Procedures Manual for Polymer Selection in Water Treatment Plants

American Water Works ; Association Procedures Manual for Polymer Selection in Water Treatment Plants Subject Area: Water Treatment PROCEDURES MANU...
Author: Bethany Atkins
2 downloads 0 Views 3MB Size
American Water Works ; Association

Procedures Manual for Polymer Selection in Water Treatment Plants

Subject Area: Water Treatment

PROCEDURES MANUAL FOR POLYMER SELECTION IN WATER TREATMENT PLANTS

Prepared by: Steven K. Dentel Beth M. Gucciardi Todd A. Sober Prasanna V. Shetty and John J. Resta Department of Civil Engineering University of Delaware Newark, DE 19716

Prepared for: AWWA Research Foundation 6666 West Quincy Avenue Denver, CO 80235

October 1989

Published by the AWWA Research Foundation and the American Water Works Association

DISCLAIMER

This study was funded by the American Water Works Association Research Foundation (AWWARF). AWWARF assumes no responsibility for the content of the research study reported in this publication, or for the opinions or statements of fact expressed in the report. The mention of tradenames for commercial products does not represent or imply the approval or endorsement of AWWARF. This report is presented solely for informational purposes. Although the research described in this document has been funded in part by the United States Environmental Protection Agency through a Cooperative Agreement, CR-811335-01, to AWWARF, it has not been subjected to Agency review and therefore does not necessarily reflect the views of the Agency and no official endorsement should be inferred.

Copyright 1989 by AWWA Research Foundation American Water Works Association Printed in U.S.

ISBN 0-89867-481-6 11

TABLE OF CONTENTS

Page TABLES

.............................. v

FIGURES ............................. vii FOREWORD

............................

ACKNOWLEDGMENTS

I.

II.

III.

IV.

ix

......................... x

EXECUTIVE SUMMARY ........................

xi

INTRODUCTION A. Purpose of this Manual ................... B. Polymer Types, Terminology, and Testing ........... C. Improving Water Quality with Organic Polymers ........ D. Possible Drawbacks to Polymers ............... E. Effect of Chemical Additives on Sludge Handling Processes . . F. References. .........................

1 3 14 16 17 19

HOW A. B. C. D. E. F. G.

TO USE THIS MANUAL Examining the Plant's Treatment Processes .......... Locating Potential Suppliers and Products .......... Assistance from Polymer Suppliers .............. Preparing to Evaluate the Products ............. The Module Format: Determining Modules to Use ....... Quality Control and Analysis of Results ........... References. .........................

21 24 26 27 28 36 38

MODULES FOR SPECIFIC SELECTION PROCEDURES A. Preparation of Polymer Solutions .............. B. Rapid Screening Procedure for Flocculants .......... C. The Jar Test ........................ D. Turbidity Measurement for Jar Test Assessment ........ E. Particle Size Analysis for Jar Test Assessment ....... F. Paper Filter Test ...................... G. Analytical Methods for Filter Tests ............. H. Determination of Sludge Volume Following Jar Tests ..... I. Preparation of Sludge Samples ................ J. Time to Filter Test ..................... K. The Capillary Suction Time Test ............... L. Sludge Jar Test ....................... M. Charge Density Determination. ................ N. Viscosity and Molecular Weight Determination. ........

39 59 62 87 91 95 119 125 131 135 145 155 159 181

USING MODULE RESULTS FOR CHEMICAL AID SELECTION A. Results Evaluation ..................... 203 B. Cost Estimating ....................... 204

111

LIST OF TABLES Table

Page

1

Descriptions of Specific Polymers ............ 7

2

Polymeric Structures and Chemical Abstract Service Numbers ....................... 9 General Description of Polymers by Usage Application Type .................. 10 Stock Solution Preparation. Final Volume: 1 L Stock Solution .................. .43 Working Solution Preparation. Final Volume: 200 mL Working Solution ................ .43 Jar Test Solution Preparation. Final Volume: 2 L Solution for Jar Testing ............. .44 Jar Test Solution Preparation. Final Volume: 1 L Solution for Jar Testing ............. .44 Example Tabulation of Jar Test Results Varying Flocculant Dose ................... 75 Example Tabulation of Jar Test Results Varying Coagulant Dose with Constant Dose of Flocculant ..................... 77 Blank Data Sheet for Tabulation of Jar Test Results When Varying Flocculant Dose with Constant Coagulant Dose .............. .81 Blank Data Sheet for Tabulation of Jar Test Results When Varying Coagulant Dose with Constant Dose of Flocculant ............. 82 Example Tabulation of Paper Filter Test Results Varying Filtration Aid Dose ............ 105 Blank Data Sheet for Tabulation of Paper Filter Test Results When Varying Filtration Aid Dose . . . 107 Example of Tabulation of Paper Filter Test Results When Varying Flocculant Dosage in Preceding Jar Tests ..................... 113 Blank Data Sheet for Tabulation of Paper Filter Test Results When Varying Flocculant Dosage in Preceding Jar Tests ................ 116 Suggested Sludge Volume Test Data Recording Sheet . . . 128 Suggested Time to Filter Test Data Recording Sheet . . . 140 Suggested Capillary Suction Time Test Data Recording Sheet .................. 149

3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18

LIST OF TABLES (Cent.)

19 20 21 22 23 24 25 26 27 28 29 30

Example Capillary Suction Time Test Data Recording Sheet ....................... Data Sheet for Standardization of PVSK Titrant .... Sample Data Sheet for Standardization of PVSK Titrant ...................... Data Sheet for Titration of Cationic Polyelectrolyte with Standardized PVSK Titrant . . . Sample Data Sheet for Titration of Cationic Polyelectrolyte with Standardized PVSK Titrant . . . Data Sheet for Back-Titration of Anionic Polyelectrolyte with Standardized PVSK Titrant . . . Sample Data Sheet for Back-Titration of Anionic Polyelectrolyte with Standardized PVSK Titrant . . . Data Sheet for Determining the Average Pure Solvent Flow Time, to ............... Data Sheet for Reduced Viscosity Determination. .... Sample Data Sheet for Reduced Viscosity Determination. . . Chemical Cost Estimating Worksheet .......... Chemical Cost Estimating Example Worksheet. ......

VI

150 166 167 172 173 177 178 188 190 195 207 212

LIST OF FIGURES

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17

Points in Water Treatment for Addition of Coagulants, Flocculants, Filtration Aids, and Sludge Conditioning Aids ............. Typical Water Treatment Process Diagram, and the Modules Which Simulate Chemical Effects on These Processes .................. Procedure for the Selection of Coagulants or Flocculants .................... Procedure for the Selection of Filtration Aids .... Procedures for Evaluating the Effect of Coagulants, Flocculants, or Sludge Conditioning Aids on Sludge Properties ...... Dimensions of 2-L "Square" Jar ............ Example of Graphed Results of Jar Tests Varying Flocculant Dose .................. Example of Graphed Results of Jar Tests Varying Coagulant Dose with Constant Dose of Flocculant .................... Example of Graphed Results of Multiple Jar Tests with Flocculant Dose Varied in Each Set, and Coagulant Dose Varied Between Sets ........ Blank Graph for Plotting Jar Test Results When Varying Flocculant Dose with Constant Coagulant Dose .................. Blank Graph for PlottingsJar Test Results When Varying Coagulant DoseNwith Constant Dose of Flocculant .................... Paper Filter Test Assembly .............. Example of Graphed Results of Paper Filter Tests ......... with Filtration Aid Dose Varied Blank Graph for Plotting Paper Filter Test Results When Varying Filtration Aid Dose ......... Jar Type Needed for Filtration Studies Subsequent to the Jar Test .................. Example of Graphed Results of Paper Filter Tests Varying Flocculant Dosage in Preceding Jar Tests ....................... Blank Graph for Plotting Paper Filter Test Results When Varying Flocculant Dosage in Preceding Jar Tests .....................

vii

4 30 33 34 35 64 76 78 79 83 84 100 106 108 110 114 117

LIST OF FIGURES (Cont.) Figure 18 19 20 21 22 23 24 25 26 27

Imhoff Cone ...................... Large Volume Time to Filter Equipment ......... Small Volume Time to Filter Equipment with Side Arm Adapter .................... Sample Time to Filter (TTF100) Data Plot ....... Capillary Suction Time (CST) Apparatus ........ Sample Capillary Suction Time Data Plot ........ Assembled Apparatus for Titration ........... Proper Vertical Orientation of Viscometer ....... Graph Used for Determining Intrinsic Viscosity (IV) ....................... Sample Graph Used for Determining Intrinsic Viscosity (IV) ..................

viii

126 136 138 142 146 151 164 184 193 198

FOREWORD The AWWA Research Foundation is a non-profit corporation that is dedicated to the implementation of a research effort to help local utilities respond to regulatory requirements and traditional high priority concerns of the industry. The research agenda is developed through a process of grass roots consultation with members, utility subscribers, and working professionals. Under the umbrella of a Five-Year Plan, the Research Advisory Council prioritizes the suggested projects based upon current and future needs, applicability, and past work; the recommendations are forwarded to the Board of Trustees for final selection. This publication is a result of one of those sponsored studies and it is hoped that its findings will be applied in communities throughout the world. The following report serves as a means of not only communicating the results of the water industry's centralized research program but also as a tool to enlist the further support of the non-member utilities and individuals. Projects are managed closely from their inception to the final report by the Foundation's staff and large cadre of volunteers that willingly contribute their time and expertise. The Foundation serves a planning and management function and awards contracts to other institutions, such as water utilities, universities and engineering firms. The funding for this research effort comes primarily from the Subscription Program. The program, through which water .utilities subscribe to the research program and make an annual payment proportionate to the volume of water they deliver, offers a cost-effective and fair method for funding research in the public interest. A broad spectrum of water supply issues are addressed by the Foundation's research agenda: resources, treatment and operations, distribution and storage, water quality and analysis, toxicology, economics, and management. The ultimate purpose of the coordinated effort is to assist local water suppliers to provide the highest possible quality of water economically and reliably. The true benefits are realized when the results are implemented at the utility level. The Foundation's Trustees are pleased to offer this publication as a contribution toward that end. This guidance manual is the successor to an earlier Foundation publication, Procedures Manual for Selection of Coagulant. Filtration, and Sludge Conditioning Aids in Water Treatment. This improved new edition contains modifications based on utility use of the original procedures. Several new modules on polymer characteristic determinations have also been added.

es F. Manwarlng, P.E. xecutive Director AWWA Research Foundation

Richard P. Me Hugh Chairman. Board of Trustees AWWA Research Foundation ix

ACKNOWLEDGMENTS

The authors would like to thank the American Water Works Association Research Foundation for the financial support of this project. Helpful suggestions were provided by Jon DeBoer and Deborah Brink as Project Officers, as well as by the Project Advisory Committee consisting of Keith Carns, Ray Letterman, Kim Fox, and Gary Logsdon. This project could not have been completed without the invaluable assistance obtained from water treatment plant personnel at numerous facilities throughout the U.S.

x

EXECUTIVE SUMMARY

PROCEDURES MANUAL FOR POLYMER SELECTION IN WATER TREATMENT PLANTS

Prepared by: Steven K. Dentel Beth M. Gucciardi Todd A. Bober Prasanna V. Shetty and John J. Resta Department of Civil Engineering University of Delaware Newark, DE 19716 OBJECTIVES An increasing variety of chemical additives have become available for im proving the removal and management of particulate matter in water treatment. Many of these products are polymers, also termed organic polyelectrolytes. Used as coagulants, flocculants, filtration aids, and sludge conditioning aids, they are usually proprietary formulations, and their chemical compo sition (and other fundamental properties) are not provided to the user. Consequently, making a selection from these products to provide optimal treatment is quite difficult. Empirical, small-scale tests are often used in the laboratory to predict the full-scale performance of these additives as well as possible. However, these methods have numerous shortcomings: most are non-standardized, with differing versions presented in scattered literature references; some appear only in obscure references or in sources seemingly unrelated to water treatment; many have been poorly validated by comparison to full-scale results, or only validated when selecting inorganic coagulants rather than organic polyelectrolytes. Not only are polyelectro lytes more difficult to prepare and handle; they also vary widely in characteristics such as charge density and molecular weight. The first edition of this guidance manual, also published by the AWWA Research Foundation (Dentel et al., 1986), selected, assessed, and developed a bank of methods for evaluation of polymers used in water treatment. These methods were placed in a modular format to provide greater flexibility in combining and using them. However, the large number of methods and procedural variables limited the extent to which these methods could be evaluated, especially in full-scale use by treatment plant personnel. The methods were also limited to empirical means of simulating unit operations rather than characterizing inherent properties of the products.

xi

Thus, in developing this guidance manual, the following objectives were addressed: disseminate the original guidance manual to water treatment plant personnel interested in using and evaluating it. utilize comments from personnel at these plants to improve the utility of the manual. obtain correlations of product efficacy at plant scale with performance predicted by laboratory tests, and modify procedures in the manual appropriately. develop additional procedures to enable assessment of inherent polymer properties such as charge density and molecular weight. by comparing product characteristics, performance of products at various plants, and the raw water and process characteristics in these plants, develop relationships for predicting the product type(s) most likely to successfully improve processes under given conditions. RESULTS This guidance manual does not report specific findings in pursuing the above objectives, but rather incorporates them in an improved document. (A supporting document, available through the Research Foundation, provides more detailed results from participating water treatment facilities.) Although data obtained from these plants was insufficient to establish meaningful statistical correlations, many comments and qualitative results were utilized in making these improvements. For example, many users found the terminology in referring to polymeric products confusing; the document was therefore retitled and broadened in scope to include all polymers rather than solely the high molecular weight products previously termed coagulant, filtration, and sludge conditioning aids. Product terminology was clarified throughout the manual. The guidance manual includes the following sections: I.

INTRODUCTION A. Purpose of this Manual B. Polymer Types, Terminology, and Testing C. Improving Water Quality with Organic Polymers D. Possible Drawbacks to Polymers E. Effect of Chemical Additives on Sludge Handling Processes F. References

II.

HOW A. B. C. D. E. F. G.

TO USE THIS MANUAL Examining the Plant's Treatment Processes Locating Potential Suppliers and Products Assistance from Polymer Suppliers Preparing to Evaluate the Products The Module Format: Determining Modules to Use Quality Control and Analysis of Results References

xii

III.

MODULES FOR SPECIFIC SELECTION PROCEDURES A. Preparation of Polymer Solutions B. Rapid. Screening Procedure for Flocculants C. The Jar Test D. Turbidity Measurement for Jar Test Assessment E. Particle Size Analysis for Jar Test Assessment F. Paper Filter Test G. Analytical Methods for Filter Tests H. Determination of Sludge Volume Following Jar Tests I. Preparation of Sludge Samples J. Time to Filter Test K. The Capillary Suction Time Test L. Sludge Jar Test M. Charge Density Determination N. Viscosity and Molecular Weight Determination

IV.

