Electrochemical Water Processing

Electrochemical Water Processing Scrivener Publishing 3 Winter Street, Suite 3 Salem, MA 01970 Scrivener Publishing Co...

Author: Ralph Zito

139 downloads 1212 Views 12MB Size Report

This content was uploaded by our users and we assume good faith they have the permission to share this book. If you own the copyright to this book and it is wrongfully on our website, we offer a simple DMCA procedure to remove your content from our site. Start by pressing the button below!
Report copyright / DMCA form

DOWNLOAD PDF

Electrochemical Water Processing

Scrivener Publishing 3 Winter Street, Suite 3 Salem, MA 01970 Scrivener Publishing Collections Editors James E. R. Couper Richard Erdlac Pradip Khaladkar Norman Lieberman W. Kent Muhlbauer S. A. Sherif

Ken Dragoon Rafiq Islam Vitthal Kulkarni Peter Martin Andrew Y. C. Nee James G. Speight

Publishers at Scrivener Martin Scrivener ([email protected]) Phillip Carmical ([email protected])

Electrochemical Water Processing

Ralph Zito

Scrivener

WILEY

Copyright © 2011 by Scrivener Publishing LLC. All rights reserved. Co-published by John Wiley & Sons, Inc. Hoboken, New Jersey, and Scrivener Publishing LLC, Salem, Massachusetts. Published simultaneously in Canada. No part of this publication may be reproduced, stored in a retrieval system, or transmitted in any form or by any means, electronic, mechanical, photocopying, recording, scanning, or otherwise, except as permitted under Section 107 or 108 of the 1976 United States Copyright Act, without either the prior written permission of the Publisher, or authorization through payment of the appropriate per-copy fee to the Copyright Clearance Center, Inc., 222 Rosewood Drive, Danvers, MA 01923, (978) 750-8400, fax (978) 750-4470, or on the web at www.copyright.com. Requests to the Publisher for permission should be addressed to the Permissions Department, John Wiley & Sons, Inc., Ill River Street, Hoboken, NJ 07030, (201) 748-6011, fax (201) 748-6008, or online at http://www.wiley.com/go/permission. Limit of Liability/Disclaimer of Warranty: While the publisher and author have used their best efforts in preparing this book, they make no representations or warranties with respect to the accuracy or completeness of the contents of this book and specifically disclaim any implied warranties of merchantability or fitness for a particular purpose. No warranty may be created or extended by sales representatives or written sales materials. The advice and strategies contained herein may not be suitable for your situation. You should consult with a professional where appropriate. Neither the publisher nor author shall be liable for any loss of profit or any other commercial damages, including but not limited to special, incidental, consequential, or other damages. For general information on our other products and services or for technical support, please contact our Customer Care Department within the United States at (800) 762-2974, outside the United States at (317) 572-3993 or fax (317) 572-4002. Wiley also publishes its books in a variety of electronic formats. Some content that appears in print may not be available in electronic formats. For more information about Wiley products, visit our web site at www.wiley.com. For more information about Scrivener products please visit www.scrivenerpublishing.com.

Library of Congress Cataloging-in-Publication ISBN 978-1-118-09871-4

Printed in the United States of America 10

9 8 7 6 5 4 3 2 1

Data:

Contents Preface Acknowledgements Introduction

xi xvii xix

1. Water Contaminants and Their Removal 1.1 Introduction 1.2 Technology, History, and Background 1.3 Application Areas: Electrochemical Technology Water Processing

10

2. Basic Electrochemical and Physical Principles 2.1 Introduction 2.2 Acidity and Alkalinity, pH 2.3 Activity and Activity Coefficients 2.4 Equilibrium and Dissociation Constants 2.4.1 Degree or Percentage Dissociation 2.5 Electrode, or Half Cell Potential 2.6 Chemical Potential Definition 2.7 Concentration Potential 2.8 Equivalent Conductance 2.9 Free Energy and Equilibrium 2.10 Dissociation Constants 2.11 Ionic Conductance and Mobility 2.12 Osmotic Pressure 2.13 Diffusion (Flick's Law)

15 15 17 19 19 20 20 21 22 23 23 24 24 26 26

3. Systems Description: General Outlines of Basic Approaches 3.1 Electrodialysis 3.1.1 Performance Characteristics 3.1.2 General Purpose Processor 3.1.3 Additional Details for Appropriate Application - Desalinator for Small Boats v

1 1 9

29 29 31 33 36

vi

CONTENTS

3.2 pH Control: Analytic Development 3.2.1 Introduction 3.2.2 Some Technical Background 3.2.3 Sample Processes for pH Control 3.2.4 Application Possibilities 3.2.4.1 Swimming Pool Water 3.2.4.2 Cooling Towers 3.2.4.3 Regeneration of Ion Exchange Resins 3.2.5 Current and Electrical Energy Requirements 3.2.6 Shielded (Limited Ion Access) Positive Electrode Operation 3.2.6.1 Double Barrier 3.2.6.2 Close Spacing 3.2.6.3 Porous Barrier Design 3.2.6.4 Etched Electrode Surfaces 3.3 Biociding Technology 3.3.1 Electrolytic Production of Free Halogens 3.3.2 Chlorination Process Description 3.3.3 Bromination Process Description 3.4 Ion Exchange Resin Regeneration System 3.4.1 General 3.4.1.1 Present Regeneration Methods 3.4.1.2 Electrochemical Regeneration Method 3.4.2 Equipment Comparison 3.4.2.1 Performance Characteristics Comparisons 3.5 Metals Reclamation 3.5.1 Electrochemical Process for the Removal of Iron in Acid Baths 3.5.2 Technical Approaches 3.5.3 Technical Approaches 3.5.4 Laboratory Feasibility & Data Study Suggestions

39 39 39 42 45 46 46 46 50 51 51 52 52 52 55 55 57 61 62 62 63 65 65 66 71 71 72 73 76

CONTENTS

3.5.5 3.5.6

Experimental Methods 3.5.5.1 Approach B Tests 3.5.5.2 Approach A Conclusions & Recommendations

Mathematical Analysis & Modeling Electrodialysis Systems 4.1 Electrodialysis: Descriptions and Definitions 4.2 Basic Assumptions and Operating Parameters 4.2.1 Electrolytic Conductivity 4.2.2 Solute Concentration & Electrical Conduction 4.2.3 Electric Charge Equivalence 4.2.4 Coulombic Efficiency 4.2.5 Coefficients of Performance 4.3 Parametric Analysis: Flow-Through Configuration 4.3.1 Performance Analysis of Electro-dialytic Systems, Part I 4.3.1.1 First Approximation 4.3.1.2 Design Assumptions 4.3.1.3 Equation Development 4.3.1.4 Resistance of a Cell 4.3.2 Further Definition Of Terms 4.3.2.1 Average Current Density 4.3.2.2 Entrance & Exit Current Densities 4.3.2.3 Water Flow Rate in Processed Chamber 4.3.2.4 Solute Concentration Along the Length of the Cell 4.3.2.5 Figure of Merit 4.3.3 Numerical Evaluation Program 4.3.3.1 Second Approximation, Part II 4.3.4 Multiple Cells in Parallel

vii

76 79 81 91 95 95 102 102 106 108 109 109 110 110 110 111 111 112 114 114 115 115 116 121 122 123 127

viii

CONTENTS

4.3.5 4.3.6

4.4

4.5

4.6

4.7

4.8

General Characteristics Total Electric Current through the Electrodes and Membranes 4.3.7 Coulombic Efficiency Variation 4.3.8 Further Considerations Flow-Through Design Exercises 4.4.1 Exercise #1 4.4.2 Exercise #2 4.4.2.1 Predetermined Independent Variables 4.4.3 Exercise #3 4.4.4 TDS Removal Rate Capacity 4.4.5 Stacked Cell Configuration 4.4.6 Expanded Analysis Batch Process Analysis: Re-Circulating or Static Water Processing System 4.5.1 Coulombic Efficiency 4.5.2 Single Cell Analysis 4.5.3 Single Cell - Special Case 4.5.3.1 Ohmic Energy loss and Water Temperature Rise Design Exercises for Water Re-Circulation Systems 4.6.1 Exercise #1 4.6.2 Exercise #2 Cell Potential and Membrane Resistance Contributions 4.7.1 Membranes 4.7.2 Electrodes 4.7.3 Opposing Voltages Diffusion Losses of Ions and Molecules Across Membranes

System Design Exercises & Examples 5.1 Electrolytic Generation of Bromine and Chlorine: Design Procedures 5.1.1 Design Geometry Comments 5.1.1.1 Example

127 127 130 138 138 140 142 142 144 145 146 146 149 154 155 157 158 160 161 162 163 164 164 169 171 177 177 184 188

CONTENTS

5.2 5.3

Simple Estimate of Capital Equipment and Operating Cost of Electrochemical Desalination Apparatus Cost Estimates Outline for an Electrodialysis De-ionizing System

Applications Discussion 6.1 Demineralizer: Electrodialysis 6.1.1 Advantages of Electrodialysis 6.1.1.1 General Characteristics 6.1.2 Desalination System - Module Specifications 6.1.3 Performance Characteristics 6.1.4 Cost Factors 6.2 Reseidentialwater Softener 6.2.1 Product Design Description 6.2.2 Physical Description of the System 6.2.3 Operation 6.2.4 Design Example 6.2.5 Competitive Methods 6.3 Electrical Water Processor Portable Design 6.3.1 Present Solutions 6.3.2 Operation of an ED System 6.3.3 Design Prototype 6.3.4 Description

ix

189 191 195 195 196 196 196 199 201 202 202 203 205 205 210 211 211 212 212 214

Appendix A: Some Physical Constants and Conversion Factors

217

Appendix B: Conductance and Solubility B.l KC1 Ionization Constants

219 219

Appendix C: Feeder Tube and Common Manifolding Losses

225

Appendix D: Variable Current Density D.l Current Density Variation

231 231

x

CONTENTS

Appendix E: Mathematical Analysis: Water pH Control Cell and Ion Exchange Resin Regeneration E.l Analytic Approach E.2 Special Case Evaluation - No Resins Present in System E.2.1 Non-Constant Electrochemical Generation Rates for H + and OH" E.2.2 Dimensions and Units E.2.3 Variable Electric Current Densities E.3 Estimation of Resin Constants E.4 Electrolytic Resistance of the System Water E.5 Solution of the Simultaneous System Equations E.6 Sample Solution of Operating System

235 238 248 250 254 256 257 261 263 268

Appendix F: Industrial Chlorination and Bromination Equipment Cost Estimates E l Bromination Equipment List F.2 Capital Cost Analysis E3 Operating Cost Analysis F.4 Conclusions and Comments

271 275 277 280 281

Appendix G: Design Mathematics in Computer Format G.l Case A G.2 Case B G.3 Case C

285 286 294 299

Appendix H: Mathematics for Simple Electrochemical Biociding

303

Bibliography Index Also of Interest

309 311 313

Preface In recent years, the awareness of water needs and processing requirement has become an increasingly important topic. As the earth's population increases the demand for "clean" water has become an even larger factor in residential as well as industrial and commercial costs. There are now almost no natural water sources that do not require some purification of one form or another to render them potable sources. If the water impurities are ionic in nature, i.e., inorganic salts such as sodium chloride, calcium chloride, or iron sulfate, or inorganic acids such as sulfuric and hydrochloric etc., then the most effective method of moving these components about is by electrochemical means. Most substances dissolved in water do not lend themselves to be removed by filtration, as usually such ionic materials have been leached out of the ground supply. If the contaminants are non-ionized organic substances in solution, then there must be other means for their removal, and we will not treat the subject of their removal by chemical, distillation or filtration means. The technology that will be described here is well known but it is hoped that some of the quantitative and systems fabrication aspects covered here will contribute to the increasing practicality of electrochemical methods in water treatment applications. Much of the methods discussed here are a direct outgrowth of our research and development work in energy storage. Electrode design and construction methods for single and multiple cell devices were first addressed in the development of energy storage cells and multi-cell modules.

XI

xii

PREFACE

As the title of this book indicates, the following pages represent a summary of the results of a number of years of R&D effort directed to a better understanding of some basic processes for water treatment as well as the development of practical methods for design of useful hardware. The information contained herein is presented in an informal manner, in much the same fashion that it was generated in the laboratory studies. It is sincerely hoped that this compendium of technical notes will add meaningfully to the body of information associated with electrochemical approaches to water treatment, and will encourage others to pursue these avenues more intensively. Except for the review of some very basic mathematical relationships associated with chemical and physical processes, involving electric potential, molecular diffusion and solution pH, all of the material presented in this book is original. The major purpose of this book is the presentation of a body of analytical and design information that the reader may find useful in the exploration of electrochemical technology as applied to water processing. Hopefully, the contents of this text will encourage and promote further development of electrochemical processing systems for consumer as well as industrial and commercial applications. In our attempt to accomplish this end, the book has been organized into three main sections. They are: 1. General description of electrochemical processing and their application areas 2. Mathematical analysis of operating systems for design and optimization purposes 3. Design examples and procedures. Most of the experimental work on various water treatment projects was done at GEL/TRL during the period between 1978 and 1988. A short introduction to some very important and basic concepts is included at the beginning of the book and

PREFACE

xiii

in Chapter 2 as a convenient review for the reader. The Appendices are also offered for some additional analytical and ancillary application information, without disrupting the arguments and descriptions in the main portion of the text. The thread of the discussions to follow is to first identify the nature of the technology and to describe its various formats and uses. Then the book proceeds to develop an approach to mathematically treat the essential parameters associated with the principle mechanisms of ionic transport, electrolysis and diffusion. These developed mathematical tools are then applied to the preliminary design of a number of systems, serving as exercises for those that wish to carry on with their application in practical water processing problem-solving. The fundamental equations that evolve in the analyses are listed in a reasonably convenient and accessible form so that the reader can place them into appropriate computer software programs for easy solution and graphing of data. I have tried to establish an obvious rationale throughout the book to minimize confusion and the obscuring of direction and purpose. The Technology Research Laboratories, Inc. sponsored most of the analytical and experimental developments presented in the following pages. Some of the laboratory hardware and prototype systems were part of development projects funded by other industrial organizations for evaluation purposes. Our purpose in writing this book is to present sufficient basic applications and information about electrodialysis so that the reader is able to develop his own designs and approaches to solving water management problems. We will treat a number of forms of electrochemical water treatment from pH control to desalination along with their many application potentials. This book is concerned with the development of the basic principles and engineering design aspects of electrochemical

xiv

PREFACE

water processing. The intent in writing this volume is to serve as handbooks for the further development of related products. It should provide most of the necessary physical and chemical background to enable the reader to proceed with his own mathematical computations and engineering designs and mathematical computations. He should be able to arrive at sizes, performance characteristics and some preliminary cost factors on the basis of the information presented in this volume. No attempt is made in this book at covering the entire field of water treatment or reviewing the various competitive technologies associated with these methods. There are many excellent texts that have treated these subjects extensively, and have reviewed and summarized the work of numerous other investigators. There have been many texts that present excellent reviews of the state of the art in electrodialysis, reverse osmosis, filtration and distillation systems. Some of the specific terms and physical constants employed in the analytic approaches are covered in Chapter 2. However, it is assumed that the reader is familiar with the basic concepts of physical chemistry and elementary inorganic chemistry. For those who are not so well versed in these scientific disciplines, the resultant equations developed in Chapter 4 and elsewhere in the book are still useable for purposes of calculating the design parameters for the water processing devices under discussion. TRL, Inc. has performed the background work, a portion of which is covered by this book, over a ten-year period. The author, with over 35 years of research and development experience in related technical areas, has been principle investigator in most of the work represented in these pages. Water treatment with the minimal use of chemical reagents will increasingly become the goal of most systems in the future. In some instances, the elimination entirely of chemical agents is possible.

PREFACE

XV

A family of systems that provide means for controlling pH, biocide level and dissolved solids concentration in water have been studied as a result of the many years of electrochemical developments in the energy storage area at TRL. A larger portion of our attention here is devoted to the direct removal of dissolved, ionized materials in water via electrochemical (electrodialysis) separation. As a final comment, it is important to note that this book is concerned primarily with methodology rather than the specifics of any one design or system configuration. Very little empirical data or materials' properties information is contained herein. The specifics of component characteristics such as membranes, electrodes, and materials properties will be treated in another reference that is presently in preparation. This future text will also contain some empirical data on system performance as well as design and fabrication methods. Many application possibilities are still available that are eminently suitable to electrochemical techniques. I would like to express my sincere appreciation to Donald Morris, who performed many of the experiments and prototype design and fabrication, and for his invaluable assistance in organizing this information. Much of the discussion on biociding processes, and especially the sections that treat large-scale industrial water cooling systems was prepared by Catherine Middelberg as part of an application study at TRL, Inc. Special gratitude is due to my wife, Min, for her encouragement and endurance with the labors of organizing the technical material for these volumes. Ralph Zito Port Orange, FL January, 2011

Acknowledgements The work represented as a summary of laboratory investigations, as well as the design, building and field testing of potentially viable and useful water treatment products took place over a period of a few years at TRL, Inc. in Durham, NC. A number of different talents of various individuals participated in critical manners to bring about the net results contained in this book. Only a few of these people can be credited here. It required not only some scientific perceptivity, but also engineering and fabrication knowhow to persistently pursue these projects. It is hoped that some of these developments and knowledge gained by the activities of these people will be useful to others in the commercializing of systems in the future. Generally, water must be treated in some manner due to the many problems that beset its use in applications where "uncontaminated water" is critical to our civilization. Among the numerous contributors to this project, I would like to particularly cite Dale Jones for his steadfast support in overcoming some difficult situations, Don Morris for his unparalleled contributions to design and hardware fabrication. And many thanks to Patricia Pearson and Sara Tortora who provided order to the laboratory; orderly enough to maintain the necessary continuity for any work to succeed with a staff of usually over a dozen individual contributors. Also, without their imposed discipline and encouragement we almost certainly would not have completed these tasks.

XVll

Introduction Water has been in plentiful supply on this planet since it long ago cooled down and the oceans were formed. Despite the fact that over 70% of the Earth's surface is covered by water, the water needed for various life sustaining purposes is either unavailable at the required locations, or it is too contaminated for practical use. Obtaining fresh water is almost invariably a costly endeavor. The natural pools and lakes have gone to a great extent, relatively free of contaminants, and scattered everywhere in the communities of this country. High population densities coupled with increased demands and pollution by industry and residents have left these idyllic scenes far behind us. Now, we must either transport water from a few remaining sources at higher elevations such as melting snow on mountains or lakes at or near mountaintops.

