuals
(e.g., filters, spent adsorbents, dewatered sludges) may constitute RCRA
corrosive wastes.
There is no data available which summarizes waste characteristics for the
diverse industries which utilize metal finishing and cleaning processes.
However, Table 3.1.8 gives an indication of the general range in corrosive
waste constituent concentrations which are generated by industries involved in
the manufacturing and coating of metal products. Detailed waste
characteristics for these operations can be found in the literature devoted to
specific industries. •
3.1.5 Electric Utilities
Electric utilities (SIC 49) generate corrosive wastes primarily during
boiler and preheater cleaning and water treatment operations. Alkaline boiler
cleaning is used to remove oil-based compounds from fireside tubes and
metallic copper deposits. Organic and inorganic acids are used to remove
boiler scale (e.g., iron and copper oxides) resulting from corrosion. Of
these, only spent inorganic acids tend to meet pH requirements for RCRA
corrosive wastes. Hydrochloric acid is the most widely used boiler cleaning
27
compound since sulfuric acid produces highly insoluble calcium sulfate.
Typical cleaning solutions contain 5 to 10 percent HC1 with inhibitors to
27
reduce attack on boiler surfaces. Copper complexers (e.g., thiorea) may
also be used to prevent copper chloride from reacting with iron to form copper
deposits. Noncoraplexed waste stream sampling data showed an average pU of 1.1
with high chemical oxygen demand (2,867 mg/L), low suspended solids (45 mg/L)
and oil and grease (15 mg/L), and high concentrations of iron (2,625 mg/L),
3-20
-------
TABLE 3.1.8. RAW WASTE CHARACTERISTICS - VARIATIONS AMONG PLANTS INVOLVED
IN THE MANUFACTURING AND COATING OF METAL PRODUCTS
Material coating
Parameter*
PH
Turbidity (JTU)
Temperature
Dissolved oxygen
Sulfide
Cyanide
Total solids
Total suspended solids
Settleable solids
Cadmium
Chromium, total
Chromium, hexavalent
Copper
Fluoride
Iron, total
Iron, dissolved
Lead
Oil, grease '
COD
Total phosphates
Zinc
Boron
Mercury
Nickel
Silver
Minimum
1.5
0.300
9.0
1.0
0.010
0.010
35.0
0.200
0.200
0.002
0.005
0.005
0.011
0.130
0.103
0.003
0.006
0.500
3.7
0.200
0.020
0.050
0.002
0.007
0.002
Maximum
11.3
3800.0
63.0
12.0
24.0
1.6
63090.0
28390.0
40.0
60.9
400.0
36.4
1060.0
110.0
422.2
367.7
102.8
13510.0
40000.0
62.4
86.5
21.3
0.055
0.950
0.100
Chemical and
electrochemical
operations
Minimum
0.9
1.2
10.0
4.0
0.100
0.019
151.0
1.6
0.150
0.002
0.005
0.005
0.017
0.150
0.023
0.138
0.018
0.40
1.0
0.200
0.153
0.120
0.008
0.006
0.002
Maximum
7.4
500.0
65.0
12.0
1.2
0.20
54210.0
16560.0
18.0
0.29
119.1
89.6
155.4
7.4
600.0
26.6
2.0
1730.0
6040.0
17.5
164.3
1.8
0.008
84.5
0.010
Assembly
Minimum
1.8
2.2
2.0
1.0
0.010
0.048
165.0
6.3
0.002
0.002
0.005
0.005
0.013
0.110
0.070
0.030
0.007
0.400
11.6
0.250
0.020
0.030
0.002
0.004
0.002
operations
Maximum
11.5
3800.0
63.0
12.0
4.8
0.192
28770.0
28390.0
60.9
60.9
0.026
0.007
184.6
325.0
95.4
77.5
102.8
13510.0
40000.0
62.4
33.9
17.0
0.055
93.5
0.052
Metal forming
(excluding
plastics)
Minimum
1.5
2.2
12.0
1.0
0.010
0.020
155.0
2.0
0.20
0.002
0.005
0.005
0.016
0.120
0.110
0.030
0.-010
0.500
1.0
0.20
0.020
0.030
0.002
0.004
0.002
Max imum
12.0
3800.0
65.0
12.0
6.4
1.8
63090.0
11990.0
40.0
0.43
0.417
0.030
145.0
1.8
600.0
250.0
103'. 0
8056.0
19170.0
45.4
146.4
16.3
0.002
165.2
0.44
"All parameters measured in mg/L except pH, turbidity (Jackson Turbidity Units)
Source: Adapted from Reference 26.
and temperature (°C).
3-21
-------
nickel (174 mg/L), calcium (53 mg/L), zinc (43 mg/L) and copper
27
(19 mg/L). Toxic metals which were sometimes present in high enough
concentration to qualify as RCRA wastes included lead (up to 5.2 mg/L) and
27
chromium (up to 8.8 mg/L). Coraplexed wastes showed higher levels of
suspended (2,375 mg/L) and dissolved solids (30,980 mg/L), iron (4,078 mg/L),
zinc (415 mg/L), copper (392 mg/L), nickel (240 mg/L), and chromium
27
(16.8 mg/L). Similar waste stream constituents are found in corrosive
air-preheater cleaning wastes."
Other potentially corrosive wastes from electric power generating
facilities include feed water demineralizer/ion exchange regeneration waste
and, in coal-fired facilities, fly ash and coal cleaning waste leachate. Ion
exchange wastes are very low (H«SO,) or very high (NaOH) in pH, have high
dissolved solids, and have copper, iron and zinc levels of 20 to
28
200 mg/L. One sample of fly ash transport water showed a high pH with
28
high levels of calcium (99 mg/L), but low levels of toxic metals.
Leachate from coal cleaning wastes can have pH of 2.0 or less and contains
high concentrations of iron (3,310 mg/L), aluminum (370 mg/L), calcium
(350 mg/L), magnesium (54 mg/L) and zinc (16 /mg/L), but does not exhibit the
29
characteristics of EP toxicity for heavy metals. - ,
3.1.6 Other Industries Which Generate Corrosive Wastes
The textile industry consumes significant quantities of corrosives (e.g.,
sodium hydroxide with smaller amounts of acetic acid) in textile treating and
finishing operations. Sodium hydroxide is used in cotton mercerizing
(15 to 30 percent solution) and scouring (1 to 5 percent) and is frequently
. . 20
recovered for reuse.
The paper industry consumes large quantities of sodium hydroxide and
lesser amounts of sulfuric acid and sodium carbonate. Sodium hydroxide is
used in chemical pulping (e.g., kraft sulfate process) in which wood chips are
cooked in digesters in highly alkaline solutions. It is also used to recycle
waste paper and for pulp bleaching. Some of these uses are non consumptive
and result in the generation of corrosive wastes.
Other miscellaneous industries which generate corrosives include dye
works and printing facilities which generate spent process solutions or
3-22
-------
20
equipment cleaning wastes. The largest source of waste corrosives from
small quantity generators is vehicle maintenance shops which dispose of used
batteries. Much of these wastes are recycled by secondary lead
Q
smelters.
3-23
-------
3.2 RCRA CORROSIVE WASTE GENERATION AND MANAGEMENT
The most recent, comprehensive source of data regarding corrosive
hazardous waste generation and management is a study performed by Camp,
Dresser and McKee, Inc. (September, 1984) for the USEPA. This study was
based on the results of the National Survey of Hazardous Waste Generators and
Treatment, Storage and Disposal Facilities (TSDFs) Regulated Under RCRA in
1981. Details on the survey methodology and limitations can be found in the
reference. The CDM analysis differed from the original survey results in
that an adjustment was made for a large volume (8.4 billion gallons) waste
stream which was found to be exempt from regulation under RCRA. Highlights of
the survey results are as follows:
• The total quantity of corrosive waste (D002 and K062) generated in
the United States in 1981 was 21.8 to 25.6 billion gallons. This
represents approximately 40 percent of all hazardous waste
generation. However, it includes mixtures of corrosive and
non-corrosive wastes which significantly inflates the total
estimate. *
• The number of generators of D002 corrosive wastes was 4,705 which
" represents over one-third of all RCRA waste generators. These
wastes were handled at 513 TSDFs in 1981. In addition, 64 TSDFs
handled K062 corrosive wastes.
• The total quantity of corrosive waste that was reportedly land
disposed, and thus affected by the proposed land disposal ban, was
3.6 to 4.2 billion gallons.
• Nearly 95 percent of land disposed corrosives were liquid acidic
wastes.
Disposal by deep-well injection accounted for 87 percent of land
disposed corrosives with another 6 percent disposed in surface
•imtvninHmpnf-R and S nereent in landfills.
U^9LTUOCU WV^&lXOAV^O w*fc»» «uv^»*«»» w £»«»&«»**
impoundments and 5 percent in landfills.
3-24
-------
3.2.1 Corrosive Waste Generation Estimates
As indicated, CDM presented a range for waste quantity generated and
managed at TSDFs. The upper bound of corrosive waste generation was
determined by summing the quantity of onsite generation at TSDFs, as reported
in the National Survey TSDF Questionnaire, and total generation at facilities
31
without onsite TSDFs, as reported in the Generator Questionnaire. This
quantity exceeded total waste quantity which had an identified management
practice by a factor of 1.235. Thus, this was used as a scale-up factor to
estimate maximum quantities handled by various management practices.
As a result of misinterpretation of the survey questionnaire, this
procedure resulted in an overestimation of hazardous waste generation,
31
particularly for corrosives. Most significantly, waste stocks carried
over from storage in 1980 and wastes handled in RCRA-exempt processes were
inadvertently included in the data. Exempt processes include discharges to
POTWs, wastes treated exclusively in tanks and discharged under NPDES permits,
and wastes which are 100 percent recycled. A significant, but unknown,
14
fraction of corrosives are handled in these ways, and thus may have
inflated the waste quantity estimates. '
The lower bound was taken as the summation of the highest quantity
reported for specific waste codes at individual facilities in either
treatment, storage, or disposal processes. This procedure eliminated possible
double counting for waste volumes reported under multiple management methods.
However, at the same time, it excluded fractions of some waste streams which
were legitimately handled in lower volumes via other management methods. The
resulting quantity exceeded total quantity with known management practices by
a factor of 1.05. Consequently, this was used as the scale-up .factor for
3
generating minimum waste quantity estimates. However, this data was still
inflated, but to a lesser extent, due to the inclusion of exempt wastes.
It becomes clear that the majority of corrosive wastes reported in the
National Survey must be dilute aqueous streams when waste generation
quantities are compared to acid/alkali production figures (Table 3.1.1).
1 2
Total production is approximately 19 billion gallons ' (100 percent
acid/alkali basis) versus waste generation estimates of 21.8 to 25.6 billion
gallons. These figures correspond roughly to the following waste profile.
3-25
-------
Five percent of acid/alkali (e.g., nonconsuraptive applications) results in the
generation of concentrated waste (10 percent concentration) at a rate of
0.1 gallons of waste produced per gallon of acid/alkali consumed. The
remaining 95 percent (e.g., consumptive applications) results in 0.01 gallons
of waste (1 percent concentration) generated per gallon consumed. Acid
1 • 3 21
consumption and waste generation in steel pickling (K062) * corresponds
well to the concentrated waste profile described above. However, there are no
available data with which the dilute waste profile can be compared.
Two other surveys which were national in scope support the waste quantity
estimates derived in the EPA National .Survey. The U.S. Congressional Budget
Office collected waste quantity data through a survey of TSDF and industrial
facility manifests. These data were extrapolated to national totals for each
SIC code based on percentage of employees in the surveyed sample. The CBO
estimated that 27 billion gallons of acidic and basic waste are
32
generated. This accounted for 38 percent of all hazardous waste
generation reported in the survey (versus 40 percent reported in the National
Survey).
Another source, based on a survey of waste generation in 21 states,
concluded that acid/alkali wastes (liquids only) accounted for 32 percent of
all hazardous waste generated. Spent pickle liquor generation was
estimated to be nearly 1.4 billion gallons (36 percent H_SO,, 58 percent
HC1, 6 percent combination acid). These data also compare favorably with the
National Survey results.
These waste quantity estimates must be interpreted in terms of the types
of wastes included in the analyses. These surveys were designed to assess
waste quantity handled by the facility as opposed to waste quantity initially
produced at the source; i.e., prior to subsequent mixing or treatment.
Thus, waste quantity estimates were significantly inflated as a result of the
mixture rule (40 CFR 261.3(a)(2)(iii)); i.e., the largest volume of diluted
corrosive wastes which still met the definition of corrosivity (40 CFR 261.22)
could have been reported. As a basis of comparison, the ISDB survey of the
Chemicals and Allied Products Industry (SIC 28) projected a corrosive waste
generation of 9.8 billion gallons/year versus 15.5 to 18.3 billion gallons
o
as estimated by the National Survey. The ISDB survey was specifically
designed to assess waste quantities and characteristics at the point of
3-26
-------
TABLE 3.2.1. MANAGEMENT PRACTICE SUMMARY FOR CORROSIVE WASTES
NATIONAL ESTIMATES (Million Gallons)
High quantity estimate3
Handled0
Disposed0
Injection well
Landfill
Land treatment
Surface impoundment
Other
Treated0
Tanks
Surface impoundment
Incineration
Other
Stored0
Tanks
Containers
Surface impoundment
Waste piles
Other
Recycled*1
Ons ite:
Generator
TSDF
Off site:
Generator
TSDF
Corrosive
waste
D002
24,596
3,970
3,635
85
18
206
26
16,127
7,040
5,614
6
3,252
10,122
1,542
9
6,530
6
2,007
373
330
42
288
43
14
29
Spent
pickle
liquor
K062
1,048
236
28
131
-
56
20
220
139
39
-
40
320
211
-
47
7
—
354
34
6
28
320
170
150
Total
25,644
4,206
3,663
217
18
262
46
16,347
7,180
5,653
6
3,292
10,441
1,754
9
6,577
13
2,007
727
364
48
316
363
184
179
Low quantity estimate'7
Corrosive
waste
D002
20,912
3,375
3,090
73
15
175
22
13,711
5,986
4,773
5
2,764
8,605
1,311
8
5,552
5
1,706
317
280
35
245
36
11
25
Spent
pickle
liquor
K062
891
200
24
112
-
48
17
187
118
33
-
34
272
180
-
40
6
—
301
30
6
24
272
143
129
Total
21,803
3,576
3,114
184
15
223
39
13,898
6,104
4,806
5
2,799
8,877
1,491
8
5,592
11
1,706
618
310
41
269
308
154
154
a1.235 x base data.
b1.05 x base data.
°Source of base data: Reference 3.
dSource of base data: Reference 35.
3-27
-------
production, thereby excluding the effects of subsequent mixing or
32
treatment. As a result, the waste generation estimate is 37 to 46 percent
lower than that given in the National Survey.
3.2.2 Corrosive Waste Management Practices
High and low waste quantity estimates handled in different management
practices are summarized in Table 3.2.1, as extrapolated from the National
3 35
Survey. ' However, since the data includes mixtures of wastes, actual
quantities may be less, particularly for wastewater handling processes in
which mixing was most likely to occur. For example, surface impoundment and
tank quantities are expected to be more inflated than land disposed
quantities, since waste segregation is more likely to be practiced in the
latter.
This assertion is supported by comparison of the data with the ISDB.
Overall, the Chemical and Allied Products Industry accounts for 71 percent of
corrosive waste generation. The ISDB, when extrapolated to national totals
(factor of 2.1), shows that the chemical industry accounts for virtually all
of the deep well injected corrosives. Thus, these wastes are probably
infrequently mixed with other (e.g., nonhazardous) waste streams prior to
disposal. The data also suggest that the National Survey underestimated
recycled quantities of D002 by nearly a factor of three (6 percent of waste
generation versus 2 percent in the National Survey). This is not surprising
since the National Survey was not designed to assess recycling, particularly
31
methods which may be exempt from RCRA regulations. Finally, the ISDB also
identified higher quantities of waste being incinerated although this
management practice still remains infrequently used.
The National Survey understated actual recycling (particularly onsite
recycling) since many of these activities are exempt from regulation. Total
quantity recycled was estimated to be -618 to 727 million gallons in 1981
20
(Table 3.2.1). Of this, 48.7 percent was spent pickle liquor. Although
pickle liquor (K062) accounts for only 4.1 percent of total corrosive waste
generation, it is frequently recovered since it is concentrated (5 to
21
15 percent) and used in high volumes. Roughly ten times as high a
percentage of pickle liquor is recycled as compared to DOU2 wastes.
3-28
-------
Wastes accepted for offsite recycling tend to be highly concentrated
(i.e., greater than 10 percent) and are generally not accepted if they contain
high levels of contaminants such as toxic metals. Conversely, wastes
recycled onsite are not restricted by contaminant concentration unless it
interferes with the recovery process. Onsite recycling has been shown to be
economical for acid/alkali concentrations of lower concentrations
(Section 5.0).
One source estimated offsite shipments of sulfuric acid destined for
recycling was 527 million gallons in 1981 (743 MG in 1982), 60 percent of
2 . .
which.was generated by the petroleum industry. The remainder originated
from steel picklers, soap and detergent producers, industrial chemicals, and
phosphate fertilizer manufacturers. In addition, large quantities of
hydrochloric and sulfuric acid pickle liquors, corrosive textile finishing
solutions, pesticide wastes, and metal plating baths are currently recycled in
onsite facilities (Sections 3.1 and 5.0). The most frequently recovered waste
constituents in the chemical industry are sulfuric acid, hydrochloric acid,
14
caustic soda and sodium carbonate. Out of 29 recovered liquid wastes
reporting concentration data, all reported acid/alkali concentrations in
. 14
excess of 10 percent.
3.2.3 Corrosive Waste Generation by Industrial Classification (SIC)
Table 3.2.2 summarizes National Survey waste quantities handled and
recycled by various industrial classification codes. Table 3.2.3 shows the
number of facilities handling and disposing corrosives. As shown, the
Chemical and Allied Products Industry accounts for 71.5 percent of the waste
quantity. This industry tends to generate a high fraction of its waste in the
form of large-volume aqueous streams relative to other industries. Electric,
Gas, and Sanitary Services is the second largest waste generator (9 percent),
followed by the Primary Metal Industries, Petroleum Refining, and Paper and
Allied Products, each with approximately 4.5 percent of the total. The
remaining corrosive waste is primarily generated in other metal-related
industries.
The data shown previously in Tables 3.2.2, 3.2.3, and 3.1.1 permit
comparisons to be made between corrosive raw material consumption, waste
generation, and waste management practices. The chemical industry accounts
3-29
-------
TABLE 3.2.2. CORROSIVE WASTE QUANTITY HANDLED AND RECYCLED BY INDUSTRIAI
CLASSIFICATION (MiUion gallons/year)
Waste quantity
I1C
code
28
49
29
33
26
36
35
32
34
37
20
42
11
50
95
97
handled*
(Hill ion gallona/year)
Industry description
Chemicals and allied product!
Electric, gas and sanitary scrvicsi
Petroleum refining
Primary metals
Paper and allied products
Electric and electronic machinery,
equipment and auppliet
Machinery, except electrical
Stone, clay, glaai, concrete
Fabricated netals
Transportation equipment
Food and kindred products
Motor freight, transportation, warehousing
Agricultural production - crop!
Wholesale trade - durable gooda
Administration of environmental
quality programs
National aecurity and international
affairs
Other Industries
Total:
High
18,337
2,305
1,150
1,143
1,126
581
417
190
183
136
27
10
7
6
' 2
2
23
25,645
Low
15,590
1,960
. 978
972
957
495
354
162
156
115
23
8
6
5
2
1
. 20
21,803
Percent
71. J
9.0
4.5
4.5
4.4
2.3
1.6
0.7
0.7
0.5
0.1
-
-
-
-
-
O.I
100.0
Waste recycled1"
(Million gallons/year)
High
377
0.2
11
350
26
14
25
2
9
14
0.1
0.0
0.0
0.1
0.0
0.9
1.3
829
tow Percent
320 45.4
0.2
9 1.3
297 42.2
22 3.1
12 1.6
21 3.0
2 0.2
7 1.0
12 1.6
0.1
0.0
0.0
O.I
0.0
0.7 0.1
1.2 0.2
705 100.0
Percent
onsite
95
0
6
87
97
0
98
100
59
0
0
NA
NA
100
NA
75
30
56
Source: National Survey, Reference 3.
blnclud«. all corrosive waitei of which D002 and
Source: National Survey, Reference 20.
K062 represented 87.7 percent.
-------
TABLE 3.2.3. NUMBER OF FACILITIES HANDLING AND DISPOSING CORROSIVE WASTE
BY INDUSTRIAL CLASSIFICATION (million gallons/year)
SIC
code
28
49
29
33
26
36
35
32
34
37
20
42
11
50
95
97
Mo. of facilities handling
corrosive waste*
Industry description
Chemicals and allied products
Electric, gas and sanitary
Petroleum refining
Primary metals
Paper and allied products
Electric and electronic machinery,
equipment and supplies
Machinery, except electrical
Stone, clay, glass, concrete
Fabricated metals
Transportation equipment
Food and kindred products
Motor freight, transportation, warehousing
Agricultural production - crops
Wholesale trade - durable goods
Administration of environmental
quality programs
National security & international affairs
Other industries
Total :
High
503
141
80
136
17
290
98
17
178
122
5
25
1
6
2
52
207
1,880
Low
427
120
68
116
15
247
83
15
151
104
4
21
1
5
2
44
173
1,596
Percent
26.7
7.5
4.3
7.2
0.9
15.4
5.2
0.9
9.5
6.5
0.3
1.3
0.1
0.3
0.1
2.7
11.0
100.0
No. of facilities
disposing corrosive waste
High
58
35
15
24
-
2
-
1
15
1
-
1
-
-
_
2
16
170
Low
49
29
13
20
-
2
-
1
13
1
-
1
-
-
_
2
14
145
Percent
34.1
20.4
8.7
13.8
-
1.4
-
0.7
8.7
0.7
-
0.7
-
-
_
01.4
9.4
100.0
•includes. D002 and K062 only.
Source: Adapted from Reference 3.
3-31
-------
1 2
for nearly SO percent of known acid/alkali consumption ' and generates
3
nearly 72 percent of the total corrosive waste volume. However, it only
accounts for 34 percent of waste disposal and 45 percent of total
20
recycling. Thus, most corrosives in .SIC 28 are large volume, dilute
wastewater streams which are treated onsite. Of that which is eventually
disposed, over 99 percent is deep-well injected and the remainder is
14
landfilled. Organic chemicals industries generate the majority of SIC 28
wastes (86.6 percent) followed by plastics and resins (5.8 percent),
pesticides (5.3 percent), dyes and pigments (2.1 percent), and inorganic
4 34
chemicals industries (0.2 percent). '
Primary metal and metal-related industries (SIC 33 through 37) account
1 2
for less than 3 percent of strong acid/alkali use, ' but these are
primarily applied in concentrated form in noneonsuraptive applications such as
metal cleaning. As a results these industries account for a disproportionate
amount of waste generation (9.6 percent), waste disposal (24.6 percent),
20
and waste recycling (49.5 percent). The metals industry also accounts for
3
the largest percentage of waste generators (44 percent). Thus, these
facilities tend to generate significantly smaller volume waste streams than
the chemical industry. These wastes also tend to be more highly concentrated •
and are more frequently disposed in landfills. In addition to neutralization,
many of these will require treatment to remove or immobilize heavy metals when
the land disposal restrictions become effective.
The electric and gas industries (SIC 49) consume less than 3 percent of
all corrosive reagents for water and stack gas treatment and boiler
1 2
cleanout. ' The latter results in a concentrated, metals contaminated
waste stream which is infrequently recycled. As a result, these industries
account for a disproportionately high percentage of waste volume generation
3 3
(9.0 percent) and facilities practicing waste disposal (20.4 percent) .
In addition, they account for very little corrosive waste recycling (less than
20
0.1 percent). Many of these disposed wastes will require treatment for
both neutralization and metals removal to comply with proposed land disposal
restrictions (see Section 3.1.5).
The petroleum refinery industry accounts for 5.1 percent of known
1 2
corrosive raw material consumption, * and generates 4.5 percent of
corrosive wastes. The National Survey suggests that this industry performs
3-32
-------
little onsite recycling. However, other data indicates a large quantity of
spent sulfuric acid (330 million gallons in 1980, 440 million gallons in 1982)
is shipped offsite for reclaiming. This is spent sulfuric alkylation acids
which tend to be highly concentrated (95 percent acid or more).
Finally, the pulp and paper industry (SIC 26) consumes 3.8 percent of
1 2
known acid/alkali consumption, ' and generates 4.4 percent of the total
3 3
waste. However, almost none of this waste is land disposed. Instead,
it is handled in onsite wastewater treatment facilities or recycled. Other
industries which use large quantities of corrosives, but generate little
waste, are the glass manufacturing industry and food products. Together,
1 2
these industries consume 7.2 percent of known end uses, ' but generate less
3
than 1 percent of total corrosive wastes. Acids and alkalis in these
industries are largely used in consumptive applications.
3.2.4 Recent Changes in Corrosive Waste Generation
Total waste generation estimates may have changed from the 1981 estimates
provided by the National Survey. Since that time, corrosive raw material
1 "
consumption has been essentially stagnant, however, small quantity
generators (100 to 1,000. kg/month) have been included under RCRA regulations,
waste management practices have shifted toward increased compliance, waste
minimization and recycling, and wastes with free liquids have been banned from
landfills.
A survey of the eight largest commercial hazardous waste disposal firms
by ICF^ showed a 40 percent increase in landfilling between 1981 and 1984.
However, much of this increase was due to disposal of bulk liquids in advance
of the ban on these wastes and the disposal of remedial action clean-ups. It
is unlikely that currently landfilled quantities of corrosives exceed levels
reported in the National Survey. Instead, solidification and stabilization
requirements and increased transport and disposal costs would have shifted
management practices to increased volume reduction prior to disposal.
Deep well injection at the surveyed commercial firms showed a 24 percent
decrease while chemical treatment increased 33 percent between 1981 and
1984. Corrosive waste management practices have probably shifted in
similar fashion. Additionally, recycling practices have become more
3-33
-------
widespread in response to increased disposal costs. However, comprehensive
data does not exist which would allow the extent of this practice to be
quantified.
On August 1, 1985, the EPA proposed lowering the small quantity generator
38
exclusion limit from 1,000 to 100 kg/month. The effect of this regulatory
change (effective 22 September, 1986) will be to add 15.4 million gallons/year
30
of corrosive wastes to that already managed under RCRA. Of this,
approximately 52 percent is used lead-acid batteries. Another 72.3 million
gallons of spent batteries (90 percent of the total) are currently reclaimed
(e.g., by secondary lead smelters) and are thus exempt from regulation under
RCRA. It is likely that much of the currently disposed batteries will
also be reclaimed when these generators become subject to regulation. The
remaining corrosives are primarily small volume streams, a high percentage of
which will probably be landfilled. Overall, small quantity generators will
account for no more than a 5 percent increase in the quantity of corrosive
wastes which are landfilled.
In summary, corrosive waste generation subject to RCRA-handling
requirements has probably not changed significantly since the 1981 National
Survey. It is likely that waste quantity affected by the land disposal ban
has decreased somewhat, primarily due to the decrease in waste volume which is
deep well injected. Increased waste minimization efforts (e.g., waste
segregation) and recycling have probably also contributed to a decrease in the
quantity land disposed. For example, offsite sulfuric acid recovery
reportedly increased 35 percent from 1980 (2.3 million tons) to 1982
(3.1 million tons). The Congressional Budget Office projected an overall
increase in waste reduction of 20 percent from 1985 to 1990 for nonmetallic
32
inorganic liquids and 40 percent for metal-containing liquids. Thus, with
stagnant demand for corrosive raw materials, it is likely that corrosive waste
quantity will decline in the next few years.
3.2.5 Corrosive Waste Characteristics Summary
Waste characterization data for corrosive wastes is limited. CDM
presented physical profile data for disposed corrosives, as summarized
below. Other sources are limited to specific waste management practices or
industries, and thus cannot be extended to include the universe of corrosive
wastes.
3-34
-------
Available data indicate that, overall, acidic wastes are generated in
significantly higher quantity than alkaline wastes. A survey of waste
generation in 21 states estimated that 65 percent of all corrosives generated
33
were acidic. This estimate is consistent with acid/base consumption
figures which show 68 percent represented by acids (Table 3.1.1). Other data
show that the majority of corrosives (80.6 percent) do not contain heavy
32
metals. In particular, only 3 percent of the waste streams reported in
14
the chemical industry contain heavy metals. However, since the chemical
3
industry generates 72 percent of all corrosives, this suggests that over
40 percent of other industry wastes contain metals.
Waste characterization data for land disposed corrosives which will be
affected by the land disposal ban, is also limited. The National Survey data
base provided waste characteristic information for 14 percent of land disposed
D002 wastes. Of this, 92.5 percent was characterized as liquid, 7.4 percent
were sludges, and only 0.1 percent were solids. The majority of wastes were
acidic (82 percent), inorganic (82 percent), and characterized as dilute
(liquids only, 94.3 percent). Solids, sludges, and concentrated wastes had a
significantly higher tendency to be characterized as organic relative to
combined wastes (48, 93, and 31 percent, respectively). The most predominant
sludges were waste alkylation acid and wastewater treatment sludge, while
solids were dominated by spent catalysts, adsorbents, and filter residue.
Other waste characterization data generated from the survey was based on a
very limited number of responses, and thus cannot be used to identify overall
trends in land disposal. However, the data presented above show good
agreement with the ISDB.
Other data sources are either not specific to land disposal or do not
cover the entire range of waste generating industries. A review of waste
manifests by the State of California showed 38 percent of landfilled
corrosives containing metals at levels which exceeded proposed land disposal
O£
ban treatment standards. The most common toxic metals in decreasing order
were Cr, Ar, Ni, Pb, and Cd. The National Survey data showed 60 percent of
landfilled waste was spent pickle liquor (K062) which would have
characteristics similar to those presented previously in Table 3.1.3. Of the
remainder, only wastes from the chemical industries have been characterized.
These are described below.
3-35
-------
The ISDB summarized characteristics of landfilled wastes from
14
SIC 28. The most common constituents were water, caustic soda, sulfuric
acid, calcium chloride, sodium chloride, and phosphoric acid, often present in
concentrations ranging up to 75 percent. Sludges and solids were reported as
frequently as liquids, but accounted for only 8.0 percent of the total waste
volume. These wastes included spent adsorbents (e.g., activated carbon),
filters or filter acid (e.g., diatomaceous earth), and alumina. Very few
wastes contained oil or metals.
Wastes which are deep well injected originated almost exclusively from
the chemical industries, and thus are well characterized by the ISDB. Of
these, 75.9 percent of the total volume was characterized as aqueous liquids,
16.6 percent as sludge, and 7.5 percent as organic liquid. Other data
describing constituent types and concentrations were not available at the time
this document was published.
Other disposed methods for which waste characterization data are
available include incineration and use as a fuel. Incineration is
infrequently applied to corrosive wastes. The most commonly incinerated
14
corrosives are organic acids or wastes contaminated with solvents. Data
also show some incineration of concentrated inorganic acids such as sulfuric
(up to 85 percent concentration) and hydrochloric (up to 30 percent).
However, these account for only a small fraction of incinerated corrosives.
Similarly, few corrosives are disposed through fuel blending in boilers.
Those that are managed in this manner contain high organic concentrations
(25 percent or more) of constituents such as adipic, acetic, butyric,
14
propionic and formic acids as well as other organics.
3-36
-------
REFERENCES
1. Mansville Chemical Products Corporation, Cortland, N.Y. Chemical
Products Synopsis. 1983.
2. Chemical Marketing Reporter. Chemical Profiles. Schnell Publishing
Company. 1982 through 1986.
3. Camp, Dresser & McKee, Inc. Technical Assessment of Treatment
Alternatives for Wastes Containing Corrosives. Prepared for U.S. EPA
under Contract No. 68-01-6403. September 1984.
4. Huppert, M. Science Applications International Corporation. Telephone
conversation with M. Breton, GCA Technology Division, Inc., regarding the
Industry Studies Data Base. September 1986.
5. U.S. EPA Supplement for Pretreatment to the Development Document for the
Inorganic Chemicals Manufacturing Point Source Category. U.S. EPA
440/1-77/087A. July 1977.
6. U.S. Environmental Protection Agency*. Treatability Studies for the
Inorganic Chemicals Manufacturing Point Source Category. U.S. EPA
Effluent Guidelines Division. EPA 440/1-80/103. July 1980.
7. Sittig, M. Fertilizer Industry Processes, Pollution Control, and Energy
Conservation. Noyes Data Corp., Park Ridge, N.J. 1979.
8. Coleman, R.T., Radian Corp. Sources and Treatment of Wastewater in the
Nonferrous Metals Industry. Prepared for USEPA IERL under Contract
No. 68-02-2608. EPA-600/2-80-074. April 1980.
9. Mooney, G.A., CH2M Hill. Two-Stage Lime Treatment in Practice.
Environmental Progress, Vol. 1, No. 4. November 1982.
10. Ricker, N.L., and C.J. King. Solvent Extraction of Wastewaters from
Acetic Acid Manufacture. Prepared for USEPA ORD, EPA-600/2-80-064.
April 1980.
11. Federal Register. 40 CFR Part 261- Hazardous Waste Management System;
Identification and Listing of Hazardous Waste. Federal Register Vol. 49,
No. 90, pg. 19608. May 8, 1984.
12. U.S. EPA Development Document for Expanded Best Practicable Control
Technology, Best Conventional Pollutant Control Technology, Best
Available Technology, New Source Performance Technology, and Pretreatment
Technology in the Pesticide Chemicals Industry. U.S. EPA Effluent
Guidelines Division. EPA-440/1-82-079B. November 1982.
3-37
-------
13. Shelby, S.E., and R.W. Me Co Hum. A Case Study for Che Treatment of
Explosives Wastewater from an Army Ammunitions Plant. Presented in the
39th Industrial Waste Conference Proceedings. Purdue University, West
Lafayette, IN, Ann Arbor Books. May 8, 9, 10, 1984.
14. Science Applications International Corporation. Industry Studies Data
Base. August 1985.
15. PEDCo Environmental, Inc., Arlington, XX. Petroleum Refinery Enforcement
Manual. Prepared for U.S. EPA Division of Stationary Source Enforcement
under Contract No. 68-01-4147. EPA-340/1-80-008. March 1980.
16. Bider, W.L., and R. G. Hunt. Industrial Resource Recovery Practices:
Petroleum Refineries and Related Industries (SIC 29). Prepared for U.S.
EPA OSW under Contract No. 68-01-6000. June 1982.
17. Booze, Allen & Hamilton, Inc. Enhanced Utilization of Used Lubricating
Oil Recycling Process By-Products. Performed for U.S. DOE/BETC under
Contract No. DE-AC19-79BC10059. June 1981.
18. Franklin Associates, Ltd. Industrial Resource Recovery Practices: Metal
Smelting and Refining (SIC 33). Prepared for U.S. EPA OSW under Contract
No. 68-01-6000. January 1983.
19. U.S. EPA Development Document for Effluent Limitations Guidelines and
Standards for .the Nonferrous Metals Forming and Iron and Steel, Copper
Forming, Aluminum Metal Powder Production and Powder Metallurgy Point
Source Category. U.S. EPA Effluent Guidelines Division.
EPA-440/1-84/019B. February 1984.
20. Versar, Inc. National Profiles Report for Recycling/A Preliminary
Assessment. Prepared for U.S. EPA Waste Treatment Branch under Contract
No. 68-01-7053. July 1985.
21. U.S. EPA Development Document for Effluent Limitations Guidelines and
Standards for the Iron and Steel Manufacturing Point Source Category.
Vol. V. U.S. EPA Effluent Guidelines Division. EPA-440/1-82-024. May
1982.
22. U.S. EPA Development Document for Proposed Effluent Limitations
Guidelines and Standards for the Iron and Steel Manufacturing Point
Source Category. Volume 1. U.S. EPA Effluent Guidelines Division
EPA-440/l-79-024a. October 1979.
23. U.S. EPA Development Document for Effluent Limitations Guidelines and
Standards for the Iron and Steel Manufacturing Point Source Category.
Vol. VI. U.S. EPA Effluent Guidelines Division. EPA-440/1-82-024. May
1982.
24. Environmental Protection Service, Fisheries and Environment Canada.
Proceedings Technology Transfer Seminar on Waste Handling, Disposal and
Recovery in the Metal Finishing Industry. Toronto, Ontario,
November 12-13, 1975. Report No. EPS 3-WP-77-3. March 1977.
25. Kirk-Othmer Encyclopedia of Chemical Technology. John Wiley & Sons,
New York, N.Y. Third Edition. 1978.
3-38
-------
26. U.S. EPA. Controlling Pollution from the Manufacturing and Coating of
Metal Products. Water Pollution Control. U.S. EPA Environmental
Research Information Center. EPA-625/3-77-009. May 1977.
27. U.S. EPA. Development Document for Effluent Limitations Guidelines and
Standards, and Pretreatment Standards for the Steam Electric Point Source
Category. U.S. EPA Effluent Guidelines Division. EPA-440/1-82-029.
November 1982.
28. Benjamin, M.M. et al. Removal of Toxic Metals from Power Generation
Waste Streams by Adsorption and Coprecipitation. Journal of Water
Pollution Control Federation, Vol. 54, No. 11. November 1982.
29. Wanger, L.E., and J.M. Williams. Control by Alkaline Neutralization of
Trace Elements in Acidic Coal Cleaning Waste Leachates. Journal of Water
Pollution Control Federation, Vol. 54, No. 9. September 1982.
30. Ruder, E., et al., ABT Associates. National Small Quantity Generator
Survey. EPA-530/SW-85/004, U.S. EPA, Washington, D.C. February 1985.
31. Deitz, S., et al., Westat, Inc. National Survey of Hazardous Waste
Generators and Treatment, Storage, and Disposal Facilities Regulated
Under RCRA in 1981. Rockville, MD. U.S. EPA/OSW. April 1984.
32. U.S. Congressional Budget Office. Hazardous Waste Management - Recent
Changes and Policy Alternatives. CBO Congress of the United States. May
1985. m • --
33. Noll, K.E. et'al. Recovery, Recycle and Reuse of Industrial Wastes.
Lewis Publishers, Inc., Chelsea, MI. 1985.
34. U«St EPA. Report to Congress on the Discharge of Hazardous Wastes to
Publically-Owned Treatment Works. U.S. EPA Office of Water Regulations
and Standards. EPA-530/SW-86/004. February 1986.
35. DPRA, Inc. Written Communication to M. Arienti, GCA Technology Division,
Inc., regarding analysis of Recycling Data from the National Survey Data
Base. Data Request No. M850415W. June 10, 1985.
36. Radinsky, J. et al. Recycling and/or Treatment Capacity for Hazardous
Wastes Containing Dissolved Metals and Strong Acids. California
Department of Health Services. October 1983.
37. ICF, Inc. Survey of Selected Firms in the Commercial Hazardous Waste
Management Industry: 1984 Update. Prepared for U.S. EPA Office of
Policy Analysis. September 1985.
38. Federal Register. 40 CFR Part 261 (50 FR 31278). August 1, 1985.
3-39
-------
SECTION 4
. NEUTRALIZATION TREATMENT TECHNOLOGIES
All neutralization processes operate under the same fundamental chemical
principles and utilize similar types of equipment and process configurations.
Additionally, pretreatment requirements and residual post-treatment options
are comparable, regardless of the specific neutralization method under
investigation. Therefore, in an effort to minimize redundancy, similar
aspects of neutralization systems are addressed prior to discussion of
specific reagent/waste combinations. Section 4.1 serves as introduction to
the basic theory of acid-base chemistry and proceeds to identify
considerations in reagent selection, pretreatment requirements, neutralization
equipment, process configurations, post-treatment and disposal of residuals.
The remaining subsections (Sections 4.2 through 4.7) are organized according
to specific neutralization reagents. These highlight the unique aspects of
each, including compatible waste types, treatment costs, sludge generation and
special considerations in equipment design and reagent handling practices.
The reagents include:
• Other acid/alkali wastes;
• Limestone;
• Lime slurry (lime, waste carbide lime, cement kiln dust);
• Caustic soda (sodium hydroxide, sodium carbonate);
• Mineral acids (hydrochloric and sulfuric acids); and
• Carbonic acid (carbon dioxide, boiler flue gas, submerged
combustion).
4-1
-------
Each of the reagent subsections covers the following topics:
• General process description including typical operating
characteristics;
• Case study data which identifies the range in potential
applications, processing equipment, and system configurations;
• Capital and operating costs; and
• Status of the technology.
4.1 GENERAL CONSIDERATIONS
4.1.1 Acid Base Theory
The principle mechanism of neutralization involves the reaction between
an acid and a base. An acid is any substance that dissociates in solution to
produce a proton (H+), and a base is any substance that combines with or
accepts a proton. Strong acids or bases are characterized by complete
disassociation, while weak acids or bases will disassociate only slightly. In
general, acid-base reactions form a salt and water as illustrated in the
following reaction between hydrochloric acid and potassium hydroxide:
HC1 + KOH—» K+:Cl" + HjO + Heat (1)
acid base salt water
Neutralization reactions are exothermic in nature, but with careful reagent
addition and adequate mixing, the excess heat can be safely dissipated. The
concentrations of H and OH ions and the equilibrium constants are
usually very small numbers, therefore, it is convenient to express them in
logarithmic terms. The conventional pH scale, which represents the negative
logarithm of the hydrogen ion concentration, employs numbers ranging from 0 to
14 to indicate relative acidity or alkalinity. The pH of acidic solutions is
less than 7, while the pH of alkaline solutions is greater than 7.
4-2
-------
Other means of expressing acidity or alkalinity include the basicity
' factor and concentration. The basicity factor is a measure of the theoretical
neutralizing power of alkali reagents. It is calculated the total weight of
the reagent's potential hydroxyl ions. A neutralizing value of 1.0 is
assigned to pure calcium oxide (CaO) and the values for other alkali reagents
are expressed in relative terms. For example, to neutralize 98 Ibs of
sulfuric acid, it requires either 80 Ibs of sodium hydroxide or 56 Ibs of
quicklime (CaO) based on stoichiometry. Thus, the basicity factor of sodium
hydroxide becomes 56/80 or 0.69. The alternate method of expressing pH is to
report the concentration of the acid or alkali in milligrams per liter. For
example acidity can be expressed as mg/L of H.SO, while alkalinity can be
expressed as mg/L of CaCO~ (limestone).
Most neutralization applications consist of adjusting an acidic or
alkaline waste stream with the appropriate reagent to a final pH of 6 to 9
which meets surface water discharge requirements established under the Clean
Water Act. However, it is sometimes only necessary to adjust the pE to
approximately 5 to 6 (i.e., partial neutralization) to achieve certain
treatment objectives. In other applications it may .be necessary to neutralize
an acid to pH 9 or higher to precipitate metallic ions or to completely
clarify a waste for acceptable discharge. These techniques are called under-
and over-neutralization, respectively .
With few exceptions, reactions involving pH go to completion in a fairly
short time. However, an understanding of the reactivity of neutralization
reagents is necessary for proper design and sizing of equipment, particularly
tankage and space. Reagent reactivity and kinetics are discussed individually
in later sections to provide a more detailed characterization for specific
reagent/waste combinations.
In contrast, one general process variable that affects reaction kinetics
and reagent consumption is the buffer capacity (the ability of a solution to
resist change in pH) of the waste stream. If a certain amount of acid (or
alkali) is added to a specified volume of buffered process solution, the
change in pH will be much less than if the same addition were made to a
completely ionized, unbuffered solution. All solutions have some buffering
capacity, but the presence of organic salts, salts of strong acids and weak
bases, or salts of weak acids and strong bases will increase buffering
capacity.
4-3
-------
4.1.2 Reagent Selection
Table 4.1.1 summarizes several of the more prevalent neutralization
reagents and their characteristics. The selection of the appropriate reagent
for wastewater neutralization processes is site specific and dependent on the
following considerations: wastewater characteristics, reagent costs and
availability, speed of reaction, buffering qualities, product solubility,
costs associated with reagent handling and residual quantities and
characteristics. Typically, the first step in reagent selections is to
characterize the wastewater. General parameters of interest include flow
(rate, quantity), pH, pollutant loading, physical form of waste, and
waste/reagent compatibility. These characteristics narrow the range of
reagents and treatment configurations available for consideration.
Following the selection of candidate reagents, the quantity of reagent
required to neutralize the waste to the desired end point must be determined.
Reagent quantity is usually calculated by developing a titration curve for
each candidate reagent using representative wastewater samples. Figure 4.1.1
illustrates a typical titration curve developed for the neutralization of a „
ferric chloride etchant using high calcium hydrated lime as an alkaline
4
reagent. These data determines toe quantity of reagent required to bring
the sample volume of wastewater to the desired pH.
The next step in the experimental procedure is the preparation of
reaction rate curves and development of kinetic rate equations for each
candidate reagent. Reagent reactivity is an important factor in determining
retention time and consequently the size of the treatment facility, the final
effluent quality, and the ease or difficulty of process system control. These
parameters, in turn, will affect both capital and operational costs associated
with the wastewater treatment system. Reaction rate curves for various
quantities of residual reagent (i.e., excess above stoichiometric
requirements) are determined by plotting pH as a function of time. Other
variables which should be monitored include temperature rise, agitator speed,
density, viscosity, color, sludge volume, and settleability.
4-4
-------
TABLE 4.1.1. ACID/ALKALINE NEUTRALIZATION AGENT CHARACTERIZATION
01
Miilociilflr Chnntc.it
H.-i|;.nit tnrmt\t on**
' Hir.lt '-tit iiim i:.tlU>| fulfill*
Itiwrrttom1 Carhonntr
I
i
: Mli)ii Male In* Vi(VH)i Celcliw
Hv.lr.il.vJ l.lnt! llydroalde
"i :.ilcU« Cal) Calcine)
'Mckliin' Oilde
i:irh..n COj Carbon
»i ••al.li! Olnilde
CfloaMM for* Hillk
Hnlecnler end coemmrclel denelty
KB 1 «hl itrenght '*/•
Mill. 1 r.nnlrr 20110-2100
9H C.iCII,
grenulei
M.I fow.lft 400-040
72-7*1 CeO
V6.I tebbli /70-llJO
9J-98J CaO
)4.0 flu Ilifulfled
under prenure
fnlubllltv
g/ 100 g
weter
O.OIII*2*
(• 2VC)
O.U)0
it )o*c)
Converted
to Ce(OH>]
~
Trpleel uie
Ac III M»ut ra-
il tat Ion
Acid Heirt ri-
ll i. clou
Ac lit Nput ri-
ll lit Ion
Alkali Hiutri-
Iliitloej
Advent agee
•elatlmly
lelatlvelf
leUtleel*
Ineapenalve
Nay bl
avillifcle
Cqulvilent ApprmiiMte Coil/inn
Ihlpecnt bulclty ' eon/ton kailclty*1 CnulvaU
DUedvantagee fora (actor* (I) (|) ""i«ht
Contain! lopurltlei lulk 0.419 6 12.27 Ml.)
alow reacting
Contilni lupurltlal lulk 0.710 46 M.79 11, m
le^ulcei dry itorage lulk 0.941 19 41.44 ZB.IIS
Tank cer — 100 — j;
1 from Hue gee
i l>..l™»ltlc Ca(nH)1 Nonuil
llv.lratrd LI** MgO Dolinltle
t
! iJnlcklUe Hegneilm
i
. S.nlj Alh Ne]CO] Sodlim
) U.iiiillc 3...I. Ne»ll Sodlm.
Hydroiida
njn.»
Converted
to Ce(oll)]
end Hg(OH),
(«')0'C)
(t lOO'C)
O.OOM10
(» lo'c)
Coeailete
Complete
Acid Hentre-
II lit Ion
1
AcM Neiitre-
II let Ion
Acid Hoot ri-
ll lit Ion
Acid Neutre-
II lit Ion
.
Acid Hcut ra-
il lit Ion
Alklll Neutre-
Alkel'l Hentra-
llaatlon
leletlvely
Ineapemlve
Irletlvely
ineiipenilve
reict Ive
eotuklt
Highly reac-
tive, tea*
handling
Highly
reactive
Highly
Ineapenllve
Nllhlr
reectlve
teei reactive then lagged 0.912 46 50.44 11.9
high calcium II*.
Leaa reactive then lulk I.IIO 39 )S. 14 2*9
high calcium Haw
Higher coltl then lulk 0.50? 11 163.11 il.O
caleliM reaglnte
High coil, requlree Tank cer O.tll 201 2tn.lt «0
heeted atorege
High Coit li|(«d 0.919 16» ill.ltt 20.11,
'
•omi cilclua Tank car — if — 44.05
More eepcniive then Tink car 66 11, s
eul Curie acid 74
* " 1--»""""""***^------'
-------
5
pH
4
I .
• TRIAL I
O TRIAL 2
A TRIAL 3
x TRIAL 4
0 TRIAL 5
B 10
MASS OP Co (0 H ) 2l 9)
12
Figure 4.1.1. Neutralization of Ferric chloride etcbant waste by
calciua hydroxide.
Source: Reference 4
4-6
-------
From the titration and reaction rate curves, kinetic rate equations can
be developed. One simplified method numerically approximates the relation
between the residual reagent (B) and time (t), as follows:
- V
Where r is the reaction rate, k is the rate constant and n is a constant
which expresses the order of the reaction (typically between 1 and 2). If the
differences (B^^ - B2> and (t, - t-) are small enough, adequate
accuracy is obtained provided that the general kinetic model assumed (i.e.,
r-kBn) fits well with the experimental data results. More difficult, but
also more accurate methods for determining the kinetic rate equation (e.g. ,
integral or statistical analysis) are discussed in the literature.
Similarly, the reader is referred to standard engineering texts for reactor
and costing methodologies based. on flow parameters and kinetic rate
6
equations.
In the final selection, the optimal reagents and reagent/waste feed ratio
will be those which incur the least overall cost, including not only the cost
of the reagent itself, but also the cost of purchasing and maintaining the
reagent and neutralization systems, and the costs associated with residual
handling. The combination of all such factors may make a slightly more
expensive reagent less expensive overall. For example, limestone is the least
expensive alkali reagent available on a unit cost basis, but the added
expenditure for complex grinding and feeding equipment combined with its slow
reactivity and insoluble sludge products, make it the least utilized.
4.1.3 Pretreatment Requirements
Pretreatment of corrosive wastes prior to neutralization typically
consists of gross solids removal (e.g., filtration), flow equalization, or
treatment of individual waste streams prior to combination with other process
wastes. These treatments of segregated wastes result in economic benefits
4-7
-------
from reduced reagent costs and smaller equipment sizing. Common pretreatment
processes include cyanide destruction, chromium reduction, metals
precipitation from highly chelated waste, and oil removal.
Cyanide wastes cannot be mixed with acid due to formation of toxic
hydrogen cyanide gas. Instead, cyanide is first oxidized to carbon dioxide
and nitrogen gas through alkaline chlorination. In two-stage chlorination, pH
is typically maintained around 11.0 in the first reaction vessel and 8.0 to
8.5 in the second vessel through addition of NaOH, as required.
Chromic acid wastes contain hexavalent chromium which must be reduced to
the trivalent form prior to precipitation. Reduction typically occurs at
pH 2.0 to 3.0 through addition of acid (e.g., sulfuric) and a reducing agent
(e.g., sulfur dioxide, ferrous sulfate, sodium meta-bisulfate-, and sodium
a
bisulfite). However, alkaline reduction (pH 7 to 10) using ferrous iron
has also been demonstrated. It has proven to be cost effective for highly
buffered alkaline waste and the treatment of mixed metal wastes containing
q
less than 10 mg/L of hexavalent chrome.
Chelated metal-containing wastes typically require pH adjustment into the
highly alkaline range in order to effectively precipitate metals. Commonly
used flocculants/coagulants include aluminum sulfate, aluminum chloride,
dithiocarbamate, sodium hydrosulfite, ferric chloride, ferrous sulfate, and
various polyelectrolytes and anionic polymers. Selection'of flocculant/
coagulant and use of sodium hydroxide versus lime is dependent on the types
and quantities of chelators in the waste. Common chelators include ammonia
and its derivatives, phosphates, EDTA, citric acid, amines and thiourea.
Removal of oil through emulsion breaking, dissolved air flotation
skimming or coalescing may also be performed prior to combination of oily
corrosives with other wastes. Traditionally, emulsified oils have been
treated at low pH (e.g., pH of 2.0) with alum. However, this form of
treatment is giving way to the use of more effective emulsion breaking
coagulants such as cationic polymers and other specialty chemicals.
While in some cases filtration or segregated waste pretreatment may be
utilized prior to neutralization, the most prevalent form of pretreatment is
flow equalization. It is generally used in facilities which experience a wide
variation in the flow or pollutant concentration of the wastewater.
Figure 4.1.2 illustrates a number of ways that flow equalization can be
achieved.
4-8
-------
(a)
Legend:
FT = flow transmitter
FC = flow controller
IT - level transmitter
(b)
Reactor
(0
(d)
Figure 4.1.2 Alternative concepts for wastewater equalization:
(a) batch reactor system, (b) batch equalization
continuous processing, (c) side-stream equalization,
and (d) flow-through equlization.
Source: Reference 5.
4-9
-------
Flow equalization through batch processing involves the storage and batch
- neutralization of the wastewater in the same vessel. The system is designed
with the reactor vessels arranged in parallel. When one vessel is full, the
flow is switched to another vessel while neutralization takes place in the
first. Equalization through continuous processing from batch storage involves
two storage tanks arranged in parallel, operating on a fill-and-draw cycle.
While one tank is being filled, the other is discharging its contents to the
neutralization system. In this manner, stream segregation may be achieved and
the neutralization facility receives a waste stream with uniform
characteristics.
In facilities that regularly dump concentrated solutions, sidestream
equalization is sometimes used. In this method, the concentrated waste or
flow surge is diverted to a storage tank prior to processing. This technique
is particularly useful in acid/alkali mixing or in cases where the solution
must be slowly bled into the wastewater system. Finally, flow-through
equalization attenuates sudden changes in the wastewater characteristics that
would adversely affect process control. This can be done through any device
that will provide flow resistance such as a sump, valve, weir, or surge tank.
In all methods of flow equalization care must be exercised during the
wastewater analysis to completely characterize any peak flows or
concentrations that might overload the system. In addition, flexibility in
system design should be provided for any future expansion, change in location,
or deviation in flow rates.
4.1.4 General Neutralization Processing Equipment
A wide variety of treatment options and configurations are available,
however, fully-engineered component neutralization systems generally consist
of the following equipment:
• Neutralization System
Tanks
Mixers j
- pH control instrumentation
4-10
-------
• Chemical Feed System
Tanks
- Mixers .
Level instrumentation
Metering equipment
• Miscellaneous
- Flow monitoring
Effluent pH recorder
Electrical and mechanical fit-up
Incremental engineering requirements
In addition, there is a need for facilities and equipment to collect and
segregate the wastewaters, transport the wastewaters to equalization sumps,
pump the wastewaters to the treatment system, perform liquid/solid separation,
and convey the treated wastewaters to the point of discharge.
4.1.5 Neutralization System
Neutralization tanks are fabricated from a wide range of construction
materials such as masonry, metal, plastic, or elastomers. Corrosion
«
resistance can be enhanced with coatings or liners which prevent the premature
«
decomposition of tank walls. For example, concrete reactors susceptible to
corrosion can be installed with a two-layer coating of a 6.3 mm base surface
(glass-reinforced epoxy polyamide) covered by a 1.0 mm coating of polyurethane
elastomer to extend service lifetimes.
Vessel geometries can be either cubical or cylindrical in nature with
agitation provided overhead in line with the vertical axis. While cubical
tanks need no baffling, cylindrical vessels are typically constructed with
suitable ribs to prevent swirling and maintain adequate contact between the
reactants. A general rule of thumb in the design of neutralization reactors
is that the depth of the liquid should be roughly equivalent to the tank
diameter or width.
Reactors can be arranged in either single- or multi-stage configurations
and operate in either batch or continuous mode. Multi-stage continuous
configurations are typically required to neutralize concentrated wastes with
variable feed rates. In these units most of the reagent is added in the first
vessel with only final pH adjustments (polishing) made in the remaining
4-11
-------
reaction vessels. This is particularly true when using'sluggish.reagents
which require extensive retention time. Single-stage continuous or batch
neutralization is suitable for most applications with highly buffered
solutions or dilute wastewaters not subject to rapid changes in flow rate or
pH.
A holdup period is required to provide time for the neutralization
reaction to go to completion. This factor is especially critical where a dry
feed (lime or limestone) or slurry is used as the control agent. In these
systems, the solids must dissolve before they react, increasing the required
holdup time and tank capacity. For example, liquid reagents used in
continuous flow operations generally require 3 to 5 minutes of retention time
in the first tank. Three minutes corresponds to the absolute minimum size
that will not cause considerable splashing or trapping of air. In comparison,
solid-based reagent systems such as lime or limestone typically require 30 and
45 minutes retention time, respectively.
Agitation serves the purpose of equalizing the hydrogen or hydroxide
concentration profile within the reaction vessel as the influent is dispersed
in the tank. Vessels with large stagnant areas provide little mixing between
•
reactants and causes large disturbances when concentrated materials are
released into the system. For accurate pH control, Hoyle has suggested that
agitator capacity should be measured as a ratio of the system dead time (the
interval between the addition of a reagent and the first observable pH change)
to the retention time (volume of the vessel divided by the flow through the
vessel). A ratio of dead time to retention time of 0.05 approaches an optimum
value.
The pH control systems for batch neutralization processes can be quite
simple with only on-off control provided via solenoid or air activated
valves. Control system designs for continuous flow neutralization systems are
more complicated because the wastewater feeds often fluctuate in both flow and
concentration. Systems currently available include; proportional, cascade,
feedforward, or feedback pH control. Each system has distinct advantages and
disadvantages which are discussed in detail in the literature.10'11'12'13
The pH control equipment usually consists of a pH probe, monitor, and
recorder. In addition, there is typically a control panel with an indicator,
starters and controls for metering pumps, all relays, high/low pU alarms,
switches, and mixer motor starters. .
4-12
-------
Chemical feed apparatus is similar to that of neutralization systems in
that they require storage tanks, agitation, level instrumentation, and
metering pumps. Storage tanks should be sized according to maximal feed rate,
shipping time required, and quantity of shipment. The total storage capacity
should be more than sufficient to guarantee a chemical supply while awaiting
delivery. Storage containers must be individually suitable to the reagent
being used. For example, hydroscopic reagents such as high calcium quicklime
or sulfuric acid must be stored in moisture-proof tanks to prevent atmospheric
degradation. Others like sodium hydroxide must be heated or carefully diluted
because they will freeze at temperatures slightly below room temperature or
when stored at concentrations greater than 40 percent. Agitation and feed
equipment specifications are particular to each reagent and are, therefore,
covered in detail in the following sections.
In addition to the chemical feed and neutralization systems, both flow
monitoring and effluent pH recording equipment are necessary to prevent
discharge of insufficiently treated waste resulting from surges or upsets.
Also, spare parts such as pH probes, pH controller circuit boards, metering
pump ball valves, o-rings, and strainers should be kept on hand to prevent any
excessive downtime.
4.1.6 Post-Treatment
Most treatment processes for corrosive wastes are typically preceeded by
neutralization or pH adjustment in order to enhance processing (e.g., metal
precipitation, emulsion breaking), prevent interference with downstream
treatment (e.g., biological treatment, carbon adsorption), or to minimize
corrosion of subsequent processing equipment. These processes are required
because the effluent from neutralization frequently fails to meet NPDES or
POTW discharge specifications as a result of the presence of toxic metals,
priority organics or high levels of oil, grease or solids. As a result,
physical, chemical, and biological treatment processes are commonly employed
to improve overall quality.
As discussed in Section 3.0, the vast majority of land disposed corrosive
wastes are aqueous liquids. These typically contain low concentrations of
organics and/or heavy metals which will frequently require post-treatment to
4-13
-------
meet discharge requirements. A smaller percentage of land disposed RCRA
corrosives consists of concentrated acid/alkali solutions and only 7.5 percent
consists of solids or sludges. However, these wastes cover the entire
spectrum in organic constituent concentrations and will be most effectively
handled by a correspondingly wide range in post-treatment processes.
Post-treatment options are, therefore, most conveniently discussed in
terms of the physical and chemical characteristics of the raw waste. Previous
14
investigators have categorized corrosive waste treatment options as
follows:
• Aqueous treatment of inorganic acids and bases which do not contain
toxic organics or heavy metals at levels which require treatment
(Waste C, Figure 4.1.3);
• Aqueous treatment of corrosive liquids with trace organics (Waste A,
Figure 4.1.3);
• Aqueous treatment of dilute organic corrosive wastes (Waste 0,
Figure 4.1.3);
• Treatment of heavy metal sludges (Waste B, Figure 4.1.3);
• Incineration of combustible sludges (Waste E), solids (Waste F), and
liquids (Waste G)-with optional neutralization (Figure 4.1.3); and
• Recovery of concentrated liquid organics (e.g., oils, solvents)
which contain acids or bases (Waste H) which is corrosive
(Figure 4.1.4).
Treatment trains for these waste categories are summarized below. This
is followed by a more in-depth discussion of clarification and sludge
consolidation, since these unit process operations will be applied in the
majority of corrosive waste neutralization treatment trains.
4.1.7 Treatment Trains
The generic process for neutralization of corrosives without metals or
organics (Process C) consists of two-stage neutralization followed by
solid/liquid separation (e.g., in a clarifier) to remove insoluble salts. The
sludge would then be dewatered (e.g., filter press) and disposed in a secure
4-14
-------
Ul
ASIC ACID/ADC All
tfltH HO OMUAMIOI,
HETALI OH CTAllll't
c.
COIIUSTIIILC
(LUDCCt AMI
HCAV> HCTAL
IHIDCtl HIT
MCANICI—'
cwiutTiiLC
CAIIIC ACIM All!
IASCI, mCANIC
IIQUIDI HITII
HIMUL ACIM
Figure 4.1.3. Treatment trains for corrosive wastes.
i
{Source: References 15 and 16.
-------
ucunii/
UWI
IMtVI
iion
rucTion
tci*
/ tiutn
\ WAIILUtld
\ (tHUUIOWl
tl OUT \ I
uuKi I n
JIOWI/ L
CMUIIIM
MCMIIH!
H
-1 miMTioi I *• Kuiunuriiw I — »J •imuATim
OIL
IUW
MU... |
I .11 »»"• .1 C«ICAL
| riOTOflON rwi[ | CMGVUIIM
IMTTCM
IIKUIIUD 1 *"* - jjjyjjt
lOUIIII
WATU
TO AQUIOV
nuctn i
--| ,»,«„«,.. |-*| . ££m \—
i
IHUT
itcwi
ui»rui
ON
riLTIAIC
Figure 4.1.4. Treatment of concentrated organica and
oily wastewater emulsions.
Source: References 15 and 16
-------
landfill. The clarified aqueous stream would be filtered as required to meet
NPDES or POTW discharge requirements. Filter backwash would be recycled to
the clarifier inlet.
Aqueous corrosives with trace organics (less than 500 ppm) would require
additional treatment of the clarified effluent. Applicable technologies for
trace organic removal include adsorption (e.g., activated carbon, resin), air
stripping, and ozone oxidation with UV radiation. Dilute organics (up to
10,000 ppm) are often more economically treated via biological-degradation as
indicated in Figure 4.1.3 (Waste Stream D). Neutralization is required
primarily to prevent a microorganism kill in the bioreactor. The aqueous
effluent concentration from biological treatment will contain organic levels
of approximately 10 to 50 ppm which can be filtered and polished (e.g.,
activated carbon) prior to discharge. Sludges would contain toxic organics
and require disposal such as incineration.