USING MODULE RESULTS FOR CHEMICAL AID SELECTION A. Results Evaluation B. Cost Estimating

Section I.B, Polymer Types, Terminology, and Testing, was added to more clearly define terms used in describing polymeric additives; it also includes tabulations of generic chemical structures with physical and chemical properties for the most common polymeric products, and a listing of the different physical forms of such products along with appropriate dilution and concentration data. Section III.A, the module describing preparation of polymer solutions, was substantially revised and enlarged as a result of both plant and laboratory work. It now gives separate makeup procedures depending on the physical form of the additive as received. An earlier module utilizing a small bench-scale filter was eliminated since plant data showed that it provided erratic results not in agreement with full-scale filter performance. As also demonstrated in previous research, the use of pilot-scale filters gave much better prediction of full-scale results. Such pilot-scale units should include a cross-section of the actual filter media and depth but otherwise may be highly variable in design. Two modules were developed and added to the manual. These enable the charge density and intrinsic viscosity of polymers to be determined. The charge density procedure was derived from a method often referred to as a "colloid titration" in previous literature, but substantial modifications were necessary in order to assure ease of use, accuracy, precision, and use of readily available reagents of required quality. Intrinsic viscosity was adopted as a surrogate measure for molecular weight; although the procedure requires more technical expertise than do some of the other modules, it is much easier than other methods such as light scattering and HPLC (high performance liquid chromatography). The charge density and intrinsic viscosity (or molecular weight) information allow identification of polymers

xiii

by chemical type, comparison of charge concentration (e.g. as a function of cost), and provide an indication of whether two products are composed of similar or identical polymer types, even at different dilutions. This guidance manual is designed for ease of use in the laboratory of a water treatment facility. Introductory material orients the user with an introduction to polymers, their uses and benefits, and explains the modular format of the manual. Each module includes a list of required equipment and reagents. Data sheets and graphs are provided in a form which can be photocopied for use, and example tabulations and graphs are given. The manual will thus guide the user through the steps necessary to select the most appropriate polymer for coagulation, flocculation, filtration, or sludge conditioning while allowing the necessary flexibility for various plant configurations and levels of laboratory skill.

xiv

I.

A.

INTRODUCTION

PURPOSE OF THIS MANUAL

An increasing variety of chemical additives have become available for im proving the removal and management of particulate matter in water treatment. However, many of these coagulants, flocculants, filtration aids, and sludge conditioning aids are proprietary products, and their chemical composition (and other fundamental properties) are not provided to the user.

Conse

quently, making a selection from these products to provide optimal treatment is quite difficult.

In fact, due to an inadequate understanding of the

involved processes, even knowing the product characteristics does not enable a rational selection of chemical products and dosages. selection procedures are necessary.

Instead, empirical

These would ideally involve the appli

cation of each possible chemical in the full-scale plant under controlled conditions.

Obviously, this is not practical, especially considering the

large number of such products on the market.

The alternative is to utilize

small-scale tests which will predict the full-scale performance of these additives as well as possible. This manual sets forth recommended procedures for performing such smallscale tests.

They provide a practical means of selecting the most effective

coagulant, flocculant, filtration aid, or sludge conditioning aid for a given water and treatment scheme. maximum flexibility.

These procedures are organized for

For example, an entire sequence of methods can be used

to comprehensively evaluate the impact of a flocculant on sedimentation, filtration, and sludge handling processes.

Or, just one procedure can be

selected for use in an initial screening of products under consideration, and the most promising ones can then be evaluated in the plant. The procedures given in this manual will be familiar in some cases; for example, Section III.C covers use of the jar test.

However, since this

manual is primarily oriented towards selection of organic polyelectrolytes or "polymers", adjustments have been made which address handling and testing difficulties associated with the evaluation of these materials.

In other cases, test procedures have been adopted either from wastewater treatment applications or from methods more common in research laboratories. In these instances, the recommended test methods have been selected or developed based on extensive comparisons with other variations and proce dures, and after assessing the method's capability for adequately predicting scale-up performance. This manual deals primarily with polymers because of their increasing use in water treatment applications even though there is common confusion about their characteristics, selection, and proper use.

In order to facilitate a

better understanding of polymers, specific test methods have been included in this manual for characterizing some of their properties. At the same time, many of the other test methods also apply to products such as alum, and allow comparison of such products with competing organic polymers. It should be stressed, however, that none of the procedures in this manual are foolproof.

Unavoidable effects of scale and the inevitable complexities

of actual plant operation (for example, the possibility of hydraulic shortcircuiting in the sedimentation basin) will always limit the extent to which laboratory test results can be trusted in predicting plant results. Some compromise has also been necessary in keeping to the goals of utility and relative simplicity for these procedures.

In general, the results of these

tests can be relied upon to give a reasonable prediction of which chemical additive will give the best results, and at what dose.

Quantitative predic

tion of full-scale performance, unfortunately, is beyond the ability of these procedures.

For example, the methods for selecting a filtration aid

will indicate a filtration aid dose that should do the best job of reducing the filtered turbidity, but can't be relied on to give a good prediction of what this improved turbidity will be. The remaining sections of this Chapter provide a brief introduction to coagulants, flocculants, filtration aids, and sludge conditioning aids. This includes some consideration of when their use may or may not be benefi cial.

More in-depth information can be found in references such as the AWWA

publications Polvelectrolytes - Aids to Better Water Quality (AWWA 20121) and Use of Organic Polvelectrolytes in Water Treatment (AWWA 20173) . Chapter II then gives a general idea of how to go about evaluating these chemical products for use in a treatment plant, and how this manual may be of assistance.

The detailed testing procedures are then presented in

Chapter III, with Chapter IV explaining how results from several different procedures can be integrated into an overall analysis of how a given polymer will affect plant processes. B.

POLYMER TYPES, TERMINOLOGY, AND TESTING

1.

Product Terminology

Figure 1 gives a flow chart for "conventional" water treatment that includes coagulation, filtration, and solids handling processes.

Coagulation, floc-

culation, and sedimentation first remove the bulk of the suspended matter contained in the raw water; filtration then takes out most of the remaining particulates.

The removal of other components such as color or THM precur

sors may also be an important objective of these processes.

In some cases,

a raw water is low enough in turbidity that sedimentation--and sometimes flocculation--can be eliminated; this is then termed direct or in-line filtration.

Plants which remove hardness by lime precipitation may add lime

at the point shown in Figure 1 for coagulant addition, or may employ a more complex flow diagram, but the same processes and additives shown in Figure 1 are generally applicable.

No matter how the suspended impurities are

removed, however, they must then be concentrated and disposed of in some manner; the solids handling processes for this purpose can include lagoons, drying beds, vacuum filters, centrifuges, and a variety of other mechanical processes. Efficiency of the processes shown in Figure 1 can often be improved by the use of various additives called coagulants, flocculants, filtration aids, and conditioning aids.

This manual is concerned with all of these.

Figure

1 also indicates where these substances are typically added in a water

RAPID MIX

I

DEWATERING

SEDIMENTATION

FIGURE 1. Points in Water Treatment for Addition of Coagulants, Flocculants, Filtration Aids, and Sludge Conditioning Aids

Conditioning Aid

FLOCCULATION

Flocculant

Coagulant

FILTRATION

Filtration Aid

treatment plant.

Strictly speaking, the point of addition shown in this

diagram determines the name applied to a given polymer; for example, a product applied in rapid mix is thus a "coagulant."

However, the way that

the product performs its function -- and therefore the product's chemical characteristics -- also must be considered in applying the proper terminology.

This knowledge is important in developing an appropriate

chemical additive strategy as well. A coagulant is usually added prior to or into the rapid mix.

Whether it is

alum, a ferric salt, a polymer, or otherwise, its usual function is to neutralize the charge on particles and other raw water constituents. additive is sometimes called a primary coagulant.

This

When polymers are used

for this purpose, they are typically low molecular weight and positively charged (cationic). A flocculant then functions by bridging or enmeshing the neutralized particles into larger floe particles.

Coagulants such as alum and ferric

salts also act as flocculants at high enough dosages.

Polymers used

specifically for this purpose, however, differ from those used as primary coagulants: they possess a much higher molecular weight and may have a charge that is positive, negative (anionic), or neutral (nonionic). flocculant may also be referred to as a coagulant aid.

A

Other materials may

also be defined as coagulant aids without being considered as flocculants, such as activated silica, sodium alginate, and bentonite clay. Filtration aids are polymers added to water prior to filtration.

They have

much in common with flocculants in terms of both structure and function.

In

direct filtration processes, a coagulant may also be utilized prior to filtration, but this addition is generally prior to or into the rapid mix step. Chemicals added to a sludge to improve its dewaterability may be termed sludge conditioners or conditioning aids.

Although ferric chloride and lime

are employed for the conditioning of wastewater treatment sludges, polymers are more commonly used in conjunction with water treatment sludges.

Such

polymers are similar in structure and function to polymeric flocculants, and may be sometimes be called flocculants. 2.

Polymer Characteristics

The organic polymers used in water treatment are chains of individual monomer units, linked together in a linear or branched configuration, with functional groups attached periodically along the chain. The functional groups may possess a negative charge (anionic polymers), positive charge (cationic polymers), or an overall neutral charge (nonionic polymers). These products are also referred to as "polyelectrolytes" although the term strictly applies only to the cationic and anionic types (since it concerns the many charges on these molecules).

The length of the polymer chain (in

cluding branches) is indicated by the molecular weight of the polymer, but it should be remembered that this is actually an overall average for the many individual polymer molecules. Tables 1, 2, and 3 provide detailed information on the types of polymers used in given water treatment applications. Table 1 provides eight chemical formulas that essentially describe the available polymers for use in water treatment.

Table 2 provides the chemical structures for each of these. The first two polymer types on these tables, PDADMAC and Epi/DMA, are used in the greatest quantities, and AWWA standards for these types are being developed.

Table 3, which groups the polymers by application, shows that these are used most commonly as primary coagulants. Even though the number of polymer types is small, other characteristics may also vary such as molecular weight and product activity, as shown by Tables 1 and 3. These tables will be referred to in later sections of this manual as well. In spite of the apparent complexity of these three tables, polymers basically may be grouped into two general types: (1) cationic low molecular weight (coagulants), and (2) high or very high molecular weight, regardless of charge (flocculants, filtration aids, and conditioning aids).

very strong; (+) only at low pH, (0) at high pH if quaternized, none; if not, strong: (0) at intermediate pH, (-) at high pH, (+) at low pH

-PDADMAC, PDMDAAC, quaternary amine -Epi/DMA (epichlorohydrin/dimethyl amine) , quaternary amine -PEA

-Mel/Form (melamine/ formaldehyde)

-PDADMAC/ DMA, mixture of quaternary amines

1. Poly(di'allyldimethyl ammonium chloride)

2'. Poly(2-hydroxypropyl-l,N,Ndime thy 1 ammonium chloride)

3. Poly (ethyleneamine

4. Poly [N- (dimethylam inomethyl) acrylamide] «

5. Poly(diallyldime thy 1 ammonium chloride/dimethylamine) copolymer

none

none

none

Synonyms

Chemical Name

Influence of pH on Charge

10-50 % (v/v)

6-10 % (V/V)

15-50 % (v/v)

25-50 % (V/v)

10-40 % (v/v)

»

'

»

"

20-5,000

Product Viscosity, cp Activity

Descriptions of Specific Polymers:

TABLE 1

dark yellow liquid

cloudy, faint blue liquid

clear liquid

amber-colored liquid

straw-colored or pale yellow liquid

Physical Description

oo

(Continued)

methacrylate)

methylaminoethyl-

8. Poly(N,N-di-

7. Poly(acrylamide/ acrylic acid) copolymer

6. Poly(aerylamide)

TABLE 1

quaternized polyamine

-PAM/PAA, poly(aery1amide/acrylate salt) copolymer, hydrolyzed PAM

-PAM

none

75-100 % (w/w)

200-700 25-30 (5-75) % (w/w)

5,0002-10 % (w/w) 70,000 or (v/v)

strong; slightly (-) at intermediate pH, (0) or slightly ( + ) at low pH, (+) at very low pH, (-) at very high pH

very strong; (0) only at intermediate pH, (-) at high pH, (+) at low pH

dry solid

dry solid white opaque viscous liquid or emulsion

dry solid, white opaque viscous liquid or emulsion

TABLE 2 Polymeric Structures and Chemical Abstract Service Numbers 2.

1. CAS #26062-79-3

-E--

CAS #25988-97-0, or 39660-17-8, or 42751-79-1

•CH, OH

H,C

/

\

r

-f

L_

CH,

CH,

CH.-CH - CH,N, I a"

CH,

3. CAS #9002-98-6, or 26913-06-4, or 25988-99-2

4.

CAS #9003-08-1

-.3—In

C

5. CAS #26590-05-6

6.

CH.

CAS #9003-05-8

CH,-CH

/ \ CH, H,C

C-NH,

C — NH,

I

J 0

O

7. CAS #31212-13-2, or 25085-02-3 -i-

L.

CH,-CH

J———U

i _!„

L_

CHj-CH

8.

CAS #25154-86-3 CH,

-I-

i _L

C — NH,

C — O'Na*

I

8 o

o

C

-TcH.C-LL.

|_Jn

C— OCH,CH,N CH,

Inverse -PAM/ > 30 or Latex acrylate million Emulsion copolymer (very Coagulant (hydrolyzed high) Aid PAM)

Anionic (-)

Nonionic (o)

Cationic (+)

Charge Type

0.5-2% (v/v)

5-20% (v/v)

none required

0.005-0.05% (v/v)

0.05-0.2% (v/v)

0.05-5% (v/v)

(v/v)

1-10%

Concentrations Resulting From: Primary Secondary Dilution Dilution

Coagulant -PAM > 30 Nonionic (o) Solid Aid (bead, -Hydrolyzed million Anionic (-) 0.1-0.8% Flocculant powder, PAM (very flake, -Quaternized high) Cationic (+) (w/v) granular) Polyamine

Liquid Emulsion

1 to 30 million (high)

-PAM

Relative Molecular Weight

Viscous Coagulant Liquid Aid Flocculant

Polymer Name

-PDADMAC 1 to 500 -Epi/DMA thousand -PEA (low) -Mel/Form

Synonyms

Liquid Organic Solution Primary Coagulant Coagulant

Polymer Type

0.00001-0.0001%(v/v ,0.1-1.0 ppm

0.00001-0. 0001% (V/V) 0.1-1.0 ppm

0.00001-0.001% (v/v) 0.1-10 ppm

0.00005-0.001% (v/v) 0.5-10 ppm

Typical Final Concentrations Used Full Scale or in Jar Tests

General Description of Polymers by Usage Application Type:

TABLE 3

The charge and molecular weight are important performance factors for both types of polymers.

Marketers of these products will often provide the type

of charge and the molecular weight range (medium, high, very high) for a given polymer product.

However, if the user is able to perform laboratory

measurements related to these properties, this may provide several benefits: First, the characteristics of concern can be quantified. for meaningful comparison of products.

This may be needed

For example, one supplier's

definition of "high" molecular weight may differ from another's.

And,

although hundreds of polymer products are marketed, many are repackaged and available through multiple sources.

Thus, polymer analyses might avoid

duplications in bench, pilot, or plant evaluations. Second, such measurements may be useful in conjunction with other polymer evaluation methods.

For example, charge or viscosity measurements can

indicate whether a polymer has been properly mixed and diluted (activated) in the laboratory.