Roman cross country aqueduct. XIX

xx

INTRODUCTION

The ancient Romans took advantage of such sources in Europe by building aqueducts (viaducts) to provide water to remote farms and villages by gravity feed. They spanned obstructions and valleys with gradually diminishing height to sustain the driving force of the mass of water being transported. There was little expense beyond the initial capital investment of the masonry, and of course, some continuing maintenance costs. These facilities were durable and had very long lives, as is evidenced by their existence today as functioning water lines. Today, we must resort to other transportation mechanisms other than gravity. Now, pipelines with water pumps are more common, but trucking and shipping is even used in some areas where water is critical. Such means of provision are quite costly. The only alternative to transporting water of good quality to other places of need is a purification or decontamination process of some sort. These methods, too, are costly depending upon the extent of processing necessary and the source of materials and energy As in the case of transportation, energy for separation is necessary to produce useable water from contaminated sources. There are only a limited number of mechanisms that can be employed to remove unwanted substances from water, whether they are dissolved or simply as particulate matter in suspension. In the latter case, filtration of one form or another or sedimentation can solve these problems. For those materials that are in solution, other methods must be employed, such as reverse osmosis, electrodialysis, or the old standby, distillation. Both distillation and RO will remove all solid matter in solution regardless of whether they are organic in nature, or inorganic, ionic materials. Distillation is the oldest method in existence, but it can be quite inefficient in terms of energy required per unit quantity of condensed water produced. Maintenance of boilers and evaporators can also be costly because of the solids left behind. The same is

INTRODUCTION

xxi

true for ED systems whose micro-porous membranes can become clogged with solid matter. ED is a cleaner system in the sense that no solids are collected as such, but the process will only remove substances that produce ions when dissolved so that they will respond to electrical forces for the separation process. Hence, ED is limited to removing only inorganic materials. Depending upon the nature of the situation one or more of these systems might be employed in treating a single body of water. It should be kept in mind that a minimum amount of energy is required to extract dissolved materials at the very least equivalent to their heats of solution. However, in practical systems that amount of energy is usually quite small compared to the dissipative factors of electrical resistance in ED devices, and mechanical resistances of RO devices. Desalination of sea water presents the greatest of all problems because of the very high concentrations of salt involved. Depending upon the form of energy available at a particular location where the fresh water supply is a problem, one of the preceding approaches is usually employed. If, for example, solar energy or cheap fuels are in great supply, distillation may very well be the method used in that location. Desalination should be receiving greater attention as a means of providing "fresh water" especially because seawater is so plentiful, and the demands are increasing so rapidly. The future will undoubtedly bring more conservation measures, but we may still be hard-pressed for better solutions to this ever-increasing problem. In recent years, especially, many devices and gadgets have been offered for sale on the open market, claiming to be capable of "purifying" water for drinking and cooking purposes. Some of these offered products are, indeed, genuine and perform as advertised. Products such as those based upon the use of ion exchange resins and distillation are based upon real science.

xxii

INTRODUCTION

However, numerous electrical devices on the market do not perform the tasks of which vendors claim they are capable. For example, ineffective devices include those that supposedly operate on the basis of some sort of magnetic field imposed on the water system via coils of wires that "polarize" or otherwise change the character of ionic species. These products are not based upon any known mechanism that would, in fact, separate unwanted materials in the water supply, or prevent mineral deposits from accumulating on the insides of pipes or hot water tanks in home or industrial water systems. There are some simple rules that one can follow to determine whether these proffered products do indeed operate. One should consider the amount of energy required to remove a given quantity of maters, i.e., dissolved substances, from a given volume of water. Generally one can employ such simple estimates to determine the probability of successful operation. Most of these important issues are covered thoroughly in the text to follow. There isn't much question about the importance of water in every facet of our lives, specifically the necessity for good quality water. The issues covered here are largely concerned with the removal of unwanted dissolved substances in water—substances that are ionic and are usually inorganic salts, acids and bases of soluble materials. The most common of these is salt, i.e. sodium chloride, because of its abundance and global availability. Various competing methods for removing materials of this nature are reviewed, but the main emphasis is electrochemical approaches such as electrodialysis. Much of the book is an analysis of the performance, efficiency and configurations of these types of systems. Some information is provided about the materials of construction, but the main theme of the book is analytic in format.

Electrochemical Water Processing by Ralph Zito Copyright © 2011 Scrivener Publishing LLC.

1 Water Contaminants and Their Removal 1.1

Introduction

This book is intended both as a tutorial presentation of basic principles of electrochemical water processing as well as a short working manual for the design and operation of electrochemical deposition cells and for electrodialysis devices. Water quality for direct and indirect human uses has always been an important concern in the past, and continues on into the future. With the ever-increasing concentrations of population centers and the demands of the industry, that concern is growing continuously throughout the world. The conditions that determine acceptable water quality are very dependent upon the use to which the water is intended. The factors involved are numerous, to say the least, and range from one high purity extreme to non-potable irrigation water. Water intended for farm 1

2

ELECTROCHEMICAL WATER PROCESSING

irrigation, for example, can contain a high level of foreign substances at the level of thousands of parts per million as long as these are not damaging to crops. At the other end of the water spectrum is the need for super pure, or "polished" demineralized water for pharmaceutical or semiconductor production uses. Among the more common types of foreign materials present that determine the "purity level" of water are listed below. Their removal from the body of water for each type is identified. 1. Solid matter in agitation or suspension Removal by: filtration, decanting 2. Dissolved organic substances Removal by: distillation, adsorption 3. Dissolved inorganic, ionic substances Removal by: distillation, reverse osmosis (RO), ion exchange resins, electrodialysis (ED) 4. Bacteria and other living organism contaminants Removal by: reverse osmosis, chemical addition, heating 5. Gasses present, in solution or otherwise Removal by: heating, ED, RO, adsorption 6. Other liquids miscible or non-soluble Removal by: fractional distillation, adsorption, RO,ED The removal method selected depends upon the economics of the situation for the intended applications, the specifics of the contaminants, and other unwanted substances present in the water. In the future, water treatment with minimal use of chemical reagents will become the goal of most processing systems. This approach is stimulated by the desire to minimally disturb the existing chemical conditions in water, and to reduce the amount of chemical correction needed as one introduces a reagent to fix one problem, only creating

WATER CONTAMINANTS AND THEIR REMOVAL

3

another problem. In some water conditioning instances, the elimination of all chemical agents is possible. A family of systems is possible, which will provide means for controlling pH, biocide level and dissolved solids concentration in water. One such form of technology is the direct removal of dissolved, ionized materials in water via electrochemical, (electrodialysis), separation. In the ensuing pages, one such class of processes will be described in some considerable engineering detail. In some instances the need to introduce bulk chemical agents to the water system can be eliminated entirely. This system removes dissolved substances such as salts, mineral compounds, acids and alkalis through the application of an electric field impressed across an array of electrodes and ion selective membranes. An electric power source is only required as input to the water. No chemicals or consumable materials are introduced into the water system. Dissolved substances that are removed from the main body or mainstream of water are carried over into a waste water stream and eventually discarded. A maximum of only a small percent of the incoming water is "wasted" in this manner, and no materials are put back into the drainage that were not present initially. The waste- water just has a higher concentration of the same dissolved materials than it had when first introduced into the systems. There are only a limited number of other methods, which can be employed to perform this task. They are: 1. 2. 3. •

Distillation Reverse osmosis, RO Ion exchange resin beds Distillation is simply the evaporation of solvent from solids and other contaminants present in the original body of water. The disadvantages are the life of equipment, relatively high maintenance in cleaning residues

4


from evaporating surfaces (heat exchangers), and high temperatures and high-energy consumption. • Ion exchange resins operate on the basis of the displacement of one ionic species for another as a function of relative concentrations. A mixed resin bed (cation and anion), regenerated in the hydrogen and hydroxide forms, respectively, will remove all other species of ions upon passing through the bed, and replace them with hydrogen and hydroxide (water as net product). High quality water can be obtained via this method. However, problems include high cost of regeneration and contamination. Usually this process is suitable for bringing good quality input water to high quality (polished, ultra pure) water for pharmaceutical and semiconductor uses. • During osmosis, solvent passes through a semipermeable membrane separating two solutions. Solvent—or water in this case—passes from the dilute solute side to the more concentrated side. This migration of solvent molecules to the concentrated side will continue until the solute concentrations are equal on both sides of the membrane. Reverse osmosis or migration of solvent to the dilute side can occur if a sufficiently large hydraulic pressure differential is established across the membrane in the appropriate direction (see Chapter-2, Section 2-12). RO is a system in which the water (solvent) is forced through the membrane, leaving behind ionized solutes as well as solids and organic materials. This filtration aspect of the process is an advantage in terms of ridding the water of most of the unwanted contaminants. However, the problems of membrane damage and clogging or blocking and fouling are significantly increased.


5

The electrodialysis, (ED), method offers some distinct advantages over all three of the above alternative methods. A comparison of attributes shows these to be some of the distinct benefits in practical use. Problem areas with Distillation a. b. c. d.

Costly in energy consumption Small systems are usually very inefficient Maintenance and scale accumulation problems Low production rate of water

Problem areas with Reverse Osmosis a. Fouling of membranes b. High pressure requirements for high TDS differences and their potential hazards. c. Costly systems d. Large size systems for large water flow rates make them impractical for many consumer applications Problem areas with Ion-Exchange Resins a. Resins are costly b. Need for regeneration is inconvenient and costly c. Two resin bed systems require use of hazardous acids and alkalis for regeneration d. Practical considerations of these de-ionize systems render them impractical for consumer and many commercial applications Problems and Limitations of ED Methods 1. No filtration provided for particulate matter 2. ED will separate out only ionized chemical species, no organic 3. Possible corruption of membrane by crystallizing materials within the membrane structure

6


The ED system employs long-term, inexpensive electrodes in conjunction with durable membranes that make for a low capital cost apparatus. The equipment operates at standard conditions of temperature and pressure, and requires no special precautions regarding quality of incoming water. The system can be designed to handle a full flow of water on a "once through" basis or it can be made very small and used in conjunction with a storage tank as a batch processor. Because the system is very simple in structure and operates through direct input of electric power, it can be made very small for portable applications. Surprisingly, ED is not employed in many areas where its applicability has distinct advantages. For various reasons of cost, unavailability of durable ion exchange membranes, and perhaps complex manufacturing requirements in the past, ED is not as popular for desalination and demineralizing as reverse osmosis, RO, systems. In a similar fashion, RO and cation exchange resin bed methods have been preferred for performing the tasks of water softening. Residential water softening presents an interesting example of a use where one method almost exclusively predominates the field. Cation resin bed devices have become the primary choice of the industry. RO and ED are competing systems that offer dramatic advantages with regard to the condition of the emergent, treated water over cation resins for softening water. Neither RO nor ED requires the consumption of sodium chloride for its operation resulting in brine waste water effluents into the ground water table. Softening by cation resins results in virtually the same concentration of dissolved substances in the treated water, except that sodium ions have been substituted for the unwanted "hardening cations" such as calcium and iron. RO and ED systems soften and demineralize input water by the removal of both cations and anions, thus


7

lowering the total dissolved solids concentration without ionic substitutions. RO devices are presently available as small units for residential applications, processing a few gallons per day. A full-scale unit that has the capacity to handle peak flows of 2 to 10 gallons per minute, without storage tanks, would be too large, impractical, complex and costly for individual home use. However, ED systems can be fabricated with high flow rate capacities that would be relatively simple and low cost. It is the intention here to review the operation of ED systems and show how economically practical ED water treatment systems might be designed to solve more problems than they are presently employed. Perhaps we can encourage a closer examination of the salient features and applicability of ED and electrochemical methods in general. The main attractiveness of the ED approach to water treatment is its inherent simplicity and compatibility with ambient conditions of standard temperature and pressures. This simplicity of hardware is illustrated in the photograph below (Figure 1.1) of an early field test prototype water demineralizer. The principle component parts of the system shown in Figure 1.1 are listed below. Processing Module (ED stack of electrodes and membranes) Processed water reservoir Waste water reservoir DC electric power supply Circulation pumps (two) We hope to accomplish our goals by going through numerous design exercises, and by establishing straightforward analytical methods, as well as describing some approaches to the practical problems of equipment life, manufacturing designs and costs.

8


Figure 1.1 Water demineralizer system components.

We are concerned in this book only with electrochemical processes for the control of water quality. Filtration, active carbon adsorption, and reverse osmosis systems are not the subjects of this text. Our primary purpose is to present the results of research and development conducted and some of the resultant engineering prototypes, along with some useful analytic methods for the design and performance optimization of an array of electrochemical systems. These potential systems range in function from that of demineralizing and desalting streams or bodies of water to the generation, in situ, of oxidizing agents for biocide purposes. The latter would eliminate the need for introduction of biocides from external sources. Some basic physical chemical principles and electrochemistry are briefly reviewed in Chapter 2 to provide a convenient reference for the reader when going through the analytic portions of this book. For a thorough understanding or tutorial presentation of the physics and


9

thermodynamics of the relationships employed in the parametric developments, it is suggested that the reader consult any of the numerous and excellent texts on the subject matter. Some suggested sources are given in the accompanying Bibliography section. ED systems can be designed to handle full flow of water on a "once through" basis or they can be made quite small and used in conjunction with a storage tank. Many portable applications are also possible due to the relative simplicity of its construction and the absence of any need for non-standard ambient conditions, including high pressures or temperatures. 1.2

Technology, History, and Background

TRL, Inc. is an independent Research & Development company. It was formed in Cambridge, Massachusetts in 1971 for the purpose of developing its own proprietary technology in energy conversion and storage; energy storage for utility load leveling and standby power. Most of the laboratory work over the years has been in the area of electrochemical processes, electrode development for Redox and halogen batteries, along with research in the physics of surfaces and transport phenomena. During the 25 years of applied research and prototype designing, a body of electrochemical technology was developed that lends itself to many applications totally unrelated to energy storage, batteries or fuel cells. Figure 1.2 shows a typical test set up for evaluating single cell electrodialysis systems for either pH control or de-ionization applications. Pictured are power supplies, timers, TDS and pH meters along with small fluid pumps and data acquisition connections for computer filing. A resulting potential application for some of the developmental work at TRL is in the field of water processing.

10


Figure 1.2 Laboratory test stand for water electrochemical studies.

Water treatment via non-hazardous methods afforded by electrochemical processes is an attractive and sensible product area possibility. The laboratory proceeded to develop, field test and engineer a class of new products for pool and spa use (labeled SimPool systems), and manufactured systems for field testing over a year. Over that time period, systems were developed and successfully installed in over 100 pools and spas in various geographic areas of the United States. 1.3

Application Areas: Electrochemical Technology Water Processing

The following is a list of some immediate areas of application, which are possible with ED and electrochemical


11

technology. These product applications would make use of well-established hardware designs and know-how. 1. Electrochemical Generation of Free Halogen (chlorine or bromine) in "salted" water for the purpose of disinfection. Product types are: 1. Swimming pool and spa halogenators 2. Water cooling tower processing 3. Potable water supply treatment 2. Free Halogen, Hypochlorous or Hypobromous Acid Injection Electrochemical generation of halogens and halogen acids via concentrated salt solution are injected into electrode reaction chambers. "Salting" an entire body of water to be treated is thus avoided. This method is particularly suited to large bodies of water and in evaporative cooling water systems, which require frequent "blow-down". 3. pH Control of bodies of water without the addition of acids or bases from an external chemical source. Only electrical power input is required to perform the conversion process instead of introducing additional chemical species. Handling, transporting, and storing corrosive and hazardous chemicals is eliminated. This is suitable for: 1. Swimming pools and spas 2. Industrial water supplies for the industry 3. Cooling water for machinery 4. Reverse osmosis system pH control

process

4. Water Softening 1. Residential and commercial water softening systems use the precipitation of dissolved minerals as scale on the negative electrode surfaces.

12


A system employing a membrane separator will perform this function without the introduction of sodium ions into the water stream, as is presently done. 2. A counter top version of this system is possible, which would "purify" a small quantity of water at a time (perhaps 1 qt. to 1 gal.)- The design may resemble that of a typical drip-type coffee maker. 5. Oxygénation Oxygen bubbles can be generated for the purpose of replenishing the supply in relatively confined and stagnant bodies of water, including aquariums. Such a unit would have no moving parts, generate no noise or vibrations, last indefinitely and be quite inexpensive. Electrolysis of the water produces the needed oxygen. "Starved" electrode designs would have to be employed in this instance to guard against production of free halogens. 6. Ion Exchange Resin Regeneration 1. Regeneration of resin beds normally employed in residential and commercial water softeners requires the periodic replenishment of salt (sodium chloride) in a reservoir. Frequently unwanted sodium ions are then substituted for the heavy, hardness producing metals, which are removed from the water. An ED system can be substituted for the salt reservoir and hydrogen ions used to regenerate the resin instead of sodium. Electrolysis of the water through an ED unit produces the H + ions. Normal pH is restored by the proper use of the staged electrolytic unit. In this approach, no salt is needed and no sodium ions are produced in the water.


13

2. Regeneration of de-ionizer resin beds at the operating site via electrochemical means. This would eliminate the need to remove equipment and transport it to service depots. Costs would be reduced and operational interruption minimized; plus, there would be no handling of hazardous chemicals, such as concentrated acids and alkalis. 7. Organic Contaminant Control In some cases where unwanted organic molecular structures can be altered and rendered acceptable via high acid or alkali exposure, a pH control cell may provide a "local" environment to achieve the needed reaction without the introduction of chemical reagents.


2 Basic Electrochemical and Physical Principles 2.1

Introduction

There are innumerable excellent texts available that cover the various subjects of thermodynamics, physical chemistry, and physics for students and researchers in the physical sciences. No attempt will be made here to present in a tutorial fashion any of these disciplines. Instead, we will outline and summarize some of the more pertinent concepts and mathematical artifacts that are basic to the design of the mathematics of operating electrochemical systems. These principles are the foundations of all of the calculations and performance assessments we make, whether or not they are consciously employed. In order to benefit from the design information presented here, and to be able to maximize the use of the associated mathematical developments, it is recommended that the reader have a general background in physical chemistry and college level calculus. 15

16


When generating the design parameters for an electrodialysis system, or for any other electrochemical process that is related to the subject matter in this text, the descriptions of the systems are in terms of these concepts and physical parameters. The major question here is simply "what are we interested in knowing about the operation and performance of an electrochemical system so that we can optimize designs and make effective use of the system?" The primary issues are the following: 1. Statement of the water treatment problem to properly identify the method of solution 2. Effectiveness of the system as applied to the problem in terms of function 3. Cost of initial capital equipment 4. Operating costs including life of equipment, labor, maintenance, and power. 5. Secondary problems that must be addressed as a result of the electrochemical system application, such as waste water disposal. 6. Physical size of system 7. "Efficiency" of operation 8. Reliability of apparatus To answer these questions, it is necessary to go through a design procedure that addresses the specific problem under consideration. The design process involves performing numerous calculations and making certain assumptions and technical decisions, again, in terms of the particular application needs. The procedures compiled in this book are the results of a series of analytical approaches in conjunction with empirical data available from much experience with related electrochemical processes, as well as the vast resources in the general scientific literature of materials properties data. Certain basic physics and chemistry tools are needed to make use of this information.

BASIC ELECTROCHEMICAL AND PHYSICAL PRINCIPLES

17

What follows is a very brief review of the background concepts that underlie much of the work that has been directed toward these water treatment application processes. • Chemical potentials at surfaces of electrodes and membranes • Diffusion and transport of molecular and ionic species • Electronic and electrolytic conduction processes • Acidity or alkalinity of water solutions, pH • Chemicals produced at the electrodes Some of the tools and concepts employed in the development of the information contained in this book are summarized in this chapter. We will review some of the chemical and thermodynamic properties and concepts that are particular to those employed in the analyses developed here. They are: Ionic equilibrium

pH

and dissociation Activity coefficients

Ohm's Law

Ionic mobility

Coulombic equivalents

Chemical potential

Figure of merit

Oxidation and reduction 2.2

Acidity and Alkalinity, p H

The concept of pH was first introduced by Sorensen in 1909, and provides a convenient measure of the level of acidity of aqueous solutions. The term is defined as the logarithm or exponent to the base 10 of the activity, aH+. Thus, 10"PH = a„. M

or (

pH =-log

aH+=

log

1^

vVy

(2.1)

18


The activity, a, of ions as employed in electrochemical theory is the ratio or percentage of the active (effective) concentration for reaction participation versus the total ionic specie concentration present in solution. In order to develop the necessary relationships for design and analysis of operating system parameters, it is necessary to define and qualify certain other basic concepts such as free energy, equilibrium constant and ionic mobility. The following relationships and concepts can be found in innumerable texts on electrochemistry or physical chemistry. For the sake of convenience, these are the basic tools needed to quantify the performance of electrolytic processes having to do with H+ and OH~ ion production or ion transport in general. The purpose of this exercise is to establish the relationships necessary to estimate the amount of electrical current flow or electrical power to change an aqueous solution by a pH point—thus, the relationships between pH, electrode potential and concentration of hydrogen or hydroxyl ions. Pertinent to the subject matter of this book is the ability to estimate the concentration of H+ ions. In very dilute solutions the activities are equal to the ionic concentrations. In term of ionic concentrations scale, pH can be expressed as üH = - l o g | H + | = l o g r ^ - T

(2.2)

where \H+1 is the concentration of hydrogen ions. The scale is from near zero to 14. On this scale, 7 is neutral (equal numbers of OH~ and H+ ions), with 14 being maximum alkalinity or minimum H+ ion concentration. For example, a pH of 2.5 would correspond to a hydrogen ion concentration of log|H + | = -2.5 \H+1 = 3.16 x 10"3 gram-ions/liter

(2.3)


19

2.3 Activity and Activity Coefficients The term activity as introduced by G.N. Lewis was directed at measuring the escaping (vaporizing) tendency of a volatile liquid. The activity term, a, is defined by the equation (2.4)

AF = FA-F°A=RT\naA

for a solvent substance, A, with a dissolved solute present. In this case, a= p / p o , where p o is the vapor pressure of the pure solvent and p is the depressed vapor pressure of the solvent with a dissolved substance present. In the case of ions in solution, the activity is defined as the effective concentration of solutes. Ideally, the activity and actual concentration are equal. And indeed, as the dilution becomes infinite, they become equal. In an infinite dilution situation the expression

X = M^ = 1

(Z5)

a means a = activity of the solute molecule.