Metal-containing wastes must undergo precipitation at elevated pH (e.g.,
pH of 9.0) in the clarifier and the supernatant will require pH adjustment
prior to additional treatment or discharge. In addition, clarifier sludge
will require stabilization/encapsulation prior to disposal in a secure
landfill if the Toxicity-Characteristic Leaching Procedure indicates that the
waste exceeds maximum permissible metal concentrations in the leachate (see
Section 1.0, Table 1.1).
Concentrated organic corrosive wastes, typically containing organic
acids/bases, solvents or oils, will frequently be amenable to recovery
processes following neutralization and phase separation (Figure 4.1.4). If
emulsions are present, they can be broken through addition of a highly ionized
soluble salt of an acid followed by phase separation; e.g., dissolved air
16 '
flotation. Aqueous effluent will be treated as previously described for
wastes with low organic concentrations. '
The concentrated organic phase can be recovered usingj distillation, steam
stripping, solvent extraction or thin film evaporation, following solids
removal and neutralization. Alternatively, if recovery isinot judged to be
j
cost-effective, the organic waste can be destroyed via incineration or some
other form of oxidation; e.g., use as a fuel, chemical oxidation, wet air
oxidation, or supercritical fluid processes.
4-17
-------
Selection of an appropriate recovery or disposal technology for organic
residuals is dependent on the waste characteristics and volume, since these
factors will generally determine overall economic feasibility. Approximate
ranges of applicability of various treatment/disposal techniques as a function
of organic concentration are shown in Figure 4.1.5. Table 4.1.2 summarizes
applicable waste characteristics, development status of the technology,
performance capability and residuals generation for organic waste treatment
processes. Additional details on. these processes can be found in the
literature.
Neutralization of organic streams prior to incineration is optional,
provided the incineration combustion chambers are lined with an acid resistant
refractory or fire brick. In particular, combustible solids would
probably not be neutralized since this would require addition of water which
would reduce the waste's heating value. Incineration flue gas would require
scrubbing, followed by aqueous treatment of the resulting liquid waste
(Waste C).
Neutralization of concentrated organic corrosives would generally be
required prior to handling in other treatment or destruction processes. Lime
is .not recommended as a neutralization agent due to scale formation.
Similarly, sodium-based alkalis are not recommended when treating in thermal
processes, due to formation of eutectic solids which create ash and clinkering
problems. Instead, ammonium hydroxide has been recommended as a suitable
16
reagent.
4.1.8 Clarification and Sludge Consolidation
As discussed previously, clarification and ^ludge consolidation unit
t
operations will be applied to the majority of corrosive wastes which are
affected by the land disposal ban. •
Typically, wastewaters undergo chemical treatment and enter a clarifier
where the flow is decreased to a point at which solids with a specific gravity
1!
greater than that of the liquid settle to the bottom. For liquid/solid
mixtures with a slight density difference, an organic polymer (flocculant) can
be added to allow the solids to agglomerate and improve the settling
characteristics. The supernatant in the overflow is drawn off and residual
4-18
-------
Drying
Thin Film Evaporation
r- — — — — — — — — — | |
Fractional DUlllloflon
Chemical Oxidation
Steam Stripping
SO
Incineration
Solvent Extraction
"™ «••* T- -.1. m**m «• L m^ ^ m _
Air Stripping
Resin Adsorption i
Carbon
Adsorption . ' •
•H- •*• 1
Ozone/UV Radiation
I — — — •
UPEND
COMMERCIALLY APPLIED
POTENTIAL EXTENSION
^ 1
Wet Air Oxidation
1 ,
Supercritical Water
-I 1 1—' ' ' ' '
J 1
J 1 1—I I I I
0.01
0.05
0.5 . 1.0
INITIAL % OR3ANICS
10
BO
Figure 4.1.5. Approximate ranges of applicability of treatment techniques as a
function of organic concentration in liquid waste streams.
Source: Reference 15
100
-------
TABLE 4.1.2. SUMMARY OF ORGANIC RESIDUAL TREATMENT PROCESS
rrocee*
Inclnnratinn
Liquid injection
incineration
Rotary kiln
incineration
rluldliad bed
incineration
rl*ed/«tUipl*
hearth*
Applicable waata atr*aiaa
All pu-peble liquid*
provided waatea can be
blended to llu lava! at
8500 Itu/lb. SIMM aolU.
reattval awy be naceaeary
to avoid plugging noatlea.
Alt waataa provided Btu
level la ataintainad.
liquid* or nonbulky
aoltda.
Can handle • vide
variety of waatat.
Stage ot development
EettMtad that over 21*
unite are in uae. Moat
widely uaed Incineration
technology.
Over 40 unite la aervlcei
•on veraatil* (or waata
deetruction.
Nine unite rtportedljr
In operatton-elretiUtlni
bed unit* under
davalopMnt.
ApproiUitelr J0 unite
In uae. Old techno loiy
for Municipal waata
conbuition.
Nrfomance
Cicellent deatructlon
eflteleney (>9f.9«S).
•lee)dln( can avoid
probiaaa eaaocteted
with realduala, e.g., IICl.
Ixellent daatructlon
efficiency (>99.»9t).
Bicellent daatructlon
efficiency (=-99.991).
Performance eiey be
aurglflel for haiardoua
waataa, particularly
helogeneted organic uaatea.
Railduala generated
TSP, poaalblx ao»» PICa and
IICl. Only Minor aah if
aotlda removed In pratreatnent
proceaaea. Scrubber water
•HI at b« neutral lied.
Raquiraa APCDa. Reaiduala
ehould be accept able if
charged properly and
neutralised.
Aa above.
Aa ebove.
Uae Aa A rue I
Induttrial kllna
High temperature
inOuatrial boiUra
Generally all waatai, but
Itu level, chlorine content,
and other (aipurity content
•ay require blending to
control charge characterise lea
and product quality.
Alt Durable flulda, but
' ahould blend HCl and
halogeneted organica. Solida
reewval particularly important
to enaure atable burner
operation.
Only a few unite now
burning hasardou* waete.
Several unit* In uae for
hatardoua waatea.
Uaually excellent
deatruction efficiency
(-99.99X) becauae of
long realdence tiawe and
high te«paratur«a.
Moat unite teated hive
deennatreted high ORE
<-99.99t).
unlaae bollera equipped
Requirei APCOa.
Realduala ahould be
acceptable.
Waete* aniat be blended
to met eniaalon
etamlarda for TSP and IICl
with APCDa.
(continued)
-------
TABLE 4.1.2 (continued)
Prooeaa
Applicable wait* stream*
Stage of development
Performance
Residual* generated
Physical Treatment Method*
Diitillation
ro
Evaporation
Steam Stripping
Air Stripping
Liquid-Liquid
Kit faction
Carbon Absorption
Heain Adsorption
Thla la • procea* uaed to
recover and separate concen-
trated organic*. Fractional
distillation will require
sol Ida reawval to avoid
plugging column*.
Agitated thin film unit a
can tolerate higher levela
of aolida and higher
viacoaltle* than other
typea of atilla.
A alaiple dlatltlation
proceaa to remove volatile
organlca froaj aqueous eolu-
ttona, Preferred for low
concentratlona and organlce
with low aolubilltlea.
Generally uaed to treat
low concentration aqueous
at ream*.
Generally aultable only for
liquids of low aolld content.
Suitable for low folld,
low concentration
aqueoua waate streams.
Suitable for low aolid
waste etreana. Consider
for recovery of valuable
organice.
Technology well developed
and equipment available
from many suppliers;
widely practiced technology.
Technology la well developed
and equipment la
available from several
suppliers; widely
practiced technology.
Technology well developed
and available.
Technology well developed
and available.
Technology well developed
for Industrial proceaalng.
Technology well developed;
ueed aa polishing treatment.
Technology well developed
in Industry for apeclat
reiin/aolvcnt combination*.
Applicability to waate
streams not demonatrated.
Separation depends upon
reflux (99* percent
achievable). Thla la
a recovery process.
This la an organic recovery
proceaa. Typical recovery
of 60 to 70 percent
depending on initial
volatile content.
Not generally conaidered
a final treatmeat, but
can achieve low reaidual
organic levela.
Hot generally considered
• final treatment, but
may be effective for
highly volatile waatea.
Can achieve high efficiency
aeparatlona for certain
waate combinations.
Can achieve low levela of
reaidual organic* in I
effluent.
Can achieve low level* of
reaidual organica in
effluent.
Bottom* will uaually contain
level* of organic in exces*
of 1,000 pp«; condeniate
may require further treatment.
Bottom* will contain
appreciable organic*.
Generally suitable
for incineration.
Aqueous treated at ream
will probably require
poliahing. Further
concentration of over-
head steam generally
required.
Air emlaslon* may
require treatment.
Organic or metal solubility
in aqueoua phaie should be
monitored.
Adsorbate nuat be
proceaaed during
regeneration. Spent
carbon and waatewater
may also need treatment.
Adaorbate matt be
proceased during
regeneration.
(continued)
-------
TABLE 4.1.2 (continued)
ri
ro
Proceaa
Other Ttwraal Tec lino 1 Of ie I
Circulating bed
combiietur
Holten |Un
incineration
Molton tilt
dot ruction
Furnace pyrolyila
unit!
.
Plaima ire
pyrulyila
Fluid wall
advanced
ulvctrlc
reactor
In altu
vitrification
Applicable waate at ream*
Llquide or nonbulky
lolldi.
Almoat ill waatai, provided
moletura and ••tit impurity
lavali art within
llmitetiona.
Not aultable for high
( 20X) a ah content
waatea.
Hoat deelgna aultablt
(or all waitae.
Praaint deelgn aultabla
only (or llqulda.
Suitable (or all waataa
if aolidt pr«tr«atad to
eniure free (low.
Technique for traatini
contiainatad aolla, could
poaaibly ba «>t*nd«d to
alurriaa. Alao uaa aa
•olldif Icatlon proccaa.
Stag* of davatopawnt
Only ana U.S. nanufac-
turar. Ho unita treating
haaariloua waata.
Technology developed
for glaa* laanufacturlng
Not available yat aa a
hatardoua waate unit.
Technology uodar davalop-
aMnt alnce 1969, but
further development on'
hold.
One pyrolyalc unit RCRA
peraitted. Certain
deaigna available
comurclally.
Coanarcial dealgn appeara :
imlnent, with future
•odlf icationa planned
for treatment of aludgee
and aollda.
Ready for conMercial
development. Teat unit
permitted under RCRA.
Not commercial, further
work planned.
Par (OHM nee
Manufacturer report a
high efdcienciea
<»9*.m>.
Ho performance data
evelleble. but DREe
ahould be high
<>«».»«).
Very nigh deatruction
efficleneiea for
orginiei ( ill nlnee
(or PCIa).
Very high deatruction
efflclenciaa poiaible
(>99.»9X). Poialbllity
of PIC formation.
Efficiencies eceeeded
all nlnea in teata with
eolvente.
Ifficlenciee have
exceeded al« nlnea.
No data available, but
DREa of over alx ninea
reported. .
Realduela generated
Bed material eddltlvea
can rmluce IIC1 emlialoni.
Reaiduala aliould be
acceptable if neutrittied.
Will need APC device for HC1
and poaaibly PICa; aotida
retained (encapsulated) in
molten gleaa.
Naeda aoew APC devlcea
to collect material not
retained in aelt. Aah
dlapoaal may be e
problem.
TUP emlaalona lower than thuaa
fron conventional coaibuatlon
will need APC device! for HC1.
Certain waatee may produce en
unacceptable tarry reaiduat.
Require! APC devicea for
IICP and TSP, needa flare
fur II2 and CO
deetruction.
Requlrea APC devicea for
TSP and IIC1; Chlorine
renoval may be required.
Off gaa ayaten needed
to control emiaalona
to air. A>h contained
in vitrified aoil.
(continued)
-------
TABLE 4.1.2 (continued)
Croc
Applicable watte streams
Stage of development
Performance
Reaidiiali generated
Chumical Treatment Proceuea
l!.
to
U.99.99Z) for
all organic conatltuenta.
Not likely to achieve
reildual organic levels in
the low ppn range for
•oat waatei.
Not likely to achieve
reaiduat levela in the low
ppn range for noit wastes.
Some reiiduel likely which
need further treatment.
Roslduali not likely to
be a problem. Neutralization
can be accomplished in
proceia.
Residual contamination
likely; will require
additional proceaiing of
off gaaei.
Heiidual contamination
likely; will require
additiomil proceaiing.
Biological Treatment Method!
Aerobic technology lultable
for dilute waitei although
some constituent! will be
resistant.
Conventional treatments
have been uaed fur(yeara.
Hay be uied aa final
treatment for specific
waitei, may be pretreat-
nent for reiiitunt species.
Source: Referance
Residual contamination
likely; will usually
require additional
proceaiing luch ai
absorpt ion.
-------
solids are removed in a final polishing step such as carbon filtration or ion
exchange. The solids in the underflow can then be discharged to a holding
tank for subsequent dewatering.
Figure 4.1.6 shows investment costs for flocculation/clarification units
as a function of flow rate. The unit is assumed to have a separate
flocculation tank, a polymer feed system, a slant-tube separator, and a zone
in which sludge collects prior to discharge.
The sludge generation and solids settling rates are usually determined by
laboratory analyses conducted by equipment vendors. Table 4.1.3 is an example
of a laboratory analysis of the neutralization characteristics of 3 alkali
18
reagents applied to a 1.75 pH spent acid plating waste. While sodium
hydroxide yielded the lowest sludge volume (due to the extreme solubility of
sodium salts) it was almost seven times as expensive ($15,170/million gallons)
to utilize as hydrated lime ($2,200/million gallons). A limestone/lime dual
alkali reagent system was ultimately selected for this application since it
yielded the lowest effective sludge volume (5,480 mg/L) at the least overall
cost ($3,170/million gallons). Since the application was a one-time
neutralization of a contaminated surface impoundment, the reagent selection
criteria were*limited to reagent cost, sludge-dewatering characteristics and
*
volume requiring disposal and the added cost of grinding and slurrying
equipment.
In addition to different degrees of quantity sludge generation, each
reagent imparts to the sludge variable settling characteristics, thereby
affecting the sizing parameters of downstream equipment. For example, lime
neutralized sludge exhibits a granular nature that settles fairly rapidly and
2
dewaters effectively (4 to 20 Ib of dry solids/hr/ft yielding a
3/16 to 3/8 in. cake). Conversely, sodium hydroxide sludge results in a
19
fluffy gelatinous precipitate with low settling rates. Figure 4.1.7 shows
the results of three settling tests conducted on power plant effluents with
both lime and sodium hydroxide. In all three cases sodium hydroxide settled
more slowly and in subsequent filtration tests, dewatered about half as
Q
effectively. However, the use of lime or limestone generates greater
sludge weight and volume. This is primarily due to insoluble acid salts and
calcium sulfates formed when neutralizating sulfate containing wastes such as
4-24
-------
so
z
1
in
- 20
20
Tout installed east
40 SO 80
FLOW RATE I gal/mm I
Notts:
Instilled cost • 1.25 x n»rOw»re cost.
Con incudes slatt -ryot eiiriflcr with flocculating
cnamoer ind ceivmer teea system.
100
120
'Figure 4.1.6 Investment cost for flocculation/clarification units;
Source: Reference 17
4-25
-------
TABLE 4.1.3 LABORATORY ANALYSIS OF THE NEUTRALIZATION
OF SPENT ACID PLATING WASTE
Pretreatment
Raw waste
10 g/L CaC03
10 g/L CaC03
Treatment
None
10 g/L CaC03
8.0 g/L Ca(OH)2
6.5 g/L NaOH
3 g/L Ca(OH)2
2 g/L NaOH
Final
PH
(S.U.)
1.75
3.0
9.5
11.5
10.5
10.3
Supernatant
TSS
(mg/L)
41
1400
162
-
130
90
Zn
(mg/L)
191
72
0.15
0.25
0.20
0.03
Pb
(mg/L)
0.4
0.3
0.1
0.1
0.1
0.1
Sludge3
TSS
(mg/L)
-
3100
6200
-
5480
5700
Volume
Z
-
10
27
-
25
40
*Sludge volume measured after 15 hours.
Source: Reference 18
4-26
-------
mt •*» mltHW tao"
•flvtOmM*!
10 n
so
Figure 4.1.7 Settling rate curves.
Source: Reference 19
4-27
-------
sulfuric acid. Therefore, as landfill and hauling costs become more
significant, sodium hydroxide becomes more competitive with lime and limestone
as a neutralization agent.
Few, if any, sludges settle at a rate sufficient to utilize only
clarifiers or thickeners to accumulate sludge for disposal on land.
Therefore, the underflow from the clarifier is typically concentrated through
the use of mechanical dewatering equipment such as centrifuges, rotary vacuum
filters, belt filters, drying ovens, and recessed-plate filter presses. The
obtainable degree of cake dryness can be determined by bench-scale tests by
the equipment vendor to identify the suitability of a particular dewatering
device (see Table 4.1.4). The low solids content of sodium hydroxide after
20
sedimentation (3 to 10 percent) requires the use of a filter press.
Conversely, supspended solids removal from lime neutralized sludges can be
accomplished through use of a wider range of equipment including rotary vacuum
or continuous belt filters. Figures 4.1.8 and 4.1.9 present the unit costs
for both sludge storage/thickening units and recessed plate filter presses,
respectively. Items that were not included in these figures, but will add to
the cost of installation include; high pressure feed pumps, filtrate return
lines (to clarifier) and cake solids handling equipment. The feed volume
capacity of the unit is based on a feed solids concentration of 2 percent, a
cake solids concentration'of 20 percent and a press cycle of 8 hours.
4.1.9 Land Disposal of Residuals
Installation of wastewater treatment systems inevitably results in the
problem of sludge disposal. The cost of hauling the sludge to a licensed
hazardous waste landfill will depend on the volume of sludge, the distance
hauled, and the aludge composition. Figure 4.1.10 illustrates the annual
disposal costs for 100 Ibs of sludge generated over a range of sludge
concentrations and unit disposal costs. Sometimes it is possible to
dispose of calcium based reagent sludges through agricultural or acid pond
lining. In one neutralization application, over 200,000 Ibs/acre of lime
neutralized waste pickle liquor sludge was applied onto Miami silt loam to
21
improve overall crop yields.
4-28
-------
TABLE 4.1.4. SUMMARY OF SLUDGE DEWATERING DEVICE CHARACTERISTICS
*".
NJ
Pnr.-i motor
('ike Solids
Z
<)|ior,it ioiml
Vnrinbles
Ailvmit.igRs
Dii.iilvantages
Gravity
(low pri'SBtirc)
16 - 24
-Rate of sludge
feed
-Polymer
concentrat ton
-Belt speed
-Depth of sludge
in cylinder
-Low 1'iiorjjy &
ciiplt.il cunt
-Low apace
requirements
-Requires little
operator skill
-Limited capacity
-Low sol Ida
concentration
-Requires large
quantity of
conditioning
chemicals
Basket
cent rl fiif.o
20 - 30
-Bowl apucil
-Time at full
speed
-Depth of
skimming
-Sludge fend
rate
-Smim machine
for thickening
& dowatering
-Very flexible
-Little operator
attention
-Unit in not
continuous
-High ratio of
capital cost to
capacity
-Requires
complex controls
-Requires noise
control
Solid Bowl
crnt r 1 fugu
30 - 42
-Bowl/conveyor
differential
• peed
-Tool depth
-Sludge
feed rate
-Easy to install
-Low space
requirement
-Either
thickening or
dewntering
-High rate of
fend
-Can operate on
highly variable
feeds
-Requires
prescreening
-Very noisy
with high
vibrat Ion
-High power
connumpt ion
-Requires high
maintenance
skills
Vacuua
filter
30 - 40
-Quantity
of II20
-Drum speed
-Vacuum level
-Conditioning
chemicals
-Filter media
-Continuous
operation
-Long media life
-Low maintenance
-Easy operation
-High power
requirement
-Vacuum pumps
are noisy
-Requires at
least 31 feed
solids for
operation
Belt filter
press
36 - 46
-Belt speed
-Belt tension
-Washwater
flow and
pressure
-Belt type
-Polymer
conditioner
-Only filter
press produces
drier cake
-Low power
-Low noise &
vibration
-Continuous
operation
-Very sensitive
to Incoming feed
-Short media life
-Greater
operational
attention and
polymer dosage
Recessed
filter
press
50 - 60
-Feed pressure
-Filtration time
-Use of precoat
-Cloth washing
frequency
-Filter cloth used
-High aolids filter
cake
-High solids capture
-Only mechanical device
capable of meeting
some landfill
requirements
-High capital
cost
-Batch discharge
-High polymer
usage
-Media replace-
ment costs are
high
Reference 20
-------
30 r-
Touii
B*Md on carbon suet construction.
Costs inchKM filMr-r«inforcM ootv«t«r
Mot «oa aupftctgm pump.
1000
2000 3000
VOLUME le*ll
SOW
m
Figure 4.1.8 Investment cost for sludge storage/thickening units.
I300
2
i
200
gwo
-iSOO
375
o
Uf
B:
cl ir*i^M* ooty*
propyMnt PIMM and fiiwr ctom*.
tzs
CM* vdunw MMO on 1 V tniek smog*
Fs*dvokim*c
sofios - 2%: eaasoMs - 20%: crctt
10
20 X «
EQUPieiT COST 81.000)
(Figure 4.1.9 Hardware cost for recessed plate filter presses.
Source: Reference 17
4-30
-------
1000 r-
^ 100
W
5
i'°
I I I I T I I I I
0.1
1.0 10
CONCENTRATION SOLIDS IN SLUDGE (wt%)
S2.00/gal sludge
dtsoosal cost
Sl.OO/gai sludge
disposal cost
SO.50-gal sludge
disposal cost
100
Figure 4.1.10 Annual cost for disposal of industrial sludge
(per 100 Ib dry solids generated per day).
Source: Reference 17
4-31
-------
In Che absence of Che possibility of land disposal of waste in a location
that will be demonstrably supportive of human health and the environment,
another option is to treat the waste Co immobilize the waste constituents for
as long as they remain hazardous. This method of treatment, based on fixation
or encapsulation processes, is a possibility for some corrosive wastes;
however, it is more likely a treatment that will be undertaken to ensure that
residuals from other treatment processes can be safely disposed. Certain of
Chese residuals could be found hazardous for reasons other than pU
characteristics; e.g., their heavy metal content may lead to positive tests
for EP toxicity. In such cases, encapsulation may be needed to eliminate this
characteristic.
The following discussions will summarize available information concerning
immobilization techniques, namely solidification/fixation or encapsulation.
Chemical fixation involves the chemical interaction of the waste with a
binder; encapsulation is a process in which the waste is physically entrapped
within a stable, solid matrix.
Solidification can be used to chemically fix or structurally isolate
metallic species or trace organics which may be present in neutralization
sludges to a solid, crystalline, or polymeric matrix. The resultant
monolithic solid mass can then be safely handled, transported, and disposed of
using established methods of landfilling or burial. Solidification
technologies are usually categorized on the basis of the principal binding
media, and include such additives as: cement-based compounds, lime-ba.sed
22
pozzolanic materials, thermoplasts, and organic polymers (thermosets).
The resulting stable matrix produces a material that contains the waste Ln a
nonleachable form, is nondegradable, cost-effective, and does not render the
land it is disposed in unusable for other purposes. A brief summary of the
compatibility and cost data for selected waste solidification/stabilization
systems is presented in Tables 4.1.5 and 4.1.6.
CEMENT BASED SYSTEMS
These systems utilize type I Portland cement, water, proprietary
additives, possibly fly ash, and waste sludges to form a monolithic, rock-like
23
ma'ss. In an EPA publication, several vendors of cement based systems
4-32
-------
TABLE 4.1.5
COMPATIBILITY OP SELECTED WASTE CATEGORIES WITH DIFFERENT WASTE
SOLIDIFICATION/STABILIZATION TECHNIQUES
u>
to
TrenlHenl Type
Uaate
component
Cement
baaed
l.lm*
baaed
Thermoplaalle
aolUlf leal Ion
Orgenlc
polymer
<«!•)•
Surface
encepaulatlon
• Self-
cement Ing
leclmli)u*e
Glaaalflcallon end
eynlhellc mineral
formal Ion
Orgenlcn
1.
2.
Organ Ic
aolvenla and
olll
Solid organ-
Ice (e.g..
plaillce.
realm, tin)
Hay imped*
aattlni, my
«acap« it
vapor
Good—of Ian
Increeaea
durability
Many Impede eet-
lln|, aiay aacapa
aa vapor
Good— often
Incrtaaaa
durability
Or|anlca aay
vapor It* on
dealing
foaalble uie ae
binding agenl
Hey retard eet
of polymara
Hey retard eet
of |iolyiaere
Huat flrat.be
ebeorbcd on
aolld aulrlx
Compatible— many
encepaulatlon
etalerlele ere
pleetlc
Fire danger
.on healing
Fire danger
on heating
Uaalta deconpoae at
high leuperalurea
Waatea decompoae at
high lempereluree
Inorjanlcai
1.
2.
).
I.
S.
6.
Acid veetee
Onldliert
Sullatea
llalldn
Heavy me t ale
Radioactive
•at cr lal a
CcMltt will
neutrellte
• Cldl
Compatible
«
Hay ratard aet-
llng and
cauaa apallln|
unlaaa apiclal
cement It uaid
taally leached
fro* cement,
•ay raiard
•citing
Cooipa tibia
Compatible
Compatible
Coaipatlbla
Compatible
Hay retard set,
•oat are
eaally leeched
Compatible
Compatible
Can be> neutral-
lied before
Incorporation
Hey cauaa
Mtrli break
down, fire
Hey dehydrate
and rehydratc
cauelng
epllttlng
Hey dehydrate
Compatible
Compatible
Compatible
Hay cauae
•elrlx break
down
Compatible
Compatible
Acid pll eolu-
bllliee Betel
hydr«xldee
Compatible
Cen be neutrel-
Ited before
Incorporation
Hey cauae
deterioration
of encepeu I el-
Ing aiaterlele
Compatible
Compatible
Compatible
Compatible
Hay be neu-
tral lied to
form eul-
fele aelte
Compatible If
eulfetei
ere preeent
Compatible
Compatible If
eiilfelee
ere el eo '
preeent
Compatible If
eulfetca
ere preaent
Compatible If
eulfetee
ere preeent
Cen be neutral lied
and Incorporated
High tempereturee
•ay cauae unde-
eble reectlone
Compatible In many •
ceeee
Compatible In Many
caaee
Coapetlble In Many
caeee
Compatible
Source: Reference 23
-------
TABLE 4.1.6 PRESENT AND PROJECTED ECONOMIC CONSIDERATIONS FOR WASTE SOLIDIFICATION/
STABILIZATION SYSTEMS
Type of treatment
system
CeMent-based
• Posiolanlc
Thtraoplaitlc
(blluiien-based)
~ Organic polyner
1 (polyester systeei)
U>
Surface encapsulation
(polyethylene)
S*tf-ce«entlng
Classlf (cat lon/silneral
synthesis
Aeioiint of *u- Cost of m»-
Itnlt terlal required lerlal required
Major cost of to 1 rent 100 Ibs tu treat 100 Ibs Equipment Energy
•starlets required siatbrliil of raw waelu of rau wants Trends In price costs uuu
Portland Ceaer.t $0.01/lb
LlM Fly ash JO.OJ/lb
BttuMn JO.OS/lb
DruM $27/drti«
Polysster $0.4S/lb
Catalyst }l.ll/lb
Druais $l7/dru«
Polyethylene) Varies
Cypsusi (fro* waste) **
Feldspar fO.Ol/lb
100 Ib
100 Ib
100 Ib
0.0 drini
4) Ib of
polyester-
catalyet •!*
Varies
10 Ib
Varlea
$ 1.00 Rctilile |.ou Low
9 3.00 Stnhlu Low Low
$18.60 Keyed to oil Very high High
prices
$27.70 Keyed to oil Very high High
prices
$ 4. SO* Keyed to oil Very high High
pricus
** Stable Moderate Moderate
-- Stable High Very high
* Based on the full cost of |9l/ton.
*• Negligible but energy cost for calcining are appreciable.
Source) Rtftrinc* 23
-------
reported problems with organic wastes containing oils, solvents, and greases
not miscible with an aqueous phase. For although the unreactive organic
wastes become encased in the solids matrix, their presence can retard setting,
cause swelling, and reduce final strength. These systems are most
commonly used to treat inorganic wastes such as incinerator generated wastes
and heavy metal sludges from neutralization/precipitation processes.
LIME BASED (POZZOLANIC) TECHNIQUES
Pozzolanic concrete is the reaction product of fine-grained aluminous
siliceous (pozzolanic) material, calcium (lime), and water. The pozzolanic
materials are wastes themselves and typically consist of fly ash, ground blast
furnace slag, and cement kiln dust. The cementicious product is a bulky and
heavy solid waste used primarily in inorganic waste treatment such as the
25
solidification of heavy metal and flue gas desulfurization sludges.
THERMOPLASTIC MATERIAL
In a thermoplastic stabilization process, the waste is dried, heated
(260-450°F), and dispersed throug-h a heated plastic matrix. Principal binding
media include asphalt, bitumen, polypropylene, polyethylene, or sulfur. The
resultant matrix is somewhat resistant to leaching and biodegradation, however
the rates of loss to aqueous contacting fluids are significantly lower than
those of cement or lime based systems. However this process is not suited to
wastes that act as solvents for the thermoplastic material. Also there is a
risk of fire or secondary air pollution with wastes that thermally decompose
22
at high temperature.
ORGANIC POLYMERS (THERMOSETS)
Thermosets are polymeric materials that crosslink to form an insoluble
mass as a result of chemical reaction between reagents, with catalysts
sometimes used to initiate reaction. Waste constituents could conceivably
enter into the reaction, but most likely will be merely physically entrapped,
within the crosslinked matrix. The crosslinked polymer or thermoset will not
4-35
-------
soften when heated after undergoing the initial set. Principal binding agents
or reactants for stabilization include ureas, phenolics, epoxides, and
polyesters. Although the thermosetting polymer process has been used most
frequently in the radioactive waste management industry, there are
formulations that may be applicable to certain neutralization sludges. It is
important to note that the concept of thermoset stabilization, like
thermoplastic stabilization, does not require that chemical reaction take
place during the solidification process. The waste materials are physically
trapped in an organic resin matrix that, like thermoplastics, may biodegrade
over extended periods of time and release much of the waste as a
leachate.26 It is also an organic material that will thermally decompose if
exposed to a fire.
Encapsulation is often used to describe any stabilization process in
which the waste particles are enclosed in a coating or jacket of inert
material. A number of systems are currently available utilizing
polybutadiene, inorganic polymers (potassium silicates), portland concrete,
polyethylene, and other resins as macroencapsulation agents for wastes that
have or have not been subjected to prior stabilization processes. Several
different encapsulation schemes have been described in References^,28. xhe
resulting products are generally strong encapsulated solids, quite resistant
to chemical and mechanical stress, and to reaction with water. Wastes
successfully treated by these methods and their costs are summarized in
Tables 4.1.7 and 4.1.8. The technologies could be considered for stabilizing
neutralization sludges but are dependent on the compatibility of the
neutralized waste and the encapsulating material. EPA is now in the process
of developing criteria which stabilized/solidified wastes must meet in order
to make them acceptable for land disposal.29
4-36
-------
TABLE 4.1.7. ENCAPSULATED WASTE EVALUATED AT THE U.S. ARMY WATERWAYS
EXPERIMENT STATION
Code Me.
100
200
300
noo
500
500
700
800
900
1000
Source of Waste
SO scrubber sludge, lime process, eastern
*coal
Electroplating sludge
Nickel - cadmium battery production sludge
SO scrubber sludge, limestone process
eastern coal
SO scrubber sludge, double alkali process
eastern coal
SO scrubber sludge, limestone process,
western coal
Pigment uroduction sludge
Chlorine production brine sludge
Calcium fluoride sludge
SO scrubber sludge, double alkali process,
western coal
Pajor Contaminants
Ca, SOj,"/SC3"
Cu, Cr, Zn
Nl, Cd
Cu, S0j,"/S03"
Na, Ca, S0j,"/S03*
Ca, SOjf/SC^"
Cr, re, CN
Ma, Q~, Hg
Ca, F~
Cu, Na, SO^'/SO-j*
Source: Reference 28
TABLE 4.1.8. ESTIMATED COSTS OF ENCAPSULATION
Process Cation Estimated Cost
Resin Fusion:
Uneonflned waste SllO/dry ton
55-Gallon drums *O.H5/gal
Pesin spray-on Not determined
Plastic Welding $253/ton - s63-lO/drum
* (80,000 55-gal druns/year)
Source: Reference 28
4-37
-------
REFERENCES
1. Lewis, C.J., and R.S. Boynton. Acid Neutralization with Lime for
Environmental Control and Manufacturing Processes. National Lime
Association, Bulletin No. 216. 1976.
2. Kirk-Othmer Encyclopedia of Chemical Technology. Vol. 14, 3rd, Edition.
John Wiley & Sons, New York, NT. 1981.
3. Chemical Marketing Reporter. Week ending July 18, 1986.
4. Oberkrom, S.L., and T.R, Marrero. Detoxification Process for a Ferric
Chloride Etching Waste. Hazardous Waste and Hazardous Materials, Vol. 2,
Ho. 1. 1985.
5. MITRE Corporation. Manual of Practice for Wastewater Neutralization and
Precipitation. EPA-600/2-81-148. August 1981.
6. Levenspiel, 0. Chemical Reaction Engineering. 2nd Edition, John Wiley &
Sons, New York, NY. 1972.
7. GCA. Case Studies of Existing Treatment Applied to Hazardous Waste
Banned from Landfill, Phase 11 Case Study for Facility D. July 1986.
8. Kiang, Y.H., and A.A. Metry. Hazardous Waste Processing Technology.
Ann Arbor Science Publishers, Inc. 1982.
9. Biggins, T.E., and B.R. Marshall. CH2MHILL, Reston, VA. Combined
Treatment of flexavalent Chromium with Other Heavy Metals at Alkaline pH.
Presented in Toxic and Hazardous Wastes: Proceeding of the 17th
Mid-Atlantic Industrial Waste Conference. Lehigb University.
10. Boyle, D.L. Designing for pH Control. Chemical Engineering, November 8,
1976.
11. Cushnie, G.C. Removal of Metals from Wastewater: Neutralization and
Precipitation. Pollution Technology Review No. 107. Noyes Publication,
Park Ridge, NJ. 1984.
12. Hoffman, F. How to Select a pH Control System for Neutralizing Waste
Acids. Chemical Engineering. October 30, 1972.
13. Jungek, P.R., and E.T. Woytowicz. Practical pH Control. Industrial
Water Engineering. February/March 1972.
14. Camp, Dresser, and McKee. Technical Assessment of Treatment Alternatives
for Wastes Containing Corrosives. Contract No* 68-01-6403. September
1984.
15. GCA. Technical Resource Document: Treatment Technologies for Solvent
Containing Wastes. Contract No. 68-03-3243. August'1986.
.4-38
-------
16. Warner, P.H., et al. Treatment Technologies for Corrosive Hazardous
Wastes. Journal of the Air Pollution Control Federation. April 1986.
17. U.S. EPA. Reducing Water Pollution Control Costs in the Electroplating
Industry. EPA-625/5-85-016. September 1985.
18. Hale, F.D., et al. Spent Acid and Plating Waste Surface Impoundment
Closure. Management of Uncontrolled Hazardous Waste Site. October 31 -
November 2, Washington, D.C. 1983.
19. Mace, G.R., and D. Casaburi. Lime vs. Caustic for Neutralizing Power.
Chemical Engineering Progress. August 1977.
20. U.S. EPA. Dewatering Municipal Wastewater Sludges. EPA-625/1-82-014,
October 1982.
21. Berger. Land Application of Neutralized Spent Pickle Liquor. 17th
Industrial Waste Conference, Purdue University. 1962.
22. GCA Technical Resource Document: Treatment Technologies for Dioxin-
Containing Wastes. Contract No. 68-03-3243. August 1986.
23. Guide to the Disposal of Chemically Stabilized and Solidified Waste,
EPA SW-872. September 1980.
24. Environmental Laboratory U.S. Army Engineer Waterways Experiment Station,
Survey of Solidification/Stabilization Technology for Hazardous
Industrial Wastes, EPA-600/2-79-056.
25. McNeese, J.A., Dawson, G.W., and Christensen, D.C., Laboratory Studies of
Fixation of Kepone Contaminated Sediments, in "Toxic and Hazardous Waste
Disposal", Vol. 2 Pojasek, R.B. Ed., Ann Arbor Science, Ann Arbor,
Michigan. 1979.
26. Stabilizing'Organic Wastes: How Predictable are the Results? Hazardous
Waste Consultant. May 1985 pg. 18.
27. Thompson, D.W. and Malone P.G., Jones, L.W., Survey of Available
Stabilization Technology in Toxic and Hazardous Waste Disposal, Vol. 1,
Pojasek, R.B., Ed. Ann Arbor Science, Ann Arbor Michigan. 1979.
28. Lubowitz, H.R. "Management of Hazardous Waste by Unique Encapsulation
Processes." Proceedings of the Seventh Annual Research Symposium.
EPA-600/9-81-002b, March 1981.
29. C. Wiles. Hazardous Waste Engineering Research Laboratory, U.S. EPA,
private communication; and Critical Characteristics and Properties of
Hazardous Waste Solidification/Stabilization, HWERL, U.S. EPA, Contract
No. 68-03-3186 (in publication).
4-39
-------
4.2 MIXING OF ACID AND ALKALI WASTES
4.2.1 Process Description
The process of acid/alkali mixing (mutual neutralization) may be the most
economical method of neutralization available, particularly in cases where
compatible acid/alkaline wastes are present in the same plant. Prior to
implementation, data are typically collected on the volume and concentration
of each waste stream and their respective flow patterns (batch or
continuous). In addition, waste stream mixing characteristics are usually
investigated in order to pre-determine possible waste incompatibilities that
would prevent or limit the use of the technology. For example, the
precipitation of metal hydroxides or other insoluble species (e.g., calcium
sulfate) may result in increased sludge generation or plugging of the
transport or dewatering equipment. If the sludge generation is considerable,
the increased dewatering, disposal, and maintenance costs could possibly
outweigh the benefit of any savings realized on reagent costs. Also, if
incompatible wastes produce a reaction that is too sluggish or difficult to
control', or generate reaction products that are toxic (i.e., hydrogen cyanide)
or highly exotbemmic, then implementation may not be feasible.
As with most neutralization processes, acid/alkali mixing can be operated
in either a batch or continuous mode. Operational type depends primarily on
the variation in flow rate or concentration of the divergent influent
streams. Batch operations are typically utilized in treating concentrated
batch dumps or intermittent flow applications. Reactor configurations can be
either single- or multi-stage. However, unit-processing of the wastes
considered in this document (pH less than 2 or greater than 12.5) generally
require multi-stage continuous operation.
Mutual neutralization finds its widest application in waste treatment
systems where reagent consumption can be reduced prior to primary
neutralization through the averaging of flow rates and pollutant loadings. In
cases where the intermixing of the acid/alkali wastestreams result in an
effluent suitable for discharge, mutual neutralization can be utilized as a
primary neutralization process. General equipment for pretreatment systems
typically consists of: storage tanks, metering and pH control equipment,
4-40
-------
attenuation vessel, pumps, and segregation and collection equipment.
Continuous, acid/alkali mixing systems operating as primary neutralization
processes usually consist of: flow equalization basin, neutralization
vessel(s), emergency reagent storage and feed system, effluent pH recorder and
controller, associated pumps and piping.
Figure 4.2.1 illustrates a. mutual neutralization pretreatment system
which is commonly employed in the metal finishing industry. The influent to
the system consists of three primary streams; effluent from cyanide oxidation
(pH 9 to 11) chromium reduction (pH 2 to 3), and various metal cleaning and
o
plating operation waste streams (variable pH). The three streams are
combined into a collection vessel where hydrogen ion and flow rate averaging
occurs through mechanical agitation. The resultant aggregate (pH 5) is
adjusted with sodium hydroxide or lime to pH 9 or 9.5 in the main
neutralization/precipitation reactor to precipitate the heavy metals as
hydroxides.
The method of acid/alkali mixing is a highly site-specific wastewater
treatment process. It requires a detailed collection of process data (flow
rate, batch dumping frequency, etc.), as well as comprehensive laboratory
analysis to determine waste stream characteristics and variability. Once this
is completed and the reaction kinetics, products, and processing requirements
are determined to be favorable, the two types of waste may be combined
approximately in stoichiometric equivalents, to neutralize the component which
is above an acceptable level for discharge. For example, a 20,000 gpd
concentrated (8 percent sulfuric acid) spent pickling liquor waste stream
would require approximately 13,050 gpd of a general purpose alkaline cleaning
solution (10 percent sodium hydroxide) to achieve hypothetical
neutralization. In actual practice, a greater quantity of sodium hydroxide
solution will be required due to excess hydroxide demand for metallic species
such as iron present in the pickle liquor.
Typically, in continuous mutual neutralization operations where
acid/alkali mixing is the primary treatment, provisions are made for storage
tanks if either waste stream is produced in excess of required quantities. In
addition, an emergency reagent feed system is sometimes required if either
stream may be generated in less than the specified quantity or concentration
to ensure a uniform discharge. Retention time is also usually greater than
for other neutralization systems due to the dilute nature of the reagent
streams 7
4-41
-------
1
WuMwawr
RoccuM&on
**mg*g
Oanfmtion
Skidg*
Manning
nr 11
. JLW/J *
Figure 4.2.1 Conventional wastewater treatment system for electroplating.
Source: Reference 2
4-42
-------
Mutual neutralization systems are not limited to liquid/liquid
application. They can be operated with either liquid/gaseous (e.'g., boiler
flue gas, Section 4.7) or liquid/solid (e.g., waste carbide limes,
Section 4.3) reagents. Semi-solids such as battery paste have also been
utilized in recent mutual neutralization applications.
When onsite waste streams are not available, they might possibly be
obtained from a nearby plant that is producing compatible acid/alkaline
wastes; e.g., waste pickling liquor from a metal finishing operation and waste
carbide lime from an acetylene production plant. Such waste exchanges can be
conducted either privately or through the use of a commercial waste exchange
firm. A more detailed discussion of-waste exchanges can be found in
Section 5.9.
4.2.2 Process Performance
The following case studies contain performance data on both simple
(two streams) and complex (three streams or more) acid/alkali mixing
applications. In addition, the data have been gathered from a variety of
industries and manufacturing operations and is intended to illustrate the
versatility of applying mutual neutralization to corrosive waste stream
treatment.
A battery paste consisting of lead oxide, free lead, and lead sulfate is
currently used in the treatment of sulfuric acid wastewater produced in a
lead-acid battery manufacturing plant. The paste (a battery manufacturing
byproduct) is fed into the system in the form of a dilute slurry
(1 to 5 percent solids) at typical strengths of 4 to 10 Ibs. of paste/pound of
sulfuric acid, depending on the lead oxide content of the paste.
The reaction takes place continuously (16,500 to 22,000 gpd) in
multi-stage reactors provided with mechanical agitation. With an influent
stream pH of 1, the effluent pH averaged about 6.9 with 6.0 as the lowest
recorded pH and 8.4 as the highest in a one month period. In addition, the
lead oxide treatment had the added benefit of lowering the sulfate ion content
in the effluent to 323 to 450 mg/L. By comparison, a similar plant using
sodium hydroxide to regulate the acidity of the wastewater produced sulfate
ion concentrations of 1,000 to 2,000 mg/L or more.
4-43
-------
A second example of mutual neutralization involves a high-volume, two
' stream integration at a major automobile manufacturing plant. The wastes to
be neutralized include a 200,000 gpd alkaline waste stream and a 60,000 gpd
acid waste stream. An engineering study indicated that the combined flow
would be of sufficient nature to correct the pH to a range of 6.0 to 10.5
prior to discharge to the sewer system. To prevent any high alkaline
conditions that could raise the pH above discharge limits, an auxiliary
neutralization system was installed. The system consisted of a pH control
system, 2-Milton Roy acid pumps, and a 10,000 gallon sulfuric acid storage
tank. The sulfuric acid tank was located immediately adjacent to a private
railroad spur, so that acid could be purchased in tank car lots for price
advantage.
A third case of simple mutual neutralization examined the feasibility of
combining selected acid and alkaline wastes from coal-fired power plants.
The 'goals of this research program were to minimize hazardous waste generation
and effectively isolating any hazardous constituents, through mutual
neutralization and subsequent precipitation. The two waste streams of
interest consisted of the supernatant from an aqueous fly ash suspension
(pH 12.4) and an acid-iron boiler cleaning'solution (pH 0.7). When the two
wastes were combined, the neutralization reaction was practically
instantaneous (pH 9.0) with both major and minor trace metallic contaminants
being effectively adsorbed (see Table 4.2.1). The authors suggest that
co-neutralization of acidic wastes with fly ash is readily applicable to other
industrial waste streams since, with a snail dosage of polymeric coagulant,
liquid/solid separation is easily achieved.
The fourth case study involves a slightly more complex mutual
neutralization application at a metal finishing facility. In this
application, the waste treatment system combined a metals-laden waste acid
stream with two highly alkaline waste cleaner streams in a continuous
acid/alkali pretreatment operation. In this manner the facility was able to
economically reduce neutralization reagent costs prior to final pH adjustment
and effectively remove metals (primarily iron) through hydroxide
precipitation. The acid waste stream consisted primarily of a spent pickle
liquor solution which averaged 14,000 gpd and pH 2.1 or less. The second
stream consisted of an industrial detergent with pH 12.9, average flow rate of
4-44
-------
TABLE 4.2.1 CHARACTERIZATION OF MUTUAL NEUTRALIZATION
WASTE STREAMS FROM COAL-FIRED PLANT APPLICATION
Constituent
PH
Alkalinity (mg/L)
Cl mg/L
804 mg/L
Ha mg/L
Kmg/L
Mgmg/L
Camg/L
Fe mg/L
Cu Ug/L
Zn ug/L
Cd Ug/L~
Cr ug/L
Se ug/L
As ug/L
Pb UR/L
Hi us/I
HH3 mg/L
Fly ash
solution
12.4
878
2.5
5.3
1.4
0.6
99.4
0.4
2.0
119
0.1
5
^•^•^
Boiler cleaning
waste
0.7
—
40,000
130
43
0.31
3.7
3.3
5,130
159,000
15,890
22
1,620
. 5
7
20.5
2,910
— -^ ^^— •.^^•^^
Combine da
system
9
—
1.2xlO~2
1. 5xlO~5
2.4xlO~4
3.4xlU~5
2.4xlO~5
2.4xlO~3
l.OxlO'3
2.0X1U-5
2.7xlO~6
1.5xlO~9
2.5xlO~6
_:__
1.3xlO"u
—
1.5xlO~6
Not
measured
* Concentrations express in M/L.
Source: Reference 5
4-45
-------
12,000 gpd and contained 74 mg/L of calcium. The third stream, a highly
alkaline CpH 13.1) scrap cleaner had an average flow rate of 19,000 gpd. The
resultant flow stream had a pH of 12.3 which was subsequently adjusted by the
addition of sulfuric acid to bring the effluent within discharge limits. A
summary of waste stream characteristics is provided in Table 4.2.2.
The fifth case study involved a high volume, complex acid/alkali mixing
application at a major manufacturer of household appliances in
Allentown, PA. The process operates as a primary neutralization system and
consisted of the integration of an acid, alkaline, and mixed waste stream.
The acid waste stream (60,000 gpd) contained wastes from pickling solutions,
acid rinses, and chromium stripping solutions. The alkaline waste stream
(43,200 gpd) contained primarily spent cleaning solutions. The mixed waste
stream (63,000 gpd) consisted of process spills, demineralizer regeneration
waters and miscellaneous acid/alkali rinses. Following sodium
bisulfite-hexavalent chrome reduction, the three streams were combined into
one of two 12 foot reactoz—clarifiers. After clarification, solids were
removed through the use of a belt filter and then disposed in an offsite
landfill. The characteristics of each of Che three waste streams and the
final effluent are summarized in Table 4.2.3.
4.2.3 Process Costs
The costs associated with two stream acid/alkali mixing can be described
in terms of capital expenditures, operation, and maintenance costs. Specific
requirements will depend on the specific nature of the waste streams
involved. An example cost analysis is provided herein for a mixed waste
system which requires solids removal and sludge consolidation. Operation and
maintenance costs for this system include; depreciation, interest, taxes,
insurance, maintenance, labor, and sludge disposal. Capital costs include;
collection sumps and associative piping, a two stage continuous reactor
system, emergency pH control system, a flocculation/clarification unit, sludge
storage, and a high pressure filter press. Equipment sizing is based on
combined acid/alkali flows of 400, 2,500, and 3,500 gallons per hour, 8 hours
a day, 300 days per year. The acid stream is assumed to be 60 percent of the
4-46
-------
TABLE 4.2.2 SUMMARY OF APPLICATION OF MUTUAL NEUTRALIZATION
TO METAL FINISHING WASTES
Parameter
Flow rate (gpd)
Percent of total
Alkalinity (rag/L CaC03
Acidity (mR/L)
pH
Calcium concentration (mg/L)
Stream 1
14,000
31
—
1,030
2.1
—
Stream 2 Stream 3
12,000 19,000
27 42
7,600 10,950
—
12.7 13.1
74
Combined
stream
45,000
100
6,400
—
12.3
31
Source: Reference 6
4-47
-------
TABLE 4.2.3 AVERAGE QUALITY OF ACID/ALKALI WASTE STREAMS
Parameter*
Suspended Solids
Total Alkalinity5
Total Acidity
Total Cr
Cr*6
Fe
Cl
S04
P04
Hi
Cu
Al
PH
Acid Alkali Mixed
Stream Stream Stream
26
—
7,400
307
270
45
1,200
660
575
35
1.7
39.6
1.2
1,400
~
1,730
9.7
2.8
—
130
16
176
2
0.1
7.6
12.4
67
900
—
2.9
—
0.2
200
12
191
51
—
3.3
2.4
Combined
Effluent
Discharge
5
125
25
0.75 .
—
0.1
56
loO
70
0.2
0.1
0.05
7.4
•All concentrations expressed in ug/L.
Expressed in nut/L
Source: Reference 7
4-48
-------
total flow and contain 8 percent sulfuric acid and 24 mg/L of ferrous iron.
The alkaline stream is assumed to be 40 percent of the total flow and contains
10 percent by weight of sodium hydroxide.
Prior to acid/alkali mixing, each waste type must be segregated in
separate pipe systems and drained into the appropriate collection sump. Each
system consists of two sumps sized for a maximum of 30 minutes of detention
time. Each sump is a preformed PVC inground tank with sump pump, level
control, and steel grating cover. In addition, provision has been made for
two wastewater collection systems, each consisting of; 60 feet of 6 inch
diameter pipe with two bends and six connections. The following costs are
based on updated (July, 1986) costing information contained in "Reducing Water
Pollution Costs in the Electroplating Industry":8
• Collection Conduit
Linear Runs 4" - $1.55/ft
6" - $2.40/ft
Each Connection 4" - $210.00
6" - $360.00
•
Each Bend 4" - $210.00
6" - $360.00
• Collection Sumps 100 gallons - $2,100
300 gallons - $2,600
500 gallons - $3,100
1,000 gallons - $3J600
The costs for the two stage continuous reactor system are based on a
first stage flow concenCration/equalization tank and a second stage final pH
adjustment tank equipped with an emergency pH control system. Figure 4.2.2
presents mixed reactor construction costs for the first stage reinforced
concrete tank applicable to vessel sizes of 14.2 m to 70.8 m .
Figure 4.2.3 presents mixed reactor construction and installation costs for
the second stage continuous neutralizer complete with pll control and emergency
reagent storage and feed systems. The reagent feed and storage system is
sized for a 1-week supply and uses caustic soda and sulfuric acid as the
neutralizing agents.
4-49
-------
Ul
o
40
—to
ss
oe
z
o
o
20
10
I
I I I I M I
5 10 20
VOLUME, M3
SO
[00
Figure A.2.2 Construction costs for reinforced concrete reactor.
Source: Reference 1
-------
42
o
o
X 35
8
26
•- 0>
co —
21 -
14
MINIMUM UNIT
SIZE
40
60
80
100
FLOW RATE , gol/min
Figure 4.2.3 Investment cost for continuous
neutralization unit.
Source: Reference 8
120
4-51
-------
The flocculater/clarifier, sludge storage, and filter press unit costs
are based on information contained in Figures 4.1.6, 4.1.8, and 4.1.9,
respectively. Sludge generation is assumed to be 0.68 gallons of
sludge/gallon of clarified underflow processed. The density of the sludge is
8.34 Ibs/gallon of sludge based on a 2 percent solution which is subsequently
dewatered to 20 percent solids in the filter press.
Yearly operation and maintenance costs are based on percentages of the
total annualized capital. Taxes and insurance are assumed to be 7 percent,
8
while general maintenance and overhead is 5 percent. Labor is assumed to
be 4 hours/day, 300 days/year, at a rate of $20/hour. Sludge disposal costs
are based on the operation of the filter press for the 400, 2,500, and 3,500
gallon/hour systems at rates of 5, 28, and 40 gallons/hour of clarifier
underflow. Approximately 16.7 Ibs of dry solids are assumed to be generated
per 100 gallons processed. At 20 percent solids and assuming a disposal rate
of $2.0/gallon of sludge, disposal costs can be estimated using
Figure 4.1.10. A summary of mutual neutralization costs is presented in
Table 4.2.4 for each of the three flow rates. As shown, significant economies
of scale are acheive, particularly between the 400 and 2,500 gpm flow rates.
Labor requirements are essentially the same regardless of system size. As
capacity increases, disposal costs and sludge handling equipment show the most
significant increase relative to total treatment cost. Thus, it is more
critical for high flow systems to select wastes which result in minimal sludge
generation.
4.2.4 Process Status
The mixing of acid/alkali wastes is both technically feasible and widely
applied. Its primary advantage is reduced cost since neutralizing reagent
requirements are minimized. The main disadvantage is that mixing two waste
streams, each with its own variability in composition and flow, may require
more conservative system design; i.e., larger equalization and neutralization
tanks and back-up reagent grade neutralization systems. Additionally, care
must be exercised when combining waste streams or accepting wastes from
another firm to prevent any hazardous by-products or releases in the
neutralization reaction.
4-52
-------
TABLE 4.2.4. MUTUAL NEUTRALIZATION COSTS
Capital investment ($)
Collection sumps (2)
Piping systems (2)
1st stage reactor
2nd stage reactor with
emergency pH control system
Flocculation /c lari f icat ion
unit
Sludge storage unit
Filter press
Total capital costs
Annualized capital*
Operating costs (S/yr)
Taxes and insurance (7Z)
Maintenance and
overhead (5Z)
Labor and overhead
<*20/hr)
Sludge disposal ($2.0/gal)
Total cost/yr ($)
Cost/1,000 gallons ($)
•annual ized capital costs derived by
CRF -
400
6,200
5,676
17,000
3Q.OOO
22,000
2,000
11.000
93,876
16,616
1,163
830
24, 000
2,000
44,609
47
using a capital
Flow rate (.gph)
2,500
7.2UO
5,676
19,000
32,000
32,000
10,000
13,500
119,376
21,129
1,479
1,056
24, 000
11,190
58, 854
10
factor:
7,200
3,67o
21,000
35,000
40,000
12,000
18.000
138,076
24,581
1,721
1,229
24, 000
10,008
67,539
8
where: i • interest rate and n " life of the investment. A CRF of 0.177 was
used to prepare cost estimates in this document. This corresponds to
an annual interest rate of 12 percent and an equipment life of 10 years.
Does not include utilities.
Source: Adapted from References 1, 2. and 8 using July, 1986 CPI index.
4-53
-------
The environmental impact of this technology is similar to that of other
neutralization technologies provided that incompatible waste constituents do
not generate toxic by-products. However, as with most neutralization
technologies! a sludge product is generated through the formation of insoluble
o
by-products and the precipitation of metals. Any technical developments
which reduce the amount of sludge and enhance its filtering and settling
characteristics will improve the acceptance of this type of neutralization.
4-54
-------
REFERENCES
1. MITRE Corporation/Manual of Practice for Wastewater Neutralization and
Precipitation. EPA-600/2-81-148. August 1981.
2. U.S. Environmental Protection Agency. Economics of Wastewater Treatment
Alternatives for the Electroplating Industry. EPA-625/5-79-016. June
1979.
3. Yehaskel, A. Industrial Wastewater Cleanup, Recent Development. Noyes
Data Corporation, Park Ridge, NJ. USA 1979, pp. 250-251.
4. Besselievre, E.B. The Treatment of Industrial Wastes. McGraw Hill Book
Company, New York, NY. 1967.
5. Benjamin, M.M. Removal of Toxic Metals from Power Generation Waste
Streams by Absorption and Co-precipitation. Journal of the Water
Pollution Control Federation. November 1982.
6. Price, F., et al. Report on Business and the Environment. McGraw-Hill
Book Company, New York, NY. 1972.
7. Anderson, J.S. Case History of Wastewater Treatment in a General
Electric Appliance Plant. Journal of the Water Pollution Control
Federation. October 1968.
8. U.S. EPA. Reducing Water Pollution Costs in the Electroplating
Industry. EPA-625/5-85-016. September 1985.
9. Arthur D. Little, Inc. Physical, Chemical, and Biological Treatment
Techniques for Industrial Wastes. U.S. EPA SW-148. November 1976.
4-55
-------
4.3 LIMESTONE TREATMENT
4.3.1 Process Description
Limestone treatment is a well-developed and established technology for
the neutralization of acidic waste streams. Limestone is a particularly
effective reagent for the neutralization of dilute acid waste streams
containing low concentrations of acid salts and suspended solids. With
modifications, it may also be a good neutralizing agent for many of the acidic
waste streams considered in this document, either as a primary treatment for
weak acids or as a pretreatment for other processes i.e., partial
neutralization. However, in most applications, limestone has been replaced by
more cost-effective reagents such as lime slurry and caustic soda which
eliminate solids handling problems. In addition, caustic soda results in
reduced sludge generation.
Limestone is available in either high calcium (CaCO.) or dolomitic
(CaCO. MgCO.) form. A summary of physical and chemical properties is
provided in Table 4.3.1. Both types of limestone are available as either a
powder or crushed stone. . Crushed stone diameters are typically 0.074 mm
(206 mesh) or less since both the reactivity and completeness of the reaction
increase proportionately to the available surface area. High calcium is
most commonly used because of its greater reaction rate and its more
widespread availability. Dolomitic limestone reactivity-will increase if
finely ground and sludge production will be minimal due to the formation of
soluble magnesium sulfate. However, its reactivity is generally too slow even
with grinding, and hence not suitable for most applications.
Various configurations are available for limestone treatment applications
as illustrated in Figure 4.3.1. Influent characteristics, effluent criteria,
operational and economic constraints will provide the basis for design
selection. Waste streams are introduced to the neutralization media in an
upflow direction with the application rate dependent on the stone size. For
example, at rates of 50 gal/f t2/min., any stone smaller than 30 mesh will be
swept from the bed. Therefore, when utilizing limestone powder (200 mesh),
flow rates are lowered to 1 to 5 gal/ft /min, thereby limiting waste
4-56
-------
TABLE 4.3.1 SUMMARY OF HIGH CALCIUM AND DOLOMITIC LIMESTONE PROPERTIES
Parameter
Chemical Analyses
CaO
MgO
C02
Si02
A1203
Other
Bulk Density
Specific Gravity
Solubility
Molecular Weight
Specific Heat
Stability
Basicity Factor
pH
Unit
High calcium
wt (%)
58.84
0.26
43.26
1.14
0.41
—
kg/m3 2,000 - 2,800
H 0 - 1 2.65 - 2.75
g/100 g water 0.0014
g/gmole 100. I
Cal/g/ C °-19 * °-26
«C 898
CaO - 1 0.489
H20 -7.0 8-9
Dolomitic
30.07
20.75
46.02
0.14
1.90
1.12
* 2,050 - 21,870
2.75 - 2.90
0.0245
184.39
0.19 - 0.294
725
0.564
8.5 - 9.2
Source: References 2 and 3
4-57
-------
SINGLE FLOW REACTOR
UPFLOW M SERIES
I
I
UPFLOW IN PARALLEL
Figure 4.3.1. Upflow limestone bed neutralization process
configurations.
4-58
-------
4
throughput. The upflow expanded configuration permits the use of smaller
particle size, limits channeling, increases neutralization rates, and
decreases required bed size.
There are two basic modes of operation for limestone beds; fixed and
•oving. In the fixed bed mode, the entire bed is removed from service when
the stones become inactive. In moving beds, a high rate of application washes
away any insoluble particles which might coat the limestone surface, extending
the bed life indefinitely. For arrangements in series, when the first bed
becomes saturated, effluent quality is maintained in the second. This method
is generally applied to waste streams with high solids content and low
application rates. Conversely, beds in parallel are primarily used for high
volume applications where suspended solids content in the influent is low.
Figure 4.3.2 is a schematic of a single bed, upflow treatment system
utilizing a settling and mixing section prior to neutralization. In this
system, a uniform, dilute waste is fed to the bed through a large holding
capacity, perforated pipe feed apparatus. Influent dilution with recycled
treated product or more neutral process streams is necessary since some strong
mineral acids such as sulfuric acid will react with limestone forming
insoluble calcium sulfate. Generally, sulfuric acid -concentrations greater
than 1.3 percent (although greater than 5,000 mg/L is not recommended) will
4
result in the sulfurization of the stone particles. When this occurs, the
neutralization capacity of the stone particles is greatly reduced and the bed
is eventually rendered inactive.
Aeration of the treated waste stream is frequently desireable since
carbon dioxide is evolved during the neutralization process. Stripping of
CO. through some form of mechanical aeration will prevent the formation of
carbonic acid which would otherwise depress the effluent pH.
Process operation data for full-scale applications are incomplete for
three primary reasons; either the essential data have not been generated, is
considered proprietary, or is outdated due to the increasing preference of
alternate alkali reagents in recent years. The available information does,
however, contain material gathered from a wide variety of data sources;
including manufacturers, trade associations, industrial users, and available
literature. Table 4.3.2 gives a composite summary of several important
parameters in a "typical" upflow limestone treatment bed operation. These
parameters will be covered in more detail in later sections.
4-59
-------
v>
WASTE AGIO STREAMS
WATER and VENT
OASES
RECYCLE EFFLUENT
O
n
LU.L
n
'RECYCLE
PUMP
SETTLING
SECTION
MIXING
SECTION
LIMESTONE
BED
GRIT
REMOVAL
COLLECTING DISCHARGE
SECTION
SUMP
Figure 4.3.2. Single bed, upflow limestone treatment system schematic.
Source: Reference 5.
-------
TABLE 4.3.2 SUMMARY OF TYPICAL OPERATING PARAMETERS
Parameter
Application Rate
Type of Stone
Temperature
Incoming Mineral
Acidity
Bed Depth
Stone Size
Aeration Time
Unit(s)
gal/ft2/min
Z Magnesium
Oxide
°C
mj?/L
ft
. Mesh
Minutes
Operating
range
1 -80
High Calcium -
52 MgO -
15 -
5.000 -
2 -
80 -
0 -
Dolomitic
402 MgO
30
7,000
4
200
10
Ideal range
'1-5
5
15
5,000
3
200
2
Source: References 3, 4, 5, 6
4-61
-------
4.3.2 Process Performance
In evaluating the effectiveness of limestone as a neutralization agent,
parameters of interest are; type of bed, type of limestone, flow rate,
influent mineral acidity, bed depth, effluent pH, rate of reaction and pre-
and post-treatment requirements. This section contains several case studies
in which limestone treatment was evaluated for neutralizing dilute acidic
waste streams. A summary of performance data and process parameters for
limestone treatment beds is presented in Table 4.3.3. Data for the individual
waste streams are, with few exceptions, for influent concentrations of 5,000
to 10,000 mg/L of mineral acidity.