A jar test performed with polymer that was inadequately

activated would certainly not give a good prediction of plant performance. Likewise, measurements can indicate how well activation is being achieved at the plant scale. Third, product specifications can be developed or checked to assure that the polymer being purchased has uniform characteristics.

Even with reputable

suppliers, such measurements might be useful; for instance, polymer characteristics can be checked if the allowable storage life is being approached. Finally, such tests can even be used to select products.

If it has been

established (perhaps by jar test comparisons) that polymer charge is the most important performance criterion, then polymers might be evaluated on the basis of "equivalents of charge per dollar." Two polymer characterization procedures are therefore provided in this manual.

The first of these is to determine the amount of charge, or charge

density, of a given organic polymer.

This method is a modification of the

11

"colloid titration" procedure which has been used in research, and is applicable to cationic or anionic charge. The second is a viscosity measurement, which can be related to molecular weight. these procedures are given in the appropriate modules.

Further details on

However, it should also be noted that other product characteristics may be important, such as the type of molecule acting as the "backbone" of the polymer chain, the actual molecular formula of the functional groups, the spacing of the functional groups along the chains, and the product form (solid, liquid, or emulsion). 3.

Laboratory Tests of Polymer Performance

Unfortunately, even knowing all of the polymer characteristics described in the previous section does not make it possible to predict which polymer will work the best under a given set of conditions.

This means it is necessary

to perform on-site tests of polymers for each specific application. In addition, an optimum dosage level should exist for a given polymer, with lower doses having less effect and higher doses either giving unfavorable results or leading only to slight improvements in water quality which are not worth the additional chemical expense. It is then important not only to find the best polymer but also to find its best dosage level. This will also provide a cost estimate for use of the given polymer, which then can be compared to the cost of alternative products. Thus the proper procedure for polymer selection should ultimately allow both improved water quality and reduced operational costs to be realized. Unfortunately, an important polymer characteristic is that they are partic ularly difficult to work with. They may come as powders, liquids, or emulsions (which appear as milky liquids) and must be prepared, diluted, and dosed in the proper manner. Some are particularly sensitive to light, chlorine, freezing, or biodegradation. Extreme turbulence (such as flash mixing) may cause them to break apart and lose their effectiveness.

12

Therefore, specific procedures are called for when testing polymers for possible use in a water treatment plant.

The test methods in this manual

have been developed to meet the particular requirements for polymer evalua tion, and can be used to assess polymer effectiveness in a number of applications.

The procedures can also be used to evaluate other non-

polymeric products (e.g. alum, lime, bentonite) in which case much of the procedure addressing polymer preparation and application can be ignored. Evaluation of filtration aids involves additional difficulties.

These

function much like flocculants (coating or bridging of particles) but also may affect the way in which particles interact with the filter media (sand, carbon, etc.) surfaces.

This includes the way that the waterborne particles

adhere to the media and the extent to which they penetrate or clog the filter. Consequently, in the case of filtration aid selection, the difficulties associated with handling and dispensing polymers are coupled with the problems of how to duplicate the filtration process and the effect of the filtration aid upon it.

Also important in any evaluation procedure for

direct or contact filtration is the simulation of rapid mix and/or flocculation processes.

As with coagulants, a dosage level should exist for a

given filtration aid which will provide the needed effluent quality at the most favorable price.

(Excessive filtration aid doses will usually decrease

the filter run length by increasing headless through the filter.) All of these considerations suggest that exacting procedures should be fol lowed if a filtration aid is to be evaluated in small-scale tests.

Adding

to the difficulty in this case is the fact that filtration mechanisms are not completely understood, so the similarity between laboratory and fullscale filter performance cannot always be assured.

However, experiments

have shown that trends in performance are usually duplicated.

What this

means for the evaluation of filtration aids is that polymer dosages indi cated by lab tests will be adequate for initial product selection, but that some "fine tuning" may be necessary when plant application is implemented.

13

C.

IMPROVING WATER QUALITY WITH ORGANIC POLYMERS

The procedures in this manual should be useful in a variety of applications, but they are mainly intended for the evaluation of polyelectrolyte additives which have not been previously used at a given treatment facility.

For such

evaluations, it will be advantageous if one first identifies the primary ob jectives of a possible change in chemical usage.

This in turn will enable

the most appropriate test procedures to be chosen for the polymer selection process. The personnel at a well-run water treatment plant are always on the lookout for ways of upgrading the operation; thus the objective of evaluating dif ferent additives might be the general desire to improve finished water quality and cost-effectiveness of the treatment processes.

However, it may

also be the case that specific problems exist at a plant, and different polymers are being evaluated in order to alleviate these particular dif ficulties . For example, if the coagulant is functioning poorly at low water tem peratures, flocculants might be evaluated in their ability to accelerate and complete flocculation under these conditions.

Here, the proper test pro

cedure would be a series of jar tests with the water temperature kept low using a water or ice bath. In fact, there are numerous situations in which polymer application might be beneficial.

These include:

Settled or finished turbidity consistently too high (for example, due to inadequate retention time in the flocculator or sedimentation basin). Polymers can help to produce larger, tougher floe which is easier to settle or filter.

If finished water quality is inadequate because the

plant is operating above design capacity, use of a flocculant or filtration aid can therefore avoid the capital expenses of expansion by allowing acceptable functioning at increased loading rates.

14

Turbidity periodically too high due to fluctuations in influent flow rate, turbidity, alkalinity, or hardness.

Because polymeric coagulants

do not act as acids the way inorganic coagulants do, and are also less pH-dependent in their effectiveness range, they may assist in maintain ing good water quality under such conditions.

Flocculants can also

clean up a poorly coagulated water in a manner that is fairly indepen dent of solids concentration.

For the same reasons, a flocculant may

lessen coagulant and pH control difficulties during fluctuating conditions. Inadequate removal of color or THM precursors.

Some polymers improve

removal of these, and jar or filtration tests followed by appropriate measurements (color, UV absorbance, etc.) can be used to select the most appropriate product for this purpose.

Specific methods are given in

this Manual for such an application. Treatment costs are too high due to the use of high dosages or too many additives.

If a flocculant can reduce the required dose of primary

coagulant (and also the amount of lime or other chemicals which may be needed for pH control), a net cost savings may be achieved. Excessive frequency of required filter backwashing caused by premature turbidity breakthrough.

A flocculant can alleviate this situation by

either reducing the amount of suspended material reaching the filters, or by favorably altering the properties of these solids.

A filtration

aid can also provide the second of these effects. Excessive sludge production or poor sludge properties.

Since a smaller

dose of flocculant can sometimes replace a larger portion of the aluminum or ferric coagulant dose, the volume of sludge produced can also be decreased.

The thickening and dewaterability characteristics of

such sludges are frequently improved as well.

Better water quality may

be realized as well due to improved sedimentation performance (e.g. less scouring or less down time for sludge removal) or cleaner recycle flows.

15

It should be understood, however, that polymers are not a "cure all" for all treatment problems.

Other changes in a water treatment plant's facilities

or practices may prove to be more effective at solving particular problems. Other AWWA publications should be consulted for additional means of im proving plant operations, such as Upgrading__Water Treatment Plants to Improve Water Quality (AWWA 20153) and Upgrading Existing Water Treatment Plants (AWWA 20126). D.

POSSIBLE DRAWBACKS TO POLYMERS

The use of polymeric coagulants, flocculants, and filtration aids can improve water treatment processes in a number of respects, as detailed in the previous section. However, there may also be disadvantages to the use of these additives which should be considered before the decision is made to employ them. Most of these involve the handling and operational aspects of polymer use: Storage, mixing, and feeding systems for polymers are significantly different than those used for inorganic coagulants. These facilities must be purchased and installed. Some polymers are supplied at very high or low pH and therefore appropriate plastic, fiberglass, or stainless steel con tainers must be specified. In some cases, the polymer must be mixed with warm water and an appropriate hot water heater will be required. Mixing with water containing a high chlorine residual will degrade some types of polymers, so a provision for dilution water other than finished water may also be necessary. equipment.

Highly viscous polymers will require heavy-duty pumping

(These materials also have the disadvantage of being quite

hazardous when spilled because they are extremely slippery.)

More

information on proper handling facilities for polymers can be obtained from polymer suppliers.

See also Use of Organic Polyelectrolvtes in Water Treatment (AWWA 20173).

Difficulties may be encountered in insuring product consistency when utilizing synthetic polymers, due both to the complexity of the polymeri zation processes used in manufacturing them, and to the vulnerability of

16

these products to various aging reactions.

Added to this difficulty is the

fact that many suppliers provide only minimal technical data on their prod ucts and often are not directly responsible for the manufacture of them.

It

may be necessary to perform laboratory analyses whenever a new batch of polymer is received in order to assure product consistency.

Such analyses

might include polymer viscosity determination or jar testing of a control suspension.

Another safeguard is to inquire of other area water treatment

plants as to the reputation of polymer suppliers under consideration. Trace levels of possibly harmful impurities are present in many polymer formulations.

Acrylamide monomer, for example, has been shown to be

mutagenic by the Ames test.

Concentrations of this impurity likely to be

created in finished water are not believed to pose any threat to human health if applied concentrations are less than maximum recommended doses. If a polymer is to be used, make sure that it is certified by the National Sanitation Foundation (NSF) and approved by the state for use in potable water treatment, and that any dosage limits are not exceeded.

This also

applies to sludge conditioning aids if any recycle flows exist. Filtration aids have particular disadvantages if the dose is not closely controlled.

Overdosing may increase the adhesiveness of solids to such an

extent that they are difficult to remove from the filter in backwashing. Additional scour may be required in such cases.

If a longer backwash period

is found necessary, this may offset any advantage gained in attaining longer filter run time.

Long term overdosing may also contribute to the formation

of mudballs and consequent short-circuiting in the filter. E.

EFFECT OF CHEMICAL ADDITIVES ON SLUDGE HANDLING PROCESSES

The treatment and disposal of the sludges and solids generated by water treatment processes is an area of water treatment technology long ignored by regulators, design engineers, and utility managers.

The typical practice in

the past was to discharge these sludges to the nearest surface water down stream from the water intake.

However, such disposal methods have recently

come under more stringent Federal and State regulation.

17

Consequently, the

handling and disposal of water treatment sludges has been receiving considerable attention. Frequently, the removal of water from these sludges is essential for costeffective disposal. This reduces both the weight and volume of sludge to be removed to a disposal site.

In fact, current federal regulations prohibit

the disposal of sludges that contain free moisture in sanitary landfills ("free moisture" being defined as liquid that will drain under the influence of gravity). Chemical additives such as polymers can have a substantial impact on sludge handling in water treatment. Two types of effects are of importance: (1)

the substitution of a polymeric coagulant or flocculant for an inorganic'coagulant can reduce the amount of sludge to be dealt with, and/or improve the sludge's dewaterability characteristics, and (2)

use of a polymer immediately prior to a dewatering process can sig nificantly increase the amount of water removed in the process. In this case the polymer is defined as a sludge conditioner or conditioning aid. The terms sludge flocculant or dewatering aid may also be used. If two coagulants (or flocculants) are being compared for possible use in a water treatment plant, and it is found that the two products provide similar performance in clarification and filtration (and are of similar cost), then the one having the most favorable effect on sludge characteristics would be the optimal choice. A comprehensive procedure to evaluate and compare such products might therefore include the option of assessing the amount and behavior of the resulting sludges. This manual provides several such methods, with the most appropriate method for a given situation depending on the time and equipment available, as well as on the plant's dewatering and disposal practices. The choice of a conditioning aid for use in full-scale dewatering processes can also have a major impact on plant costs. Laboratory procedures which

18

can be used to select the best chemical for such a purpose are given in this manual, although they are probably less familiar in the water treatment field than are many of the procedures recommended for coagulant, flocculant, or filtration aid selection.

Nonetheless, some of the methods are similar--

for example, a "sludge jar test" can be employed to help locate the most promising polymer for dewatering processes.

Furthermore, the initial

polymer measuring, dilution, and addition procedures are the same regardless of the specific polymer application. An in-depth reference which may be consulted for further information on the design and cost of dewatering operations is "Water Treatment Plant Waste Management," prepared by the AWWA Research Foundation and available through the AWWA. In summary, it may be possible to improve the finished water quality or reduce water production costs through the selective use of synthetic organic polymers.

Three specific aspects of water treatment may be upgraded with

polymer addition:

coagulation, filtration, and sludge handling.

A

preliminary polymer selection should be made based on laboratory product evaluation, and appropriate test methods are given herein for this purpose. The next chapter gives a general overview of the way in which these proce dures might best be applied for a given set of plant circumstances. F.

REFERENCES

AWWA Seminar Proceedings-Polyelectrolytes-Aids To Better Water Quality. American Water Works Association Publication # 20121 (1972). AWWA Seminar Proceedings-Upgrading Existing Water Treatment Plants. American Water Works Association Publication # 20126 (1974). AWWA Seminar Proceedings-Upgrading Water Treatment Plants To Improve Water Quality. American Water Works Association Publication # 20153 (1980).

19

AWWA Seminar Proceedings-Use of Organic Polvelectrolvtes in Water Treatment. American Water Works Association Publication # 20173 (1983) .

20

II.

A.

HOW TO USE THIS MANUAL

EXAMINING THE PLANT'S TREATMENT PROCESSES

As discussed in Section I.C, the main objective(s) of a change in chemical additives should be identified before proceeding with any evaluations. Frequently the objective is to remedy an obvious under-performance of a specific process, in which case the possible improvements listed in Sections I.C and I.E would suggest whether a polymeric coagulant, flocculant, filtration aid, or sludge conditioning aid should be evaluated for use. Sales representatives for the polymer products may also be helpful in making this determination.

The appropriate procedure Modules in this manual can

then be consulted, and the chemical additive selected which will best improve finished water quality. If the goal is a more general attempt to decrease plant operating costs, then the first step should be to examine plant processes to decide where a different chemical addition might prove most cost-effective.

Making this

determination properly could save much time and effort in later evaluation work.

The paragraphs below give a basic description of how cost savings

might be estimated for chemical additives.

If these estimates are made,

they can then be compared to give an indication of what type(s) of chemical additives could provide the most savings, and the proper procedures Modules can be consulted for actual evaluations of such products. Coagulants and Flocculants:

The first step in cost estimation is

gathering of the required data.

Here, costs of present coagulant and

1.

accompanying additives which might be affected (lime, chlorine for reduction of ferrous iron, etc.) are needed, along with average dosages employed. Straightforward multiplication of each chemical's unit cost times its dosage will give chemical cost per volume of water treated; sum these up to get total chemical cost.

This should be in units such as dollars per million

gallons of water produced.

Tables and conversion factors to use in these

21

"baseline" calculations are given in Section IV.B. A check on this figure can be obtained if it is multiplied by the average flow rate to give cost per day, and then compared with purchase records for the chemicals. A similar calculation of total chemical cost can be done for projected conditions utilizing a different coagulant and/or flocculant, again using the tables in Section IV.B.

Additional data are required for each polymer

under consideration:

probable dose level, cost, and the effect on other dosages (such as a possible decrease in the coagulant dose due to the use of a flocculant). These figures can be estimated by prospective suppliers of the respective products, although a better source of such information would

be other treatment plants using the same product and treating a similar raw water. The above approach does not consider savings which may accrue through improved filter performance (longer filter runs and thus less backwash), or decreased cost of solids handling (less sludge or a more dewaterable sludge). However, it will give rough cost figures using an alternative coagulant or flocculant. 2.