Activity coefficient, y, is defined as the activity divided by the molarity, m, of the specific ion, or a

(2.6)

r=— m

2.4 Equilibrium and Dissociation Constants For a binary electrolyte, i.e. a single solute consisting of two ionic components, A and B, the reaction AB^A+

+ B~

(2.7)

is the dissociation of the molecule into its oppositely charged ions. The equilibrium, or dissociation constant, K, is expressed in terms of the activity coefficients as B

K=^

-

a

AB

(2.8)

20


2.4.1

Degree or Percentage Dissociation

A more directly useful parameter for calculating performance of equipment is the degree of dissociation, ß, of an ionic solute. That would also enable us to better estimate pH values of solutions 2 . ß is defined as the ratio of the concentrations of the ionized solute species to the total solute concentration, or

__[£] +[ £ ] _ ß=

[AB]+[A+]

+

[B-]

a 9 )

In dilute solutions the activity coefficients can be approximated by aM~[AB] = c(l-ß) ~[A+] = cß

A+A+

a

a„-

(2.10)

[B~] = cß

where c = total concentration of the solute of ionic as well as undissociated molecules. Continuing on, we may see how the pH of a dilute solution can be approximated as pH = - l o g [ A + ] = - l o g [ H + ] = -log[cß]

(2.11)

on the basis of calculations made from the equivalent conductance, Ao, of completely ionized pure water. 2.5

Electrode, or Half Cell P o t e n t i a l

The electrical potential, E, of an electrochemical cell is simply stated as AF = -nß

(2.12)

where n = electric charge transfer per ion and f = faraday's number, 96,000 coulombs per gram, the equivalent weight of monovalent charge transfer.


2.6

21

C h e m i c a l Potential D e f i n i t i o n

In order to appreciate the origin and definition of the term chemical potential as a mathematical concept, other energetic relationships associated with chemical processes are outlined below. Internal energy, E, of a system is usually a measure of its energy state in terms of its thermal energy content and chemical or physical configuration. Or, simply the change, AE, in internal energy in going from one condition to another is AE = E 2 - E 1

(2.13)

Next is the concept of enthalpy, which is the sum of internal energy change and any change in volume or pressure of a system during some energetic transition by either internal means or input or extraction of energy by means external to the system. This is described as AH = AE + Apv

(2.14)

The free energy, F, of the system is another useful parameter employed in calculating chemical potentials of electrochemical cells. It is the sum F = E + pv-TS = H-TS

(2.15)

where S = entropy and T = absolute temperature. The concept of entropy is best described mathematical as ,_ CdT dS = , or T S=\ ^

(2.16)

where C is the appropriate specific heat of the system (substance), and Q is the quantity of heat transfer absorbed or expelled by the system.

22


Willard Gibbs introduced the concept of chemical potential as a means of identifying chemical energy changes in multi-component systems. The internal energy of a system with n-components is then described as dE - TdS - pdv + Hydnx + jL^d^ +

+ ^dnn

(2.17)

and *

K^aJ

2.7

(2.18) S,V,n

C o n c e n t r a t i o n Potential

If the standard electric potential, Eo, of a cell is defined as that potential when all reactants are at unit concentration, (activities are all unity), then RT E0=—InK

nf

(2.19)

and K is the quotient of the reaction products multiplied together, divided by the reactant concentrations product. This ratio of products is defined by the above as the equilibrium constant, and Eo can be considered the "half cell" potential of an electrode in equilibrium with that electrolyte. If two electrolytes, separated for example by a porous barrier, contain the same chemical species but at different concentrations, a potential is realized between two electrodes, each immersed in opposite solutions. That net potential, A£, is given as RT A£ = — - [(In K - In Q, ) - (in K - In Q2 )] nj

(2.20)


23

Simplifying we obtain "/

Oi

And, most frequently this can be further related to a difference in concentration of the reactant specie in the two electrolytes, or RT, c, AE = — I n 2 nf cx 2.8

(2.21)

Equivalent Conductance

This is employed for determining electrical conductivity (or resistivity) of working solutions of electrolytes in the processes associated with the ED treatment of water. The term equivalent conductance refers to the electrical conductance of an equivalent weight of electrolyte. It is found by multiplying the specific conductance, L, (mho- cm 4 ) by the solution volume, V, that contains 1 gram equivalent of solute, or A = VL = \000L/c

(2.22)

and c is the solute concentration in the actual situation. Units of specific conductance are usually cm2 equiv 1 ohm-1. These are useful in determining actual conductivity or resistivity for very highly diluted electrolytes. 2.9

Free Energy a n d E q u i l i b r i u m

The free energy change, AF in any process may also be represented as AF = nRT In K

(2.23)

The equilibrium constant, K, for a process described between two reagents and two reactants A + B-+C + D

(2.24)

24


can be given as KAK = KAB=^-

lAllB ! , or as \C\\D\ (2.25) acaD

The terms \A\, and |ß|, etc. are concentrations of the respective components, or the expression employing activities may be used. 2.10

Dissociation Constants

The dissociation constant, K d , of a solute electrolyte can be given as

«£^c (l-a)c

A \-a

(226)

where a is the fraction dissociated of the solute, and c is the stoichiometric concentration of the electrolyte. These terms are employed to compute equilibrium conditions for electrolyte systems.

2.11 Ionic Conductance and Mobility The movement of ions through a solution under the influence of an electric field is the basis of all of the subject matter in this book. Electrodialysis processes, pH control, electrodepostion of materials, transport of ionic and molecular species across porous and ion-exchange barriers are all results of moving ionic substances in an electrolyte by means of electric charge conduction. The mobility, u, of an ion in a conductive solution is defined as follows.


25

(2.27)

u = x\t-—\ dx

where x is the distance an ion moves per unit time, t, under an electric field gradient, d E / d x , of volts per unit distance. The specific conductance, L, of an electrolyte is certainly dependent upon mobility, concentration of ions (or charge carriers), or in general, (2.28)

L = k-u-Ni

where k is some constant of proportionality, and N. is the population density of that specific ion in terms of numbers per unit volume of solution. The conductance in terms of mobilities of cations and anions in solution is represented as follows. L = f(caua + ccuc)/1000

(2.29)

The terms c, above, are concentrations of anions and cations respectively. The equivalent conductance, A, also discussed above, can be related to the specific conductance as follows. A = f(ua + uc)

(2.30)

Transference numbers through electrolytes or membranes for the anions and cations are simply n = —-— na + n=l

and (2.31)

if the concentrations do not change significantly with time as the ions are transported out of an electrolyte region.

26


2.12

O s m o t i c Pressure

When a semi-permeable membrane separates two fluids with different concentrations of the same solute, the solvent transports itself across the membrane from the dilute to the concentrated side until the concentrations of the solute are equalized. The pressure necessary to prevent this migration of solvent applied higher on the concentrated side is known as the osmotic pressure. This is an important phenomenon to be considered under some circumstances in the design and operation of an electrodialysis cell because large differences in water solution concentrations are established during its operation. Conventionally the osmotic pressure is represented by the symbol %. The value of this pressure is given as 7c=mRT

(2.32)

where m = molarity of the solution. For dilute solutions, solution volume is essentially the same as the solvent volume. Hence, the relationship may be written in the form nRT n=^± = CRT

(2.33)

so!«

where C = concentration of the solute in moles per liter. 2.13

D i f f u s i o n (Flick's Law)

Diffusion because of thermal motion or agitation (random collisions) of molecules can be quantified by applying Ficks's law in the form described below. The number of molecules passing through or across an area, A, normal to the path of molecular migration is given as D = -QLaTAC

(2.34)


27

The terms in the above equation are: D = diffusion coefficient (specific diffusion rate) 1 = diffusion distance through the area, A AC = concentration change over the distance, 1 t = time interval over which the diffusion is measured Another form of the expression is simply ^

= D-A-^

(2.35)


3 Systems Description: General Outlines of Basic Approaches 3.1 Electrodialysis Electrodialysis as a water processing mechanism has applications ranging from separation of ionic species in industrial processes to the production of potable water from sea water and demineralizing water sources. The principles of operation are quite simple and straightforward to comprehend, apply and manipulate for the purposes of achieving certain application criteria. In essence, a dialysis cell consists of four basic components: two electrodes and two membranes. Electrodes are, of course, opposite in electrical polarity, and the membranes have ion-selective properties. The positive ion selective, which permits (+) ions to be transported more easily than negative ones, is usually fabricated of carboxylic group, or sulfonic acid compounds, and is referred to as a cation membrane. 29

30


The negative ion selective, which permits (-) ions to be transported more readily than positive ones, is usually an amine structure or styrene-quaternary ammonium base material, and it is referred to as the anion membrane. A cell is comprised of these three compartments. The middle compartment is the "processed water" compartment. Ionic materials are removed from the middle chamber to the two sides adjacent to the electrodes. The cation membrane is on the side of the (-) electrode, and similarly, the anion membrane is adjacent to the positive electrode. This general configuration is shown in Figure 3.1. A cell can be designed specifically as a desalinator to produce potable water from concentrated saline solutions such as

Cation membrane

Anion membrane

Positive electrode

Negative electrode Na+

cr

Waste water

Waste water

Processed water Figure 3.1 Single cell electrodialysis.

GENERAL OUTLINES OF BASIC APPROACHES

31

seawater. Such a unit may also be used to process hard water from ground sources as a demineralizer (water softener). TDS of water at 40,000 ppm is reduced to 50 ppm or less within one hour of operation at the rate of 10 gallons per hour. An ED system is usually comprised of a series of electrodialysis cells that separates dissolved ionic materials from processed water compartments into waste water sections by means of selective transfer through ion exchange membranes. The ions are transported across these membranes by passage of electric current from a dc power supply. The process of ion transfer is shown in a single cell of Figure 3.1. Sodium and chloride ions move out of the middle compartment into the two waste water chambers where they are expelled as effluents. Over a period of time, the center (processed water) chamber is relieved of most of the ionic materials in solution and eventually becomes drinkable water. Module waste water is returned to the sea. Water is desalinated in a "batch process" manner. The time required for the desalination is dependent upon the volume of the processed water reservoir. The process takes place without application of internal high pressures or high temperatures as in Reverse Osmosis and Distillation systems, respectively. Life of components is long and maintenance is low. Electric pumps circulate the water through water reservoirs (tanks). By means of a controlled, variable voltage power supply, dissolved substances in the water are reduced by a factor of 1000 to 1 per hour of operation. 3.1.1

Performance Characteristics

Some typical performance data and time dependence of desalination are shown in Figure 3.2 presented here on the basis of a 10 gallon reservoir. The system is a recirculation configuration in which the processed water is

32


Controlled, variable stack voltage 10 cell stack, 120 volts DC max

Figure 3.2 Recirculating water system small boat desalinator.

a predetermined volume and is circulated many times through the processor in the desalting operation. A typical application for this size system is for small boat water processing. Power input to the unit is variable and dependent upon the TDS concentration of the processed water volume at that time during the cycle. Operation involves filling the 10 gallon reservoir with sea water at the beginning of a cycle. The graph shows the manner in which the TDS of a 10 gallon reservoir of processed water diminishes with time. Initially, the input power level is highest because that is when the electrical conductivity of the processed water has the maximum values. As the TDS is lowered, the electrical current drawn from the Power Supply also diminishes and eventually approaches zero power consumption. This occurs when the TDS arrives at the level of high quality potable water. The power level is about 2 KW initially and rapidly sinks to only a few watts within 30 to 40 minutes under normal


33

circumstances. The total energy consumed during the entire time for desalinating 10 gallons of sea water is less than 1 KWH. Depending upon the source of electric energy, the operating cost for desalinating can be less than $0.01 per gallon. 3.1.2

General Purpose Processor

To provide the reader with some notion of the size and configuration of ED systems and their performance capabilities, the following example is offered. The unit in Figure 3.3 is a general purpose system capable of processing 10 gallons of water from 1,000 ppm TDS to 50 ppm within one hour. Electric pumps circulate the water through water reservoirs (tanks). Dissolved substances in the water are reduced by a factor of 10 to 1 per hour of operation. Depending upon the concentration of minerals in the water initially and the power supply capacity, the rate of purification can be greater even than 10 to 1.

Figure 3.3 Operational schematic diagram.

34


The ED system consists of four major components. They are: Processor Module DC voltage Power Supply Processed Water and Waste water Reservoirs Two Circulation Pumps Assembly of a working system on a platform can be accomplished within an hour if no special installation requirements are encountered. In Figure 3.3, the principle components of the ED system are shown connected in an operational manner. In this arrangement, water is introduced from a source into both reservoirs. After the tanks are filled, the pumps are switched on, and water is circulated through filters into the module. Table 3.1 gives some typical values for the weight and volume contributions of the different system components shown in Figure 3.3. Because the water flow is over the surfaces of the membranes, and not through them as in a reverse osmosis device, no filtering of particulate matter is involved in the ED systems as part of their operation. Hence, separate filters must be provided to perform that task, if needed. Filters-A

Table 3.1 ED component weight & size. Component

Weight

Size

Processor Module

22 lbs

20"h x 6"w x 4"d

Power Supply

12 lbs

5"h x 12"w x 6" d

Reservoir Tanks(s)

10 lbs

1.5 cubic feet

Circulation Pumps

8 lbs

Power Input

10 to 500 watts

10 to 50 watts


35

are mechanically porous solids gatherers to prevent accumulation of materials in the module and reservoirs. Adsorption type filters can be provided if necessary to remove unwanted odor- and taste-producing organic substances from the body of processed water. Water processed in this manner results in the equivalent of de-mineralized (desalted) water approaching the purity of distilled water. The graph below, Figure 3.4, shows the manner in which both the TDS and power input varies with time at constant 50 volts operation from the power supply. At $0.10 per KWH as the cost of electrical energy, the cost per gallon of processed water is less than half a cent per gallon. The system is self-regulating, as seen here, because the electrical resistance of the processed water rises toward the end of the cycle, shutting off any electrical current from the power supply.

TRL electric demineralizer performance standard configuration

1

I 30 Volts constant stack voltage operation 5 cell stack, 200 sq in electrodes

Figure 3.4 Performance data. Module has five series cells and operates at 50 volts.

36


Some of the pertinent system data are: Cell stack Stack voltage Maximum Power supply rating TDS readout Hydraulic connections 3.1.3

5 cells in series electrically and in parallel hydraulically 30 volts 500 watts (for 1000 ppm input water) meter & indicator light polyethylene or Tygon for maximum versatility

Additional Details for Appropriate Application - Desalinator for Small Boats

An eminently applicable area for ED is for desalination in the small and medium boat sizes. Technology Research Laboratories, Inc. has explored these possibilities and has designed and prototyped some systems specifically for this application. Figure 3.5 is a photograph of an early desalinator prototype. This unit will produce up to 10 gallons per hour of potable water in the range of 50 ppm of salt from seawater at 40,000 ppm. ED water desalination designed specifically for small boats offers a number of significant advantages. These are: • • • • • • • •

Self Contained No High Pressures (RO) Noiseless, No Vibration No Boiling Water Low Maintenance No Installation Economical Operation Portable

It operates at standard temperature and pressure conditions. There are no high pressure pumps, clogged membranes, or noise and vibrations with which to contend


37

Figure 3.5 TRL Processor.

as in an RO unit. The main components of a desalinator are shown schematically in Figure 3.6, and are listed as: Processor Module Power Supply Processed Water Reservoir Circulation Pumps Initially water is drawn from the sea and pumped into the reservoir as shown until filled to the level desired, and then both pumps circulate water. Processed water is circulated through the module numerous times during any one period of desalination. Waste water is slowly expelled back into the sea as the processing takes place. Filters are provided at the inputs to the module to extract any solids from the original water. If desired, the filter in the processed water tank circuit can be an adsorption unit for the removal of undesirable microorganisms or other organic substances.

38

ELECTROCHEMICAL WATER PROCESSING To the dc voltage power supply

^

Filters

Sea water Pumps

Figure 3.6 Electrolytic desalination system schematic diagram.

Table 3.2 Weight and size of system components. Component

Weight

Size

Power Input

Processor Module

50 lbs

28"hxl4"wxl0"d

10 to 2,000 watts

Power Supply

30 lbs

8"h x 12"w x 10"d

10 to 2,000 watts

Reservoir Tank

12 lbs

1.5 cubic feet

Circulation Pumps

10 lbs each

25 to 50 watts

A convenient size tank for most applications is probably about ten gallons, requiring about one hour of processing to produce less than 50 ppm potable water from the initial 35,000+ ppm source water. The table 3.2 lists the salient features of the system in terms of size and weight.


39

There are many other examples of ED hardware application, some of which will be discussed in some detail later in this text.

3.2 pH Control: Analytic Development 3.2.1

Introduction

The control and maintenance of pH of aqueous solutions is another important consideration in water treatment of virtually all bodies of water for industrial, commercial or personal use. The pH of water in industrial processes will determine reaction types and rates. In cooling systems, low pH will frequently result in unwanted corrosion of metal parts and heat exchange surfaces. Skin irritation of promotion of bacterial growth may be encountered depending upon the level of pH in swimming pool water. High pH tends to promote growth of algae and other microorganisms detrimental to health. High pH will also result in scale accumulation and microorganism deposits on working parts of chemical treatment systems or heat transfer surfaces. A method of controlling pH levels in aqueous systems is described herein and depends upon electrochemical reactions. This approach substitutes the introduction of electrical energy to the water system to generate and inject the H+ and OH' needed instead of the chemical addition of an acid or base. No external supply of chemicals are required or added to the controlled body of water. 3.2.2

Some Technical Background

Returning to our definition of pH (see Chapter 2, Section 2.1), we have pH = log

1 a +,

V H

(3.1) J

40


The activity, a, of ions as it is employed in electrochemical disciplines is the ratio or percentage of the active (effective) concentration for reaction participants versus the total ionic specie concentration in solution. The free energy of reaction, F, is represented in terms of activity as F = F0 + RT In (aA)

(3.2)

or substance, A, undergoing a change from its standard state to another. The Debye-Huckel theory provides for a relationship between activity coefficient, y, and ionic strength, u. This is expressed as log ft =0.51zi VM for a solution, y+ = 0.51z + z_ Vw = ^y+Y-,

(3.3)

where u = sum of the products of concentration of each ionic specie in solution times the square of their electric charge. Or,

u = y^cxz\

+ c2zf +

)

(3.4)

Now, since it has been established that for all cases a

=1(T 14 ,

-a +

H

and water dissociates as

OH"

H20-+H+

(3.5)

'

+ OH-,

(3.6)

the equilibrium constant, K, is represented as K=

\H+ OH~

(3.7)

HiO

and because the concentration of water is nearly constant: K

w = \H+

OR-

i4 l(T-14

(3.8)


41

Reference electrodes are employed experimentally to obtain values of pH from measurable electric potentials. The standard hydrogen electrode, with high purity H2 gas at 1.0 atmospheres on a platinum probe and surrounded by H+ ions with unit activity, is assigned a potential of 0.00 volts. This reaction is simply

Y2H2^H+ + e-.