One case study involves a bench-scale evaluation of the neutralization of
a dilute nitric and sulfuric acid waste stream from the manufacture of
A
nitro-cellulose (Table 4.3.4). The parameters evaluated were rate of
application, influent mineral acidity, type of stone, effect of aeration, and
• ludge product formed. The experimental unit employed a bed of fine stone
(approximately 8 to 30 mesh) in which the waste was applied at high upflow
rates. The rates of application ranged* from 20 to 80 gal/ft /min. The
waste acidity.was varied from 2,950 ppm to 12,400 ppm, but 5,000 ppm of
sulfuric acid acidity was judged optimal due to calcium sulfate formation.
Bed depths were also varied, but sufficient depth was allowed for
decomposition of the stone by the acid. Depths of 2 to 4 feet were
recommended so that expansion was not excessive,
Results of the study indicated that both amorphous and crystalline
high-calcium limestone were satisfactory reagents while dolomitic limestone
did not achieve satisfactory performance. Wastes containing appreciable
concentrations of acid salts were not neutralized at high rates, possibly due
to bed clogging or fouling. Finally, acid wastes relatively free of solids
can be continuously neutralized by an upflow rather than downflow application
to beds utilizing limestone gravel. Since limestone is limited in its ability
to neutralize acidic solutions above pH 7.0 (over-neutralization), it
generally cannot be used as the sole reagent in applications where metal
precipitation is required.
4-62
-------
TABLE 4.3.3 SUMMARY OF LIMESTONE BED TREATMENT PERFORMANCE"DATA
OJ
Application
Resin
Manufacturing
Waste3
Silicon
Products
Wastes
Acid Mine
Drainage0
Nitro-Cellulose
Wasted
Nitro-Cellulose
Waste
Nitro-Cellulose
Waste
Type
of
bed
Upflow
Upflow
Upflow
(pilot)
Upflow
(lab)
Upflow
(lab)
Upflow
(lab)
Type of
limestone
Unspecified '
94% Soluble
High calcium
Dense High
Calcium
(England)
Calcite
Calcite
Amorphous
Flow rate ; Concentration
(gal/ft* /m) ; (mg/L)
13.6 g/ft2/m 1% HC1
20 - 30 <1% HC1
6.1 5,000 mg/L
21.6 ' 5,660 mg/L
34% NHC-3
66% H2S04
16.9 2,950 mg/L /
15.8 2,950 mg/L
Bed Bed
depth area
, (ft) ; (ft2)
3 ft 113.5 ft2
3 27
10 .785
2 .083
1 .083
1 .083
Effluent
PH
Unspecified
4.0 - 6.0
5.8 - 6.5
5.4
4.7
4.8
aSource: | Reference 7
Source: Reference 4
C8ource: Reference 8
"Source: j Reference 5
-------
TABLE 4.3.4 CASE STUDY 1. BENCH SCALE LIMESTONE TREATMENT DATA SUMMARY
Rate of application
(gal/ft2/min)
a
a
a
21.6
52.8
60.0
16.9
21.2
34.0
15.8
21.6
44.0
b
b
b
Influent
Mineral
acidity
(ppm)
12,400
6,200
3, 100
5,660
5,660
5,660
2,950
2,950
2,950
2,950
2,950
2,950
b
b
b
Aeration
Effluent Type of time
pH , stone (min)
— High calcium —
— —
— —
5.4 High calcium —
4.9 —
4.0 —
4.7 Calcite —
4.2
2.7 —
4. 8 Amorphous —
4.2 —
2.6
4.3 High calcium 0
7.4 3
8.0 10
CaS04
formed
(ppm)
11,084
5,542
2,271
—
—
—
—
—
'. —
—
—
—
—
—
—
•Liter portions of waste were stirred rapidly after addition of limestone.
bAeration of effluent having a pH of 4.3 with diffused air.
Source: Reference 4
4-64
-------
Another case study describes a treatment train in which limestone is used
in conjunction with biochemical oxidation for the removal of acid salts. The
evaluation consisted of several bench-scale studies followed by a pilot plant
g
application (Table 4.3.5). The waste to be treated was acid mine drainage
vfaicb contained less than 5,000 mg/L of sulfuric acid (pH 2.8) and a high
concentration of acid salts (primarily ferrous sulfate). The pilot plant was
designed to convert ferrous-to-ferric salts 'through biochemical oxidation,
followed by neutralization with limestone (see Figure 4.3.3). The
2
neutralization system consisted of two 10 foot (0.785 ft diameter)
limestone grit reactors arranged in series. The system was operated for
28 months at an average temperature of 20°C with effluent pH ranging from
5.8 to 6.5.
The lower limit of application for the system was expected to be 10 to
20 mg/L of dissolved iron and a total acidity of approximately 25 mg/L of
•ulfuric acid. The upper limit was assumed to be 5,000 mg/L H^SO^, but
higher limits may be achieved at lower operating temperatures (15°C) and
through the use of aeration. Peak flow rate for an operational system was
estimated at approximately 832,000 gpd with an average flow rate of
approximately 240,000 gpd.
The inherent problem with this type of neutralization/precipitation
system is that it is only effective for reducing metallic species such as
trivalent chrome and iron in its operational (pH) range. In addition, the
high solubility of Cr+ (approximately 8 mg/L) in comparison to Fe+
(0.0075 mg/L) at pH 6.0, limits this treatment in potential applications to
primarily the removal of acid iron salts.
While it has been demonstrated in the previous two examples that
limestone neutralization is possible, the inhibition of the stone particles in
the presence of high quantities of sulfate and/or metallic ions make it less
attractive than other reagents. Limestone is a solid-based reagent that
liberates CaO for neutralization through surface dissolution. The inihibition
of the particle surface through calcium sulfate precipitation increases
retention times, reagent purchases, equipment sizing, and lowers waste
throughput. Improved reaction kinetics can be achieved by increasing the
available solid surface area through greater limestone loading. However, both
reagent purchase and sludge disposal costs will increase proportionately with
the excess limestone applied.
4-65
-------
TABLE 4.3.5 SUMMARY OF LIMESTONE TREATMENT PILOT PLANT DATA
Teat Series
Fowler
first Grit
Second Grit
Third Grit
Fourth Grit
Fifth Grit
Sixth Grit
Lines tone
particle aize
Ezpertsints
<60 I.S.S.
3/16 x 1/4 in
1/2 x 3/4 in
12 I.S.S. to 1/2 in
12 to 4 I.S.S. and
l/« - 1/4 in
1/16 to 1/4 in
60-22 I.S.S.
60-16 I.S.S.
Height of
1 lne«toae
<«*>
0.05 - 1
35
300
1.3
1.7
105 - 150
0.5
7
Acid nine
drainage
feed
Aliquot
Continuous
Continuous
Aliquot
Continuous
Continuous
Aliaoot
Continuous
Bate of
neutralization
of strong
acid salts
Faat
Faat initially
Faat initially
Faat
Faat
Faat
1-4 nia
4 **jft
Flow
Configuration Application
*nA nixing rate
Agitated —
Down 20 gpn
Horizontal 20 gpn
Aerated —
Don or vp 25 nL/nln
Don 750 nL/mln
tu-blins -
Op 3.3 ft/sun
Loss of activity
or blocking of
bed
Blockad in 48 hours
Blockad in few days
Sot dateninad
Blocked in 70 hours.
reatored by up flow
expansion
Blockad SO - 150 hours.
restored by upflov expansion
•one • •
25 hours to loas of activity
laconditT fapariaanu
Filot Flant
60 - approxisiualy
6 «.S.S.
40
Cantinaeai 4 sda
3.3 ft/»in
•o less of activity with
attributors
Soarea: tafarasca 8
I.S.S. • Srltlah slavt aixa.
4-66
-------
Air
Limestone
Grit
Acid
Mine
Drainage
FLOW
BALANCING
BIOCHEMICAL
OXIDATION
LIMESTONE
NEUTRALIZATION
Active
Sludge
Figure 4.3.3 Flow diagram of complete biochemical oxidation
and limestone neutralization process.
Cake to
Waste
TREATED
EFFLUENT
Source: Reference 8
-------
In 1981, Volpicelli et al. sought to increase limestone dissolution rates
o
by decreasing Che mean particle size of the stone. Crushed stones of
various diameters were applied in stoichiometric amounts in a controlled
stirred reactor. The waste stream that was neutralized was a sugar plant
effluent (pH 0.9) containing 17,500 mg/L of sulfuric acid. The experimental
results, which are presented in Table 4.3.6, demonstrated that stone sizes
greater than 400 mesh (38 microns) were only effective when applied in
excesses of 220 to 550 percent. However reaction rates were slow
(35 to 240 minutes) and an excessive quantity of reagent was wasted
(54 to 82 percent). When the mean particle size was reduced to 400 mesh or
less, complete neutralization (pH 7.0) was achieved in 15 minutes due to the
increased dissolution rate. More significantly, only a 10 percent excess of
limestone powder was required, resulting in only 9 percent of the available
surface area being rendered inactive.
Figure 4.3.4 illustrates a two-stage, backmix flow reactor system
proposed by Volpicelli for the continuous application of limestone powder.
With optimum agitation, the reagent loading should approach the solid surface
area to liquid volume ratio determined ;in the laboratory. Provision is made
in the system design for an aeration vessel prior to sludge thickening to
raise the final pH from 5.75 to 7.0. Aeration is required since carbon
dioxide is evolved as a by-product in the limestone neutralization process.
The carbon dioxide will often combine with free hydrogen to form carbonic acid
which will depress the pH endpoint. Therefore some sort of desorption
apparatus such as an aerator or a drop from, a weir is employed to strip the
C02 prior to final discharge.
4.3.3 Process Costs
Table 4.3.7 summarizes the process costis for the construction and
installation of a continuous limestone powder neutralization system.
Equipment sizing and operating costs were based on the neutralization of a
2 percent sulfuric acid waste stream with various flow rates. The flow rates
are assumed to be 400, 2,500, and 3,500 gallons/hour, operating
2,400 hours/year. The capital costs include; a collection sump and
associative piping, a two-stage continuous reactor, a dry powder feed system,
4-68
-------
TABLE 4.3.6 SUMMARY OP LIMESTONE NEUTRALIZATION EXPERIMENTS
Run
1
2
3
4
5
I
7
8
9
10
11
Liquid Particle Limestone Reaction
volume size weight Limestone time Final
(L) (mm) (g) Excess8 . (min) pH
.75
.50
.50
.50
.50
.50
.50
.25
.25
.25
.25
1-1.25
1-1.25
.42-. 6
.2-. 42
.2-. 42
.09-. 125
Powder
Powder
Powder
Powder
Powder
16
50
50
50
25
20
10
5
2.5
3.5
4
1.18
5.49
5.49
5.49
2.76
2.20
1.10
1.10
.55
.77
.88
90
130
240
35
180
35
15
15
15
15
15
1.12
1.60
7.00
6.25
2.00
6.90
6.55
6.55
1.30
1.70
2.70
Surface
area
(cm2/L)
420
1,960
4,300
7,200
3,600
8,200
12,000
12,000
6,000
8,400
9,600
Limestone
conversion
16.4
9.4
18.0
18.0
25.0
46.0
91.0
91.0
98.0
96.0
98.0
aRatio between loaded and stoichiometric weight of limestone.
Source: Reference 9
-------
MS-
LIMESTONE
COAGULANT
THICKENER
ACID WASTE
WATER
pH-Q9
TO WASTE
SLUDGE
Figure 4.3.4 Continuous limestone powder neutralization process.
Source: Reference 9
-------
TABLE 4.3.7 CONTINUOUS LIMESTONE POWDER TREATEMENT COSTS
Flowrate (gph)
400
2500
3500
Captial lnvestment($)
Collection sump
Piping system
2-Stage reactor
Feed system
Aeration vessel*
Flocculation/Clarification
Sludge storage
Filter press
Total Capital Cost
Annual ized Capital
Operating Costsb(Z)
Taxes and Insurance (72)
Maintenance (52)
Labor ($20/hr)
Sludge Disposal ($2007 ton) c
Reagent Cost ($84/ton)
Total Cost/Year ($)
Cost/1,000 gallons
3100
2838
34,000
37,078
18,419
22,000
8,000
11,000
136,435
24,149
1,690
1,208
24,000
64,514
37,596
153,157
159
3600
2838
38,000
118,648
20,419
32,000
8,000
16,500
240,000
42,480
2,974
2,124
24,000
456,809
98,864
627,251
104
3600
2838
42,000
133,479
22,419
40,000
10,000
21,000
275,336
48,735
3,412
2,437
24,000
565,499
107,795
751,878
90
•Does not include cost of sparger pumps.
''Does not include utilities.
^oes not include gypsum refund.
Source: Adapted from References 10, 11, 12 using July, 1986 CPI index.
4-71
-------
and- sludge separation and handling equipment. Operational expenses were
assumed to include taxes, insurance, maintenance, labor, sludge disposal, and
reagent costs.
The wastewater collection system is comprised of one PVC in-ground tank,
a sump pump, a level controller, and 30 feet of 6-inch pipe with one bend and
three connections (for cost estimate see acid/alkali mixing). The costs for
the reactor system are based on Figure 4.2.2 and consist of two reinforced
concrete reactors sized for 30 minutes of retention time. The cost data for
the feed system is based on Figure 4.3.5 and adapted construction costs for
dry feed systems. A 5 minute detention time is required in the dissolving
tank and water is used at a rate of 2 gallons/pound of limestone. Conveyance
from the solution tanks to the point of application is by dual head diaphragm
retaining pumps. . The limestone is stored in mild steel storage hoppers
located indoors and directly above the storage tanks (hopper facilities also
include dust collectors).
The flocculation/clarification, sludge storage, and filter press unit
costs are based on information contained in Figures 4.1.6, 4.1.8, and 4.1.9
respectively. Sludge generation is assumed to be 1.322 Ibs of calcium sulfate
precipitate per pound of sulfuric acid neutralized based on a calcium sulfate
solubility of 0.241 grams/100 milliliters of solution. Final sludge volumes
will contain 50 to 60 percent- solids, although cost estimates are based on
60 percent as a maximum value attainable by a recessed plate filter press.
The aeration vessel is assumed to be a standard reinforced concrete tank
provided with 30 feet of sparger system piping. Cost estimates are based on
Figure 4.2.2 but does not include purchase and installation of pumps and
compressors which will add slightly to the final figure.
Operating costs for: (1) taxes and insurance; and (2) maintenance, are
based on percentage of the total annualized capital investment
(7 and 5 percent, respectively).11,12 Labor costs are based on 1,200 hours
of manual operation per year at a basic labor rate of $20/hour, including
overhead. Sludge disposal costs are based on dry sludge generation rates of
976, 8,078 and 11,310 Ibs/day for the 400, 2,500, and 3,500 gph systems,
respectively. Final sludge volumes are assumed to be 60 percent dry solids
and disposed of at a cost of $2007ton. Due to the relatively pure state of
the neutralization reaction byproducts, the precipated calcium sulfate
(gypsum) is suitable for a variety of uses such as agricultural liming or
•
4-72
-------
lOOOr
-I
§3
gg
P«
o •
—\ o>
100
w
I
* •
10
100
IOOO 3000
LIMESTONE POWDER FEED RATE (kg/h)
Figure 4.3.5. Construction cost for Limestone Powder Feed System.
Source: Reference 10.
4-73
-------
gypsum board manufacture. Therefore, the possibility of waste exchange or
reuse can significantly reduce or even eliminate the costs associated with
disposal. Reagent costs are based on a quoted price of &84/ton
(Lee Lime, Lee Massachusetts, 1986) for extremely fine (200 mesh or less)
limestone powder. Reagent demand was assumed to be 1.284 Ibs of high calcium
limestone powder per pound of sulfuric acid neutralized. This figure is based
upon a basicity factor of 0.489 and a 10 percent excess due to insoluble
limestone feed.
As evidenced by Table 4.3.7 sludge disposal consitutes as much as
75 percent of the total yearly expenditures for this treatment application.
While more efficient methods of sludge dewatering such as drying ovens may
reduce the overall costs of disposal, the implementation of a land disposal
ban in the near future will make any process generating significant quantities
of sludge product, both economically and environmentally unattractive.
4.3.4 Process Status
The primary advantage of limestone neutralization is that limestone is a
low coat and widely available reagent. However, limestone is limited in its
ability to neutralize over pH 6.0 or acid concentrations greater than
5,000 mg/L. While the process of limestone powder treatment enhances
neutralization, it also increases reagent costs nearly eight fold. The added
cost of grinding and sifting to mesh sizes of 200-400 to achieve the same
neutralizing power as hydra ted lime makes it twice as expensive. This,
combined with the rising costs and environmental awareness in disposing of
solid hazardous waste residues, typically makes limestone treatment
unattractive except under special circumstances.
There have been attempts to use limestone in combination with lime in a
dilute, dual alkali mode. The limestone is used as a pretreatment to raise
the pH to approximately 3.0 or 6.0 with lime completing the process of
neutralization. The limes tone/lime process is usually more complicated than a
simple line slurry process, resulting in higher projected costs and limited
application. However, in large volume applications (see Section 4.4.2) the
large savings in reagent (when used in pebble form) may offset any increases
in capital expenditures.
4-74
-------
REFERENCES
1. Camp, Dresser, and McKee. Technical Assessment of Treatment Alternatives
for Wastes Containing Corrosives. Contract No. 68-01-6403. September
2. Kirk-Othmer Encyclopedia of Chemical Technology. Vol. 14, 3rd. Edition.
John Wiley & Sons, New York, NY. 1981.
3. Boynton, R.S. Chemistry and Technology of Lime and Limestone.
Interscience Publishers, New York, NY. 1966.
4. Gehm, H.W. Neutralization of Acid Waste Waters with Upf low Expanded
Limestone Bed. Sewage Works Journal 16:104-120. 1944.
5. Tully, T.J. Waste Acid Neutralization. Sewage and Industrial Wastes
30:1385. 1985.
6. Levine, R.Y., et al. Sludge Characteristics of Lime Neutralized Pickling
Liquor. 7th Industrial Waste Conference, Purdue University. 1952.
7. Arthur D. Little, Inc. Physical, Chemical, and Biological Treatment
Techniques for Industrial Wastes. U.S. EPA SW-148. November 1976.
8. Gloves, H.G. The Control of Acid Mine Drainage Pollution by Biochemical
Oxidation and Limestone Neutralization Treatment. 22nd Industrial Waste
Conference, Purdue University. 1967.
9. Volpicelli, G. , et al. Development of a Process for Neutralizing Acid
Wastewaters by Powdered Limestone. Environmental Technology Letters,
Vol. 3, pp. 97-102. 1982.
10. MITRE Corporation Manual of Practice for Wastewater Neutralization and
Precipitation. U.S. EPA-600/2-81-148. August 1981.
11. U.S. EPA. PB-220-302. Processes Research, Inc. Neutralization of
Abatement-Derived Sulfuric Acid.
12. Peters, M.S., et al. Plant Design and Economics for Chemical Engineers.
3rd Edition, McGraw-Hill Book Company, New York, NY. 1980.
4-75
-------
4.4 LIME SLURRY TREATMENT
4.4.1 Process Description
Lime Slurry—
Lime slurry treatment of corrosive waste streams is analogous to that of
limestone neutralization. It is one of the oldest and perhaps most prevalent
of all industrial waste treatment processes. It is -used extensively as an
alkaline reagent in the neutralization of pickling wash waters, plating
rinses, acid mine drainage, and process waters from chemical and explosive
234
plants. ' ' Lime slurry has replaced limestone in many applications as a
low-cost alkali due to its greater available surface area, pumpable form,
continuous application, and greater effectiveness in removing Ca salts from
the process. However, similar to the use of limestone, a major
disadvantage of the process is the formation of a voluminous sludge product.
Limes are formed by the thermal degradation of limestone (calcination),
and are available in either high calcium (CaO) or dolomitic (CaO-MgO) form
(Table 4.4.1). The pure, oxidized calcium product is referred to as
quicklime. Quicklime varies in physical form and size, but can generally be
obtained in lump (63 to 255 mm), pebble (6.3 to 63 mm), ground (1.45 to 2.38
mm) or pulverized ( 0.84 to 1.49 mm) form. As with limestone, experimental
evidence has shown an increase in dissolution as the size of a lime particle
diminishes. For example, a 100 percent quicklime of 100 mesh (0.149 mm) will
dissolve twice as fast as one of 48 mesh (0.35 mm).
Although lime can be fed dry, for optimal efficiency it is slaked
(hydrated) and slurried before use. Slaking is usually carried out at
temperatures of 82 to 99°C with reaction times varying from 10 to 30 minutes.
Following slaking, a wet plastic paste is formed (lime putty) and then
slurried with water to a concentration of 10 to 35 percent.
While most lime is sold as quicklime, small lime consumers often cannot
economically justify the additional processing step that slaking entails.
Therefore, high calcium and dolomitic lime are also available in hydrated form
(either Ca(OH)2 or Ca(OB)2 MgO). This product is made by the lime
manufacturer in the form of a fluffy, dry, white powder. It is supplied
either in bulk or in 23 kg (50 Ib) bags. Hydrated lime is suitable for dry
4-76
-------
TABLE 4.4.1 SUMMARY OF HIGH CALCIUM AND DOLOMITE QUICKLIME PROPERTIES
Parameter Unit High calcium lime
Molecular
Formula —
Molecular
Weight g/gmole
Components Z by weight
CaO
MgO
Fe203
A1203
H20
C02
Bulk Density Kg/m3
Specific Gravity H20 » 1.0
Solubility R/lOOg water
Specific Heat Cal/g/°C
Stability °C
CaO
56.1
93.25 - 98.0
0.30 - 2.50
0.20 - 1.50
0.10 - 0.40
0.10 - 0.50
0.10 - 0.90
0.40 - 1.50
770 - 1120
3.2 - 3.4
converts to
Ca(OH)2
0.175 - 0.286
Stable at any temp-
Dolomite lime
CaO - MgO
96.4
55.50 - 57.50
37.60 - 40.BO
0.10 - 1.50
0.05 - 0.40
0.05 - 0.50
0.10 - 0.90
0.40 - 1.50
801 - 1165
3.4 - 3.6
converts to
Ca(OH)2 MgO or
Ca(OH)2 Mg(OH)2
0.19 - 0.294
Same
Basicity Factor CaO-1
pH H20 » 7.0
^^^•i
Source: Reference 5
erature but will
readily hydrolyze
0.941
11.27 - 12.53
1.110
Same
4-77
-------
feeding or for slurrying and the storage characteristics, purity, and
uniformity are generally superior to slaked lime prepared onsite. High
calcium hydrate is far more reactive than dolomitic hydrate, neutralizing
acids in minutes instead of hours. It is also more efficient with dilute
acids and in cases where over-neutralization is required. Dolomitic hydrate,
which possesses greater basicity (approximately 1.2 times), is a much slower
reactant, although heat (76.4°C) and agitation can accelerate its inherently
3
slow reactivity. Generally, dolomitic hydrate is most efficient with
strong acids where complete neutralization (above a pH of 6.5) is not
necessary.
Both quicklime and bydrated lime deteriorate in the presence of carbon
dioxide and water (air-slaking), therefore prior to dry feeding, lime is
generally stored within moisture—proof containers and consumed within a few
weeks after manufacture. The storage characteristics of dry hydrated lime are
superior to quicklime, but carbonation may still occur causing physical
•welling, marked loss of chemical activity, and clogging of discharge valves
and pipes.
Dry chemical feed systems consist of either manual addition of 50 Ib bags
or, in large operations where lime is stored in bulk, automatic mixing and
feeding apparatus. Automatic feeders are often positioned directly at the
base of the bulk storage bins to minimize potential clogging due to excessive
dry lime transport distance. Two types of automatic feed systems are
available. Volumetric feed systems deliver a predetermined volume of lime
while gravimetric systems discharge a predetermined weight. Gravimetric
feeders require more maintenance, are roughly twice as expensive, but can
guarantee a minimum accuracy of 1 percent of set rate versus 30 percent for
volumetric feeders.
Figure 4.4.1 is an example of a typical lime slurry system with storage
and slaking equipment. A slurry tank with agitator is used followed by a
slurry recirculation line. The process flow lines bleed off a portion of
the recirculation slurry to reactors arranged in parallel. The process line
is as short as possible (to prevent caking) and the control valves are located
close to the point of application. Research has indicated that the most
successful control valves used have been automatic pinch valves.
Figure 4.4.2 is a schematic of a small volume, high calcium, hydrated
lime application for the neutralization/precipitation of electroplating
4-78
-------
v\
AIR
SUPPLY
SLURRY
LOOP
1
— •
1
— 1
|l SLURRY
f
1 •- - -'.-,:
""• i-
* "
FLU9MINC
CONTROL
pRoerci
r LOW
PROCESS
REACTOR
r
7
AIR SUPPLY
I
7
--©
—X-
PROCESS
FLOW
PROCESS
REACTOR
•oo-:
BLURRY
PUMP
STASIL.IZATION
ft
6TORACE
Figure 4.4.1 Flowsheet of a lime slurry system.
Source: Reference 7
4-79
-------
imiuuti
tllCtllllltl
iitr-
item
Hif rn
cmimmi iu»
iinti minim
eiiciiiiittt
iMtiiiit iiiiiD HI
Illtllll mr Itlll
tune nun
(IIIK UUil
•Hlltt
p
— Ill
r>
^•i
tflllUIIIIII lilt
ti cm
UICI
StJHl
linn mi
(CIIIIIIIIS IMIIIIII)
Figure 4.4.2. Small volume, high calcium, hydrated lime
slurry treatment system.
Source: Reference 8
4-80
-------
g
wastes. The system operates in either batch or continuous mode and
includes a sedimentation tank for liquid/solid separation and a lagoon for
sludge disposal. A pH probe controls the supply of a lime slurry to either a
sump for the neutralization of concentrated bath dumps (high in acids and
metals) or a continuous mixing tank (57 gpm) for the neutralization of less
acidic wastes; e.g., electroplating rinses. Mixing is provided by both
compressed air and a mechanical agitator. The sedimentation tank is operated
in a semi-continuous manner with sludge being pumped to a lagoon once or twice
a month and the supernatant being discharged to the sewer.
Lime slurry operations are typically conducted under atmospheric
conditions and room temperatures. The neutralizing unit is usually a
reinforced tank with acid-proof lining and some sort of agitation to maintain
intimate contact between the acid wastes and the lime (slurry) solution.
Vertical ribs can be built into the perimeter to keep the contents from
swirling instead of mixing.
During operations, adequate venting may have to be provided due to the
possible evolution of heat and noxious gases. Calcium hydroxide will evolve
31,140 calories/gram molecular weight of sulfuric agid neutralized, and ,
.approximately 27,900 calories with hydrochloric acid. Table 4.4.2 presents '
a summary of process parameters gathered from various lime slurry
neutralization systems. However, while these provide an indication of typical
system design, testing under actual or simulated conditions is the only sound
basis for the determination of individual waste treatment parameters.
Waste Carbide Lime —
One alternative to the purchase of virgin Lime for neutralization
applications, is the use of carbide lime. Calcium carbide and water are
reacted together in the acetylene manufacturing process. To produce acetylene
and by-product carbide lime:
CaC2 + 2H20 - C2H2 + Ca(OH)2
(1)
'4-81
-------
TABLE 4.4.2 SUMMARY OF TYPICAL LIME-SLURRY OPERATING PARAMETERS
Parameter
Unit(s)
Operating range
Optimum range
Type of Stone Z MgO
Stone Size mm
Slaking Temperature °C
Slurry Solids I
Retention Time Min
Sedimentation Time Min
Mineral. Acidity Mg/L
5-40
0.149 - 255
82 - 99
5-40
5 - 15
15 - 60b
It), 000 - 100,000
5
0.149
Same
a
5
15 - 30
20,000
•is dependent on site specific factors
^High calcium lime will settle in 15 minutes with 1-2Z acid wash streams and
30-60 minutes with 3-10 percent acid streams. Dolomite will typically take
15-60 minutes.
Source: References 2, 3, 4
4-82
-------
However, since commercial calcium carbide is usually produced by burning
coke and limestone in an electric furnance, all of the nonvolatile impurities
in the coke and limestone are retained in the calcium carbide. Many of these
impurities are in the reduced oxidation state and, therefore, non-reactive.
Table 4.4.3 contains an example of two typical carbide lime samples which
illustrates the typically high levels of such impurities. In addition, a
study by the National Lime Association showed that the physical
characteristics of carbide lime, such as degree of porosity and surface area
were poorer than commercial lime.
When carbide lime is utilized-in dry powder form, feeding, slaking, and
neutralization process requirements are equivalent to that of hydrated lime.
However, the presence of large quantities of reduced impurities will increase
reagent requirements, sludge volumes, and processing equipment dimensions.
For example, to neutralize 100 Ibs of reagent grade (98 percent pure) sulfuric
acid, stoichiometric requirements for a 98 percent pure high calcium hydrated
lime are approximately 87 Ibs, (10 percent insoluble feed and 2.2 Ibs of
inerts). In comparison, when an 84 percent Ca(OH). - carbide lime is used
to neutralize the same acid, 117.7 Ibs is required (10 percent excess and
'18.8 Ibs of inert impurities)". Storage, slaking, and feeding equipment
dimensions in this case would increase by approximately 35 percent to handle
the greater volume of reagent required on an equivalent neutralization basis.
Similarly, sludge handling equipment and disposal costs will increase by a
minimum of 14 percent based on a. greater quantity of insolubles in the feed.
Consequently, before choosing between virgin reagent and carbide lime wastes,
one should compare the 20 percent price advantage enjoyed by carbide lime
with increased capital expenses, slower reactivity, and variability in the
quality of the waste materials.
Cement Kiln Dust—
Cement kiln dust has also been proposed as an alternative to lime for the
neutralization of acidic waste streams. Cement kiln dust is a cement
aanufacturing byproduct derived from reacted or partially reacted raw
materials (limestone) and fuel used in cement rotary kiln operations. The
dust particles (6 microns or less) become entrained in the rotary kiln hot
exhaust gases and are subsequently removed through collection devices such as
4-83
-------
TABLE 4.4.3 WASTE CARBIDE LIME COMPOSITION
Compound
Ca(OH)2
CaC03
CaS03
CaS04
Si02
A1203
Fe203
CO
CaS
CNS
Weight
Sample A
84.3
6.9
- 1.8
1.0
1.7
0.7
l.l
-
-
-
Percent (Z)
Sample B
96.30
-
-
0.34
1.41
1.33
0.12
0.14
0.08
0.01
Source: Reference 9
4-84
-------
baghouses. Kiln dust is similar to quicklime in that it contains calcium
oxide, is a dry powder, and must be hydrated (i.e., slaked and slurried) prior
to use.
The primary advantage of kiln dust is that it costs about 66 to
75 percent less than virgin lime and consistently displays a pH in the range
of 11.2 to 12.1. However, physical and chemical characteristics vary
widely according to collection method or kiln efficiency. Table 4.4.4
presents elemental and anion variation in U.S. cement kiln dust from 113
different cement kiln operations. As shown,the maximum calcium concentration
analyzed was 36.7 weight percent. Therefore, depending on the degree of
calcination in the kiln exhaust stack, the maximum theoretical CaO
concentration would be approximately 51.3 percent with a median value of
42.6 percent. Facilities replacing virgin quicklime with cement kiln dust
will require a minimum increase in reagent of 93 to 128 percent. In addition,
both feed and sludge handling facilities will have to be expanded in
proportion, with attendant capital and sludge disposal cost increases.
Similar to waste carbide lime, savings in reagent purchase costs will be
balanced against these increases.
4.4.2 Process Performance
Table 4.4.5 presents a summary of bench-scale performance data on the
lime neutralization of six acidic manufacturing wastes conducted by
Faust.13 Both high calcium and dolomitic quicklimes were examined in
various dosages to evaluate effects on final effluent pH and sludge
characteristics. The results showed that while dolomitic formed up to
44 percent less sludge (on a dry weight basis), it was less effective in
'treating concentrated acids in a timely manner. For example, after treating a
paint manufacturing waste stream of 12,910 ppm mineral acidity for 6 minutes,
dolomitic lime only achieved a final pH of 4.5. Conversely, when high calcium
quicklime was applied to dye manufacturing waste containing 10,410 ppm of
mineral acidity, a final pH of 9.2 was reached in the same time frame.
Table 4.4.6 summarizes the operating characteristics and performance of
an automatic, continuous two-stage neutralization system used by a
manufacturer of automobile bumpers.14 The wastewater flow to the plant was
4-85
-------
TABLE 4.4.4 - Elemental and anion variation in IT.S. cement kiln dust, ug/g
Element or anion
Al
As
Ba..
Be
Bi
Ca.. •..-..
Co
Cr . i . .
Cu. ...............
Fe............. ...
Hr .... .^.i..
Mo
-------
TABLE 4.4.3 SUMMARY OF LIME-SLURRY NEUTRALIZATION DATA
Parameter
PH
Mineral Acidity (ppra)
Type of Lime
Lime Dosage (g/L acid)
pll (6 min)
oo % Sludge8 (Volume)
Dry Sludgeb (g/L)
% Dry Solids
(Neutralization Mix)
% Dry Solids
(Settled Sludge)
Dye
0.5
30,500
Dolomite
Quicklime
17.3
3.9
38.0
21.4
1.9
5.7
Pharma-
ceutical
0.7
19,900
Dolomite
Quicklime
11.9
4.1
22.4
12.1
1.1
5.4
Type of
Paint
1.5
12,910
Dolomite
Quicklime
10.0
4.5
33.0
13.7
1.25
4.16
waste
Paint
1.0
12,430
High
Calcium
Quicklime
8.2
10.8
38.4
21.4
1.98
5.6
Chemical
1.0
12,530
High
Calcium
Quicklime
7.8
10.2
24.0
15.9
1.5
6.7
Dye
1.5
10,410
High
Calcium
Quicklime
7.9
9.2
17.0
14.7
1.38
8.63
Acid Volume » 500 ml.
Acid Temperature » 25°C.
aExpressed as percent of original acid volume after one hour of sedimentation.
"From neutralization of one liter of acid.
Source: Reference 13
-------
TABLE 4.4.6 FULL-SCALE AUTOMATIC LIME NEUTRALIZATION SYSTEM CHARACTERISTICS
Parameter
Unit
Value
Line Type
Stone Size
Influent Feed Rate
Slaker Feed Rate
Slaker Tenerature
Slurry Make-Up Ratio
Dry Solid*
Concentration
Slurry Holding
Ibs/hr lime
°C
H^/Lime
High calcium quicklime
6.3 - 63
100,000 - 125,000
750
57 - 68
3.5
10
Capacity
Influent pH
Effluent pH
gals
S.U.
S.U.
400
2-.0
6.3 - 6.7
Source: Reference 14
4-88
-------
100,000 gal/hour, with a peak capacity of 125,000 gallons. Approximately
11,500 Ibs of sulfuric acid (spent pickle liquor) were discharged to the
system daily, primarily in the form of intermittent, concentrated (2 Ibs
acid/gal) bath dumps. To prevent reagent wastage due to acid surges, the
reactors were arranged in series and provided with high-speed agitation. A
10 percent lime slurry is added into the first tank (2,500 gallons), regulated
by an automatic pH probe. This is followed by a second 2,500 gallon agitated
reactor and a 6,000 gallon tank where the reaction mixture is allowed to
settle. The supernatant flows over a weir into a receiving sewer. While
solids are collected and concentrated by vacuum filtration prior to
incineration.
A third case study, conducted by Mooney et al., focused on two lime
neutralization systems for the treatment of acidic cooling waters from the
production of phosphoric acid in the fertilizer industry. One system
(Plant A) treated highly-acidic process cooling water, and other (Plant B)
treated reduced strength seepage waters.
At Plant A, use of a combination of calcium carbonate and lime was
selected, due to the high acidity, to reduce chemical costs. First-stage
facilities were newly constructed, including lime and calcium carbonate.
storage and feed systems, reactors-, and an earthen-basin thickener clarifier
of 48.8-meter diameter. Calcium carbonate was added at a fixed dosage in the
first reactor, and lime was added to the second reactor based on pH control to
approximately pH 5.5. Sludge was pumped from the collection sump directly to
the gypsum-slurry area for disposal in the gypsum stack. Overflow from the
thickener clarifier was partially reused for the slurry of calcium carbonate
and lime, and the remainder was pumped to the second-stage system.
An original liming station was revamped for use as second-stage
facilities, where lime was added based on pH control to about pH 9.5. An
existing plate settling device was relocated to provide clarification.
However, as predicted from pilot testing, only 30 to 40 percent of the
overflow nameplate capacity could be achieved due to the gelatinous nature of
the solids. At very high polymer dosages, plugging of the plates occurred.
Thus, settling and sludge storage was conducted in an existing 4050-m
pond. Pond overflow is monitored for effluent limits, and settled sludge was
periodically transferred to the gypsum stack.
4-89
-------
At Plant B, lime was added in the first and second stages to about
pH 5.5 to 6.0 and pH 9.0 to 9.5, respectively. Settling was conducted in
identical 27.5 meter diameter earthen-basin thickener clarifiers, and sludge
was separately pumped from each to the gypsum-slurry area for disposal in the
gypsum stack. No reuse of effluent was practiced, and the effluent was
monitored for permit limits. An existing cooling pond was modified to serve
as an equalization/storage basin to provide consistent influent flow and
strength. This system can also treat direct cooling pond water at reduced
rates due to higher strength. Table 4.4.7 contains a summary of nominal
design criteria, while Table 4.4.8 contains typical operating data.
These studies demonstrated that two-stage lime treatment of phosphoric
acid process cooling water was capable of achieving EPA guidelines for
fluoride (25 mg/L) and phosphorus (35 m/L). Major advantages of the two-stage
method are minimum lime requirements, separate handling of readily dewaterable
first-stage solids, and effective removal of numerous other constituents.
Addition of calcium carbonate in conjunction with lime for treatment of
high-acidity cooling water allows optimisation of chemical efficiency,
minimizes sludge protection, and .ensures the proper first stage pH for optimum
fluoride removal.
4.4.3 Process Costs
Table 4.4.9 summarizes process costs developed for the construction and
operation of a continuous, high calcium, hydra ted lime neutralization system.
Equipment sizing and operational cost were based mainly on precepts presented
in the two previous costing sections. ' ' However, based on the
increased reactivity of lime as compared to mutual neutralization or
limestone, the 400 gph reactor system was assumed to be a single-stage
reinforced concrete reactor sized for 30 minutes of retention time. In
addition, both the 2,500 and 3,500 gph systems were assumed to be two-stage,
reinforced concrete reactors sized for a 15-minute retention time in the first
stage and 30 minutes in the second stage.
The cost of the hydrated lime slurry feed system is based on
Figure 4.4.3. The equipment includes a slurry tank, pH control, recycle loop,
electrode assembly, signal transmitter, pH recorder-controller, controller
4-90
-------
TABLE 4.4.7 NOMINAL DESIGN CRITERIA FOR TWO STAGE LIME NEUTRALIZATION SYSTEM
Parameter
Plant
Flow, nrVmin.
Stage I
CaC03 Reactor Detention, Min
Ca(OH)? Reactor Detention, Min
Thickener-Clarifier Dia., Meters
Stage II
Ca(OH)2 Reactor Detention, Min .
Thickener-Clarifier Dia., Meters
CaC03 Storage, Metric Tons
CaO>3 Feed Capacity, Metric Tons/Hr
CaO Storage, Metric Tons
CaO Feed Capacity, Metric Tons/Hr
Operating Power, kW
3.8
30
30
48.8
30
27.5
30
(4050-m2 Pond)
563
13.6
454
7.3
224
35
27.5
454
7.3
231
Source: Reference 15
4-91
-------
TABLE 4.4.8 TYPICAL OPERATING DATA FOR THE TWO-STAGE LIME
NEUTRALIZATION OF ACIDIC PROCESS COOLING WATER
Parameter
Plant
Influent Flow, m^/min
Raw Water
Fluorine, mg/L as F
Phosphorous, mg/L as P
pH, std. units
CaC03 Kg/m3
CaO Kg/m3
Net Effluent Flow, m3/min
Treated Water
Fluoride, mg/L as F
Phosphorous, mg/L as P
pH, std. units
3.8
11,1000
6,730
1.5
10
24
2.7
1 - 14
1-15
9.2 - 9.7
3.4
1,460
2,230
2.6
—
6.8
2.7
6 - 12
5-70
9.1 - 9.6
Source: Reference 15
4-92
-------
TABLE 4.4.9 CONTINUOUS HYDRATED LIME NEUTRALIZATION TREATMENT COSTS
Capital investment
Collection sump
Piping system
Reactor system
Feed system
Flocculation/clarification
Sludge storage
Filter press
Total capital cost
Annual ized capital
Operating cost3 ($)
Taxes and insurance (7%)
Maintenance (52)
tabor ($20/br)
Sludge disposal C$200/ ton) b
R»Acrpnt cost (&40/ton)
lro»|^~ll V WwOU \*^wf ww»*^
Total cost/year ($)
Cost/ 1,000 gal ($)
400
3,100
2,838
17,000
74,000
22,000
8,000
11,000
137,938
24,415
1,709
1,221
24,000
64,514
4,615
102,474
126
— ^— — — — ^— ^
Flow rate (gph)
2,500
3,600
2,838
36,000
125,000
32,000
8,000
15,000
222,438
39,371
2,756
1,969
24,000
435,809
28,892
532,798;
89
3,500
3,600
2,838
38,000
150,000
40,000
10,000
18,000
262,438
46,452
3,252
2,323
24,000
565,499
40,449
681,975
81
•Does not include utilities.
not include gypsum refund.
Source: Adapted from References 6, 16, 17, and 18 using July |l986 CPI index.
4^93
-------
600
,-» 400
200
100
80
60
40
M
o
u
o
o
O
20
X/
WITH pH CONTROL
SYSTEM ,
WITHOUT pH CONTROL
•SYSTEM
10
2O
40 60 80 100
200
400 600 800 1000 2000
4000
Cd(OH)2 FEED (Kg/h)
Figure A.4.3. Investment costs for hydrated lime feed systems.
Source: Reference 16.
-------
valves, instrument panel, miscellaneous hardware and installation. The
reagent demand for the neutralizer system was based on a basicity factor of
0.710 and a 10 percent insoluble feed. Sludge handling equipment sizing was
estimated on a sludge generation factor of 1.322 Ibs of dry solids/pound of
sulfuric acid neutralized.
As with limestone, hydrated lime sludge generation and subsequent
disposal costs are a significant portion of the overall treatment
application. In the advent of a land disposal ban, the increased costs and
liability in disposing of wastewater treatment sludges may outweigh any
possible benefit gained from the reduced chemical costs associated with
calcium based reagent systems. Therefore, when evaluating neutralization
.systems careful consideration of options such as acid recovery or reagents
with soluble end products should be performed to reduce total sludge volumes.
4.4.4 Process Status
Lime slurry treatment is a widely used technology for neutralizing dilute
and concentrated acidic waste streams. Its ability to treat a wide variety of
manufacturing waste streams has been well demonstrated in bench, pilot, and
full-scale systems. Environmental impacts can result from emissions during
the neutralization process and the production of large volume of potentially
18
'hazardous sludge. Exit gases can be scrubbed by using a control system,
however, sludge reduction methods (seeding, dilution, vacuum filtration,
19 20
etc.), ' have only partially offset the problems associated with sludge
generation. Therefore, new methods of sludge disposal and reduction and
recycle/reuse options (such as agricultural liming) should be considered.
The advantages and disadvantages of lime-slurry neutralization are
summarized in Table 4.4.10.
4-95
-------
TABLE 4.4.10 ADVANTAGES AND DISADVANTAGES OF LIME-SLURRY NEUTRALIZATION
Advantages
Proven technology with documented neutralization efficiencies
No temperature adjustments normally necessary
Modular design for plant expansion
Can be used in different configurations
Able to overneutralize and precipitate metals lowering space requirements
Low-cost and widely available reagent
Disadvantages
Relatively high capital and operating costs when operating slaking
equipment
Forms voluminouft sludge product
Requires large and moisture-proof storage facilities when using quicklime
Will foul and clog process equipment unless continually recirculated
Source: References 1, 3, 4, 12, 21
4-96
-------
REFERENCES
1. Camp, Dresser, and McKee. Technical Assessment of Treatment Alternatives
for Wastes Containing Corrosives. Contract No. 68-01-6403, September
1984.
2. Besselievre, E. B. The Treatment of Industrial Wastes. McGraw-Hill Book
Company, New York, NY. 1967.
3. Boynton, R. S. Chemistry and Technology of Lime and Limestone.
Interscience Publishers, New York, NY. 1966.
4. Cushnie, G. C. Removal of Metals from Wastewater: Neutralization and
Precipitation. Pollution Technology Review, No. 107. Noyes
Publications, Park Ridge, NJ. 1984.
5. Kirk-Othraer Encyclopedia of Chemical Technology. Vol. 14, 3rd Edition,
John Wiley & Sons, New York, NY. 1981.
6. U.S. EPA. Process Design Manual: Sludge Treatment and Disposal.
EPA-625/1-79-011. September 1979.
7. Mace, G. R. Lime vs. Caustic for Neutralizing Power. Chemical Engineering
Progress. August 1977.
8. HSU, D. Y. , et al. Soda Ash Improves Lead Removal in Lime Precipitation
Process. 34th Industrial Waste Conference, Purdue University. 1977
9. USEPA. Characterization of Carbide Lime to Identify Sulfite Oxidation
Inhibitors, Interagency Energy/Environment RPD Program Report. U.S.
EPA-600/7-78-176. September 1978.
10. USEPA. Disposal and Utilization of Waste Kiln Dust From Cement
, Industry. U.S. EPA-670/2-75-043. May 1975.
11. U.S. Department of Transportation. Kiln Dust-Fly Ash Systems for Highway
Bases and Subbases. Report No. FHWA/RD-82/167. 1983.
12. Haynes, B. W., and G. W. Kramer. Characterization of U.S. Cement Kiln
Dust. Bureau of Mines Information Circular No. 8885. 1982.
13. Faust, S. D. Sludge Characteristics Resulting from Lime Neutralization of
Dilute Sulfuric Acid Wastes. 13th Industrial Waste Conference, Purdue
University. 1958.
14. Hugget, et al. Automatic Continuous Acid Neutralization. 23rd Industrial
Waste Conference, Purdue University. 1968.
4-97
-------
15. Mooney, G. A., et al. Two-Stage Lime Treatment in Practice.
Environmental Progress, Vol. 1, No. 4. November 1982.
16. MITRE Corp. Manual of Practice for Uastewater Neutralization and
Precipitation. EPA 600/2-81/148. August 1981.
17. Peters, M. S., and K. D. Timmerhaus. Plant Design and Economics
for Chemical Engineers. McGraw-Hill Book Company, New York, NY. 1980.
18. U.S. EPA. Reducing Water Pollution Control Costs in the Electroplating
Industry. U.S. EPA-625/5-85/016. September 1985.
19. Levine, R. Y., et al. Sludge Characteristics of Lime Neutralized Pickling
Liquor. 7th Industrial Waste Conference, Purdue University. 1952.
20. Dickerson, B. W., et al. Neutralization of Acid Wastes: Industrial
Engineering Chemistry 42:599-605. 1950.
21. Arthur D. Little Inc. Physical, Chemical, and Biological Treatment
Techniques for Industrial Wastes. U.S. EPA SW-148. November 1976.
4-98
-------
4.5 CAUSTIC SODA TREATMENT
4.5.1 Process Description
Pure anhydrous sodium hydroxide (NaOH) or caustic soda is a. white
crystalline solid manufactured primarily through the electrolysis of brine.
Caustic soda is a highly alkaline, water soluble compound especially useful in
reactions with weakly acidic materials where weaker bases such as sodium
carbonate are less effective.1 It is also useful in the precipitation of
heavy metals and in neutralizing strong acids through the formation of sodium
salts. Table 4.5.1 lists the properties of pure anhydrous sodium hydroxide.
Although available in either solid or liquid form, NaOH is almost
* *}
exclusively used in water solutions of 50 percent or less. The solution is
marketed in either lined 55-gallon drums or in bulk; i.e., tank car or truck.
As a solution, caustic soda is easier to storej handle, and pump relative to
either lime or limestone. In comparison to lime slurries, caustic soda will
not clog valves, form insoluble reaction products, or cause density control
problems. However, when sodium hydroxide is stored in locations where the
ambient temperature is" likely to fall below 12°C, heated tanks should be
provided to prevent reagent freezing.
After lime, sodium hydroxide is the most widely used alkaline reagent for
acid neutralization systems. Its chief advantage over lime is that, as a
liquid, it rapidly disassociates into available hydroxyl (OH-) ions. Holdup
time is minimal, resulting in reduced feed system and tankage requirements.
3
Caustic soda's main disavantage is reagent cost. As a monohydride, in
neutralizing diprotonated acids such as sulfuric, two parts base are required
per part of acid neutralized. In contrast, dihydroxide bases such as hydrated
lime, only require one part base per part of acid neutralized:
This increase in reagent requirements combined with a higher cost/mole
(approximately five times that of hydrated lime), makes caustic soda more
expensive on a neutralization equivalent basis. Generally in high volume
'applications where reagent expenditures constitute the bulk of operating
expenses, lime is the reagent of choice. However, in low volume applications
where low space requirements, ease of handling, and rapid reaction rates are
4-99
-------
TABLE 4.5.1 PHYSICAL CONSTANTS OF PURE SODIUM HYDROXIDE
CAS Registry No. 1310-73-2
Molecular Weight 39.99
Specific gravity 20°/4°C 2.13
Melting Point °C 318
Boiling Point °C 1390
Index of Refraction 1.358
Latent Heat of Fusion (Cal/g) 40.01
Beat of Formation (kcal/nol)
Alpha Form 100.97
Beta Form 101.96
Transition Temperature °C 299.6
Solubility at 20°C (g/100 g water) 109
Source: Reference 2
A-100
-------
the deciding factor in reagent selection, caustic soda is clearly superior.
Also, in any system where sludge disposal costs will be high, caustic soda
will compete more favorably with lime.
The higher solubility of NaOH in water (approximately 100 times that of
lime at 25°C) reduces or eliminates the need for complex slaking, slurrying,
or pumping equipment. In a typical system, caustic is added through an
air-activated valve controlled by a pH sensor.4 Reagent is demanded as long
as the pH of the acidic waste stream remains below the controller setting.
Agitation is provided by a mechanical mixer to prevent excessive lag time
between the addition of the reagent and the first observable change in the
effluent pH. The neutralized solution is then pumped to a large settling tank
for liquid/solid separation.
Caustic soda is a particularly versatile alkali reagent that can be used
in either over or under-neutralization applications. The neutralization
reaction is typically carried out under standard operating temperatures and
pressures. The reaction is almost instantaneous since caustic soda reacts
vigorously with water. At concentrations of 40 percent or greater, the heat
generated by dilution can raise the temperature above the boiling point.
Handling precautions, are required when performing dilution or other reagent
handling since even moderate concentrations of NaOU solution are highly
corrosive to skin.
Process configurations for caustic soda treatment are a function of waste
type, volume, and raw waste pH level and variability. For example, the
neutralization of concentrated acidic waste streams with low dead times
depends on pH as follows: one reactor system for feeds with pH ranging
between 4 and 10, a reactor plus a smoothing tank for feeds with pH
fluctuations of 2 and 12, and two reactors plus a smoothing tank for feeds
with pH less than 2 or greater than 12.5 Retention times vary with the rate
of reaction and mixing, however, 15 to 20 minutes appears to be optimal for
complete neutralization in most systems. The interval between the addition
of sodium hydroxide and the first observable change in effluent pH (dead time)
should be less than 5 percent of the reactor residence time in order to
naintain good process control. A summary of typical operating parameters
is provided in Table 4.5.2.
4-101
-------
TABLE 4.5.2. SODIUM HYDROXIDE NEUTRALIZATION: SUMMARY OF TYPICAL
OPERATING PARAMETERS
Parameter
Sodium Hydroxide
Concentration
Dead Time
Influent pH
Retention Time
Batch Treatment
Throughput
Continuous Treatment;
Throughput
Suspended Solids
Storage Temperature
(40 - 50Z NaOH)
Unit(s) Operating range
Z NaOH 12 - 50
Z Retention 3-10
Time
H20 - 7.0 0.1-6
Min 5-30
gal/min 1-20
gal/min >15
Weight Z 3-10
°C 12 - 20
Ideal range
40 - 50
3-5
2-4
15 - 20
20
20
10
16 - 20
Source: References 3, 5, 7, 8
4-102
-------
A typical caustic system is designed to add most of the reagent in a
preliminary neutralization stage, while a second stage acts as a smoothing and
finishing tank. In this manner, the second reactor is able to compensate for
pH control overshoots or concentrated batch dumps which may temporarily
overwhelm the primary neutralization system.
Overshoot is due primarily to the lack of sodium hydroxide solution
buffering capacity. For example, Figure 4.5.1 illustrates the titration curve
for the neutralization of a ferric chloride etching solution (pH 0.5) with a
5 Molar caustic soda solution. The steep slope of the titration curve
beginning at pH 2.0 combined with a strong demand for alkali prior to that
point, often make over- or under- correction unavoidable. For continuous
neutralization applications of greater than 20 gpm, pH control in the portion
of the titration curve which is nearly vertical (between pH 2.0 and 9.0) is
achieved in a second reactor to prevent excess reagent usage or effluent
discharge violations.
«
Sodium carbonate (soda ash) is an alternative sodium alkali for acidic
wastestreams lacking buffering capacity such as deionized acid-bath
rinsewaters. The use of sodium carbonate (a weak base) with strong acids,
such as sulfuric, will impart a buffer to the wastewater stream, thereby
facilitating pH control within the neutral range. These buffering reagents
will produce a smaller change in pH per unit addition than comparable
unbuffered, strong bases such as high calcium lime or caustic soda. This
phenomenon can be seen in Figure 4.5.2, which illustrates the neutralization of
a 1 percent sulfuric acid solution with caustic soda and soda ash. A small
incremental addition of caustic soda caused the pH to change from 2 to 11
standard units. Alternatively, approximately 3 times the quantity of soda
ash, resulted in a modest pH change from 6 to 9 units.
Due to its carbonate-based reaction mechanism, the neutralization of
acidic rinsewaters with soda ash (as with limestone) proceeds at a much slower
pace than comparable hydroxide-based reagent systems such as lime or caustic
soda. Accordingly, continuous flow reactors must be sized to provide a
minimum of 45 minutes hydraulic retention in each stage. In addition, soda
•sh is commercially available only in a dry form. Consequently, onsite batch
mixing and solution preparation facilities, similar to those of hydrated lime,
are manadatory when using this chemical as a neutralizing agent. The
4-103
-------
PH
0 TRIAL I
A TRIAL Z
0 -10 20 30 40
VOLUME OF S MOLAR NaOH (ml)
50
Figure 4.5.1 Neutralization of ferric chloride etching waste by
sodium hydroxide
Source: Reference 9
4-104
-------
13.0
NoOH 50%
e
o
2.0
1.0 -
4.0 5.O 6.0
GRAMS OF REAGENT AOOEO
7.0
8.0
9.0
No CO
2 3
10.0
Figure 4.5.2
Titration curve for the neutralization of a 1%
sodium carbonate.
Source: Reference 6
solution with sodium hydroxide and
-------
solubility of*soda ash also limits its use since a chemical solution feed
strength of only 20 percent by weight can be maintained at ambient
temperatures without salt recrystallization. Continuous mixing of the
prepared solution is recommended to maintain homogeneity.
An advantage of soda ash is the generation of less sludge since
sodium—based end products are more soluble than calcium-based products.
However, sodium-based sludges do not filter as readily or to as high solids
4
content as calcium based sludges. In addition, the clarified liquid
effluent may not be as low in metals content or total dissolved solids as
insoluble end product systems such as lime. All these factors must be
carefully weighed before selecting sodium carbonate or any other alkaline
reagent as a neutralizing agent.
4.5.2 Process Performance
While design constraints such as deadtime, reagent splashing, and pU
control system overshoot will determine minimum system retention time,
parameters such aa influent flowrate and concentration will usually establish
overall system volume and configuration. This is partially as a result of the
availability of caustic soda. pH control systems which rapidly respond to
instantaneous "changes in acid concentration. Thus, the following three case
studies summarized in Table 4.5.3, are intended to illustrate caustic soda
neutralization system design as a function of influent flowrate.
The first application involved two-stage continuous neutralization of
acidic wastewater from semiconductor wafer processing operations.
Rinsewaters and process bath dumps are collected in an equalization sump prior
to the first stage neutralization as shown in Figure 4.5.3. The influent
averages a pH of less than 2.0 and a flowrate of 30 gal/minute. The
i
neutralization system components'include a neutralization tank with pH control
instrumentation, a mixer (1/2 hp) and mounting bracket, one chemical feed tank
with mixer, one chemical metering pump, and an electrical control panel.
In the first neutralization tank (150 gallons), the pH is raised to 5 or
6 with a 3 to 5-minute retention time. In the second neutralization tank
tj
(750 gal), the pH is raised to 9.0 with a 15 to 25-minute retention time.
Upon complete neutralization, the effluent is checked by a final pH probe and
then discharged to the sewer.
4-106
-------
TABLE 4.5.3 SUMMARY OF SODIUM HYDROXIDE NEUTRALIZATION DATA
Parameter
Type of waste
(manufacturer)
Treatment system
Printed circuit3
board
Integrated ciruits'3
(semi-conductors)
Cutting oilsc
(automobile)
Influent pH
Influent Flow rate
(gpm)
NaOH Concentration (Z)
Mode of Operation
2.0
60
45
Continuous
2.0
30
50
Continuous
1.5
700-4000
50
Batch
Process
Sump
Capacity (gal) . 250
Retention Time (Min) 4
Effluent pH 5.0
Primary Neutralization
Capacity (gal) 1500
Retention Time (Min) 25
Effluent pH 7.0
Secondary neutralization
Capacity (gal) 1500
Retention Time (Min) 25
Effluent pH 8.5
••MMHHM^^B^^^"^^*
Sources: a. Reference 8
b. Reference 6
c. Reference 10
150
3-5
5.0-6.0
750
15-25
9.0
500,000
125
6.0-9.0
4-107
-------
O
CO
ACID
ALKALI
ALKALI
fill
7
®
FIRST STAGE
NEUTRALIZATION
TO
FIRST
STAOC
10
SECOND
STAGS
TO
rinsr
3TA«C
SeCOND STAOE
NEUtRALIIAIION
TO
SECOND
STAOC
COUAI IZATIOfl
SUMP AMD DUPLEX
IRAHirCR PUMPS
KEY
LEVEL
INDICATOR
fit RECORDER
LEVEL
COMIROILER
©
|>H CONIROLI.ER
FLOW
HONIIOn
CHEMICAL
mo
(ACID)
CHEMICAL
FEED
(ALKALI)
Figure 4.5.3 Process schematic: two-stage neutralization system.
Source: Reference 6
-------
The second application of caustic soda involves the continuous
neutralization of acidic wastewater from a large printed circuit board
Q
facility. The facility operates on a 24-hour, 7-day/week basis. The "
average influent flow rate is 60 gal/minute with a pH of less than 2.0. The
reagent solution is stored inside in a 3,500 gallon tank at 45 percent
strength to prevent freezing. Reagent usage is approximately 3,000 gallons
each week. The system configuration is very similar to the previous case
study, however, a third stage and increased retention times have been added to
accommodate a flow rate which has essentially doubled. This expanded syste-ji
allows for a greater safety margin in insuring effluent compliance and minimal
reagent wastage. It also provides more flexibility in the event that future
expansion is required. _
The influent collects in a 250 gallon sump where the pH is raised to 5
during a four minute retention time. Primary and secondary neutralization
takes place in two 1,500 gallon tanks (25-minute retention time) and the pH is
raised to 7 and 8.5, respectively. Agitation in provided in all tanks with a
3/4 hp mechanical agitator and reagent is regulated through feedback control.
Following caustic neutralization, liquid/solid.separation is achieved through
two activated carbon filters arranged in series which remove suspended solids
and trace organics. Following separation, the supernatant is discharged to
.the sewer while the absorption columns are backwashed and the resulting
solution is dewatered in a filter press.
The third application consists of the large scale neutralization of
acid-emulsion breaking wastes. The physical plant consists of a
44 ft x 60 ft treatment building, three 500,000 gallon treatment tanks, and
10,000 gallon reagent storage tanks. The waste stream (pH 1 to 2) consists of
an oily waste that has been acidified with sulfuric acid, heated to 120°F, and
then skimmed to remove the oil in a batch emulsion-breaking process. The
remaining contents are clarified with a flocculating agent (alum), neutralized
by the addition of caustic soda or lime, and then discharged to the city sewer
system (pH 6-9). Air for agitation is provided by a six-stage, 60 hp blower
«t 7 psig. Plant air is available for standby through a pressure reducing
Station. When the control panel starts from the bottom for the blower is
activated, a control valve closes to prevent surging.
4-109
-------
The reagent system (Figure 4.5.4) consists of 10,000 gallon liquid
caustic storage tank and a 100 gal/minute recycle pump system. Reagent is
added as needed and overdose is prevented by time interlocking of the reagent
injection operation. Most of the piping is fiberglass reinforced polyester
and all controls are pneumatic.
While the influent flow rate is greater than any of the other case
studies (700-4,000 gpm), the flow from the emulsion breaking system requires
that the neutralization process be carried out in a batch mode.
4.5.3 Process Costs
Table 4.5.4 presents the process costs developed for the construction and
operation of 50 percent sodium hydroxide continuous neutralization systems.
Capital investment and operational costs are based on the methodologies that
were presented in previous sections. ' ' Reactor sizing was based on
influent flow rate. The individual treatment configurations for the 400,
2500, and 3,500 gpm systems consisted of respectively; a single-stage reactor,
a primary reactor in series with one finishing tank, and a primary reactor in
series with two finishing tanks.
Reagent demand was estimated to be 1.18 Ibs of 50 percent caustic
solution per pound of suIfuric acid neutralized. The reagent usage ratio is
based on a basicity factor of 0.69 and a 50 percent molar excess of water.
Sludge generation was assumed to be negligible since the neutralization of
pure sulfuric acid will result in the formation of a soluble
(44 grams/100 milliliters) sodium sulfate dihydrate reaction product.
However, the presence of metallic species or other reducible compounds in the
wastestream would necessitate the additional investment of sludge handling
equipment. Table 4.5.5 presents sludge generation factors for seven common
metallic contaminants.
4.5.4 Process Status
Sodium hydroxide neutralization is widely applied technology for treating
corrosive waste streams. The rapid reaction rate of NaOH treatment
facilitates, the use of smaller tanks and shorter retention times. Also,
4-110
-------
AGITATING AIR
EFFLUENT FROM
ELPON AVENUt
AXIE PLANl
TREATMENT
DECANT TANK
NO. I REMOTE NO. I LIF I PROCESS LIFT
LIFT STATION
CHEMICAL RECYCLE SYSTEM
CHEMICAL RECYCLE SYSTEM
TREATED WATER TO CITY SEWER
LIQUID CHEMICALS
TRUCKS UNLOADING
CONTROLLER
FOR CHEMICAL INVENTORY
AND TITRATION INTO
CHEMICAL RECYCLE SYSTEM
Figure 4.5.4 Process schematic sodium hydroxide batch treatment system.
Source: Reference 10
-------
TABLE 4.5.4. CONTINUOUS CAUSTIC SODA NEUTRALIZATION TREATMENT COSTS
Capital investment (J)
Collection sump
Piping system
Reactor system
Total capital cost
Annualized capital
Operating costsd ($)
Taxes and insurance (7Z)
Maintenance (5Z)
Labor ($20/hr)
Reagant cost ($205/ton)
Total cost/yr ($)
Cost/1000 sal
400
3,100
2,838
19,000a
24, 938
4,414
309
221
24,000
35,493
64,439
67
Flow rate (gph)
2,500
3,600
2,838
48,UOOb
54,438
9,635
675
482
24, 000
221,908
256, 700
43
3,500
3,600
2, 638
74,000C
80,438
14,238
997
712
24, 000
311,146
351,093
42
'Includes single stage continuous reactor with feed system.