Filtration Aids:

The use of a polymer prior to filtration will general ly decrease effluent turbidity and increase head loss through the filter. This trade-off has a number of implications when considering the use of these products. Most importantly, filtration aids will be of most use when filtered tur bidity is to be improved or turbidity breakthrough limits filter run length. If head loss development determines when filters must be backwashed, then it is unlikely that introduction of a filtration aid will increase run length unless a currently used filtration aid is to be replaced by a more satis factory product. For conventional treatment configurations, decreased head loss may best be accomplished by improvements in coagulation, flocculation, and/or sedimentation processes which will reduce the amount of turbidity to be handled by the filters.

22

However, if filtered turbidity is seen to increase long before a limiting head loss is observed, then a filtration aid may be successful in balancing these two trends.

A preliminary estimate of expected improvements might be

made as follows. If it is feasible to observe the performance of a filter in the plant beyond the run length normally dictated by turbidity breakthrough, then the run length limit due to head loss may be found.

This is the maximum possible

run length, i.e.. the run length which would be achieved using a polymer that significantly decreased the effluent turbidity but had no effect on head loss. If such an extended filter run would lead to unacceptable deterioration of overall finished water quality, then a rough prediction of this run length may be done graphically.

Plot head loss vs. time for the period available

and extend the line forward until the maximum acceptable head loss is reached.

Read off the time at which this occurs.

This estimate is valid

for constant rate filters; for declining rate systems, plot flow rate vs. time instead, extrapolating to the minimum acceptable average flow rate. Actually, it is unlikely that this maximum run length will be attained because a filtration aid usually causes an increased rate of head-loss development.

The amount of this increase will depend on the polymer used

and many other factors as well.

As an estimate, a run length halfway

between the head-loss and turbidity-based run lengths might be attained with a filtration aid.

Literature reports have shown a doubling of run length in

some cases when a filtration aid was utilized. The estimated increase in run length can then be converted into a projected cost reduction by the computational method given in Section IV.B.

If this

is estimated in dollars per million gallons, the potential savings can be compared to the potential savings using a different coagulant or flocculant. The greater possible savings suggests which type of polymer application might save the most money, and thus the procedures to use from this Manual.

23

3.

Sludge Conditioners:

The proper use of a sludge conditioner will

increase the solids concentration of the dewatered sludge.

This can have

obvious implications for sludge disposal costs, which typically are in rough proportion to the sludge volume disposed.

Increasing the solids

concentration in a dewatered sludge by a given factor will decrease the sludge volume and the sludge disposal costs (dollars per day) by the same factor.

For example, if a better conditioner will enable a dewatering

process to achieve 24% solids instead of the present 20%, the factor is 24/20 or 1.2, so the present sludge disposal costs divided by this factor gives the projected sludge disposal costs.

Subtraction gives the savings,

and dividing by average plant water production rate (MGD) gives approximate savings in dollars per million gallons.

Again, this figure can be compared

to those calculated for possible savings with flocculant and filtration aid use, to indicate which type of polymer would save you the most money.

The

projected solids concentration is the difficult quantity to get in this calculation, and should be approximated based on opinions of technical representatives and treatment plant personnel who use conditioners in comparable processes and facilities. Obviously, these are all very rough calculations and should be interpreted with caution.

However, the results will suggest which polymers are likely

prospects for laboratory evaluation using the procedures in this Manual. For example, if the calculations indicate that a certain polymer will substantially increase treatment costs, then it may be dropped from laboratory evaluation.

B.

LOCATING POTENTIAL SUPPLIERS AND PRODUCTS

It is not generally difficult to locate sales representatives for the larger polymer distributors, who maintain occasional contact with most water treatment facilities.

However, if a wider range of suppliers is preferred,

additional effort will be necessary in locating further product sources. The AWWA can provide a list of polymers currently approved for use in water treatment, including the names and addresses of distributors.

24

The maximum

recommended dose for each polymer is also given.

The NSF maintains a

certification program for polymer additives used in potable water treatment (P.O. Box 1468, Ann Arbor, Michigan 48106). Additional "word of mouth" information on polymers and suppliers might be obtained from other water utilities in the area.

National and Section

Conferences of the AWWA may provide good opportunities for this. Hundreds of polymeric formulations are available from suppliers, for a wide variety of applications.

Among these, however, only some have been approved

by the NSF for use in potable water production.

Therefore, when requesting

polymer samples from suppliers, be sure to indicate that this specification must be met.

This may ease selection, and avoid the testing of products

should not be used.

(If sludge conditioning aids are being screened for

use, this only applies if the products might come into contact with potable water due to recycle streams from sludge handling processes.) Once a particular product that has been approved for use in potable water is located, check to see if a maximum dosage level has been specified.

If such

a limit exists, do not use the product in excess of this amount for testing since it cannot be applied full-scale at such high concentrations.

(The

product's percent activity may be needed to determine this; see Module A for details.) Another way to ease the difficulties experienced in polymer selection is to assess only those polymers which are suited for the desired application. For instance, primarily low molecular weight cationic polymers should be Refer to Table 3 as an aid for making

tested for use as primary coagulants. such determinations.

(If no charge or molecular weight information is

available, check the "physical description" category in Table 1.

For the

above example, solid products would not be tested, but pale yellow liquids would be.)

It may also be necessary to refer to Table 3 to determine which

of the preparation procedures in Module A is appropriate for the products being evaluated.

25

If a utility is restricted to operating within specific pH and temperature ranges, choose a polymer type which is able to function within these limits. As an example, PEA is intended to be a cationic primary coagulant (as indicated in Table 1), but it retains its positive charge only at low pH's. If this is not a feasible operating pH, a cationic polymer with a pH independent positive charge should be selected, such as PDADMAC. C.

ASSISTANCE FROM POLYMER SUPPLIERS

A primary role of sales representatives is advising potential customers with the obvious goal of increasing product sales. Assistance may therefore be available from such personnel towards determining which polymers in their product line would be most successful in a particular application. In some cases this service may include comparison testing of their own products, either on site or at the company's laboratories. Such assistance might therefore be of considerable use in the initial screening of chemical additives under consideration. On the other hand, evaluations done by treatment plant staff may be more useful because: the use of consistent procedures will allow quantified comparisons between polymers from different suppliers' product lines, including cost comparisons. tests can be performed during various raw water conditions (turbidity, temperature, color, etc.) to assess the range of situations in which a given polymer will be helpful. plant personnel know about any unusual aspects of their treatment processes, and can incorporate this into the testing procedures (for example, if short residence times are a problem at high flow rates, the time used for each stage of the jar test can be reduced correspondingly).

26

Consequently, the most productive approach may be to utilize recommendations of the suppliers only to develop an initial list of potential products, and then to proceed with evaluations performed by plant staff.

Small samples of

the additives to be tested will usually be provided for evaluation purposes at no charge. product.

Suppliers will also furnish technical literature for each

Check the specifications for (a) EPA or NSF approval for use in

potable water treatment, (b) maximum recommended dosage level, (c) instruc tions concerning product storage and preparation, and (d) safety considera tions in handling and storing the substance, which should be given on the Manufacturer's Safety Data Sheet ("MSDS").

All of these specifications

should be heeded, although preparation procedures are also given in this Manual in case none are provided.

Of course, the approximate unit cost of

each polymer should also be obtained.

Finally, polymer distributors can be

of assistance once a preliminary polymer selection has been made.

They can

usually offer advice on proper storage, mixing, and feed facilities and may be able to loan equipment to the plant for a final full-scale polymer evaluation, such as pumps and dose control systems. D.

PREPARING TO EVALUATE THE PRODUCTS

Before actual testing begins, a number of preliminary actions are necessary: 1)

As suggested in the previous section, discuss with sales representatives

which of their products would be most successful in the application at hand. List these additives as candidate products. 2)

Using cost and technical data provided, go through preliminary cost

estimates for these products as described in Section II.A and Chapter IV. Rank the product list according to each additive's estimated cost. 3)

Obtain samples of all these polymers.

Testing should not begin until

all products are on hand, since comparisons of test results are only possible when raw water conditions were fairly similar.

27

4)

According to what is known of problem conditions in the plant, determine what "worst case" conditions should also be used in testing. For instance, some testing may be advisable under high raw water turbidity conditions if this is when plant operation is usually least satisfactory.

If seasonal

difficulties are encountered, it may be necessary to delay some evaluations until this time period. 5)

Determine which evaluation procedures should be used (see Section II.E). If any equipment must be purchased or fabricated, allow sufficient time for this to be done.

Become familiar with the appropriate Module in Chapter III for each procedure, and perform the quality control steps recommended in Section II.F. 6)

Some test methods for specific objectives may be desired beyond what is presented in this Manual. For example, if removal of THM precursors is of concern, an appropriate method for measuring these would be needed, such as TOG (Total Organic Carbon) or UV absorbance. Acceptable methodologies for these additional tests must then be selected. 7)

Finally, assemble the required data sheets for all required tests.

The

Modules in this Manual include suggested formats, and these tables can simply be photocopied as required. If other procedures or information are desired, more appropriate data sheets should be designed. E.

THE MODULE FORMAT:

DETERMINING MODULES TO USE

As indicated in Chapter I, the procedures in this Manual have been divided into separate sub-procedures or "Modules." Each of these is presented in Chapter III as a separate Section:

for example, Section III.C, "The Jar

Test," is also referred to as "Module C." The division into Modules has been done to allow maximum flexibility of use. There are many ways of combining these Modules into a sequence which will be of most use in a given situation. Each user of this Manual can select a proper set of Modules to use based upon the specific:

28

type of additive being considered process(es) to be improved water quality objectives equipment available expertise and available time of the personnel to be performing the tests. The type of additive being considered, if not already determined, can be decided upon by referring to Sections I.C, I.E, and II.A. also relates to the process(es) to be improved:

This decision

for example, better

filtration might best be achieved with a filtration aid.

This might seem

obvious, but as suggested in Figure 2, a flocculant might also affect filtration--and sludge characteristics as well. These types of interactions may be quite important, and the Module arrange ment enables them to be investigated.

For instance, if a flocculant is to

be evaluated, assessment of the possible impacts on downstream processes (filtration and sludge handling) may also be of interest.

Figure 2 indi

cates which Modules are designed to simulate the various water treatment processes and the effect of polymers on them; for the present example, this Figure shows that Module A could be used to prepare and dose the flocculant in a manner similar to that in full-scale application, Modules B, C, D, and/or E would predict the effect on flocculation and sedimentation.

If the

impact on filtration is of concern, Modules F or G could be used to evaluate this.

Finally, any of Modules H through L might be appropriate in analyzing

changes to be expected in sludge thickening, dewatering, or disposal.

Thus,

depending on the situation, only one Module might be used, or several might be combined. Most of the Modules shown in Figure 2 describe procedures designed to simulate the corresponding water treatment processes.

However, several also

present water quality analysis methods which can be performed on the water following these simulation procedures.

For example, Module D gives

instructions on how to properly measure turbidity following a jar test, and

29

o

CO

t SEDIMENTATION

T

DEWATERING

(Modules H, I, J, K, and L)

Conditioning Aid (Module A)

(Modules B, C, D, and E)

FLOCCULATION

(Modules F and G)

FILTRATION

I

Filtration Aid I (Module A)

FIGURE 2 . Typical Water Treatment Process Diagram, and the Modules Which Simulate Chemical Effects on These Processes

RAPID MIX

Flocculant I (Module A)

Coagulant

would be used along with Module C which describes the jar test method it self.

Obviously, other references such as turbidimeter manuals and Standard

Methods are already available for this purpose, and the main reason for such a Module is to relate such methods to the most suitable sampling techniques following a jar test (in this case). It would not be possible to include a Module for every water quality anal ysis which might be of interest to a user of this Manual.

Instead, the user

should decide on these depending on the specific water quality goals and then locate appropriate methods.

The same sampling techniques suggested in

this Manual's Modules (such as for turbidity) should be incorporated into these methods.

For instance, if improved color removal is hoped for through

the use of a filtration aid, then a method for true color from Standard Methods or the AWWA Introduction to Water Quality Analyses (1982) might be appropriate. Other important factors in selecting Modules to use are the equipment, time, and expertise requirements.

A description of necessary apparatus is given

in each Module along with an indication of the test's relative level of com plexity, and an initial reading of the Modules under consideration will thus help in making the best selection from among them for the situation at hand. The user should not be immediately dissuaded by the specific equipment or a significant testing effort required in some Modules.

The possible cost sav

ings estimated in Section II.A may indicate that such an investment will be amply returned.

Moreover, once testing procedures are developed, they will

be available in the future for maintaining optimal plant operation.

Speci

fications which may seem arbitrary have usually been made with good reason; as an example, testing has indicated that square "plexiglass" jars are superior to the more traditional glass beakers in several respects when evaluating flocculants. Figures 3, 4, and 5 illustrate possible Module sequences when evaluating polymers for various applications.

Note that Module A is used in all cases

for the proper preparation of polymer working solutions.

31

Figure 3 also

indicates that an optional Module B (a "rapid jar test") may be used if initial screening of a large number of polymeric flocculants is desired. The more elaborate jar test (Module C) can then be used on the most promising polymers.

As previously discussed, many other Modules can then be

used to more fully characterize coagulant aid performance as it may affect filtration or sludge management. The Modules shown in Figure 4 for filtration aid evaluation include a paper filter test (Module F) and analytical methods to accompany it (Module G). Bench-scale and pilot-scale evaluations may also be employed, although this Manual does not include specific Modules for this purpose.

The paper

filtration test is applicable to a variety of process configurations, including the evaluation of filtration aids in conventional treatment, contact (or in-line) filtration, or direct filtration with prior flocculation.

The test can also be used following a jar test if a

coagulant's or flocculant's effect on filtration is to be assessed. Figure 5 depicts a variety of sludge characterization methods.

A change in

coagulant or flocculant use can alter sludge properties, and diagram A in this figure shows how such effects can be evaluated.

Module H enables

changes in sludge volume resulting from a coagulant aid to be measured if this is of concern; it is also a required procedure in obtaining sludges for Modules J or K if one of these is to be run.

Module J provides information

on the dewaterability of a sludge through a "Time To Filter" (or Buchner funnel) test.

Module K is an alternate Capillary Suction Time test for

dewaterability which is faster than Module J but requires more sophisticated equipment. Diagram B suggests possible Module sequences when polymers are to be anal yzed for direct use as sludge conditioning aids, in which case Module I is needed for proper sludge sample preparation.

Module L describes a sludge

jar test, which differs from the conventional jar test in that it is used as a quick, qualitative method for determining sludge thickenability or dewaterability.