(3.9)

A commonly employed pH measurement apparatus is the glass electrode. A solution of constant pH next to an Ag-AgCl electrode with a glass membrane separates the solution. The observed potential is measured between the glass electrode and the reference calomel, (Hg2Cl2) electrode. The voltage measured between these two electrodes will vary in accordance with pH or activity of H+ ions in the intervening solution. About 0.59 volts per pH unit is obtained at 25° C. It must be remembered that the pH is a measure only of the activity concentration and not the total available hydrogen ion concentration in solution. The latter value can be obtained by direct titration with a base or acid as the case may be with low or high solution pH, respectively. In some instances it is possible to work backwards from the pH measurements (if sufficient a priori information is available), to the values of aH+ and fl0H- and y+ and y_. Concentration, m, of the electrolyte could be estimated then from the equation, log y- -0.51-v/m - bm,

(3.10)

where b = some constant. Similarly, for dilute solutions, the concentrations, cT and c2, for a "two ion kind" electrolyte such as HC1, H 2 S0 4 , etc. may be evaluated from the above formula where

42


log y= log J ^ _ = log

{m^âl)

z 1 +c 2 z 2 + ...) 3.2.3

(3.11)

Sample Processes for p H Control

Alkalinity and acidity of a test body of water can be controlled via a single electrochemical cell with a separating membrane between electrodes as shown in Figure 3.7. In the illustrations that follow the membrane is a microporous structure (semi-permeable) that has no ionic transport preference. The H+ ions and the OH' ions are supplied by the water. However, other anions and cations must be supplied from some solute, which will enable a balance of charge on either side of the membrane. One excellent choice of solutes for this purpose is sodium sulfate, Na2SOA. Na2SOA is very soluble, and the sulfate radical will not be decomposed at the positive electrode. No chemical other than oxygen will be evolved at the (+) electrode. Effluent at high pH OH" ions

Low pH output H+ ions

*G-

Main water supply input Figure 3.7 pH control cell.

Oxygen

Hydrogen

S0 4 =

Na+

OH-

H+

-►

■€>'


43

Water is electrolyzed at the electrode surfaces, and the principle reaction at the negative and positive electrodes, showing the unaltered salt ionic components as charge carriers: Negative side - basic: Na+ + H20 + e~ ^Na++OH~+y2H2

(pHup)

(3.12)

Positive side - acid: SOl + H20^2H++SOl+y202

+ 2e-

(pHdown) (3.13)

The decomposition of water with the accompanying evolution of hydrogen and oxygen is essential to the generation of hydrogen and hydroxide ions in the pH down and up chambers, respectively. OH~ ions are generated on the negative side of the membrane with the attendant rise in pH. H+ ions are produced in the positive side, and pH is driven downward. Depending upon the function desired, the main test body of water is directed through the appropriate side of the cell, or more specifically, the electrical polarity of the electrodes is established on the basis of pH drive direction needed. For example, if we wish to drive the pH downward in the main body of water, the hydraulic system configuration shown in Figure 3.8 would be employed. Sulfuric acid, H2S04, would be produced in the positive (+) cell side along with the evolution of oxygen being expelled into the water stream and vents. The generated H2S04 enters the main tank and the pH is consequently lowered. A corresponding amount of base is generated in the negative (-) side, in this case as NaOH, and it is slowly discarded from the main water body in order to keep the net pH change in the downward direction. Because the membranes are not ion selective in the configurations discussed here, there is significant loss of the generated H+ and

44

ELECTROCHEMICAL WATER PROCESSING Filter

ItTW Switch H 2 S0 4 H 2 0

pH Cell

Jr-n-

'Til

H+ ion injection main body of water

-O

V

V

Water pump

€f

Power H supply

Figure 3.8 Single cell control - pH down.

OH~ ions by diffusion and transport via electric field across the membrane. These processes are outlined and analyzed in some detail in Appendix D. Loss of sodium (as basic NaOH), in this case, is replenished by make-up water impurities added to the system. Scale and heavy metal deposits on electrodes are removed by periodic electric polarity reversal at the electrodes, and are then subsequently eliminated in the drain. A simple switching and valving via timer circuit controls the polarity reversal and periodic discharge. A second example is shown in Figure 3.9 where the pH is driven upward in the main water system. Main flow is directed through the negative side of the cell, which is actually the same physical side of the cell is utilized, but the operating voltage polarity is opposite that of the first case above in Figure 3.8. Sodium hydroxide, NaOH, is produced in the negative compartment, and


45

Filter

Î

1

Hydraulic switch NaOH + H 2 0 pH cell

*

H 2 S0 4 + H 2 0

A

OH" ion injection

-O

Main body of water

V

V

Water pump

& Power supply

Figure 3.9 Single cell control - pH up.

H2 gas is evolved. NaOH entering the stream raises the pH of the water system. As in the above first case, to maintain a net upward change in pH, the acidic solution in the positive side of the cell is expelled as unwanted low pH effluent. Some provision must also be made for cleaning scale off the negative electrodes and out of the (-) cell compartments. The solid effluents must then also be filtered out of the main water stream. The filter shown in Figures 3.8 and 3.9 is connected through a manual application of an electrically actuated hydraulic switch, permitting periodic filter back-flushing and rinsing. 3.2.4

Application Possibilities

Of the various applications possible, the following are listed as the more obvious candidates. Some configurations will be offered that are appropriate for pH control.

46


3.2.4.1

Swimming Pool Water

Pool water is usually treated with a strong oxidizer such as CaClö3 as disinfectant and for prevention of growth of algae, etc. Continual addition of such agents results in everincreasing pH, and necessitates periodic addition of acid to lower pH to acceptable levels. Loss of Cl2 and Br2 from the pool water, with the corresponding increase of Ca(OH)2 and NaOH will push pH upwards and encourage further water haziness and promote scale accumulation. However, in some instances it may be necessary to raise the pH due to the natural water source conditions and acid rainfall. 3.2.4.2

Cooling Towers

Mineral deposits that accumulate in commercial watercooling towers as a result of addition of hard water to make up for evaporation losses in the cooling process will tend to increase pH of the system and further promote the deposition of scale and solid matter on the heat exchanger surfaces. 3.2.4.3 Regeneration of Ion Exchange Resins Regeneration of resin beds for de-ionizers, ion exchange apparatus and water softener systems is another important area of application. Figure 3.10 shows a schematic of a proposed system for a single ion (cation) resin bed regeneration process. Instead of the normal or conventional brine reservoir employed to recharge the Na+ ions for Ca+, Mg+2, Fe+3, Cu+2, etc.ions during the recharging cycle, a source of hydrogen ions may be provided. These H+ ions are supplied electrically from a pH cell as described in Figures 3.7 and 3.8. Initial H2SOA may be introduced into the water stream to initiate the process and supply charge carriers.

GENERAL OUTLINES OF BASIC APPROACHES Softened output for use

47

R Regenerative mode position N Normal operation mode position Mechanically coupled valves

Input from water supply source

Trickle drain of un-needed NaOH

Figure 3.10 Cation regeneration without effluent recycling hydraulic circuit ■ water softener.

As the regeneration cycle proceeds, some portion of the effluent from the resin bed that is rich in Na+, Ca+1, etc. ions is permitted to recycle through the pH cell to lower the electrical resistance and then supply positive ions for migration to the negative electrode for OH~ ion production charge balance. Make-up water added during recycling will also contain a significant amount of minerals as charge carriers, thus assisting the electrical conduction. Let us review the principle mechanisms of the process as follows: When the resin bed is "depleted", the cation sites are largely occupied by the unwanted heavy metal ions that were extracted from solution during the working or softening portion of the cycle. Step 1 - Regeneration mode is established by turning off the two-way valves to the main water so that the main supply is shut off and no longer flows through the resin

48


bed. Instead, water is caused to flow through the pH cell and resin bed as shown in Figure 3.10. Step 2 - Electric power, dc current, is applied to the pH cell with the electric polarity as shown. Step 3 - H+ ions (H2S04, etc., acids) are generated in the positive compartment and fed through the resin bed resulting in substitution of H+ ions for the Ca+, Mg+2, Fe+3, Cu+2, etc. ions, which at this time occupy the loaded sites. Step 4 - These re-dissolved minerals are flushed as drainage from the system. However, some portion may be re-circulated through the pH cell to provide charge carriers for generation of additional H+ ions. Most of the heavy metals passing through the pH cell will migrate to the negative side and be discarded as shown. Step 5 - As the process continues most of the cation sites will be occupied by H+ ions and the bed will be regenerated and made ready for the normal work mode wherein A and B are switched to open the main water lines and close off the lines to and from the pH cell. Step 6 - If the negative electrode has much scale accumulation the power supply, polarity can be reversed for a period of time between regeneration cycles to de-plate the scale and discharge the solids through the pH drain. Actually, the de-plating operation may take place during the regular regeneration cycle by having the electric current reverse for a period of time long enough to perform the needed clean-up action, but for a period that is still so much shorter than the standard regeneration length of time that the net effect on the total regeneration process is negligible. Another configuration is shown in Figure 3.11 in which both outputs from the pH cell flow through the resin beds. The anion resin bed is supplied with base, OH' ions, to regenerate the sites from Cr,CO^,SOÏ,etc. which were removed during the normal water treatment mode.

GENERAL OUTLINES OF BASIC APPROACHES Processed and de-ionized water for use

N

N

Cation resin bed Drain during regeneration

49

Anion resin bed NaOH

HoSO,

*®«-|

$>Note:

Input from water supply source

-e

u

pH cell

Figure 3.11 De-Ionizer system regeneration without effluent recycling hydraulic circuit.

Operation of the water system shown in Figure 3.11 is virtually identical to that of Figure 3.10, except for the existence of two regeneration loops, each through their respective resin beds. An electrode or "scale clean-up" mode may also be utilized as before, by voltage polarity reversal for brief periods of time. Application to a de-ionization system takes better advantage of the attributes of the pH control cell because both streams are directly utilized instead of using only the acidic side as in the water softener set-up. Also, residential, as well as commercial use possibilities for de-ionizers, are opened up due to the simplicity and low cost of on-site regeneration via electrolysis—without the necessity of obtaining and handling bulk acids and bases. Those areas where low sodium water is needed can be served very effectively by the de-ionizer systems in a low cost manner.

50


3.2.5

Current and Electrical Energy Requirements

To compute typical electrical energy requirements for the creation of a 0.001 N acid solution in 1,000 gallons of water, the following estimates can be made. According to the Faraday equivalent, about 96,500 coulombs, or 26 amp-hours of electrical charge will produce lgram equivalent weight of product. In a 1,000 gallon volume of water, a 0.001N solution of acid has a total of about 4 equivalent weights of acid. That would correspond to about a 100 amp-hour of electric charge, or about 500 to 1,000 watt-hours of energy if the electrolysis took place at a potential of 5 to 10 volts per cell. If the regeneration cycle takes place within a period of one hour, then a power source of about 1 kw is needed to perform the task; assuming 100% coulombic efficiencies. Let us now also examine what this total energy input to a water softener resin system would do in terms of its effectiveness at replacing heavy metal cations. If 4 equivalent weights of CaC0 3 or FeCl3, etc. were replaced at an average equivalent weight of 100 grams per molecule, then 100 ppm in 1,000 gallons of water is 0.8 lbs of mineral compounds, or about 4,000 grams. This amount is about a 4 gram equivalent weight of the materials. To displace the heavy metal ions in the resin bed by means of an acid wash would require at least a 4 gram equivalent weight of the acid or H+ ions. Generation of such an amount of acid as calculated earlier requires about 1 kwH of energy if the coulombic efficiencies are between 50% and 100%. Most residential water softeners will handle much less than the 1,000 gallons between regenerations. Hence, less than 1 kwH of energy is needed for most home use situations for the regeneration cycle.


3.2.6

51

Shielded (Limited Ion Access) Positive Electrode Operation

In many instances the water stream may contain a fairly high concentration of anions, such as chlorides and bromides that will produce chemical species other than oxygen at the (+) electrode when electrolyzed. These cited anions, for example, will result in free chlorine or bromine, respectively, at the inert (+) electrode surface under most electrolysis conditions. Operation of the pH control cell may necessitate minimizing these reactions that are dissipating to raising pH levels by decomposition of water and subsequent evolution of oxygen, e.g. 2Cr -> C/2 + 2e~ and

2Br~ -> Br2 + 2e~

(3.14)

because they will reduce coulombic efficiency for generation of H+ ions. Oxygen evolution is necessary for the production of H+ ions in the positive compartment. A method developed at TRL that has proven quite effective is the intentional "starvation of the positive electrode". To achieve this end, a means of inhibiting the flow or diffusion is always present from the dissociation of water, hence the availability of Cl~ and Br~ ions in favor of the OH'. These methods are listed below. 3.2.6.2

Double Barrier

This is an extra or second separator between the cell that compartmentalizes membrane and the positive electrode. A relatively stagnant region of electrolyte is established in which the Cl~ and Br~ ions are depleted and remain in low concentration. Hence, their availability for oxidation to their elemental forms at the positive electrode would be considerably more limited. Ion membrane or micro-porous membranes may also be employed as shown in Figure 3.12.

52


H+ ¡on, acidic output

OH" ion, alkaline output Membranes

+

-O

G-

Input water

Slow flow .

Normal flow rates

Figure 3.12 Starved posode pH control cell double barrier configuration.

3.2.6.2

Close Spacing

The space between the positive electrode and the cell membrane can be made very small. Low fluid circulation in this intervening space of low availability of specific ions can be achieved. See Figure 3.13. 3.2.6.3 Porous Barrier Design A diffusion-limiting barrier in intimate contact with the positive electrode can also be used to create a stagnant electrolyte region to achieve specie starvation conditions. Thick, almost sponge-like nonconductive polypropylene, unwoven fiber structure placed against the electrode will impede the availability of dissolved anions in preference to OH' ions provided by the water. See Figure 3.14. 3.2.6.4

Etched Electrode Surfaces

Use of carbon-polymer composites as electrode structures provides the opportunity to develop limited electrolyte availability conductive surfaces. The surfaces of the carbon composites are cavities whose walls are largely

GENERAL OUTLINES OF BASIC APPROACHES H+ ions acidic output

53

OH" ions alkaline output

Membrane

+

Electrodes

o

-0' Slow flow

Input water

Normal flow


H+ ion, acidic output

OH ion, alkaline output

Electrodes

'©-

-©'

Porous layer Membranes

Input water


non-conductive polymer materials with conductive carbon located deep within the cavities. This configuration results in recessed regions of electrolytic conduction into which migration of Or and Br~ ions is severely restricted. Their slow diffusion from the main electrolyte volume renders them less available as compared to the ever-present OH~ ions. This is illustrated in Figure 3.15.

54


Figure 3.15 Starved posode pH control cell etched posode configuration.

The rate of generation of Or, for example, at the positive electrode surface is not only dependent upon current density, but also upon Or availability at the electric current density demand rate. Thus, if the current density is high and the obstruction to Or migration to that electrode is also great, then the electrode will be starved for those ions, and more 0 2 will be generated via the electrolytic decomposition of water as the preferred reaction. Current efficiencies are greatly dependent upon population density of the specific ion in the immediate vicinity of the electrode. Experience with carbon-plastic electrodes has shown that the efficiency of C/2 and Br2 production on these electrodes diminished to very low values as the surface erosion (spallation of carbon particles) proceeded with the generation of oxygen during cell operation. The drawing of Figure 3.15 illustrates this effect by showing caverns of non-conductive plastic walls with conductive carbon bases at the bottom of the caverns and as the working surface of the electrode. Availability of 0~ ions from the moderate concentration in the surrounding solution is severely limited within these caverns.


55

hr ions OH ions

Posilyte line

+

&

K

Negalyte line

-€>"

-

Porous barriers

Input from the water reservoir Figure 3.16 Series bipolar array - separate manifolding.

An interesting illustration of acid/alkali generation is shown in Figure 3.16. This is an array of cells specifically designed to produce pH effluents of wide separation. 3.3 3.3.1

Biociding Technology Electrolytic Production of Free Halogens

Production of strong oxidants such as free chlorine and bromine by means of electrolytic processes is well known and has been employed for over two hundred years. In fact, an early method of producing sodium hypochlorite, NaClO, a strong bleaching agent, was by electrolysis of NaCl. The use of elemental chlorine and bromine as a biocide is widespread. At present the most popular method for chlorination of pools is the use of Ca(ClÖ)2 (calcium hypochlorite, HTH trade name) as a source of available chlorine for sanitization. Cost of the product is over $1.00 per pound.

56


Because the available chlorine is about 50% by weight, the cost of the biocide is over $2.00 per pound. This is a reliable standard of cost comparison for sanitizing small bodies of water. Larger water volumes such as cooling water for utilities, municipal supplies and commercial pool employ elemental chlorine in the form of compressed gas. In most instances, these reagents are produced in a manufacturing plant and transported to the use on site as elemental substances with all the attendant hazards and associated costs. However, it is possible to produce these substances at the end point of use by electrolytic means. The electrolysis process that takes place in a brominator electrode assembly is described as follows. Direct current flow through the electrodes and electrolyte with NaBr salt in solution generates these net chemical reactions at the electrode surfaces (see Figure 3.17).

Figure 3.17

o,

H,

Br-

Na*

OH-

H*


57

At the negative electrode: 2H++2e~ ^ H

2

t

(3.15)

At the positive electrode: 2Br~-^Br2 + 2e~

(3.16)

The total chemical reaction in the water is 2NaBr + 2H20 -> 2NaOH + Br2 + H21

(3.17)

Hence, free bromine is produced at the positive electrode and becomes solubilized in the water stream. Meanwhile, NaOH is generated at the negative electrode and slightly raises the pH level of the body of water. If the electrode separation is wide and the water flow rate between the electrode plates is sufficiently rapid, there are very few immediate secondary reactions. However, prolonged interaction of sodium hydroxide and bromine will produce the hypobromite, NaBrO, or 2NaOH + Br2 -> NaBrO + H20 + NaBr

(3.18)

resulting in an immediate 50% loss of Br2 back to the original NaBr compound. The hypobromite is as effective as free bromine for germicidal purposes. About 0.5 to 2.0 ppm of Br2 is adequate to ensure useable water condition. For a quantitative assessment of design procedures in constructing electrolytic halogen generators see Chapter 5, Section 5.1. This approach offers opportunities for sanitizing water at significantly lower costs. 3.3.2

Chlorination Process Description

Water disinfection through chlorination is practiced in industrial, municipal and private water supplies to destroy harmful water-borne organisms. In private and municipal waters, disinfection processes have been employed to mainly eliminate pathogenic organisms, while industrial

58


processes require the destruction of fouling, and sometimes corrosive, micro-and macro-organisms inhabiting their water systems. Though water disinfection can be accomplished through physical or chemical means, chlorination has achieved wide acceptance because of its low cost and adaptability in water treatment operations. This section will review chlorination and an alternative chemical treatment method of disinfection—bromination. It will also compare the capital and operating costs in a specific industrial application of open, re-circulating water systems. Bromine disinfection has been successfully employed in the private sector for pools and spas since the 1940s1, utilizing the organic compound, l-bromo-3-chloro-5dimethyhydantoin, known commercially as Aquabrom or Dihalo. More recently, industrial cooling water treatment with this biocide has shown the advantages of bromination over chlorination, though the operating costs were reported to be significantly higher 2 . This is attributed to the cost differential between chlorine gas and Aquabrom. However, when chlorine is not an acceptable treatment method because of safety and high maintenance costs, l-bromo-3chloro-5-dimethylhydroin has proved to be a cost effective treatment method as compared to other available or current technologies. Other combined forms of bromine and the chemical oxidizing agents necessary to produce available bromine residuals in water have been evaluated as disinfection treatment alternatives. These include chlorine—or persulfate-activated bromine salts. Each of these activation methods has limitations that hinder its use on a wide scale. However, more recently, TRL, Inc. has developed and fieldtested a bromination system based upon electro-chemical

See1,2 Appendix F references on page 283.