^Includes single stage continuous reactor in series with 25 minute finishing
reactor with feed systems.
cInclude single stage continuous reactor in series with two 25 minute
finishing reactors with feed systems.
dDoes not include utilities.
Source: Adapted from References 5, 11, 12 using August 1986 CPI index cost
data.
4-H2
-------
TABLE 4.5.5 SODIUM HYDROXIDE SLUDGE GENERATION FACTORS
Ib dry solids generated
Metal ion Ib of metal precipitated
Cr 1.98
Ni 1.58
Cu 1.53
Cd 1.30
Fe 1.61
Zn 1.52
Al 2.89
Source: Reference 12
4-113
-------
sodium hydroxide minimizes the usual requirement to install large flow/pH
equalization basins due to the flexibility of proportional control. It also
results in lower quantities of sludge generated relative to calcium based
reagents. However, these sludges may be more difficult to dewater. The
advantages and disadvantages of caustic soda neutralization are summarized in
Table 4.5.6.
4-114
-------
TABLE 4.5.6 ADVANTAGES AND DISADVANTAGES OF CAUSTIC SODA NEUTRALIZATION
Advantages
- Proven technology with documented neutralization efficiencies
'.'- Strong base with rapid reaction rate
- Smaller tanks and retention times than comparative reagents
- Inventory and storage handling procedures are less complicated due to liquid
form
^Storage does not require continuous agitation to maintain homogeneity
- Does not require complex slaking or slurrying equipment
- Produces more soluble by-products in low pH applications
Disadvantages
• Chemical costs are significantly higher ($205/ton vs. $46/ton for
bydrated lime)
-Does.not impart any buffering capacity to industrial waste streams •
• Close attention must be given to the design of the pU control
-Caustic soda precipitation will result in a fluffy gelatinous floe
increasing the size of the clarification chambers and sludge dewatering
equipment.
-Cannot effectively precipitate sulfate waste streams due to solubility of
" sodium sulfate. ;
Source: References 1, 2, 3, 6, 13
4-115
-------
REFERENCES
1. Cushnie, G.C. Removal of Metals from Wastewater: Neutralization and
Precipitation. Pollution Technology Review, No. 107, Noyes Publication,
Park Ridge, NY. 1984.
2. Kirk-Othmer Encyclopedia of Chemical Technology. Vol. 1, 3rd. Edition,
John Wiley & Sons, New York, NY. 1981.
3. Camp, Dresser, and McKee. Technical Assessment of Treatment Alternatives
for Wastes Containing Corrosives. Contract No. 68-01-6403. September
1984.
4. Mace, G.R. Live vs. Caustic for Neutralizing Power. Chemical
Engineering Process. August 1977.
5. MITRE Corp. Manual of Practice for Wastewater Neutralization and
Precipitation. EPA-600/2-81-148. August 1981.
6. Mabbett, Cappacio & Associates. Industrial Wastewater Pretreatment
Study: Preliminary Engineering Design Report. January 1982.
7. Hoffman, F. How to Select a pH Control System for Neutralizing Waste-
Acids. Chemical Engineering. October 30, 1972.
8. Leedberg, T., Honeywell Corporation. Telephone conversation with
Steve Palmer, GCA Technology Division, Inc. August 13, 1986.
9. Okey, R.W. Neutralization of Acid Wastes by Enhanced Buffer. Journal of
Che Water Pollution Control Federation. July 1978.
i
10. Hoad, J.G. How One Pollution Problem was Solyed - Simply Industrial
Waste. Novenber 1971. •
I
11. Peters, M.S., and K.D. Timmerhaus. Plant Design and Economics for
Chemical Engineers. McGraw-Hill Book Company,'New York, NY. 1980.
i
12. U.S. EPA. Economics of Wastewater Treatment Alternatives for the
Electroplating Industry: Environmental Pollution Control Alternatives.
EPA-625/5-79-016. 1979. jj
M
13. Arthur D. Little, Inc. Physical, Chemical, and1 Biological Treatment
Techniques for Industrial Wastes. U.S. EPA SW-148. November 1976.
4-116
-------
4.6 MINERAL ACID TREATMENT
4.6.1 Process Description
Mineral acid treatment of corrosive waste streams is the most widespread
of neutralization processes for alkaline wastes. The acid-base reaction is
essentially the analog of those discussed in previous sections; i.e., a proton
(H ) is donated by the neutralization reagent. The two primary mineral acid
reagents are sulfuric and hydrochloric which are characterized by their highly
reactive nature, complete miscibility with water, and rapid disassociation
rates. In concentrated form, application may result in the generation of an
acid mist or toxic fumes. Therefore, the choice of an acidic reagent is
typically based on ease of handling, as well as cost per unit basicity, and in
some cases such as food processing; end product characteristics.
Sulfuric acid (H,SO,) is the most widely used of all mineral acid
1
reagents . Ease of manufacture, diprotonated reaction chemistry and
concentrated nature, combine to make it the least expensive mineral acid on a
neutralization equivalent basis. It is supplied in concentrated liquid form,
(93 to 98 percent), is highly reactive, strongly hygroscopic and presents a
burn hazard to personnel. Dilute solutions are highly corrosive to iron
and steel, whereas concentrated solutes (greater than 93 percent) are not
corrosive. Protection from freezing during storage and transport is
required since sulfuric acid exhibits a maximal freezing point of 8°C
at a concentration of 85 percent. In addition, pH control overshoot and
fuming characteristics frequently require diluting the acid to 30 percent
concentration prior to application . In diluting operations, the acid
should be slowly added to the water, with provisions for agitation, adequate
ventilation and protective clothing. A summary of sulfuric acid physical data
is provided in Table 4.6.1.
When sulfuric acid is uneconomical or otherwise inapplicable as a
neutralization reagent, hydrochloric acid (HC1) is often used as a
substitute. Hydrochloric (Muriatic) acid is supplied in aqueous solutions
of 35 to 37 percent acid. Although it is somewhat more reactive than
sulfuric, the most concentrated commercial grade contains at least 63 percent
water, increasing transportation costs and limiting most major uses to a
4-117
-------
TABLE 4.6.1 SULFUR1C ACID PHYSICAL PROPERTIES
Chemical name
Sulfuric acid
Reagent grade concentration
CAS Mo.
Formula
Molecular weight
Boiling point
Melting point
Specific gravity
Nonvolatilea (max) Z
As (max) ppa
SC<2 (max) ppm
Fe (max) ppm
Nitrate (max) ppm
AFHA turbidity (max)
AFHA color (max)
78 - 96.52
7664-93-9
98.08
310°C
-14 to -9°C
1.842
0.025 - 0.05
1.0 - 2.5
40 - 80
50 - 100
5-2.0
100 - 150
100 - 200
Source: Reference 2
4-118
-------
radius of 300 to 500 km from the producing source . In addition, its higher
unit cost and monoprotonated reaction chemistry result in an overall reagent
cost approximately double that of sulfuric acid on a neutralization equivalent
basis. Hydrochloric acid is highly disassociated (a 10,000 mg/L HC1 solution
will result in pH of 0.9), and extremely corrosive, attacking most metals
through surface dissolution. Commonly used plastics and elastomers are
recommended as materials of construction when designing neutralization systems
using hydrochloric acid as reagent.
The main advantage of using HC1 is that it generates soluble end
products, However this attribution may not be beneficial in some cases, since
it may cause the waste to exceed dissolved solids and metal effluent
standards. As with sulfuric, hydrochloric acid will react vigorously with
water, sometimes evolving an acid mist which can destroy the mucous membrane
and cause choking, coughing, headache and dizziness. In addition,
hydrochloric acid will decompose in the presence of heat into toxic hydrogen
chloride gas. Therefore, caution must be exercised when storing or handling
concentrated hydrochloric acid to minimize splashes, spills or mist
generation. A summary of hydrochloric acid physical property data is provided
in Table 4.6.2.
* Conventional mineral, acid treatment systems dispense a highly reactive
aqueous reagent and are, therefore, similar to those of caustic soda.
Equipment consists of chemical storage and reaction vessels, mixers,
chemical-feed metering pumps or valves, and a pH control system. Typically
the reagent storage vessel is constructed of fiberglass reinforced plastic
(FBP) and sized for one week to one months supply, depending on reagent usage
and/or reagent delivery schedules. For low volume systems, the reagent can
be netered directly from the barrel to the reaction system. Reaction vessels
can be constructed of FKP or reinforced concrete, depending on corrosion
resistance and flexibility for operation and expansion. Since mineral acids
disassociate rapidly into solution, reaction vessels are generally sized for
15 to 20 minutes of retention time.
Mixers should be of the type that generate a top-to-bottom flow pattern
with fluid redistribution outward from the center to the walls of the
tank1. This helps avoid "short circuiting", in which the wastewater exits
the tank without sufficient dispersion and treatment. Metering pumps should
4-119
-------
TABLE 4.6.2 HYDROCHLORIC ACID PHYSICAL PROPERTIES
Trade name Muriatic acid
Chemical name Hydrochloric acid
Reagent grade concentration 35 - 37Z
CAS Ho. 7647 - 01-0
Formula HC1
Molecular weight 36.46
Boiling point 110°C
Melting point —
Specific gravity 1.1885
Evaporation rate
(Butyl acetate " 1) 10
Vapor pressure (20°C) 160 mm Hg
Vapor density (air - 1) 1.25
Percent volatiles by volume 37
pH 1.0 N (aqueous) 0.1
0.1 N (aqueous) 1.1
0.02 N (aqueous) 2.02
0.002 N (aqueous) 3.02
Source: Reference 5
4-120
-------
be of the positive displacement type, (large turndown ratio) which allow for
accuracy and flexibility in regulating the chemical dosage to the system. In
systems where precipitation reactions occur, solids adhere to the pH sensing
electrode .and eventually build up a film that interferes with the
measurement. Proper cleaning and maintenance are thus necessary to maintain
accurate pH control.
Due to the similar nature of mineral acid and sodium hydroxide treatment
systems, process equipment and operating parameters for each system are
essentially the same and, in most cases, interchangeable. The primary
difference between these treatment systems is the inclusion of a ventilation
and an acid mist scrubber. The scrubbing system is necessary when
neutralizing concentrated alkaline wastes with strong mineral acids. A
significant amount of heat can be generated raising the temperature of the
neutralization tanks and causing emissions of acid fumes and hazardous gases.
The scrubbing system captures these fumes and neutralizes them in a packed
tower. A caustic scrubbing solution is typically used in the packed tower,
since a lime slurry would rapidly coat the packing medium and plug the
tower.
4.6.2 Process Performance
When utilized as a final treatment, mineral acid systems are used
singularly or in tandem with an alkali reagent depending on the variability of
influent pH. They are also used in pretreatment systems to reduce the
alkalinity of highly caustic waste streams prior to secondary treatment. In
addition, mineral acids are used as an emergency pH control for mutual
neutralization systems in which the concentration and flow rate of the alkali
vaste stream can exceed the neutralization capacity of the acid waste stream.
Finally, although sulfuric acid is usually chosen as the acid reagent, there
are situations for which hydrochloric acid achieved better overall
performance. The following case studies demonstrate the aforementioned used
•ineral acid neutralization systems and provide an example where hydrochloric
acid was found to be a more suitable reagent than sulfuric acid.
The first case study illustrates a neutralization system in which
sulfuric acid was used singularly to partially neutralize a highly alkaline
4-121
-------
waste stream prior to activated sludge treatment (Figure 4.6.1). The pH
reduction vas necessary to prevent toxic shock and maintain optimal growth
conditions for the bio-culture.
The coal-liquefication process wastewater was" segreated into a high
strength caustic waste stream of pH 9.8 to 13.5 and a low strength liquid
waste steam of pH 3 to 11.
Waste characteristics are summarized in Table 4.6.3. The high strength
waste stream, which has been non-cbemically treated for oil removal, is
treated on a fill and draw basis in one of two 10,000 gallon chambers in the
pretreatment tank. The solution is batch catalyzed with manganese sulfate
(400 mg/L) and diffused-air (2,500 SCFM) to oxidize the sulfide groups. The
chambers are used alternately to maintain aeration in one while the other is
being filled. When the sulfide content falls below 20 mg/L, the pH is
adjusted to between 9.5 and 10.5 with sulfuric acid. The pH is controlled in
the moderately basic range since the bio-reaction system tends to drive the pH
to the acid side. The high strength waste stream is then combined with the
low strength waste in a 55,000 gallon, concrete equalization storage tank.
The drawoff point is protected by a baffle to allow a 2.5 ft buildup of sludge
in the bottom. In_addition, a 55,000 gallon emergency storage basin is
provided to hold, and dilute high-strength spills and retain bio-food during
extended shutdowns.
As previously stated, singular mineral acid neutralization systems are
usually only applicable when alkaline influent pH characteristics are fairly
uniform in nature. However, the wide fluctuations in pH normally encountered
in industrial manufacturing wastewaters often necessitate the use of dual
reagent system. The following case study illustrates the most commonly used
form of modular, dual reagent neutralization systems.
In 1986, Alliance Technologies Corporation evaluated a 2-stage,
22,000 gpd wastewater treatment system for the neutralization and
precipitation of a variable pH effluent from a printed circuit board
a
manufacturing operation. The wastewater streams consisted of alkaline
etching solution (pB 11.8 to 13.7), acidic plating rinsewaters (pH 2-3), or a
combination of the two, depending on the process operations. The wastewaters
were neutralized to a final endpoint of 8.5 with 93 percent sulfuric acid or
50 percent sodium hydroxide, as required to facilitate sodium borohydride
4-122
-------
~« — . .
LOW
STRENGTH
WASTE
SUMP
HIGH
STRENGTH
WASTE MnSI
i
l~ " '
* »EAD * CK!M
i™* TALI i/ ^^ oi\in
M TANK MCI,
u> ntK
r*"
H2SO/(
3J| 1 AIR
1 f
'RETREAT
MENT
TANK
NO. 1
NO. 2
1
EQUAL 1 -
ZATION
STORAGE
TANK
DO—
i-CX)-
-oo-
EMERGENCY
HOLDING
BASIN
BYPASS
ACTIVATED CARBON
"A" BIO-REACTOR
(AERATION BASIN)
1
"B" BIO-REACTOR
(AERATION BASIN)
— I r
L SET-
TLER
r
SURFA
— **• oi sen
CLEAR
WELL
GRADIENT
FILTER
' s~*\
pfCLARIFIERj 1
Figure 4.6.1 Coal-liquefaction facility wastewater treatment system.
Source: Reference 7.
-------
TABLE 4.6.3 SUMMARY OF COAL-LIQUEFACATION FACILITY
WASTEWATER CHARACTERISTICS
Flow, gpd
pH
BOD, rng/L
COD, mg/L
Phenolics, mg/L
Oil Grease, mg/L
Sut fides, mg/L
NH3, N, mg/L
High strength
waste stream
4,000-5,000
9.8-13.5
250-2,500
900-5,500
30-800
20-110
70-3,400
75-200
Lov strength
waste stream
5000-21,000
3-11
7-23
NA
0.1-1.0
1-15
1.0
—
Combined3
9, 000-26, 000
9.5-10.5
100-1, 100
400-8,000
20-600
5-70
1.0
15-75
•Streams combined after catalyzed air oxidation of CWS and flow equalization.
Source: Reference 7
4-124
-------
metals reduction. On the day of testing, the influent to the waste treatment
system primarily consisted of dilute alkaline etching solution (average pH
12.28). In addition, the metals content of the waste stream was 790 mg/L of
which approximately 99 percent was divalent copper. After neutralization,
metals reduction occurred through the addition of a 12 percent sodium
borohydride solution. Liquid/solid separation of the precipitate was achieved
through the use of a Memtek ultrafiltration unit with subsequent dewatering in
a Delta unifilter low-pressure filter press. A schematic of the wastewater
treatment system is provided in Figure 4.6.2. Sludge production was
approximately 185 Ibs/day of dry solids, of which approximately 78 percent was
reduced copper. Table 4.6.4 contains a summary of alkaline wastestream and
final effluent characteristics.
The wastewater treatment system consisted of a 250 gallon polypropylene
collection sump and two 825 gallon carbon steel, FRP lined reactors in
series. The wastewater collection sump was automatically emptied into.the
neutralization tank through an adjustable 35 gpm Gould feed pump. The
reaction tanks were vertical, rectangular, and flat bottomed with open tops.
agitation was provided through an electric propeller mechanical mixer with
stainless steel shaft and props, driven by a 2 hp motor. The neutralization
pH controller/recorder was a single position, three set point unit which
controlled pH and sulfuric acid/caustic soda addition. The chemical feed
system consisted of a duplex pump system which drew reagent directly from
55 gallon barrels to minimize hazardous transport. The chemical feed pumps
Here positive displacement, diaphragm type with adjustable output of 0 to
20 gph.
Since industrial processes rarely produce only one type of corrosive
vaste stream, the dual reagent system tends to predominate over the singular
reagent system. Regardless of the alkaline reagent selected, sulfuric acid is
the overwhelming mineral acid of choice, except in specialized cases. An
exception to these general rules of thumb is the neutralization of food
processing wastewaters from caustic peeling operations, as described below.
A peeling operation to remove skin materials is a necessary step in the
processing of numerous fruits and vegetables. ' ' One such process
involved submerging tomatoes for about 30 seconds in a 200°F, 16 to 20 percent
sodium hydroxide bath. In the subsequent washing step which removed the
Softened outer peel and residual caustic, the pH of the washwater ranged from
4-125
-------
t
1
SUMP PUHP
RECIRCULATION STREAM
(CONCENTRATE)
ETCHING
WASTES
PLATING
WASTES
DISCHARGE TO
PERMEATE" SEWER >
REACTION CONCENTRATION
TANK TANK
FILTRATE
OFFSITE
DISPOSAL
SLUDGE
Figure A.6.2. Process schematic showing plating/etching waste treatment system.
Source: Reference 8.
-------
TABLE 4.6.4. SUMMARY OF NEUTRALIZATION/PRECIPITATION TEST DATA
Parameter
pB
Total organic carbon
Total organic halide
Total trace metals:
'Cu
Hi
Pb
Zn
EP toxic metals:
Ar
Ba
Cd
Cr
Pb
Dg
Se
Ag
Stream 3
Influent waste
(mg/L)
12.28
40.0
1.76
786.0
0.055
0.57
3.86
__
—
—
' —
--
——
^*^*
Stream 5
Effluent wastewater
(mg/L)
8.5
36.1
1.75
1.49
0.03
0.10
0.028
__
—
—
—
—
— ™
~
Stream 7a
Sludge3
(ug/g)
NA
184.8
—
780,000
58.7
300
1,430
0.03
0.522
0.002
0.003
1.8
6.0U02
0.04
0.5b
•Results given on a dry weight basis for sludge.
Source: Reference 8.
4-127
-------
13 to 14. The caustic peeling operation processed 35,000 tons of
tomatoes/year, generating approximately 24 gallons/minute of caustic
sludge.11 In addition over 100 gallons/minute of washwater containing
9.1 Ibs of 50 percent sodium hydroxide had to be subsequently neutralized.
Sulfuric acid was originally used as the neutralizing agent (4 Ibs/min),
but resulted in the loss of approximately 120 tons/day of tomato
constituents. By substituting food grade hydrochloric acid for sulfuric acid,
the company expects to realize an estimated 50 percent reduction in sludge
production as shown in Table 4.6.5. The use of hydrochloric results in the
formation of sodium chloride (table salt) since this is permitted as a food
ingredient under the FDA standards (21 CFR 53-10, 53-20, and 53-30), a
substantial fraction of the neutralized waste can be recycled. A flowsheet
for the proposed recovery of the waste peeling sludge is presented in
Figure 4.6.3.
4.6.4 Process Costs
Table 4.6.6 presents the process costs developed for the construction and
installation of two 93 percent sulfuric acid neutralization systems and two
37 percent hydrochloric acid systems. Each mineral acid system was evaluated
on the basis of its ability to cost-effectively neutralize: (1) a 2 percent
sodium hydroxid wastestream; and (2) a 2 percent calcium carbonate
wastestream. The reaction chemistries for each mineral acid neutralization
reaction are presented below:
NaOH + H2S04 * NaS04 + H20 (1)
NaOH + 2 HC1 * NaCl + H20 (2)
CaC(>3 + H2S04 * CaS04 + H20 + C02 (3)
NaC03 * 2 HC1 * CaCl * H20 + C02 . (4)
Capital costs for each of the four waste treatment scenarios have been
adapted from cost information presented previously for sodium hydroxide
neutralization (Section 4.5.3) due to a similarity in process equipment and
operational parameters. Reagent demand was estimated to be 1.32 Ibs of
4-128
-------
TABLE 4.6.5. SUMMARY DATA ON CAUSTIC TOMATO PEELING
OPERATION AND' ACID USAGE
Parameter
Caustic
waste
stream
Reagent system
riCl
Number of caustic peelers
Loading rate, tons/min, each
Caustic solution in peeler, Z
Total season throughput, tons
Caustic usage, tons
Unit caustic usage, Ibs/ton
Acid usage, tons
Sludge production, tons/day
4
0.33
7-10
35,000
121
6.9
53
120
99
64
Source: Reference 11
4-129
-------
sot/o
FOOD S*AO£
PCELING
SLuoee
(pH 11.3 - lg.fi
PULKR - FINISHER
MCVT**UZCQ
JUMX
SURGE TANK
pH ADJUSTMENT TANK
pH PRO8E
STORAGE TANK
neunuuz£o peeLiMS juice
TO MOTB*C*K
Figure 4.63 Flowsheet for proposed recovery of waste peeling sludge.
Source: Reference 11.
4-130
-------
TA.BL.B A. 6.*.
Co
OOMMHUOUS MXHBRAJ. AOX» T*BA*MBNX
ACIDS AS THE NEUTRALIZING AGENT
UBXW* SUUTURItS AMD HYDROCHLORIC
Treatment cost ($)
Waste
stream
2% NaOII
2X NaOll
22 CaC03
2Z CaC03
Reagent
(gph)
U?SOA
400
2,500
3,500
IIC1
400
2,500
3,500
H2S04
400
2,500
3,500
IIC1
400
2,500
3,500
Total Annual! zed
capital capital
24,938
54,438
80,438
24,938
54,438
80,438
59,938
109,938
148,438
24,938
54,438
80,438
4,414
9,635
14,238
4,414
9,635
14,238
10,609
19,459
26,273
4,414
9,635
14,238
Taxes &
insurance
OX)
309
675
997
309
675
997
743
1,362
1,839
309
675
997
Maintenance Labor
(5X) (*20/hr)
221
482
712
221
482
.712
530
973
1,314
221
•482
712
24,000
24,000
24,000
24,000
24,000
24,000
24,000
24,000
24,000
24,000
24,000
24,000
Reagent
costa/yr
15
99
139
31
194
271
16
100
141
31
197
275
,943
,639
,496
,085
,286
,999
.833
,942
,318
,492
,094
,556
Disposal Total
U200/ton) coat/yr 1
44
134
179
60
227
311
104,475 157
626,504 773
877,100 1,071
60
231
315
,887
,431
,443
,029
,078
,946
,190
,240
,844
,436
,886
,503
Cost/
,000 gal
47
22
21
63
38
37
163
129
128,
63
39
38
Source: Adapted from References 13 and 14 using September 1986 CPI index cost data,
-------
93 percent sulfuric acid and 2.47 Ibs of 37 percent hydrochloric acid/pound of
sodium hydroxide neutralized. Similarly, for the 2 percent calcium carbonate
wastestream, reagent demand was estimated to be 1.05 Ibs of 93 percent
sulfuric acid and 2.02 Ibs of 37 percent hydrochloric acid/pound of calcium
carbonate neutralized* The reagent usage ratios were based on stoichiometric
equivalents with a 7 percent molar excess of water for sulfuric acid and a
63 percent excess for hydrochloric.
Based on end-product solubilities and reagent requirements, sludge
generation is negligible in all cases except for the reaction involving
sulfuric acid and calcium carbonate. The reaction product (calcium sulfate
dihydrate) was estimated to generate sludge at approximately 0.75 pounds of
sludge/pound of calcium carbonate neutralized.12,13 xhe 60 percent solids
and disposed of in a secure landfill at a cost of $200/ton. In addition,
the cost of both a sludge storage unit and filter press were include in
total capital cost for this system.
As shown in Table 4.6.6, sulfuric acid is most cost-effective than
hydrochloric on a neutralization equivalent basis in situations where sludge
generation does not occur or is not required; e.g.j.precipitation of heavy
metals. However, in situations where sludge generation-cannot be avoided,
hydrochloric may be the mineral acid of choice.
4.6.4 Process Status
Mineral acid treatment is the most widely used and demonstrated
technology for the neutralization of corrosive alkaline waste streams. Both
sulfuric and hydrochloric have very high acidities, so that quantities
required for neutralization are relatively low in comparison to other
acids. Consequently, reactor volumes and handling/storage facilities are
smaller. Sulfuric acid, being the most widely available and lowest in cost on
a neutralization equivalent basis, is the most prevalent acidic reagent. It
is typically used in combination with an alkali reagent to control pH
fluctuations in both the acidic and alkaline ranges. Hydrochloric acid is
generally used in situations requiring rapid reaction rates and soluble
reaction products.
4-132
-------
The primary environmental impact from the use of these regeants is the
generation of potentially hazardous sludge (sulfuric acid) or generation of an
acid mist or toxic/hazardous fumes (hydrochloric acid). The highly corrosive
nature of mineral acids presents a burn hazard to personnel and increases the
likelihood of a possible catastrophic release during bulk transport or
storage. A summary of both the advantages and disadvantages of sulfuric and
hydrochloric acid are presented in Tables 4.6.7 and 4.6.8, respectively.
4-133
-------
TABLE 4.6.7 ADVANTAGES AND DISADVANTAGES OF SULRJRIC ACID NEUTRALIZATION
Advantages •
- Highly reactive, concentrated acid with rapid disassociation rate
- Widely available and low in cost
- Unlikely to evolve noxious or hazardous fumes
- Proven technology with documented neutralization efficiencies
- Easy to store and can be used in a variety of configurations
Disadvantages
• Strongly bydroscopic and presents burn hazard to personnel
- Highly corrosive to metals when in dilute form
- Will freeze when stored at temperatures below 47°F and concentrations
greater than 85 percent
- Can possibly form voluminous, insoluble end-products
- Can introduce sulfates into the effluent wastestream
Source: References 1, 2, 3, and 6.
4-134
-------
TABLE 4.6.8 ADVANTAGES AND DISADVANTAGES OF HYDROCHLORIC ACID NEUTRALIZATION
Advantages
- Has a more rapid reaction rate than sulfuric -acid
- Will form insoluble end-products and thereby minimize sludge production
- Is a proven technology which can be applied in solution form
- Not as hygroscopic as sulfuric acid
Disadvantages
- Highly corrosive and represents burn hazard
- Higher unit cost than sulfuric acid
- Can evolve both noxious and hazardous fumes
- Can possibly exceed effluent standards due to solubility of reaction
products
- Will decompose in the presence of heat to form hydrogen chloride gas
Source: References 3, 5, and 6.
4-135
-------
REFERENCES
1. Capaccio, R.S., and R. Sarnelli. Neutralization and Precipitation.
Plating and Surface Finishing.- September 1986.
2. Kirk-Othmer Encyclopedia of Chemical Technology. Volume 12, 3rd Edition.
pp. 983-1011. John Wiley & Sons, New York, NY. 1981.
3. Cushnie, G.C. Removal of Metals from Wastewater: Neutralization and
Precipitation. Pollution Technology Review No. 107, Noyes Publication,
Park Ridge, NJ. 1984.
4. Francini, F. Honeywell Corporation. Telephone conversation with
Stephen Palmer, GCA Technology Division, Inc. September 12, 1986.
5. Kirk-Othmer Encyclopedia of Chemical Technology. Volume 22, 3rd Edition,
pp. 226-229. John Wiley & Sons, New York, NY. 1981.
6. Camp, Dresser, & McKee. Technical Assessment of Treatment Alternatives
for Wastes Containing Corrosives. Contract No. 68-01-6403. September
1984.
7. Sapp, J.B. Wastewater Treatment at a Coal-Liquefaction Facility.
Environmental Progress (Volume 2, No. 3). August 1983.
8. Palmer, S. Case Studies of Existing Treatment Applied to Hazardous Waste
Banned from Landfill, Phase II.- U.S. EPA Contract No. 68-03-3243. July
1986. '
9. Graham, R.P. et al. Double-Dip Caustic Peeling of Potatoes. Proceedings
of the Sixth National Symposium on Food Processing Wastes.
EPA-600/2-76-224. December 1976.
10. Schultz, W.G., Graham, R.P., and M.R. Nant. Pulp Recovery from Tomato
Peel Residue* Proceedings of the Sixth National Symposium on -Food
Processing Waste. EPA-600/2-76-224. December 1976.
11. Fernbach, E. et al. Wastewater Management of Hickmott Foods Inc.
Proceedings of the Sixth National Symposium on Food Processing Wastes.
EPA-600/2-76-224. December 1984. 1
I
12. CRC Handbook of Chemistry and Physics. 54th Edition 1973-1974, CRC
Press, Cleveland, Ohio. ;
13. MITRE Corp. Manual of Practice for Wastewater Neutralization and
Precipitation. EPA-600-81-148. August 1981.
>';
14. US EPA Economics of Wastewater Treatment Alternatives for the
Electroplating Industry. EPA 625/5-79-016.
4-136
-------
4.7 CARBONIC ACID TREATMENT
4.7.1 Process Description
Carbonic acid neutralization of alkaline waste streams is a relatively
old, but as of yet, undeveloped treatment technology. As early as 1931,
Curtis and Copson patented a process using a reaction product carbonic acid to
neutralize a cotton waste (Rier liquor) treated with caustic soda. The
inherent problem with carbonic acid treatment is that carbonic acid, a weak
acid disassociates slowly in solution, retards reaction rates, and limits pH
2
reduction applications to the pH 7 to 8 range. In addition, carbonic acid
reaction products are slightly alkaline in nature and tend to act as buffers
in the neutralization of concentrated alkaline wastes. For example:
H2C03 + NaOH 7t NaHC03 + H20 (1)
Carbonic Sodium Sodium
acid hydroxide bicarbonate
pH 8.4
(0.1 normal)
H2C03 + Ca(OH)2 * CaC03 + 2H20 (2)
Carbonic Hydrated Calcium
acid lime carbonate
pH 9.4
(saturated)
Typically, carbonic acid is generated directly in the neutralization
chamber by injecting carbon dioxide into the wastewater solution. Upon
hydration, the carbon dioxide will form carbonic acid and neutralize excess
tlkalinity. Carbon dioxide is available as either a compressed gas or the
by-product of a combustion process. Table 4.7.1 contains a summary of carbon
.dioxide physical property data.
Compressed (liquid) carbon dioxide is stored and transported at ambient
temperatures in cylinders containing up to 22.7 kilograms. Larger quantities
tre stored in refrigerated, insulated tanks maintained at -18°C and
20 atmospheres.3 Transportation is by insulated tank truck and rail car.
4-137
-------
TABLE 4.7.1. SUMMARY OF CARBON DIOXIDE PHYSICAL PROPERTY DATA
Trade name Carbon dioxide
Molecular formula CO2
CAS nunfcer 124-38-9
Sublimation point (°C) -78.5
Latent heat of vaporization (Btu/lb at 0°C) 101.03
Gas density (g/L) 1.976
Liquid density (g/L) 914
Viscosity (cp) 0.015
Heat of formation (Btu/mol at 25°C) 373.4
pH (saturated solution)
I atm 3.7
23.4 atm 3.2
Source: Reference 3
4-138
-------
The standard method of applying compressed carbon dioxide for pH control
is to vaporize carbon dioxide in a heat exchanger or across a flash vaive.
The pressurized gas is forced through porous diffuser tubes placed along the
bottom of a batch treatment tank. Carbon dioxide gas is released from the
diffusers as fine bubbles (15 microns) which are preferentially absorbed by
the surrounding wastewater. This type of treatment requires a slow-moving
effluent stream with a treatment tank of sufficient depth to ensure that the
carbon dioxide is fully absorbed before reaching the surface. Figure 4.7.1
shows the solubility of carbon dioxide in water as a function of temperature
and pressure. Since hydration of carbon dioxide forms carbonic acid, it is
recommended that the diffuser assembly be constructed of a corrosion-proof
material.
The primary advantages of compressed carbon dioxide are minimal capital
requirements, uncomplicated piping, and the inability to over-acidify the
wastewater. Its primary disadvantages are a low dissolved oxygen content
(4.5 percent) at the point of injection, and a high reagent cost on a
neutralization equivalent basis (approximately $200 to $300/ton). However,
•
for large volume users of 200 tons or more per year, the unit cost per ton of
compressed carbon dioxide drops to $90 to- $100/ton.
A secondary' source of carbon dioxide is flue or stack gas from furnaces
using fossil fuel. The resultant flue gas, which may contain up to 14 percent
carbon dioxide, can be used to neutralize waste caustic solutions in a manner
analogous to that of compressed carbon dioxide. In plants with a ready
source of available boiler flue gas, dramatic savings in reagent purchases can
be realized. At a minimum, seven times the SCFH of boiler flue gas will be
required on a neutralization equivalent basis in comparison to compressed
carbon dioxide. However, since flue gas is usually available in sufficient
quantities, this does not generally create a problem.
i
One problem with using boiler flue gas as the neutralization reagent is
that it! is typically available only at low pressure and high temperature
(450°F)^ It contains moisture, carbon dioxide, carbon monoxide, oxygen,
nitrogei^ and sulfur dioxide in varying amounts, depending on the fuel and the
efficiency of the combustion process. Hot flue gas is not especially
corrosive, but when the gas is cooled, the water vapor forms liquid droplets
in which the various gases can dissolve and form compounds which are corrosive
6
to metallic surfaces.
4-139
-------
8
0 10 20 30 40 50 60 70 80 90 100 110 120
Temperature, °C
Figure 4.7.1. Solubility of carbon dioxide in water,
Source: Reference 3.
4-140
-------
Since boiler flue gas must be pressurized prior to use in order to
overcome the liquid head of the treatment tank, a compressor must be
supplied. This adds substantially to system capital costs since the
compressor, sparger, and associative piping must be constructed of corrosion
resistant materials. Alternatively, gas can be dissolved in a pressurized
sidestream prior to introduction into the waste treatment tank (see
Section 4.7.2).
The primary drawback which has limited the use of flue gas
neutralization to a few industrial applications is the presence of both
reduced and oxidized sulfur groups in the gas. These sulfur groups will
result in an atmospheric pollution hazard from hydrogen sulfide evolution.
They also create a potential water pollution hazard due to the incorporation
of sulfate groups into the neutralized effluent stream. Therefore, at this
point in time, boiler flue gas neutralization of alkaline wastestreams is not
considered to be a technically viable treatment process.
A third method of producing carbon dioxide is the underwater combustion
of natural gas (see Figure 4.7.2). In 1951, this method was used to reduce
the alkalinity of a sulfur dye waste through a submerged combustion
process. However, due to high fuel costs, this type of treatment is more
economically attractive as an evaporative acid-recovery process than as a
neutralization technology.
4.7.2 Process Performance
Although research involving carbonic acid neutralization of alkaline
wastestreams has been ongoing for over 50 years, practical applications and
literature citations are limited. Only recently have compressed carbon
dioxide neutralization systems become competitive with sulfuric acid in select
applications. These include plants which use over 200 tons/year of reagent or
Q
have flow rates greater than 100,000 gpd.
The following case study (see Table 4.7.2 for summary of process data),
illustrates an automatic, compressed carbon dioxide neutralization process
which reduces the pH of a 17 MGD chlor-alkali plant effluent from 11.8
to 8.3.4 The caustic wastestream is fast flowing (1 ft/sec) and shallow
(3 ft by 12 ft) and is composed primarily of waste sodium hydroxide. Reagent
usage is approximately 45 tons every 2 to 3 weeks. Compressed C02 is stored
onsite in a leased liquid storage tank.
/ 4-141
-------
FEED
RESERUOIR
! l.:'.--'-'--^'
i :.:;::JO;.=
H£
*!;*;•
1
;__: — •-.::••• -i
\ f N /»
HIKING
BUSTLE
EURPORRTION
UESSEL
OUERFLOLU
Figure 4.7.2. Submerged combustion piloc unit.
Source: Reference 2
4-142
-------
TABLE 4.7.2. SUMMARY OF AUTOMATIC CARBON DIOXIDE
NEUTRALIZATION PROCESS DATA
Parameter Range
WasCewater flow rate (MGD) 17
Wastewater pH (S.U.) 11.8
Sidestream flow rate (MGD) 0.85
Sidestream pH (S.U.) 3.2 - 8.3
Carbon dioxide (psig) vapor pressure 60 - 120
Carbon dioxide (tons/yr) usage 700
Hunter of discharge orifices 20
Diameter of discharge orifices (in.) 1/8 - 5/15
Depth of discharge manifold (ft) 3
Effluent pH 8.3
.Source: Reference 4
4-143
-------
A sidestream pH control system is used to increase reagent efficiency
(Figure 4.7.3). Five percent of the total waste stream flow is pressurized to
50 to 150 psig and mixed with CO. gas. The high pressure in the sidestream
greatly increases the carbon dioxide solubility. This increased solubility
allows the total CO. requirement for the system to be completely dissolved
in a small sidestream flow. The sidestream water ranges in pU from 8.3 when
no CO. is required, to as low as 3.2 at the maximum gas feed rate. Since
waste effluent pH is controlled with a supersaturated CO. water solution,
little gas is released from the discharge manifold into the effluent stream.
Thus, losses are negligible.
The carbonated sidestream water is injected into the mainstream through a
submerged injection manifold. The orifices in the manifold can serve two
purposes. They restrict the flow of the sidestream pump so that the correct
water pressure and flow rate are maintained, and they thoroughly mix the acid
sidestream water with the caustic mainstream.
The high absorption efficiency of the system results from the fact that
the CO. gas is in a supersaturated solution and can only be lost to the
atmosphere by coming out of solution in the form of bubbles. In the event of
an unusually high alkalinity or reagent system electronic failure, an operator
can manually open the CO. control valve and overtreat the effluent. The
excess gas will then slowly bubble out downstream and neutralize the effluent
to an 8.3 pH equilibrium.
The sidestream pH control system includes two water pumps (20 and 40 hp)
which are automatically operated on the basis of required C02 flow. The
smaller pump is sized to handle the average carbon dioxide flow requirement of
0 to 7 Ib/min., while the larger pump is sized to handle peak flows of up to
35 Ib/min. The controls monitor the C02 valve position and turn on the
pumps as required for a particular effluent flow.
4.7.3 Process Costs
Table 4.7.3 summarizes the process costs for the construction and
installation of a continuous, automatic, compressed carbon dioxide
neutralization system. Equipment sizing and operating costs were based on the
4-144
-------
CO, Vaporizer
Caustic Water In Top-Tube Side of Heal Exchanger
Carbonated Water to Discharge Menilold Out Bottom
CO2 Liquid Inlet-Bottom-Shell Side Heal Exchanger
COjVapor Outlet-Top-Shell Side Heat Exchanger
Annunciator System
Common Alarm
Normally Opon
Contacts
To Plant
(Alarm Logic)
h-1
•P*
Ol
Hir.MCO TANK I'RESSlinE
towco
Pressurized
~T~ Caustic
Sidestream
Water
375 psig
Helief Valve
Heat Exchanger
CO, Vaporizer
Pressure Gauge
0-150 psig
Check
Valves
300 opm-286 ft nd
Pump
250gpm-138lthd
Pump
Discharge
Manifold
14 II Wide
20 Orifices
Caustic
17mgd »•
Effluent Plow
Figure 4.7.3. Compressed carbon dioxide treatment system.
Source: Reference 4.
-------
TABLE 4.7.3. COMPRESSED CARBON DIOXIDE TREATMENT COSTS
NaOH waste Ca (OH)2 waste
Flow rate (gph)a Flow rate (gph)b
12,500 15,000 20,000 12,500 15,000 20,000
Capital investment ($)
Liquid C02 storage tank 32,728 43,125 48,650 32,728 43,125 48,650
pH control system 6,800 6,800 6,800 6,800 6,800 6,800
Pump system . 1,976 2,233 2,576 1,976 2,233 2,576
Heat exchanger 5,000 6,000 8,500 5,000 6,000 8,500
Sparger system 574 656 820 574 656 820
Flocculation/clarification — — — 100,000 112,000 120,000
Filter press — — — 600,000 630,000 700,000
Total capital cost 47,078 58,814 67,346 747,078 800,814 887,346
Annualized capital 8,333 10,410 11,920 132,233 141,744 157,060
Operating cost (S)c
Taxes and investment (72) - 583 729 834 9,256 9,922 10,994
Maintenance & overhead (5Z) 417 521 596 6,612 7,087 7,853
Labor & overhead (S20/hr) 48,000 48,000 48,000 48,000 48,000 48,000
Sludge disposal ($200/ton) — — — 416,667 500,000 666,667
Reagent cost (SlOO/ton) '55,140 66,165 88,220 55,140 66,165 88,220
Total cost/year 112,473 125,825 149,570 667,908 772,918 978,794
Cost/1,000 gallon . 4 3.5 3 22 21 20
*0.1 N NaOH wastestream.
b1.7 percent Ca(OH)2 wastestream.
eDo«s not include utilities.
Source: Cardox quote September 1986, Reference 9, 10 using
September 1986 C?I index cost data.
4-146
-------
neutralization of a 0.1 N (4 g/L) sodium hydroxide wastestream and a
1.7 percent bydrated lime wastestream. The flow rates were assumed to be
12,500, 15,000, and 20,000 gal/hr, operating 2,400 hrs/yr. Capital costs
include: a refrigerated, liquid carbon dioxide storage tank, a pH control
system, sidestream pump, liquid carbon dioxide vaporizer, diffuser assembly,
and sludge separation and handling equipment. Operational expenses were
assumed to include: taxes and insurance, maintenance and overhead, labor and
overhead, sludge disposal, and reagent costs.
The liquid C02 storage tank is sized to hold a 3-week supply under
normal operating conditions. The pH control system includes a 30-day
recorder, pH probe, alarm system, control valves, and switches. The quoted
a
price for this particular pH control system was $6,800, although costs for
comparable equipment may be as high as $12,000. The sidestream pump is
assumed to remove 5 percent of the total effluent flow at a pressure of
150 psi. The pumps are API-610, cast steel casing, horizontal, in-line
centrifugal pumps with in-line vertical motors. Pump costs are based on
9
updated pump capacity curve data contained in the literature. The heat
exchanger/vaporizer system costs are based on those supplied for
fixed-tube-sheet heat exchangers with 3/4 in. O.D. x 1 in. square pitch with
9 •
carbon-steel shells operating at .150 psi. The diffuser system cost is
represented by vendor-supplied manifolds, each of which can displace
600 ft /hr of carbon dioxide gas at $82/sparger.
Flocculation/clarification unit costs are based on information contained
in Figure 4.1.6, with multiple units provided when maximum capacities are
exceeded. Sludge generation for the sodium hydroxide system is assumed to be
negligible based on a sodium bicarbonate formation of 8.4 g/L and a minimum
solubility of 69 g/L. Sludge generation for the hydrated lime system is
assumed to be 1.35 Ibs of calcium carbonate precipitate per pound of lime
neutralized, based on a maximum solubility of 0.018 g/L. Clarifier
underflow is based on a sludge containing 2 percent solids which is
subsequently dewatered to 60 percent solids in a recessed plate filter press.
Filter press capital investment costs are based on Figure 4.7.4 which
includes: the filter press (maximum capacity 224 ft ), feed pumps
(including one standby), a sludge conditioning and mixing tank, an acid wash
system, and housing. Housing costs are for a two-story, concrete block
4-147
-------
r
e
4
3
2
0.000
1
€
i
*
3
2
1 .000,
9
a
6
9
4
3
2
100
—
6
S
4
3
2
,000
1
000
,000
' 1 ' : ' 1 1 ' ! ' ! ' ! ' i ' ' — ! ' ; ;
I 1 1
|
SIN61L
UNIT
J
!
1
MULT PLE!
i
i
I
»
i^
UN
X
its
X
x
xf
— -, .i *-.. i I . . ...
' ' ^****~~f^
_^--"
. ^ef****~
^*
•
!
2 3458789)0 2 34567 8 9CO 2 3 436789
tooo
TOTAL FILTER PRESS VOLUME -ft 3
0.1 1-0 10.0
TOTAL FILTER PRESS VOLUME -*3
Figure 4.7.4. Construction cost for recessed plate filter press,
Source: Reference 10.
4-148
-------
building with the filter press located on the upper floor, and discharging
through a floor opening to a disposal truck. Also included in the
Figure 4.7.4 cost estimates are a lime storage bin and feeders and ferric
chloride solution storage and feeders. However, these costs have been
deducted from the final cost analysis since they are not applicable to the
example waste'streams. Cycle times were estimated to be 2.25 hours with a
20-minute turnaround time between cycles.10
Operating and maintenance costs are based on precepts presented previously,
except that labor has been increased to 8 hrs/day at a rate of $20/hr, and
reagent costs are $100/ton based on reagent usage greater than
Q
200 tons/yr. Reagent consumption is assumed to be 0.25 percent above
stoichiometric requirements due to atmospheric losses.
Table 4.7.3 demonstrates that, in high volume applications, reagent costs
can constitute up to 50 percent of total costs in non-sludge generating
systems. In non-carbon dioxide producing regions such as the east coast, the
cost of compressed carbon dioxide can effectively double. Sulfuric acid has a
30 percent minimum price advantage over carbon dioxide on a neutralization
equivalent basis. Combined with its proven performance and widespread
availability, sulfuric acid is typically the reagent of choice in the majority
of applications. However, when an available source of carbon dioxide is
nearby, space limitations are critical, and rapid reaction rates are necessary
(e.g., in-line neutralization), liquid carbon dioxide may be the reagent of
choice.
4.7.4 Process Status
Liquid carbon dioxide treatment of alkaline wastestreams is a promising
but not widely applied technology. Improved facilities for the
transportation, storage, and handling of liquid carbon dioxide have
contributed to the recent emergence of this technology as a viable treatment
in the last 5 years. While application of this process is limited to less
than 300 facilities nationwide, increasing concern over the hazardous
aspects of mineral acid treatment (burn dangers, acid mists, etc.) may
increase utilization of this technology in high volume applications. A
summary of the advantages and disadvantages of liquid carbon dioxide treatment
is provided in Table 4.7.4.
4-149
-------
TABLE 4.7.4. ADVANTAGES AND DISADVANTAGES OF CARBON
DIOXIDE NEUTRALIZATION
Advantages:
- Capital requirements are minimal.
- Wastewater cannot be over-acidified
- No complex piping required.
1 to 1-1/2 minute retention time allows for in-line
neutralization of alkaline wastes which are fairly
uniform in nature.
- No danger of acid burns as with sulfuric or
hydrochloric
Disadvantages;
- Low dissolved oxygen content (4.5 percent) at point
of injection.
- High reagent cost limits applications to facilities
with minimum flow rates of 100,000 gpd or minimum
reagent usage of 200 tons/year.
- May be an asphyxiant at high concentration.
- Care must be exercised when contacting liquid C02
with wastewater in the vaporizer to prevent freezing.
Source: References 2, 3, 4, 5, and 8.
4-150
-------
REFERENCES
1. Curtis, H.A., and R.L. Copson. Treating Alkaline Factory Waste Liquors
Such as Kier Liquor from Treating Cotton with Caustic Soda. U.S. Patent
No. 1,802,806. April 28, 1931.
2. Camp, Dresser, and McKee. Technical Assessment of Treatment Alternatives
for Wastes Containing Corrosives. Contract No. 68-01-6403. September
1984.
3. Kirk-Othmer Encyclopedia of Chemical Technology. Vol. 4, 3rd Edition.
pp. 725-741. John Wiley & Sons, New York, NY. 1981.
4. Griffith, M.J. et al. Carbon Dioxide Neutralization of an Alkaline
Effluent Industrial Waste. March 1980.
5. Ponzevik, D. Liquid Air Products. Telephone conversation with
Stephen Palmer, GCA/Technology Division, Inc. September 6, 1986.
6. Beach, C.J., and M.G. Beach. Treatment of Alkaline Dye Waste with Flue
Gas. Proceeding - Fifth SMIWC. 1956.
7. Murdock, H.R. Stream Pollution Alleviated-Processing Sulfur Dye Wastes.
Industrial and Engineering Chemistry. 43:77A (1951).
8. Berbick, D., Cardox Corporation. Telephone conversation with
Stephen Palmer, GCA Technology Division, Inc. September 25, 1986.
9. Peters, M.S. et al. Plant Design and Economics for Chemical'Engineers.
3rd. Edition. McGraw-Hill Book Company, New York, NY. 1980.
10. U.S. EPA. Design Manual: Dewatering Municipal Wastewater Sludges.
EPA-625/1-82-014. October 1982.
11. CRC Handbook of Chemistry and Physics. 58th Edition. CRC Press, West
Palm Beach, FL. 1978.
4-151
-------
SECTION 5
RECOVERY/REUSE TECHNOLOGIES
5.0 INTRODUCTION
The processes discussed in this section are used to recover or reuse
corrosive wastes. Information is presented on process description,
performance, costs, and current status for the following technologies:
• Evaporation/Distillation
• Crystallization
• Ion Exchange
• Electrodialysis
• Reverse Osmosis
• Donnan Dialysis & Coupled Transport
• Solvent Extraction, and
i
i
i
• Thermal Decomposition.
i
• Crystallization and evaporation/distillation involve the use of
temperature changes to ejffect a separation of contaminants and recovery of
corrosive solutions. Ion exchange methods are based on the use of an anionic
or cationic selective resin to remove ionic contaminants (i.e., metal ions)
5-1
-------
from corrosive wastes. Electrodialysis, reverse osmosis, Donnan dialysis, and
coupled transport processes involve the use of a membrane to separate
contaminants from corrosive solutions. Solvent extraction uses the
differential distribution of constituents between the aqueous phase waste and
an organic phase solvent to separate constituents from a mixed solution of
metal salts and acid wastes. Thermal decomposition involves decomposing metal
salts (present in spent acid wastes) in a roaster and collecting vaporized
acid in a condenser.
In addition, a brief discussion of the role of waste exchanges in reuse
of corrosives is presented at the end of this section. Waste exchangese
involve the transfer of unwanted corrosive waste material generated by one
company to another company capable of using it.
5-2
-------
5.1 EVAPORATION AND DISTILLATION
5.1.1 Process Description
Evaporation is a concentration process used in the metal finishing and
electroplating industries to recover plating solutions, chromic acid, nitric
acid/hydrofluoric acid pickling liquors, and metal cyanides from spent baths
123
and rinsewaters. ' ' Distillation techniques are commonly used in
conjunction with evaporation to recover water vapors by condensing them and
returning them to the rinse tank, and to recover acids for return to acid
baths.
Evaporation can be performed using atmospheric or vacuum techniques.
Atmospheric evaporation occurs by boiling the liquid at atmospheric pressure
(14.7 psi). The evaporation temperature can be lowered by spraying the liquid
1 4
on a heated surface, and blowing air over this surface. ' Thus,
1 4
atmospheric evaporation occurs by humidification of the air stream. ' With
vacuum evaporation, the system pressure is lowered which causes the liquid to
boil at a lower temperature. The vapor is subsequently condensed. A vacuum
pump is used to maintain the vacuum condition during this process. The basic
types of evaporation systems include-the following:
.*
1. Spray Evaporator;
2. Rising (or Climbing) Film Evaporator;
3. Submerged Tube Evaporator; and
4. Atmospheric Exhaust Evaporator.
Process flow diagrams for these systems are presented in Figure 5.1.1.
The spray and rising film evaporators involve covering the heating surface
with a thin film of waste liquid, whereas the submerged tube and the
atmospheric exhaust evaporation systems transfer heat to a reservoir (tank) of
waste liquid. The distillate is returned to the rinse tanks for each of
these systems, with the exception of the atmospheric exhaust evaporator which
5-3
-------
IXHAUST
VATHVAPQir.
PACKtO TOVIM
AIM •
VASUVATIR '
VACUUM tm
L/"
HtAIIXCNADCIII
ntAM
ecneiNt»Ati
R. HTHOSPHtHIC [UHPOHHTOH
W»*T|W*TW
CONCtHIIATI
,'•&
VACUUM PUMP
COOIWC
\rAIH
-> CONWNSAU
• (T(AH
> *TI »M
CONOCN5AIC
B. SUBHtRCtO TUBt tUHPOHHTOB
-LOUC MMTUMC
IVAPCWATOII
MOARATOR CONDINCCK
V All* VAPOR
VAiTIVAUR
C. CUMBINCfKM tllRPORfllOR
Figure 5.1.1. Process flow diagrams for commonly used evaporation systems.
Source: References 1 and 4.
-------
vents the distillate to the air. The surface film evaporators (flos. 1 and
2 above) are more efficient due to their higher heat transfer
coefficients. Therefore, smaller corrosive-resistant surface areas are
required for surface film evaporators, which in turn reduces capital costs.
Heat energy requirements for the evaporation system will depend upon the
method used to supply heat. Heat is generally supplied in the form of steam.
According to the basic laws of thermodynamics, it takes 970 BTU to evaporate
1 Ib of water at atmospheric pressure. However, the amount of energy required
to produce sufficient steam can be reduced if heat is reused at successively
lower temperatures by multi-effect evaporation. For example, a
double-effect evaporator uses half the heat of a single-effect evaporator.
However, capital costs will increase with increasing number of effects. The
use of low—grade waste heat achieves the optimum amount of energy savings.
If sufficient waste heat is not available, a vapor compression (VC) evaporator
can be employed.
The VC technique involves the use of a mechanical compressor to heat the
plating solution and to increase the pressure and temperature of the separated
water vapor. VC units operated under atmospheric conditions are only
»
applicable to alkaline wastes due to the corrosive carryover of acid fumes
under these conditions. However, VC systems operated under vacuum
conditions have lower temperature requirements, and can also make wider use of
le
3
waste heat. Thus, the vapor compression (VC) system reduces the operating
costs of the evaporation process by lowering energy consumption.'
Although, distillation can be performed using either atmospheric or
vacuum techniques, the high temperatures associated with atmospheric
3
techniques can cause degradation in plating chemicals. Also, carryover of
corrosive fumes can occur with atmospheric techniques. Vacuum evaporation
systems do not require cooling towers or external steam sources, and therefore
3
can be more cost-effective. Additionally, vacuum systems allow the use of
lower distillation temperatures and therefore have the advantages of greatly
2 3
reduced corrosive action and lower-cost construction materials. '
5-5
-------
Evaporation/distillation techniques are typically used in the recovery of
acids and bases from spent rinsewaters. The spent solution is heated to
evaporate water from the solution, thereby increasing the concentration of
solute in the remaining solution. During distillation, the water vapor
resulting from the evaporation process is condensed. The distillate is
returned to the rinse tanks and the concentrate is used in the fresh bath
make—up.
Currently, the only application of vacuum evaporation-distillation
techniques for the recovery of corrosive wastes directly from the spent bath
is in the recovery of spent nitric acid/hydrofluoric acid pickling liquor.
The process can be operated using either a one-stage evaporation system or a
two-stage system. The one-stage system is suited for treating undiluted
wastes, and the two-stage system is applicable to diluted waste acids obtained
during pickling.
A two-stage system is diagrammed in Figure 5.1.2. With the two-stage
system, the spent pickle liquor is initially vacuum distilled to reduce the
total liquid volume as much as possible before appreciable loss of acid in the
2
distillate occurs. Following this step, sulfuric acid is added to the
2
residual liquid causing precipitation of metal sulfates. Hydrofluoric and
nitric ccids have a-higher vapor pressure than sulfuric acid and will
8
therefore boil off with the remaining water. The hydrofluoric and nitric
acids are later condensed and recycled to the pickling tank (Stephenson et
al., 1984). The sulfuric acid/metal sulfate slurry is passed through a
solid-liquid separator, which generates a liquid consisting of 50 percent
2
sulfuric acid and a metal sulfate sludge. The 50 percent sulfuric acid
'solution can be reused in the second stage of the evaporation process. The
metal sulfate sludge must either be disposed or subjected to further treatment.
A one-stage system is diagrammed in Figure 5.1.3. With this process, the
waste liquid flows continuously into the evaporation system without
concentration and evaporates under vacuum. Concentrated sulfuric acid is
fed with the waste acid into the evaporator. Upon boiling, more sulfuric acid
is added to maintain the liquid level in the evaporator. Nitric and
5-6
-------
IlKICllUGt
(KtlUl
U1
I
IrtHl
FUMIN
ACID
(OttnNl
V
IVi.'CiilCK
«•-
V
V
•.ilia
ACID
Figure 5.1.2. Process flow diagram for a two-stage vacuum evaporation/distillation
system to recover spent HNO /HP pickling liquors.
Source: Reference 7.
-------
PRESSURE
GAUGE
WASTE LIQUID
00
FILTER
METERING
CONTROL
DEVICE
/ METERING
I CONTROL DEVICE
WATER
PUMP
CIRCULATING WATER POND
Figure 5.1.3.
Process flow diagram for a single-stage vacuum evaporation/distillation (
system to recover spent HNCL/HF pickling liquors.
Source: Reference No. 6.
-------
hydrofluoric acid -evaporate continuously and condense in the condenser. The
distilled liquid consists of the regenerated acid. Metal salts in the waste
liquid are continuously converted into metal sulfates which remain in the
evaporator. As with the single-stage operation, the metal sulfates must
either be disposed or treated.
Operating Parameters—
Important parameters which affect the operation of an
evaporation/distillation system include: temperature, pressure, acid
concentrations, and construction materials.
For direct treatment of spent acid, the acid concentration is a critical
control factor in the proportions of constituents distilled. For a two-stage
system, it is important to drive off most of the water while retaining the
acid in the residual liquid. As shown in Figure 5.1.4, when the mixed-acid is
being distilled, the acid concentration in the distillate remains relatively
low until approximately 50 percent of the original acid volume is
a
distilled. As distillation continues beyond this point, the concentrations
of HF and HN03 in the distillate increase sharply.
Sulfuric acid is added to the system to transform the nitrates and
fluorates to sulfates. The amount of sulfuric acid added to the system will
affect the acid recovery ratio. If the concentration of circulating sulfuric
acid is too low, the acid recovery ratio will decrease. Also, if the
sulfuric acid concentration is too high, iron salts will precipitate too soon,
which will lower the acid recovery ratio. The concentration of sulfuric acid
should be maintained at 12.5 N for proper control.
The pressure of the system also effects distillate losses. As can be
seen from Table 5.1.1, the distillate losses decrease with decreasing
pressures. The permeability and compression strength of the graphite heater
determine the pressure of the heating steam. It usually does not exceed
2 kg/sq cm, and is generally controlled at 1 kg/sq cm.
Lower operating temperatures are desirable in order to minimize corrosion
of evaporation/distillation equipment. Operating temperatures should not
exceed 65°C. By lowering the pressure the boiling point
5-9
-------
I/I
0 10 20 40 60 80
Per Cent of Originol Volume Distilled
Figure 5.1.4. Nitric and hydrofluoric acid concentrations In cumulative
distillate fraction versus original volume distilled.
Source: Reference 8.
-------
TABLE 5.1.1 EFFECTS OF PRESSURE ON DISTILLATE LOSSES.
Pressure
(psia)
14.7
14.7
1.8
Final
Temp. (°F)
224
221
150
Percent-Original
Volume Distilled
51.2
50.0
49.0
Distillate
Nitrate
4.0
3.5
1.8
Losses (%)
Fluoride
6.4
4.6
1.3
Source: Reference 8.
5-11
-------
1 of a liquid decrease. Thus, the use of a vacuum allows lower operating
temperatures. The vacuum pressure typically ranges from 660 to 680 mm Hg.b
Materials used in the evaporation/distillation system need to be able to
withstand the corrosive properties of the waste. Commonly used materials
include titanium, tantalum, borosilicate glass, fiberglass-reinforced plastic,
^
and polyvinyl chloride. The materials used will depend upon the specific
application. Metallic materials will be dissolved by the waste acid
solution. Ceramics will be degraded by the hydrofluoric acid. Certain
plastics will withstand the corrosive properties of the waste if the
temperature is not too high. Fiber-reinforced PVC plastics are typically used
in the construction of the evaporator and peripheral piping (Delu, et al.,
1980). However, plastics are not good heat conductors. Therefore, the use of
an impermeable graphite heat exchanger is recommended due to its good heat
conductivity, impermeability, strong oxidation resistance, and non-fouling
property. It has been successfully used in several plants since 1974.
Additional factors to consider when designing an evaporation system for
the recovery of corrosives from rinsewaters include: rinse ratios, mixing
techniques, rinse flow rates, and rinse concentrations. Each of these
parameters must be optimized in order to design an efficient, properly-sized
evaporation system. *
The rinse ratio quantifies the amount of rinse water available to a
Q
particular load. For example, if the dragout amounted to a workload of
2 gal/hr, and the flow rate of the rinse water was 100 gal/hr, the rinse ratio
would be 50 (100/2). With a lower rinse ratio, a smaller evaporator is
Q
required, which in turn lowers the costs. However, the rinse ratio must be
high enough to assure product quality.
Effective mixing in the rinse water tank will aid in maintaining
uniformly good quality when fresh water additions are made. Some methods of
achieving proper mixing include: maintaining a high water flow rate in the
rinse tank, air agitation using a low-pressure blower, mechanical mixing using
electric or air driven stirrers, and the use of countercurrent rinsing (using
two or more rinse tanks in series).
5-12
-------
If effective rinsing techniques are used, the evaporator can generally be
9
sized to approximately 15 times the dragout rate from the plating tank. The
evaporator size is equivalent to the water flow rate in th'e rinse ratio
formula. In order to determine if a rinse ratio is appropriate for a
particular application, the following formula can be used:
ci
R.
where Ci=concentration in rinse tank, Co=concentration of the plating tank,
9
and Ri*rinse ratio raised to a power equal to the number of rinse tanks.
The total quantity of concentrate that can be recovered can be calculated
using the following formula:
Z Capture - (1 - 1/R.) x 100
These two formulas can be used to optimize an evaporative recovery system for
various rinse ratios, rinse tank numbers, and plating bath concentrations.
It is also important to consider the energy efficiency of a system when
designing an evaporation system. Energy requirements contribute the most to
operating costs for these systems. Efficiencies are typically measured using
the coefficient of performance (COP). The COP is equal to the energy output
divided by the energy input. Higher heat transfer efficiencies will
contribute to a higher thermal efficiency. In addition, effective reuse of
waste heat will lower the overall energy requirements.
Pre-Treatment —
A filtration system may be required to remove suspended particulates
present in the spent solution to avoid clogging of the distillation system.
Also, if any impurities are present in solution or suspension in the spent
solution, pretreatment prior to evaporation may be necessary because these
impurities will be concentrated in the evaporator. Chemical treatment or
filtration techniques may be employed to remove these impurities. For
5-13
-------
continuous operation, it is recommended that deionized water be used in makeup
additions in order' to avoid buildup of impurities.
Post-Treatment—
The evaporation/distillation process generates a metal sulfate sludge.
The sludge can either be treated and disposed, or subjected to further
treatment via roasting techniques to recover the metal oxides (see section
5.8).
Also certain impurities may be recovered along with the concentrate. A
filtration system may be required prior to returning the concentrate to the
bath. A Freon-cooled crystallization/filtration system is commonly required
to remove carbonates from the concentrated solution.
5.1.2 Process Performance
The performance of an evaporation/distillation system is typically
evaluated on the basis of the percent removal of contaminants, the percent
product concentration achieved, the product .quality, the amount of waste
generated, the energy requirements, and the economics of the process.