It can be used independently or as a screening procedure

32

co

CO

III. F, G Determine coagulant aid's impact on filtration

impact on sludge management

IV

Selection of coagulant aid

III.H, I,J, K,L

Optional analytical methods

III. E Particle size analysis

Determine coagulant aid's

FIGURE 3 Procedure for the Selection of Coagulants or Flocculants

Preparation of working solutions

III. B Preliminary screening (rapid jar test)

III.D Turbidity

CO

III. A Preparations of working solutions

FIGURE 4 . Procedure for the Selection of Filtration Aids

Other tests

III. G Particle size analyses

III. G Turbidity measurements

III. F Paper filter test

IV. Evaluation of results

Pilot filter evaluation

III. A Preparation of Working Solutions

lit. K Capillary Suction Time Test

III. J Time to Filter Test

III. L Sludge Jar Test

III. K Capillary Suction Time Test

III. J Time to Filter Test (small volume)

FIGURE 5. Procedures for Evaluating the Effect of Coagulants, Flocculants, or Sludge Conditioning Aids on Sludge Properties

Sludge Sample Preparation

B. Evaluation of Sludge Conditioning Aids:

Detailed Jar Test

III. H Sludge Volume Determination

A. Impact of coagulant aids on sludge properties:

IV. Evaluate Results

IV. Evaluate Results

prior to the use of Module J or K. All evaluation sequences then utilize Chapter IV, "Using Module Results for Chemical Aid Selection."

It gives means of comparing results and developing

cost estimates which will indicate the chemical aid of most benefit to the treatment plant. Once a choice of Modules has been made, a flow chart similar to these can be drawn which will only show the methods to be used.

(A copy of Figure 1 may

be used for this purpose by just crossing out the Modules which will not be used.)

This may help in organizing the equipment and procedures which will

be required.

Also see the following section for important considerations

regarding quality control and analysis of results. Following the evaluation of all prospective products, it may appear that one product is the most promising.

Full-scale evaluations in the plant are then

recommended in order to validate the product's success prior to the purchase and installation of permanent storage, mixing, and feeding devices.

If more

than one product exhibited good performance at the same cost in small-scale tests, they may also be compared on a full-scale basis before a final choice is reached.

Due to the wide variation in process and plant characteristics,

a method of implementing full-scale evaluations is beyond the scope of this manual.

Consult "System Design for Polymer Use" in Use of Organic Poly-

electrolytes in Water Treatment (AWWA 20173) and consult with technical representatives of the polymer suppliers. F.

QUALITY CONTROL AND ANALYSIS OF RESULTS

It is of utmost importance that the procedures in this Manual be performed in a correct and reproducible manner.

Otherwise, variability in results may

prevent any meaningful conclusions from being drawn.

The following steps

should be taken prior to actual evaluation of polymers in order to assure that rigorous and methodical experimental techniques are employed:

36

1)

Personnel who are to run the tests should practice each procedure until

sufficiently familiar with it. 2)

Data sheets should be completed in full, dated, and stored in a loose-

leaf notebook.

Additional tables, graphs, and related records should be

kept in this notebook as well. 3)

All analytical methods used should be calibrated with accepted standards

for the test. accuracy. 4)

This should be done periodically to insure consistent

Standard Methods is the preferred reference in this regard.

Precision of analytical methods should be assessed by periodically

performing replicate analyses.

This includes replicate sampling.

For

example, multiple jar test turbidity measurements would be done by taking different supernatant samples from the same jar and comparing measured turbidities. 5)

Tests on replicate polymer doses should be done.

Continuing with the

jar test example, this would mean also running several jars with the same polymer dose to check for consistent final turbidities.

Variability

indicates inconsistent procedure. 6)

The ability to reproducibly prepare polymer solutions should be assessed

in a similar manner.

Again, this would be done in jar testing by comparing

jar test turbidities after using equal polymer doses taken from different preparations of polymer stock and working solutions. 7)

If more than one person is to be performing tests to be compared, each

person should run an identical test.

If a comparison of results indicates

significant deviation, each person's technique should be examined and compared. 8)

Calculations done in analyzing results should be performed with care.

Units associated with each value should always be indicated.

37

It is ad-

visable to initially have two persons use the required computation method and compare answers to catch any methodical errors. G.

REFERENCES

AWWA Seminar Proceedings-Use of Organic Polyelectrolytes in Water Treatment. American Water Works Association Publication # 20173 (1983). Introduction to Water Quality Analyses (Vol. 4). American Water Works Association Publication # 1931 (1982). Standard Methods for the Examination of Water and Wastewater, APHA, AWWA (Publication* 10035), and WPCF. Washington, B.C. (1985).

38

III.

A.

MODULE A.

MODULES FOR SPECIFIC SELECTION PROCEDURES

INSTRUCTIONS FOR THE LABORATORY PREPARATION OF POLYMER SOLUTIONS

1.

General Discussion

After a number of potentially useful polymeric product samples have been acquired, it is desirable to dilute the pure (or neat) polymers before each comparative test.

In order to simulate full-scale polymer use, dilution is

achieved in stages. During the first stage of dilution, a small amount of pure polymer is mixed into a large volume of water.

This step is referred to as either primary

dilution or primary make-up, and the mixture which results is known as a stock solution. Similarly, the second dilution stage is termed secondary dilution or secon dary make-up.

To perform this dilution, a small volume of stock solution is

mixed into another large volume of water.

The mixture or solution which

results from secondary dilution is called a working solution.

The primary

and secondary polymer dilution stages are both meant to simulate off-line processes. The third or final dilution stage simulates the in-line feeding of a polymer solution into an influent raw water stream.

To achieve this third dilution

in the laboratory, a small portion of the working solution is added (or dosed) to the appropriate volume of raw water for analysis in a jar test. The final polymer concentration in the jar is the same as that concentration which would be used for full scale polymer application. In summary, the order of decreasing concentration of these solutions is

39

pure product

primary ^ ilution

working solution

'

, , . . stock solution

-r-r^

:

>

secondary dilutior/ '

plant scale/jar test concentration.

It is important to note that any measurement errors made while performing a consecutive (or serial) dilution will be greatly magnified in the final solution.

Also, high degrees of error are associated with the measurement

of small quantities.

(When measuring out a 1 L sample, a measurement error

of one drop will not be very significant, whereas when measuring out a 0.0001 L sample, a measurement error of one drop is a very large portion of the total amount sought, and the resulting error will be huge.) Therefore, extreme care must be taken to accurately obtain and transfer the small quantities of pure polymer and polymer solution which are required. Pure polymer products and their solutions are sensitive to degradation (or activity loss) with the passage of time. For this reason many polymer suppliers will indicate expiration dates for their pure polymer products. If no expiration date is supplied, assume a six month period of effective ness. If possible, record the arrival and expiration dates directly onto the polymer sample container. discarded.

Polymer samples which have expired should be

It is recommended that stock solutions be kept no longer than 2

days (48 hrs.), and working solutions be kept no longer than 2 hours.

These

time periods include preparation time; that is, if a stock solution is aged overnight, it may only be used for 24 more hours.

The more dilute a polymer

solution is, the quicker it will be degraded. Polymer preparation and concentrations to be tested may be dictated by the maximum dosage level set by the NSF. If such a limit exists, do not use the product in excess of this amount for testing since it should not be applied full-scale at higher concentrations. At times, polymer suppliers use the term "percent activity" when describing their product in polymer specification sheets.

For example, a liquid

product which is described as 40% active can be said to contain 40 mL of

40

effective product for every 100 mL of neat polymer sample.

(The same

information is conveyed if the phrase "60% volatiles" is used; it indicates that 60 mL out of every 100 mL of neat polymer sample is water. implies that 40% is the active fraction.)

This

Understanding and using this

concept is most important if the polymer specification sheet expresses the EPA maximum allowable concentration as 2% active.

To determine how much

actual polymer sample (or polymer solution) is required to produce a final 2% active solution, first locate the general value for percent activity, (or 100 minus the percent volatiles) on the specification sheet.

For illus

trative purposes, assume this value is 40% active, or 40 mL active per 100 mL pure.

To determine the equivalent percentage of neat polymer which would

contain 2% active, simply invert the 40% activity fraction and multiply by 2%. 100 mL neat 40 mL active

_ c& x 2% active = 5% neat

Now, by using the 5% neat polymer value as a maximum allowable concentra tion, compliance with EPA standards is a more straightforward process.

(See

Example 1 at the end of this Module for details on how to dilute a neat polymer product to achieve a specific final concentration.) For the following procedures, all glassware used in polymer preparation should first be thoroughly cleaned, especially if polymer characterization procedures (Modules M or N) are to be performed.

Glassware should first be

washed with hot soapy water, then rinsed with copious amounts of distilled, deionized, filtered water.

A subsequent cleaning with acetone, followed by

further rinsing, may also be necessary. The polymer solution preparation instructions which follow are intended for use in cases where specific dilution instructions (for laboratory-scale use) are not provided by polymer suppliers.

If such information is provided it

should be used, since it may incorporate considerations specific to the product.

If there is any question of whether the procedure being used is

achieving sufficient product activation, Modules M or N may be used to

41

assess properties of the preparation.

The more effective the activation is,

the greater and more consistent the charge density and viscosity characteristics of the polymer solution will be. Tables 4, 5, 6, and 7 provided with this Module provide additional flexibility in preparing polymer concentrations other than those specifically given in the procedures. Table 4 applies to makeup of stock solutions; Table 5 applies to working solution preparation; and Tables 6 and 7 are used to obtain desired final chemical concentrations in either a 2-L or 1-L container, respectively. 2.

Apparatus

1)

Several 1 and 2 L glass beakers.

2)

Tall, narrow, cylindrical plastic vessel (about 5.5 in. tall, with a 3.5 in. diameter). 3)

Calibrated disposable syringes (sizes 50, 20, 5, & 1 mL). A digital piston-type pipette can also be used. Conventional glass pipettes are not recommended for accurate dispensing of viscous polymer solutions; if one must be used, the tip should be broken off after scoring with a file. The end should be flame-polished to remove any sharp edges. 4)

1 L volumetric flask, to measure out dilution water for preparation of stock solutions.

5)

100' mL graduated cylinder to measure out dilution water for preparation of working solutions. 6)

Analytical balance or scale with an accuracy of at least 0.01 grams, for weighing out dry polymers. 7) Glassine weighing paper or weighing vial for use in weighing out and transporting dry polymers.

42

TABLE 4 Stock Solution Preparation: Final Volume:

1 L Stock Solution

Desired Stock Solution Concentration, Volume %

0.5

0.1%

Neat Polymer Dosage, grams

50

10

g

10

15

20

100

150

200

TABLE 5 Working Solution Preparation: Final Volume:

200 mL Working Solution

Desired Working Solution Concentration, Volume % and g/L Concentration of Stock Solution 0.001% Being Diluted, O.Olg/L Volume % 0.1% 0.5 1.0 5.0 10 15 20

2 mL 0 .4 0 .2 0 .04 0 .02 0 .013 0 .01

0.005 0.05

10 2 1 0 .2 0 .1 0 .067 0 .05

0.05 0.5

0.01 0.1

20 4 2 0 .4 0 .2 0 .133 0 .1

100 20 10 2 1 0.67 0.5

0.1

40 20 4 2 1.33 1.0

0.5 5

10

100 20 10 6.67 5.0

40 20 13.33 10.0

5 50

10 100

100 66.67 133.33 50.0 100.0

Milliliters of stock solution to be diluted for the preparation of a 200 ml working solution. 100% (neat polymer)

0.002mL

0.01

0.02

0.1

0.2

1.0

2.0

10.0

20.0

Milliliters of neat liquid coagulant polymer to be diluted for the preparation of a 200 mL working solution.

43

TABLE 6 Jar Test Solution Preparation: Final Volume:

2 L Solution for Jar Testing

Concentration of Working Solution with which Jars are Dosed, Volume %, g/L 0.001%, 0.005%, 0.01% , 0.05% , 0.1% , 0.5% , , 1% , 5% , 10%

0.01 g/L 0.05 g/L 0.1 g/L 0.5 g/L 1 g/L 5 g/L 10 g/L 50 g/L 100 g/L

Desired Concentration in Jar, ppm (or mg/L), and Volume % 0.1 ppm 0.00001%

0.5 0.00005

15 0.0001 0.0005

200 40 20 4 2 0.4 0.2 0.04 0.02

100 20 10 2 1 0.2 0.1 0.02 0.01

20 mL 4 2 0.4 0.2 0.04 0.02 (0.004) (0.002)

1000 200 100 20 10 2 1 0.2 0.1

10 0.001

400 200 40 20 4 2 0.4 0.2

50 0.005

1000 200 100 20 10 2 1

Milliliters of Working Solution used to dose raw water for jar testing.

TABLE 7 Jar Test Solution Preparation: Final Volume:

1 L Solution for Jar Testing

Concentration of Working Solution with which Jars are Dosed, Volume % , g/L 0.001%, 0.01 g/L 0.005%, 0.05 g/L 0.01% , 0.1 g/L 0.05% , 0.5 g/L 0.1% , 1 g/L 0.5% ,5 g/L , 10 g/L 1% , 50 g/L 5% , 100 g/L 10%

Desired Concentration in Jar, ppm (or mg/L) , and Volume % 0 . 1 ppm 0.00001% 10 mL 2 1 0.2 0.1 0.02 0.01 (0.002) (0.001)

0.5 0.00005 50 10 5 1 0.5 0.1 0.05 0.01 (0.005)

1 0.0001 100 20 10 2 1 0.2 0.1 0.02 0.01

5 0.0005

10 0.001

500 100 50 10 5 1 0.5 0.1 0.05

200 100 20 10 2 1 0.2 0.1

50 0.005

500 100 50 10 5 1 0.5

Milliliters of Working Solution used to dose raw water for jar testing.

44

8)

Small size hand-held blender; for example, the "MR-7" type manufactured by "BrAun". NOTE: Magnetic stirrers DO NOT activate polymer products adequately or reliably.

These are NOT RECOMMENDED for stock or working

solution preparation. Bench or gang stirrers may be substituted for the blender if necessary. Be sure to follow the specific procedures for bench stirrer used in the follow ing sections.

Bench stirring will be adequate if lower polymer concentra tions and smaller volumes of solution are prepared, so select the smallest feasible quantity listed in Table 3 (see Section II.B). The stirrer paddle may be lowered into a smaller liquid volume by placing a flat board under the beaker to raise it. Bench stirrers are particularly ineffective for preparing dry solid polymers. 9)

Stopwatch for timing polymer mixing.

10)

pH meter, calibrated as near as possible to final desired polymer solution pH. 3.

Reagents

1)

Polymer samples are typically available on request from most commercial suppliers of water treatment chemicals, at no charge. When communicating with such suppliers, be sure to request that polymer specification sheets and any safety information available be sent to the utility with the polymer samples, as these can be valuable sources of information. See Chapter II, Sections B, C, and D for more information on obtaining samples. 2)

Dilution water used for polymer solution preparation should be the same water as that which would be used to prepare polymer solutions for full scale implementation. Be sure to avoid freshly chlorinated water for both laboratory and full-scale use, as high chlorine doses can degrade some polymers, and thus reduce their effectiveness. Also, the temperature of the dilution water used should be close to that of the water used full-scale.