59

decomposition of bromine salts to produce free bromine residuals. For over two decades, the TRL involvement in bromine technology and electrochemical research in the form of zinc/bromine and redox batteries has initiated interest in similar processes for water treatment. As an outgrowth of this experience in halogen chemistry, a series of successful electrochemical disinfection devices have been constructed for swimming pools and spas. The following is a description of the economic and safety considerations and attributes that an electrochemical bromination system offers the water treatment field over conventional chlorination practices. Many industrial operations require cooling water to reduce process heat, such as in petroleum refining, power and steam generation chemical production, metals recovery and refining, air conditioning and refrigeration. Whenever water is used in a cooling system, microorganisms, plants and animals will reproduce and inhibit the rate of heat transfer through fouling, and decrease process flow rates as well as corrode process equipment. The goal of chlorination is to minimize microbial and macro-organism growth to avoid expensive system shutdowns for cleaning, repair and maintenance. Liquid chlorine is stored in this facility in a single chlorine tank car, and is unloaded through a flexible metal connection to the process piping under its own pressure. If necessary, air padding is provided to deliver the chlorine from the storage tank to the chlorination equipment. When a tank has been emptied, air purging for safe inspection and repair is also required. The system piping has both a liquid and gas header— one for gas withdrawal and one for liquid. Both headers feed the evaporator and are provided with expansion tanks in the event of liquid chlorine vaporization or hydrostatic pressure build-up in the supply line.

60


Liquid chlorine enters the evaporator, which is essentially a water-jacketed pressure vessel equipped with electric immersion heaters. The liquid chlorine absorbs heat through the vessel walls from the hot water until it boils at its vaporization temperature. The chlorine gas absorbs superheat as it is withdrawn from the evaporator through contact with the hot upper walls of the evaporator vessel. This prevents liquefaction in the process lines. The gas passes from the evaporator through a gas filter to remove impurities. The chlorine pressure-reducing valve is downstream of the filter, which further reduces the liquefaction temperature by reducing the chlorine gas pressure. This valve also serves as a safety device to automatically shut off the chlorine supply to the evaporator in the event that the water bath temperature falls below the set point. Downstream from the relief valve, the gas is drawn through the chlorinator, which is the flow metering and control equipment for the supply system. The components of this piece of equipment include an inlet gas pressurereducing valve, a rotameter, a metering orifice, a vacuum differential-regulating valve, a pressure-vacuum relief valve and an injector. The injector creates a vacuum, which draws the gas through the equipment and ultimately mixes the chlorine with water to supply the required dosage to the cooling water circuit. The inlet gas-regulating valve controls the density and velocity of the gas entering the chlorinator. The variable orifice meter controls the chlorine feed rate and is controlled automatically through chlorine residual feedback control. The rotameter is a local feed rate indicator. Because of the possible chlorine supply failure, vacuum relief valves are provided within the chlorinator to allow air to enter the system and prevent "suck-back" of water from the injector into the chlorine supply lines and to prevent a high vacuum rise within the chlorinator. As previously described, the injector mixes the chlorine gas with a water stream to provide the dosage required by the treatment system. From the injector, the chlorine


61

solution is applied at the designated points near the heat exchanger equipment and in the tower basin. 3.3.3

Bromination Process Description

It has been reported that bromine disinfection efficiency is lower than that achieved with chlorination. When tested under standard methods of disinfection evaluation, bromine, at 1.0 ppm was as effective as 0.6 ppm of available chlorine to control the bacteria (see E. Coli). This can be attributed to the higher oxidation potential of chlorine. However, one must be careful in its correlating ppm between chlorine and bromine by weight because the molecular weight of bromine is twice that of chlorine. Also, bromamines, formed when bromine reacts with nitrogenous compounds, are superior disinfectants in comparison with the corresponding chloramines. The dissociation constant, pK, for hypobromous acid is approximately one pH unit higher than the pK for hypochlorous acid. This is significant in systems operating over a wider pH range of 7 to 8.5. Recent field trials with Aquabrom (a bromine biocide) in open, re-circulating water-cooling systems have demonstrated bacteriological, algae and slime control, while minimizing the maintenance and safety hazards encountered with contemporary chlorination practices. While the only disadvantage to current bromination practices is chemical costs, TRL has developed an industrial bromination process with operating costs equivalent to or less than a chlorination process. The bromine vehicle to the process is the salt, sodium bromide (NaBr), which is stored in a feed bin. Capacities and sizes have been equated with a chlorination system to compare costs. The NaBr salt is intermittently metered to the solution tank and dissolved to a concentration of about 31b per gallon of water. This concentrated solution is metered to the cooling water circuit to maintain the salt at

62


about 1,000 ppm. Due to windage losses, it is estimated that a make-up of 50 lbs per hour of NaBr will be required, or approximately 15 gallons per hour of salt solution. Control of salt feed rate is maintained by the volumetric, vibrating feeder and bin load cell. Cooling water is re-circulated to the heat exchanger equipment or other points with the cooling tower recirculation pump(s). Though it is common to employ several pumps, it has been assumed that one pump will handle the entire flow of 100,000 gallons per minute. On the downstream side of the pump, the flow will pass through the bromination tank, equipped with 12 modules of ten cells each, with a five square foot cell area. Though 600 square feet of cell area is provided, only 500 square feet are required to maintain 2 ppm bromine residual. The cells are arranged so that 2 spare modules are available should a failure occur. Power to the cells is provided by a 150 amp, 240 volt supply. Current density to the cells is 100 ma per sq. in. From the bromination tank, the cooling water is distributed to its specific application points. Because the bromination tank is a pressure vessel, modules cannot be removed or repaired without bypassing the tank. For this reason the spare modules have been added.

3.4 Ion Exchange Resin Regeneration System 3.4.1 General Of the various processes employed in water treatment, e.g. reverse osmosis, dialysis, ion-exchange, direct chemical treatment, filtration, etc., the ion-exchange process offers the highest purity water in terms of de-ionized content, especially the mixed bed version. Cation and anion exchange resins such as those produced by Rhom & Haas, Diamond Shamrock, and Dow are employed to remove the respective ions from solution. In a


63

de-ionization apparatus there are either two separate beds of ion-exchange resins (one cation and one anion) or a single mixed resin bed. Our attention is directed here to the two separate resin bed system. Input water to be treated is first passed through the cation stack wherein most unwanted cations such as Ca++, Mn++, Fe+3, Mg++, etc. are removed and replaced by H + ions. Subsequent passage of the water through the second stack (anion resin) removes the negative ions such as C0 3 ~ 2 , Cl~, S 0 4 2 , etc., which are replaced by OH" ions. Thus, dissolved solids are removed from solution and replaced by water as the resultant product. After some usage, the resin sites become increasingly occupied by the removed ions in place of the H + and OH" in the cation and anion beds, respectively. As this replacement proceeds, the resin becomes less effective in removing the above ionic species dissolved in the input water and it must be "regenerated". Regeneration entails exposing the cation resin to a concentration of H+ ions (acid bath) and the anion resin to a concentration of OH" ions (alkali bath). Water processed in this manner is used for various industrial and science laboratory purposes as well as for potable water in some instances. Some of the reasons for removal of these dissolved solids include: • Improve taste characteristics • Reduce hardness and accumulation of scale • Remove coloration and staining properties of the water • Remove salt where low sodium levels are required in potable water 3.4.2.1

Present Regeneration Methods

Regeneration of resin stacks is presently accomplished by treating with appropriate chemical solutions (see Figure 3.18).

64

ELECTROCHEMICAL WATER PROCESSING Treated water Cation resin

Anion resin

Input water In use mode Input water

Acid

Cation

Anion

Waste

Water

Base

Regeneration mode

Figure 3.18 Functional block diagram conventional deionizer - on site regeneration.

Cation resins are washed in a 1 N H 2 S0 4 or HCl (sulfuric or hydrochloric acid) solution. Anion resins are washed in a 1 NaOH (sodium hydroxide) solution. These standard solutions are obtained by diluting much more concentrated reagents (30% to 50% concentration) purchased in bulk quantities. Regeneration is either performed at the water treatment site or at some regeneration station as a service to the user. In the first case, it is periodically necessary to transport the acids and alkalis to the water-processing site. In the latter case, transportation of resin tanks and dismantling and reinstallation operations are necessary. Transportation and handling of chemical reagents also involves the required procedures for hazardous materials and their disposal.


3.4.1.2

65

Electrochemical Regeneration Method

Resins can be regenerated electrochemically and on site or at established regeneration depots. The needed H + ions (acid) and OH" ions (base) are produced from the waste via the electric charge carriers within the resins as they are replaced by H + and OH - via electrolysis. Water is electrolyzed with the accompanying evolution of hydrogen and oxygen gasses to produce acid and alkali on opposite sides of an ion conductive membrane. This regeneration process can be implemented in numerous ways. They include: • Regeneration of resins during no-use times • Regeneration of a standby stack, which is made ready for interchanging at the appropriate times The acid and alkali electrochemically generated in the electrode assembly (conversion reactor) are circulated through the respective resin stacks during the regeneration cycle period. 3.4.2

Equipment Comparison

Component parts of the currently employed regeneration method are compared with the electrochemical system in Table 3.3. In conventional, on site regeneration, concentrated solutions are diluted and introduced into the resin stacks via eductor units. Because no concentrated reagents are employed in an electrochemical regenerator, no eductor, controller or programmer circuitry is needed as presently found in bulk chemical agent methods. Instead, a solenoid actuated hydraulic switching apparatus is employed. A timer or a water conductivity sensor can control the switching. Dilute acid and base are produced electrochemically in a hydraulic circuit in which the standby (resin stack being

66


Table 3.3 Conventional System

Electrochemical Regeneration

1 Cation Resin Bed

2 Cation Resin Beds

1 Anion Resin Bed

2 Anion Resin Beds

Controller Circuitry

Timer and / o r Sensor Switching

Eductor Heads

Manifold Assembly

Acid Reservoir

none

Alkali Reservoir

none

none

Electrolytic Unit

none

D.C. Power Supply

none

2 Circulation Pumps

regenerated) is situated. Input water is slowly passed through the regeneration circuit during the reprocessing. The water consumption rate can be significantly reduced in the regeneration loop by employing a recirculation circuit (2 optional pumps). 3.4.2.1 Performance Characteristics Comparisons There are a number of significant differences in the performance characteristics between the more conventional systems and an ECR approach. These are simply represented in the list of attributes given below. • • • • •

Reprocessing waste water is minimized Lower capital equipment costs Lower operating costs Reduction in maintenance and personnel time Less complex apparatus—greater unattended reliability • Minimum chemical hazard • No chemical reagents to purchase or transport


67

• No objectionable fumes or stored chemicals • No addition of chemical reagents to effluent Electric energy is the substitute for chemical reagents in an Electrochemical Regeneration (ECR) system. Hence, the consumption of electrical energy is greater in the ECR approach then in present systems. Two overall design versions of an ECR system will be briefly described here. They are: 1. Self contained regeneration configuration within the resin bed, (in situ regen system) (see Figure 3.19). 2. External regeneration system, which employs a physically separate electrochemical reprocessing unit (see Figure 3.20).

Processed water

Input water Cation

&

In use unit

anion

"l

(

r

'r

i

Cation

&

Power supply controller

anion

Standby (regen) unit

—►

■4

Waste

Waste No valvlng and switching show

Figure 3.19 Functional block diagram system A: normal operating configuration.

68

ELECTROCHEMICAL WATER PROCESSING Input water

Cation resin

Anion resin

In use water

y

Input water

Electrolytic regenerator

Power supply

J~L Cation resin

Input . water

13 Anion resin

Standby units version A

£5

0

Standby units version B

rft Cation resin

Anion resin

Waste

0

Waste

Power supply Figure 3.20 Functional block design system B: normal operating configuration.

In the first design, A, the entire resin bed configuration is revamped. Electrode assemblies, separator membranes and fluid manifolds all become an integral part of one total unit. This design necessitates a completely new system, and would make use of only some ancillary control devices of present de-ionizer systems. Design B necessitates only the replacement of the regeneration portion of present systems, which have an on site regeneration capability. The approach could also serve as a substitute method for regeneration stations in service companies where chemical handling problems need to be eliminated. Performance of the above two approaches are outlined below. System-A Two pairs of resin beds are employed. A resin pair for processing input water and a pair undergoing a slow electrochemical regeneration process. The regeneration time available could be a minimum of 20 hours or


69

as long as 100 hours. The reactor processor and the power supply are sized by the dissolved solids removal capacity of the system and the time available for regeneration. At the end of the "use" time, a simple switching of the in-useresin beds to the newly regenerated beds takes place, and the overall processes continues with minimal interruption. During regeneration of the "standby stacks" the necessary washing and back-flushing operations are carried out as they are normally conducted at present. Provision is made to exhaust the generated H 2 and 0 2 gasses. In most cases these gasses are discharged along with the waste water during the regeneration mode. A single pair resin tank system could be employed. However, unless there is a large non-use period of time available, the required regeneration equipment can become quite large and expensive. System-B This approach enables the use of most of the equipment presently employed in de-ionizers except for eductor heads, chemical reagent tanks and some of the controller and valving specially designed for the chemical treatment. The significant advantage of this design is its high degree of compatibility with existing ion-exchange hardware. Three versions of design B are possible. They are: 1. Recirculation of water through resin beds during regeneration via a small reservoir and circulation pump (see Figure 3.21). 2. Once through water flow with no recirculation (see Figure 3.21). 3. Only one pair of resin stacks is needed if the acid and alkali are electrochemically generated continually and stored in reservoir tanks for treatment of the resin bed during regeneration. In this manner, the equivalent of the present system is available wherein the regeneration process occurs periodically and may take only

70


Cation resin

Use mode

Anion resin

Processed water

Input water Cation

Anion

Ï—€ Acid tank

^

Regeneration mode Alkali tank

Power supply

_©

Figure 3.21 System B, Version 1.

Deionized water output

Valve

Deionizer in use

Water input

Deionizer in regen. mode

T^ Drain

Figure 3.22 System B, Version 2.

the standard time of about 1 hour. Circulation pumps and reagent storage tanks are required in this case (see Figure 3.22).


71

In the first of the above versions, water economy is maximized, and in the second version, the equipment is somewhat simplified, but the water consumption rate is considerably higher. The third case is essentially a direct replacement of the acid /alkali wash without the necessity of supplying, handling, diluting and chemically feeding reagents from an external source. All of the above designs use the same regeneration power supply and electrochemical capacity. As in the system A design, the H 2 and 0 2 gasses produced during the regeneration as a result of the electrolytic decomposition of water in design versions 1 and 2 are expelled with the waste water. Special provision must be made for discharging the gasses in design version 3 because there is no waste discharge or drain during the electrolytic production mode of acid and base in the reservoirs. The real potential merit of the ECR approach is in the reduction of materials handling labor and administrative costs along with the attractiveness of a cleaner facility. 3.5 3.5.1

Metals Reclamation Electrochemical Process for the Removal of Iron in Acid Baths

In this section we will explore the practicality of removing iron and other heavy metals from solutions such as "spent" pickling baths by electrochemical means. There are, in fact, a few electrochemical approaches to achieve a low cost and safe method with minimal associated pollution hazards. As a test case we will explore some of these means of continuously removing iron from sulfuric acid solutions. All the proposed methods are electrochemical processes and they require special ion exchange membranes and electrodes for their practical operation. The membranes need to be chemically resistant to the acidic and somewhat oxidizing environment. They must have adequate selectivity

72


and not be "fouled" by iron or other heavy metals, as well as have low permeability for the prevention of excessive molecular (thermal) diffusion. Electrodes must be chemically inert, inexpensive enough and have a long operating life to make them useable in practical systems. As the iron accumulates within the pickling bath during a stripping process, the pickling solution capability diminishes. Eventually, the solution must be replaced with fresh sulfuric acid solution. The problem that exists is the disposition of the "spent" pickling solution. Ideally, the continuous restoration of the bath via a method not requiring either the discarding or the re-supply of toxic or hazardous chemicals is desired. We will briefly explore the possibility of removing dissolved iron in the form of ferrous and ferric sulfate from the bath as a solid. The acid level in the bath is controlled solely by means of an electrochemical process. The technical goal is creating circumstances in which no chemical reagents, other than iron and iron compounds (Fe(OH)2, Fe(OH)3) in solid form, are brought to or removed, respectively, from the pickling process. 3.5.2

Technical Approaches

Three different electrochemical cell configurations, identified as A, B & C, are identified as potential solutions. Approach B as the first for this study. Cell design B appears to be the most attractive because it might result in the most current efficient method, if other parameters prove practical. Approach A, another very likely possibility, has been selected as a close alternative method. Approach C is a strong contender, but probably more complex in structure and operational management. One additional approach to removing heavy metals in solution from a water stream is one in which insoluble


73

compounds are precipitated out of solution into a collection chamber, or in which the metal(s) are electrodeposited onto a negative electrode. Passage of soluble iron compounds such as the sulfates through a cation membrane into an electrode cell chamber in which the pH is maintained high by means of electrolysis, for example, will result in removal of these metals as solid hydroxide and oxide compounds. As an example of a realistic problem that may be solved, we will select the following. We will assume a pickling bath is a 10% H 2 S 0 4 / H 2 0 solution with Fe2(S04)3 at levels up to 5% in concentration. It is desired to remove the iron component from the bath, or at least reduce the level significantly for bath reuse, and return the bath solution to its original 10% acid concentration. To achieve a practical and workable approach, it is necessary that this entirely electrochemical method of iron removal and acid level restoration of the pickling bath have certain characteristics. They include: a. Reasonably high coulombic efficiencies of separation b. Long life of components (estimates) c. Acceptable membrane transport number and conductivity values d. High percentage removal of iron, and low loss factors for sulfate ion 3.5.3

Technical Approaches

There are three basic approaches outlined as possible contenders. These are described below. Approach A. Figure 3.23 is a very simple cell schematic showing the type of process to be investigated. The membrane is an anion structure with excellent iron compatibility. Modified RAI cation membranes have yielded very encouraging properties in terms of iron transport numbers (Lee, Zito, and D'Agostino).

74


H 2 S0 4

Filtered OHNa+ so;

H+ Fe+3

ÍJ*

H

Fe2(S04)3 Na 2 S0 4

(+)

Anion membrane

H 2 S0 4 Fe2(S04)3

Figure 3.23 Approach A.

Depleted bath solution is circulated through the negative side of the cell and iron is plated out and iron oxides are formed in the increasingly basic environment. Iron compounds are precipitated out of solution and collected for discarding. Sulfate ions are pulled through the membrane to the positive side and sulfuric acid is generated as the new, reconstituted pickling bath. Disadvantages: Higher coulombic need for moving S0 4 = ions across the cell to generate acid. An auxiliary stage of cells may be of assistance in further restoration of acid level. Approach B Figure 3.24 is a simple cell diagram showing the operation of a device employing cation membranes. In this case the iron is being transported for removal from the main bath rather than the sulfate to form acid. As the iron is accumulated in the negative side (basic), it will be precipitated out of solution as oxide-hydroxide compounds. The sulfuric acid solution will also be reconstituted to its original concentration via loss of Fe+3 ions and replacement


75

Fe ' Solid ! «= ! Fe(OH)2 ;

1

1 1 i

MaHH

i

OHNa+

!

■H*r'rrr-

1

A

; Cation Membrane

1

H+

Fe+3

1

|1

i

(+)

|

H 2 S0 4 ' Fe 2 (S0 4 ) 3

Figure 3.24 Approach B.

of H + ions with the evolution of oxygen gas at the positive electrode. Disadvantages: Some loss of S0 4 = ions from the (+) side due to membrane diffusion, and coulombic loss due to H + ions transport from (+) to (-) side. Advantages: Better coulombic efficiency due to the fact that Fe is being removed from the bath rather than removing the larger quantity of sulfuric acid to another solution. Investigate: Membrane performance, electrode life and sludge removal. Approach C Another possibility is the use of membranes, which do "foul" in the presence of iron ions. There may be an opportunity to preferentially remove hydrogen ions from solution. Figure 3.25 shows the type of process contemplated as a means of separating iron from the acid solution. Each of the above has limitations and certain advantages. The optimum choice would depend greatly upon the nature of the problem to be solved as well as the properties of the electrode and membrane components available.

76


o,

Neg

Pos

Cation

Anion

Figure 3.25 Approach C.