Evaporation/distillation systems can be used effectively to recover
acidic waste streams. A system using a vapor compression evaporator operated
under vacuum conditions was tested at the Naval Facilities Engineering Command
(NFEC) Charleston, South Carolina to recover chromic acid from a hard chrome
plating line rinse tank. Testing was performed over a 9-month period with
more extensive data collected between March 23 and April 23, 1984. Typical
operating parameters and results during this period are summarized in
able 5.1.2. As shown in this table, the unit was able to successfully recover
chromic acid at relatively low operating temperatures (i.e., lower energy
requirements). An economic evalution determined that the system was only
cost-effective for processes generating more than 100 gal/hr dragout. A
smaller capacity system was then developed which could operate economically at
boiling temperatures in the range of 100° F.
5-14
-------
TABLE 5.1.2.
SUMMARY OF OPERATING PARAMETERS AND RESULTS DURING
TESTING OF HIGH VACUUM VAPOR COMPRESSION EVAPORATION
SYSTEM AT THE CHARLESTON NAVY YARD.
Parameter
Result
Compressor Efficiency
Coefficient of Performance (COP)
Adiabatic Efficiency
Capacity
Total Chrome Recovered
(70 gals x 54,900 mg/l)/7484
Dragout Rate
(32.1 lb/320 hrs/month)
x (1 gal/2 Ib Cr+6)
Rinse Ratio
(Ratio of plating bath concentration
to final rinse concentration using
3 countercurrent rinse tanks)
Rinse Flow Rate
Evaporator Capacity
(Required Rinse Rate)
27 gph x 0.05 gph - 1.35
Recovered Process Water
Quantity
Conductance
Operating Temperatures
Electrical Requirements
10.3
25 %
25 gph @ 700 rpm speed
40 gph @ 1170 rpm speed
32.1 Ib
(513.5 oz.)
0.05 gal/hr
20,000
27 gph
per 1 gph dragout
1.35 gph
8.75 gph
(33,600 gpy)
10 mono
95 - 122°F
9 kw
Source: Reference 5.
5-15
-------
The performance of Che smaller capacity system was tested at the Superior
Plating Division of Florida Plating, Inc. in St. Petersburg, Florida for
recovery of a cadmium cyanide plating solution. The process line consisted of
a manual hoist and barrel (10 x 18 in.) with two plating tanks and three
countercurrent rinse tanks. The average dragout was 0.44 gal/hr.
The evaporation system employed is diagramed in Figure 5.1.5. The
system consists of single-effect, high-vacuum, climbing-film evaporator, a
20-ton Freon heat-pump, and a Freon crystallizer. Operating and design
parameters are summarized in Table 5.1.3. The evaporator uses bayonet
augmented tube (BAT) heat exchangers to recover heat from the heat pump, which
allows the evaporation temperature to be maintained below 110°F so that
cyanide breakdown is minimized. The Freon crystallizer also employs BAT
heat exchangers to extract heat from the cadmium cyanide concentrate before
returning it to the plating bath in order to raise the temperature of the
incoming rinsewater from the cadmium cyanide operation. The purpose of
the crystallizer is to remove carbonates (impurities that interfere with
product quality) from the bath by chilling the concentrated cadmium cyanide
complex to 28°F, which precipitates sodium carbonate crystals that can be
removed by settling. The distilled rinsewater is returned to the last
'countercurrent rinse tank).
• •
Superior Plating Division is satisfied with the performance of the system
for their application. Detailed monitoring of the performance of the system
was conducted for an 11-day period in April 1985.l0'11»12»13»14 it was
found that during 90 hours of operation, 40 gal of cadmium cyanide complex
solution was recovered. The average concentrations in the recovered solution
were 2.14 oz/gal cadmium and 15.3.oz/gal sodium cyanide. The system is
currently operated during the day shift. Approximately 1 to 2 hours per day
are required for start-up, data collection and recording, and shut-down.
An economic evaluation of the system is presented in Table 5.1.4.
Performance data on the use of evaporation/distillation system to recover
corrosives directly from the spent solutions (as opposed to recovery from
rinsewaters) is limited to research conducted on the recovery of nitric and
hydrofluoric acids from spent pickling liquors. The performance of this
5-16
-------
PLATE
TANK
WORK
CONOENSOR
e
2
UJ
Ui
HORIZONTAL
CLIMBING FILM
EVAPORATOR
3-STAGE COUNTERCURRENT
RINSE TANKS
FEED
COOLING WATER
CRYSTALLIZE?
FREON
EVAPORATOR
FREON
CONOENSOR
FREON
OESUPERHEATER
COOLING
LOOP
HEATING
LOOP
INDUSTRIAL
WATER
SOURCE
HEAT
PUMP
POWER
CONCENTRATE"
SUPPLY
SODIUM CARBONATE
CRYSTALS
WARM WATER
RINSING
Figure 5.1.5.
Flow diagram of Evaporative Recovery System
installed at Superior Plating, Inc.
Source: Reference 5.
5-17
-------
TABLE 5.1.3.
Parameter
SUMMARY OF OPERATING PARAMETERS AND RESULTS USING
EVAPORATION TO RECOVER A CADIUM CYANIDE PLATING BATH
AT SUPERIOR PLATING, INC.
Result
Heat Pump Capacity
Heat Pump Exit Temperature
Evaporator Capacity
Evaporation Temperature
Chiller Exit Temperature
Freon Condenser Exit Temperature
Coefficient of -Performance (COP)
Recovered Cadmium Cone.
Recovered Sodium Cyanide Cone.
300,000 BTU/hr
125°F
200 - 250 Btu
per Ib water distilled
110°F
95°F
140°F
4.35
2.14 oz./gal.
15.3 oz./gal.
Source: Reference 10.
5-18
-------
TABLE 5.1.4,
ECONOMIC EVALUATION OF VACUUM COMPRESSOR EVAPORATOR
UNIT EMPLOYED AT SUPERIOR PLATING, INC.
Item
Cost
Capital Equipment Cost
Operating Costs
Cost Savings
Net Savings
(Cost Savings - Operating Costs)
Payback Period
(Capital Costs/Net Savings)
$ 24,000/yr
$ 2,000/yr
$ 11,560/yr
$ 9,560/yr
2.5 yrs
Source: Reference 10.
5-19
-------
process is typically evaluated on the basis of the percent of nitric and
hydrofluoric acid recovered, the percent of metal contaminants removed, and
acid additions required in the makeup solution. Pilot-scale • ~
evaporation/distillation systems for the recovery of nitric/hydrofluoric acid
pickling liquors have not been installed in the United States.
A single-stage system was employed at a steel plant in China.
Monitoring of this pilot-scale system was conducted over a 3—year period as
part of an acid recovery research project. Typical results are shown in
Table 5.1.5.
A two-stage system developed by Rosenlew has been installed at the NYBY
Steel Works in Sweden. Typical operating parameters and results are
summarized in Table 5.1.6. In all tests, the evaporation of hydrofluoric acid
sr<
7
went to near completion. However, the percent recovery of nitric acid was
dependent upon the sulfuric acid addition.'
Commercial-scale evaporation/distillation systems for recovery of
corrosives directly from the spent bath have not been tested in the United
States. Thermal decomposition (see Section 5.8) appears to be a more
cost-effective process for this application.
5.1.3 Process Costs
Capital costs for an evaporative recovery system will vary with the waste
type, waste quantity, process flow rates, type of heat exchanger employed, and
system size. Table 5.1.7 presents costs for various system sizes.
Operating costs generally include 1-2 hours labor for system maintenance
and operation (labor requirements will be reduced if the system is operated
continuously), electrical and fuel energy requirements for heat supply, taxes
and insurance, and depreciation costs. Approximately 10 Ibs of low pressure
9
steam (15 psig) is required for every gallon of liquid evaporated.
Evaporation/distillation processes require large amounts of heat energy,
which can make the process quite costly. However, efficient use of energy
systems can lower these costs significantly. Waste heat from other industrial
processes (diesel generators, incinerators, boilers, and furnaces) within the
plant can be recovered for use in the evaporation/distillation system. The
use of multi-effect evaporators and vapor compression systems can also improve
thermal efficiencies.
5-20
-------
TABLE 5.1.5. SUMMARY OF RESULTS OF PILOT-SCALE UNIT
INSTALLED AT A CHINESE STEEL PLANT.
Item
Volume
(liters)
Concentration (g/1)
F-
Fe+2 Ni+2 Cr+3
Spent Acid 1140
Residual Liquid 350
Regenerated
Acid 858
Sulfuric Acid 150
Recovery
Ratio (2)
2.58 2.67 1.70 21.50 3.58 4.27
13.47 0.88 0.23 80.05 12.30 13.7
3.23 2.12
32.4
92.9 93.9
Source: Reference 6.
5-21
-------
TABLE 5.1.6. TYPICAL OPERATING PARAMETERS DURING TESTING OF THE
PILOT-SCALE EVAPORATION/DISTILLATION SYSTEM AT THE
8YBY STEEL WORKS IN SWEDEN.
Parameter
Result
Design Capacity
Equipment Size
Vacuum Evaporation Temp.
Nitric Acid Concentration
in Recovery Evaporator
Hydrofluoric Acid Cone.
in Recovery Evaporator
Sulfuric Acid Cone.
1.5 cu. meters/hr
9m x 9m x 15m
80°C
10 to 20 wt-Z
• 2 to 8 wt-Z
60 wt-Z
Source: Reference 7.
5-22
-------
TABLE 5.1.7 TYPICAL CAPITAL EQUIPMENT COSTS FOR VARIOUS
EVAPORATION SYSTEM CAPACITIES
Evaporator Capital Costs
Capacity Cgph) ($)
20 25,000
40 33,800
55 39,199
120 44,129
300 115,000
Source: References 15 and 16 (August 1986).
5-23
-------
Cost savings will be realized in reduced neutralization costs, reduced
sludge disposal costs, and reduced purchase requirements for fresh bath makeup
solutions.
5.1.4 Process Status
Evaporation/distillation is one of the oldest recovery techniques, and is
widely used in industry. _ Over 600 units are currently in operation in the
9 15
United States. ' They are most commonly used in metal finishing and
electroplating industries to recover plating solutions, chromic acid and other
concentrated acids, and metal cyanides. In addition, water recovered from the
evaporation process is of high purity and can be reused in process waters.
The percentage of these units used in various plating applications is
presented in Table 5.1.8. These systems are most effective in recovering
acids, bases, and metals from rinsewaters. Systems can be designed
9
cost-effectively with capacities ranging from 20 gph to 300 gph. These
systems are cost-competitive with conventional neutralization and disposal
technologies. Greater cost savings are realized with larger operations.
The use of evaporation/distillation systems to recover concentrated
streams directly from the spent solution is limited. Pilot-scale
evaporation/distillation systems for recovery of nitric/hydrofluoric acid
pickling liquors have been tested at facilities in Europe. However,
cost-effective systems for direct recovery of spent solutions via
evaporation/distillation have not been developed at the commercial-scale for
application in the United States. Other technologies, such as thermal
decomposition (see Section 5.8) appear to be more cost-effective for this
purpose.
In summary, evaporation/distillation systems are effective in recovering
corrosives from rinsewaters. Cost-effective systems are commercially
available for a wide range of spent rinsewater constituents and process
sizes. Application to direct treatment of spent corrosive solutions is
limited by costs.
5-24
-------
TABLE 5.1.8. PERCENTAGE BREAKDOWN BY PLATING TYPE OF
EVAPORATION UNITS CURRENTLY IN OPERATION.
Plating Chemical Percent of Units
Chrome 50
Chrome Etch 10
Nickel 20
Cyanide 10
Other 10
Source: Reference 9.
5-25
-------
REFERENCES
1. Cushnie, G. C. Centec Corporation, Reston, Virginia. Navy
Electroplating Pollution Control Technology Assessment Manual. Final
report prepared for the Naval Civil Engineering Laboratory, Port Uueneme,
California. NCEL-CR-84.019. February 1984.
2. Stephenson, J.B., J.C. Hogan, R.S. Kaplan. Recycling and Metal Recovery
Technology for Stainless Steel Pickling Liquors. Environmental Progress,
(3)1: 50-53. February 1984.
3. Chacey, K., L. Mellichamp, and W. Williamson. Chrome Electroplating
Waste BAT. Pollution Engineering. April 1983.
4. Camp, Dresser, and McKee, Inc. Technical Assessment of Treatment
Alternatives for Wastes Containing Corrosives. Prepared for the U.S. EPA
Office of Solid Waste under EPA Contract No. 68-01-6403 (Work Assignment
No. 39). September 1984.
5. Williamson, R. C., and W. R. Williamson. Li con, Inc., Pensacola,
Florida. Energy Effective Systems for Closed Loop Treatment of
Electroplating Waste Water. Final. Report prepared for the U.S.
• Department of Energy under DOE Contract No. DE-AC07-79CS4029P. March 17,
1986.
fr. Delu, H., L. Xiuchung, and W. Chingwen. The Regeneration of Nitric and
Hydrofluoric Acids From Waste Pickling Liquid. In: Symposium on Iron
and Steel Pollution Abatement Technology for 1980 held in Philadelphia,
Pennsylvania. November 18-20, 1980.
7. Solderman, J. New Method for Recovery of Spent Pickling Acids. In:
Third International Congress on Industrial Wastewaters and Wastes,
Stockholm, Sweden. February 6-8, 1980.
8. Dasher, J., and D. Goldstein. Recovery of Nitric and Hydrofluoric
Acids. Metal Finishing, 61(5): 60-63. May 1963.
9. Constantine, D. Corning Process Systems, Corning, New York. Technical
Data Sheet No. RT-1: Rinse Theory. May 12, 1980.
10. Williamson, R. C., and S. Natof. Evaporative Recovery for Cadmium
Cyanide Plating. Plating and Surface Finishing. November 1985.
5-26
-------
11. Brandywine, P. Mixed Rinses Treated by Evaporation. Product Finishing,
August 1984.
12. Industrial Finishing Staff. Recovering Brass Cyanide Plating Solution.
Industrial Finishing. June 1978.
13. Industrial Finishing Staff. Evaporator/Chiller System Recovers Brass
Plating Solution. Industrial Finishing. January 1980.
14. Rose, B. A. Design for Recovery. Industrial Finishing. May 1979.
15. Licon, Inc. Pensacola, Florida. Product Literature. Received August
1986.
16. Constantine, D. Corning Process Systems. Letter to J. Spielman, GCA
Technology Division, Inc. August 6, 1986.
5-27
-------
5.2 CRYSTALLIZATION
5.2.1 Process Description
Crystallization is a recovery technique whereby metal contaminants in a
spent corrosive solution are crystallized and removed by settling or
centrifugation. Crystallization techniques for the recovery of corrosive
wastes are most applicable to spent acid pickling liquors, and spent caustic
soda aluminum etching solutions.
Crystallization techniques are similar differing only in the methods used
to crystallize the metal salts and to separate the crystals from the recovered
solution. Crystallization can be induced by cooling or evaporating water from
wo.
1,2
the pickling solution, or a combination of the two. The process may be
operated in either a batch or a continuous mode.
Cooling crystallization techniques are used for the recovery of sulfuric
acid pickling liquors by removing iron contaminants. Typically, the process
involves crystallizing the iron salts in a cooling chamber, removal of the
crystals in a. settling/drainage chamber, and addition of fresh sulfuric acid
to restore the pickling solution to its original strength. ' -Thus, the
free sulfuric acid remaining in the spent pickling solution is able to be
reused.
A batch-mode process using cooling to induce crystallization is
diagrammed in Figure 5.2.1. Spent pickling solution is pumped to the
crystallizer. Recovered sulfuric acid from a previous batch operation
replaces the spent solution. Fresh sulfuric acid is added to the crystallizer
to increase the yield of ferrous sulfate crystals. Cooling water flows
through teflon-cooling coils around the crystallization chamber and is cycled
through a refrigeration unit. The temperature of the spent liquor is slowly
reduced (over a 6 to 10 hour period) to a temperature of 35 to 50°F, causing
1 2
ferrous sulfate heptahydrate crystals to form. '
The resulting crystal/acid slurry is then transferred to a drainage
chamber which retains the ferrous sulfate crystals, but allows the acid
solution to pass through to an acid recovery tank . The recovered
5-28
-------
r
_5PENT SULFUR 1C ACID PICKLE LIQUOR
CHILLED WATER
SFRIGERATION
UNIT
HEATED YATER
F*S04« 7H2
CRYSTALS
CRYSTALLIZES
DRAINAGE
TANK
RECOVERED
ACID
TANK
RECOVERED
ACID
PICKLING TANK
AIR
STEAM
I
CONDENSATE
Figure 5.2.1. Flow diagram of crystallization system for recovery
of stilfuric acid pickling liquor.
Source: Reference 1.
5-29
-------
acid is preheated wich steam and then returned to the pickling liquor tank
during Che next batch cycle. The ferrous sulfate crystals are washed with
water to remove any remaining free acid, air-dryed, and then removed for
disposal (or marketed, if possible).
Greater contaminant removal can be achieved using a combination of
4 5
evaporation and cooling techniques in a two-stage system . ' A flow
diagram of a two-stage process to recover nitric hydrofluoric acid pickling
liquors is presented in Figure 5.2.2.
During the first stage, approximately half of the waste pickle liquor is
evaporated in order to concentrate the dissolved metals (supersaturated
solution). The vapor resulting from the evaporation process is condensed, and
the acid condensate (the initial portion of regenerated acid) is directed to a
4
storage tank.
During the second stage of the process, the remaining solution is sent to
a crystallizing chamber. Cooling is used to induce crystallization of metal
fluoride crystals. The crystals are then separated from the solution in a
drainage chamber. The filtered concentrate, which contains 90 percent of the
nitric acid and the free hydrofluoric acid, is directed to the recovered acid
storage tank. The metal fluoride crystals can be disposed, or thermal
decomposition techniques' (discussed in Section*5.8) can be used to regenerate
hydrofluoric acid from the crystals and to form reusable metal oxides. The
recovered acid in Che storage tanks is Chen mete red to the pickling bath in
appropriate proportions.
A crystallization process has also been developed to recover sodium
hydroxide from spent aluminum etch solutions. Aluminum contaminants need .to
be removed in order to recover the sodium hydroxide (caustic soda).
Figure 5.2.3 presents a flow diagram for a continuous crystallization process
for recovery of caustic soda. In this process, a vacuum chamber is used to
induce crystallization of aluminum hydroxide. The pressure in the chamber
is reduced, which causes cooling of the spent etchant and some vaporization of
the water. The recovered acid, which has a high caustic concentration and a
low aluminum concentration, is returned to the etching tank. A centrifuge
is used to separate and wash the crystals formed in the crystallization
chamber. ' ' The filtrate from the centrifuge can also be returned to the
etchant tank. The dewatered crystals are of commercial-grade quality and
5-30
-------
PICKLING
TANK
SPENT ACID
STORAGE
TANK
EVAPORATOR
RECOVERED
ACID
STORAGE
TANK
J,
CRYSTALLIZATION
TANK
CONDENSER
DRAINAGE/
FILTER
TANK
r
CONDENSER
OPTIONAL
ROASTER
o
a.
o
METAL
OXIDE
STORAGE
O
H
a
o
OPTIONAL
I
Figure 5.2.2. Flow diagram of two-stage recovery system for
nitric-hydrofluoric acid pickling liquor.
Source: Reference 4.
-------
STEAM
WATER
REGENEP-flTED
ETCHRMT
SPEMT
ETCHRMT
o
ALKALINE
ETCHANT
\
ETCHING PUMP
TANK
CRYSTALLIZER
TANK
CENTRIFUGAL
SEPARATOR
FILTRATE PUMP
TANK
Figure 5.2.3. Flow diagram of crystallization system for the recovery
of caustic soda aluminum etching solution.
Source: Reference 1.
5-32
-------
can be traded or sold depending on the available market. Additions of fresh
sodium hydroxide are only required to replace dragout losses.
Operating Parameters—
Crystallization techniques for sulfuric acid recovery are based on the
solubility of ferrous sulfate decreasing with decreasing temperatures.
Figure 5.2.4 shows that the solubility of ferrous sulfate also decreases with
3
increasing acid concentrations. Therefore, more efficient operation of a
crystallization process is achieved when pickling lines have a high acid
concentration and a relatively low temperature.
Sulfuric acid recovery systems using crystallization techniques typically
cool the pickle liquor to approximately 40°F under controlled conditions to
3 8
form the ferrous sulfate heptahydrate crystal. ' At this low temperature,
most of the ferrous sulfate will be in the heptahydrate form rather than the
monohydrate form, which allows for easier removal. This crystallization
system works most efficiently in treating solutions with high iron
concentrations and high free acid concentrations. High acid concentrations
minimize the solubility of ferrous sulfate in the pickling solution, which
produces a lower iron content in the recovered acid. Also, the high iron
concentration allows less operating time for the acid recovery system and/or
the use of a smaller recovery system.
Typical operating parameters for a two-stage evaporation-
crystallization system to recover nitric-hydrofluoric acid pickling liquors
are presented in Table 5.2.1. The metal salt crystals are formed during the
cooling stage. The metal removal efficiencies achieved with this process
are directly dependent on the availability of fluoride ions in an amount
9
sufficient to combine with the metal ions. Therefore, instead of adding
the makeup hydrofluoric acid at the end of the process (to achieve the proper
pickling concentrations), it is desirable to add the makeup hydrofluoric acid
9
following the evaporation stage and prior to the crystallization stage.
However, nitric acid should not be added at this stage because a portion of
9
the acid will be lost with the crystals. The effect of nitric and
hydrofluoric acid concentrations on the solubility of iron fluoride crystals
over various temperature ranges is shown in Figure 5.2.5. Also, as illustrated
5-33
-------
PHASE CHANGE
10
7.5
2.5
50
88
86
104 122 140
TEMPERATURE
158
176
194
Figure 5.2.4. Solubility of ferrous sulfate in various
sulfurie acid concentrations.
Source: References 3 and 7.
5-34
-------
TABLE 5.2.1. EVAPORATION-CRYSTALLIZATION SYSTEM FOR
RECOVERY OF NITRIC-HYDROFLUORIC ACID
Parameter
Result
Operating temperatures
Spent pickle liquor
Evaporator
Vaporization
First crystallizer
Second crystallizer
Residence times
Evaporator
First crystallizer
Second crystallizer
40 to 60°C
105°C
107.5°C
70 to 788C
40 to 60°C
15 minutes
4 to 12 hrs
4 to 12 hrs
Source: References 9 and 10.
5-35
-------
\
~
-
c
--
TEMPERATURE (°C)
68 86 104 122 I4Q 158 176 194 212
60 70 80 90 100
40 50
TEMPERATURE (°P)
Figure 5.2.5 Solubility of iron in mixed acid containing
150 g/L (12.5 percent) nitric acid and
different amounts of free hydrofluoric acid
at varying temperatures.
Source: Reference 4.
5-36
-------
in Figure 5.2.6, the rate of crystallization of the iron fluoride crystals is
dependent on the initial concentration of iron present in the solution, which
is controlled by the evaporator.
Equipment used in each of these crystallization processes should be
resistant to corrosion. Commonly used materials include Teflon,
8 9
polyfluorohydrocarbons, sintered corundum, and structural graphite. '
Pretreatment Requirements—
For efficient operation of the crystallization recovery process, the
amount of water added to the pickling or etching system should be minimized.
Under optimum conditions, the water added to the system should not
significantly exceed the water exiting the system through evaporation, free
aoisture, and water of hydration of the crystals. Otherwise, the
recovered solution becomes too dilute for pickling/etching purposes. Some
methods which can be used to reduce water in the pickling/etching system
include: counter-current rinsing operations, indirect steam heating of the
pickling/etching tank, and air agitation of the bath to promote
evaporation.
Post-Treatment Requirements—
Upon completion of the acid recovery process, the purified acid or
etchant is returned to the pickling bath. The removed metals are in the form
of crystal salts, which will require further handling.
The iron removed in the single-stage cooling crystallization system for
recovery of sulfuric acid pickling liquor is in the form ferrous sulfate
heptahydrate crystals. These crystals can either be disposed, traded locally,
or marketed. The value of ferrous sulfate crystals on the market varies
widely. Crown Technology will purchase the iron sulfate crystals from their
clients.3'8 Crown dries the crystals, bags them, and then ships them for
use in water and sewage treatment facilities (flocculation processes),
8
fertilizer industry, and the animal feed industry.
The metal fluoride crystals formed in the two-stage evaporation-
crystallization system for recovery of nitric-hydrofluoric acid can be treated
by thermal decomposition techniques (discussed in Section 5.8) to recover
additional hydrofluoric acid. The thermal decomposition process generates a
•etal oxide product which can be reused in the steel process.
5-37
-------
ISO
50
4.5
234
TIME DELAY (HOURS)
m
Figure 5.2.6. Delay of formation of visible crystals in
oversaturated mixed acid versus initial
concentration of iron (150 g/L HNO3 and
20 g/L free hydrofluoric acid).
Source: Reference 4.
5-38
-------
5.2.2 Process Performance
The performance of a crystallization system is typically evaluated on the
basis of percent metal removal, percent acidity loss, product quality,
processing time, and economics.
Case studies demonstrating the performance of single-stage cooling
crystallization systems for the recovery of sulfuric acid are being prepared
by Acid Recovery Systems, Inc. in Lenexa, Kansas, but they are currently not
2
available. Typical performance data for single-stage acid recovery systems
are summarized in Table 5.2.2. Improvements are generally noted in product
quality when a continuous recovery process is employed due to the consistency
of the bath concentrations. Although the acid recovery process is typically
operated in a batch mode, interruptions to the pickling line are minimal since
bath dumping is no longer required.
Typical performance data for a two-stage evaporation-crystallization
system for the recovery of nitric-hydrofluoric acid are presented in
Table 5.2.3. Better performance is achieved with this system than the
evaporation-distillation systems described in Section 5.1 for
nitric-hydrofluoric acid recovery because of reduced sludge generation (i.e.,
sulfuric acid additions with subsequent neutralization "treatment are not
required).
Limited performance data is available for crystallization systems used to
recover caustic soda. However, recoveries of up to 80% have been reported for
this process.
5.2.3 Costs
Sulfuric acid recovery systems are available with throughput rates
3 8
ranging from 600 gpd to 30,000 gpd. ' Capital costs will include
crystallization equipment, two tanks (customer-supplied); one to hold the high
iron content solution, and one to hold the low iron content solution.
Additional requirements include: connections for hooking up the system to the
product line, electrical requirements, and heating coils for the tanks.
5-39
-------
TABLE 5.2.2. TYPICAL OPERATING PARAMETERS AND RESULTS
FOR SOLFURIC ACID RECOVERY SYSTEM USING
CRYSTALLIZATION
Parameter Result
Optimum iron content in the waste feed 10 to 14Z
Iron removal efficiency 80 to 85Z
Acidity losses in recovered acid 2 to 31
Average cycle time 6 hrs
Source: Reference 2.
5-40
-------
TABLE 5.2.3. TYPICAL PERFORMANCE OF A TWO-STAGE CRYSTALLIZATION
SYSTEM FOR THE RECOVERY OF NITRIC-HYDROFLUORIC ACID
Concentration, weight-percent (Ibs/hr)
Parameter
Feed to evaporator
Feed to crystallizer
Condensed vapor
Residue ' from
'crystallizer
Filtrate from
crystallizer
Total concentration
recovered
Total required
- additions
Fe
3.4
(26.5)
6.5
(26.5)
-
25
(20.0)
2.0
(6.5)
0.9
(6.5)
—
Cr
1.1
(8.6)
2.1
(8.6)
-
4.6
(3.7)
1.5
(4.9)
0.7
(4.9)
—
Ni
1.6
(12.5)
3.1
(12.5)
-
0.8
(0.6)
3.5
(11.9)
1.7
(11.9)
—
N03
12.0
(93.6)
22.1
(89.9)
1
(3.7)
6.0
(4.8)
26.0
(15.1)
12.7
(88.8)
(43)
F
6.0
(46.8)
10.1
(41.2)
1.5
(5.6)
30.8
(24.6)
5.1
(16.6)
3.2
(22.2)
(32)
Water
75.9
(592)
56.1
(228.3)
97.5
(363.7)
32.9
(26.3)
61.7
(202.0)
80.8
(565.7)
(261.1)
Source: Reference 9.
5-41
-------
Table 5.2.4 presents cost estimates for three acid recovery systems with
varying throughput rates. The economics.of sulfuric acid recovery varies
considerably with the costs of acids, the market for ferrous sulfate, and the
costs for disposal. '
Two—stage systems are generally not cost-effective for recovering
sulfuric acid pickling liquors due to the high capital equipment costs.
However, two-stage systems can be cost-effective for recovering hydrofluoric-
trie acid pickling liquors, because of the higher acid purchase costs for
12
these acids. An economic evaluation of the two-stage system for recovery
of nitric-hydrofluoric pickling liquors is presented in Table 5.2.5. Costs
are given for a system with a regeneration capacity of 2,000 L/hr
(530 gal/hr), which was designed for a steel plant with a pickling capacity of
40 mL/hr (400 gal/hr).
5.2.4 Process Status
Crystallization is a demonstrated and commercially available technology
for the recovery of acid pickling liquors and caustic etching
solutions. ' ' ' ' Due to the large capital investment costs, the process
is more economically feasible for use in operations that generate larger
quantities of spent solutions.
The use of crystallization techniques for the recovery of sulfuric acid
pickling liquors and caustic aluminium etching solutions is limited by
economics due to the small quantities of these solutions used by individual
manufacturers, the costs for plan modifications, and the varying demand for
the crystal product. * Despite these limitations, these processes are
currently being used in the metal finishing industry.
Nitric-hydrofluoric pickling liquors are used in larger quantities by
individual manufacturers in the steel industry than sulfuric acid pickling
liquors. Therefore, crystallization techniques would have wider application
for this waste type.
Due to economics, acid and alkaline wastes are normally treated by
neutralization techniques and land disposal. With increasing restrictions on
land disposal, recovery using crystallization techniques may become more
economically viable.
5-42
-------
TABLE 5.2.4. ECONOMIC EVALUATION OF ACID RECOVERY SYSTEM
USING CRYSTALLIZATION TECHNIQUE
Item
Flow rate (gal/day)
CAPITAL COSTS
Equipment
Tank (2 tanks @ $1.25/gal)
Installation (10% of investment)
Total capital costs:
OPERATING COSTS
Maintenance (62 of investment)
Taxes & insurance
(0.52 of investment)
Utilities (@ $0.02/KW-h)
Depreciation (102 of investment)
Total operating costs:
COST SAVINGS
Neutralization savings
Disposal savings
Process water savings
Acid makeup savings
Total cost savings:
NET SAVINGS:
(Gross savings-Operating costs)
PAYBACK PERIOD
(Capital costs/Net savings)
Small unit Medium unit Large unit
(S) ($) ($)
2,400
175,000
5,000
17,500
197,500
10,500
875
8,000
17,500
36,875
22,653
51,025
2,633
16,250
92,560
55,685
3.59 yrs
(43 months)
16,000
460,000
40,000
46,000
546,000
27,600
2,300
10,000
46,000
85,900
139,400
314,000
16,200
100,000
569,600
483^700
;
;
j
i.16 yrs
(14 months)
30,000
850,000
75,000
85,000
1,010,000
51,000
4,250
12,000
85,000
152,250
261,375
588,750
30,375
187,500
1,068,000
915,750
1.14 yi
(14 months)
Source: References 1, 2, 3 and 12 (July, 1986 cost data).
5-43
-------
TABLE 5.2.5. ECONOMIC EVALUATION OF TWO-STAGE SYSTEM
FOR RECOVERY OF NITRIC-HYDROFLUORIC ACID
Item
Cose ($)
Capital costs
Operating costs
Cost savings
Net savings
(Gross savings-Operating costs)
Payback period
(Capital costs/Net savings)
10,000,000
2,064,000/year
5,222,000/year
3,158,000/year
3 years
Source: References 4 and 12.
5-44
-------
REFERENCES
1. Camp, Dresser, and McKee, Inc. Technical Assessment of Treatment
Alternatives for Wastes Containing Corrosives. Prepared for the U.S. EPA
Office of Solid Waste under EPA Contract No. 68-01-6403 (Work Assignment
No. 39). September 1984.
2. Luhrs, R. Acid Recovery Systems, Inc., Lenexa, Kansas. Telephone
conversation with L. Wilk, GCA Technology Division, Inc. Re: Sulfuric
Acid Recovery System. September 4, 1986.
3. Crown Technology, Inc. Product Literature: Crown Acid Recovery
Systems. Received July 1986.
4. Krepler, A. Total Regeneration of the Waste Pickle Liquor for Stainless
Steel. Ruthner Industrieanlagen Aktiengesellschaft Technical Report
No. 3, Vienna, Austria. 1980.
5. Smith, I., Cameron, G. M., and H. C. Peterson. Chemetics International
Co..Toronto, Canada. Acid Recovery Cuts Waste Output. Chemical
Engineering. February 3, 1986.
6. Versar, Inc. National Profiles Report for Recycling/A Preliminary
Assessment. Draft Report prepared for the U.S. EPA Waste Treatment
Branch under EPA Contract No. 68-01-7053, Work Assignment No. 17. July
8, 1985.
7. Peterson, J. C. Crown Technology, Inc. Closed Loop System for the
Treatment of Waste Pickle Liquor. Prepared for the U.S. EPA-Industrial
Environmental Research Laboratory, Research Triangle Park, North
Carolina. EPA-600/2-77-127. July 1977.
8. Peterson, J. C. Crown Technology Inc., Indianapolis, Indiana. Telephone
conversation with L. Wilk, GCA Technology Division, Inc. Re: Sulfuric
Acid Recovery System. July 10, 1986.
9. Rrepler, A. Apparatus for Recovering Nitric Acid and Hydrofluoric Acid
from Solutions. U.S. Patent No. 4,252,602. Assigned to Ruthner
Industrianlagen-Aktiengese11schaft, Vienna, Austria. February 24, 1981.
10. Krepler, A. Process for Regenerating a Nitric Acid-Hydrofluoric Acid
Pickling Liquor. U.S. Patent No. 4,144,092. Assigned to Ruthner
Industrieanlagen-Aktiengesellschaft, Vienna, Austria. March 13, 1979.
5-45
-------
11. Micheletti, W. C., Nassos, P. A., and K. T. Sherrill. Radian
Corporation. Spent Sulfuric Acid Pickle Liquor Recovery Alternatives and
By-Product Dses. Draft Report prepared for U. S. EPA-IERL, Research
Triangle Park, North Carolina. DCN-80-203-001-17-07. November 3, 1980.
12. Chemical Marketing Reporter. Spot Market Prices. Volume 230, No. 3,
Pages 32-40. July 21, 1986.
5-46
-------
5.3 ION EXCHANGE
5.3.1 Process Description
Ion exchange has been used to recover corrosive wastes from the metal
finishing, electroplating, and fertilizer manufacturing industries by removing
metal contaminants and recycling the treated solution.1 Industrial process
waters, plating baths, and acid pickling baths contain dissolved metal salts
which dissociate to form metal ions. Ion exchange is a reversible process
which involves an interchange of these ions between the solution and an
essentially insoluble, solid resin in contact with the solution. Ions from
the solution are exchanged for similarly charged ions attached to the solid
resin.
The maximum quantity of exchanges per unit of resin is set by the number
of mobile ion sites (exchangeable ions) attached to the resin. This is
dependent on the type of resin, which can be composed of either naturally
occurring materials (e.g., clays or zeolites) or synthetic organic polymers.
'Synthetic resins are more frequently employed .because they can be designed for
specific applications.
Typical equipment used in an ion exchange treatment system includes: a
waste storage tank, prefilter system, cation and/or anion exchanger vessels,
2
and caustic or acid regeneration equipment. The ion exchange system may be
operated in a batch or flow-through (column) mode; the latter mode is
generally preferred due to greater exchange efficiencies.
With the batch mode of operation, the ion exchange resin and the waste
solution are mixed in a batch tank. Upon completion of the exchange reaction
(i.e., equilibrium is reached), the resin is separated from the treated
solution by filtering or settling. The spent resin is then regenerated and
reused. Unless the resin has a very high affinity for the contaminant ion,
the batch mode of operation is chemically inefficient and thus has limited
applications.
Flow-through operation involves the use of a bed or packed column of the
exchange material (resin). These systems are typically operated in cycles
consisting of the following four steps:
5-47
-------
1. Service (Exhaustion) - Waste solution is passed through the ion
exchange column or bed until the exchange sites are exhausted.
2. Backwash - The bed is washed (generally with water) in the reverse
direction of the service cycle in order to expand and resettle the
resin bed.
3.- Regeneration - The exchanger is regenerated by passing a
concentrated solution of the ion originally associated with it
through the resin bed or column; usually a strong mineral acid or
base.
4. Rinse - Excess regenerant is removed from the exchanger; usually by
passing water through it.
A flow-through (column) system can be designed with cocurrent or
countercurrrent flow of the waste and regenerant (steps 1 and 3 above). In
cocurrent systems, the feed and the regenerant both pass through the resin in
a downflow mode. Figure 5.3.1 illustrates the cocurrent flow process. Each
ion exchange unit consists of a cylindrical vessel having distributors or
collectors at the top and bottom. Resin is loaded into approximately half of
the vessel to accommodate resin expansion during the backwash cycle.
Cocurrent-systems are most cost-effective for weak acid or base exchangers
which do not require highly concentrated regenerant solutions. However,
regeneration of strong exchangers (high exchange capacity) requires strong
acid and base solutions which can be more costly.
Often it is too costly to fully regenerate a bed. In order to avoid
carry over of contaminants into the next service run, two or more sets of
fixed columns arranged in parallel series can be used. Also, to avoid
excessive down—time during the regeneration cycle, dual sets of fixed columns
are generally used. While one set of columns is being regenerated, the second
set of columns will be switched on line. This technique allows continuous
operation of the system. As illustrated in Figure 5.3.2, improved regenerant
efficiency can also be accomplished by reusing the last portion of the
regenerant solution that flows through the resin. For example, if 5 Ib/cu ft
(80 g/L) of regenerant were used for the system shown in the figure, the first
50 percent of spent regenerant would only contain 29 percent of the original
acid concentration, whereas the remaining regenerant would contain 78 percent
4
of the original acid. If the last portion of the regenerant is
5-48
-------
WASTE
SOLUTION
X
vO
TREATED
EFFLUENT
CONTAMINATED
BACKWASH
I— WATER
STRONG
ACID/ALKALI
CONTAMINATED
REGENERATE
WATER
CONTAMINATED
RINSE
.SERVICE
BACKWASH
REGENERATION
RINSE
Figure 5.3.1. Cocurrent ion exchange cycle.
Source: Reference 3.
-------
100 r-
80
K
o
g 60
i
o
V
40
O
55
tr
o
o 20
LEGEND:
•••• with acid reuse
— ——without acid reuse
I
2468
REGENERANT CONSUMED (Ib. HCl/ft3)
Based on strong acid resin in calcium form.
10
Figure 5.3.2. Effect of acid regeneration on chemical efficiency.
Source: Reference 4.
5-50
-------
reused in the next cycle before the resin bed is contacted with fresh HC1, the
•exchange capacity would increase from 60 to 67 percent at equal chemical
doses.
Ion exchange systems generate a waste stream of spent regenerant, which
will typically require neutralization and disposal. The use of countercurrent
flow between feed and regenerant uses regenerant chemicals more efficiently
than is possible with cocurrent systems. Counter current systems also achieve
a higher product concentration than is possible with conventional cocurrent
flow. A counter current system that is widely used for chemical recovery from
plating rinses is the reverse or reciprocating flow ion exchanger (RIFE). An
example of an RF1E system used for recovery of chromic acid from a dilute
solution is shown in Figure 5.3.3.
Eco-Tech, Ltd. (in Pickering, Ontario) has developed an acid purification
system that uses the RFIE technology to efficiently remove high concentrations
of metal contaminants from acid baths. Although the acid purification unit
(APU) uses countercurrent flow techniques, the bed is not fixed. As shown in
Figure 5.3.4, the acid purification process consists of an upstroke adsorption
cycle and a downstroke regeneration cycle. During the upstroke cycle, spent
acid is forced by air pressure through the resin bed. The acid is adsorbed
and the de-acidified metallic salts are collected from the top of the resin
bed. During the downstroke regeneration cycle, compressed air is used to
force water into the top of the resin bed. The water passes countercurrently
through the bed, displacing the adsorbed acid from the resin. The purified
acid product is collected from the bottom of the resin bed and recycled into
the process operation. The only waste product from the acid purification
process is a metallic salt sludge. ' '
Typical components of the APU include the resin bed, metering tanks, an
electrical control panel, plastic piping, and pneumatically operated plastic
control valves. The size of the unit varies from 6 to 24 square feet with a
6
height of less than 6 feet.
Although the conventional cocurrent flow ion exchange systems can be used
effectively to recover process rinsewaters, RFIE systems have wider
application in the recovery of corrosive wastes.
5-51
-------
1 —
ON STREAM ILOAOINBI
- REOENERATION 1
RINSE
WATER
j
CATKM
V
&
" H"
1
4 ANION
1
t
Biiatrirn
1
f
riLTER
zi
f
CATION
EXHAUST WATER
PRODUCT
t
CATION
L_
EXHAUST
TREATMENT)
t
ANION CATION
.1
ITO WASTE
t
j
CATION
L
~1 t
ANION CATION
t
—^ri?*
TREATMENT! -^CHBH "• SHE:
jyy~l —*-szr
NiOH
NoOH
Figure 5.3.3. Schematic of a fixed bed reverse flow ion exchange (RFIE) system
for the recovery of chromic acid from a dilute solution.
Source: Reference 4.
-------
WATER
UPSTROKE
1
COMPRESSED AIR
*
4 4 t
1
RESIN BED
4
r -^
\ \ t
SPENT
ACID
(FEED)
COMPRESSED AIR SPENT ACID (FEED)
I DOWNSTROKE
\ \ \
WATER
RESIN BED
PURIFIED ACID (PRODUCT)
4 4 4
Figure 5.3.4. Basic operation of the acid purification unit (APU)
using a continuous bed RFIE system.
Sources: References 7 and 9.
5-53
-------
Operating and Design Parameters—
The most significant design parameter in an ion exchange system is the
selection of an appropriate resin. Resin selection is based on the type of
ion exchanger, the strength of the resin, its exchange capacity and
selectivity, and the volume required.
Resins can be classified as acid cation exchangers or base anion
exchangers. Cation exchangers have positively charged exchangeable ions, and
anion exchangers have negatively charged exchangeable ions. Heavy metal
selective chelating resins are weak acid cation resins that exhibit a high
affinity for heavy metal cations.
• The exchange capacity of a resin is generally expressed as equivalents
per liter (eg/L), where an equivalent is equal to the molecular weight of the
4
ion in grams divided by its electrical charge or valence. For example, a
resin with an exchange capacity of 1 eq/L could remove 37.5 g of divalent zinc
+2
(Zn , molecular weight " 65 g) from solution.
The strength of a resin is determined by the degree to which its
exchangeable ions dissociate in solution. The exchangeable ions of strong
acid and strong base resins are highly dissociated and therefore are readily
available for exchange* and are unaffected by solution pH. Conversely, the
exchange capacity of weak acid and weak base resins is strongly influenced by
solution pH.
Ion exchange reactions are stoichiometric and reversible. A generalized
form of an ion exchange reaction can be depicted as follows:
R-A+ + B* ^ R-B+ + A*
where R is the resin, A is the ion originally associated with the resin,
and B is the ion originally in solution. The degree to which the exchange
reaction proceeds is dependent on the preference (or selectivity) of the resin
for the exchanged ion. The selectivity coefficient, K, is used to measure the
preference of a resin for a particular ion. It expresses the relative
distribution of ions when a charged resin is contacted with solutions of
different (but similarly charged) ions. For example, in the generalized ion
exchange reaction presented above, the selectivity coefficient (K) is defined
as follows:
5-54
-------
[B ] in resin [A ] in solution
K
[A ] in resin [B ] in solution
The selectivity coefficient of a resin will vary with changes in solution
characteristics and the strength of the resin. Table 5.3.1 shows the
selectivites of strong acid and strong base resins for various ionic species.
Operating parameters will vary with the particular application. The
following factors will influence the selection of a resin type, pretreatment
requirements, flow rates, cycle times, and the sizing of a system for a
particular application:
• Types and concentrations of constituents present in the feed
• Rate of metal salt accumulation in the bath
• Flow rate
• Number of hours of operation.
The types and concentrations of constituents present in the spent solution
will determine the type of resin selected. For corrosive wastes, these
constituents will typically be metal salts. Weak cation exchangers can be
used for spent solutions containing low concentrations of metal ions. For
solutions containing high concentrations of metal ions, a strong anion
»
exchanger can be used effectively in an APU.
The constituent concentrations will also determine the resin volume
needed to treat the stream; larger concentrations will generally require a
larger volume of resin to reduce regeneration frequency. The system size will
also increase along with the rate of contaminant accumulation in the bath and
the volume of solution to be treated per unit time. Smaller systems can
generally be used if the process is operated for a longer period of time due
to stabilization of the bath. Commericially available systems are able to
process wastes at throughput rates ranging from 38 to 6,700 liters per
hour.11 Cycle times for RFIE systems generally range from 5 to 15
minutes.11'12, whereas for cocurrent systems they can be as much as 1 to 2
hours because of the time needed to regenerate the column. As a result,
dual sets of columns are typically used in cocurrent systems to avoid
excessive downtime.
5-55
-------
TABLE 5.3.1.
SELECTIVITIES OF ION EXCHANGE RESINS IN ORDER OF
DECREASING PREFERENCES3
Strong acid
cation
exchanger
Strong base
anion
exchanger
Weak acid
cation
exchanger
Weak base
anion
exchanger
Weak acid
chelate
exchanger
Barium (+2)
Lead (+2)
Mercury (+2)
Copper (+1)
Calcium (+2) '
Nickel (+2)
Cadmium (+2)
Copper (+2)
Cobalt (+2)
Zinc (+2)
Cesium (+1)
Iron (+2)
Magnesium (+2)
Potassium (•*•!)
Manganese (+2)
Ammonia (•*•!)
Sodium (+1)
Hydrogen (+1)
Lithium (+1)
Iodide (-1)
Nitrate (-1)
Bisulfite (-1)
Chloride (-1)
Cyanide (-1)
Bicarbonate (-1)
Hydroxide (-1)
Fluoride (-1)
Sulfate (-2)
Hydrogen (+1)
Copper (+2)
Cobalt- (+2)
Nickel (+2)
Calcium (+2)
Magnesium (+2)
Sodium (+2)
Hydroxide (-1) Copper (+2)
Sulfate (-2) Iron (+2)
Chromate (-2) Nickel (+2)
Phosphate (-2) Lead (+2)
Chloride (-1) Manganese (+2)
Calcium (+2)
Magnesium (+2)
Sodium (+1)
aValence number is given in parentheses.
Source: References 4 and 10.
5-56
-------
Pretreatment Requirements/Restrictive Waste Characteristics—
Pretreatment of the waste stream (usually via filtration) is necessary to
remove any constituents which would adversely affect the resin. Certain
organics (e.g., aromatics) become irreversibly sorbed by the resin. Oxidants
(such as chromic or nitric acid) can also damage the resin. Sodium
oetabisulfite, which converts hexavalent chrome to its triavalent state, can
be added to the solution to prevent damage to the resin. Eco-Tec has
stated that resin degradation is not a problem with the RFIE process due to
the short duration of contact (approximately. 1.5 mins) between the acid and
9 1A
the resin. 'iH
Ion exchange units are only able to treat contaminants in solution.
High concentrations of suspended solids which can foul the resin bed are
typically pretreated through some form of filtering. Examples of filters
which may be employed include activated carbon, deep bed, diatomaceous earth
precoat, and resins. The filters eventually become clogged with particulates,
and are replaced when overall cycle time decreases due to the increased time
required for passage of the solution into the ion exchange unit. The
filter size and replacement frequency will depend on the quantity of suspended
particulates passed through the filter per unit time.
For large volume systems, which require more frequent changing of the
filter cartridges, it may be more cost-effective to use a multimedia sand
filter with a backwashing system to regenerate the filter. Although initial
capital costs for this type of filtration system are higher, savings in filter
replacement costs will be realized.
The use of weak acid and weak base exchangers for treating corrosive
wastes will require additional pretreatment. The exchange capacity of weak
acid exchangers is limited below pH 6.0, and weak base exchangers are not
effective above pH 7.4 Therefore, a pH adjustment system must be
incorporated prior to feeding the waste stream through weak ion exchangers.
Although weak exchangers cannot be used directly for the treatment of
corrosive wastes, they can be used to reduce the costs of neutralization
treatment by eliminating the need for over-neutralization (discussed in
Section 4) to remove the metals.
Ion exchange using cocurrent flow is not suitable for removal of high
concentrations of exchangeable ions (above 2,500 mg/L expressed as calcium
carbonate equivalents). The resin material is rapidly exhausted during the
5-57
-------
4 15
. exchange process and regeneration becomes prohibitively expensive. '
However, the reverse is true for acid purification units since they are
capable of recycling the regenerate. In addition, higher concentrations of
metal contaminants present in the solution will improve removal efficiencies
(see Section 5.3.2).
Post-Treatment Requirements--
Waste streams from the ion exchange process include: spent regenerant
solution, byproduct stream, and filtrate from the pre-filtering system. Spent
regenerant, produced when cocurrent flow ion exchange is used, generally
requires neutralization and disposal. Optimal neutralization methods will
depend upon the waste type (refer to Section 4). The byproduct may require
treatment and disposal, or in some cases the recovered metal can be reused or
marketed. Filtrate from the pre-filtering system can generally be land
disposed without further treatment. The quantities of wastes generated
will depend on the types and concentrations of contaminants present in the
solution being treated. However, the total volume of wastes from the ion
exchange system requiring treatment or disposal will be less than if a
recovery system was not employed. Thus, an overall savings in treatment and
disposal costs is realized.
5.3.2 Process Performance
The performance of an ion exchange system will be predominantly
influenced by the characteristics and quantity of the waste stream being
treated. Parameters which need to be considered when evaluating the
applicability of the system for a particular waste stream include: types and
concentrations of constituents present in the waste stream, acidity of the
spent stream versus acidity of the fresh stream, and required quality of the
recovered stream.
Additional factors which can be used to evaluate the performance Of an
ion exchange system include: the quantity of byproduct streams generated,
cycle times,- product concentration, process modifications required, flow
rates, processing capabilities (system size), and costs. •
5-58
-------
Cocurrent flow generates an additional corrosive waste stream which
requires treatment and disposal. Cocurrent flow methods are inefficient for
the treatment of corrosive wastes and are generally not used for this
purpose. Reciprocating Flow Ion Exchange (RFIE) units are more cost-effective
than cocurrent fixed-bed systems for use with corrosive wastes. RFIE units
use smaller resin volumes, and therefore have lower capital costs and space
requirements. Also, operating costs are lower due to the reuse of
regenerant. Automation of the process operation provides additional savings
in labor requirements. In addition, these units are able to generate a higher
product concentration when used for chemical and acid bath recovery.
Acid purification systems using RFIE have been commercially demonstrated
to be effective in the recovery of acids from aluminum anodizing solutions,
acid pickling liquors, and rack-stripping solutions. Acid purification
systems are most effective in recovering acids which have high concentrations
of metal ion contaminants.
An acid purification unit using RFIE technology was installed at the
Continuous Colour Coat, Ltd. plant in Rexdale, Ontario, to recover sulfuric
acid from a steel pickling process. During pickling, as iron buildup
begins to cause poor product quality, acid additions are made to the bath or
the tank is drained and a fresh acid solution is prepared. As an alternative,
to neutralization and disposal of spent baths, Colour Coat, Ltd. employed an
APU to remove the iron so that the solution could be recycled.
Typical operating parameters and results for the APU application are
presented in Table 5.3.2. As shown, iron concentration in the reclaimed acid
was reduced by 80 percent. Also, acidity and acid volume losses were
minimal. Occasional replenishment of the bath was necessary, but draining of
the tank (an expensive process) was no longer required. Also, improvements in
product quality were noted due to the consistency of the bath concentration
that was maintained during the APU operation.
Savings were realized in neutralization and disposal requirements. Also,
there were net reductions in labor requirements due to reduced bath
maintenance. An economic evaluation of the system (Table 5.3.3.) showed an
estimated payback period for the unit of less than 2 years. -
5-59
-------
TABLE 5.3.2. TYPICAL OPERATING PARAMETERS AND RESULTS FOR THE APU
INSTALLED AT CONTINUOUS COLOUR COAT, LTD. IN
* REXDALE, ONTARIO
Parameter
Flow Rate (gal/min)
Temperature (°F)
2-Sulfuric Acid
2- Iron
Feed acid
19
119
15.8
3.0
By product
N/A
N/A
0.5
2.6
Return acid
19
119
15.7
0.6
Note: N/A - Not applicable.
Source: Reference 11.
5-60
-------
TABLE 5.3.3. ECONOMIC EVALUATION OF THE APU INSTALLED AT
CONTINUOUS COLOUR COAT, LTD.
Item
Cost
CAPITAL COSTS*
(includes costs for equipment & installation)
OPERATING COSTS
Resin Replacement
(every 4 years at $58/liter)
Utilities
(0.5 KW x 16 hrs/day x 250 days/yr x $0.055/KWH)
Taxes and Insurance
(12 of TIC)
TOTAL OPERATING COSTS
COST SAVINGS
Reduction in Acid Purchase @ $92.40/ton
Reduction in Neutralization Costs (Lime) @ $80/ton
Reduction in Sludge Disposal Costs
TOTAL COST SAVINGS
RET COST SAVINGS
(Gross Savings - Operating Costs)
PAYBACK PERIOD
(Capital Costs + Net Savings)
$100,000
$3,770/year
S 110/year
Sl.OOO/year
$4,880/year
$25,875/year
$18,000/year
$20,000/year
$63,875/year
$58,995/year
1.7 years
(approx. 20 months)
Capital Equipment included APU Model No. AP30-24, multimedia sand
filter, water supply tank., and piping.
Source: Reference 11 (As Ecotech cost quote, August 1986).
5-61
-------
Another common application in which the APU works effectively is the
recovery of acids from aluminum anodizing solutions. An APU system was
installed at the Mod in e Manufacturing Company in Racine, Wisconsin to recover
nitric acid from an aluminum etching process. The APU was connected
directly on-line, which allowed for continuous process operation. The APU
generated a more concentrated solution of recovered nitric acid than it was
fed; thus a slightly lesser volume of acid was returned to the tank. After a
certain period of time, the total acid volume in the tank was reduced enough
so that additional concentrated acid could be added without draining the
tank. Table 5.3.4 presents a summary of the results and operating parameters
of the APU'at the Mod in e plant. Improvements in product quality and savings
in neutralization, disposal, and fresh acid makeup were noted by Modine
personnel. An economic evaluation of this system (Table 5.3.5.) shows
nitric acid recovery with an APU to be very cost-effective.
A full-scale demonstration of the APU in recovering sulfuric acid from an
aluminum anodizing .solution was performed at Springfield Machine and Stamping,
• Inc. of.Warren, 'Michigan. typical operating.parameters and results during
the 6 month testing period are-summarized in Table 5.3.6. The system proved
to be cost-effective due to the high aluminum removal efficiency, retention of
acid strength, and reductions in raw material purchase, disposal and labor
costs.
•
A pilot scale unit for recovering hydrochloric acid from an
electroplating pickling liquor was tested at Electroplating Engineering, Inc.
in St. Paul, Minnesota. The results were not as successful as with the
other cases presented. Reduced metal removal efficiencies and high acidity
losses were experienced.
Typical concentrations of metals in the new, intermediate, and spent
pickling solution at this plant are presented in Table 5.3.7. As shown, zinc
and iron are the primary constituents of concern. A different system
configuration was required for this application because the zinc was present
in the form of zinc chloride bomplexes. As described in Section 5.3.1, the
resin used in an APU shows a/preferential affinity for acid anions as opposed
to metal cations, which causes metals to pass through the resin while the acid
is retained. However, instead of passing through the resin, zinc chloride
complexes will also be retained by the anion resin.
5-62
-------
TABLE 5.3.4. TYPICAL OPERATING PARAMETERS AND RESULTS DURING TESTING
OF THE APU FOR RECOVERY OF NITRIC ACID AT MODINE
MANUFACTURING COMPANY IN RACINE, WISCONSIN
Parameter Result
Feed to APU from etch tank 6.2 N
Product returned to etch tank 6.5 N
By-product going to waste treatment • 0.6 N
Level of aluminum contamination:
Coming into APU from etch tank 793 mg/L
Returning to the etch tank 231 mg/L
Average cycle t i~e 12-7
Volume of watar rsnoved from etch tank/ APU cycle 0.89 gal
Mass balance
Equivalents : : nitric acid into APU from etch tank 251
Equivalents returned to etch tank and waste -257
Source: References 6 and 7.
5-63
-------
TABLE 5.3.5. ECONOMIC EVALUATION OF THE APU INSTALLED AT MODINE
MANUFACTURING CO. IN RACINE, WISCONSIN FOR THE
RECOVERY OF NITRIC ACID
Item
Cost
CAPITAL COSTS
(includes costs for equipment and installation)
COST SAVINGS
Reduction in nitric acid purchase
Reduction in neutralization costs
Reduction in disposal costs
Reduction in labor
TOTAL COST SAVINGS
OPERATING COSTS
Resin replacement
(every 4 years at $58/1iter)
Utilities
(0.5 KW x 16 hrs/day x 350 days/yr x J0.055/KWH)
Taxes and insurance
(1Z of TIC)
TOTAL OPERATING COSTS
NET COST SAVINGS
(Gross savings - Operating costs)
PAYBACK PERIOD
(Capital cost-7-Net savings)
$37,234
520,064/year
$ 6,276/year
$ 7,236/year
$ 2,400/year
$35,976/year
* 1,305/year
$ 132/year
* 372/year
* 1,809/year
$34,167/year
1.1 years
(approx. 13 months)
Source: References 7 and 11 using August 1986 cost data.
5-64
-------
TABLE 5.3.6. TYPICAL OPERATING PARAMETERS AND RESULTS FOR THE APU
INSTALLED AT SPRINGFIELD MACHINE & STAMPING, INC.
IN WARREN, MICHIGAN FOR SULFURIC ACID RECOVERY
Parameter Feed Product By-product
Flow rate (liters/hr) 298 296 175
Sulfuric acid concentration (g/L) 183.8 175.0 13.0
Aluminum concentration (g/L) 12.2 4.2 12.0
Source: Reference 8.
5-65
-------
TABLE 5.3.7 AVERAGE CONSTITUENT CONCENTRATIONS IN THE NEW,
INTERMEDIATE, AND SPENT BATH SOLUTIONS
New Intermediate Spent
Parameter solution solution solution
Acidity, mg/L CaC03 246,333 168,667 112,833
Iron concentration (mg/L) 219 1,333 1,650
Zinc concentration (mg/L) 1,052 3,693 4,283
Nickel concentration (mg/L) ND ND 1.3
Copper concentration (mg/L) ND ND 0.6
Chrome concentration (mg/L) ND ND 6.3
Source: Reference 13.
5-66
-------
In order to remove both the zinc chloride complex and the iron
contaminants, it was necessary to operate the system in two stages.
Initially, the spent solution was passed through one resin to remove zinc
(termed the inverse mode since the acid ions are not retarded). Then the acid
ions which have passed through the first resin are passed over a second resin
in the normal mode of operation, retaining the acid while allowing the iron
ions to pass through the resin. As with typical AFU processes, the acid is
recovered during the regeneration cycle. Figures 5.3.5 and 5.3.6 illustrate
these two modes of operation.
Pickling solutions from three different HC1 pickling liquors were used to
test the performance of the acid purification system. Analysis of these spent
solutions yielded the following:
Parameter Range of Concentrations
Acidity 77,000 - 284,000 mg/L
(expressed as CaCC^)
Zinc Content 640 - 52,000 mg/L
Iron Content 1,100 - 7,000 mg/L
Several test runs were performed using the acid purification unit as
summarized in Table 5.3.8. The results show that good zinc removal
efficiencies (99.3 percent)-were achieved during the inverse mode of operation
with minimal losses in acidity (3.5 percent). However, during the normal mode
of operation, only an average of 60 percent iron removal was achieved and
acidity losses were quite high (averaging 38 percent). The results showed
that increased iron removal could only be achieved at the expense of greater
reductions in acidity. It was determined that the ratio of iron to acidity in
the feed has to approach 1:15 in order to achieve effective performance. The
iron to acidity ratio for the feed used during these tests was 1:67, which
contributed to the poor performance results. ;
It is possible that the intermediate byproduct solution generated after
the inverse mode may be of sufficient quality to be returned to the pickling
bath.13 Iron content of the byproduct solution was comparable to the iron
concentrations measured in the bath during its intermediate solution age
(Table-5-3.7). Additional testing would be required in order to determine
whether batn quality would be acceptable under these conditions.
5-67
-------
STEP ONE - WATER DISPLACEMENT
WATER.
4
»
&ESIIN
&SI8
I
SPENT
>—
ACID
Compressed air
Spent Acid (SA) displaces
water from resin void
volume.
STEP TWO - INTERMEDIATE BYPRODUCT GENERATION
Water Replenished
Intermediate Byproduct (IB) to reservoir
•Compressed air
SPENT ACID Intermediate Byproduct (IB) is
routed to IB Reservoir.
Water reservoir is refilled
at beginning of step.
STEP THREE - SPENT ACID DISPLACEMENT
Compressed a! r
••«••
WATER
1
^
»
ESIN
9 ED
:-:^-|
SPENT
ACID
Water displaces
Spent Acid from resin void
volume.
STEP FOUR - INTERMEDIATE WASTE PRODUCT GENERATION
Compressed air-
II ^
—
WATER
i
r
IRES IN
&B.ED
F*=
SPENT
1
is produced.
Kl° Spent Acid reservoir is
refilled at beginning
of step.
Intermediate Waste Product (IU)
LEGEND
Open valve
Closed valVe
Direction jof flow
1X3
Figure 5.3.5. Schematic of inverse/mode of operation.
Source: Reference 13.
5-68
-------
STEP ONE - WATER DISPLACEMENT
HXJ
L WATER.
-»•*•
& BED
A
M DO
-Compressed ai r
INTER-
MEDIATE
BYPRODUCT
(IB)
Intermediate Byproduct (IB)
displaces water from resin
void volume.
Uater Replenished
STEP TWO - FINAL WASTE GENERATION
Final Waste (FW) to Final
waste reservoir.
Compressed air
Final Waste Product (FW) is
routed to FW reservoir.
Water reservoir is refilled
at beginning of step.
_ „
WATER
3
t
b
-y.v--.-n 4
•RES 1 N
S.BED,
i
H
>
rv^
was
|— Cc
—^INTER-
MEDIATE
BYPRODUCT
OB)
STEP THREE - INTERMEDIATE BYPRODUCT DISPLACEMENT
Compressed air
rCXr
-M-
:RESIN
I O
M CXI 1—X
INTER- Water displaces Intermediate Byproduct from
MEDIATE resin void volume.
BYPRODUCT
(IB)
STEP FOUR - RECLAIMED ACID GENERATION
Compressed air-
— BH^W
WATER
V
ta>
INTER-
MEDIATE
BYPRODUCT
(IB)
Rec 1 a i med Ac i d
(RA)
Reclaimed Acid (RA) is produced.
Intermediate Byproduct
reservoir is refilled.
Intermediate
Byproduct
Replenished
LEGEND
Open valve
Closed valve
Direction of flow
Figure 5.3.6. Schematic of normal mode of operation.
Source: Reference 13.