45

These general considerations will suffice for most polymers, but when employing PAM or hydrolyzed PAM type polymers, more specific dilution water characteristics are necessary. For PAM polymers, the dilution water utilized should have a hardness < 100 ppm (or mg/L). For hydrolyzed PAM polymers, dilution water alkalinity should be > 50 ppm (or mg/L), and the conductivity of the dilution water should be > 150 /zmho/cm. 3)

1 M NaOH solution for raising the pH of polymer solutions before use in jar testing, only as needed. See Table 1 in Section B of Chapter II. 4)

1 M HC1 solution for lowering the pH of polymer solutions before use in jar testing, only as needed. See Table 1 in Section B of Chapter II. Solutions of NaOH and HC1 may be prepared as indicated in Standard Methods for the Examination of Water and Wastewater: extreme degrees of preparation accuracy are not necessary. Sections 4 to 7, which follow, are for preparation of polymer solutions according to the original product form. Be sure to use the appropriate section only. 4.

Preparation of Liquid Solution Coagulants

a.

Overview

These types of polymers are generally of low molecular weight, with a high positive charge. Their functioning is similar to that of traditional metal salt coagulants such as alum or ferric chloride, as they too neutralize the negative charge associated with colloidal particles present in raw waters. Since these polymers are low in molecular weight, they are the least difficult to solubilize and activate. Therefore, stock solution preparation is not required, and working solutions may be made directly from the original neat polymer product. (NOTE: All other types of polymeric formulations require stock solution preparation.)

46

b.

Working Solution Preparation

Notes: Always add polymer to water, not vice versa. Prepare this solution just prior to experimentation as it only lasts 2 hours. 1)

Shake the polymer sample until homogeneous.

2)

Calculate the volume of 2 g of polymer.

Polymeric density is usually

given on the data sheets provided by polymer suppliers, but it is recommended that this value be checked since the exact product specifica tions may be slightly different.

(To determine polymeric density in g/mL,

simply weigh 1 mL of polymer in a tared syringe or weighing vial.

Make this

measurement several times; use the average of the weights obtained as the density figure.)

The equation used to calculate the volume of 2 g of

polymer is: , 7 = volume , , T N = ——————Q————————— 2 grams V (mL) polymer density g/mL If density is supplied as "specific gravity," the value may be directly compared to the average density figure obtained as described above. density is given in Ibs/gal, divide by 8.345 to convert to g/mL.

If

If density

is given in Ibs/ft , divide the value by 62.43 to convert to g/mL. 3)

Subtract this calculated volume from 200 to determine the volume of

dilution water required.

Measure out this volume of water with a graduated

cylinder and place it in the tall, narrow plastic vessel. 4)

Draw out this calculated polymer volume into a syringe.

Avoid the

measurement error associated with drawing up bubbles, as well as polymer, into the syringe. 5)

Insert the BrAun hand-held mixer (or bench stirrer paddle) into the

plastic vessel which contains the dilution water.

47

Turn on the mixer to full

speed (at least 100 rpm for bench stirrer), activate the stopwatch, and SLOWLY evacuate the syringe; BE CERTAIN that the polymer is entering the dilution water at a point of high turbulence. If the polymer "stream" leaving the syringe touches either the vessel walls or the blender "neck" (or the stirrer paddle), the solution must be discarded and the process repeated. Continue mixing for 30-60 seconds, or until the solution is homogeneous. (For bench stirrer use, mix for 1 hour at maximum rpm set ting.) 6)

Working solutions do not require aging; in fact, they should be used as quickly as possible after they are prepared. 7)

The resulting 200 mLs of solution is a 10 g/L (or 1%) working solution. In order to prepare other working solution concentrations, refer to Table 5. 8)

Dosage:

0.1 mL of this working solution added to 1 L of raw water gives a polymer dosage of 1 mg/L (or 1 ppm) for jar testing. Note: for the preparation of other working solution concentrations, consult the last line of Table 4 (100% stock solution - neat polymer product). For an example of use of Table 4, see Example 2 at the end of this Module. Example 3 describes how polymer concentrations can be prepared which are not listed in the Tables. For preparation of other final jar test concentrations, consult Table 6 or 7. 5.

Preparation of Liquid Solution Flocculants

a.

Stock Solution Preparation

1)

Shake the polymer sample until homogeneous.

2)

Calculate the volume of 10 (ten) grams of polymer. Polymeric density is given by most major polymer suppliers in their data sheets. (Or preferably, the density can be measured as described in step (2) of Preparation of Liquid Solution Coagulants.) The equation used to calculate the volume of 10 grams of polymer is:

48

TT = Volume TT T / T \ = ———————°——— 10 grams V (mL) polymer density (g/mL) If density is given as "specific gravity", it may be directly compared with the value obtained as described in step (2) of Preparation of Liquid Solu tion Coagulants.

If it is given in Ibs/gallon, divide the value by 8.345 to

convert to g/mL.

If it is given in Ibs/ft , divide by 62.43 to convert to

g/mL. 3)

Subtract this calculated volume from 1000 to obtain the volume of water

required for dilution.

Add this amount of water to a 2 liter beaker.

(To

do this accurately, use a 1 liter volumetric flask to measure out 1 liter of water, then use a syringe or glass pipette to draw out this calculated volume of water.) 4)

Measure out this calculated volume of polymer into a syringe.

With the

hand-held BrAun mixer (or the gang stirrer paddle) immersed in the dilution water and set at its maximum setting, SLOWLY evacuate the syringe.

Take

extra care to make sure that the polymer leaving the syringe is entering the dilution water in a region of high turbulence.

Should the polymer not enter

this region, and instead make contact with the beaker walls or blender surface (or bench stirrer paddle), discard the solution and begin again. Continue mixing for 30-60 seconds, or until the solution appears homo geneous.

Let the solution age for 15 minutes, then blend for another 30-60

seconds before further dilution.

(For bench stirrer use; mix 1 hour, age

overnight, and mix again for 1 hour).

The total volume of the resultant

solution is 1000 mL; it is a 10 g/L (1%) stock solution.

Consult Table 4

when stock solutions of different strength are desired. b.

Working Solution Preparation

This is prepared just prior to experimentation by diluting a portion of the stock solution described above.

49

1)

Using a syringe, SLOWLY add 20 mL of the stock solution to 180 mL of dilution water which is being agitated by the BrAun hand-held mixer (or a bench stirrer paddle) at maximum speed, and is contained in the tall, narrow plastic vessel. Be sure that the polymer solution is entering the dilution water in a highly turbulent region, and that it is not making contact the vessel or the blender (or the paddle) surfaces. If such contact is made, the solution will have to be discarded and the procedure repeated. Continue mixing for 30-60 seconds (for bench stirrers, mix for 1 hour). Do not age this solution; use it as soon as possible after it is prepared. The resultant 200 mL of polymer solution is a 1 g/L (0.1%) working solution. Consult Table 5 for guidance concerning the preparation of other working solution concentrations. 2)

Dosage:

5 mL of this working solution added to 1 L of raw water yields a polymer dosage of 5 ppm (or mg/L) for jar testing. To prepare other final dosages, consult Table 6 or 7. 6.

Preparation of Liquid Emulsion Polymers

Notes: Use syringes when dispensing viscous polymer solutions. Always add polymer to water, not vice versa. a.

Stock Solution Preparation

1)

Shake polymer sample thoroughly until homogeneous.

2)

Calculate the volume of 10 (ten) grams of the polymer. Polymeric densities are provided by most suppliers in their data sheets. (It is suggested that this value be checked since product variation is common.

The

polymeric density may be determined by simply weighing out 1 mL of polymer in a tared syringe or weighing vial. Make this measurement several times; use the average of the weights obtained as the density figure, in grams/mL.) The equation used to calculate the volume of 10 grams of polymer is:

50

,, = Volume ,7 n / T N = —————c——— 10 grams V (mL) density (g/mL) If the density is given as "specific gravity", the value may be directly compared with the value obtained as described above.

If it is given in

Ibs/gallon, divide the value by 8.345 to convert to grams/mL.

If it is

given in Ibs/ft , divide the value by 62.43 to convert to grams/mL. 3)

Subtract this calculated volume from 1000 to obtain the amount of water

(mL) required.

Add this amount of water to a 2-liter beaker.

(To do this

accurately, use a 1-liter volumetric flask to measure out 1 liter of water, then use a syringe or glass pipette to draw out this calculated volume of water.) 4)

Measure out the calculated polymer volume using a syringe.

Insert the

BrAun hand-held mixer (or bench stirrer paddle) into the 2-liter beaker which contains the dilution water.

Turn on the mixer to full speed (for

bench stirrers, at least 100 rpm), and SLOWLY evacuate the syringe; BE CERTAIN that the polymer enters the dilution water at a point of high turbulence and that no contact is made with vessel or mixer (or stirrer) surfaces.

Continue mixing for 30-60 seconds, or until the solution is

homogeneous.

Allow the stock solution to age for 15 minutes, then blend for

another 30-60 seconds before further dilution.

(For bench stirrers; mix the

solution for 1 hour, let it age overnight, then mix it again for 1 hour.) The resultant total volume, 1000 mL, is a 10 g/L (1%) stock solution.

Refer

to Table 4 for information on preparing stock solutions of other strengths. b.

Working Solution Preparation

This is prepared just before experiments are run by diluting a portion of the above described stock solution. 1)

Using a clean syringe, SLOWLY add 10 mL of the stock solution to 190 mL

of water which is being mixed at the maximum hand-held mixer (or bench stirrer) speed, and is contained in the tall, narrow plastic vessel.

Again,

be sure that the polymer solution enters the dilution water at a region of

51

high turbulence.

Care must be taken to ensure that the polymer does not

touch the blender or vessel (or stirrer) surfaces as. it leaves the syringe if this happens, start over. Continue mixing for 30-60 seconds (for bench stirrer, mix for 1 hour), or until homogeneous. 2)

The resultant 200 mL of solution is a 0.5 g/L (0.05%) working solution. Working solutions do not require aging. In fact, these solutions should be used as quickly as possible after they are prepared.

Consult Table 5 for

guidance when other working solution concentrations are desired. 3)

Dosage:

2 mL of the working solution added to 1 liter of raw water

gives a polymer dosage of 1 mg/L, or 1 ppm.

Additional dosages may be

calculated accordingly, or may be read directly from the dosage chart (Table 6 or 7). 7.

Preparation of Dry Solid Flocculants

Note:

Solids are generally PAM or hydrolyzed PAM type polymers, which have special dilution water requirements; see Section 2 (Reagents). These also

have pH requirements; see Table 1 in Section B of Chapter II. a.

Stock Solution Preparation

1)

Accurately weigh out 5.00 grams of solid polymer using either a tared weighing paper or a tared weighing vial. Make this measurement as quickly as possible since solids can be hygroscopic and may draw water from the air, thus altering the measurement. 2)

Using a 1 L volumetric flask, obtain 1 L of dilution water. dilution water into a 2 L glass beaker. 3)

Place

Insert the BrAun hand-held mixer (or bench stirrer paddle) into the

water and turn it on to full speed.

Simultaneously, activate the stopwatch

and begin SLOWLY sprinkling the polymer grains into highly turbulent regions of the dilution water. BE SURE that the grains separately contact the

52

water; if solid particles are not "wetted" individually they will stick together in clumps.

If such clumps (or "fish eyes") form, discard the

solution and begin again.

Also, BE SURE that the solid particles do not

touch the beaker walls or the blender neck (or the stirrer paddle) as they are being sprinkled into the dilution water.

Since these surfaces will be

wet, the grains will stick where they land and will not dissolve further. If this occurs, discard the solution, thoroughly wash the blender (or stirrer) and beaker, and start again. 4)

Continue mixing for 60-90 seconds or until the solution appears homo

geneous .

Then allow the solution to age for 15 minutes, then blend for

another 30-60 seconds before further dilution.

(For bench stirrers; mix for

1 hour, allow to age overnight, and mix again for 1 hour before further dilution). If several attempts do not provide a homogeneous mixture (large particles of solid polymer visible), repeat the procedure using dilution water heated to 100°C.

Allow the polymer solution to cool before proceeding to the next

step. 5)

Insert the pH probe(s) into the polymer solution and record its pH.

If

pH adjustment is necessary to attain the desired charge (as given in Table 1 in Chapter II), add 1 M acid or base as required.

If more than 1 mL of acid

or base is used, recalculate the polymer concentration (e.g. 5 g/1.001 L). During acid or base addition, continue mixing intermittently to ensure that accurate pH readings are being obtained. 6)

The resulting 1 L of solution is a 5 g/L (0.5%) stock solution.

Consult

Table 4 for information about preparing stock solutions of other strengths. b.

Working Solution Preparation

1)

Using a syringe, SLOWLY add 2 mL stock solution to 198 mL of dilution

water which is contained in the tall, narrow plastic vessel, and is being mixed by the hand-held blender (or bench stirrer) at its maximum speed.

53

BE

CERTAIN that the polymer solution enters the dilution water at a region of high turbulence, and that the solution leaving the syringe does not make contact with the vessel walls or with the blender neck (or with the stirrer paddle).

If such contact is made, discard the solution and begin again.

Continue mixing for 30-60 seconds or until the solution is homogeneous. (For bench stirrers, continue mixing for 1 hour.) 2)

The resulting 200 mLs of solution is a 0.05 g/L (0.005%) working solu

tion.

This solution does not require aging, and should be used as soon as

possible after its preparation is completed.

To prepare working solutions

of different strength, refer to Table 5. 3)

Dosage:

10 mL of the working solution added by syringe to 1 L of raw

water yields a polymer dosage of 0.5 ppm (or mg/L) for jar testing.

Tables

6 and 7 list information concerning the preparation of other jar test dos ages and final volumes.

54

.Example 1.

Preparation of a Specific Final Polymer Concentration From a Neat Polymer Product For this example, assume a 1 L jar test requires a viscous liquid flocculant dose of 10 ppm (or 10 mg/L, or 0.001% (v/v)). How should this be accomplished? It is first necessary to refer to Table 3. Locate the phrase "viscous liquid flocculant" under the heading "polymer type"; then read across this row and note the recommended concentration ranges for primary dilution (520%), secondary dilution (0.05-5%), and final concentration (0.1-10 ppm; our choice of 10 ppm is acceptable since it falls within this range). These concentration ranges will be employed, but the specific values will be determined in reverse order.

That is, we mentally "backtrack" from the final desired jar test concentration of 10 ppm, to the concentration of the working solution with which the jar is dosed, to the concentration of the stock solution from which the working solution is prepared, and finally to the original amount of neat polymer product from which the stock solution is prepared. (That is, from Table 7, to Table 5, to Table 4.) Now turn to Table 7 since a 1 L final jar test volume is being sought. Locate the vertical column headed by "10 ppm"; this is found under the section labelled "Desired Concentration in Jar." (This column is the second vertical column when counting from the right.) The numbers in the column below this heading are volumes (in mL) of working solution with which the jar will be dosed. Next choose a working solution concentration from within the range given in Table 3; as previously noted, this range is 0.05-5% (v/v). This illustra tion will utilize a 0.1% (v/v) working solution concentration. Now, find the value 0.1% in the vertical column on the far left of Table 7 which is headed "Concentration of Working Solution with which Jars are Dosed." Read across horizontally from this value of 0.1% until the vertical column headed

55

JU

"10 ppm" is intersected.