Another possible cell design that could be a practical solution to removal of iron and other heavy metals would entail a modified pH control cell wherein the cations to be removed are transported into a region of high pH in the (-) compartment of a control cell. These ionic substances would be either plated out onto the (-) electrode or precipitated as hydroxides and oxides. 3.5.4

Laboratory Feasibility & Data Study Suggestions

All the proposed methods are not only electrochemical processes, but they also require special ion exchange membranes and electrodes. The membranes need to be resistant chemically to the acidic and somewhat oxidizing environment. As was stated earlier, they must have adequate selectivity—not be "fouled" by iron—and have low permeability to prevent excessive molecular (thermal) diffusion. 3.5.5

Experimental Methods

In order to conduct the necessary series of experiments, data collection and quantitative evaluations,


77

appropriate tests were designed. Apparatus, chemical analysis equipment and electrical circuits must be available, while demountable electrochemical cells need to be constructed. Electrodes with carbon polymer compositions and external wire connections were fabricated to accommodate the test cells. Ion exchange membranes from various sources were assembled, and some membranes were modified for ferrous and ferric ion transfer. Figure 3.26 shows a simple circuit employed in the experiments. A constant current power supply was used to simplify computation of total amp-hours transferred. A coulometer in series with the cell was employed. Wet chemistry analysis was employed to determine the concentrations of ferric and ferrous ions, and the level of acidity (normality) of the solutions at different periods during cell operation.

Constant current power supply

/»

\._

Ammeter

1 ^ 1 I / 1 Voltmeter

Electrochemical cell

Figure 3.26 Electric circuit cell operation.

Coulometer

78


The chemical methods employed for these measurements are identified below. 1. Acidity determination - Titration with a 1 normal solution of NaOH and with phenolpthlalein as indicator. 2. Ferric concentration - Oxidation of iodide method. A solution of Nal is used with a starch indicator. Titration is accomplished with a 0.10 N solution of sodium thiosulfate. 3. Ferrous concentration-An ortho-phenanthroline ferrous complex is used as indicator, (changing color from red to faint blue when ferrous is oxidized to ferric), with a 0.1 N solution of ferric sulfate as titrant. Figure 3.27 is a diagram of the type of open top test cells that were made for and employed in these experiments. The cell is constructed of rigid PVC in two half sections. Each section contains an electrode and a flat rim around the frame. The two half-cell sections are clamped together with a membrane sandwiched between the frames. RTV is used as a liquid gasket to make a water-tight seal.

(+)

Carbon composite electrode

Cell container

Figure 3.27 Basic test cell construction box, electrode, & membrane.


79

3.5.5.1 Approach B Tests A cation membrane, SYBRON lonac, was placed in the cell shown in Figure 3.23. Solutions of sulfuric acid and ferric sulfate mix were prepared to anticipate a worst-case condition for pickling baths. The first solution is as follows: 100 ml of 66 Baume H 2 S0 4 50 grams of Fe 2 (S0 4 ) 3 5H 2 0 Total volume of solution is 1200 cc Titration of the mix gives the net concentrations below. 1.9 ml of thiosulfate per 2 ml of bath sample = 0.095 N ferric salt, or approximately 4% by weight solution of ferric sulfate. 7.1 ml of NaOH solution per 2 ml sample = 3.55 N sulfuric acid, or approximately 16% by weight solution. The first cell tested employed a SYBRON cation membrane, and was configured as cation| (+) 0.095 N F e + 3 + 3.35 N acid membrane 210 ml

0.10 N NaOH

(-)

210 ml

with 7.5 square inch active area electrodes in the cell. The cell operation and planned ion transfer are shown in Figure 3.24. Ferric ions are to migrate from the (+) side to the (-) side with an increase in pH. They precipitate as hydroxides, or come out as iron plating. The cell ran for 2 hour intervals at 1.5 amps on five separate occasions with no measurable net transfer of Fe+3 out of the positive (+) side compartment. It appeared that all the charge transfer was via H + ions.

80


Other similar series of tests were conducted employing cation membranes. Membranes'were tested from a variety of sources such as HCI, Ionics, and DuPont, NAFION types. The same results of very little or no significant Fe+ ion transfer were obtained. An ESC (RAI) homogeneous polyethylene membrane was converted from the normal Na form to the Fe+3 form at TRL and tested in similar circumstances. The short Table 3.4 below gives typical results. Let us review the magnitude of the Fe+3 concentration change we should expect from this charge transfer if the mechanism was entirely attributed to iron transport. A200 ml volume of 0.095 N Fe+3 solution in the cell chemically configured above would have these electrode reactions: (+) side ... 3(S04"2) +3 H 2 0 = 3H 2 S0 4 +3/2 0 2 + 6e" (-) side ... 2Fe+3 + 3H 2 0 + 6e~ = Fe(OH)3 + 3/2 H 2 Six electrons are drawn from the external circuit per two Fe ions, or about 78 amp-hours of charge are needed to +3

Table 3.4 Iron Form ESC membrane test (2 ml test sample sizes). Time, hrs

Volts

Amps

Thiosulf ate, ml

0

10.7

0.64

1.8

0.2

10.7

1.73

1.7

0.4

4.7

1.5

1.8

0.6

4.7

1.6

1.7

0.8

4.7

1.6

1.7

1.0

5.4

1.9

1.7

1.5

5.4

1.9

1.65

2.0

5.4

1.9

1.6

2.5

5.4

1.9

1.65

3.0

5.4

1.9

1.65


81

transfer iron from the (+) to the (-) side at 100% coulombic efficiency. If the coulombic efficiency were 100%, total transfer of iron from 200 ml of 0.095 N Fe+3 solution would require about 1.5 amp-hours of charge. In the experiments above, more than 1.5 AH of charge flowed with no measurable net transfer of iron. In the case of the ESC membrane, its coloration after conversion to the iron form was brown. Its normally colorless dry and white when wet. During operation of the cell, the membrane became white, showing that it had reconverted to the hydrogen form via displacement of iron by H + ion migration through the membrane as the sole charge carrier in the cell. A quick look at the relative specific ion conductivities of H + and Fe+3 at infinite dilution will show the significant preferential nature of the processes in the cell. H+ l/3Fe + 3 OHNa + 1/2SO; 2

320 mho-cm" Vmilli-equiv. 50 172 44 68

And, because the concentration of the H + ions is 3.55 N, or 35 times as great as that of the iron, one can expect the transport numbers for these ions to be at least 100 times greater for hydrogen. It appears that the only way in which the cation membrane configuration of Approach B, shown in Figure 3.24 can be useful is in very much more dilute acid solutions. The competition for hydrogen transport over iron is not as pronounced in lower acid concentrations. 3.5.5.2 Approach A The results of the experiments of method B were not very promising, so it was decided to explore the possibilities of approach A. The configuration is essentially the same as in

82


B except that an anion membrane was employed and the principle ion for transfer is the sulfate radical. Figure 3.24 shows the principle processes involved. S04~2 ions are transferred from the (-) side to the (+) side with no net movement of iron, sodium or hydroxide ions. SYBRON anion membranes were employed in all these tests. The approach involves the idea that as the bath circulates through the (+) side of the cell, its H 2 S0 4 content rises because of the transfer of S04~2 ions from the (-) side of the cell, and the liberation of oxygen at the (+) electrode. Meanwhile, hydrogen is evolved at the (-) electrode with the continuous formation of hydroxide ions for the precipitation of insoluble iron compounds. Some iron also comes off as plating on the (-) electrode. These solid materials are removed as sludge or by filtration in a continuously moving bath. Various tests show the behavior of this system, and provide some quantitative data for further design purposes. A series of tests with solutions of composition listed below were made. (+) side ... 16% H 2 S0 4 solution, about 3.35 N concentration. 4% Fe+3 solution, about 0.10 N concentration. (-) side ... 2% Na 2 S0 4 solution initially, 0.3 N Concentration. The cell was operated at 2 amps with an electrode effective area of 7.5 in2. A 2 ml sample was withdrawn from the solutions periodically and analyzed for reagent concentrations. Table 3.5 shows a typical run. The purpose of these experiments is to determine if the acidity in the (+) side can be increased at the expense of S04~2 ions from the (-) side, and if the ferric and ferrous ion concentrations move in the direction desired. The acidity in the (+) side increases steadily to that of the (-) side initial value.


83

Table 3.5

Time-hrs 0

(+) side

(+) side

(-) side

1.0 N NaOH

Fe+3

Fe+2

1.9

0

Volts

Amps

Titration

6.56

2

0.05

0

0.05

0.08

0.15

0.16

0.15

0.33

0.25

0.67

0.60

1.5

4.8

2

1.7

2.16 2.67

1.3

3.56

2

0.05 3.3

1.95

The Fe+3 concentration diminishes and the Fe+2 concentration increases as the iron is reduced at the (-) electrode. A quantitative assessment of the transfer yields the following information. 1.95 ml of a 1 N N a O H / 2 ml sample corresponds to a concentration of 1.0 N acid transfer in 2.67 hrs. via 2 amps x 2.67 hrs = 5.34 AH input. Only 5 AH are needed to do transfer for 200 ml of 1.0 N for 100% coulombic efficiency. Figure 3.28 is a graph of this data, and shows how close the cell performance is to the maximum. Another new experiment was devised to explore the possibility of introducing small quantities of Fe+3 from pickling bath into the (-) side to maintain a steady flow of freed solution back into the increased acidity bath of increasing acidity as a continuous restoration process. The following composition was prepared as the initial solutions, and a

84

ELECTROCHEMICAL WATER PROCESSING Approach A Data from table I 120

0.25

I

I

0.15

c a 0.05

3 LU

2

3 4 Amp-hour input

■ Actual equivalents

Theoretical Max. Equiv.

- Coulombic efficiency Equivalents of sulfate in 200 ml volume Figure 3.28

small quantity of depleted pickling bath was introduced periodically into the (-) side for iron removal. (+) side ... 200 ml 3.35 N H 2 S0 4 + 0.095 N Fe2(S04)3 (-) side ... 200 ml H 2 0 + 5 ml of (+) side composition solution The (-) side is plain water with a very small quantity of the pickling solution to provide for initial conductivity to get the cell started. The approach contemplated here is that while the main body of the pickling bath circulates through the (+) side, a small percentage is introduced slowly into the (-) side. The (-) side is at a much higher pH, enabling the iron to be removed as filterable solids. The liquid from the (-) side is then returned to the main bath as essentially a Na 2 S0 4 water solution. The pH in the (+) side is continuously being lowered due to acid regeneration via S04~2 ion transfer from the (-) side.

G E N E R A L O U T L I N E S OF B A S I C A P P R O A C H E S

85

All these observations are made in the tests conducted with these cells. If the pH in the (-) side is kept above 7, there is little Fe+2 or Fe+3 in solution. Table 3.6 below lists some typical data obtained from such a cell. Plots of this data are presented in Figure 3.29.

Table 3.6 Hydrogen ion concentration figures are given as ml NaOH titration data. Time, hrs

Volts

Amps

0

0

0.08

14.6

2

0.16

16.2

1.68

0.25

16.4

1.24

0.33

16.1

0.97

0.42

16.1

0.60

0

(+)H +

(+) Fe+2

0.4

0.05

0.5

0.05

(-) side Fe+

3

0.3 0.1

0.3 0.2

precipitate noted in (-) side, reddish brown iron 0.5

16.1

0.40

introduced 2 ml of pickling, (+) side solution, into (-) side 0.59

13.1

2

0.1

0.67

16.1

1.3

0.2

0.75

16.1

0.61

0.25

orange colored ppt. in (-) side appeared again 0.83

16.1

0.41

added 3 grams Na 2 S0 4 salt to (-) side 0.92

8.3

2

1

7.6

2

1.08

8.3

2

0.3

(-) side H+

86


Table 3.6 (cont.) Hydrogen ion concentration figures are given as ml NaOH titration data. Time, hrs

Volts

Amps

(+)H +

8.4

2

0.4

1.13 1.25

0.45

added 2 ml 1.28

7.8

2

added 2 ml 1.72 added 2 ml 1.8

0.8

added 2 ml 2.0 2.04

0.95

2 ml 2.25

1.15

2.42

1.25

2 ml 2.52

1.25

2 ml 2.72

1.4

2 ml 2.83

1.5

2 ml 3.13

1.6

2 ml 3.83

1.9

(+) Fe+2

(-) side Fe+3

(-) side H+


87

Table 3.6 (cont.) Hydrogen ion concentration figures are given as ml NaOH titration data. Time, hrs

Volts

Amps

(+)H +

(+) Fe+2

(-) side Fe+3

(-) side H+

12 ml 4.08

2.0

4.27

2.25

10 ml 2.35

4.45 4.55 4.68

5.6

2

2.45

Some simple calculations and general observations follow. • Every time the pH of the (-) side became basic, a large quantity of brown-orange precipitates were seen. • Coulombic efficiency for generating acid in the (+) side can be estimated as: ° Constant current @ 2 Amp for 4.2 hrs is 8.4 AH charge transfer ° Acidity change in that time is 0.3 to 2.45 ml = 2.15 ml/sample, or about 1.1 N acid in vol. of 200 ml giving about 1/5 equivalent S04~2 transported to (+) side. At 100% coulombic efficiency, about 5 AH are needed, giving the cell a current efficiency of 5/8.4 = 60%. Considering the very low concentration of S04~2 present in the (-) side during the processing, this efficiency is quite surprising. • To estimate the efficiency of maintaining a basic environment in the (-) side, we may take the

88


readings between 1.72 and 2.0 hrs. This provides the following: 2.0-1.72 = 0.28 hrs, @ 2 Amps the charge is 0.56 AH 4 ml @ 4.2 N contains 0.0168 equiv. weights of acid 26 x 0.0168 = 0.44 AH required @ 100% eff. to transport SO~ Coulombic efficiency of conversion to precipitates is the found as 0.44/0.56 = 79%. Figure 3.29 is a graph of the salient data from Table 3.6. A different composition batch of test solution was prepared that may better represent the spent pickling bath. Experiments were established to observe the behavior of the cell when initially at the condition where the solution on both sides is the same.

Approach A Data from table I

0.25 to

o u 7 77c

a

dt

(4.42)

Without pursuing the mathematical manipulations further, one can see that the form taken by the expression for r/c is that of an exponential with Q N appearing in the exponent.

134


As an approximation, and as a first compensation for the variation in r¡c, we will consider only the effect of ion concentration. In general, as the concentration or availability of specific ions to be transported across membranes becomes lower, the coulombic efficiency for transporting that specific ion will diminish. The decrease in 77c is simply due to the fact that the competition of H + and OH" ions becomes greater because their availability as charge carriers becomes proportionately greater than the less populous ions. Hence, the functional dependence of r¡c upon concentration, cx, at some point x within a cell must be determined. A simplified, but useful, form of the functional dependence that could be taken would be the exponential relationship i% = (l-e^),

(4.43)

where \]/ has a value such that the coulombic efficiency is nearly unity when cx was at some level where the competition for transport by other ions was negligibly small. Such values of cx can be determined experimentally. To define the function completely, at least one empirical data point is needed. Some simple measurements should be made for each of the membranes to be employed at more than one concentration of dissolved substances. We have graphed Equation (4.43) to show how coulombic efficiency varies with concentration for different values of the constant, \\f. Empirical data provides the information needed to quantify \\f. To properly account for the dependency of r¡c upon cx it is necessary to insert the expression for r¡c into the preceding formulas for cx and ix as seen in Equations (4.31) and (4.35). Because of the additional variable terms, the resultant expression becomes more complex mathematically when accounting for a non-constant r¡c. For example, the solute concentration takes the form

ANALYSIS «fe MODELING ELECTRODIALYSIS SYSTEMS

pvdcx/? =

Ec

X

r¡- À- dx

135

(4.44)

substituting for r/c from the equation rç = ( l - e ^ ) ,

(4.45)

E-c,

(4.46)

and we get p-v-dcx-ß:

(l-e-^)-A-dx,

H tu

n >

a

n

O

n

M

r

m 1

178.6 4.00

72

2

Hf

He

Helium

Tellurium

Tantalum

Te

Ta

S

Hafnium

197.2

Sulfur

79

Au

Gold

162.5

63.6

92.9

58.9

52

72

16

127.6

180.9

32

87 38

Sr

Strontium

72.6

32

Ge

Germanium

23

11 Na

Sodium

69.7

31

Ga

Galium

107.9

47 Ag

Silver

156.9

64

Gd

Gadolinium

28

14

Si

Silicon

19

9

F

Fluorine

79

34

Se

Selenium

152

63

Eu

Europium

45.1

21

Sc

Scandium

167.2

68

Er

Erbium

150.4

62

Sm

Samarium

66

Dy

Dysprosium

44

Ru

Ruthenium

29

Cu

101.7

Copper

85.5

37

Rb

Rubidium

41

Cb

Columbium

102.9

45

Rh

Rhodium

27

Co

Cobalt

186.3

75

Re

Rhenium

52

24

Cr

Chromium

222

86

Rn

Radon

35.5

17

Cl

Chlorine

z

H

t->

> a (X) o

M

d

a o Z n

Z

o

n

ça

X

o

M

>

Thallium Thorium

Uranium vanadium

1.01 114.8 126.9 193.1 55.9 83.7 138.9 207.2 6.9 174.9 24.32 54.9 200.6

1

49

53

77

26

36

57

82

3

71

12

25

80

H

In

I

Ir

Fe

Kr

La

Pb

Li

Lu

Mg

Mn

Hg

Hydrogen

Indium

Iodine

Iridium

Iron

Krypton

Lanthanum

Lead

Litium

Lutecium

Magnesium

Manganese

Mercury

Zirconium

Zinc

Yttrium

Ytterbium

Xenon

Tungsten

Titanium

Tin

Thullium

Terbium

164.9

67

Holmium

Ho

Element

Atomic Wt.

Symbol

Element

Atomic No.

Table B.4 (cont.) Atomic weights and numbers.

232.1 169.4 118.7 47.9

90 69 50 22

Th Tm Sn

131.3

54

Zr

40

30

39

Y Zn

70

Yb

Xe

51

23

V

91.2

65.4

88.9

173

238.1

92

u

183.9 74

W

Ti

204.4

81

Tl

159.2

65

Tb

Atomic Wt.

Atomic No.

Symbol

o

z

on

ta

o n

ta TI

5

>

ñ

g

M

n

O

!«

H

n

M

r1

m


Appendix C: Feeder Tube and Common Manifolding Losses In designing an array of cells, there are many additional factors to consider for the purposes of optimization. Among these are such considerations as the shape and size, manner in which cost changes with geometry, electrical power supply requirements and electrical current losses encountered in the common manifolding of cells that are supplied with water from a common source. Cells are interconnected by common manifolds that are so large in cross-sectional area that they present little electrical resistance to short circuiting currents. The only significant resistances to the common connections are those in the smaller diameter, longer length feeder tubes. In electrodialysis systems, there are four such circuits in parallel electrically. Two manifolds and corresponding feeder tubes for the entering water, and two manifolds for the separate exiting water processed and waste water 225

226


channels. This situation is true regardless of the specifics of the design geometry and manufacturing methods of module construction. Figure C.l shows the general configuration with only two of the four alternate manifolds illustrated. Manifold cross sectional area is much larger than the individual feeder tubes for the individual cells. The following is the development of the equations describing the common loop current losses, (parasitic currents), that an ED array experiences. Figure C.2 shows the simple mesh current model that represents the leakage paths through the parallel hydraulic circuits model employed in arriving at the expressions and consequent estimates of coulombic losses. In other words, electrical current that does not contribute to the separation of dissolved substances from the processed water stream. Each cell is represented as a voltage source, E, and the electrical resistance path due to the feeder tubes or channels from each cell to the common manifold is represented as the constant, R. If all the fluid supply channels are the Exit manifold

Electrodes Membranes

Feeder tubes

Entrance manifold Figure C.l Manifold configuration.

FEEDER TUBE AND COMMON MANIFOLDING LOSSES

R:

R -I E

^

E

227

R:

Hh E

-I E

Figure C.2 Equivalent circuit for common manifold.