5-69
-------
TABLE 5.3.8. SUMMARY OF RESULTS OF RESEARCH PERFORMED AT
ELECTROPLATING ENGINEERING, INC.3
Parameter
VOLUME TREATED/GENERATED (LITERS):
Spent acid
Intermediate by-product
Reclaimed acid
Intermediate waste product
Final waste product
INVERSE MODE LOADINGS TO RESIN:
Zinc (grains /cycle)
Volume (bed volumes/cycle)
Feed rate (liters/hour)
NORMAL MODE LOADINGS TO RESIN:
Acidity (grams CaC03/cycle)
Volume (bed volumes /cycle)
Feed rate (liters/hour)
STREAM CONCENTRATIONS OF ACIDITY:
(expressed as g/L CaC03 equivalents)
Spent acid
Intermediate by-product
Reclaimed acid
Intermediate waste product
Final waste product
Preliminary
runs
(No.2,3,4)b
98
86
.70
64
67
34.2
4.05
11.4
43.4
0.40
6.0
242
217
116
39
90
Preliminary
runs
(No.5A,5B,6)c
174
174
162
96
121
4.4
4.05
11.4
35.2
0.40
6.0
156
151
94
15
57
Final
runs
(No. 7A-7P)d
189
189
4.6
106
3.3
2.6
4.05
11.4
14.6
0.34
5.5
77
74
46
5.9
30
(continued)
5-70
-------
TABLE 5.3.8 (Continued)
Parameter
Preliminary
runs
(No.2,3,4)t>
Preliminary
runs
(No.5A,5B,6)c
Final
runs
(No. 7A-7P)d
STREAM CONCENTRATIONS OF ZINC (mg/L):
Spent acid 40,333
Intermediate by-product 34,667
Reclaimed acid 19,233
Intermediate waste product 13,000
Final waste product 7,366
STREAM CONCENTRATIONS OF IRON (mg/L):
Spent acid 4,700
Intermediate by-product 4,400
Reclaimed acid 1,450
Intermediate waste product 1,277
Final waste product 3,367
STREAM CONCENTRATIONS OF CHROME (mg/L):
1,100
8
18
2,000
0.61
2,600
2,433
920
357
1,733
*The results of Run 1 were discarded due to improper installation.
zinc loadings for Runs 2, 3, and 4 were above the recommended
18 grams/cycle maximum loading recommended for the system.
cThe objective for Runs 5A, 5B', and 6 were to process a sufficient
quantity of acid for reuse and to optimize loadings.
dThe objective for Runs 7A through 7P were to optimize loadings for
the normal mode of operation.
Source: Reference 13.
'640
1.4
5.47
1,200
0.25
1,100
1,100
439
120
728
Spent acid
Intermediate by-product
Reclaimed Acid
Intermediate waste 'product
Final waste product
43
42
12
4.6
28
2.7
2.6.
1.6
0.51
3.4
NA
NA
NA
NA
NA
5-71
-------
Based on the results of these pilot-scale tests, Pace Laboratories (1986)
conducted an economic evaluation of the APU system for this application. They
determined that the APU could only be cost-effective for this application if a
large volume of spent solution is processed.
In summary, available performance data suggest that the technical and
economic feasibility of acid purification systems will mainly depend on the
types and concentrations of metal ions present. These systems work well in
recovering solutions with highly positively charged contaminant ions
(e.g., aluminum, iron) because these ions pass rapidly through the strong base
anion exchanger res'in. Solutions containing low concentrations of contaminant
ions are not efficiently recovered using the APU. Recommended minimum
concentrations for efficient results are presented in Table 5.3.9. Although
lower concentrations may be treated, removal efficiencies will be lower unless
larger systems are employed. However, cost-effectiveness is compromised due
to the increased capital costs for larger systems. Generally, for low
concentrations of metal ions, ion exchange methods using weak
cation-exchangers to retard the metals are recommended.
5.3.3 Process Costs
An economic evaluation of countercurrent (RFIE) systems is presented in
this section. Cocurrent flow methods are generally not technically or
economically feasible for the treatment of corrosive wastes and will therefore
not be discussed.
Costs for RFIE systems vary with the specific application. -Factors that
affect the costs include: quantity and quality of constituents recovered,
production rates, volume of spent solution to be treated, concentration of
metal salts present in the spent solution, rate of build-up of metal ions in
the bath, concentration of the bath, and number of hours of process operation.
Capital costs, which include equipment, installation, and peripheral
costs, increase with system size. These costs are offset by savings which are
realized through reduced volumes of wastes requiring treatment
(e.g., neutralization) and disposal, and reduced purchase requirements for
bath reagents. Operating costs will include replacement of filter cartridges,
resin replacement (approximately every-5 years), and utilities.
5-72
-------
TABLE 5.3.9. RECOMMENDED MINIMUM CONCENTRATIONS (g/L) FOR EFFICIENT
METALS REMOVAL USING THE EGO-TECH APU
Total
Solution Iron Zinc Aluminum Copper metals
Hydrochloric acid 30-50 130-150 -
Sulfuric acid 30-50 5 20
Nitric/hydrofluoric acid - - - - 30
Nitric acid rack stripping - - - - 75-100
Note: The APU can be used for solutions with lower concentrations of-these
metals, but the metal removal efficiencies will be lower unless a
larger unit is used. Metal removal efficiencies average 55% for
typical systems.
Source: Reference 16.
5-73
-------
Capital costs for acid purification systems typically range from $15,000
to $100,000 depending on the system size (Fontana, 1986). Typical equipment
costs for systems with various throughput rates are presented in
Table 5.3.10. These costs include installation, equipment and peripherals,
and a pre-filter system. Costs presented in this table are for the recovery
of sulfuric acid from aluminum anodizing solutions. Costs may be slightly
higher for other applications (Fontana, 1986).
Typical operating costs are presented in Table 5.3.11. Savings will
depend on the specific application. Table 5.3.12 presents an economic
evaluation of several hypothetical systems.
5.3.4 Process Status
Cocurrent ion exchange systems are generally not employed for direct
treatment of corrosive wastes. Cocurrent systems using weak exchangers have
inefficient exchange capacities in the corrosive pH ranges, and are generally
only used as polishing systems following other treatment operations
(e.g., neutralization). Cocurrent systems using strong exchangers are
technically feasible for the treatment of corrosive wastes, but they are not
cost-effective because of the high costs for column regeneration.
Ion exchange systems, using the reverse or reciprocating flow mode
(countercurrent), have been shown to be effective in the treatment of
corrosive wastes. The process has been demonstrated commercially for chemical
recovery from acid copper, acid zinc, nickel, cobalt, tin, and chromium
plating baths as well as for purification of spent acid solutions.
Chemical recovery systems using fixed bed RFIE have been used to recover
chromic acid and metal salts. It has also been used to deionize mixed-metal
rinse solutions for recovering process water and concentrating the metals for
4
subsequent treatment. Commercial units are available from several vendors.
Acid purification systems using continuous RFIE have been used to remove
aluminum salts from sulfuric acid anodizing solutions, to remove metals from
nitric and rack-stripping solutions, and to remove metals from sulfuric and
4
hydrochloric acid pickling solutions. The APD is primarily used for
4
recovering aluminum anodizing solutions. Acid purification systems are more
cost effective for removing high concentrations of contaminants than other ion
exchange systems. Demonstrated applications are listed in Table 5.3.13.
5-74
-------
TABLE 5.3.10. TYPICAL CAPITAL COSTS FOR ECO-TECH APU
Item
APU Model No.
Flow rate
Capital cost
Small
unit
AP-6
38 L/hr
$14,000
Medium
unit
AP-24
500 L/hr
$37,000
Medium
unit
AP-54
800 L/hr
$116,000
Large
unit
AP-72
6700 L/hr
$184,000
Notes: Capital Costs include equipment, installation, peripherals, and
cartridge-type prefilter system.
Costs presented in this table are for application to recovery of a
sulfuric acid anodizing solution. Costs for other applications may be
slightly higher.
Twelve different size units are available from Ecb-Tech, Ltd. The
model numbers, which indicate bed -diameters, for these units are:
AP-6, AP-12, AP-18, AP-24, AP-30, AP-36, AP-42, AP-48, AP-54, AP-60,
AP-66, and AP-72.
Source: References 16 and 18 (Ecotech quote July 1986).
5-75
-------
TABLE 5.3.11. TYPICAL OPERATING COSTS FOR ACID PURIFICATION
USING CONTINUOUS COUNTERCURRENT ION EXCHANGE (RFIE)
Item Cost
Filter cartridges for prefilter system $10.00/month
Utilities:
(0.5 KW x 16 hrs/day x 20 days/month
x 0.055 i/KHH) i8.80/month
Resin replacement
(specific cost depends on system size) $58/liter every 4 years
Source: References 11 and 13 (Based on August 1986 cost data).
5-76
-------
TABLE 5.3.12. ECONOMIC EVALUATION OF ACID PURIFICATION PROCESS
Description
30,000 gpy
throughput
100,000 gpy
throughput
500,000 gpy
throughput
Case 1 - Purification of Sulfuric Acid Anodizing Solution: Previous
approach used caustic acid neutralization: New approach uses
APU with caustic neutralization.
Approx. APU Cost
Previous treatment cost
Previous acid cost
Annual savings
Payback (months)
$ 6,000
$ 9,690
$ 2,349
$ 8,427
9
$11,000
$32,300
4 7,830
$28,891
5
$ 25,000
$161,500
$ 39,150
$140,455
2
Case 2 - Purification of Sulfuric Acid Anodizing Solution: Previous
approach used lime neutralization: New approach uses APU with
lime neutralization.
Approx. APU cost
Previous treatment cost
Previous acid cost
Annual savings
Payback (months)
$ 6,000
$ 2,250
$ 2,349
$ 3,216
22
$11,000
$ 8,500
$ 7,830
$10,731
12
$ 25,000
$ 38,500
$ 39,150
$ 53,655
6
CaSe 3 - Purification of Sulfuric Acid Anodizing Solution: Previous
approach used waste haulage: New approach uses APU with
caustic neutralization.
Approx. APU cost
Previous treatment cost
Previous acid cost
Present treatment cost
Annual savings
Payback (months)
$ 6,000
$ 3,000
$ 2,349
$ 2,907
$ 1,737
41
(continued)
$11,000
$10,000
$ 7,830
$ 9,690
$ 5,791
• 23
$ 25,000
$ 50,000
$ 39,150
$ 48,450
$ 28,955
10
5-77
-------
TABLE 5.3.12 (Continued)
Description
30,000 gpy
throughput
100,000 gpy
throughput
500,000 gpy
throughput
Case 4 - Purification of Sulfuric Acid Anodizing Solution: Previous
approach used waste haulage: New approach uses APU with
lime neutralization.
Approx. APU cost
Previous treatment cost
Previous acid cost
Present treatment cost
Annual savings
Payback (months)
$6,000
33,000
$2,349
$ 675
$3,969
18
$11,000
$10,000
$ 7,830
$ 2,250
$13,245
10
Case
_5 - Nitric Acid Recovery: Previous approach used caustic
neutralization: New approach uses APU with caustic
neutralization.
$ 25,000
$ 50,000
$ 39,150
$ 11,250
$ 66,155
5
Approx. APU cost
Previous treatment cost
Previous acid cost
Total previous cost
Annual savings
Payback (months)
$9,400
$7,575
$8,775
$16,350
* 9,810
11
$11,300
$30,300
$35,100
$65,400
$39,240
3
$ 18,400
$ 50,500
$ 58,500
$109,000
$ 65,400
3
Source: References 6, 8, 9, and 14. (Based on July 1986 Ecotech cost data)
5-78
-------
TABLE 5.3.13
DEMONSTRATED APPLICATIONS OF ECO- TECH ACID
PURIFICATION UNIT USING RFIE
Application/
bath components
Typical bath
concentration
(g/L)
Typical product
concentration
(g/L)
Typical by-product
concentration
(g/L)
Sulfuric acid
Aluminum
Sulfuric acid
Iron
Nitric acid
Nickel and copper
Sulfuric acid
Hydrogen peroxide
Copper
Hydrochloric acid
Iron
Nitric acid
Hydrofluoric acid
Iron
Nickel
Chrome
Sulfuric acid
Sodium
190
10
127
36
514
99
128
41
13.3
146
34
150
36
29
7.02
7.33
61.3
7.8
182
5.5
116
10.5
581
47.5
113
35
5.9
146
25
139
28.8
8.7
2.1
2.2
54.9
0.8
13
6
10
21
10
70.8
18
7
9.2
10
15
4.5 '
7.2
20.3
4.9
5.1
5.88
5.56
Source: Reference 9 (Based on July 1986 Ecotech cost data)
5-79
-------
Although the use of ion exchange for acid purification is currently under
investigation by several ion exchange vendors (e.g., Alpha Process Systems;
Illinois Water Treatment Company; Ionics, Inc.; etc.), Eco-Tec, Ltd. is the
13 19 20
only vendor with commercial units currently in operation. ' *
Table 5.3.14 compares the advantages and disadvantages of the various ion
exchange alternatives. Cocurrent fixed beds are generally used in wastewater
treatment; they are not cost-effective for treatment of corrosives due to the
high costs associated with regenerant purchase and treatment. The
countercurrent (RFIE) continuous mode of operation, which is utilized by the
APO, is generally more efficient in the treatment of solutions containing high
concentrations of metal ion contaminants. The countercurrent (RFIE) fixed
mode is most effective in the recovery of dilute solutions. Overall, RFIE
systems may be more cost-effective for the treatment of corrosives than
conventional neutralization, particularly when enactment of land disposal
restrictions increase costs associated with land disposal of residuals.
5-80
-------
TABLE 5.3.14. COMPARISON OF ION EXCHANGE OPERATING MODES
Criteria
Cocurrent
fixed bed
Counts rcurrent
(RFIE)
fixed bed
CounCe rcurr en t
(RFIE)
continuous bed
Capacity for high feed
flow and concentration
Effluent quality
Kegenerant and rinse
requirements
Equipment complexity
Equipment for
continuous operation
Relative costs
(per unit volume)
Investment
Operating
Application for
corrosive wastes
Least
Fluctuates with
bed exhaustion
Highest
Middle
High, minor
fluctuations
Somewhat less
than cocurrent
Simplest; can
use manual
operation
Multiple beds,
single
regeneration
equipment
Least
Highest
chemicals and
labor; highest
resin inventory
Useful only as
a polishing step
for neutraliza-
tion treatment
More complex;
automatic
controls for
regeneration
Multiple beds,
single
regeneration
equipment
Middle
Less chemicals,
water and labor
than cocurrent
Effective in
the treatment
of more dilute
solutions (i.e.,
corrosive
plating rinses)
Highest
High
Least, yields
most concen-
tration
regenerant
waste
Most complex;
completely
automated
Provides con-
tinuous service
Highest
Least chemical
and labor;
lowest resin
inventory
Effective in
the treatment
of more con-
centrated
solutions
(i.e., direct
treatment of
acid baths).
Source: References 16 and 21.
5-81
-------
REFERENCES
1. Camp Dresser & McKee, Inc. (CDM). Technical Assessment of Treatment
Alternatives for Wastes Containing Corrosives. Prepared for the U.S.
EPA-OSW, Waste Treatment Branch, Washington, D.C., under EPA Contract No.
68-01-6403, Work Assignment No. 39. September 1984.
2. Uiggins, T.E., CH2MHILL. Industrial Processes to Reduce Generation of
Hazardous Waste at DOD Facilities - Phase 2 Report, Evaluation of 18 Case
Studies. Prepared for the DOD Environmental Leadership Project* and the
U.S. Army Corps of Engineers. July 1985.
3. GCA Technology. Industrial Waste Management Alternatives And Their
Associated Technologies/Processes. Prepared for the Illinois
Environmental Protection Agency, Division of Land Pollution Control,
Springfield, Illinois. GCA Contract No. 2-053-011 and 2-053-012.
GCA-TR-80-80-G. February 1981.
4. U.S. EPA, Industrial Environmental Research Laboratory, Cincinnati,
Ohio. Summary Report: Control and Treatment Technology for the Metal
Finishing Industry - Ion Exchange. EPA-625-8-81-007. June 1981.
5. Hatch, M.J,, and J.A. Dillon. Acid Retardation: A Simple Physical
Method for Separation of Strong Acids from Their Salts. I&EC Process
Design and Development, 2(4):253-263. October 1963.
6. Brown, C.J., Davy, D., and P.J. Simmons. Recovery of Nitric Acid from
Solutions Used for Treating Metal Surfaces. Plating and Surface
Finishing. February 1980.
7. Robertson, W.M., James, C.E., and J.Y.C. Huang. Recovery and Reuse of
Waste Nitric Acid From An Aluminum Etch Process. In: Proceedings of the
35th Industrial Waste Conference at Purdue University. May 13-15, 19«0.
8. Brown, C.J., Davy, D., and P.J. Simmons. Purification of Sulfuric Acid
Anodizing Solutions. Plating and Surface Finishing. January 1979.
9. Eco-Tec, Ltd. Product Literature: Acid Purification Unit (APU).
Bulletin No. ET-4-84-5M. Received July 1986.
10. U.S. EPA, Industrial Environmental Research Laboratory, Cincinnati,
Ohio. Sources and Treatment of Wastewater in the Nonferrous Metals
Industry. EPA-600/2-80-074. April 1980.
11. Fontana, C., Eco-Tech, Ltd. Telephone Conversation with L. Wilk,
GCA Technology Division, Inc. Re: Acid Purification Unit.
August 21, 1986.
12. Dejak, M. Acid Recovery Proves Viable in Steel Pickling. Finishing,
10(1): 24-27. January 1986.
5-82
-------
13. Face Laboratories, Inc. Final Report: Reclamation and Reuse of Spent
Hydrochloric Acid, Hazardous Waste Reduction Grant. Prepared for the
Minnesota Waste Management Board on behalf of Electro-Plating Engineering
Company, Inc. February 14, 1986.
14. Eco-Tec, Ltd. Product Literature: Ion Exchange Systems. Bulletin
No. ET-11-83-3M. Received July 1986.
15. GCA Technology. Corrective Measures for Releases to Ground Water from
Solid Waste Management Units. Prepared for U.S. EPA-OSW Land Disposal
Branch, under EPA Contract No. 68-01-6871, Work Assignment No. 51.
GCA-TR-85-69-G. August 1985.
16. Fontana, C., Eco-Tech, Ltd. Telephone Conversation with L. Wilk,
GCA Technology Division, Inc. Re: Acid Purification Unit.
August 26, 1986.
17. Chemical Processing Staff. Spotlight: Pickling Acid Recovery Unit Saves
$40,000/year, Purifies Spent Sulfuric Acid. Chemical Processing,
49(3): 36-38. March 1986.
18. Fontana, C., Eco-Tech, Ltd. Telephone Conversation with L. Wilk,
GCA Technology Division, Inc. Re: Acid Purification Unit. July 7, 1986.
19. Parcy, E., Ionics, Inc. Telephone conversation with J. Spielman,
GCA Technology, Inc. August 14, 1986.
20. Jain, S.M., Ionics, Inc. Telephone conversation with J. Spielman, GCA
Technology Division, Inc. August 12, 1986.
21. U.S. EPA, Office of Research and Development, Washington, D.C.
Treatability Manual, Volume III: Technologies for Control/Removal of
Pollutants. EPA-600-8-80-042c. July 1980.
5-83
-------
5.4 ELECTRODIALYSIS
5.4.1 Process Description
Elect rodialys is (ED) uses an electric field and a semi permeable
ion-selective membrane to concentrate or separate ionic species in an aqueous
1 2
solution. ' Its primary application in the treatment of corrosive wastes
is for use in the recovery of acids from plating solutions, pickling
solutions, and etchants. Three types of configurations may be used in the
design of ED units: concentrating-diluting, ion-substituting, and
electrolytic.
The concentrating-diluting configuration, shown in Figure 5.4.1,
typically contains a series of alternating cation and anion membranes arranged
in parallel between two electrodes (cathode and anode) to form an ED
2 4
"multicell" or membrane "stack." ' Cationic membranes are only permeable
to positive ions and anionic membranes are only permeable to negative
ions. ' Spacers (gaskets) are used to separate the adjacent membranes into
4
leak-tight compartments.
An electrical potential is .applied across the ion exchange membrane which
causes the migration of cations (e.g., copper, nickel, and zinc) toward the
negative electrode and anions (e.g., sulfates, chlorides, and cyanides) toward
the positive electrode. The cathode is typically comprised of stainless
4
steel, and the anode usually consists of platinized titanium. Thus, under
the influence of an electric potential applied across the membranes,
alternating cells of concentrated and dilute solutions are formed between the
cationic and anionic membranes.
Electrodialys is units are generally characterized by the number of
"cell-pairs" comprising a multicell unit. One concentrating and one diluting
cell comprise a cell pair. A cell-pair consists of a cation-selective
membrane, a diluting spacer, an anion-selective membrane, and a concentrating
4
spacer. A stack or multicell refers to the entire arrangement of repeating
cell-pairs between two electrodes. . Industrial stacks typically have 50 to
300 cell-pairs.4
The ion-substituting configuration is similar to the concentra ting-
diluting configuration, but it involves a transfer of ions. As illustrated in
Figure 5.4.2, the waste feed is passed between two membranes of like charge.
5-84
-------
TRANSFER
STREAM
PROCESS £
STREAM y
ELECTRODE
STREAM
n
©
(ANODE)
s
s
s
s
s
s
V
V
V
s
s
s
s
s
V
s
s.
^
X2,°2
HX
JA
*-
c
x-
M +
o*
A
^•i
_^^
M+
"?
x-
^^^~
•^
c*
! .
x-
M+
D*
A
i
.^^
m.^^
M +
7
x-
^"™*
^^
c*
C A
x-
M+
o*
^^^
^^^
c
x-
M*
0*
A
^^fr
^^»
M+
~~?
x-
c™1
*^
c*
ELECTRODE
STREAM
e-
x-
M+
0*
-»
L
s
V
s
V
s
v;
^
N
^
v
s
>
©
(CATHODE)
* ^*2, MOH
^ DILUTED
PROCESS
STREAM
A • ANION-TRANSFER MEMBRANE
C ? CATION-TRANSFER MEMBRANE
X-s ANION
M+: CATION
0*: DILUTING CELL
C*« CONCENTRATING CELL
CONCENTRATED
TRANSFER
STREAM
Figure 5.4.1.
Schematic of a concentrating-diluting
electrodialysis process .
Source:
Reference 3.
5-85
-------
AC1O STREAM
ANODE
I CAUSTIC PRODUCT FEED
I WASTE STREAM
1 ELECTRODE
rn i M i i 11 F--
H +
Mt
OH-
>M»
X'
H +
x-
M+
OH-
"7
X'
x-
Mt
OH-
V
~7|
x-
X-
i A
C
C
A
C
C
-
A
°
.
ACID
C
•
NEU1
WAST
'RALI2
STREAM (
CATHODE
ELECTRODE
STREAM-
LEGEND:
A : ANION MEMBRANE
C = CATION MEMBRANE
Figure 5.4.2. Schematic of an ion transfer/ion substituting
electrodialysis process.
Source: Reference 3.
5-86
-------
A solution containing the ions which are to be substituted is passed through
an adjacent compartment. Upon application of an electric potential, an ionic
substitution occurs across the membranes.
Electrolytic ED units are similar to conventional electrolysis cells,
except that a membrane is placed between the electrodes. The membrane acts to
separate the products of the electrode reactions.3 In typical applications
for corrosives, the diaphragm cell is fitted with cation selective
membranes. Figure 5.4.3 shows a schematic representation of a diaphragm
cell containing one anode chamber and two cathode chambers. Spent etchant is
fed into the anode chamber. Upon application of an electric potential, metal
contaminants migrate to the cathode chambers. The anolyte (etchant solution)
is regenerated and reused, and the metals can be recovered at the cathode.
Operating Parameters--
Important design factors which affect the operation of an ED unit include
membrane characteristics, anode and cathode materials, voltage density,
current density, temperature, and the number of cells. Parameters used to
select an appropriate membrane for a specific application include transport
number for the appropriate ions and electrical, physical and chemical
resistance of the membrane.
^Higher current densities will improve the product concentration. Also,
increased solution concentrations will require higher voltages to maintain
current densities. However, increases in the current density also cause
increases in temperature. Operating temperature is a limiting factor in that
high temperatures (above 100°F) can cause damage to the ED unit. A cooling
system may be required to prevent heat damage.
The number of cells determine the maximum throughput rate of the system.
Selection of the appropriate number of cells is based on spent bath ion
concentration and required reduction, rate of metal salt accumulation in the
bath, volume of spent bath to be treated per unit time, and the number of
hours of process operation.
Pre-Treatment Requirements/Restrictive Waste Characteristics—
Spent finishing and/or plating solutions often contain small metal parts
and extraneous metal chips. Buildup of these particulates can foul the
membranes and lower the cell resistance. High organic levels can also foul
5-87
-------
Cation-selective
Copper
cathode
\
; Cathode
chamber
— — membrane —
Lead-based
anode
t
X
/
X
M/
Anode
chamber
\
//
X
'/
Copper
cathode
^
i
c
^
Cathode '
chamber >
Figure 5.A.3. Diagram of an electrolytic electrodialysis cell.
Source: Reference 6.
5-88
-------
the membrane. Thus, filtration generally needs to be employed; as a
pretreatment step to avoid plugging of the membrane. In addition,
oxidizing substances will attack the membranes and must be chemically
removed.
Post-Treatment Requirements—
Post-treatment requirements for electrodialysis systems are minimal
because it is a closed-loop operation. Electrodialysis generally creates a
concentrated solution such that an evaporation/concentration unit is not
0)
1
1
required. ED units do not need to be regenerated and are able to operate
continuously.
Electrodialysis removes ionic species nonselectively. This may cause
ionic impurities to be returned to the plating baths in cases where there is
more than one contaminant ion present in the solution. ' Therefore,
periodic treatment of the plating baths using an ion exchange technique may be
required to remove impurities.
As discussed in Section 5.4.2, the applicability of the electrolytic ED
process may be limited if the contaminant metal ions do not electrodeposit
Q Q
homogeneously on the cathodes. ' Difficulty in removing the metal from the
cathode may make the process.economically inefficient.
5.4.2 Process Performance
Factors to consider when evaluating the applicability of an ED system for
a particular use include types and concentrations of constituents present in
the spent solution, required product concentration, required throughput rates,
and spent bath temperature. Factors which affect the performance of a system
for a particular application include: the chemical, physical, and electrical
resistance of the membrane for the spent solution; the number, type, and
concentrations of metal ion contaminants present in the spent solution; the
ability of the contaminant ions to electrodeposit on the cathode; the ease of
removal of the metal contaminants from the cathode; and the ability of the
system to maintain satisfactory product concentration.
5-89
-------
Research was conducted using a concentrating-diluting electrodialysis
4
unit to recover chromic acid from chrome plating rinses. Preliminary
testing was performed using a laboratory-scale (five cell-pairs) ED unit to
determine the chemical resistance of the membrane and to optimize current
4
densities for maximum product concentration.
Chemical resistance was tested by immersing the membranes at room
temperature in a chromic acid plate solution and removing sections after 7,
4
14, 42, and 69 days. The membranes appeared to be unaffected by the acid,
with the exception of a slight roughening of the surface. The membranes did
not exhibit a loss in physical strength or exchange capacity and remained
leak-tight.
Electrodialysis was performed on a simulated chromic acid rinse solution
(chromium trioxide dissolved in tap water) over a range of current
4
densities. Samples were collected after several hours operation at each
operating condition. The results are presented in Table 5.4.1. The data
indicate that chromic acid from the rinsewater could be concentrated to
70 percent original strength. The maximum chromic acid concentration achieved
was approximately 175 g/L. As expected, the product/feed ratio decreased
with increasing feed concentration.
Following the laboratory testing, a 50 cell-pair demonstration module was
constructed for a chrome plating line at the Seaboard Metal Finishing Company
in West Haven, Connecticut. Figure 5.4.4 presents a flow diagram for the
system. Operating parameters for this series of tests are presented in
Table 5.4.2. Objectives of the demonstration were to test the chemical
resistance of the membranes, determine proper operating parameters for optimum
4
chrome recovery, and to familiarize company personnel with its operation.
Results similar to those obtained in the laboratory-scale tests were
obtained. However, problems were encountered with high rinse temperatures due
4
to heat generated from the electrical current and pumps. During much ot
the operation, rinse temperatures were higher than 100°F, the maximum
recommended operating temperature. The high temperatures caused extensive
slippage of the stack spacers and membranes in the ED cells which led to
external and internal leakage problems. A significant drop in product
concentration occurred as a result of the leakage.
5-90
-------
TABLE 5.4.1. LABORATORY-SCALE ELECTRODIALYSIS OF SIMULATED CHROMIC
ACID RINSES OVER A RANGE OF CURRENT DENSITIES
Run
1
2
3
4
5
Current
density
10
12
14
16
18
Feed
concentrat ion
g/L Cr03
0.37
0.32
1.24
0.98
0.70
Product
concentration
g/L Cr03
61
106
143
167
174
Product/
rinse ratio
165
33-1
115
170
249
Source: Reference 4.
5-91
-------
Workplaces
Concentrated
Traduce
Figure 5.4.4. Flow diagram for electrodialysis treatment of chrome
line at Seaboard Metal Finishing Company.
Source: Reference 4.
5-92
-------
TABLE 5.4.2. OPERATING PARAMETERS FOR ELECTRODIALYSIS
DEMONSTRATION UNIT FOR RECOVERY OF CHROMIC
ACID AT SEABOARD METAL FINISHING COMPANY
Parameter
Measurement
Operating time
Maximum current
Product concentration
Rinse concentration
Product/feed concentration
ratios
Maximum observed rinse
temperature
Maximum recommended rinse
temperature
250 hrs over> 2 month period (day shift)
20 to 21 amps
160 to 212 g/L chromic acid
50 to 70 g/L Cr03
2 to 4
118°F
100 °F
Source: Reference 4.
5-93
-------
Figure 5.4.5 shows the product concentrations achieved during operation
of the ED unit as a function of operating current. As expected, produce
concentrations increased with increasing current densities. However, as shown
in Figure 5.4.6, the increasing rinse temperatures and the resulting damage to
the ED cells caused a sharp drop in product concentrations as the testing
progressed.
As a result of the leakage problems, operations were temporarily ceased
to evaluate the system and to make improvements. It was determined that
product concentrations could be increased by: (1) running the ED unit
concurrently with plating shifts; (2) increasing the membrane area to increase
recovery capacity of the stack; and (3) adding a cooler to the rinse so that
4
higher current densities could be explored. However, results from the
second phase of tests are not available because the project was never
completed.
An electrodialytic process has recently been developed by Aquatech
Systems (Bethel, New Jersey) for the recovery of hydrofluoric/nitric acid
11 12
pickling liquors. ' With this process, electrodialysis is used following
neutralization of the waste acid with potassium hydroxide. Thus, membrane
degradation is minimized. The potassium hydroxide neutralization forms a
potassium fluoride/nitrate solution which can be filtered with diatomaceous
earth to remove metal hydroxides. The filtered solution is then directed
to the electrodialysis stack where, upon application of an electric potential,
a 3 M stream of mixed hydrofluoric/nitric acid and a 2 M potassium hydroxide
solution are produced. ' The HF/HNO- can be recycled back to the
pickling bath and the KOH solution can be reused in the neutralization step.
Typical operating parameters for this process are summarized in Table 5.4.3.
Innova Technology, Inc. has developed ion transfer electrodialysis units
for commercial application in the recovery of chromic acid from plating rinse
baths. Typical results are presented in Table 5.4.4. Although more
concentrated solutions (such as the actual plating bath) can be fed to the
unit without causing degradation of the resin, much larger systems would be
14
required to achieve satisfactory product concentration. At the present
14
time, these systems would not be cost-effective.
The Bureau of Mines has conducted research using an electrolytic membrane
cell system to recover chromic acid and sulfuric acid from spent brass
5-94
-------
200-
n 17S
u
°° 150 .
H 125
z
u
8-
H 100
i
10
IS
20
CURRENT (AMPERES)
Figure 5.4.5. Relationship between current and product
concentration during operation of ED
unit for chromic acid recovery.
Source: Reference 4.
5-95
-------
DAYS
Figure 5.4.6.
Production concentrations achieved during
operation of the pilot acale ED unit at
Seaboard Metal Finishing Company.
Source: Reference 4.
5-96
-------
TABLE 5.4.3. TYPICAL OPERATING PARAMETERS AND RESULTS FOR
THE AQUATECH ED PROCESS
Parameter
Result
Operating temperature
Current density
Pressure drop
Voltage
Feed solution
Conductivity
Maximum particle size
Maximum metal concentrations
Chromium
Nickel
Iron
Calcium
Magnesium
Molybdenum.
Product concentrations
Hydrofluoric/nitric acid
Potassium hydroxide
80 to 100°F
70 to 100 amp/ft2
5 psi
2.0 to 2.5 volts/cell
>100,DOO mohms
<10 microns
<1 to 2 ppm
<1 to 2 ppm
<1 to 2 ppm
<0.1 ppm
<0.1 ppm
<70appm
3 M
2 M
Source: Reference 13.
5-97
-------
TABLE 5.4.4. TYPICAL OPERATING PARAMETERS FOR ION TRANSFER
SYSTEM DEVELOPED BY INNOVA TECHNOLOGIES, INC.
Parameter Result
Cell size 18 in. x 27.6 in. x 2.5 in.
(46 cm x 69 cm x 6.4 cm)
Voltage 25 Vdc
Product concentration 152
Bath temperature 112°F (45°C)
Rinse temperature 80 to 85°F
(27 to 30°C)
Source: References 14 and 15.
5-98
-------
etchants. Their pilot-scale unit, termed the Process Research Unit (PRU),
is designed for online regeneration. . The PRU is able to use one to five anode
membrane cells in single catholyte tank.
Spent etchant solution is pumped from the bottom of the etchant tank
through a filter into a holding tank. Etchant from the holding tank is pumped
into the anode chambers of the PRU. An electric potential is applied, causing
migration of copper and zinc through the cation-selective membranes and into
the catholyte (180 g/L sulfuric acid). Trivalent chromium (Cr ) is
oxidized to hexavalent chromium (Cr ) in the anode chamber. Metal
contaminants are recovered at the cathode, and the etchant solution is
regenerated in the anode chamber to a level comparable to fresh etchant. The
regenerated etchant is then pumped directly back to the etchant tank.
The condition of the PRU is monitored by a control panel, which includes
an ampere-hour meter, a voltage meter, and an ampere meter. A fume hood
located directly above the diaphragm cell is used to remove hydrogen, oxygen,
and chromic acid mist.
Preliminary testing of the PRU was conducted with samples from brass
etchant solutions. Two anode chambers were used for Tests 1 and 2, and
five anode chambers were used for Test 3. The test results, presented in ••
Table 5.4.5, indicate that sodium dichromate production increases with
increasing flow rate.
Following the preliminary tests, online testing of the Bureau of Mines
PRU was performed at the Valve and Fitting Division of Gould Manufacturing,
17 7 fi 1Q 20
Inc. (now known as Impervial Clevite, Inc.) in Niles, Illinois.
The PRU was operated over a 17-day period, which included 11 days when two
10-hour shifts were operated and three weekends when the plant was shut down.
Operating parameters recorded during the testing period and test results are
summarized in Table 5.4.6. The PRU proved to be technically and economically
effective. The PRU regenerated and recycled the chromic acid/sulfuric acid
etching solution at an acceptable level of performance and reduced sodium
dichromate consumption, disposal volumes, and process operating costs.
The Bureau of Mines research led to the development of a commercial-
scale electrolytic ED unit, which is currently available from Scientific
Control, Inc. in Chicago, Illinois.21'22 The unit is applicable to the
recovery of chromic acid/sulfuric acid etching solutions in which copper is
5-99
-------
TABLE 5.4.5. SUMMARY OF RESULTS OF PRELIMINARY TESTING OF PRU -
Parameter
Number of anode chambers used
Flow rate (liters/hr)
Cr+^ oxidation (percent)
Cu removal (percent)
Zn removal (percent)
Or lost to catholyte (percent)
Energy consumption (KWH/kg
sodium dichr ornate)
Sodium dichromate produced
(kg/hr)
Anode efficiency (percent)
Duration of run (days)
Test 1
2
2.6
>96
41
41
11
8.6
0.43
19
3
Test 2
2
5.7
88
20
23
4
5.7
0.70
31
3
Test 3
5
7.3
92
31
28
5
8.0
0.48
21
7
Source: Reference 5.
5-100
-------
TABLE 5.4.6. SUMMARY OF OPERATING PARAMETERS AND RESULTS OF
BUREAU OF MINES TESTS USING ELECTROLYTIC ED TO
TREAT SPENT CHROMIC ACID-SULFURIC ACID ETCHANTS
Parameter
Result
Total operating period
Flow rate
First 5 days
Remaining 12 days
Etchant (working) volume
Residence time
First 5 days
Remaining 12 days
Cathode current density
Average applied current
Average applied potential
Total electric power consumption
e
Products generated
Sodium dichromate
Copper
Zinc
Percent removal
Copper
Zinc
Percent conversion of trivalent
chromium to hexavalent chromium
Overall current efficiency
17 days (220 hrs)
26.5 L/hr (7 gal/hr)
41.6 L/hr (11 gal/hr)
378.5 L (100 gal)
15 hours
9 hours
215 amp/sq meter (20 amp/sq ft)
1,181 amps
3.77 volts
1,518 KW-hr
40.0 kg (88.3 Ib)
19.8 kg (43.8 Ib)
8.6 kg (18.9 Ib)
40.9 percent
21.1 percent
81.2 percent
11.2 percent
Source: Reference 17.
5-101
-------
Che primary metal contaminant. Companies using the systems have reported
21
product quality equivalent to that of fresh make-up solution. Typical
operating parameters for the system are listed in Table 5.4.7. The system
operates most efficiently with acid baths containing copper concentrations of
2 to 4 oz/gal.21
Additional studies have been performed using the Electrolytic ED unit to
8 9 23
recover acids contaminated with iron. ' ' At present, difficulties in
removing the iron metal from the cathode have made the process economically
infeasible. However, reseajrch is currently being performed at the U.S. Bureau
co
8
a 23
of Mines to address this problem. ' Preliminary results are expected in
late 1986.
In summary, at the present time the concentrating-diluting and the ion
transfer types of electrodialysis units are only technically feasible for
treatment of dilute solutions; e.g., plating rinse waters. A recently
developed ED process is able to recover HNO. and HF from pickling liquors
that have been neutralized with KOH. However, limited performance data is
available for this process. Electrolytic ED units are capable of treating
concentrated solutions. However, the application of electrolytic ED units is
currently limited to the recovery of solutions which contain one predominant
metal ion contaminant that can be homogeneously electroplated and easily
removed from the cathode. Performance data have demonstrated electrolytic ED
units to be effective in the removal of copper ions from acid baths. However,
difficulties have been encountered with the removal of iron metal from the
cathode.
5-102
-------
TABLE 5.4.7. TYPICAL OPERATING PARAMETERS FOR SCIENTIFIC
CONTROL, INC. ELECTROLYTIC ED UNIT FOR
RECOVERY OF BRASS ETCHANTS
Parameter Typical value
Temperature 70°F
Voltage requirements 4 to 6 volts
f\
Current density 30 amps/ftz
Optimum copper concentration 2 to 4 oz/gal
Source: Reference 21.
5-103
-------
5.4.3 Costs
Typical costs for electrodialysis systems to treat plating rinsewaters
range from $30,000 to $45,000, depending on the application.1 Capital costs
for the "Chrome Napper" system available from Innova Technology, Inc. in
Clearwater, Florida, range froa $9,900 to $30,000, including installation and
14
power supply. Systems are sized according to bath temperature, dragout
concentrations, number of rinse tanks, concentration of the bath, and the
volume of spent solution to be treated per unit time.
The ED system developed by Aquatech Systems is generally only
cost-effective for large quantity generators. An economic evaluation for
a typical application is presented in Table 5.4.8. Systems are also available
for leasing at the costs listed in Table 5.4.9.
Scientific Control, Inc. sells an electrolytic ED unit to recover chromic/
21
suIfuric acid brass etchants. System sizes are based on the amount of
copper the system is capable of removing per unit of time. Available unit
sizes range from 0.05 to 0.5 Ib copper removal/hour. Capital equipment costs
21
for these units range from $24,000 to $80,000. These costs do not include
installation which would- include a hoist, plumbing, and a ventilation/exhaust
system. Additional costs for the exhaust system cost could range from $5,000
to $15,000, depending on the size required. Operating and maintenance costs
are relatively low. Membranes will need to be replaced approximately every
9 months, depending on usage, at a replacement cost of approximately 10 to
15 percent of the original equipment costs. Additional maintenance costs will
include approximately $10/month for replacement of filter cartridges (a
pre-filter system is incorporated into the unit). The estimated payback
period for the system is approximately 2 years, based on savings in treatment
21
and disposal costs.
The Bureau of Mines conducted an economic analysis of electrolysis versus
conventional neutralization/disposal for treatment of spent chromic/sulfuric
24
acid etching solutions. Capital costs will depend upon the size of the
system employed. Table 5.4.10 presents fixed capital and installation labor
costs for 500 and 1,000 gallon catholyte holding tank capacities.
5-104
-------
TABLE 5.4.8. ECONOMIC EVALUATION OF AQUATECH ED SYSTEM TO
. REGENERATE 1.5 MILLION GALLONS OF HF/HN03
PICKLING LIQUOR PER YEAR
Item Cost ($)
Capital costs (installed system) 1.5 to 2.0 million
Operating costs 415,000/year
(includes electricity, chemical -
make-up, and cell maintenance)
Cost savings • 808,000/year
(includes HF and HN03 purchase costs,
and waste disposal savings)
Net savings 393,000/year
Payback period 3.8 to 5.1 years
Source: Reference 13 (Aquatech cost data August 1986).
5-105
-------
TABLE 5.4.9. COSTS FOR LEASING AQUATECH SYSTEM
Description
Size
Minimum
lease period Cost
Bench-scale 6 in. x 8 in., 8 cells, 3 months
cell stack effective area » 1 sq ft
Pilot-scale 1 ft x 2 ft, 50 cells, 6 months
cell stack effective area » 50 sq ft
Pilot plant Full scale, 4 months
effective area « 50 sq ft
1,000/month
5,000/month
10,000/month
Source: Reference 25 (Aquatech cost data January 1986).
5-106
-------
TABLE 5.4.10. ESTIMATED FIXED CAPITAL COSTS FOR
ELECTROLYTIC RECOVERY UNIT
Item
Electrolytic cell
Hoist
Blower
Ductwork
Electrical
Piping
Total :
1,000 gal cell
(*)
65,700
3,400
1,000
100
1,800
200
72,200
500 gal cell
(*)
42,000
3,400
1,000
100
It 100
200
47,800
Source: Reference 24 (Based on 1983 cost data).
5-107
-------
Estimated annual operating costs for the Bureau of Mines electrolytic
.process are presented in Table 5.4.11. These costs are based on an average of
350 days of operation per year over the life of the equipment. It is assumed
that maintenance activities, such as copper removal and occasional system
checks, will require a minimal amount of time and can be performed by existing
employees. Equipment depreciation is based on a straight-line 20-year
period. The following cost- savings were noted in the Bureau of Mines economic
24
evaluation:
• better product performance,
• reduced waste solution treatment and disposal costs,
• reduced sodium dichromate consumption,
• reduced drag-out losses, and
• copper recovery.
Table 5.4.12 presents quantitative estimates of the cost savings for a
1,000 gallon electrolytic unit operating 350 days per year. Figures 5.4.7 and
5.4.8 show the effects of variable costs on the operating costs for 500 and
1,000 gallon systems. The cost savings presented in Table 5.4.12 do not
include the value of recovered copper, which may be marketed, or benefits
resulting from improved product quality. Inclusion of these factors would
24
decrease the payback period slightly.
5.4.4 Process Status
Currently, electrodialysis has a limited area of application in the
recovery/reuse of corrosive wastes. Concentrating-diluting and ion transfer
ED units have been successful in the recovery of chromic acids from dilute
solutions. The electrolytic ED unit has shown success in removing copper ions
from spent chromic/sulfuric acid brass etchants and bright dipping solutions.
Another recently developed application uses electrodialysis in conjunction
with neutralization to recover spent hydrofluoric/nitric acid pickling liquors.
5-1Q8
-------
TABLE 5.4.11. ESTIMATED ANNUAL OPERATING COSTS FOR
ELECTROLYTIC RECOVERY UNIT
1,000 gal cell 500 gal cell
Item (*) (*)
DIRECT COST
Utilities
Electric power (@ 0.045 fc/KW-h) 5,500 2,800
Total Utilities: 5,500 2,800
Plant Maintenance
Labor 600 300
Materials 600 400
Total plant maintenance: 1,200 700
Payroll overhead 200 100
(33Z of payroll)
TOTAL DIRECT COST 6,900 3,600
FIXED COST
Taxes ' 700 500
(1Z of total capital cost)
700 500
Insurance
(1Z of total capital cost)
Depreciation 3,600 2,400
(20-year life)
TOTAL OPERATING COST U.900 7,000
Source: Reference 24 (Based on 1983 cost data).
5-109
-------
TABLE 5.4.12. COST COMPARISON OF ELECTROLYTIC UNIT
VS. CONVENTIONAL TREATMENT
Item
Cost
Electrical consumption/lb of sodium
dichromate regenerated
Sludge reduction/lb of sodium
dichromate regenerated
Reduction in sodium dichromate
consumption for a 1,000 gal system
Sludge reduction for a 1,000 gal system
Cost for sludge treatment and disposal
Cost for sodium dichromate purchase
Total cost savings
Electric power cost for electrolytic unit
Electric power cost for 1,000 gal system
(not including taxes and depreciation)
Net savings
Payback period for 1,000 gal system
3.6 KW-hr
9.4 gal
100 lb/day'
940 gal/day
0.20 i/gal
0.68 S/lb
256 $/day
0.045 $/Kw-hr
20 a/day
236 $/day
0.84 years
( 10 months)
Source: Reference 24 (Based on 1983 cost data).
5-110
-------
The primary application of electrodialysis is in the treatment of metal
.wastes. The only commercially available electrodialysis system for the direct
recovery/reuse of corrosive wastes is an electrolytic ED unit sold by
Scientific Control, Inc. in Chicago, Illinois.21 Development of the unit
was based on research performed by the U.S. Bureau of Mines. It is only
capable of treating brass etchants or bright dipping solutions where copper is
the primary metal contaminant. As a result of its limited proven area of
application, only four units are currently in operation. The companies using
the system have reported satisfactory results.
The Aquatech ED process, developed for the recovery of hydrofluoric/
nitric acid pickling liquors that have been neutralized with potassium
hydroxide, has recently been made available for purchase or lease. However,
no commercial systems have been installed to date. Additionally, the system
12 13
is only cost-effective for large quantity generators. '
Innova Technology, Inc. in Clearwater, Florida sells an ED system which
uses ion transfer technology to recover chromic acid from chromic acid plating
14
rinsewaters. The system, called the "Chrome Napper", has been employed
cost-effectively by a number of companies to recover chromic acid plating bath
of rinse water. However, this system is not cost-effective for direct
treatment of the plating bath.
Current research in the application of electrodialysis to the treatment
of corrosive wastes is directed at using the electrolytic ED configuration.
The U.S. Bureau of Mines in Rolla, Missouri is experimenting with techniques
to recover other types of acid baths. They are currently bench-testing
various techniques for the removal of iron from nitric acid/hydrofluoric acid
pickling liquors. '
In addition. Ionics, Inc. in Watertown, Massachusetts is currently
developing improved ED membranes for use with corrosive wastes. However,
their research is only at the bench scale at this time, and preliminary
results have not yet been released.
In summary, the predominant use of ED units is in the recovery/reuse of
o
rinsewaters rather than concentrated wastes. Electrodialysis can be used
to concentrate acids for recovery from acidic process wastewaters to reduce
5-111
-------
neutralization and disposal costs. Research and commercial demonstrations
•have shown the technical and economic feasibility of regenerating and
recycling spent brass etchants using electrolytic ED units. This research
has also indicated that ED units show potential applications in the recovery
of other corrosive wastes.
5-112
-------
REFERENCES
1. Higgins, T. C. CH2MHILL. Industrial Processes to Reduce Generation of
Hazardous Waste at DOD Facilities - Phase 2 Report, Evaluation of 18 Case
Studies. Prepared for the DOD Environmental Leadership Project and the
U.S. Army Corps of Engineers. July 1985.
2. Radimsky, J., Daniels, D. I., Eriksson, M. R., and R. Piacentini.
California Department of Health Services. Recycling and/or Treatment
Capacity for Hazardous Wastes Containing Dissolved Metals and Strong
Acids. October 1983.
3. Jain, S. M. Ionics, Inc. Electrodialysis and Its Applications.
American Laboratory. October 1979.
4. Eisenmann, J. L. Membrane Processes for Metal Recovery From
Electroplating Rinse Water. In: Proceedings of the 2nd Conference on
Advanced Pollution Control for the Metal Finishing Industry, Co-sponsored
by the American Electroplaters Society and the U.S. EPA, Kissinmee,
Florida, February 5-7, 1979. EPA-600/8-79-014. May 1979.
5. George, L. C., Soboroff, D. M., and A. A. Cochran. Regeneration of Waste
Chromic Acid Etching Solutions in an Industrial. Scale Research Unit.
In: Third Conference on Advanced Pollution Control for the Metal
Finishing Industry, sponsored by the U.S. EPA, Cincinnati, Ohio.
EPA-600/2-81-028. 1981.
6. Soboroff, D. M., Troyer, J. D. , and A. A. Cochran. Regeneration and
Recycling of Waste Chromic Acid-Sulfuric Acid Etchants. - Bureau of Mines
Report of Investigations No. 8377. 1979.
7. Laughlin, R.G.W., Forrestall, B., and M. McKim. Ontario Research
Foundation. Technical Manual: Waste Abatement, Reuse, Recycle, and
Reduction Opportunities in the Metal Finishing Industry. Environment
Canada, Toronto, Ontario. January 1984.
8. Horter, G. L. U.S. Bureau of Mines, Rolla, Missouri. Telephone
conversation with Lisa Wilk, GCA Technology Division, Inc. August 29,
1986.
9. Southern Research Institute, Birmingham, Alabama. Electromembrane
Process for Regenerating Acid from Spent Pickle Liquor. Project No.
12010-EOF, submitted to the U.S. EPA Water Quality Office. March 1971.
10. Wheeler C. Seaboard Metal Finishing Company, West Haven, Connecticut.
Telephone conversation with Jon Spielman, GCA Technology Division, Inc.
August 1986.
11. Basta, N. JJse Electrodialytic Membranes for Waste Recovery. Chemical
Engineering. March 3, 1986.
5-113
-------
12. Rodgers, B. Aquatech Systems, Bethel, New Jersey. Telephone
conversation with Jon Spielman, GCA Technology Division, Inc. August 11,
1986.
13. Aquatech Systems. Bethel, New Jersey. Product Literature. Received
August 1986.
14. Pouli, D. Innova Technology, Clearwater, Florida. Telephone
conversation with Lisa Wilk, GCA Technology, Inc. August 26, 1986.
15. Industrial Finishing Staff.' Ion Transfer Recovers Chrome. Industrial
Finishing. March 1981.
16. McDonald, H. 0., and L. C. George. Recovery of Chromium From Surface-
Finishing Wastes. Bureau of Mines Report of Investigations No. 8760.
1983.
17. Horter, G. L., and L* C. George. Demonstration of Technology to Recycle
Chromic Acid Etcbants at Gould, Inc. In: Proceedings of the 4th
Recycling World Congress and Exposition. 1982.
18. Soboroff, D. M., Troyer, J. D., and A. A. Cochran. U.S. Bureau of Mines,
Rolla, Missouri. Regeneration of Waste Metallurgical Process Liquor.
U.S. Patent No. 4, 337, 129. June 29, 1982.
19. Herdrich, W. J. Recovery of Acid Etchants at Impervial Clevite, Inc.
In: Fourth Conference on Advanced Pollution Control for the Metal
Finishing Industry, sponsored by the U.S. EPA. EPA-600/9-82-022. 1982.
20. George, L. C., Rogers, N. L., and G. L. Horter. Recovery and Recycling
of Chromium-Bearing Solutions. In: Proceedings of the 29th National
SAMPE Symposium. April 3-5, 1984.
21. Gary, S. Scientific Control, Inc., Chicago, Illinois. Telephone
conversation with Lisa Wilk, GCA Technology Division, Inc. August 29,
1986.
22. Altmayer, F. Introducing the Cops: One-Step Regeneration and
Purification of Chromic Acid Pickling/Stripping Solutions. Plating and
Surface Finishing. March 1983.
23. Horter, G. L., Stepbenson, J. B., and W. M. Dressel. Permselective
Membrane Research for Stainless Steel Pickle Liquors. In: Proceedings
of the International Symposium on Recycle and Secondary Recovery of
Metals; sponsored by the Metallurgical Society of AIME, Warrendale,
Pennsylvania; held in Fort Lauderdale, Florida. December 1-4, 1985.
24. S potts, D. A. Economic Evaluation of a Method to Regenerate Waste
Chromic Acid-Sulfuric Acid Etchants. Bureau of Mines Information
Circular No. 8931. 1983.
/
25. Aquatech Systems. Bethel, New Jersey. Information on Leasing Program.
January 21, 1986.
5-114
-------
5.5 REVERSE OSMOSIS
5.5.1 Process Description
Reverse osmosis (RO) is a treatment technique used to remove dissolved
organic and inorganic materials, and to control amounts of soluble metals,
1 2
IDS, and TOC in wastewater streams. ' The technology has been applied in
the metal finishing industry to recover plating chemicals from rinse water,
such that both plating chemicals and rinsewaters can be reused.
RO involves passing the wastewater through a semipermeable membrane at a
pressure greater than the osmostic pressure caused by the dissolved materials
234
in the solvent. ' ' Thus, the osmotic flow, defined as the flow from a
concentrated solution to a dilute solution, is reversed due to the increase in
pressure applied to the system.
The application of RO to treatment of corrosive wastes is generally
limited by the pH range in which the membrane can operate. Only a limited
number of membranes are applicable for recovery of corrosive plating and other
solutions. Table 5.5.1 summarizes the characteristics of these commercially
available membranes.
The semipermeable membranes can be fabricated either in the form of a
>du]
10
3 9
sheet or tube, which is then assembled into modules. ' Figure 5.5.1 shows
the three basic module designs, which include:'
• Tubular - a porous tubular support with the membrane cast in place,
or inserted into the tube. Feed is pumped through the tube;
concentrate is removed downstream; and the permeate passes through
the membrane/porous support composite.
• Spiral Wound - large porous sheet(s) wound around a central permeate
collector tube. Feed is passed over one side of the sheet, and the
permeate is withdrawn from the other.
• Hollow Fiber - thousands of fine hollow fiber membranes (40 to 80 ym
diameter) arranged in a bundle around a central porous tube. Feed
enters the tube, passes over the outside of the fibers, and is
removed as concentrate. Water permeates to the inside of the fibers
and is collected at one end of the unit.
5-115
-------
TABLE 5.5.1. COMMERCIALLY AVAILABLE RO MEMBRANES APPLICABLE
TO THE TREATMENT OF CORROSIVE WASTES
Operating Module
Membrane Membrane Optimal pressure replacement
type description Source pH range3 (psig) cost ($)
RC-100 Flat Sheet Composite Fluid Systems 1 to 12 400 to 800 1,000
Membrane of Division of
Polyether/Amide on UOP, Inc.
Polysulfone, rolled (San Diego, CA)
into Cartridge
PA-300
(TFC-PA) Polyether/Amide Fluid Systems 1 to 7 400 to 800
Spiral-Wound Division of
UOP, Inc.
(San Diego, CA)
Thin Film Desalination 1 to 7 400 to 800 560
Spiral-Wound Systems, Inc.
Polyamide (San Diego, CA)
Thin Film Desalination 4 to 11 400 to 800 320
Spiral-Wound Systems, Inc.
Polyamide (San Diego, CA)
Spiral-Wound Osmonics, Inc. 1 to 12 200 to 800
Polyamide (Minnetanka, MN)
treatment of solution with pH outside of this range is possible, but will
result in more rapid deterioration of the membrane.
Source: References 5,6,7, and 8.
-------
A. TUBULAR MEMBRANE
CASING
WATER
PLOW
MEM4RANE
b. SPIRAL-WOUND MODULE
ROLL TO
ASSEMBLE
FEED SIDE
SPACER
PERMEATE OLTT
PERMEATE SIDE BACKING
MATERIAL WITH MEMBRANE ON
£ACH SIDE AND GLUED AROUND
EDGES AND TO CENTER TUBE
FEED FLOW
PERMEATE PLOW
(AFTER PASSAGE
THROUGH MEMBRANE!
c. HOLLOW-FIBER MODULE
SNAP RING
FEED
"O" RING
SEAL
CONCENTRATE
OUTLET
OPEN END
OF FIBERS EPOXY
TUBE SHEET
pcaous
SACK-UP DISC
NAP RING
POROUS FEED
DISTRIBUTOR
TUBE
END PLATE
END PLATE
Figure 5.5.1.
Reverse osmosis membrane module configurations.
Source: Reference 3.
5-117
-------
Although they are able to operate at higher pressures, tubular modules
.are not applicable for most industrial applications because of large floor
space requirements and high capital costs as compared with other modules.
Hollow fiber and spiral wound modules have similar capital costs. Hollow
fiber modules require less floor space, but spiral wound modules are not as
11 12
susceptible to plugging by suspended solids. '
Reverse osmosis systems typically consist of a number of modules
connected in series or parallel, or a combination of both arrangements. In a
series arrangement, the reject stream from one module is fed directly to
4
another module, such that greater product concentration is achieved.
Alternatively, the reject stream may be recycled to the feed stream of the
same RO unit. Series treatment may be limited in some-cases by the ability of
4
the membrane to withstand concentrated contaminants. The system capacity
can be increased through the use of a parallel arrangement of modules;
4
however, product quality will not be enhanced. Schematic flow diagrams of
series and a parallel systems are shown in Figure 5.5.2.
Operating Parameters—
The operatio.n of a reverse osmosis system is affected primarily by the
feed characteristics, operating pressure, and membrane type. These factors
will affect the flux and percent rejection, which in turn, define system size
requirements and effluent quality, respectively.
Flux determines the system size for a given waste flow rate; i.e., higher
flux permits the use of smaller RO systems. Flux is the volume flow of
permeate per unit membrane area. It is proportional to the effective pressure
driving force, according to the following relationship:
J - K (AP - AD)
where J'flux, K»membrane constant; AP"difference in applied pressure across
the membrane; and A Indifference in osmotic pressure across the membrane*
5-118
-------
FILTERED
WASTE
FEED
CONCENTRATE
(REJECT STREAM)
TO EVAPORATOR
OR
PLATING SOLUTION
FILTERED
WASTE
FEED
PERMEATE TO
RINSE TANK
CONCENTRATE
(REJECT STREAM)
TO EVAPORATOR
OR
PLATING SOLUTION
PREMEATE TO RINSE TANK
(a.) Parallel Operation
(b.) Series Operation
FILTERED
WASTE -
FEED
CONCENTRATE
(REJECT STREAM)
TO EVAPORATOR
OR
PLATING SOLUTION
PERMEATE TO RINSE TANK
(c.) Combination Parallel-Series Operation
Figure 5.5.2. Reverse Osmosis Module Arrangements.
Reference: No. 4
-------
Since osmotic pressure is approximately proportional to molar feed
.concentration, flux increases with increasing operating pressure, and it
decreases with increasing feed concentration. Thus, chemicals which form
high-molecular weight complexes will have higher flux for a given weight
percent in solution.
More concentrated solutions can be achieved by utilizing a large
effective driving pressure. Increases in temperature of the waste feed will
also increase the flux by lowering viscosity. However, although
increased operating temperatures will improve the performance of the system in
the short-term, the lifetime of the membrane will be shortened.
Percent rejection is defined as follows:
.... . (feed concentration) - (permeate concentration) , nn<*
Z - Rejection » E————: x 100%
J permeate concentration
Higher percent rejections will result in better quality (higher purity) of the
permeate and concentrated streams. Percent rejection is primarily affected by
the membrane type although rejection will decrease with increasing feed
fc fc- 10
concentration. - •
Pretreatment Requirements/Restrictive Waste Characteristics—
Colloidal and organic matter can clog the membrane surface, thus reducing
the available surface area for permeate flow. Also, low-solubility salts will
precipitate on the membrane and similarly reduce membrane efficiency.
Pretreatment techniques such as activated carbon absorption, chemical
precipitation or filtration may be required to ensure extended service life.
Operating costs for membrane systems are a direct function of the concentration
of the impurity to be removed, due in part to increased maintenance and
membrane replacement costs.
Multi-charged cations and anions are easily removed from the wastewater
with this technique* However, most low molecular weight, dissolved organics
are, at best, only partially removed with this method. The use of reverse
osmosis for recovery/reuse of corrosive process wastes is currently somewhat
limited because RO membranes are attacked by solutions with a high oxidation
5-120
-------
potential (e.g., chromic acid) or corrosive pH levels. However, future
development of membranes which are able to withstand corrosive and oxidizing
solutions is expected.
Post-Treatment—
Reverse osmosis applied to plating bath wastes is usually supplemented
with an evaporation system in order to adequately concentrate constituents for
reuse. Increases in concentration of corrosive constituents will tend to
degrade the membrane. Thus, the amount of feed concentration permitted in an
BO unit is limited by the membrane characteristics. RO units can concentrate
•ost divalent metals (nickel, copper, cadmium, zinc, etc.) from rinse waters
to 10 to 20 percent of the bath solution. Further concentration must be
achieved through the use of a small evaporator. Evaporators are especially
necessary for systems operated at ambient temperatures where atmospheric
evaporative losses are minimal.
5.5.2 Process Performance
Reverse osmosis applications i;n recycling corrosive wastes is somewhat
limited due to membrane degradation in the extreme pH regions. However,
research has been conducted using RO to recover both acidic and basic plating
rinsewaters, which has led to the development of more chemically resistant
•embranes.
The Walden Division of Abcor, Inc. (Wilmington, MA) conducted a number of
studies of RO systems for recovery of plating rinsewaters. Initially, studies
were conducted to test the applicability of membrane types to various plating
rinses. Test samples were prepared by diluting actual plating bath solutions
irith deionized water.10 Bath properties (total dissolved solids, pH) and
test solution properties (concentration, pH) for the corrosive wastes* tested
is presented in Table 5.5.2.
The RO performance was evaluated with other wastes during ^is testing
'program! however, only the data for the corroszve wastes tested 1S
presented in this section.
5-121
-------
N>
TABLE 5.5.2. SUMMARY OF THE CHARACTERISTICS OP THE CORROSIVE WASTES TREATED BY
REVERSE OSMOSIS DURING PRELIMINARY TESTING CONDUCTED BY THE
WALDEN DIVISION OF ABCOR, INC.
Bath properties
Plating bath
Chromic -AeTd"
Cadmium Cyanide
Zinc Cyanide
Copper Cyanide
Source of bath
Whyco Chromium Co.
American Electroplating Co.
American Electroplating Co.
American Electroplating Co. .
X r TDS
27.5
26.3
11.4
37.0
PH
0.53
13.1
13.9
13.3
Teat soln properties
Concentration
range (X-TDS)
0.4-9.0
0.3-10.0
0.5-4.0
0.6-8.0
pH range
0.9-1.9
11.4-12.9
12.3-13.7
11.8-12.9
Membrane
modules
tested8
A,B,C
A
A
A
Notes: aA - Dupont B-9 permeator, polyamide hollow-fiber membrane.
B " T.J. Engineering 97 H 32 spiral-wound module, cellulose acetate membrane,
C • Abcor TM 5-14 module, tubular configuration; cellulose acetate membrane.
Source: Reference 10.