The value at this point of intersection is 10 mL , indicating that 10 mL of a 0.1% working solution should be added to 1 L of raw water to obtain a final polymer concentration of 10 ppm for jar testing. Next, turn to Table 5 and locate the vertical column headed by 0.1%; this is found in the center of the section labelled "Desired Working Solution Concentration." Using the range supplied in Table 3 for acceptable stock solution concentrations (5-20%), make a choice from among those found in the far left vertical column entitled "Concentration of Stock Solution Being Diluted." This illustration will utilize the value 10%. Follow horizontally across from this 10% value until the vertical column headed by 0.1% is intersected. The volume value at this point of intersection is 2 mL. Therefore 2 mL of a 10% stock solution, when added to 198 mL of dilu tion water, will yield 200 mLs of a 0.1% working solution. For the final stage of "backtracking", go to Table 4. Locate the 10% stock solution value found in the first horizontal row. Directly below this number is the weight in grams of neat polymer required to prepare 1 L of a 10% stock solution. This weight is 100 grams. Directions for the preparation of this stock solution (as well as the other solutions described) are found in Section 5 of this module. The actual procedure followed in the laboratory is the reverse of this "backtracking" process. First, Table 4 is consulted to find out what quantity of neat liquid solution flocculant is required for the preparation of a 10% stock solution. This solution is then prepared as described in Section 5.a. The amount of 10% stock solution required for the preparation of 200 mLs of a 0.1% working solution is then located in Table 5. The desired 0.1% working solution is prepared as described in Section 5.b. "$C

Since a range of working solution concentrations is acceptable, the analyst is offered a number of choices. If possible, a concentration should be selected which calls for the addition of < 50 mL of working solution to the raw water so that the final jar test volume will still be approximately 1 L.

56

Finally, the amount of 0.1% working solution necessary for the preparation of a 10 ppm polymer concentration for jar testing is found in Table 7 (or in Table 6 for a jar test on a 2 L raw water sample).

This volume of polymer

solution is added to the raw water during the "rapid mix" phase of jar testing.

The final polymer concentration in the jar is 10 ppm, as initially

desired. Example 2.

Direct Working Solution Preparation from Neat Liquid Solution

Coagulant Type Polymers As pointed out in the introduction to Section 4, these polymers are prepared in a manner unique to their type.

Therefore, the tables provided will be

utilized differently, and an example of such use is called for. First of all, Table 4 should not be used at all.

No stock solution prepara

tion is required; working solutions are made directly from the neat polymer products.

Dosages of neat polymer product for initial dilution are found in

the last horizontal line of Table 5.

The term "100% stock solution" is

actually another way of referring to the pure polymer product.

With this in

mind, the use of Table 5 is straightforward. The final horizontal line in Table 5 is comprised of the final entry in each of the vertical columns; that is, the volume % and g/L headings associated with these vertical columns still apply to this last horizontal row of values.

Thus, Table 5 indicates that in order to prepare 200 mLs of a 10%

(or 100 g/L) working solution directly from a pure liquid solution coagulant-type polymer, the neat polymer volume required is 20 mL. proper dilution procedure for liquid solution coagulant polymers is described in Section 4.a. of this module.

57

The

Example 3.

Preparation of Polymer Concentrations Which Are Not Listed in

the Tables Suppose it is desirable to prepare a 4% (v/v) stock solution. Since Table 4 does not list the neat polymer dosage required for such a stock solution, one of two methods must be employed; the interpolation between values given for a 1% and 5% stock solution, or the algebraic calculation of this dosage using those dosages which are given. To interpolate, a simple relationship of proportionality is employed as follows: 1% / 10 g = 4% / x g

or

x = 40 grams

To algebraically calculate this dosage, first locate those % stock solution concentrations which are closest to the one desired. In this case, those concentrations are 1% and 5%.

Then simply perform that algebraic operation

which would yield 4% from 1% and 5% on their respective dosages. examples follow: 5% - 1% = 4%

4 x 1% = 4%

50 g - 10 g = 40 g

4 x 10 g = 40 g

Some

15% - (10% + 1%) = 4% 150 g - (100 g + 10 g) = 40 g There is no single preferred method for this type of calculation, although it should be obvious from the above three cases that some are more simple than others. Just be sure that the resultant dosage is a logical one. Use common sense to predict that the dosage required for a 4% stock solution will fall between those dosages required for 1% and 5% stock solution preparations.

58

B.

1.

MODULE B.

RAPID SCREENING PROCEDURE FOR FLOCCULANTS

General Discussion

In general, all testing of flocculants will involve the jar test procedure as discussed in Module C.

However, since proper "jar testing" requires a

somewhat elaborate procedure, a simple screening test can first be utilized which is rapid and requires less equipment.

In this test, a polymer dose is

used which is likely to be much higher than the "optimal" dose; if welldefined floe are observed, smaller doses should be evaluated using the more elaborate jar test procedure given in Module C. polymer can be eliminated from consideration.

If no floe develops, the Because this is only a rough,

initial procedure for estimating polymer performance, a visual inspection of the extent of flocculation is the only "analytical method" required. This procedure is NOT appropriate for the screening of coagulants, including polymers used for this purpose.

If a polymer is described by the supplier

as a coagulant rather than flocculant, proceed to Module C.

If there is any

doubt about product type, refer to Tables 1 and 3 in Chapter II, Section B. 2.

Apparatus

1)

Magnetic mixer and stir bar.

One paddle on a bench stirrer, as describ

ed in Module C, may be substituted. 2)

2-L beaker or 2-L "square jar" as described in Module C.

If necessary,

smaller size containers can be used with appropriate adjustments to all dosages.

For example, if 1-L containers are used, all dosages in this

procedure should be cut in half. 3)

3-valve rubber pipette bulb.

59

4)

Graduated pipette, 25 mL or as appropriate for the polymer volume to be

dispensed. 5)

Small beakers (e.g. 50 mL) if chemicals are to be pre-dispensed.

6)

Plastic bucket for obtaining water samples.

3.

Reagents

1)

Working polymer solutions prepared according to the instructions in

Module A, or according to the manufacturer's instructions.

If the manufac

turer furnishes a range of recommended doses, calculate the volume of polymer solution required to provide the highest recommended dose to the The formula given in Module C, Section 3 can be used for this purpose.

jar.

If no recommended doses are given, use the volumes specified below. 2)

Water sample for polymer treatment (see below).

4.

Procedures

1)

With a bucket, obtain a. sample of the water to be tested.

This water

should be obtained from a point directly after the rapid mix chamber so that This water should be taken immedi

the coagulant has already been added.

ately prior to testing so that the temperature has not changed. 2)

Pour a 2 L sample from the bucket into the beaker.

Drop a 2-inch stir

bar into the beaker and rapidly mix the contents on a magnetic mixer.

A

1-inch vortex should be present in the beaker to assure good mixing condi tions . 3)

Dose the sample with 20 mL (or a volume calculated as described above)

of the working polymer solution, and continue mixing for 20 seconds to allow for dispersion.

60

4)

Slow down the mixing speed (so there is a 1/4-inch vortex) and continue

mixing for 5 minutes. 5)

If large floe form, the polymer should be further investigated.

For

each test, note the time at which floe form, their general size after 5 minutes, and whether the fluid between the floe appears clear. 5.

Interpreting Results

This test is useful to determine which aids will probably enhance or increase flocculation.

It does not provide conclusive results as to which

aids and doses are effective, but rather gives an indication as to those that should be tested first in the more time-consuming jar test procedure. Generally, when floe size increases due to the addition of a polymer, it can be safely concluded that flocculation is being enhanced, and that the polymer warrants further investigation.

A proper dose should also lead to

the water between the larger floe being essentially free of turbidity or smaller particles, but determination of this optimum dose is best done in subsequent testing with the more elaborate jar test procedure (Module C). 6.

References

Introduction to Water Quality Analyses. American Water Works Association Publication # 1931 (1982). Hudson, H.E. and Wagner, E.G., "Conduct and Use of Jar Tests."

AWWA Seminar

Proceedings-Upgrading Existing Water Treatment Plants. American Water Works Association Publication # 20126 (1980).

61

C.

MODULE C.

1.

General Discussion

THE JAR TEST

The jar test procedure basically involves filling 2-L containers with the water to be treated, placing them under a multipaddle stirring machine (bench stirrer), dosing the jars while at a high paddle speed (rapid mix), flocculating at a lower speed, and finally, allowing sedimentation to occur under quiescent settling conditions. The jar test has seen substantial use in the water treatment industry due to its ability to simulate rapid mix, flocculation, and sedimentation in one vessel. The key consideration in the running of a jar test is to simulate the physical conditions in the treatment plant as closely as possible. These conditions include such parameters as detention times, velocity gradient (mixing intensity), temperature, pH, and order of addition of all coagulants, pH control chemicals, and polymeric aids. Varying any of these parameters during a jar test may significantly alter the final results and produce poor correlations with actual full-scale plant processes. 2.

Apparatus

1)

Containers or "jars".

glass jar was acceptable.

When the jar test was originally developed, any Since then, however, research has shown that the

jar type can have a significant effect on the success of the test.

Although

there are a number of different jar configurations available, a 2-L acrylic plastic square jar is recommended. This type of jar is essentially unbreak able and is less heat conductive than glass. The 2-L volume enables filtra tion and sludge characterization tests discussed in later Modules to be incorporated in this procedure. Such jars can either be purchased (Hach Chemical Company, Loveland, CO or Phipps and Bird, Inc., Richmond, VA) or easily constructed. Construction requires a table saw with a fine toothed blade, 1/4 inch thick acrylic plastic ("Plexiglass"), and acrylic plastic

62

solvent.

Details as to construction techniques may be obtained from most

glass dealers.

Figure 6 gives the approximate dimensions for the jars.

They should be provided with a calibrated 2-L mark. although less preferable, may be substituted.

Other container types,

However, do not mix or

compare results between different types of jars. 2)

Variable speed multistirring machine (bench stirrer) with at least 5

paddles.

Make sure there is adequate clearance for the jars: if not, the

legs may need to be modified. 3)

Illuminator base for optional use with bench stirrer.

Enhances

observation of floe formation by providing effective glare-free illumination of floe samples. plastic plate.

Consists of fluorescent lamp mounted below translucent Test jars and support posts of stirrer rest directly on

diffuser top plate. 4)

Plastic pail to collect enough sample to fill all the jars.

5)

Syringes or glass pipettes to dispense coagulant, polymer, limewater, or

other chemical. 6)

3-valve rubber pipette bulb.

7)

Magnetic mixer, stir bars, and stir bar retriever for mixing solutions

prior to dispensing into jars. 8)

Plastic syringe, large "turkey baster", or similar device to withdraw

sample from jars for further analysis.

This device should have a mark at

some distance from the tip (e.g., 5 cm) so that the sample is taken from a constant distance below the surface in every jar.

Alternatively, a hold may

be drilled about 2.5 inches from the bottom of the jar to insert a sampling port with flexible tubing and a pinch clamp as shown in Figure 6.

The

distance at which the port is located should be identical for all jars.

63

-4.5 : 4.5

FIGURE 6. Dimensions of 2-Liter "Square" Jar. Acrylic Plastic ("Plexiglass").

64

Material is 1/4 Inch

9)

Stopwatch or timer to time the periods of rapid mix, flocculation, and

sedimentation. 10)

Beakers, 100 mL or larger to collect samples for further analysis

(turbidity, particle size analysis, color, alkalinity, etc.). 11)

The rmome te r.

12)

(Optional) A "reagent rack" can easily be constructed to facilitate

simultaneous chemical addition to all jars.

It basically consists of a

board which holds tubes or cups spaced equally to the distance between jars. Further details are provided in the ASTM jar test procedure (see Reference section.)

If syringes are used to dispense chemicals, this rack is

unnecessary. 3.

Reagents

1)

Coagulant aids to be evaluated, in proper working solution strengths.

If instructions for the preparation of such solutions are not provided for a given product, see Module A. 2)

Coagulants and Other Chemicals.

In general, all chemicals (coagulants, lime, etc.) should be identical to those used in the plant and preferably should be taken directly as supplied, since the actual concentrations are often not accurately known.

For

example, lime slurry should be obtained directly from the lime dosing line. Chemicals can also be prepared according to the procedure outlined in AWWA Introduction To Water Quality Analyses.

However, alum or ferric solutions

should not be diluted more often than necessary beforehand, since this may alter their effective dose.

Alum stock solutions should be at least 30 g/L

and ferric chloride stock solutions should have a concentration of at least gOQ g/L.

Due to the solution chemistry of these chemicals, concentrations

less than the values mentioned above may be less effective than those which

65

are more concentrated, even when identical doses are used. Consequently, liquid ferric chloride and alum should be used in the concentrated form whenever possible.

If dilution is necessary, it should be done just prior

(within seconds) to jar testing. To determine the final concentration (C 2 ) °f a solution after dilution, use the following formula:

G!

X

VI ———

=

C2

V2 where

C^ = Concentration of initial solution, (mg/L) , YI = Volume of original solution being diluted, (mL) , V2 = Total volume of new solution, (mL) , C 2 = Final concentration of new solution, (mg/L) , and 1 mg/L = 1 ppm.

For example, if the initial concentration (C^) of a stock solution is 900 ppm and 25 mL (V^) of this stock solution is added to a flask and diluted to 500 mL (V2), then the final solution concentration is (900 X 25)/500 or 45 ppm = 45 mg/L (C2) . If coagulants are purchased as liquids, information concerning their concentration may be obtained from the manufacturer. Gen eral concentration ranges are also given in AWWA Introduction to Water Treatment . 4.

Procedure

a.

General Considerations

Before performing the jar test, label all reagents and mix them thoroughly with magnetic stirrer and stir bar. Label all pipettes, syringes, and Decide on chemical dosages including any lime, activated carbon, permanganate, etc. before the jar test begins, record them on the data

beakers.

66

sheet, and put the sheet in a highly visible location.

See steps 1 and 2

below for determination of lime and coagulant aid doses. Several methods can be used to rapidly dispense chemicals into jar test containers.

When pipetting, it may be convenient to use one or two large

pipettes (which are filled prior to the start of the jar test).

Each

pipette should have a 3-valve rubber bulb so that a number of accurate doses can be dispensed before refilling the pipette.

There may be a need to

dilute a chemical to decrease the likelihood of measurement error when pipetting.

If possible, however, the chemicals used in jar tests should be

identical both in origin and concentration to those in the plant. Another method which allows rapid addition of chemicals (coagulants such as alum or ferric, lime, activated carbon, etc.) during the jar test is to predispense them into syringes or small beakers (appropriately labelled). These are then lined up for each jar test container and used at the approp riate stage of the test.

After use, they can immediately be rinsed with

distilled water which is also added to the jar.

This method helps prevent

pipetting error and is especially recommended when a large number of different chemicals are to be added to the jars.

The "reagent rack" listed

under Apparatus can also be used in conjunction with small beakers or other containers, but is not needed with syringes. Use of small beakers is not recommended when adding viscous polymer solu tions since their dispersion may be adversely affected.

Such polymers also

tend to adhere to glass and large volumes of rinse water would be required to completely remove the polymer from the walls of the small dosing beakers. If predispensing chemicals into syringes or small beakers, label them first. Dispense the proper chemicals into them, and locate them near the proper jar test container.

Analysis equipment such as turbidimeters or spectro-

photometers should be ready for use, since initial measurements such as turbidity and color should be taken prior to the initiation of the jar test.