E

E

E

Figure C.3 Single manifold - Equivalent manifold.

same in construction the resistances, R, are all equal and constant. There are a number of ways in which the analysis of the electrical current losses can be determined. Simple application of Kirchoffs Law enables us to calculate the dissipative current through any cell in an array. Figure C.3 shows the loop currents assumed in the series of circuits of the array of cells. Each small loop is treated as an independent circuit except for the interconnectivity due to currents flowing in the adjacent resistances from other loops. After establishing the basic relationships, we can construct a series of simultaneous, linear equations to be solved for calculating the currents. The form of these equations is as follows. For any one loop in the circuit array the expression has the form of equation C.l. Tracing the potential around the nth loop results in the sum of all current producing voltages passing through the respective feeder tube resistances equals the voltage of that particular cell produced by current flow through the cell. The cell voltage, E, is that expressed in Equations (4.25) and (4.87).

228


2i n -R = E + i n . 1 -R + i n+1 -R

(C.1)

where n =1 to N, if there are N number of cells in the array. The current passing through the n t n cell due to the common manifold circuit is i n . A sample calculation of the above solution can be made by taking a 10 cell array, and solving for the currents in each cell. The equations are: 2irR = E + i2-R 2i2-R = E + irR + i3-R (C.2) 2i9R = E + i 8 R + i10R 2i10R = E + i 9 R If we assign some values to both R and E that are typically in the range of operating ED systems, we can solve for the currents to obtain an idea of the dissipation current distribution. Also, in order to see what the shape and the magnitudes of the dissipation currents are we must settle upon values for the geometry of the feeder channels from the manifold to the cells. Let E = 1 volt, and the electrolyte resistivity be in the order of 200 ohm-in., and the length, 1, and diameters, d, of the feeder channels can be about 0.10 in. and 1 in., respectively. The solution of the equations above give the calculated values for the currents through cells one through ten as follows. For p=200 ohm-in.

For p=2 ohm-in.

ii = 1.96x10""4 amps \ = 3.53 i3 = 4.7 \ = 5-5 \ = 5.89

0.0196 amps 0.0353 0.047 0.055 0.0589

FEEDER TUBE AND COMMON MANIFOLDING LOSSES

For p=200 ohm-in.

For p=2 ohm-in.

\ = 5.89 \ = 5.5 i = 4.7 \ = 3.53 i10 = 1.96

0.0589 0.55 0.47 0.353 0.0196

229

Dissipation currents are highest in the middle cells, and diminish toward either end cell. The solution concentration in the waste water channels can be between 10 and 100 times greater than in the processed water side. In fact, if the ratio of waste water to processed water is to be as high as practical, these could very well be the range of concentrations encountered in any operating ED systems. The values above are, for most water demineralizing applications, quite high. Significant coulombic losses can certainly be encountered in desalination ED systems where the salt water conductivities are high. The two entering water concentrations will be the same, and at the level of the initial input water source. However, one of the manifold systems at the exit will be at the concentration of the waste water, and will be the major contributor of any parasitic current losses. Figure C.4 is a plot of the two values of currents from the above table. As is readily seen, the most severe dissipative current, or coulombic losses are in the inner most, or middle cells of an array. Another current calculation that results in the same answers as derived by the preceding calculations is shown in Figure C.4. Here, the loop currents are taken differently. Instead of independent for each small circuit in the series, the currents are assumed as all originating from the first cell and circulating through each of the succeeding circuits in turn. Thus, the first current, iv passes through only the first cell, and the second current, i , passes from the first resistance, R, and then through around through the third resistor, R, and then through the two cell sources, E.

230

ELECTROCHEMICAL WATER PROCESSING Ten cell array 0.07 0.06

E

0.05

< 0.04 "'
W ) 0.0136

(E.44)

(E.47)

169 266

It can be seen that the value of K ranges from about 160 as a minimum to 330 as a maximum over the allowable spread of qH and qc concentrations. If the acid concentration is too low the resin cannot be regenerated to the preset percentage of full capacity.

APPENDIX E: MATHEMATICAL ANALYSIS

261

E.4 Electrolytic Resistance of the System Water When evaluating the expressions for quantitative relationships it becomes necessary to specify the constants for the transport terms due to electric current flow. The RA(ED), RC(ED), RH(ED) and ROH(ED) terms are electric current dependent. Equations E.7(a), E.8(a), E.14(a) and E.15(a) on pages 197 and 198 all contain R(ED), or current terms. The constants b, c, r and s are equivalent resistance (reciprocal equivalent conductance) terms. Appendix B gives a table of some specific ion values of conductance. These resistance values are inversely proportional to the ion mobilities for that specific ion. If the equivalent conductance of a substance is M, and its solution concentration is S, in GEW per liter, then: Conductance = SM/1000, or Resistivity, R = 1000/SM ohm-cm. Equivalent Conductance, M, is given as mho/cm-equiv. Converting to the English system, R = 1000/2.54SM = 400/SM ohm-in Now, the electric current is V/R, where V is the voltage drop along the path of flow of the specific, subject ion. The units would be amperes. To convert to equivalent weights transported per unit time, we need to change dimensions. Since 26 amp-hr corresponds to about 1 gram equivalent weight of substance, The above equation becomes d q / d t = V/26R GEW per hour.

(E.48)

If we are interested in the quantity, d q / d t , transported per hour per square inch of electrode area in a particular cell, the inter-electrode or electrode-membrane spacing, W, must be taken into account, and d q / d t becomes V/26RW.

(E.49)

262


Since the q terms in the system expressions are in units of GEW per in3, it is necessary to normalize them in terms of GEW per liter. Hence, the expression d q / d t = V/26RW = V/26(400/SM)W = VSM/10400W,

(E.50)

becomes d q / d t = 60VqM/10400W = VqM/173W.

(E.51)

Typical values of M at low concentrations are: MH+ 340 mho/cm-equiv. MOH.190 M Na+ 44 M r 60 Ca M

C1-

6 5

MS04 75 Now, we may insert numerical values for the constants, b, c, r, and s as follows. b = M H+ /173W = 2/W + mho/equiv.

(E.52)

c = M OH ./173W = 1.1/W

(E.53)

r = M c+ /173W = 0.29/W +

(E.54)

s = MA_/173W = 0.40/W

(E.55)

A potential problem exists in determining the value of voltage to use in the rate equations since the division of the cell potential from the membrane to the electrodes will depend upon the ratio of the resistances. For example, if the resistance on the positive side is R+ and that of the negative side is R, then the potential drop, V , (driving voltage) on the (-) side of the cell is


V = VR./(R + R+),

263

(E.56)

and similarly for the positive side voltage, V+ = VR + /(R + R+).

(E.57)

The R's are obviously obtained by taking the reciprocal of the sum of all the conductances due to the various ionic species in either side of the cell. For example, R=(l/ROH. + l / R A + l / R / = ([SM]OH_/400W + [SMIA/400W_ + [SM]_C/400W)"1 (E.58) and similarly for the positive side, R+ = (l/R H +

+

l/R+A+l/R+c)-i

= ([SM]H+/400W+ + [SM]+A/400W+ + [SM]+C/400W+)-1 (E.59) A reasonable first approximation which may be employed in the computations associated with the rate processes is to let the total resistance on either side of the membrane be equal then the driving potential for the specific ions will be 1/2V. In most designs the spacing in the (-) and (+) cell compartments will be about equal also, or W = W+ = W.

E.5 Solution of the Simultaneous System Equations The series of equations that were described earlier now assume the forms given below. These are the modified forms of equations (E.l), (E.2), (E.3), (E.4), (E.5), (E.6), (E.7a), (E.8a), E.9, (E.14a) and (E.15a) in light of the further definitions provided the coefficients. The q and Q terms have been substituted in the following equations as shown on page 9 for the sake of brevity

264


of use of symbols, and in order to make the nomenclature more adaptable to MuMath for symbolic manipulations, FORMULA ONE and LOTUS for numerical evaluations. Q + 2q = B + C + D

(E.60)

A+C =E

(E.61)

B=G +L

(E.62)

D = Q-H

(E.63)

Q + 2q = E + L + J

(E.64)

J = Q-I

(E.65)

v W d A / d t =: V+rC + V sL - V+bA - V cG + (eCH - fAD)W2v

(E.66)

v W d G / d t == VrC + VsL - V bA +

- V cG + (gLI - hGJ)W2v EvWF/Rj. := V+bA + V+rC + V cG + V sL VbA=VsG

(E.67) (E.68) (E.69)

0 = 2V + bdA/dt + V+r(V+bA + VcG - V+rC - (eCH - fAD)W2v) + V s(V+bA + V cG - VsL - (gLI - hGJ)W2v)

(E.70)

NOTE: The quantity, Rp, appearing in equation (E.68) is the total electrical resistance of the cell, or Rj. = R + R+ Since H + ions do not exist in the negative cell side, and because OH" ions do not survive in the positive side, current is carried by these ions only in their respective sides of origination. Hence the driving potential for these ions is that of the half-cell side, or V+ for the H + ions, and V for the OH" ions.


265

The conduction process is due to all ion migration on both sides of the membrane. Hence, the total resistance is found as follows. The conductance on the positive side is the sum of all charge carriers, or l/R + = b A + r C + sE

(E.71)

l / R = c G + rB + sL

(E.72)

Since Rj. = R+ + R, the final expression for the total resistance becomes; RT = 1

bA + rC + sE

+

cG + rB + sL

(E.73)

The above value for the total resistance must then be substituted into equation (E.68) for Rj, Equation (E.70) has the form above only if Rj, is treated as constant in equation (E.68). Since it is in fact not constant, equation (E.70) must be modified accordingly to reflect this fact. Differentiating equation (E.68) as before is now a very complex operation. There are some approximations that are allowed while still preserving the basic relationships of interest. For example, the resistance on both sides of the membrane will probably be about equal throughout the operation of the system. Thus, one may let the total resistance be equal to twice that of either side, or R =2R+ = T

+

bA + rC + sE

(E.74)

Substitution of (E.74) into (E.68) gives, VvWF[bA +rC +sE]/2 = VbA/2 +VrC +VcG/2 +VsL (E.75)

266


Eliminating both C and L from equation (E.75) by substituting from (E.61) and (E.62) gives VvWF(bA +r(E-A)+ sE)/2 = VbA/2 + Vr(E - A)/2 +VcG/2 + Vs(B - G)/2

(E.76)

Differentiating equation (E.76) with respect to time yields vWF(bdA/dt + r(dE/dt - d A / d t ) + sdE/dt) = b d A / d t + r(dE/dt -dA/dt) + cdG/dt + s(dB/dt - dG/dt)

(E.77)

Now we can substitute again for the derivatives d E / d t and d G / d t and dB/dt from equations (E.12), (E.8a) and (E.13), respectively. vWF(bdA/dt + r(VsL/2 - dA/dt) +Vs2L/2) = bdA/dt + r(VsL/2 - dA/dt) + cdG/dt + s(VrC/2 - VrC/2 - VsL/2 + VbA/2 + VcG/2 -dQ A /dt) (E.78) dA/dt(bvWF - rvWF - b + r) + vVWFrsL/2 + vVWFs 2 L/2 = VrsL/2 + sVrC/2 - rsVC/2 - Vs 2 L/2 + VsbA/2 + VscG/2 - sdQ A /dt + VrcC/2 + VscL/2 - VbcA/2 + Vc 2 G/2 - cdQ A /dt Equation (E.78) becomes the new version of equation (E.70), and equation (E.68) has the form of equation (E.75). Thus, the new series of relationships are equations (E.60), (E.61), (E.62), (E.63), (E.64), (E.65), (E.66), (E.67), (E.75), (E.69) and (E.78). We may now proceed with the solution of these equations in generalized form and then in terms of specific values of the system constants.


267

Equations (E.60) through (E.65) and equation (E.75) are solved as a series of linear, simultaneous algebraic relations. Their solutions for the variables B, C, D, E, G, H and I in terms of the variables A, J and L are listed below. B = (AvWsF + AvWbF - ab + 2vqWrF + 2vqWsF + vHWrF + vHWsF -2qr - Ls + Lc - H r ) / (-r + c + vWrF + vWsF)

(E.79)

C = (-AvWsF - AvWbF + Ab + 2qc + Ls - Lc +Hc)/ (-r + c + vWrF + vWsF) D = Q-H

(E.80) (E.81)

E = (AvWrF - AvWbF - Ar + Ab + Ac + 2qc + Ls - Lc + He)/ (-r + c + vWrF + vWsF)

(E.82)

I = (AvWrF - AvWbF - Ar + Ab + Ac - 2vqWrF - 2vqWsF + vLWrF + vLWsF + 2qr - Lr + Ls + He)/ (-r + c + vWrF + vWsF)

(E.83)

J = (-AvWrf + AvWbF + Ar - Ab - Ac + vQWrF + 2vqWrF + 2vqWsF - vLWrF - vLWsF - Qr + Qc - 2qr + Lr - Ls -He)/ (-r + c + vWrF + vWsF)

(E.84)

G = (AvWscA2F + AvWbc A 2F -Abc A 2 - 2vqWrc A 2F + 2vqsWc A 2F - vLWrcA2F - vLWscA2F + vHWrc A 2F + vHWsc A 2F - 2qrc A 2 + Lrc A 2 - LscA2 - Hrc A 2)/ (vWrcA2F + vWsc A 2F - rc A 2 + cA3)

(E.85)

268


These are employed in conjunction with equations (E.66), (E.75) and (E.78) to totally evaluate A, (as it is set up at present), as a function of time. To accomplish this end the first order differential equation, d A / d t , must be solved. The relationships rapidly become extremely complex and lengthy. Hence, it serves simplicity to insert numerical values wherever possible and as early as practical to shorten the lengths of these expressions. E.6

S a m p l e S o l u t i o n of O p e r a t i n g S y s t e m

It seems that the initially most interesting relationships from which to obtain quantitative information and to graph are the following: • Time dependence of qH and Q H for different values of V, W, and perhaps v. • Maximum level of regeneration achievable in the resin for a particular set of conditions • Coulombic efficiency versus time, V, W, and v. • Different resin properties will affect the performance as well Let us now insert the first series of numerical values for the system constants as a test for the type of analytic information afforded here. The values for e,f,g and h are set to K = 200/vW in 3 / hr-GEW The values of the remainder are: b = 2 / W mho/equiv c = l.l/W r = 0.29/W s = 0.40/W V = about 10 volts, but is one of the variables


269

W = W+ = W_ = about 1 inch, but is one of the variables Q o = Q = 1.3xl0 2 GEW/in 3 for 15,000 grain/ft 3 resins qo = q = 1.3x10-* GEW/in 3 for 400 ppm input water If we substitute the above values for the constants, the equations are: B = 2.7A + 2.137xl0 4 + 0.113L +0.822H C = -2.7A + 0.436xl0"4 - 0.113L + 0.177H D = 1.3xl0- 2 -H E = -1.7A + 0.436xl0 4 - 0.113L + 0.177H I = -1.7A - 2.137xl0 4 + 0.886L + 0.177H J = 1.7A + 1.3xl0-2 + 2.137xl0"4 - 0.886L - 0.177H G = 2.7A + 2.137x1o4 - 0.886L + 0.822H and 0.3dA/dt = - 20A + 2.9C - 11G + 4L -200AD + 200CH 2A = 0.4G 10.764L + 11.628dA/dt = -7A + 1.595C + 8.25G + 1.98L +300GJ -300LI If we leave in the parameters W and V as indeterminate, the expressions are; B = (-0.00007 - 2A + 0.7L - 0.29H + 18.72AW + 0.00139W + 5.382HW)/(0.81 +5.382W) C = (0.0002 + 2A- 0.7L + 1.1H - 18.72AW)/(0.81 + 5.382W)

270


D-0.013-H E = (0.0002 + 2A- 0.7L +1.1H - 13.338AW)/(o.81 + 5.382W) I = (0.00007 + 2.81 A + O.llL + I.IH - 0.00139W - 13.338AW + 5.382LW)/(0.81 + 5.382W) J = (0.01045 - 2.81 A - O.llL - I.IH + 0.07136W + 13.338AW - 5.382LW)/(0.81 + 5.382W) G = (-0.00009 - 2.42A - 0.1331L - 0.3509H + 0.00169W + 22.6512AW - 6.51222LW + 6.51222HW)/ (0.9801 + 6.51222W) 0.3dA/dt = -200ADW -AV + 200CHW + 0.29CV -1.1GV +0.4LV 2A = 0.4G -1.71dA/dt + 1.0764LWV + 13.338WdA/dt = -0.7AV + 0.1595CV + 300GJ +0.825GV -300LI + 0.198LV The above equations can now be solved for A as f(t) for different voltage ranges and cell widths.


Appendix H: Mathematics for Simple Electrochemical Biociding In Chapter 5, Section 5.1 we outlined the salient factors in the design of an electrochemical bromine generator. The assembly of electrodes is designed to accomplish a particular bromination task, and hence require electrodes of sufficient size along with a dc power supply for the necessary current flow. The first of the following two sets of equations quantitatively relates the geometry of an array of electrodes, and voltages necessary to production of a given amount of bromine per unit time. This set of equations is named Figure H.l. The parameters in the above equations are identified and defined as follows. Bt Ec

Bromine generation rate, g m / h r Cell potential, volts 303

304

ELECTROCHEMICAL WATER PROCESSING Bt= i■ tj ■ Ae-3.08 Ec= Er+ i ■ Ae- Rc

p=1000/(/V-JJ

ppm = 103- 10 3 -/V R^s-pKN-AJ COP=Bt/P ppmBr= Bt- 106/(454 ■ V ■ 8) l=i-Ae P=l-Ec

Figure H.l Bromine generator array.

E r A

e

Î1

ppmBr

ppm N R c p n P s COP V

J

Electrode reaction potential, volts Working electrode area, sq in Coulombic efficiency Bromine concentration rate, p p m / m i n ppm of NaBr salt NaBr solution normality Cell resistance, ohms Power Input (electrical), watts Number of cells (electrodes) in parallel Specific resistivity, ohm-in Inter-electrode spacing, in. Coefficient of performance, gm of Br/ hr-watt Gallons of water being brominated, Equivalent conductance, 130 cm 2 / equiv-ohm

The second set of equations, Figure H.2, concerns the geometrical arrangement for housing an array of electrodes of a brominator. The simplest, least costly and most effective manner of housing an array to withstand high hydraulic pressures normally encountered in most application settings, is a cylinder. As was described in Section 5.1, the relationship is between the dimensions of electrodes and cylinder diameter to give the largest utilization of available space within the cylinder to the greatest total electrode area.

MATHEMATICS FOR SIMPLE ELECTROCHEMICAL BIOCIDING

305

The short set of such relationships is £>= n2 ■ u ■ sAe=L-W-n-L-W

s

4 f i 2 = W2+b2 2b2- u ■ b -4R2=0

Figure H.2 Cylindrical configuration.

The above relates the maximum area, A , and the inside radius of the cylindrical housing. The parameters are identified as follows: b n Ae L W R u s

Stack thickness, in. Number of plates (electrodes) Operating electrode area, sq in Plate length, in Plate width, in Cylinder radius, in. Plate thickness Inter-electrode spacing, in.

To illustrate the usefulness of establishing these equations into a suitable mathematics program, the following is offered as a typical computation for a convenient size of "reactor module", or array of cells needed to supply sufficient bromine for a body of water with a volume of 40,000 gallons - a large swimming pool. Strictly as a matter of illustration, let us specify that about 1 ppm of bromine must be provided to this body of water when electrolyzing a water solution of about 1000 ppm of sodium bromide. That concentration of NaBr is a 0.0097 normal solution. If the above equations are entered into either a MathCad type or spreadsheet program and solved for various lengths of electrodes and diameter of cylindrical housings, we can manipulate the variables to arrive at a maximally acceptable configuration of electrode array.

306


Let us see what the physical dimensions and electrical power requirements estimates are for the following set of operating conditions. These electrodes are assumed to be all connected electrically in parallel making a single cell with a multiplicity of electrodes. uirements: Single cell r, = l ppm = 1000 ppm ppm-Br = 1 p p m / h r

E =3 s i = 0.10 amp/in 2 = 0.016 amp/cm 2 s = 0.25 in = 0.625 cm V = 40,000 gal.