-------
The membrane performance is affected by feed constituent concentrations,
•operating pressure, operating temperature, flow rate, and pH. In tests
performed by Abcor, the effects of pH were evaluated by comparing results of
corrosive test solutions with neutralized test solutions.10 Flux and
rejection data were not affected by changes in pH, but extreme pH values were
found to decrease the membrane life. Additional tests were performed to
evaluate the effects of feed concentrations on flux and %-rejection.
Corrosive waste rinse streams that were tested included: chromic acid,
cadmium cyanide, zinc cyanide, and copper cyanide. In order to calculate
these values, samples of feed, concentrate, and permeate were analyzed for
IDS, temperature, pH, and chemical composition. The analytical test methods
used are presented in Table 5.3.3. A summary of the operating parameters and
results for these tests is presented in Tables 5.5.4 through 5.5.7. As can be
seen from these tables, both flux and rejection decrease with increasing feed
concentration.
Rejection and flux results were satisfactory for the chromic acid test
•
runs, but hydrolysis and degradation of the membranes occurred. Three types
of membranes were tested on the chromic acid test solution: hallow-fiber
'polyamide membrane, spiral-wound cellulose acetate membrane,- and tubular
cellulose acetate membrane. Although the tubular module showed the best
results, large floor space requirements and high capital costs make this
alternative less cost-effective. Only polyamide membranes could be used for
testing of the cyanide plating rinses because the cellulose acetate membranes
are rapidly degraded in the high pH range. ' Good results were observed for
rejection values.
Additional tests were conducted on a pilot-scale level at American
Electroplating Company in Worcester, MA to recover a zinc cyanide plating
rinse.
13
Since the system operated at room temperature, an evaporator was
used in conjunction with the RO unit in order to achieve the level of
concentration necessary for closed-loop operation.
A schematic «jf the pilot system used to recover the zinc cyanide plating
rinse is presented in Figure 5.5.3. Feed was pumped from the first plating
rinse tank to a series of cartridge filters in parallel. Filters ranging from
1 to 20 m were used during the testing. After filtration, the feed was
pressurized to 700 psi using a multi-stage centrifugal feed pump before being
passed-through spiral-wound RO modules arranged in series. The permeate from
5-123
-------
TABLE 5.5.3. ANALYTICAL TEST METHODS USED DURING PRELIMINARY
INVESTIGATIONS OF REVERSE OSMOSIS APPLICATIONS
CONDUCTED BY WALDEN DIVISION OF ABCOR, INC.
Constituent
Test method
Procedure
Reference No.
Total chrome
Hexavalent chrome
Copper
Zinc
Cyanide
Total organic carbon
Total dissolved solids
Atomic absorption
Atomic absorption
Colorimetrie
Atomic absorption
Atomic absorption
Titration
Combustion-infrared
Gravimetric
129
129
108
129
129
207B
138A
148B
Source: Reference 10.
5-124
-------
TABLE 5.5.4. ANALYTICAL RESULTS FOR REVERSE OSMOSIS TREATMENT
OF SPENT CHROMIC ACID PLATING RINSE
Operating conditions
Vafst-f' ft*r*r\ -.-^-
Membrane
module3
Hollow fiber
spiral
tubular
Hollow fiber
spiral
tubular
Hollow fiber
spiral
tubular
Hollow fiber
spiral
tubular
Pressure Temp.
Z-TDS Z of bath (psig) (°C)
400
0.40 1.5 600 29
800
400
1.83 6.7 600 29
800
400
4.11 15 600 29
800
400
9.43 34 600 28
800
pH of
Feed Fluxb
2.59
1.9 15.3
10.0
1.97
1.2 13.2
8.58
1.20
1.2 10.6
7.31
leak
0.9 leak
6.60
% - Rejection
Basis:
TDS
84
97
99
95
94
97
90
92
95
leak
leak
94
Basis:
Cr+6
97
96
98
87
86
91
91
92
7
leak
leak
97
aThree commercially—available membrane modules were tested:
DuPont B-9 hollow-fiber module (polyamide membrane);
T.J. Engineering 97H32 spiral-wound module (cellulose acetate membrane); and
Abcor, Inc., TM5-14 tubular module (cellulose acetate membrane).
bGallon/minute/single DuPont B-9 permeator size 0440-035.
Gallon/day/ft^ for spiral-wound and tubular modules.
Source: Reference 10.
5-125
-------
TABLE 5.5.5. ANALYTICAL RESULTS FOR REVERSE OSMOSIS TREATMENT
OF SPENT COPPER CYANIDE PLATING RINSE
Operating conditions
waste teed — — — ______
___-—-_-— ——_—__ Pressure Temp. pH of
Z-TDS Z of bath (psig) (°C) Feed Fluxa
1.93 5.2 400 26 12.2 1.20
3.71 10 400 26 12.5 0.62
7.98 22 430 27 12.9 0.076
Ballon/minute permeator. Hollow fiber membrane module
(poly amide membrane).
Source: Reference 10.
TABLE 5.5.6. ANALYTICAL RESULTS FOR REVERSE OSMOSIS
OF SPENT CADMIUM CYANIDE PLATING RINSE
Operating conditions
Unot-» FaftA ______...,... i. _ ,. i. ...._,______
waste ceea __________________________
.._._.... __________ Pressure Temp. pH of
Z-TDS Z of bath (psig) (°C) Feed Flux*
2.43 9 400 27 12.5 0.67 *
3.12 12 400 27 12.5 0.24
9.82 37 430 27 12.9 0.028
Z - Rejection
TDS Cu+
98 99+
97 99+
77 84
TREATMENT
CN-
99+
99+
92
Z - Rejection
TDS Cd+
97 99
96 99
9.2 78
CN-
95
92
10
aGa lion/minute permeator. Hollow fiber membrane module
(poly amide membrane).
Source: Reference 10.
TABLE 5.5.7. ANALYTICAL RESULTS FOR REVERSE OSMOSIS TREATMENT
OF SPENT ZINC CYANIDE PLATING RINSE
Operating conditions
Una fa fooH ____---. ^»»_LJ._L T- -• n .__—
------ •'-"»-•—•—- Pressure Temp. pH of
Z-TDS Z of bath (psig) (°C) Feed Flux*
2.43 9 400 27 12.5 0.67
3.12 12 400 27 12.5 0.24
9.82 37 430 27 12.9 0.028
Z - Rejection
TDS Zn
97 99
96 99
9.2 78
CN-
95
92
10
aGalion/minute permeator. Hollow fiber membrane module
(polyamide membrane). S-126
Source: Reference 10.
-------
DISTILLATE
(0
_ • •"«—
EVAPORATOR
CONCENTRATE
(1.0 gpm)
EVA
.24 gpm)
OVERFLOW
(0.02 gpm)
PLATING BATH
RO FEED
(2.22 gpm)
EVAPORATOR • ^
I
V^x-»--^— .
HOLDING
TANK
f
RINSE
TANK #1
fit
FEED
(1.02 gpm)
^ FEED PUMPC
k
[5 '
FILJ
(700 psl) 1
RO
onp.™ CONCENTRATE
(0.22 gpm)
RO
SYSTEM
30%
CONV.
~~| DISTILLATE
(0.22 gpm)
t,
RINSE
TANK #2
PERMEATE
(2 gpm)
ERS
-RECIRCULATION
(10 gpm)
Figure 5.5.3.
Schematic of reverse osmosis/evaporation system used
to recover zinc cyanide plating solution at American
Electroplating Company.
Source: Reference No. 13.
5rl27
-------
the RO system was returned to the first rinse tank, and the concentrate from
the last RO module was sent to the evaporator. The distillate from the
evaporator was pumped back to the second rinse tank, and the concentrate was
metered back to the plating tank. Operating parameters for the system are
presented in Table 5.5.8.
The typical composition of the plating bath is presented in Table 5.5.9.
Samples were collected and analyzed for zinc, free cyanide, total solids,
13
conductivity, and pH. The system was operated at a fixed feed
concentration in order to evaluate any trends in the flux and rejection data.
The system was operated for a total of 1,000 hours. However, the membranes
were exposed to the highly alkaline (pH 13) rinsewater for a total of
4,200 hours because the membranes remained in contact with the plating rinse
during downtime. Total exposure time is a good measure of the chemical
resistance of the membranes whereas total operating time is a better indicator
of the effect of membrane plugging due to particulates present in the feed.
Both flux and rejection values gradually declined during the operating
period as a result of membrane plugging. However, after cleaning the RO
•
modules with a 2 percent citric acid solution, rejection and flux values
improved within 24 hours. Periodic cleaning of the modules was also found to
extend their lifetime. •
Although RO membrane systems are currently available which can be used
with corrosive wastes, lifetimes for these membrane systems are a limiting
factor in their cost-effective application. Research is currently being
conducted to develop more chemically-resistant membranes.
5.5.3 Process Costs
The economics of RO systems varies with the operating parameters,
membrane type, waste feed characteristics, and desired product quality.
Capital costs include costs for RO modules, evaporation system, pumps,
and associated piping and instrumentation. Typical capital costs average
$25,000 for a RO system, and approximately $40,000 for the evaporation
system.
5-128
UM or diaclomre of data is
wbject to the restriction on
the title page of thU propoul
ALLIANCE
Technologies Corporation
-------
TABLE 5.5.8. TYPICAL OPERATING PARAMETERS DURING TESTING OF REVERSE
OSMOSIS SYSTEM AT AMERICAN ELECTROPLATING COMPANY
FOR THE RECOVERY OF ZINC CYANIDE PLATING RINSE
Item Parameter
Feed pressure 700 psi
Recirculation flow rate 10 gpm
Temperature 70 to 90°F .
Concentrate flow rate 0.2 gpm
Evaporator vacuum 26 to 27 in. Hg
Evaporator temperature 100 to 110°F
Evaporator steam pressure 5 psi
Evaporator concentrate flow rate 1 gpm
Source: Reference 13.
5-129
-------
TABLE 5.5.9. TYPICAL COMPOSITION OF ZINC CYANIDE PLATING
BATH AT AMERICAN ELECTROPLATING COMPANY
Constituent
Concentration
Zn (as metal)
CN (as NaCN)
Caustic
Brightener
Polysulfide carbonates
Total solids
20,000 mg/L
(2.7 oz/gal)
60,000 mg/L
(8.0 oz/gal)
75,000 mg/L
(10.0 oz/gal)
4 mg/L
(4 gal/1,000 gal)
155,000 mg/L
350,000 mg/L
(35Z by weight)
Source: Reference 13.
5-130
-------
Capital and operating costs will increase proportionally with system
.size. System size will be determined by the desired conversion, the feed
concentrations, operating temperatures and pressures, and membrane type.
Table 5.5.10 presents estimates of capital costs for various RO system
conversions.
Operating costs will include membrane replacement costs, filter
replacement costs, energy costs for the pumps, and evaporator, operating labor
(minimal) and maintenance costs for cleaning and replacing membranes and
filters. Table 5.5.11 summarizes the operating cost requirements.
Figure 5.5.4 shows that minimum total costs using an RO/evaporation system
combination occur when the RO system conversion is 90 percent. Typical
operating costs for an RO/evaporation system designed for 90 percent
conversion are $12,000/year.
Savings will be realized in neutralization and disposal costs, chemical
purchase costs, and process water requirements. An EPA study of the economics
of a reverse osmosis /evaporation system for the recovery of zinc cyanide from
rinsewaters showed only a $10,000 savings/year for rinsewater treatment, water
and makeup chemical costs, which was not deemed to be cost-effective. '
However, it is anticipated that the savings will increase as costs for
neutralization and disposal increase.
In summary, the cost-effectiveness of the reverse osmosis technique is
dependent upon the following factors: production rate, type and concentration
of rinsewater constitutents, water supply, wastewater disposal costs, and
useful lifetime of RO membrane. As more chemically resistant membranes are
developed RO systems will have more cost-effective applications for corrosive
waters. Also, with the implementation of the land disposal ban and the
resulting rise in sludge disposal costs, reverse osmosis will become a more
cost-effective alternative to conventional neutralization practices.
5.5.4 Process Status
Reverse osmosis systems have been widely applied in wastewater
desalination and are also currently available for recovering corrosive
wastewater streams. Research has focused on the recovery of corrosive plating
5-
-------
TABLE 5.5.10. CAPITAL COSTS FOR VARIOUS RO SYSTEM CONVERSIONS
• •
Item
Required permeate
flow (gpm)
Required membrane
area (sq ft)a
Required number
of modules5
Membrane module
RO system
0 70
0 2.575
0 371
0 6
0 3,780 5
conversion
80
3.625
522
6
,040
«>
90
5.85
842
12
7,560
95
8.91
1.283
18
11,340
cost ($)c
Housing cost ($)d 0 1,700 2,559 3,400 5,100
Total cost 0 21,780 23,890 28,560 34,040
for RO system ($)e
Required evaporator 120 66.2 54.4 39.0 28.1
capacity (gph)*
Total cost 44,129' 39,199 39,199 33,880 33,880
for evaporator ($) •
Total system 44,129 60,979 63,089 62,440 67,920
*Design flux - 10 g/f/d.
bBased on area of 70 ft2/module.
cBased on $630/module.
dfiased on $850/3-module housing.
eBased on system cost of $1,500 for system without modules, housings, and
high-pressure pump. Pump/motor cost » $1,300 for 4 gpm permeate; $2,600
for 4 gpm permeate.
fDouble-effect evaporator with cooling tower package. Based on rated
capacities of 200 gph ($44,129), 100 gph ($39,199), and 50 gph ($33,880).
Source: References 8, 13, and 15.
5-132
-------
TABLE 5.5.11. OPERATING COSTS FOR A REVERSE OSMOSIS SYSTEM
Operating requirement
Cost basis
Unit cost
Steam heat requirements
(for evaporator)
Electrical
Cartridge filters
Cleaning chemicals
No. 4 fuel oil
at $0.393/gal;
Heating value of
$140,000 BTU/gal;
80% boiler efficiency
To run pressure pump
and evaporator
4 cartridges/month
(average)
3 cleanings/year
with citric acid
$3.50/100 Ibs
$0.055/KWH
$4.68/cartridge
$0.82/lb
Source: References 13 and 16.
5-133
-------
ANNUAL OPERATING COST. I
o>
3
JL
£
*-
S
oo
o
g
IM
o
CM
ro
3
Figure 5.5.4. Annual operating costs for various RO system conversions.
Note: System costs shown in this figure include direct operating costs and capital amortization
costs (based on straight-line depreciation over 10 years with zero salvage value).
Source: Reference 13.
-------
rinses. However, cost-effective use of RO systems for this application is
.generally limited due to the shortened lifetime of membranes exposed to
corrosive wastes and the associated high costs for membrane cleaning and
replacement. However, future development of membranes which are able to
withstand corrosive and oxidizing solutions is expected.
If membrane resistance is increased, reverse osmosis would be a very
cost-effective alternative to conventional treatment technologies. Excluding
membrane cleaning and replacement costs, the only significant operating cost
is the electricity required for operation of the pump. However, current
systems are limited in the degree of attainable concentration of the reject
stream which frequently requires the use of an evaporator in conjunction with
the RO unit. The use of a combination reverse osmosis/evaporation treatment
system can be more cost-effective than evaporation alone.
Five systems designed by Osmonics, Inc. (Hinnetaka , Minnesota) are
g
currently in operation for the treatment of corrosive wastes. Although
detailed performance data is not available for these systems, performance has
0
reportedly been satisfactory. Fluid Systems Division of UOP, Inc. and
Desalination Systems, Inc. (both in San Diego, CA) currently market RO
membranes that are capable of being used in the corrosive pH ranges. '
However, membrane replacement is more frequent compared to. use with more
neutral solutions. Ionics, Inc. (Watertown, MA) is currently conducting
14
research to develop a more chemically resistant membrane. However,
preliminary test results are not available at this time.
5-135
-------
REFERENCES
1. Metcalf & Eddy, Inc. Wastewater Engineering: Collection, Treatment, and
Disposal. McGraw-Hill Book Company, NY. 1972.
2. U.S. EPA. Treatability Manual, Volume III: Technologies for
Control/Removal for Pollutants. EPA-600/8-80-042c. July 1980.
3. U.S. EPA, Office of Research and Development, Cincinnati, OH. Handbook
for Remedial Action at Waste Disposal Sites. - EPA-625/6-85-006.
October 1985.
4. U.S. EPA. Sources and Treatment of Wastewater in the Nonferrous Metals
Industry. Prepared by Radian Corporation for the U.S. EPA, Industrial
Environmental Research Laboratory, Cincinnati, OH, under EPA Contract
No. 68-02-2068. EPA-600/2-80-074. April 1980.
5. Biggins, T.E., C^M HILL. Industrial Processes to Reduce Generation of
Hazardous Waste at DOD Facilities - Phase 2 Report, Evaluation of 18 Case
Studies. Prepared for the DOD Environmental Leadership Project and the
U.S. Army Corps of Engineers. July 1985.
6. Corns tock, D. Desalination Systems, Inc. Telephone conversation with
L. Wilk, GCA Technology Division, Inc. September 19, 1986.
7. Filtwell, J. Fluid Systems Division of UOP, Inc., San Diego, CA.
Telephone conversation with L. Wilk, GCA. Technology Division, Inc.
September 19, 1986. . .
8. Osmotics, Inc., Minnetaka, MN. Telephone conversation with L. Wilk,
GCA Technology Division, Inc. September 19, 1986.
9. Sundstrom, D.W. , and H.E. Klei. Wastewater Treatment. Prentice-Hall
Inc., Englewood Cliffs, NJ. 1979.
10. Donnelly, R.G. , R.L. Goldsmith, K.J. McNulty, and M. Tan. Reverse
Osmosis Treatment of Electroplating Wastes. Plating. May 1974.
11. Crampton, P., and R. Wilmoth. Reverse Osmosis in the Metal Finishing
Industry. Metal Finishing. March 1982.
12. Cushnie, G.C. Centec Corporation, Reston, VA. Navy Electroplating
Pollution Control Technology Assessment Manual. Final Report prepared
for the Naval Facilities Engineering Command under Contract No.
F08635-81-C-0258. NCEL-CR-84-019. February 1984.
5-136
Use or disclosure of data is
subject to the restriction on
the title page of this proposal
ALLIANCE
Technologies Corporation
-------
13. McNulty, K.J., and J.W. Kubarewicz. Field Demonstration of Closed-Loop
Recovery of Zinc Cyanide Rinsewater Using Reverse Osmosis and
Evaporation. In: Proceedings of the 2nd Conference on Advanced
Pollution Control for the Metal Finishing Industry, Co-sponsored by the
American Electroplaters Society and the U.S. EPA Industrial Environmental
Research Laboratory, Kissimmee, FL, February 5-7, 1979.
EPA-660/8-79-014. May 1979.
14. Jain, S.M., Ionics, Inc., Watertown, MA. Telephone conversation with
Jon Spielman, GCA Technology Division, Inc. August 12, 1986.
15. Mahoney, F., F. Mahoney Co./Permutit Systems, Hingham, MA. Telephone
conversation with L. Wilk, GCA Technology Division, Inc. September 19,
1986.
16. Chemical Marketing Reporter. Chemical Prices for Week .Ending
July 18, 1986. 230(3): 32-40. July 21, 1986.
17. McNulty, R.J. Walden Division of Abcor, Inc., Wilmington, MA. Telephone
conversation with Jon Spielman. GCA Technology Division, Inc.
August 4, 1986.
5-137
-------
5.6 DONNAN DIALYSIS & COUPLED TRANSPORT
5.6.1 Process Description
Other membrane technologies currently being researched include Donnan
dialysis and coupled transport. Neither of these technologies is presently
commercially available. However, both show potential for cost-effective
applications in the recovery of corrosive plating solutions.
Donnan dialysis and coupled transport are similar processes in that both
employ a concentration gradient to drive ions from a spent solution across a
membrane into a stripping solution. These processes utilize a cell consisting
of a membrane separating two compartments. The waste is fed to one
compartment and the stripping solution (into which the concentrate will be
extracted) is fed to the other. As with electrodialysis, the cells are
generally arranged in stacks or multi-cells containing many membrane-spacer-
units which allows treatment of larger quantities of waste at a faster rate.
Figure 5.6.1 diagrams a typical stack unit.
Unlike other membrane technologies (i.e., electrodialysis and reverse
osmosis), energy requirements are minimal for Donnan dialysis and coupled
transport. The only energy required is the energy to pump the feed and
2
stripping solutions across the cell. The large hydraulic pressures
required for reverse osmosis and the large electric current flow required by
2
electrodialysis, are not required for these technologies.
The major difference between these processes is the type of membrane
employed, and the transport mechanisms involved. The coupled transport
membrane is highly selective and therefore has more specific process
applications, whereas the Donnan dialysis membrane has application to a wider
variety of solution constituents. However, 'greater purity can be achieved
using the coupled transport membrane.
Donnan dialysis uses an anion- or cation- selective membrane, which
functions similarly to ion exchange resins. For an anion exchange membrane,
cations in both solutions (on each side of the membrane) are prevented from
diffusing across the membrane, but anions will redistribute themselves across
the membrane until equilibrium is reached; i.e., ratios of all similarly
charged anions are equal. With a cation-exchange membrane, cations will
5-138
-------
End Plate Flow Spacer Membrane Flow Spacer End Plate
Feed
Strip
Note: The assembly bolt holes in the end plates are not shown.
Figure 5.6.1. Membrane stack unit.
Source: Reference 1.
5-139
-------
diffuse across the membrane and movement of anions will be restricted. The
driving force for these exchange reactions is the potential created by the
displacement of the system from the equilibrium ratios which can be controlled
2
by adjusting solution concentrations.
The microporous membrane used in the coupled transport process contains a
metal-complexing agent within its pores. ' Ions combine with the
completing agent and are removed from the spent solution. On the other side
of the membrane, the ion solubility is favored over that of the complex in the
stripping solution. Thus, a transfer of the ion across the membrane occurs
4
due to the coupling of these two complexation reactions. The membrane used
in the coupled transport is selective to the transport of one metal ion over
other ions; i.e., it is highly specific. Therefore, a different membrane is
required for each application. However, the greater selectivity of the
membrane allows more complete removal of the ion from the spent solution.
The two types of coupled transport reactions which may occur are
A
cotransport and countertransport. With cotranaport, ions are only
tranferred in one direction across the membrane. A typical cotransport
process is diagrammed in Figure 5.6.2. In countertransport processes, the
transport of one ion is balanced by the flow of another ion in the opposite
direction. A. generalized flow diagram of a countertransport system for
recovering metal ions is shown in Figure 5.6.3.
Operating Parameters—
The Donnan dialysis operation is based upon the Donnan equilibrium
principle, which can be described by the following equation:2'6
Constant
Where: C » activity, approximately equal to concentration,
i * any mobile ionic species,
1 • left-hand side of membrane,
r * right-hand side of membrane,
Z - valence of the ionic species.
5-140
-------
ui
I
o
e
H
n
n
&
HI
re
O
n
(C
Oi
*
H-
f*
t"t
a
Ui
•
-------
Low or moderate pH, dilute
solution of M** Ions
2RH(org,
2H(aq) * R2M(org)
MIcroporous membrane
containing completing
agent RH
Lower pll. concentrated solution
of M'+ ions
R2M(org) + 2H*(aqj
2RH. . * M*», .
(org) (aq)
Figure 5.6.3. Counter-transport process for metal cation recovery.
Source: Reference 5.
5-142
-------
The driving force which causes the ions to move from one side of the
membrane to the other is the concentration potential resulting from the
differences in concentration between similarly charged ions (i.e., both
positive or both negative) on each side of the membrane. The degree of
separation can be increased by any of the following changes in the system:
• Increasing initial concentrations on one side of the membrane; i.e.,
in the spent solution;
• Decreasing initial concentrations on the other side of the membrane;
i.e., stripping solution;
• Adding a complexing agent to remove ions from the solution on one
side of the membrane; and/or
• Using a countercurrent flow system.
This equilibrium principle is also in effect during a coupled transport
process. However a complexing agent is incorporated into the membrane which
increases the quantity of a specific ion that can be transported across the
membrane resulting in improved separation.
• The operation of both of these processes is most significantly affected
by the membrane characteristics. Important membrane properties include:
equilibrium water content, ion—exchange capacity, osmotic water transport
rate, metal complex anion transport rate, chemical resistance, and physical
resistance.
The equilibrium water content is the amount of water which the membrane
holds when the system has reached Donnan equilibrium. It is generally
expressed as the grams of water absorbed per dry gram of membrane. This
2
property measures the hydrophilicity of the membrane.
The ion exchange capacity of the membrane can be expressed as the number
of equivalents (eq) per dry gram of membrane, where an equivalent is equal to
the molecular weight of the ion in grams divided by its electrical charge or
valence. It indicates the quantity of ions that can be exchanged with the
given exchange medium. For example, a membrane with an exchange capacity of
1 eq/dry g could remove 37.5 g of divalent zinc from solution per dry gram of
membrane (Zn+2, molecular weight - 65, 65/2 - 37.5). A Donnan membrane is
5-143
-------
non-specific in that it will transfer all anions (anion-exchange membrane) or
all cations (cation-exchange membrane). However, a coupled transport membrane
is highly specific and will only transport one type of metal ion.
The osmotic water transport rate is the amount of water that can be
transported across the membrane in a given time period. It is typically
expressed as a rate per unit of membrane area; i.e., volume/unit time/membrane
area.
The metal complex ion transport rates can be determined using the
2
following first order rate equation:
In (C /C)
o
Where: k * rate constant,
In * natural log,
t « residence time in the cell,
Co * initial concentration of ion in cell inlet, and
C « concentration of ion in cell outlet.
The residence time (t) can be calculated by dividing the cell volume by
the feed flow rate. The metal ion transport,rates will be an important factor
in selecting the membrane area required for a specific application. The rate
of transport across the membrane will vary with changes in pH and solution
3 8
concentrations. '
The physical and chemical resistance of the membrane will determine the
membrane stability. Lower operating costs will be incurred with more
resistant membranes.
Pretreatment—
Pretreatment in the form of filtration may be necessary to remove any
suspended particulates present in the spent solution. Suspended particulates
can clog the membrane and reduce membrane efficiencies.
Post-treatment—
Post-treatment requirements can not be evaluated for Donnan dialysis at
this time due to the lack of available performance data. Post-treatment
5-144
-------
requirements for coupled transport will vary with the specific application.
.In some cases, the recovered product will not be suitable for direct return to
the plating bath; e.g., chrome recovery as discussed below. In these cases,
it may be necessary to employ an ion exchange treatment process to convert the
recovered product to a reusable form.
5.6.2 Process Performance
Limited performance data is available for these membrane technologies.
Donnan dialysis has only been tested in laboratory-scale experiments.
Attempts to test the process on a pilot-scale level were hindered by
mechanical problems. Coupled transported has been both laboratory and field
tested. The information presented in this section is a summary of available
findings to date.
Donnan Dialysis—
Although the Donnan dialysis process has been in existence for quite some
269
time, only laboratory-scale demonstrations have been performed. ' ' In
addition, the- Southwest Research Institute (San Antonio, Texas) attempted to
develop a pilot-scale system for field testing of recovery applications during
studies conducted in 1984 and 1985. Initial testing was conducted to
develop an appropriate membrane for use with metal finishing waste solutions.
Several membranes were evaluated for metal ion transport. A schematic of the
test system is shown in Figure 5.6.4.
Initially, the membranes were tested with relatively low feed
concentrations of about 50 ppm metal ions. Table 5.6.1 presents the results
of these tests. As expected, increased ion exchange capacity correlated
positively with increased metal ion transport rates. In addition, ion
exchange capacity increased with increasing osmotic flow rates.
The four membranes which demonstrated the highest metal ion transport
rates during the preliminary tests were subjected to further testing with
higher metal ion concentrations in the feed (approximately 500 ppm). Table
5.6.2 presents the results of these tests, which show that increasing the feed
concentration lowers the metal ion transport rate.
5-145
-------
L___. L
A - Food Reservoir
B - Strip Reservoir
t - Strip Pump
D - Feed Pump
E - Sample Points
F - Test Cells
G - Flow Meters
Waste
Figure 5.6.4. Membrane test system.
Source: Reference 1.
5-146
-------
TABLE 5.6.1 RESULTS OF PRELIMINARY MEMBRANE TESTING PERFORMED
BY THE SOUTHWEST RESEARCH INSTITUTE
Membrane No.
E11Q4
E12Q4
E12Q5
E14Q4
E15Q1
E16Q1
E16Q2
E18Q1
E18Q2
Ion exchange
capacity
(meq/dry g)
2.5
2.6
1.1
0.7
1.1
2.2
1.2
2.2
1.4
Osmotic water3
flow rate
(ml/hr/cm2)
0.040
0.053
0.010
0.010
0.012
0.102
0.016
0.054
0.009
Metal complex anion transport
rate constant (min~^»^)
Cu Cd
3.1 2.8
2.3 2.2
0
l.lc 0.3
1.0 0.4
2.5
0
2.2 1.4
0 -
Zn
1.9
2.9
-
0.2
0.1
-
-
0.6
-
a0.2N NaCl Versus Deionized H20 - cell effective area -" 122 cm2.
bFeed-50 ppm of metal; stripping solution 0.2N NaCl.
cDuplicate determination of the rate constant also showed 1.1/min.
Source: Reference 1.
5-147
-------
TABLE 5.6.2. RESULTS OF TESTS TO DETERMINE THE EFFECT OF METAL ION CONCENTRATION ON TRANSPORT RATE
00
Equilibrium Ion
• Membrane
EIIQ4
E12Q4
E16QI
E16Q1
No. Membrane type
Vlnylpyridine-grafted
polyethylene (4-VP) CH3I
Vinylpyridlne-grafted poly-
ethylene (4-VP/N-VP) CII31
Vinylbenzyl chloride-grafted
polyethylene (VBC) (CII3) 3N
Vinylbenzyl chloride-grafted
polyethylene (VBC/N-VP)
(CII3) 3N
water
content
g H20/g
1.08
2.45 '
1.52
1.00
exchange
capacity
(meq/dry g)
2.5
2.6
2.2
2.2
Oamotlc*
water
flow rate
ml/hr/cm2
0.040
0.053
0.102
0.054
Metal complex anion transport rate (I/rain)
Low cone. (50 ppm)
Cu Cd ' Zn
3.1 2.8 1.9
2.3 2.2 2.9
2.5
2.2 1.4 0.6
High cone.
Cu Cd
1.5 1.5
0.7 0.8
0.6 0.9
1.41 0.6
(5,000 ppm)
Zn
1.5
0.6
b
0.4
a0.2N NoCl verauB deionieed II2O - cell effective area - 122 cm2.
"Membrane ruptured.
Source: Reference 1.
-------
An important factor in the performance of the system is the stability of
.the membrane. During the testing conducted by the Southwest Research
Institute, it was observed that the metal ion transport rate decreased over
time. The decline in the transport rate was attributed to problems in
membrane stability, the precise cause of which was not determined. Due to
time constraints, the best available membrane was used in the construction of
a prototype Donnan dialysis stack.
Although the Southwest Research Institute intended to test the prototype
Donnan dialysis system for the recovery of various plating solutions, the
project was never completed due to leakage problems. Future studies will
focus on developing membranes with higher metal ion transport rates and also
g
on making mechanical improvements to the dialysis stack.
In summary, the performance of the Donnan dialysis system can not be
realistically evaluated at this time due to lack of available performance
data. However, preliminary testing indicated that a technically feasible
system could be developed.
Coupled Transport—-
The coupled transport process has been laboratory-tested on several types
of metal-containing solutions (primarily metal finishing wastes). The most
developed application is in the removal of hexavalent chromium from plating
rinses.3 Field tests of this application have been performed, but limited
performance data is currently available.
The hexavalent chromium application works most effectively with solutions
having a pH ranging from 0 to 4. * The process is capable of reducing the
level of hexavalent chrome in the rinsewater to less than 1 ppm. However,
unlike other membrane technologies, the product can not be returned directly
to the plating bath.11'12 The product is a 5 to 10 weight-percent solution
of-sodium chromate, which can generally be used for other in-house
processes. Alternatively, the sodium chromate product can be passed through
a cation-exchange column to recover approximately 5 weight-percent chromic
acid for return to the bath.3 The process is generally most effective for
treating solutions with concentrations greater than 50 ppm; however, more
dilute solutions can be treated.
5-149
-------
5.6.3 Process Costs
Realistic costs for the Donnan dialysis process can not be developed at
this time due to the lack of commercial-scale testing. It is expected that
the primary cost would be for the membrane unit. Operating costs would be
expected to be minimal, and savings would be realized in reduced disposal
costs and reduced purchase costs for recovered chemicals.
Preliminary cost estimates for coupled transport hexavalent chromium
treatment systems have been prepared by the developer (fiend Research
Corporation) based on pilot-scale testing.
Capital costs for the coupled transport process can vary widely with the
application. Capital costs for pilot-scale units developed by Bend
Research Corporation have ranged from $500 to $1,000,000. Standardized
process units are not available since each unit has to be custom-tailored for
a specific application. However, currently the most widespread application
for the coupled transport process is for the recovery of chrome in plating
shops. For a typical plating shop, using three countercurrent rinse tanks
(1,000 gallons each), the capital equipment cost would be approximately
$20,000. The total membrane area for this recovery system would be
500 sq ft. An increase-in solution volume to "be handled (at the same pH
and concentration) would require a proportional increase in membrane area.
Operating costs would include periodic maintenance and membrane
replacement. The membrane performance gradually deteriorates over time.
Approximately every 6 months, the coupled transport unit needs to be drained
so that the membrane can be regenerated with a proprietary regenerant
solution. Bend Research Corporation currently performs this service for the
pilot-scale units that they have installed. Costs for regeneration are
approximately $2/sq ft, plus labor. The estimated lifetime of the membrane
(with cleanings every 6 months) is approximately 2 years. However, this
estimate is conservative since membranes have not been tested for periods
longer than 2 years . The approximate unit cost for the membrane is
estimated to be $20/sq ft.3
The process requires minimal day-to-day operating maintenance. If
solution conditions are changed (e.g., pH or concentration changes),/then
modifications, such as increases in membrane area, may be required. /
5-150
-------
Savings will be realized in reduced disposal costs, and benefits from
.recovered chemicals and recovered process waters. The estimated payback
2
period for the chrome recovery system is approximately 2 years.
5.6.4 Process Status
Donnan dialysis has not yet been tested on a pilot-scale. Much of the
research that has been performed to date with Donnan dialysis has concentrated
on membrane development. The Southwest Research Institute (San Antonio, TX)
g
is currently doing research for purposes of developing a pilot scale unit.
Although preliminary research has demonstrated the technical feasibility of
the process, pilot-scale testing is needed to determine if sufficient solution
concentrations can be achieved. The Donnan dialysis process could prove to be
a cost-effective alternative to conventional treatment practices because of
its minimal operating requirements.
Bend Research Corporation (Bend, Oregon) has done most of the development
work for the coupled transport technology and has a patent pending on the
process. Although the process is applicable to the treatment of several
metal-containing solutions, the most developed application is for the
treatment of hexavalent chromium in plating rinses. The process was recently
licensed to Concept Membrane, Inc. for marketing and sales purposes. However,
3 4
commercial units are not currently available for purchase. '
Although the coupled transport process is more developed than Donnan
dialysis, both of these membrane technologies show good potential as
cost-effective alternatives to conventional neutralization and disposal
practices for metal-containing corrosive wastes.
5-151
-------
REFERENCES
1. Hamil, H.F. Southwest Research Institute, San Antonio, TX. Project
Summary: Fabrication and Pilot Scale Testing of a Prototype Donnan
Dialyzer for the Removal of Toxic Metals from Electroplating Rinse
Waters. EPA/600/s2-85-080. August 1985.
2. Hamil, H.F. Southwest Research Institute, San Antonio, TX. Fabrication
and Pilot Scale Testing of A Prototype Donnan Dialyzer for the Removal of
Toxic Metals from Electroplating Rinse Waters. Prepared for U.S.
EPA-ORD, Water Engineering Research Laboratory, Cincinnati, OH, under EPA
Cooperative Agreements CR-807476 and CR-809761. January 1985.
3. Friesen, D. Bend Research Corporation, Bend, Oregon. Telephone
conversation with Lisa Wilk, GCA Technology Division, Inc. September 25,
1986.
4. Higgins, T.E., CH2M Hill. Industrial Processes to Reduce Generation of
Hazardous Wastes at DOD Facilities - Phase 2 Report, Evaluation of 18
Case Studies. Prepared for the DOD Environmental Leadership Project and
the U.S. Army Corps of Engineers. July 1985.
5. Cushnie, G.C. Centec Corporation, Reston, VA. Navy Electroplating
Pollution Control Technology Assessment Manual. Prepared for the Naval
Facilities Engineering Command, Alexandria, Virginia. NCEL-CR-84.019.
February 1984.
6. Davis, T.A., Wu, J.S., and B.L. Baker. Use of Donnan Equilibrium
Principle to Concentrate Uranyl Ions By an Ion-Exchange Membrane
Process. AICHE Journal, 17(4): 1006-1008. July 1971.
7. U.S. EPA. Coupled Transport Systems for Control of Heavy Metal
Pollutants. EPA-600/2-79-181. August 1979.
8. Hamil, H.F. Southwest Research Institute, San Antonio, TX. Telephone
conversation with Lisa Wilk, GCA Technology Division, Inc. September 24,
1986.
9. Wen, C.P., and H.F. Hamil. Metal Counterion Transport in Donnan
Dialysis. Journal of Membrane Science, 8: 51-68. 1981.
10. Martin, M., Bend Research Corporation, Bend, OR. Telephone conversation
with Lisa Wilk, GCA Technology Division, Inc. September 24, 1986.
11. Babcock, W.C. Industrial Water Reuse with Coupled Transport Membranes,
Prepared for the U.S. Dept. of the Interior, Bureau of Reclamation.
Report No. OWR/RU-83-3. May 1983.
12. Babcock, W.C., et al. Renovation of Electroplating Rinse Waters with
Coupled Transport Membranes. In: Proceedings of the Fourth Annual
Conference on Advanced Pollution Control for the Metal Finishing
Industry. EPA-600/9-82-022. December 1982.
5-152
-------
5.7 SOLVENT EXTRACTION
5.7.1 Process Description
Solvent extraction is widely used as an analytical chemistry technique
and for the recovery of metals in the field of hydrometallurgy. 1>2 In
addition, research has shown applications for solvent extraction in the
recovery of spent nitric/hydrofluoric acid pickling liquors generated by the
steel industry. '
Solvent extraction is a separation technique which utilizes the
differential distribution of constituents between the aqueous phase and the
extractant (organic phase) to separate constituents from a mixed solution. A
generalized flow diagram of a solvent extraction process is shown in
Figure 5.7.1. The equipment used in the extraction step can consist of a
single-stage mixing and settling unit, several single-stage units connected in
series, or in a single unit, multi-stage unit operating by countercurrent
flows in a column or differential centrifuge. However, in order for the
process to be cost-effective, it is usually necessary to recover the
extracting solvent for reuse.
With a single-stage system, the extracting solvent is mixed with the
solution, allowed to settle, and separated. The extracted solute can then be
recovered by stripping from the extractant. The process may be operated as a
batch or continuous technique, although with the latter, different vessels are
required for mixing/settling and decanting.
With a multi-stage system, the extracting solvent and the solution flow
countercurrently through a vertical tower, which may contain internal devices
to influence the flow pattern and provide intimate contact between the
streams. The flow, through the tower may be stage-wise or continuous-contact
type.
Although the solvent extraction process is a separation technique, it can
also be used as a recovery technique. Use of solvent extraction for recovery
{.
involves the following general steps:
1. Extraction - Constituents are transferred from aqueous phase to
organic phase using an organic solvent as an extractant.
5-153
-------
UNTREATED
WASTE STREAM
SOLVENT
EXTRACTION
WATER+
SOLVENT
TREATED
WfcTER
RAFFINATE
SOLVENT & SOLUTE
SOLVENT
SOLVENT
RECOVERY
SOLUTE
REMOVAL
SOLUTE
SOLVENT
MAKE-UP
Figure 5.7.1. Generalized flow diagram of solvent extraction
process.
Source: Reference 5.
5-154
-------
2. Scrubbing - Impurities (e.g., metals) which are co-extracted with
the desired constituents are transferred back to the aqueous phase.
3. Back-Extraction/Stripping - The constituent to be recovered
(e.g., nitric/hydrofluoric acid) is transferred from the organic
phase to a concentrated aqueous phase.
Various methods may be used to accomplish the recovery of acid pickling
liquors including: the Republic Steel process, the AX process (also known as
the Stora process), the Nisshan process, the Solex process, and the Kawasaki
process.
The Republic Steel process was the first application of solvent
extraction to the recovery of corrosive spent sulfuric acid pickling liquor.
Since iron was the main contaminant in the pickling liquor, solvent extraction
was used to remove the iron so that the sulfuric acid could be recovered for
reuse. The process involves oxidation of ferrous ions, complexing ferric ions
with a cyanide complex, extracting the iron complex with a solution consisting
of 25 percent tributylphosphate (TBP) and 75 percent kerosene, and stripping
the iron with ammonia. Both sulfuric acid and ferric oxide (a marketable
product) are recovered. However, this process was never applied on a
commercial-scale.
The AX process was developed in Sweden for the purpose of recovering
spent hydrochloric-nitric acid pickling liquors. A flow diagram of the
process is presented in Figure 5.7.2. Sulfuric acid is initially added to the
spent liquor in order to liberate the nitrate and fluoride from their soluble
metal salts. The spent liquor then flows to the top of a column where it
is extracted countercurrently with a solution of TBP in kerosene. Nitric and
hydrofluoric acids are removed as kerosene-soluble adduct complexes and are
subsequently added to recover the nitric and hydrofluoric acids from the TBP
acid complex. Stripped through addition of water, an activated carbon filter
is used to remove traces of TBP prior to returning the acids to the pickling
tank. The solvent extraction produces a metal sulfate waste stream which
, 3
requires further treatment prior to disposal...
The Nisshan extraction process is similar to the AX process, except that
it uses hydrochloric acid instead of sulfuric acid to convert the metal
nitrates and fluorides to nitric and hydrofluoric acids in the scrubbing
5-155
-------
[
1.
•
*
A
u
fV
TB
IP (in
X.
keroienel
/ \
/ A
Pickling bath
(HNOj.HF. F«.Ni.Cr.Mol
^
5
•
•
^ :
\ !
;
I !
HZO j
•
CaCO3
Filter aid
Wa?te
NaOH
T
H,SO.
T
Me(OH).CaCO,
M
Wash
T
NaOH
Ca(N03),
-c
r
t
l-o
all. 2
ur
ii
l-o
CaMoO4
Figure 5.7.2. Floy diagram of Ax process.
Source: Reference 6.
5-156
-------
step. The use of hydrochloric acid in this step results in a sludge
containing iron, chromate, and nickel, which can be reused as raw material for
A '
steelmaking. Additionally, the Nisshan process makes use of the different
affinities of nitric and hydrofluoric acid for TBP by fractionally recovering
these products in the water stripping step.4
With the Solex process, metal ions are removed prior to extracting the
acids with the organic solvent. The Solex process uses a dialkylphosphoric
acid (e.g., diethyl hexyl phosphoric acid, D2EHPA) in an organic solution
containing a small amount of carboxylic acid and active hydrogen atoms to
•3
extract iron from the spent pickling liquor. The iron is then reduced to
its divalent state by the addition of a sodium chloride solution containing
either sodium sulfate, sodium nitrate, or ammonia. The iron complex is then
stripped from the organic extractant with a strong solution of hydrochloric
acid and the regenerated D.EHPA is recycled.
Following removal of iron,the Solex process proceeds in similar fashion
to the AX and Nisshan processes. HC1 or H-SO, is added to the spent
pickle liquor to convert any remaining metal ions to chlorides or sulfates. A
TBP solution is used to extract the nitrate and fluoride ions after which
nitric and hydrofluoric acids are stripped from the organic extractant using
water. Since the second extraction primarily recovers nitric acid, this waste
is again subjected to TBP extraction/water stripping to recover additional
hydrofluoric acid.
Kawasaki Steel modified the Solex process for application in their Chiba
Works plant in Japan. The Kawasaki process shown in Figure 5.7.3, also
extracts iron prior to extracting the acids. A solution containing 30 percent
D EHPA and 70 percent n-paraffin is used to extract the ferric ions from the
spent acid. The ferric iron is then complexed and stripped from the
extractant using an ammonium fluoride solution. The organic extractant is
recycled to the iron extraction process while the ferric ammonium fluoride
complex is precipitated as crystals. A thermal decomposition process
(discussed in Section 5.8) is used to convert the iron to ferric oxide, a
usable product in the steel process.
After iron removal, hydrochloric acid is added to the waste acid solution
to convert any remaining metal salts to metal chlorides. The waste acid
solution is then subjected to a second extraction with a 70 percent
5-157
-------
l-icklm*
Stainien nevl
hut strip
Acid extraction
,—- S-tirK j^|TBI'i
Crane .\i chi'-ridi-
I .-vaaratiimf-* Ct and Xi __ Kv yirt^Dtnt:
t. | I nver-Mrtr, I
tSHtne-
r»comr> tag*
XII :uiu.-!
[Ao*.irn!iim !^
wairr treammt equipemn
I3t Iran oxia*
-------
TBP, 30 percent n-paraffin solution to remove the nitrate and fluoride ions.
The remaining aqueous solution contains metal chlorides which are further
treated to form metal oxides for reuse in the steel process. The organic
phase is stripped with water to recover nitric and hydrofluoric acids for
return to the pickling bath and TBP/n-paraffin extractant which is recycled to
the second-stage extractor.
Operating Parameters —
Important process parameters include the distribution ratios of nitric
and hydrofluoric acids (i.e., organic phase concentration divided by aqueous
phase concentration), and the stability constants (B-values) of the acids and
metals in the aqueous phase. Both parameters vary with the ionic medium,
ir strength, and the temperature of the system. The distribution ratio
will also be affected by the solution concentration. ^>^ Additional factors
to consider include: the flow ratio of organic phase to aqueous phase,
residence times of the two phases, and number of extraction stages required.
These factors are discussed in detail below in terms of their
interrelationships and effects on extraction efficiency.
During the extractian process, it is desirable to have high distribution
ratios. Tributyl phosphate (TBP), used as the acid extractant, exhibits a
strong affinity for uncharged molecules. The preferential order of acid
extraction using TBP is as follows:
Nitric Acid > Hydrofluoric Acid > Hydrochloric Acid > Sulfuric Acid
The distribution ratios of nitric acid and hydrofluoric acid increase
linearly with TBP concentration with optimal concentration for extraction
being 75 percent TBP in kerosene.6 As shown in Figure 5.7.4, the extraction
of nitric acid is significantly improved by the addition of sulfuric acid.
However, increasing the sulfuric acid additions will increase the
post-treatment (neutralization) costs. The recommended optimum excess
sulfuric acid concentration is 20 percent. Alternatively, post-treatment
costs can be lowered by using hydrochloric acid in this step instead of
sulfuric acid since only stoichiometric amounts of hydrochloric acid are
required for stripping.7 Figure 5.7.5 shows that the distribution ratio for
5-159
-------
1-0 _
as.
PICKLING BATH
CM.-tuiph«.'0.66M
I
3
•nitrate.aq
Figure 5.7.4. Distribution of nitrate between 75Z TBP in kerosene
and water as a function of concentration of total
nitrate in the aqueous phase.
Notes: C * concentration
0 * distribution ratio
Me * metal
aq • aqueous
Source: Reference 6.
5-160
-------
0.25
0.20
0.15
en
o
2
3
§
•33.
I—I
cn
t-l
a
0.10
0.05_
\
\
I I
2 4
INITIAL CONCENTRATION OF
HYDROFLUORIC ACID
T
6
Figure 5.7.5. Distribution of H2S04 between 100% TBP and water
as a function of total added HF concentration.
Notes: C * concentration
D « distribution ratio
aq = aqueous
Source: Reference 6.
5-161
-------
hydrofluoric acid decreases with increas.ing concentration. The data plotted
in Figure 5.7.6 demonstrate that the quantity of sulfuric acid extracted is
dependent on the hydrofluoric acid concentration. Similar results are
observed when hydrochloric acid is used instead of sulfuric acid.
In a mixed solution containing both nitric and hydrofluoric acids, the
extraction efficiency for nitric acid will be better than that for
hydrofluoric acid, although the extent to which this is true is dependent upon
their relative concentrations. As can be seen from Figure 5.7.7, at high
fluoride concentrations, the distribution ratio of fluoride decreases with
increasing nitrate concentration, but at low fluoride concentrations this
effect is minimal or even reversed. In addition, at low nitric and
hydrofluoric acid concentrations, increasing concentrations of sulfuric acid
will increase the distribution ratio of fluoride. Thus, as in the Solex
process, the majority of hydrofluoric acid is extracted after the nitric acid
has been removed.
IXiring the stripping process, it is more advantageous to have lower
distribution ratios to ensure that most of the species is removed with the
aqueous phase* As shown in Figure 5.7.8, nitric acid is readily stripped with
water resulting in nearly complete recovery. • However, low hydrofloric acid
concentrations results in a high distribution ratio and, therefore, reduces
the effectiveness of stripping. However, the stripping efficiency for
hydrofluoric acid can be increased by using an additional mixer-settler
containing nitric acid, without sulfuric acid. '
The distribution ratios decrease with increasing solution
temperature. Therefore, more effective recovery: is achieved if the
solution is extracted at a lower temperature (approximately 10°C) and stripped
at a higher temperature (above 60°C), but this would require additional
equipment and energy costs for heating and cooling the feed.
• i
The organic flow in both extraction and stripping stages will be the
same. The water flow used during the stripping stage should not be greater
than 1.5 times the pickle liquor flow. Since a high distribution ratio is
good for extraction, but not for stripping, it is ^important to optimize the
' I
flow ratios according to the effects of concentration changes on distribution
ratios. For a unit capable of treating 70,000 tons/year of pickle liquor,
the optimum organic:aqueous phase ratios were calculated to be 2.3 for
extraction and 1.6 for stripping.
5-162
-------
2.5_
DHF
2.0
.1.5.
1.0
0.5
= 4M
2.0
[«<=]
aq
T~
3.0
Figure 5.7.6. Distribution of HF between 1002 TBP and water
as a function of added concentration of HF
and II2S04.
Notes: C * concentration
D = distribution ratio
aq = aqueous
Source: Reference 6.
5-163
-------
2.5 _
Dnitrate
^fluoride
2.C
PICKLING BATH
Dniiri
1.5
1.0
0.5
F)ooride.aq
=2
CFioor,de.aQ =0.6M
CRuoride.aq = 0.1 M
1.0
2.0
^nitrate,aq
Figure 5.7.7.
Notes:
C
D
Me
aq
Extraction of HN03 and HF by 75% TBP in kerosene
from aqueous solutions 0.33 M in Me2 (S04J3
as a function of total aqueous nitrate and fluoride
concentration.
concentration
distribution ratio
metal
aqueous
Source: Reference 6,
5-164
-------
Figure 5.7.8. The distribution of HNO3 and HF between 75% TBP in
kerosene and water as a function of aqueous concentra-
tion of HN03 and HF.
Notes: C * concentration
D «• distribution ratio
aq = aqueous
Source: Reference 6
5-165
-------
P re-Tre a tmen t—
Removing iron prior to extracting nitric and hydrofluoric acids, as in
the Solex and Kawasaki processes, permits greater recovery of hydrofluoric
acid since the iron complex has a greater distribution ratio than hydrofluoric
acid in the IBP extractant. The order of extraction is as follows:
HN03 > HFeCl4 > HF > HC1
As shown in Figures 5.7.9 and 5.7.10, the extraction efficiencies are
adversely affected to a significant degree by the presence of iron in the
solution.
Post-Treatment—
The solvent extraction process will generate a metal sludge that will
require further handling. The sludge can either be neutralized and disposed,
or recovered for reuse using thermal decomposition techniques as in the
Kawasaki process. Metals recovery is facilitated if hydrochloric acid is used
*
instead of sulfuric acid in the scrubbing step.
. The spent scrubbing acid used to free the nitrate-and fluoride ions will
also require neutralization. The use of hydrochloric acid (Kawasaki process)
instead of sulfuric acid (AX process) to free the nitrate and fluoride reduces
sludge generation because a caustic soda neutralization can be used instead of
lime.
Extracting solutions can be recovered and reused in the solvent
extraction process. The organic solvent exiting the stripping stage will
contain: small concentrations of acids (mostly HF), calcium, and molybdenum.
It can be recovered by washing with a 1-M solution of sodium hydroxide in a
i 6
mixer-settler prior to reuse in the extractor. The wash solution can be
i
treated'by adding calcium nitrate to precipitate calcium fluoride, calcium
molybdaCe, and a sodium nitrate filtrate which is recycled to the spent
pickling; bath feed to recover the extractable nitrate.
5-166
-------
HCl^re * 50-1000
HCl 8-65.994
After Fe
Extraction
Before Fe
Extraction
1 2
Aqueous Phase (HNOs)
Figure 5.7.9.
Extraction of nitric acid with 75% TBP in
kerosene as a function of iron content.
Source: Reference 7.
5-167
-------
2JQ
1.0
j
2 3
•
After Fe
Extraction
Before Fe
Extraction
HCI
1 2
Aqueous Phase CHNOO
Figure 5.7.10. Extraction of hydrofluoric acid with 75Z TBP in
kerosene as a function of iron content.
Source: Reference 7.
5-168
-------
5.7.2 Process Performance
Although research on the application of solvent extraction to acid
recovery was performed in 1962 by the Republic Steel Company, the first
commercial-scale application of the process was demonstrated at Stora
Kopparbergs Bergslags in Sweden in 1973.4 The AX process was developed
utilizing countercurrent multiple-stage extractions in a pulsed column mode
and a multi-stage mixer-settler arrangement. Operating and design
parameters for these two systems are presented in Tables 5.7.1 and 5.7.2,
respectively. Results of tests conducted in each of these operational modes
are summarized in Table 5.7.3. The results were successful in that they
demonstrated the potential for recovering nitric and hydrofluoric acids while
maintaining a continuous pickling process. High sludge generation, caused by
the addition of sulfuric acid (which forms metal sulfates) to free the nitrate
and fluoride ions for organic extraction, reduces the cost-effectiveness of
this solvent extraction technique.
Nisshan Steel Co., Ltd. in Japan made modifications to the AX process and
conducted commercial-scale testing of the modified process at Nisshan Steel,
4
Shunan Works, in Japan. Hydrochloric acid was added to the spent
nitric-hydrofluoric acid to free the nitrate and fluoride ions. The nitrate
and fluoride ions'were then extracted with an organic solvent consisting of
75 percent TBP and 25 percent aromatic hydrocarbons. Nitric acid was used to
strip the acids from the organic phase into the aqueous phase. The Nisshan
process was able to recover 94 percent of the nitric acid and 16 percent of
the hydrofluoric acid. The lower hydrofluoric acid recovery was attributed
4
to the fact that ferric ions form complex ions with fluoride ions.
Further research led to the development of the Solex process which was
4
modified for commercial application by Kawasaki Steel Corp. in Japan. The
typical composition of the spent nitric-hydrofluoric acid pickling liquor at
this point is presented in Table 5.7.4, As described in Section 5.7.1, the
process consisted of four stages: iron extraction, iron oxide formation, a
second extraction and back-extraction (stripping) to recover
nitric-hydrofluoric acid, and ferrite formation. The operating and design
parameters for the recovery system are presented in Table 5.7.5. As shown,
greater acid recoveries are achieved with the Kawasaki process. In addition,
iron is also recovered.
5-169
-------
TABLE 5.7.1. OPERATING AND DESIGN PARAMETERS FOR THE
MULTI-STAGE PULSED MODE COLUMN
Item
Parameter
Treatment capacity
Pulse amplitude
Frequency amplitude
Number of stages
Extraction
Stripping
Extraction column dimensions
Height
Diameter
Stripping column dimensions
Height
Diameter
600 L/hr
0 to 40 mm
20 to 120 cycles/min
10
5
9 m
25 cm
3 m
90 am
Source: Reference 6.
5-170
-------
TABLE 5.7.2. OPERATING AND DESIGN PARAMETERS FOR THE
MULTI-STAGE MIXER-SETTLERS
Item Parameter
Treatment capacity 600 L/hr
Number of stages
Extraction 5
Stripping 10
Stripping column dimensions
Height 8 m
Diameter 35 m
Source: Reference 6.
5-171
-------
TABLE 5.7.3. RESULTS OF TESTS CONDUCTED-AT STORA IN SWEDEN
USING THE AX PROCESS3
Extraction
Parameter
Flow rate
Aqueous (mL/m)
Organic (mL/m)
Organic : aqueous
Pickling bath concen-
tration (aqueous feed)
HN03 (M)
HF (M)
Iron (g/L)
Chromium (g/L)
Nickel (g/L)
Molybdenum (g/L)
Concentration of exit-
ing organic stream
HNQ3 (M)
HF (M)
Iron (g/L)
Chromium (g/L)
Nickel (g/L)
Molybdenum (g/L)
Concentration of exit-
ing aqueous stream
HK>3 (M)
HF (M)
Iron (g/L)
Chromium (g/L)
Nickel (g/L)
Molybdenum (g/L)
Percent extraction
Nitric acid
Hydrofluoric acid
Pulsed
column
Test 1
985
2,860
2.9
1.68
1.68
33
5
8
0.7
0.-54
0.41
- '
-
-
0.36
0.02
0.47
33
5
8
0.01
99
72
Pulsed
column
Test 2
1,070
2,530
2.4
2.25
1.87
44
7.2
11.5
1.13
0.91
0.35
-
-
-
0.48
0.01
1.05
-
-
-
0.02
99
44
Mixer-
settler
49
127
2.6
2.25
1.87
44
7.2
11.5
1.13
1.07
0.42
—
-
-
—
0.01
0.84
-
-
-
-
99
55
Stripping
Pulsed
column
Test 1
915
1,500
1.6
0.54
0.41
-
-
-
0..36
0.04
0.17
-
-
—
0.17
0.92
0.42
-
—
-
0.26
93
59
Pulsed
column
Test 2
1,000
1,789
1.8
0.91
0.35
-
-
-
0.48
0. 11
. 0.09
—
—
—
0.25
1.38
0.38
_
—
-
0.39
88
74
Mixer-
settler
53
102
1.9
1.10
0.39
-
-
-
—
. 0.23
0.16
_
—
—
-
1.39
0.39
_
_
-
-
79
59
* Not analyzed.
Source: Reference 6.
5-172
-------
TABLE 5.7.4. TYPICAL COMPOSITION OF SPENT
PICKLING LIQUOR AT KAWASAKI
STEEL CHIBA WORKS IN JAPAN
Constituent Concentration (g/L)
Nitric acid 180 to 200
Hydrofluoric acid 40 to 45
Iron 28 to 30 (35 maximum)
Chromium 10 to 15
Nickel 5 to 10
Source: Reference 4.
5-173
-------
TABLE 5.7.5. TYPICAL OPERATING PARAMETERS AND RESULTS
FOR KAWASAKI PROCESS
Parameter
Iron separation Nitric-Hydrofluoric
stage acid recovery stage
1.0
(24 n^/day)
Waste liquor constituents Waste acid
Treatment capacity
Extraction or recovery
percentage
Nitric acid 98.5
Hydrofluoric acid 75.3
Iron 95
Iron-free acid
-0.3 m3/hr
(7.2 m3/day)
95.7
79.5
Source: Reference 4.
5-174
-------
5.7.3 Process Costs
It is difficult to evaluate the economics of solvent extraction due to
the limited number of facilities using the process. Currently, there are no
commercial-scale solvent extraction systems being used in the United States.
Capital costs would include the equipment for the solvent extraction
process and the thermal decomposition equipment (see Section 5.9) if ferric
oxide recovery is judged to be cost effective. Operating costs would include
the costs for solvent (TBP, kerosene, D2EHPA), sulfuric or hydrochloric acid
used in scrubbing, labor, and maintenance. Solvents and acids can be recycleo
to reduce acid purchase requirements, reduce neutralization and disposal
costs, and generate savings due to metal oxide recovery.
5.7.4 Process Status
Solvent extraction has been demonstrated to be effective in the recovery
of spent nitric-hydrofluoric acid pickling liquors generated by the steel
industry. Commercial-scale systems have been tested and are in use in both
Sweden and Japan. However, no commercial-scale solvent extraction systems
have yet been employed in the United States.
•
Of the four solvent extraction processes developed for application to
acid wastes, the Kawasaki (or Solex) process has shown the most promising
results. Commercial-scale testing of the Kawasaki process has demonstrated
95 percent recovery of nitric acid, and 70 percent recovery of hydrofluoric
A i
acid. In addition, 95 percent of the iron was recovered for reuse.
Kawasaki intends to eventually market the process. '
Although the process has been demonstrated to be technically feasible
with good recovery results, limited data are available to assess its economic
viability. However, with further demonstrations this process could prove to
be an effective alternative to conventional neutralization and disposal
practices.
5-175
-------
REFERENCES
1. Clevenger, I.E., and J.T. Novak. Recovery of Metals from Electroplating
Wastes Using Liquid-Liquid Extraction. Journal of the Water Pollution
Control Federation, 55(7): 984-989. July 1983.
2. McDonald, C.W. Removal of Toxic Metals from Metal Finishing Wastewater
by Solvent Extraction. Prepared for the U.S. EPA, Industrial
Environmental Research Laboratory, Cincinnati, Ohio. EPA-600/2-78-011.
February 1978.
3. Stephens on, J.B., Hogan, J.C., and R.S. Kaplan. Recycling and Metal
Recovery Technology for Stainless Steel Pickling Liquors. Environmental
Progress, 3(1): 50-53. February 1984.
4. Watanabe, T., Hoshimo, M., Uchino, K., and Y. Nakazato. A New Acid and
Iron Recovery Process in Stainless Steel Annealing and Pickling Line.
Kawasaki Steel Technical Report No. 14, pp. 72-82. March 1986.
5. U.S. EPA. Trestability Manual, Volume III: Technologies for Control/
Removal of Pollutants. EPA-600/8-80-042c. July 1980.
6. Rydberg, J., Reinhardt, H., Lunden-, B., and P. Haglund. Recovery of
Metals and Acids from Stainless Steel Pickling Bath. In: Proceedings of
the 2nd Annual International Symposium on Hydro-Metal, Chicago,
Illinois. February 25 throuth March 1, 1973.
7. Watanabe, A., and S. Nishimura. Process for Treating an Acid Waste
Liquid. Assigned to Solex Research Corporation, Osaka, Japan. U.S.
Patent No. 4,166,098. August 28, 1979.
8. Japan Metal Bulletin. Kawasaki Recovers Iron and Acids from Pickling
Line. No. 4530. May 19, 1984.
5-176
-------
5.8 THERMAL DECOMPOSITION
5.8.1 Process Description
Thermal decomposition is a capital intensive regeneration process which
can be used to recover acid wastes, including sulfuric and hydrochloric acid
pickling liquors and sulfuric acid wastes generated by the organics industry.
Currently, the most significant area of application for this technology is the
recovery of hydrochloric acid from spent pickling liquors generated by the
steel industry. By using thermal decomposition processes, both free and bound
acids can be recovered.
The hydrochloric acid regenerating process involves precipitating hydrous
ferrous chloride and decomposing it in a roaster or furnace to form iron oxide
and hydrochloric acid.1 A general flow diagram of a HC1 regeneration
process is presented in Figure 5.8.1. The spent pickle liquor is
preconcentrated in a venturi scrubber by heat exchange with the hot gases
emerging from the reactor. The preconcentration step removes excess water
from the spent pickling liquor which reduces the energy requirements for the •
subsequent roasting step.
Following preconcentration, the acid stream is fed to the roaster. Free
*
water and hydrochloric acid evaporate in the upper regions of the roaster
while hydrolysis takes place in the lower regions, resulting in the formation
of iron oxide and hydrochloric acid. The iron oxide is discharged as a
free-flowing powder from the base of the reactor. At the high reactor
temperatures, hydrochloric acid vaporizes and exits through the top of the
reactor with the combustion gases.
Hot gases exiting the reactor will contain hydrochloric acid gas,
superheated water vapor, and inert combustion products. These gases pass
through the venturi scrubber where a direct heat exchange occurs between the
hot gases and the spent pickle liquor, such that preconcentrating the liquor
and cooling the hot gases to 100°C occurs simultaneously. The cooled gases
exit the venturi scrubber and are then directed under negative pressure to an
absorber. This is charged with make-up water or water from the fume scrubber
to recover hydrochloric acid at an approximate concentration of 20 percent.
5-177
-------
GfOfiftE
RIVE STOCK
ABSORBS! SCRUB86* SERftRATOR
Figure 5.8.1 General flow diagram of UC1 regeneration by
thermal decomposition.
Source: Reference 2.
5-178
-------
The remaining vapors from the absorber are discharged to the fume scrubber and
exhausted to the atmosphere.3 Several types of reactors may be used for the
roasting step, including the following:
• Fluidized-Bed Reactor - Concentrated pickle liquor is evaporated in
a fluidized-bed of granular iron oxide.Iron oxide product will
adhere to the fluidized grains, and must be discharged at a
continuous rate to maintain desired bed volume.
• Spray Roaster - Concentrated pickle liquor is atomized and
evaporated at high temperatures in the reactor.
• Sliding Bed Reactor - Concentrated pickle liquor is sprayed onto a
continuously circulating bed of hot iron oxide. Iron oxide is
recirculated through the system using a surge bin and bucket
elevator.
A spray roasting thermal decomposition process has recently been
developed for the regeneration of waste sulfuric acid from the production of
4
titanium dioxide pigments. In this case, the spent solution contains iron
and other metal sulfates. As with the HC1 regeneration process, free sulfuric
acid is vaporized and exits with the combustion gases to a condensor/absorber,
where the free sulfuric acid can be recovered. However, unlike HC1
• .
regeneration, the initial roasting reaction does not regenerate the sulfuric
acid directly; i.e., through hydrolysis of iron chloride. Instead, ferrous
and other metal sulfates are oxidized to sulfur dioxide and an anhydrous metal
salt precipitate. The metal sulfate salts can be further roasted to convert
nearly all the sulfates to oxides and sulfur dioxide. The sulfur dioxide
resulting from these roasting reactions is converted to sulfuric acid via
5,6
conventional vanadium—catalyzed processes.
Currently, research is underway for a thermal decomposition system which
is capable of recovering sulfuric acid from wastes which contain a much
broader range of constituents, such as those generated by organics
industries. Waste acids generated in the manufacture of herbicides, dyes,
and bactericides may contain sulphonated polynuclear aroraatics, chlorinated
hydrocarbons, and nitrates. The conventional thermal decomposition techniques
mentioned previously cannot be used for these wastes because of the hazardous
air emissions that would be generated. Waste acids and acid tars with metal
5-179
-------
impurities cannot be regenerated with conventional thermal decomposition
systems because of slag formation on the' refractory wall lining which impedes
the performance.
A process developed by the German Ministry of Research and Technology
(BMFT) uses a rotating furnace, an intermediate chamber, two additional
combustion chambers, and a waste heat boiler to recover organic contaminated
sulfuric acid streams. Figure 5.8.2 presents a flow diagram of the
process. Waste acid is fed with combustion air and sulfur to the rotating
furnace which contains a coke bed, that serves as a heat reservoir and
collects ashes. The sulfur, hydrocarbons, and hydrogen are oxidized in the
furnace • Partial coking of the incoming hydrocarbons occurs which maintains
a continuous bed of coke, and also reduces the nitrates and sulfur trioxide in
the decomposition gas. The decomposition gas then flows to the intermediate
chamber, where additional air is mixed with the decomposition gases before
directing the flow to the secondary combustion chambers. In these units,
liquefaction of the ash particles in the gas stream occurs, forming a slag bed
which causes a 1 to 2 percent excess of oxygen in the decomposition gas
mixture and results in a temperature drop. In subsequent steps, the
decomposition gas temperature is further reduced in a heat exchanger, the gas
is washed, and sulfur dioxide is condensed and recovered. The sulfur dioxide
is then converted to sulfuric acid via vanadium-catalyzed processes. '
Operating Parameters—
Operating parameters which will affect the efficiency of thermal recovery
systems include: heat requirements, process water requirements, and volume of
material processed. These parameters also determine the size of the system
required for regeneration, which in turn will affect the capital and operating
costs. Typical operating parameters for HC1 regeneration systems are
presented in Table 5.8.1.
Indirect heating of the pickling tank, using either in-tank heaters or
external circulation heaters, reduces the overall volume of waste acid to be
treated by approximately 30 percent. By reducing the volume of waste acid
to be processed, a smaller size regeneration system can be employed with fewer
capital costs.
5-180
-------
•OT3
w*^
II
secondary air
rotating
furnace
^
^J
J1"
r m
CK
1
medial*
coke
waste heal boiler
•team
Ncombuitlon chamber 2
combutHon d^amber 1
Figure 5.8.2. Flow diagram of a thermal decomposition process
using a rotating furnace.
Source: Reference 7.
5-181
-------
TABLE 5.8.1. TYPICAL OPERATING PARAMETERS FOR THERMAL DECOMPOSITION
ACID REGENERATION SYSTEM
Parameter
Result
Reactor temperature
Temperature of gases exiting reactor
Temperature of gases exiting
. venturi scrubber
Average cycle time
Specific heat consumption
Specific energy consumption
800 to 100°C
~400°C
100°C
•"10 hours
5,000 kcal/kg Fe203
0.3 to 0.4 kwh/kg Fe203
Source: References 1, 2, 3, 6, 8, 9, and 10.
5-182
-------
In order Co reduce the volume of rinse, water required for the process, a
cascade rinsing system can be employed. - Good rinsing can be achieved with
7 to 10 gallons of rinse water per ton of steel pickled by using four or five
stages of cascade rinsing.10 By reducing the volume of rinse water
generated, treatment costs are reduced.
5.8.2 Process Performance
The performance of a thermal decomposition system can be evaluated on the
basis of percent acid recovered, the purity (concentration) of recovered acid
and oxide products, processing time, and by-product generation.
Several companies currently offer thermal decomposition systems for the
regeneration of hydrochloric acid from spent pickling liquors. The most
commonly used reactor is the spray roaster. Typical results for these system
are presented in Table 5.8.2. Unlike other recovery processes which are
limited to recovery of free acid, thermal decomposition processes are able to
recover more than 99 percent of the total HC1 equivalent in the spent pickling
waste. 10 In addition, a high quality iron oxide by-product is generated,
which can be reused within the plant or marketed for use in ferrite magnets,
pigments, molding sands, glass and other industries. Table 5.8.3 lists
the typical composition and purity of the iroti oxide by-product formed.
The sulfuric acid thermal regeneration process for titanium dioxide
production waste has not been tested at the commerical-scale level. However,
pilot tests have demonstrated the potential to recover 93 to 96 percent of the
4
sulfuric acid equivalent in the spent pickling waste. The sulfuric acid
regeneration process will generate a waste stream of unreacted sulfates which
will require neutralization. However, this amount will be significantly less
than the amount that is required if no recovery process is employed. For
example, a regeneration system for a 50,000 metric-ton/year titanium dioxide
production plant will generate 19,600 to 46,000 metric-tons/year of waste
requiring neutralization as compared to the 100,000 to 200,000
metric-tons/year of waste requiring neutralization without the regeneration
4
process.
5-183
-------
TABLE 5.8.2. TYPICAL PERFORMANCE OF A THERMAL DECOMPOSITION SYSTEM
FOR REGENERATION OF HYDROCHLORIC ACID
Parameter
Result
Range of available regeneration capacities
/
Composition of spent pickling liquor
Hydrochloric acid
Ferric chloride
Composition of regenerated acid
Hydrochloric acid
Iron
Regeneration efficiency
(percent of HC1 equivalents
recovered from waste)
Purity of iron oxide by-product
5 to 75 gpm
0.5 to 5.02
20 to 25Z
20Z
0.25Z
>99Z
98.5 to 99.4Z
Source: Reference 10.
5-184
-------
TABLE 5.8.3. TYPICAL COMPOSITION OF IRON OXIDE BY-PRODUCT GENERATED
BY THERMAL DECOMPOSITION PROCESS
Constituent
Percent composition
Ferrous oxide
Aluminum oxide
Manganese oxide
Chromium oxide
Silicon oxide
Calcium oxide
Magnesium oxide
Sodium oxide
Potassium oxide
Nickel oxide
Copper oxide
Chloride
Sulfur trioxide
Water solubles
Ignition loss
•—^B————•«
Source: Reference 10.
98.5 to 99.4
0.05 to 0.12
0.35 to 0.45
0.02 to 0.06
0.01 to 0.06
0.01 to 0.06
0.01 to 0.04
0.01 to 0.06
0.01 to 0.03
0.01 to 0.05
0.01 to 0.05
0.05 to 0.20
0.02 to 0.04
0.10 to 0.70
0.25 to 1.00
J-185
-------
The rotary furnace thermal decomposition technique has also only been
demonstrated at the pilot-scale. Unlike the previously mentioned
decomposition systems, the rotary furnace is designed to handle waste acids
with a variety of constituents. The German Ministry of Research and
Technology (BFMT) tested 18 different waste sulfuric acids with varying
concentrations, composition, and consistency. Table 5.8.4 lists the
approximate compositions of these waste feeds. The performance evaluation
results are summarized in Table 5.8.5. In all cases, it was found that sulfur
dioxide could be obtained of sufficient quality to permit use in sulfuric acid
production. Performance was not affected by changing waste viscosity, carbon
content, or chemical forms of sulfur.
5.8.3 Process Costs
The economics of using thermal decomposition systems to regenerate waste
hydrochloric acids from spent pickling liquors can be evaluated on the basis
2
of the following factors:
• Cost of recovered acid if purchased;
• Cost of treatment and disposal of waste acid by other technologies;
• Pickling waste quantity (which determines regeneration system size
requirements and costs); and
• Quality and market value of "the by-product iron oxide.
These costs will vary with the particular application, but in all cases the
capital costs will be high. Typically, capital costs for an installed system
can range from Jl to 7 million for regeneration systems with capacities
ranging from 5 to 75 gpm. ' ' In some cases, the value of the recovered
hydrochloric acid may be less than the costs for regeneration. However, with
increasing disposal costs, and a developing market for the iron oxide
byproduct, this situation could be reversed.
Table 5.8.6 presents an economic evaluation of a small, medium, and large
HC1 regeneration system. As can be seen from the table, thermal decomposition
is most cost-effective for plants generating large quantities of spent acid.
5-186 -
-------
TABLE 5.8.4. COMPOSITION OF WASTES FED TO THE ROTARY FURNACE
DURING TESTING CONDUCTED BY THE BMFT
Approximate percent composition
Waste acid Sulfuric
type no. acid
1 60
2 76
3 41-48
4 50-55
5 43
6 35
7 66-70
8 90
9 50-53
10 65
11 26
12 50
Organic Difficult
carbon Water Ash Chlorine impurities
2 30 Ammonia
3-4 17.5. 0.05 Brominated
polynuclear
aromatic s
0.1-1.0 51-57 0.06 0.02 Aliphatics,
Polynuclear
aroma tics
0.1-2.0 40-45 0.06 Ammonia
0.1-0.4 54 0.005 0.6 • Aliphatics,
Chlorinated
aroma tics
0.4 56 0.8 0.07 Ammonia
15-22 10-14 0.1 Aromatics .
( alkylbenzenes)
4-5 4-6 0.01 Aromatics
1.2-5.3 42-45 0.05 Aliphatics,
Aromatics,
Ketones
2 32 0.05 0.1 Aromatic
sulfonic acids
0.2 73 0.07 0.02 Sulfonated
aromatic s,
Phosphates
O.i 50 0.1 20 ppm Chlorinated
aromatic s
(continued)
5-187
-------
TABLE 5.8.4 (continued)
Waste acid
type no.
13
14
15
16
17
18
Approximate percent composition
Sulfuric Organic Difficult
acid carbon Water Ash Chlorine impurities
75-59 1.0-3.6 19-21 0.4-4.4 Aliphatics,
Aroma tics
90-92 0.2-0.5 7.5-9.5 0.02 Aromatics,
Heterocycles
77 5.3 18 0.002 Sulfonated
branched
unsaturated
. alkanes
68 19 10 0.05 Sulfonated
branched
unsaturated
alkanes
30 0.3 69 0.01 0.02 Nitric acid,
nitrated
aromatic s,
Ke tones
30-35 30 30-33 6 Chlorinated
aromatic s,
Alkalis,
Alkaline earth
Source: Reference 7.
5-188
-------
TABLE 5.8.5. SUMMARY OF ANALYTICAL RESULTS OF THE ROTARY FURNACE
PILOT TESTS CONDUCTED BY THE BMFT
Parameter Result
Throughput Rates
Waste sulfuric acid 1,000 to 2,300 kg/hr
Acid tar 700 to 1,050 kg/hr
Sulfur 800 to 1,200 kg/hr
Average Composition of
Decomposition of Gas
Oxygen 0.8 to 2.0 2-vol.
Sulfur dioxide 10.0 to 15.0Z-vol.
Carbon dioxide 6.0 to lO.OZ-vol.
Composition of Processed
Decomposition Gas (mg/m-*)
Sulfur trioxide 220 to 640*
Elemental sulfur 2 to 45
Ammonia 0 to 20
NO* 0 to 4
N02 ND
Hydrogen sulfide ND
Hydrocarbons ND
Average sulfuri'c acid output rate 170 tons/day
*The 803 can be converted to sulfuric acid via vanadium catalyzed reactions.
Source: Reference 7.
5-189
-------
TABLE 5.8.6. ECONOMIC EVALUATION OF HYDROCHLORIC ACID REGENERATION
USING THERMAL DECOMPOSITION
Item
Capital Costs
Operating Costs
Labor
Maintenance
Fuel
Electricity
Water
Cost Savings
Acid value
Iron oxide value
Treatment and
disposal costs
Net Annual Savings
Payback Period
(years)
Cost basis
TIC* $3
1.8Z of TIC
3% of TIC
12,000 Btu/gal waste
0.10 kwh/gal waste
1 gal/gal waste
Total
50Z of PMV**
$100/ton
Caustic soda
Total
- operating
Capital+Net savings
10,000
,907,000
69,000
120,000
59,000
11,000
6,000
$265,000
457,000
187,500
12,000
$656,500
$391,500
10.0
System size
100,000
$14,974,000
266,000
460,000
225,000
40,000
24,000
$1,015,000
4,570,000
1,875,000
120,000
$6,565,000
$5,550,000
2.
(gpd)
200,000
$23,487,000
427,000
720,000
363,000
65,000
39,000
$1,614,000
9,133,000
3,750,000
240,000
$13,123,000
$11,509,000
7 2.0
* TIC * Total installed cost
**PMV » Present market value
Source: References 9, 10, and 13. Costs updated to July 1986 dollars.
3-190
-------
Costs for the two sulfuric acid regeneration systems can not be
realistically determined at this time, since both processes have not been
developed beyond the pilot-scale level. Cost savings would be realized by the
recovered acids and the reduced disposal requirements. Significant quantities
of these waste acids are currently neutralized and disposed at high costs.
For example, a typical 50,000 metric ton/year pigment plant using lime or
limestone neutralization can produce up to 360,000 metric tons/year of gypsum,
while consuming over 130,000 tons of limestone.4 Significant savings in
materials would be realized if these acids could be recovered. However,
capital costs for these systems are expected to be the same or slightly higher
than the costs for HC1 regeneration systems. In addition, according to
current market prices, the value of the recovered sulfuric acid would be less
than the value of the same quality of hydrochloric acid. Therefore, a
sulfuric acid regeneration system would not be cost-effective except for large
quantity generators or waste disposal facilities.
5.8.4 Process Status
Thermal decomposition is an effective but capital intensive process for
the recovery of- hydrochloric acid from spent pickling liquors. It is a
demonstrated technology and is currently being used by several steel
manufacturers. There are currently no commercial-scale applications of
thermal decomposition for wastes other than spent hydrochloric acid pickling
liquors. However, research has demonstrated the technical feasibility of
using thermal decomposition to regenerate waste sulfuric acid effluents from
spent sulfuric acid pickling liquors and from organic industry waste acids.
The thermal decomposition process has the advantage of being able to
recover bound as well as free acid from waste, which distinguishes it from the
previously mentioned recovery technologies. More than 99 percent of the
hydrochloric acid equivalents in waste pickle liquor can be regenerated, and
an estimated 93 to 96 percent of sulfuric acid equivalents can potentially be
4,11,12
regenerated by thermal decomposition.
However, capital costs for thermal decomposition will be prohibitive for
small volume generators. Although the total quantity of waste sulfuric acid
5-191
-------
generated by the steel industry is greater than the amount of hydrochloric
acid generated, the latter is generated by a small number of large quantity
generators. Combined with the higher purchase price of hydrochloric acid, UC1
regeneration systems may be more implementable than sulfuric acid regeneration
systems in the steel industry. However, large quantities of waste sulfuric
acid are generated by individual organic chemical manufacturing plants, and
therefore acid regeneration may have a wider application for this industry.
For large quantity generators, the use of thermal decomposition will
realize a net savings in reduced acid purchase, waste treatment, and disposal
costs. In addition, the burden to the environment will also be significantly
reduced. With increasing costs for disposal, and increasing development of
the technology for other waste types, thermal regeneration systems are likely
to find wider application in corrosive waste treatment.
5=492
-------
REFERENCES
1. Ruthner, Michael, and Othmar Ruthner. Twenty-five (25) Years of Process
Development in HC1 Pickling and Acid Regeneration. Iron and Steel
Engineer. November 1979.
2. Wadhawan, Satish C. Economics of Acid Regeneration - Present and
Future. Iron and Steel Engineer.* October 1978.
3. Bierbach, Herbert, and Klaus Hohmann. Continuous Regeneration of
Hydrochloric Acid Pickle Liquors and Other Metal Chlorides According to
the Lurgi Process. Wire World International, 15(5):161-163.
September/October 1973.
4. Smith, Ian, Gordon M. Cameron, and Howard C. Peterson. Acid Recovery
Cuts Waste Output. Chemical Engineering. February 3, 1986.
5. Franklin Associates. Industrial Resource Practices: Petroleum
Refineries and Related Industries. Prepared for U.S. EPA, Office of
Solid Waste, Washington, DC, under EPA Contract No. 68-01-6000 (1982B).
1982. .
6. Versar, Inc. National Profiles Report for Recycling - A Preliminary
Assessment. Draft Report Prepared for the U.S. EPA, Waste Treatment
Branch, under EPA Contract No. 68-01-7053 (Work Assignment No. 17). July
8, 1985.
7. Driemal, Klaus, Norbert Lowicki, and Joachim Wolf. Harmless Disposal of
Waste Sulfuric Acids Containing Polynuclear Sulphonated Aroma tics In A
Rotating Furnace with Special Further Combustion Stages. Project of the
German Ministry of Research and Technology (BMFT). In: Recycling
International, 4th edition, pp. 1078-1083. Karl J. Thome-Kozmiensky,
editor. Published by EF-Verlag fur Energie- and Umwelttechnik.
ISBN 3-924511-05-5. 1984.
8. Elliot, A.C. Regeneration of Steelworks Hydrochloric Acid Pickle Liquor.
Effluent and Water Treatment Journal. July 1970.
9. Camp, Dresser, and McKee, Inc. (COM). Technical Assessment of Treatment
Alternatives for Wastes Containing Corrosives. Prepared for the U.S.
EPA, Office of Solid Waste, under EPA Contract No. 68-01-6403 (Work
Assignment No. 39). September 1984.
5-193
-------
10. Perox, Inc. Brochure: Hydrochloric Acid Regeneration and Iron Oxide
Production. Received August 1986.
11. Wadhawan, Satish C. Ferox, Inc., Pittsburgh, Pennsylvania. Telephone
Conversation with Jon Spieloan, GCA Technology, Inc. August 6, 1986.
12. Wadhawan, Satish C. Perox, Inc., Pittsburgh, Pennsylvania. Letter to
Jon Spielman, GCA Technology, Inc. August 7, 1986.
13. Chemical Monitoring Reporter, 230(3):32-40. July 21, 1986.
5-194
-------
5.9 WASTE EXCHANGE
5.9.1 Description
Waste exchange involves the transfer of an unwanted waste material to a
company which is capable of using it in its industrial process. Corrosive
wastes can be exchanged and reused for neutralization (see Section 4.1), pH
adjustments, and cleaning solutions. Corrosive wastes which are most commonly
exchanged are listed in Table 5.9.1.
Waste exchanges are typically initiated by a third party, who uses either
passive or active means to effect the transfer. The passive approach occurs
through the use of an information clearinghouse, which is typically sponsored
by governmental agencies (e.g., regional chambers of commerce) or industry
associations. Active waste transfers are handled by materials exchange
services which actually buy, treat, process and/or store wastes, and solicit
1 2
potential users for them. '
The clearinghouse approach is the most prevalent waste exchange
technique. Clearinghouses generally issue a catalog in which generators list
the wastes they wish to transfer, and potential users- list the wastes they
desire. Information typically contained in the catalogue includes: waste
description, quantity generated, availability, and general location of the
f%
waste. Waste materials are assigned a code number so that the identity of
the listing company can be kept confidential. An interested user will send a
letter of inquiry to the clearinghouse, who will then forward the letter to
the generator. The actual waste transfer will be handled by the generator and
the user.
Clearinghouses exchange information only. They do not actively seek out
customers, negotiate transfers, set values, process materials, transport
materials, or provide legal advice; these functions are left to the
participating waste generator and waste user. ' In order to be successful,
clearinghouses must be responsive to industry needs. They typically
provide:
• Support of plant managers and engineers faced with disposal problems;
«. Industry endorsement and assistance; and
• Confidentiality of industry identities and waste generation data.
5-195
-------
TABLE 5.9.1. LIST OF MOST COMMONLY EXCHANGED
CORROSIVE WASTES
Acids
Alkalis
Hydrochloric acid
Hydrofluoric acid3
Nitric acida
Phosphoric acida
Sulfuric acid8
Picric acid3
Pickle liquors (FeCl2 or FeS04>a
Sodium hydroxide
Sodium carbonate
Acetylene sludge
3ln concentrations exceeding 15 percent.
Source: References 1 and 2.
-5-196
-------
A materials exchange takes a more active role in the waste transfer hv
actually identifying a potential match and assisting in consummating a contract
between the waste generator and the waste user. The material exchange acts
•i
as an agent or broker, seeking a buyer or seller for the waste material.
In some cases, the materials exchange will actually take possession of the
wastes, charge a fee for handling the transaction, process the waste (to make
o
it marketable), and sell the waste to a potential user. In other cases, a
materials exchange will use various information sources to match generators
with potential users. Typical sources of information include: waste exchange
catalogs, personal contacts with industry, trade associations, trade journals,
and referrals from other industries and government agencies. The materials
exchange will then actively assist in the arrangements between the generator
2
and potential user.
By taking a more active role in the waste transfer, a materials exchange
incurs greater legal risks than a clearinghouse. The legal liability for a
materials exchange is the same as that for other companies involved in the
hauling, treating, and handling of hazardous waste materials. Legal
accountability for parties involved in a w.aste exchange encompasses the
•following areas:
• Public Liability - Generator is responsible for packaging, handling,
and transportation of waste while under the control of a transfer
agent or user.
• Liability to Third Party - Personal or property injury resulting
from a waste in transit between generator and user.
• Contractual Liability - Responsibilities to users with regard to
content of the waste.
Although a materials exchange incurs greater risks and costs, it is generally
more effective than a clearinghouse in recycling industrial waste.
5.9.2 Application
Important factors in the success of waste exchange operations include:
the compatibility of generator waste material with user processes, the purity
5-197
-------
of the waste material, the quantity of specific materials, variability in
waste characteristics or availability, and the amount of processing
required.
With a clearinghouse, the responsibility of ensuring compatibility and
purity of the waste lies with the waste generator and the waste user.
However, in taking a more active role, a materials exchange also assumes this
responsibility. Laboratory analyses are performed to determine the
constituents of the waste material and the level of impurities present.
Generally, successful results are achieved when wastes are transferred
from industries having high purity requirements to those with lower purity
requirements. Both caustic and acidic wastes with relatively high quantities
of impurities can be reused as cleaning solutions (lower purity
requirements). For example, sulfuric acid wastes from pharmaceutical
manufacturers can be reused by iron and steel plants in cleaning rolling
steel. Greater purity is generally required when the corrosive wastes are
reused in a manufacturing process. Although purity is not as critical when
corrosive wastes are reused for neutralization or pU adjustments,
compatibility is very important.
Most users require large amounts of waste material that are available on a
regular basis. However, there are occasions when small-volume, one-time-only
wastes will be required. A clearinghouse will generally list all volumes of
waste, whereas a materials exchange will only list those wastes available in
sufficient quantity to guarantee a transfer. Often, industries producing
large quantities of a specific waste type will find it more economical to
process the waste in-house and reuse it, or send it to a recovery facility.
Medium and small-volume generators typically cannot afford in-house processing
equipment and are, therefore, more likely candidates for waste transfer.
Large volumes of waste materials are generated as sludge from a variety
of industries. The large volume, complex composition, and physical form of
123 '
sludge makes their transfer impractical. ' ' However, if waste streams are
segregated and processed near the point of origin (i.e., prior to sludge
generation), valuable resources (e.g., acids, alkalies, and metals) may be
recovered more cost-effectively.
Waste exchanges are more successful when the area served by the waste
exchange includes a larger, more diversified industrial base. Industries
which represent the majority of both potential users and generators that could
benefit from the use of waste exchanges are listed in Table 5.9.2.
5-198
-------
TABLE 5.9.2. INDUSTRIES REPRESENTING THE MAJORITY
OF GENERATORS AND POTENTIAL USERS
BENEFITING FROM WASTE EXCHANGE
Percent
of wastes
SIC code Industry description transferrable
2911 Petroleum refining 63
2865, 2669 Organic chemicals 22
2831, 2833, 2834 Pharmaceuticals 17
35XX Industrial machinery 17
Source: Reference 1.
5-199
-------
5.9.3 Costs
Costs of waste exchange will vary with the type of service employed, the
amount of waste to be transferred, the distance between exchanging companies,
and reprocessing costs. In order for a waste exchange to be a cost-effective
alternative for users and generators, the transfer must meet the following
145
requirements: ' '
• User's net costs for waste material must be lower than purchase
costs for feedstock;
• Generator's net costs for waste transfer must be lower than waste
disposal costs; and
• User's and generator's net gain from the waste transfer will
adequately compensate any transportation and/or reprocessing costs.
The costs to waste generators and potential waste users for using
the services of a.clearinghouse are minimal. The prevailing practice by
clearinghouses is to charge a flat service fee, regardless of the waste
* l
quantity listed. This fee generally ranges from $5 to $20 per listing.
Clearinghouses, which are not associated with an organization or trade
associationt may also charge a subscribers fee for their waste catalogs.
If a -waste generator or potential waste user chooses to use the
clearinghouse form of waste exchange, they will also incur additional costs
for laboratory analysis of the waste to be exchanged, transportation costs for
transferring the waste, and costs for any reprocessing required. For the
waste generator, these costs can be offset by reduced disposal costs and/or
revenues from the sale of the waste material. Costs incurred by the waste
user will be offset by reduced purchase costs for feedstock material. A
transfer over a distance greater than 50 miles is generally not economical if
the waste is valued at less than $0.13/lb.1)6'7
Costs for a materials exchange service would vary with the particular
waste to be transferred. These costs would depend upon the market potential
j
of the waste material, transportation costs, laboratory analytical costs, and
1 /
reprocessing costs. Laboratory analysis, reprocessing, and transportation
5-200
-------
may sometimes be subcontracted to other companies by the material
exchange. Although costs for use of a materials exchange service are
higher, the generator and user do not have to handle the waste transfer and,
therefore, incur fewer costs and risks during the actual waste transfer.
Examples of cost-effective materials exchanges are given in Table 5.9.3.
5.9.4 Status
A number of clearinghouses and material exchanges are currently operating
in Europe, Canada, and the United States. Tables 5.9.4 and 5.9.5 list the
clearinghouses and material exchanges currently operating in North America.
Waste exchange can serve the following functions: '
• Saves valuable raw materials;
• Saves energy by not having to process raw materials fo.r disposal;
• Saves purchase costs for raw materials; and
• Benefits health and environmental quality by decreasing the
procurement of raw materials and the disposal of waste.
With rising raw materials costs, and increasing restrictions on land
disposal, waste exchange is becoming an increasingly attractive alternative to
conventional practices for handling corrosive wastes. Under the USWA of 1984,
small generators will have increased requirements for waste handling. Waste
exchange may be aIcost-effective alternative for these small generators who
do not have the resources for inhouse recycling/reuse.
5-201
-------
-5-9^.--EXAMPLES OF ECONOMIC USES OF WASTE EXCHANGES FOR CORROSIVE WASTES
User/generator
Storage facility
Milling company
Manufacturer
Steel processor
Manufacturer
County agency
Castings mfgr.
Metal products
mfgr.
Tons
generated
0.11
12.10
1.20
480.00
144.00
0.44
Ongoing
Ongoing
Waste description
Sulfuric acid
Soda ash, liquids
Sodium hydroxide
Sulfuric acid
Sodium hydroxide
Oxalic acid
Sodium hydroxide
Sodium hydroxide
Savings over
disposal
costs ($)
350
1,350
NA
55,200
35,200
900
*
*
Savings over
raw material
purchase($)
25
1,320
1,600
NA
168,000
350
*
9,582
Net
• value
exchange
375
2,670
1,600
55,200
203,200
1,250
6,060
9,582
of
(*)
Notes: NA - Not applicable.
* - Not available.
Source: References 2, 8, 9, and 10.
-------
TABLE 5.9.4. CLEARINGHOUSE (INFORMATION) WASTE EXCHANGES IN NORTH AMERICA
Clearinghouse
Location
Contact
Telephone No.
Alberta Waste Materials Exchange
California Waste Exchange
Canadian Inventory Exchange*
Canadian Waste Materials Exchange
Enkarn Research Corporation*
Georgia Waste Exchange*
Great Lakes Regional Waste Exchange
Industrial Materials Exchange Service
Industrial Waste Information Exchange
Manitoba Waste Exchange
Midwest Industrial Waste Exchange
Montana Industrial Waste Exchange
Northeast Industrial Waste Exchange
Ontario Waste Exchange
Piedmont Waste Exchange
Southern Waste Information Exchange
Tennessee Waste Exchange
Wastelink, Div. of Tencon Assoc.*
Western Waste Exchange
Edmonton, Alberta, CAM
Sacramento, CA
Ste-Adle, Quebec, CAN
Mississauga, Ontario, CAM
Albany, HY
Marietta, GA
Grand Rapids, MI
Springfield, IL
Newark, NJ
Winnipeg, Manitoba, CAN
St. Louis, MO
Helena, MT
Syracuse, NY
Mississauga,- Ontario, CAM
Charlotte, NC
Tallahassee, FL
Nashville, TN
Cincinnati, OH
Tempe, AZ
Charles Wood
Robert McCormick
Philippe LaRoche
Robert Laughlin
J.T. Engster
Michael Weeks
William SCough
Hargo Ferguson
William E. Payne
Rod McCormick
Clyde H. Wiseman
Buck Boles
Lewis Cutlet
Brian Forrestal
Mary McDaniel
Roy Berndon
Sharon Bell
Mary E. Malotke
Nicholas Hild
(403)436-6303
(916)324-1818
(514)229-6511
(416)822-4111
(518)436-9684
(404)363-3022
(616)451-8992
(217)782-0450
(201)623-7070
(204)257-3891
(314)231-5555
(406)442-2405
(315)422-6572
(416)822-4111
(704)597-2307
(904)644-5516
(615)256-5141
(513)248-0012
(602)965-2975
"Operates for profit. .
Source: References 3 and 11.
5-203
-------
TABLE 5.9.5. MATERIAL EXCHANGES IN NORTH AMERICA
Material exchange Location Contact Telephone No.
Resource Recovery of America, Inc. Mulberry, FL Robert Kincart (813)425-1084*
Zero Waste Systems, Inc. Oakland, CA Trevor Pitts (415)893-8257
NYC Environmental Facilities Corp. Albany, NY Peter A. Marini (518)457-4132
Source: References 3 and 12.
5-204
-------
REFERENCES
1.
2.
3.
GCA Corporation, Technology Division. Industrial Waste Management
Alternatives Assessment for the State of Illinois, Volume IV: Industrial
Waste Management Alternatives and Their Associated
Technologies/Processes. Final Report prepared for the Illinois
Environmental Protection Agency, Division of Land Pollution Control.
GCA-TR-80-80-G. February 1981.
New York State Environmental Facilities Corporation (EFC).
Materials Recycling Act Program, 4th Annual Report. 1985.
Industrial
Jones, E.B., and W. Banning. The Role of Waste Exchanges in Waste
Minimization and Reclamation Efforts. In: Proceedings of the Hazardous
and Solid Waste Minimization Seminar, sponsored by Government Institutes,
Inc., Washington, D.C. May 8-9, 1986.
4. Terry, R.C., et al. Arthur D. Little, Inc. Waste Clearinghouses and
Exchanges: New Ways for Identifying and Transferring Reusable Industrial
Process Waste. .EPA/SW-130C. October 1976.
5. Terry, R.C. , Berkowitz, J.B., and C.H. Porter. Waste Clearinghouses and
Exchanges. ' Chemical Engineering Progress. December 1976.
6. Laugh 1 in, R.G.W. , Golomb, A., and H. Mooij. Waste Materials Exchanges
for Environmental Protection and Resource Conservation. In: Proceedings
. of the 24th Ontario Industrial Waste Exchange Conference, Toronto,
Ontario. June 1, 1977.
7. Laughlin, R.G.W. Canadian Waste Exchange Program: Successes and
Failures. In: Proceedings of the National Conference on Hazardous and
Toxic Wastes Management, New Jersey Institute of Technology. June 5,
1980.
8. New York State Environmental Facilities Corporation (EFC). Industrial
Materials Recycling Act Program, 5th Annual Report. 1986.
9. New York State Environmental Facilities Corporation (EFC). Industrial
Materials Recycling Act Program, 3rd Annual Report. 1984.
10. New York State Environmental Facilities Corporation (EFC).
Materials Recycling Act Program, 2nd Annual Report. 1983.
Industrial
5-205
-------
11. Versar, Inc. National Profiles Report for Recycling - A Preliminary
Assessment. Draft Report prepared for the U.S. EPA, Waste Treatment
Branch, under EPA Contract No. 68-01-7053 (Work Assignment No. 17). July
8, 1985.
12. Mar in i, P. A. New York State Environmental Facilities Corporation.
Telephone conversation with J. Spielman, GCA Technology Division, Inc.
July 29, 1986.
5-206
-------
SECTION 6.0
CONSIDERATIONS FOR SYSTEM SELECTION
6.1 GENERAL CONSIDERATIONS
Waste management options consist of four basic alternatives: source
reduction, waste exchange, recycling/reuse, use of a treatment
(i.e., neutralization) or disposal processing system or some combination of
these waste handling practices (see Figure 6.1.1). Recovery, treatment, and
disposal may be performed onsite in new or existing processes or through
contract with a licensed offsite firm which is responsible for the final
disposition of the waste. Selection of the optimal waste management
alternative will ultimately be a function of regulatory compliance and
economics, with additional consideration given to factors such as safety,
public and employee acceptance, liability, and uncertainties in meeting cost
and treatment objectives.
Many of the technologies discussed in previous sections can be utilized
to meet land disposal ban requirements or to achieve adequate waste recovery
rates. However, practicality will limit applications to waste streams
possessing specific characteristics. Since many processes yield large
economies of scale, waste volume will be a primary determinant in system
selection. The physical and chemical nature of the waste stream and pertinent
properties of its constituents will also determine the applicability of waste
treatment processes. Treatment will often involve neutralization and the use
of other technologies in a system designed to progressively recover or destroy
hazardous constituents in the most economical manner. Incremental costs of
contaminant (i.e., toxic organics and heavy metals) removal will increase
rapidly as low concentrations are attained.
6-1
-------
ACID/ALKALI
FEED
KEUSE OP
RECOVERED
PRODUCT
SALE OF
RECOVERED
PRODUCf
REUSE OR
RECYCLING
SYSTEM
CORROSIVE WASTE
GENERATING
PROCESS
SOURCE
REDUCTION
NEUTRALIZATION
TREATMENT/DISPOSAL
SYSTEM
N
IIAZARDO
WASTE
DISCHARGE
Figure 6.1.1. Corrosive waste management options.
-------
6.2 WASTE MANAGEMENT PROCESS SELECTION
All generators of hazardous corrosive wastes will be required to
undertake certain steps to characterize regulated waste streams and to
identify potential treatment options. Treatment process selection should
involve the following fundamental steps:
1. Characterize the source, flow, and physical/chemical properties of
the waste.
2. Evaluate the potential for source reduction.
3. Evaluate the potential for waste exchange.
4. Evaluate the potential for reuse or sale of recycled acid/alkali
streams and other valuable waste stream constituents;
e.g., recovered metals or solvents.
5. Identify potential treatment and disposal options based on technical
feasibility of meeting the restrictions imposed by the land disposal
ban. Give consideration to waste stream residuals and fugitive
emissions to air.
6. Determine the availability of potential options. This includes the
use o"f off site services, access to markets for recovered products or
waste exchange, and availability of commercial equipment and
existing onsite systems.
7. Estimate total system cost for various options, including costs of
residual treatment and/or disposal and value of recovered products.
Cost will be a function of Items 1 through 5.
8. Screen candidate management options based on preliminary cost
estimates.
9. Use mathematical process modeling techniques and laboratory/
pilot-scale testing as needed to determine detailed waste management
system design characteristics and process performance capabilities.
The latter will define product and residual properties and identify
need for subsequent processing.
10. Perform process trials of recovered products and wastes available
for exchange in their anticipated end use applications.
Alternatively, determine marketability based on stream
c haracteris t ics.
11. Calculate detailed cost analysis based on modeling and performance
results.
6-3
-------
12. Perform final system selection based on relative cost and other
considerations; e.g., safety, acceptance, liability, and risks
associated with data uncertainties.
Key system selection steps are discussed in more detail below.
6.2.1 Waste Characterization
The first step in identifying appropriate waste management alternatives
to land disposal involves characterizing the origin, flow, and quality of
generated wastes. An understanding of the processing or operational practices
which result in generation of the waste forms the basis for evaluating waste
minimization options. Waste flow characteristics include quantity and rate.
Waste quantity has a direct impact on unit waste management costs due to
economies of scale in processing costs and marketability of recovered
products. Flow can be continuous, periodic, or incidental (e.g., spills) and
can be at a relatively constant or variable rate. This will have a direct
impact on storage requirements, and waste management process design; e.g.,
continuous or batch flow.
Waste physical and chemical characteristics are generally the primary
determinant of waste management process selection for significant volume
wastes. Of particular concern is whether the waste is pumpable, inorganic or
organic, and whether it contains recoverable materials or constituents which
may interfere with processing equipment or process performance. Waste
properties such as degree of corrosivity, reactivity, ignitability, heating
value, viscosity, concentrations of toxic organic chemical constituents,
biological and chemical oxygen demand, and solids, oil, grease, metals and ash
content need'to be determined to evaluate applicability of certain waste
management processes. Individual constituent properties such as solubility
(affected by the presence of chelating compounds), vapor pressure, partition
coefficients, reactivity, reaction products generated with various biological
and chemical (e.g., neutralizing, oxidizing, and reducing) reagents, and
adsorption coefficients are similarly required to assess treatability.
The presence or absence of buffers will affect neutralization reagent and
pH control system requirements. Chelators will enhance metal solubility,
6-4
-------
requiring over neutralization to alkaline pH to effect metal precipitation.
Cyanides and chromium will require treatment through chlorination and
reduction, respectively, prior to combined neutralization with other corrosive
wastes.
Finally, variability in waste stream characteristics will necessitate
overly conservative process design and additional process controls, thereby
increasing costs. Marketability of recovered products or materials offered
for waste exchange will also be adversely affected by variability in waste
characteristics.
6.2.2 Source Reduction Potential
Source reduction potential is highly site specific, reflecting the
variability of industrial waste generating processes and product
requirements. Source reduction alternatives which should be investigated
include raw material substitution, product reformulation, process redesign and
waste segregation. .The latter may result in additional handling and storage
• «
requirements, while viability of other waste reduction alternatives may be
more dependant on differential processing costs and impact on product quality.
Many opportunities exist for firms to achieve waste minimization through
implementation of simple, low-cost methodologies currently proven in
2
successful programs. Lack of available techniques has been less of an
impediment to increased implementation than perception that these methods are
2 i
not available. Historically, management has favored end-of-pipe treatment
and has been reluctant to institute waste reduction and reuse practices. This
i
reluctance is primarily due to potential for process upsets or adverse impacts
on product quality. Other risks of installing waste reduction methods include
uncertain investment returns and production downtime required for
installation. However, in the wake of increasing waste disposal and liability
costs, source reduction has repeatedly'proven to be cost effective, while at
| 2
the same time providing for minimal ad'yerse health and environment impact.
Thus, source reduction should be considered a highly desirable waste
management alternative.
6-5
-------
6.2.3 Waste Exchange Potential
As discussed in Section 5.8, corrosive wastes have significant potential
for being managed through waste exchange. Wastes will be good candidates for
exchange if: 1) contaminant concentrations are low, consistent, and at levels
which'are compatible with user processes; 2) processing requirements are
minimal; and 3) the waste is available in sufficient volumes on a regular
basis. Wastes generated from processes with high purity requirements may
be used directly in processes with lower specifications; e.g., paint stripping
or equipment cleaning. Another reuse method with high exchange potential is
acid/alkali waste combination for mutual neutralization. Economics are
particularly favorable when these individual wastes would have required
separate post—treatment for metal precipitation or organic removal. Finally,
waste exchange may prove to be the least cost management option for firms with
wastes that have high recovery potential, but lack the waste volume or capital
to make onsite recovery viable.
Potential for waste'exchange is reduced when industries are faced with
concerns about liability, confidentiality, and quality of the waste.
Additionally, transportation costs are frequently a limiting factor in the
2
exchange of high volume, low concentration wastes.
6.2.4 Recovery Potential
As part of the waste characterization step, the presence of potentially
valuable waste constituents; should be determined. In the case of concentrated
corrosive solutions, the bulk of the waste typically has recycle potential.
I
Economic benefits from recovery and isolation of this or other materials may
result if they can be reused in onsite applications or marketed as saleable
i
products. In the former cage, economic benefits result from decreased
consumption of virgin raw materials. This must be balanced against possible
adverse effects on process equipment or product quality resulting from buildup
or presence of undesirable contaminants. Market potential is limited by the
lower value of available quantity or demand. Market potential will be
enhanced with improved product purity, availability, quantity, and consistency.
6-6
-------
Onsite reuse has several advantages relative to marketing for offsite use
including reduced liability and more favorable economics. Offsite sale is
less profitable due to transportation costs and the reduced purchase price
which can typically be charged offsite users as a result of uncertainties in
product quality. Thus, economics and liability combine with factors such as
concerns about confidentiality to encourage onsite reuse whenever possible.
In practice, recycling of corrosive wastes has been limited to recovery
of highly concentrated solutions, particularly those which are not amenable to
management in low-cost wastewater treatment systems. Recycling options have
been discussed in detail in Section 5.0. These are summarized in Table 6.2.1
with information provided on current applications, residuals generated, and
availability.
6.2.5 Identifying Potential Treatment and Disposal Options
Following an assessment of the potential for source reduction and
recycling, the generator should evaluate treatment systems which are
technically capable of meeting the-.necessary degree of neutralization and
hazardous constituent removal or destruction. Guideline considerations for
the investigation of treatment technologies are summarized in Table 6.2.2.
The treatment objectives for a waste stream at a given stage of treatment
will define the universe of candidate technologies. Treatment objectives must
ultimately assure that the waste meets appropriate specifications for surface
water discharge (NPDES, POTW), land disposal (pH, Extraction Procedure
Toxicity Characteristic leaching test for metals and organics, treatment
standards for priority organics), or reuse in an industrial process.
Restrictive waste characteristics (e.g., concentration range, flow,
interfering compounds) and technological limitations of candidate treatment
processes will reduce the list of candidate technologies to a list of
potential applications for a specific waste. Consideration must be given to
pretreatment options for eliminating restrictive waste characteristics
(e.g., cyanide or chromium pretreatment), process emissions, residuals and
their required treatment, and opportunities for recovery. System design will
be based on neutralization requirements and the most difficult compound to
recover, remove or destroy.
6-7
-------
TABLE 6.2.1. SUMMARY OF RECOVERY/REUSE TECHNOLOGIES FOR CORROSIVE WASTES
Frocaaa
Applicable waate atreaai* Stage of development
Ferfonaanco
Realduata generated
Coat
Kvaporat ton/
dlattllatlon
Metal plating rlnaaa)
acid plcktlng liquor*
Uall-eatabllahad for
treating plating
rlniea.
Ft it Ing •olHtlon recovered
(or reuae In pitting bath.
Rime water can be rauaed.
liapurltle* wilt b* concentrated,
therefore,' cryatalllaatlon/
filtration eyateai My b«
required.
Can be coat-effective
for recovering corroaivi
plating solution! froia
rinae wjt«r».
Cry*taltUatlon
Ion exchange
<*>
tlectrodlalyal*
H280i pickling liquora;
IIHOj/lir pickling llquorat
eauittc aliHilmin etch
•olutlona.
Mating riniaaj acid
pickling bat ha! elimlnuei
etching •olutionaj
II2SO^ anodltlng
eolutiona; rack-atripping
lolutioni (lir/IWOj).
Recovery of chromic/
an Ifuric acid etching
•olutlona.
lecoverjr of pitting rinaea
(particularly chronic acid
rlnae water)•
Recovery of IUIO)/lir
plcktlng liquor*.
20 to 2S eyateaia cur-
rently in operation
(fewer application*
for cauatlc recovery),
97-981 recovery for hSSO
(BO-8H awtal reanval).
991 HHOj and 501 HT
recovered.
80S recovery of NeOH.
Ferroue aulfate hetptehydeate Coat-effective if treat-
cryatala (can be traded or aold). ing large quant It lea of
waate.
Metal fluoride cryetal* (can
recover additional Hf by
theraul deeoaipoaition).
Atualnua hydroxide cryitale
(can be traded or aold).
Several R'tK unit* in Cocurrent ayateaM not tech- Cocurrent proceaa generatea
operation for treat* nlcally feaaible for direct
awnt of corroalvea.
Unit* for direct
treatment of acid
bath only available
(ton ECO-TEC, Ltd.
Unit* currently being
aold, but United
area of application.
S in operation.
Several In operation.
Marketed, none in
operation to date.
treat awnt of corroelvaa; can
be uaed in conjunction vlth
nautrallaatlon technologlea
to lower overall coat*.
RFtE unlta ahow good reautte.
Conventional RFIB perforau
beat, with dilute aolutlona.
APO perfor»a beet with high
•etal concentration
(30 to 100 g/L).
811 recovery of etching
•olutlon.
451 copper remval;
IDS line reenval.
Work* beat when copper con-
centration* are In the 2 to
4 oi/gal uaage.
3 H HF/IINO] recorded.
•pent regenerant
corroalve.
which la alao
RFU and APU are
coat-effective.
Recovered *etala which can be
reuaed, treated, dlipoaad, or
Marketed.
Hetala which can be treated,
dtapoied, or regenerated for
reuae.
Chroaiic acid can be returned to
plating bath! rinae water can
be reuaed.
Coat-effective for
apeciflc application*
(chronlc/autfate acid
etchanta).
Low capital Inveatiaent;
coat-effective for
•pcctfic application
(chromic acid rinaea).
2 M KOII Sotn which can be Coat-effective for large
recycled back to the pretreat- quantity generator.
•ent atep for thla ED application.
1(continued)
-------
TABLE 6.2.1 (continued)
Procoas Appllcnblc waste streams Stage of development
Performance
Residuals generated
Cost
Reverie osmosis Filling rinsei.
Donnan dlalyaia/
coupled
transport
Plating rlnaei; poten-
tially applicable to
acid batha.
O>
VO
Solvent
extraction
Thermal
decompualtlort
HNOj/IIP pickling liquor§.
Acid waatea.
Corroilve waite
branea marketed by
(our companies.
RD nodule lyitena
applicable to corro-
alvea available from
two coiapanlea.
Donnan analyala only
lab-icale Kited.
Coupled transport
lab and field teited.
Coupled tranaport
•ysteia It currently
being marketed.
Commercial-scale
system installed for
development purposes
In Europe and Japan.
No commercial-scale
Inatallatlona In U.S.
Well-established for
recovering ipent
pickle liquors gen-
erated by ateel
Industry. Pilot-
acale atage for
organic vastea.
90X converaion achieved
with cyanide plating rinses.
Data not available for
Donnan analysis (further
testing required).
Coupled transport has dem-
onstrated 99X recovery of
chromate from plating rlnaea.
Other plating rinses should
be applicable, but not fully
teated.
95t recovery of IINOjj
70X recovery of HF.
991 regeneration efficiency
for pickling liquors.
Recovered plating solution
returned to plating bath (after
being concentrated by an
evaporator). Rlniewater reused.
Data not available for Donnan
analysis.
For chromate plating rinse
applications, sodium cliromate
la generated; can be used else-
where in plant or subjected
to Ion exchange to recover
chromic acid for recycle to
plating solution.
Metal aludge (95X Iron can be
recovered by tha nail
deconpoaltlon).
98-991 purity iron oxide which
can be reused, traded, or
marketed.
Coat-effective for
limited applications.
Development of a more
chemically resistant
membrane would make It
very coat-effective for
a wider area of
application.
No coat data available
for Donnan analysis.
Average capital coat
for plating shop Is
$20,000. Can be coat-
effective for specific
applications.
Not available^
Expensive capital
invest itent. Only coat-
effective for large
quantity waite acid
generators.
-------
TABLE 6.2.2. GUIDELINE CONSIDERATIONS FOR THE INVESTIGATION OF WASTE
TREATMENT, RECOVERY, AND DISPOSAL TECHNOLOGIES
A. Objectives of treatment;
- Primary function (pretreatment, treatment, mutual neutralization,
residuals treatment)
- Primary mechanisms (neutralization, destruction, removal, conversion,
separation)
- Recover waste for reuse
- Recovery of specific chemicals, group of chemicals (acids, alkalis,
metals, solvents, other organics)
- Polishing for effluent discharge (NPDES, POTW)
— Immobilization or encapsulation to reduce migration (inorganic sludge)
- Overall volume reduction of waste
- Selective concentration of constituents (acids, alkalis, metals,
solvents, other organics)
- Detoxification of hazardous constituents
B. Waste applicability and restrictive waste characteristics;
- Acceptable concentration range of primary and restrictive waste
constituents
- Acceptable range in flow parameters
-. Chemical and physical interferences (compatibility with neutralization
reagent) . •
C. Process operation and design;
- Batch versus continuous process design
- Fixed versus mobile process design
- Equipment design and process control complexity (pH control)
- Variability in system designs and applicability
- Spatial requirements or restrictions
- Estimated operation time (equipment down-time)
- Feed mechanisms (wastes and reagents; solids, liquids, sludges,
slurries)
- Specific operating temperature and flow
- Sensitivity to fluctuations in feed characteristics
- Residuals removal mechanisms
- Reagent selection and requirements
- Ancillary equipment requirements (tanks, pumps, piping, heat transfer
equipment)
- Utility requirements (electricity, fuel and cooling, process and
make-up water)
(continued)
6-10
-------
TABLE 6.2.2 (Continued)
D- Reactions and theoretical considerations;
- Waste/reagent reaction (neutralization, destruction, conversion,
oxidation, reduction)
- Competition or suppressive reactions (buffers)
- Enhancing conditions (specify chemicals)
- Fluid mechanics limitations (mass and heat transfer)
- Reaction kinetics (temperature and concentration effects)
- Reactions thermodynamics (endothermic/exothermic/catalytic)
E. Process efficiency:
- Anticipated overall process efficiency
- Sensitivity of process efficiency to:
- feed concentration fluctuations
- reagent concentration fluctuations
- process temperature fluctuations
- toxic constituents (biosystems)
- physical form of the waste
- other waste characteristics
F. Emissions and residuals management:
- Extent of fugitive and process emissions and potential sources
(processing equipment, storage, handling)
- Ability (and frequency) of equipment to be "enclosed"
- Availability of emissions and residuals data/risk calculations
- Products of incomplete reaction
- Relationship of process efficiency to emissions or residuals generation
- Air pollution control device requirements
- Process residuals (fugitive/residual reagents, recovered products,
filter cakes, sludges, incinerator scrubber water and ash)
- Residual constituent concentrations and teachability
- Delisting potential
G. Safety considerations;
- Safety of storing and handling corrosive wastes, reagents, products and
residuals
— Special materials of construction for storage and process equipment
- Frequency and need for use of personnel protection equipment
- Requirements for extensive operator training
- Hazardous emissions of wastes or reagents
- Minimization of operator contact with wastes or reagents
- Frequency of maintenance of equipment containing hazardous materials
- High operating temperatures
- Difficult to control temperatures
- Resistance to flows or residuals buildup
- Dangerously reactive wastes/reagents
- _Dangerously volatile wastes/reagents
6-11
-------
A key consideration in Che choice of a corrosive wastewater
neutralization system is reagent selection. Potential reagents and their
associated advantages and disadvantage with respect to handling, processing
and sludge generation, are summarized in Table 6.2.3. Sludge characteristics
and volume will have a significant impact on ultimate reagent selection since
disposal of this material constitutes a large percentage of total treatment
costs. The presence of toxic organics in corrosive wastewaters will also
significantly add to post-treatment costs. These will increase with organic
concentration and decrease with reactivity, volatility, and
A
biodegradability.
Concentrated organics and sludges will typically be treated through
neutralization prior to handling in other equipment. Organic wastes will then
be amenable to recovery, treatment, or disposal (e.g., incineration)
technologies as summarized in Section 4.1 and discussed in detail in the
45
literature. ' Sludges will be neutralized and dewatered prior to aqueous
treatment of the supernatant. Inorganic dewatered residues can be solidified/
encapsulated prior to land disposal and organic solids can be incinerated or
otherwise thermally destroyed (Section 4.1).
Potential reagent and processing system equipment designs and
configurations have been summarized in Section 4.0. Ultimately, the selection
of a specific treatment system from the list of potentially applicable
processes will depend on cost, availability, and site specific factors. These
considerations are discussed below.
6.2.6 Availability of Potential Management Options
The availability of each component of a waste management system will
affect its overall applicability. Existing available onsite treatment process
capacity (e.g., wastewater treatment system), ancillary equipment, labor,
physical space, and utilities will have a significant impact on the economic
viability of a treatment system. Purchased equipment must be available in
sizes and processing capabilities which meet the specific needs of the
facility. Offsite disposal, recovery, and treatment facilities and companies
using exchanged materials or purchasing saleable products must be located
within-a reasonable distance of the generator to minimize transport costs. In
-6-12
-------
Process
/ TABLE 6.2.3. SUMMARY OF NEUTRALIZATION TECHNOLOGIES
Applicable waste streams Stage of development
Perform nee
Reiidualf generated
Coat
Ac i
mutual
neutralization
Li me a tone
Lime
I '
*-•
to
Caustic soda
Sulfuric acid
All acid/alkali compatible
waate streams except
cyanide.
Dilute acid waate streams
of less than 5,000 mg/L
mineral acid strength and
containing low concentre*
tlona of acid salts.
All acid wastes.
Wall developed.
Well developed.
Hell developed.
All acid wastes.
All alkaline wastes
except cyanide.
Hell developed.
Well developed.
Generally alower than com-
parable technologies due to
dilute concentrationa of
reagents. Hay evolve haz-
ardoua constituents if in-
compatible waates are mixed.
Requires atone sices of
0.074 nm or lees. Requires
45 minutes or more of reten-
tion time. Can only neutra-
lize acidic wastes to pll
6.0. Huat be aerated to
remove evolved COj.
Requires IS to 10 mlnutea
of retention time. Hust be
slurried to a concentration
of 10 to 15% solids prior
to use. Can under- (below
pll 7) or over- (above pH 7)
neutralise.
Requires ) to 1) minutes of
retention time. In liquid
form, eaay to handle and
apply. Can under- or over-
neutralise including pll
,11 or higher.
Requires 1$ to 10 minutes
of retention time. In
liquid form, but presents
burn hazard. Highly re-
active and widely avail-
able.
Variable, dependent on quantity
of inaolubles and products con-
tained In each waste stream.
Hill generate voluminous sludge
product when reacted with
sulfate-contalnlng wastes.
Stone* over 200 mesh will sul-
fonate, be rendered Inactive,
and add to aludge product.
Will generate voluminous aludge
similar to limestone.
Reaction products are generally
soluble, however, sludges do not
dewater as readily or as easily
aa lime or limestone.
Hill generste large quantities
of gypsun sludge when rescted
with calcium-based alkaline
waates.
Least expensive of all
neutralisation
technologies.
Hot cost-effective in
treating concentrated
wastes. Hay be cost-
effective in treating
dilute acidic wastes.
Hore expensive than
crushed limestone
(200 neah).
Host expensive of all
widely used alkaline
reagents (five times the
cost of line).
Least expensive of all
widely used acidic
reagents.
Hydrochloric
acid
Carbonic acids,
liquid carbon
dioxide
All acid wastes.
All alkaline wastes
except cyanide.
Well developed, but
rarely applied due
to high reagent
cost.
Requirea 5 to 20 minutes of
retention time. Liquid
form preaenta burn and fume
hazard. Hore reactive
than aulfurlc.
Emerging technology. Retention time I to 1-1/2
minutes: In liquid tons,
must be vaporized prior to
use. Can only neutralize
alkaline wastes to pll 8.3
end point.
Reaction products are generally
soluble. i
Will for* calcium carbonate pre-
cipitate when reacted with
calcium-baaed alkaline wastes.
Approximately twice as
expenaive as sulfuric
on a neutralization
equivalent basis.
Approximately 1 to 4
tiroes aa expensive as
sulfuric. Therefore,
limited to applications
using more than 200 tons
of reagents per yenr or
with flow rate greater
than 100,000 gpd.
-------
addition, they must have available capacity for the waste type and volume
generated. Finally, time constraints may eliminate certain treatment
processes from consideration as a result of anticipated delays in procurement,
permitting, installation, or start-up.
In general, neutralization systems are widely applied and readily
available. However, several recovery systems and post-treatment systems for
organic wastes have only recently been applied to corrosive waste treatment.
Availability and uncertainty in expected performance will play a significant
role in the decision to implement these technologies.
6.2.7 Management System Cost Estimation
The relative economic viability of candidate waste management systems
will be the primary determinant of ultimate system selection. This must be
evaluated on the basis of total system costs which includes the availability
of onsite equipment, labor and utilities, net value of recovered products, and
recovery/treatment/disposal processing costs. Capital equipment expenditures
and financing constraints are frequently a limiting factor in system
selection, particularly for processes which have higher uncertainties of
success.
Costs for a given management system will be highly dependent on waste
physical, chemical, and flow characteristics. Thus, real costs are very
site-specific and limit the usefulness of generalizations. The reader is
referred to the sections on specific technologies (Sections 4.0 and 5.0) for
data on costs and their variability with respect to flow and waste
characteristics. Costing methodologies have also been adequately described in
6 7 ft Q
the literature ' ' and are available in software packages. Major cost
centers which should be considered are summarized in Table 6.2.4.
6.2.8 Modeling System Performance and Pilot-Scale Testing
Following this preliminary cost evaluation, which will enable the
generator to narrow his choice of waste management options, steps must be
taken to further finalize the selection process. These could involve the use
of theoretical models to predict design and operating requirements. However,
'6-14
-------
TABLE 6.2.4. MAJOR COST CENTERS FOR WASTE MANAGEMENT ALTERNATIVES
A. Credits ;
- Material/energy recovery resulting in decreased consumption of
purchased raw materials
Sales of waste products
B. Capital costs :a
- Processing equipment (reagent addition, reaction vessel, recovery
apparatus, sludge and other residual handling equipment)
- Ancillary equipment (storage tanks, pumps, piping)
- Pollution control equipment
- Vehicles
- Buildings, land
- Site preparation, installation, start-up
C. Operating and maintenance costs:
- Overhead, operating, and maintenance labor
- Maintenance materials
- Utilities (electricity, fuel, water)
- Reagent materials
- Disposal, off site recovery, and waste brokering fees
- Transportation
.- Taxes, insurance, regulatory compliance, and administration
D. Indirect costs and benefits:
- Impacts on other facility operations; e.g., changes in product quality
as a result of source reduction or use of recycled materials
- Use of processing equipment for management of other wastes
aAnnual costs derived by using a capital recovery factor:
Where: i - interest rate and n * life of the investment. A CRF of
0.177 was used to prepare neutralization cost estimates in
this document. This corresponds to an annual interest rate
of 12 percent and an equipment life of 10 years.
6-15
-------
models generally sacrifice accuracy for convenience and are often not
sufficiently accurate for complex waste streams. Laboratory data, or
pilot-plant and full-scale data, may ultimately be needed to confirm predicted
performance. In fact, some data may be needed as model inputs for predicting
system behavior.
In many cases, models are useful in predicting behavior and can be used
in place of costly laboratory testing. Models are also useful in assessing
relative performance, costs of various approaches to treatment, and the
incremental costs of achieving increasingly stringent treatment concentration
levels. Many suppliers of neutralization and recovery equipment use models "to
optimize design and operations parameters and to scale treatment processes.
Equipment manufacturers are also often able to provide experimental equipment
and models to establish process parameters and cost, including the costs
required for disposal of residuals.
.6-16
-------
REFERENCES
1. Allen, C.C., and B. L. Blaney. Research Triangle Institute. Techniques
for Treating Hazardous Waste to Remove Volatile Organics Constituents.
Performed for U.S. EPA HWERL. EPA-600/2-85-127. March 1985.
2. Committee on Institutional Considerations in Reducing the Generation of
Hazardous Industrial Wastes. Environmental Studies Board, National
Research Council. Reducing Hazardous Waste Generation: An Evaluation
and a Call for Action. National Academy Press, Washington, D.C. 1985.
3. GCA Technology Division, Inc. Industrial Waste Management Alternatives
Assessment for the State of Illinois. Volume IV: Industrial Waste
Management Alternatives and Their Associated Technologies/Processes.
Final Report prepared for the Illinois Environmental Protection Agency,
Division of Land Pollution Control. GCA-TR-80-80-G. February 1981.
4. Breton, M. 'et al. GCA Technology Division, Inc.. "Technical Resource
Document: Treatment Technologies for Solvent Containing Wastes.
Prepared for U.S. EPA HWERL under Contract No. 68-03-3243. August 1986.
5. Surprenant, N. GCA Technology Division, Inc. Technical Resource
Document: Treatment Technologies for Halogenated Organic Wastes.
Prepared for U.S. EPA HWERL under Contract No. 68-03-3243. October 1986.
6. Peters, M. S., and K. D. Timmerhaus. Plant Design and Economics for
Chemical Engineers. 3rd Edition. Mc-Graw Hill Book Company, New York,
NY. 1980.
7. U.S. EPA Design Manual: Dewatering Municipal Wastewater Sludges. U.S.
EPA Municipal Environmental Research Laboratory, Cincinnati, OH.
EPA-625/1-82-014. October 1982.
8. Mitre Corp. Manual of Practice for Wastewater Neutralization and
Precipitation. EPA-600/2-81-148. August 1981.
9. Cunningham, V. L. et al. Smith, Kline & French Laboratories.
Environmental Cost Analysis System. 1986.
-6-17
------- |