From equations in Figure H.l, the values of the dependent variables are found to be, Bt = 145 gm/hr, A = 2950 cm2 = 461 in2 E = 11.2 volts, I = 47 amps, and P = 530 watts regardless of the number of cells or specific geometry of the unit as long as the electrode spacing, s, is kept the same. Now, in order to decide upon the actual number of cells and size and shape of the unit, we make use of the set of equations in Figure H.2. The only actual choice we have in placing an electrode assembly into a cylinder is the inside radius of the housing. The equations are set up to provide the values of L and W, and n for the maximum area in the shortest stack length, L. We will select two values of R to complete the illustrative computations, and to provide the reader with an idea of the size and power required to produce electrolytically 1 p p m / h r of bromine in 40,000 gallons of water at a coulombic efficiency of 100%.

MATHEMATICS FOR SIMPLE ELECTROCHEMICAL BIOCIDING

For R = 4 in: n = 17 plates

L = 5.1 in

W = 5.6 in For R = 2 in: n = 9 plates W = 2.8 in.

L = 21 in

307


Appendix F: Industrial Chlorination and Bromination Equipment Cost Estimates The processes and cost information described here are based upon current practices and available information. Equipment cost figures was estimated from engineering experience and cost figures was estimated from 1992 dollars using the Engineering News Record Construction and Cost Indexes (CCI). This feasibility study has produced an order-of-magnitude investment and operating cost estimate for a proposed bromination circuit. Both the chlorination and bromination water treatment processes have been sized to treat a three million gallon industrial water supply in an open, re-circulating system. It has been assumed that a recirculation rate of 100,000 gallons per minute (gpm) occurs in an induced-draft cooling tower.

271

272


Table E l summarizes the design and operating conditions chosen for this analysis and systems comparison. No forced blow-down in the bromination process design was assumed because of the water softening capabilities of the unit. Though no specific calculations or field tests have been conducted on this assumption, the experience with operating these types of devices supports this contention. Zero forced blow-down also minimizes bromide salt make-up requirements. Chemical storage facilities for chlorine gas and bromide salt have been provided to accommodate one tanker or truckload delivery of the gas or solid, respectively. No building storage or shelter, other than that indicated on the equipment list, was considered. Chlorine demand Table El Water treatment design assumptions and operating conditions. Chlorination

Bromination

Recirculation Rate - gpm

100,000

100,000

Turnover Rate - min

30

30

Tower Load - F

20

20

Evaporation - gpm

2,000

2,000

Windage Loss - gpm

100

100

Forced Blowdown - gpm

400

-

Total Blowdown - gpm

500

100

Total Makeup - gpm

2,500

2,100

Cycles of Concentration

5

21

Chemical Disinfectant

chlorine (g)

bromine (s)

Chemical Demand - ppm

3

2

Chemical Demand - lb / d a y

6,000

1,200*

*Expressed as sodium bromide.

EQUIPMENT COST ESTIMATES

273

was calculated based on average ranges reported and the industry's published "rules-of-thumb" for processes with no product contamination. 3 Bromine demand was assumed based on experience at TRL, Inc. The brominator module and power supply were sized assuming a bromine loss rate of 1 ppm per hour, requiring the generation of 25 lbs per hour of bromine. The efficiency (coulombic) of bromine generation was assumed at 50%, and the current density applied was 0.10 amps per square inch of cell area. All cells were connected in series at 5 volts per cell. The process demand is approximately 75 to 80% of the equipment design capacity. The following is a major equipment list for a chlorination circuit, which will supply 6,000 lbs/day of chlorine to an open, re-circulating cooling water system with a recirculation rate of 100,000 gallons per minute. Table F.2 is an Table F.2 Chlorination system equipment list. Equipment

Specification

Materials of Construction

Cost-1992

Cl2 Gas Storage Tank & Shield

4,800 gallons

MS

$25,000 E

Weigh Platform

30 ton

mf g std

23,900

Storage Tank Eductor

2 in. air oper.

mf g std

1,000

Air • • •

50 SCFM at 125 psi 50 hp

mfg std

32,200

Flexible Tank Connections 2 required

10 ft. L

Monel

2,000

Padding System air compressor air dryer aftercoolers receiver • air filter • moisture indicator

274


Table F.2 (cont.) Chlorination system equipment list. Equipment

Specification


Cost -1992

Expansion Tanks

20% header volume

MS

600

Injector

25 in. Hg vacuum

titanium

Chlorinator • inlet reducing valve • retameter • oriface meter • regulating valve • relief valve

8,000 lb/day

mfg std

Injector Pump

150 gpm 150 ft. 15 hp

Iron

2,100

Evaporator • temperature alarms • level & pressure switch

8,000 lb/day 12 KW

mfg std

57,900 E.

Residual Controller/ Analyzer / Monitor

-

mfg std

5,000

Safety Equipment • Cl 2 Leak Detector • Breathing Apparatus • Container Kits • Floor Level Fans (2)

10 hp @ 2500 acfm

mfg std frp

7,000

Total Erected E q u i p m e n t

$82, 900

Total Non-erected E q u i p m e n t

$86,600


275

equipment, materials listing along with their estimated costs. Below is a glossary of abbreviations found in the equipment list. dia frp hP lb gpm rpm sqft MS RL mfg std E El

diameter fiber-reinforced polyester horsepower pound, avoirdupois gallons per minute revolutions per minute square feet mild steel rubber lined manufacturers standard field erected

B r o m i n a t i o n E q u i p m e n t List

The following table is a major equipment list for a bromination circuit which will supply 1,200 lb/day of sodium bromide salt to the cooling tower system, and maintain a bromine concentration of 2 ppm to effect slime, algae and bacteriological control. The system has been specified Table F.3 Bromination system. Equipment

Specification


Cost -1992

NaBr Feed Bin w / load cell

45,000 lb

MS,RL

$24,600

NaBr Vibrating Feeder

8,000 lb/hr, 1/4 hp

MS, RL

2,900

NaBr Discharge

-

MS,RL

800

NaBr Solution Tank

4,000 gallon, 9 ft dia, 10 ft deep

MS,RL

17,000

276


Table F.3 (cont.) Bromination system. Equipment

Specification


Cost -1992

NaBr Mixing Agitator

32 in dia impeller 5 hp,125 rpm

MS,RL

5,500

Feed Water Pump

100gpm-30 ft 4hp

RL

7,400

Solution Feed Pump

variable gpm50ft,l/2hp

mfg std

2,800

Pneumatic Conveyor System

5 ton/hr capacity

mfg std

14,700

Bromination Channel Tank

3 ft x 2 1/2 ft xl9ft

frp

6,100

Bromination Modules • 12 modules/10 cells per • 1200 sq ft plates

5 sq ft per plate

carbon plates copper bus

7,500

Residual Analyzer/ Recorder /Monitor pH Analyzer Power Supply

5,000 " 150 amps 240 volts 36 KW

mfg std

Total Erected E q u i p m e n t

$24,600

Total Non-erected E q u i p m e n t

$79,700

10,000


277

for an open, re-circulating cooling water system with a recirculation rate of 100,000 gpm. F.2

Capital Cost A n a l y s i s

Capital costs were developed from the major equipment lists and preliminary capital cost estimating techniques and factors. Installation labor costs were estimated based upon purchased equipment costs. Installed instrumentation and control costs were based upon equipment and process type; 30% of the purchased equipment costs for the chlorination system and 20% of the purchased equipment costs for the bromination system. This cost difference is required when the safety precautions necessary in the chlorine supply, handling and metering system are examined. Because the chlorine is handled as a liquid and gas in duplicate header systems in the storage area, and as a gas in the metering system, the piping costs associated with the chlorination process are greater; 40% of purchased equipment costs, versus 30% for the bromination circuit. Electrical costs are estimated at 10% of equipment costs for the bromination circuit, and 5% of equipment costs for the chlorination costs. Because of the complexity of the power supply and circuitry associated with brominator modules, it is estimated that the electrical installation costs for the bromination circuit would be a significant cost. Services and platform construction is necessary for both processes, and include lighting, site preparation and access platforms to the storage facilities. Both costs are approximately 30% of the purchased equipment costs. All of the above costs, plus the equipment costs, make up the total construction costs'4-5'. As a check against estimating techniques developed in this treatment, the construction costs figures published by the US EPA in January 1978 for chlorine storage and feed systems were referenced. Escalated to 1992 dollars, the construction cost for an 8,000 lb / d a y capacity chlorine system

278


was approximately $332,0006. The total construction costs estimated for this exercise is approximately $363,000. Because the proposed bromination system is a new treatment method for industrial water systems, there are no construction data available for comparison. Indirect costs, such as engineering and supervision, construction expenses and profit, administration and legal fees, and contingency were taken as percentages of the total construction costs usually associated with water treatment facilities6. In the bromination circuit, an initial charge of bromide salt was included in the indirect costs as a one-time fee. The interest charges were calculated based on estimated construction times and an 8% interest rate. For the chlorination system, the fixed-capital investment is estimated at $607,500. The bromination system is estimated to cost $375,000, approximately 60% of the chlorine circuit. The following tables (F.4 and F.5) list the individual cost components and charges for each system.

Table F.4 Chlorination system fixed capital summary. Total Non-erected Equipment

$86,000

Total Erected Equipment

82,900

Installation Labor

13,100

Instrumentation & Controls (installed)

50,700

Freight & Site Handling

3,400

Piping (installed)

67,600

Electrical

8,400

Service, Platforms

50,700

Total Construction Costs (TIC)

362,800

Contingency @15% TCC

54,400


Table F.4 (cont.) Chiorination system fixed capital summary. Contractor Overhead & Profit @ 20% TCC

43,500

Engineering @ 13% TCC

47,200

Legal, Fiscal & Administrative @ 20% TCC

72,600

Subtotal

580,500

Interest During Construction @ 8%

27,000

Fixed Capital Investment

607,500

Table F.5 Bromination system fixed-capital summary. Total Non-Erected Equipment

$79,700

Total Erected equipment

24,600

Installation Labor

13,000

Instrumentation & Controls (installed)

20,900

Freight & Site Handling

2,000

Piping (installed)

33,000

Electrical

10,600

Service, Platforms

33,000

Total Construction Costs (TCC)

216,800

Contingency @ 15% TCC

32,500

Contractor Overhead & Profit @ 12% TCC

26,000

Engineering @ 13% TCC

28,200

Legal, Fiscal & Administrative @ 20% TCC

43,400

Sodium Bromide Initial Charge, delivered 27,000 lb

16,200

Subtotal

363,100

Interest During Construction @ 8%

12,000

Fixed-Capital Investment

375,100

279

280

F.3


O p e r a t i n g Cost A n a l y s i s

Operating costs were calculated based on factored estimates delivered from the fixed capital investment. Percentages typical of water treatment facilities were utilized for plant life and depreciation, maintenance and insurance. Raw material costs were restricted to bromide and chloride salts for each respective process. Corrosion inhibitors, chemical cleaners and flocculants were not included. Connect hp and make-up water requirements from evaporation, windage and forced blow-down estimates were used to calculate utility charges. Administrative charges include office and laboratory labor, and operation charges include labor and engineering. The following operating cost summary in Table F.6 indicates no appreciable operating cost difference between chlorination or bromination under the design basis and assumptions outlined earlier. Because of safety considerations and maintenance and insurance increases possibly Table F.6 System o Derating cost comparison. Bromination Usage/yr Depreciation @ 10% TFI

$/yr

-

60,700

1050 tons

210,000

Chlorination Usage/yr

$/yr

-

37,500

210 tons

252,000

Maintenance @ 4% TFI Insurance @ 2% TFI Chemicals (delivered) • Chlorine® $0.10/lb • Sodium Bromide @ $0.60/lb


281

Table F.6 (cont.) System operating cost comparison. Bromination Usage/yr

$/yr

Utilities • Cooling 1,260 MM Gal 88,200 water @ $0.07/ 1,000 gal • Electricity® 0.30 MM 15,000 KWH $0.05/KWH Labor • Administrative • Operation Total Operating Cost

-

37,800 109,200

Chlorination Usage/yr

$/yr

1,058 MM Gal 74,100

0.36 MM KWH

-

$557,300

18,200

37,800 89,600 $531,700

associated with the storage and handling of chlorine, a possible operating cost increase in the estimate of $35,000 to $40,000 would not be unreasonable. However, this would not be a significant difference in the operating cost comparison for this order of magnitude estimate.

F.4 Conclusions and Comments The preliminary capital investment and operating cost estimates have indicated that the bromination system for disinfection and control of micro-organisms, algae, and slime growths in re-circulating water systems is a cost effective alternative to chlorination. A capital savings of approximately 40% is realized with the installation of the bromination process versus the chlorination equivalent. The assumptions for the design of the brominator and power supply were chosen to provide a severe duty estimate for equipment selection. Because of the water-softening action, iron, manganese and other dissolved metal removal

282


achieved by the brominator unit, it is assumed no forced blow-down will be required in this system. This assumption significantly decreases make-up bromide salt requirements, but needs to be verified in an actual operating system. When demineralized water is used for tower make-up, the tower cycles are limited only by windage losses, and the assumption of zero forced blow-down would be completely valid for both operations. There are many significant advantages to bromination over chlorination in water treatment applications. In the past, use of bromine biocides were limited to uses where chlorine disinfection was unacceptable because of higher chemical costs of bromine, and the oxidizing agents needed to produce the bromine residuals. With an electrolytic bromination system, based upon electrochemical "activation" of the bromide salts, the higher operating costs of bromination have essentially been eliminated. Because bromine is supplied in relatively inert salt form as sodium bromide, the storage and handling hazards associated with elemental halogens do not exist in this system. Operator safety training, control and safety equipment, system maintenance, leak detection apparatus are not required in the storage of bromide salts. The use of this system should also favorably impact insurance costs because of reduced safety hazards to operating personnel. Producers of bromine chemicals, such as l-bromo-3chloro-5, 5 dimethylhydantoin (Aquabrome), are actively pursuing industrial cooling water applications of bromine disinfection. References in the following describe the results of field tests and operations in several water-cooling water facilities. Another bromine biocide, 2.2-dibromo-3-nitrilo proprionamide (DBNPA), found to be an effective disinfectant, with no environmental toxicity because of it accelerated decomposition upon the application of heat or the increase in pH (78) . The data taken with these compounds have shown the effectiveness and advantages of bromine disinfection in open, re-circulating water systems.

EQUIPMENT C O S T ESTIMATES

283

Table F.7 Industrial re-circulating water rates. Industry

Recirculation Rate

200 MW Power Station

120,000 gpm

900 MW Power Station

405,000

200,000 MTPY Lead Production Hydrometallurgical Process Ammonia Production Facility (Capacity Unknown) Aquitance Refinery (Capacity Unknown)

5,300 36,000 292,000

Due to environmental regulations concerning toxicity and thermal pollution, cooling systems are being built by industry in an effort to meet government standards. In the power industry, two thirds of the fuel supplied must be dissipated as waste heat, requiring large cooling systems, and many of which have recirculation rates over 200,000 gpm. Cooling water requirements and quantities vary greatly, depending upon the industry. Table F.7 outlines several industrial systems cooling tower recirculation rates and production capacities.

References 1. White, George C , Handbook of Chlorination, page 712, Van Nostrand Reinhold Co., New York, 1972 2. Colturi, T. F, and Kozelski, K.J., "Corrosion and Biofouling Control in a Cooling Tower System", Material Performance, Vol. 23, No. 8, pages 43-47, August 1984. 3. White, pages 557-558. 4. White, pages 711-712. 5. Peters, Max S., and Timmerhaus, Klaus D., Plant Design and Economics for Chemical Engineers, pages 100 -141.

284


6. Montgomery, J.J., Water Treatment Principles and Design, pages 656-673. 7. Colyuri, pages 43-47. 8. Cappeline, G.A., and Carroll, J.G., "Enhance Your Cooling Systems's Performance Through Proper Use of Microbiocides," Power, Vol. 121, No. 10, pages 56-61, October 1977.


Appendix G: Design Mathematics in Computer Format The groups of mathematical equations developed in Chapter 4 are listed here, along with their corresponding lists of variables (basic parameters), for the convenience of those readers that wish to perform design estimates of operating ED systems. The information contained here is also useful is their implementation into computer files and programs. Because all the mathematical derivations and assumptions upon which they are based are to be found in the text, the reader can modify these expressions to reflect other analytical approaches or to allow other assumptions to be made. Three basic analyses will be given in this appendix. They are based upon these types of configurations: Case-A Once through, multiple cell array Case-B Re-circulation array with constant current Case-B Re-circulation array with constant voltage 285

286


In the first case, A, the idea of either constant current or constant voltage is not relevant in the analysis since the composition of the water flowing through the cells remains constant after steady state conditions are met. The dc power supply will merely provide whatever the current demand is at the impressed cell or stack voltage - after steady state is achieved. The analysis is set up such that the cell voltage or cell current at the input end can be set to a predetermined value. For the derivation of the following listed equations for Case-A refer to Chapter 4, Sections 4.2 and 4.3. The set of relationships employed here are direct results of those developed earlier and listed in Equations (4.35). In the re-circulating configurations, B and C, the mathematical representation is quite different, and the composition, TDS, of the re-circulating water changes with time. Hence, one must decide whether to establish a constant current, constant voltage, constant current or voltage limited, etc. system before the analysis and numerical computations can begin.

G.l

Case A

A continuous water flow-through configuration through a module establishes a set of mathematical expressions that were developed in Section 4.3, and are listed below. These relationships in the form given in Table G.l are set up for simultaneous solution in programs with the capability of performing such computations. The variables and constants employed throughout are identified and defined below in Table G.2. Let us examine the set of variables and constants in Table G.2. The last two columns in the table list the typical options of either initially setting a value for a parameter such as width of membranes or stack voltage, or setting some other parameter and solving for that particular one.

APPENDIX G: DESIGN MATHEMATICS IN COMPUTER FORMAT

287

Table G.l Equation group for case-A. Once flow-through design. v = 23\-F/W-p-n x =Vc;e/V-P

{

i-Ec-ri-ix/

Electrochemical Water Processing

Electrochemical Dictionary

Electrochemical Dictionary

Electrochemical Sensor Analysis

Fundamentals of Electrochemical Corrosion

Electrochemical Toxicity Sensors

Electrochemical Process Simulation III

Fundamentals Of Electrochemical Deposition

Electrochemical Thermodynamics and Kinetics

Electrochemical Impedance Spectroscopy

Electrochemical Activation of Catalysis

Electrochemical Phase Formation and Growth

Fundamentals of Electrochemical Deposition

Electrochemical activation of catalysis

Scanning Electrochemical Microscopy

Electrochemical methods for neuroscience

Encyclopedia of Electrochemistry, Electrochemical Engineering

Electrochemical Processes in Chemical Industries

Encyclopedia of Electrochemistry, Electrochemical Engineering

Electrochemical studies of biological systems

Water

Electrochemical Aspects of Ionic Liquids

Electrochemical methods. Fundamentals and applications.pdf

Water

Water

Water

Water

Water

Electrochemical Phase Formation and Growth

Water Encyclopedia - Ground Water

Electrochemical Water Processing

Electrochemical Dictionary

Electrochemical Dictionary

Electrochemical Sensor Analysis

Fundamentals of Electrochemical Corrosion

Electrochemical Toxicity Sensors

Electrochemical Process Simulation III

Fundamentals Of Electrochemical Deposition

Electrochemical Thermodynamics and Kinetics

Electrochemical Impedance Spectroscopy

Electrochemical Activation of Catalysis

Electrochemical Phase Formation and Growth

Fundamentals of Electrochemical Deposition

Electrochemical activation of catalysis

Scanning Electrochemical Microscopy

Electrochemical methods for neuroscience

Encyclopedia of Electrochemistry, Electrochemical Engineering

Electrochemical Processes in Chemical Industries

Encyclopedia of Electrochemistry, Electrochemical Engineering

Electrochemical studies of biological systems

Water

Electrochemical Aspects of Ionic Liquids

Electrochemical methods. Fundamentals and applications.pdf

Water

Water

Water

Water

Water

Electrochemical Phase Formation and Growth

Water Encyclopedia - Ground Water

Recommend Documents