67

The 2-L jars and paddles from the multi-stirrer should be cleaned by wiping with a damp cloth and rinsing with warm tap water in between each test to remove any polymer residue.

Similarly, pipettes or syringes used for

dispensing a polymer should be thoroughly rinsed once a day or immediately if a different polymer is used. This may be accomplished by rinsing with a large squeeze bottle filled with hot water.

Occasionally, it may be

necessary to acid wash with 1 Normal acid to remove any residual polymer. Jar testing should be conducted at similar temperatures to those of the water in the actual treatment process. If the water sampled is cold (as in winter), the jar test must be performed before the sample warms appreciably. The optimal coagulant aid dose may differ substantially at different temperatures. To best simulate coagulation, flocculation, and sedimentation at colder temperatures, it may be necessary to place the jars in a water bath, i.e. a container or tray which is filled with cold water or actual plant water so that most of the jar is immersed and held close to process temperatures. If the purpose of a coagulant aid is to improve performance under difficult conditions, such as high turbidity, then an attempt should be made to run jar tests when raw water in this condition can be obtained. Regardless of the jar test procedure used, there may still be differences between jar test results and plant results. The most important consid eration is not whether bench-scale and full-scale systems provide identical results, but that the bench-scale jar tests predict the polymer dose that is optimal for the plant. The following are possible causes why optimum dose may be falsely predicted: Deposition of coagulant aid on the mixing blade of the bench stirrer or walls of the jar. Improper preparation of the reagents. Excessive aging of the reagents. Incomplete dispersion of polymers added to jars.

68

Measurement error (particularly with viscous polymers). Use of reagents from a different source than those used in full-scale plant processes. Mixing conditions during flocculation in the jar test that are drasti cally different from those in the plant. b.

Procedure Using Raw Water

1)

Determine the coagulation aid doses to be evaluated in this jar test

series.

Generally, because polymers are expensive but usually work well at

low concentrations, the maximum dose should be less than 1.5 mg/L and doses as small as 0.05 mg/L may prove effective as an aid for coagulation.

If the

supplier of a polymer suggests a range of recommended doses, try these in the jar tests.

Otherwise, a. good range of doses to try first might be 0, The proper volume of polymer solution

0.1, 0.25, 0.5, 0.75, and 1.0 mg/L.

For example, assuming that the polymer

can then be calculated for each jar.

to be tested is at a concentration of 0.1 g/L (0.1 mg/mL), the amounts of polymer to be added for the above doses would be 0, 2, 5, 10, 15, and 20 mL. Once these doses are evaluated, further series of jar tests can be performed with different dosages to more exactly locate the optimal polymer dose. 2)

iIf lime or other chemicals are used for pH control in the plant, check

that the jar test dose will produce the desired final pH. the pH in the flocculators. tested.

Usually this is

Fill a 2-L glass beaker with the water to be

Add a 2-inch magnetic stir bar, place the jar on a magnetic mixer,

and turn up the mixing speed so as to form a 1/2 inch vortex.

With a

pipette, dose the jar with coagulant (identical to the full-scale plant dose).

Insert the pH electrode into the jar and add small doses of lime or

other chemical to adjust the pH to the desired level.

(If limewater is

used, it is important to shake it up just prior to addition so that the settled lime is uniformly dispersed.)

Note the volume of lime or other

chemical added since an identical volume will be used in all subsequent jar

69

tests.

These volumes can also be pre-dispensed into syringes or small

beakers for rapid addition during the jar test. 3)

If coagulant, lime, or other chemicals (carbon, permanganate, etc.) are

to be pre-dispensed, complete this step of the procedure and organize the syringes or small beakers before beginning the jar test. water to fill the jars.

Collect enough

Measure and record all initial parameters of

interest (pH, turbidity, color, etc.) from the head of the treatment plant. 4)

Pour 2-L of sample into each square jar.

5)

Position the jars under the multi-stirrer so they are centered with

respect to the shaft of the paddles. 6)

Lower the paddles so that the top of each paddle is equidistant from the

surface of the water and the bottom of the jar. 7)

Begin rapid mix.

If no other paddle speed is otherwise specified, the

maximum RPM is recommended (about 125 RPM on most bench stirrers). 8)

Using a 25 mL graduated pipette or calibrated syringe, dispense ident

ical doses of coagulant as rapidly as possible to each of the jars.

The

coagulant dose should be the same as used in the plant although the solu tion may need to be diluted prior to the jar test.

For the purpose of

accuracy, the volume of coagulant dispensed to each jar should be at least 3 mL. 9)

For smaller alum doses use a smaller graduated pipette (e.g. 5 mL).

To each of the jars, add any lime or other chemical used for pH control.

The dose to be added was predetermined in step 2.

(Note:

Steps 9, 10, and

11 may be reordered depending on the point of lime or anticipated coagulant aid addition.) 10)

Bypassing the first jar (which is the zero polymer dose), apply

increasing coagulant aid doses to the remaining jars.

As with the alum,

dosing is achieved by using a large graduated pipette or syringe and insert-

70

ing the tip 1/2 inch below the water surface of each jar while releasing the polymer. Continue rapid mix for 15 seconds after the last jar has been dosed to facilitate good dispersion. 11)

Decrease mixing speed (e.g. 25 RPM) to simulate mixing conditions

during flocculation in the full-scale process. 12)

Flocculate for a period similar to the estimated flocculator detention

time (e.g. 20 minutes).

Visually observe floe size in the first jar (where

only the primary coagulant has been added and no polymer) and compare to the actual floe size observed in the plant. If the floe size in the jar is much larger than in the plant, the mixing speed in the jar may be too low.

If

the floe size in the jar is significantly smaller than in the plant, the mixing speed in the jar may be too high.

If the difference is substantial,

the jar test should be revised and repeated.

If the floe size is larger

than in the plant, determine a better paddle speed by increasing the paddle speed until floe size resembles that in the plant. If the floe have settled to the bottom during flocculation, readjust paddle level lower in the container and allow floe to be redispersed. (Note: More elaborate methods of establishing the times and RPMs which best simulate plant conditions may be found in Schull (1967) and Cornwell and Bishop (1983). (See References at the end of this Module.) 13)

Stop the mixer, pull up the paddles, and allow sedimentation to occur.

If this is the first jar test series to be run, this step must be adjusted to simulate plant settling as much as possible.

To do this, periodically

measure turbidity in the first jar (with no coagulant aid).

Take care not

to significantly disturb jar contents while taking these samples.

When the

jar's turbidity is approximately equal to the settled plant turbidity, this time should be recorded and used for all subsequent jar test settling periods. 14)

(Depth of sampling may be adjusted for this purpose).

Withdraw samples for desire analyses (turbidity, particle size

analysis, true color, pH, etc.).

Sample withdrawal may be accomplished

either by the use of a syringe or sampling port.

71

When using a syringe,

samples should be taken from a specific depth. For example, a piece of tape 5 cm from the end of the syringe serves as a reliable marker for depth. The syringe should be rinsed with distilled water before sampling from different jars. The first 150 mL taken from a fixed port, however, should be dis carded before using to measure for turbidity, color, or particle size analysis. c.

Procedure for Water from Rapid Mix

This procedure is applicable to water which has been sampled after coagu lants such as alum or ferric chloride have been added, but before polymer addition. This is useful only when the goal is to improve water quality using a polymer without modifying the alum or ferric dose. The procedure is similar to the one given previously, but is easier to perform since the water to be tested has already been dosed with one or more chemicals during rapid mix. All steps are the same, except Step 8 is omitted. If chemicals for pH control have been added in the plant prior to the sampling point, Steps 2 and 9 are also omitted. d.

Procedure to Determine if Alum/Ferric Dose May Be Decreased as a Result of Coagulant Aid Addition

This procedure is applicable to situations where it is desirable to approxi mate the degree to which the primary coagulant, e.g. ferric or alum dose, may be reduced as a result of adding a coagulant aid. It differs from the preceding procedures in that the coagulant dose is varied while the previ ously selected aid dose is held constant. The range of doses which will be used will be determined by the dose which is presently being used in the plant. For example, if the plant operates at 100 mg/L alum under highly turbid conditions which are being assessed in the jar tests, then a good range of doses to try might be 10, 30, 40, 60, 80, and 100 mg/L alum. In this example the first jar would be dosed with 10 mg/L and increasing doses in subsequent jars until 100 mg/L is reached.

72

If lime or any other pH controlling chemical is presently employed, this dose must also be adjusted since its addition is meant to counteract pH changes due to the coagulant.

The amount added to achieve a particular pH

should be adjusted to be proportional to the coagulant dose.

For example,

if 60 mg/L of lime was used to arrive at a certain pH for a coagulant dose of 100 mg/L, then if the coagulant dose was reduced to one half of 100 (i.e., 50 mg/L), the lime dose would also be reduced by the same factor to 30 mg/L (i.e., 1/2 x 60 mg/L).

The different lime and coagulant doses would

therefore be known before the jar test begins and could be pre-dispensed into small beakers.

The Procedure Using Raw Water (Section C.4.b.) is then

used with the above modifications. test should be measured.

In addition, the final pH of each jar

If these pH values are not within 0.5 pH units of

the settled water in the plant, readjust the lime doses and repeat the jar test. A more time-consuming procedure can be used if the precise combination of optimal coagulant and coagulant aid doses is to be located.

An entire set

of coagulant aid additions can be evaluated at each coagulant dose, thus covering many combinations of dosages.

For example, six polymer doses of 0,

0.1, 0.25, 0.5, 1.0, and 1.5 mg/L could be employed in each jar test set, with a constant alum dose.

The procedure would be repeated using alum doses

(in all six jars) of 10, 25, 50, 75, and 100 mg/L alum.

The optimal dose

combination can then be determined as illustrated in the next section.

The

disadvantage of this approach is obviously the large number of jar tests that must be run; note that a very large raw water sample should be obtained so that all of these tests are run on the same water.

This is particularly

important if coagulation is being evaluated under storm or other rapidly changing conditions. 5.

Interpreting Results

Most importantly, the quality control steps described in Section II.F should be followed before any jar test results are used to determine actual plant dosages.

This will insure that the data used for chemical selection is

statistically meaningful.

In addition, the procedure should be repeated

73

with varying raw water conditions to get some idea of the range of dosages that may be required. Use of the overall procedure will be illustrated in the following example, including the tabulation and graphical analysis of results.

To predict

optimum polymer dose for a given sample, procedure C.4.b. or C.4.c. would be used.

Table 8 shows data that might result and how it could be tabulated.

The turbidity versus polymer dose could then be plotted such as on Figure 7 connecting the points with a straight-edge.

From this graph, an optimum

dose of 0.4 mg/L of the polymer might be chosen.

This value might be

preferable to 0.3 mg/L to provide a factor of safety.

Although in this

example, the coagulant aid significantly improved turbidity removal, it would also be desirable to learn if the coagulant dose could be decreased to save chemical costs while maintaining similar turbidity removal.

Following

procedure C.4.d., the results could be recorded as on Table 9, and graphed on Figure 8.

The coagulant dose could be decreased to 90 mg/L in this case,

a 10 percent decrease in primary coagulant, while having about the same turbidity removal.

A 90 mg/L dose might be chosen rather than 80 mg/L to

provide a factor of safety. To warrant the use of a coagulant aid, it should significantly reduce the turbidity, color, or some other measure of performance.

For example, if a

coagulant aid only reduces the turbidity by about 20 percent at optimum dose it probably isn't economically feasible to use, unless it reduces the primary coagulant dose enough to save significantly on chemical costs. Other factors, such as sludge disposal costs, may also be significant factors. Figure 9 shows a different set of jar test results obtained by the more lengthy procedure of varying both polymer and coagulant dosages.

The point

is located on the graph for each coagulant and coagulant aid combination, and the resulting turbidity is noted.

After all the data is on the graph,

lines of equal turbidity (contour lines) are drawn.

This produces an

"isoturbidity topogram" which can be examined to help select a dose combina tion (log-log graph paper is most satisfactory for such graphs because a

74

TABLE 8 Example tabulation of jar test results varying flocculant dose.

DATE

HAW WATER CHARACTERISTICS

OPERATOR

COAGULANT UD

P« _________________

POLT

COM. UD COHCEHTRATIOII 0.1 AL0H/rtRRIC D03B

J ^^

a

O

i-h

0) •d (H

9L

TURBIDITY HFTER.2S. MIN. SE-D

a

^

*^ •a

In

10 -
in CL

ro

Z

-< O "t

M)

fl> o 0

n -
••

JU

ILOM/mRZC DOSE (M/L)

-

LXHZ D05Z (M/t)

rzxjo. TTOBIDITT nHii pa rixiL COLOR rZXAL TXM?.

riHZ ST1RTEO

riMS it on or rue. riKI TO TAIE S1HPLE f ••••••••••«•«•••••••• ••• t •••••••••• t«llf*llft« I Ktf l«lt«« •••••••••••••••I

OBSZRT1TIOK3

82

oo

U)

£

(E

a u .

z tH r

a u

e.B

8.8 8.6 8.4 POLYMER DOSE (mg/l) _______, RLUM/FERRIC DOSE ___mg/l) (POLYMER NRME ) (RRH HRTER TURBIDITY , FINRL pH OF JRRS -

e.2

. HHTER TEMP.

FIGURE 10. Blank Graph for Plotting Jar Test Results Varying Flocculant (Photocopy for lab use) Dose With Constant Coagulant Dose

'

, DRTE

TURBIDITY vs. POLYMER DOSE

OPERRTOR

00

2

4

48 68 BB PLUM/FERRIC DOSE (mgX|) (POLYMER NRME/DOSE __________ X ___ mg/M ) (RRH HRTER TURBIDITY____, -FlNflL pH OF JRRS - _

28

IBB

FIGURE 11. Blank Graph for Plotting Jar Test Results When Varying Coagulant Dose With Constant Dose of Flocculant (Photocopy for lab use)

a a

a >-i

n

a:

HLL.

oc u

a u

18

TURBIDITY vs. RLUM/FERRIC DOSE

OPERRTOR______________, DR'TE______, HRTER TEMP.

INTRODUCTION TO WATER QUALITY ANALYSIS (Vol. 4), American Water Works Association Publication # 1931 (1982). James M. Montgomery, Consulting Engineers, Inc., Water Treatment Principles & Design. NY: Wiley Interscience, p.455 (1985). "Practice for Coagulation-Flocculation Jar Test of Water." Annual Book of ASTM Standards: Water and Environmental Technology. 11.01:869 (1987). Schull, K.E., "Filterability Techniques for Improving Water Clarification." Journal AWWA. 59:1164 (1967). TeKippe, R.J. and Ham, R.K., "Coagulation Testing: niques." Journal AWWA. 62:594 (1970).

85

A Comparison of Tech

TURBIDITY MEASUREMENT FOR JAR TEST ASSESSMENT

D.

MODULE D.

1.

General Discussion

Turbidity measurement, in most cases, is sufficiently sensitive for evalu Limitations and

ation of coagulant and coagulant aid effectiveness.

specific procedures will vary according to the model of turbidimeter employ ed.

General points concerning turbidity for jar test assessment follow; for

more specific instructions, refer to the operating manual for the individual turbidimeter and to publications listed at the end of this Module under References.

Also refer to Section II.F regarding quality control in

analytical procedures. 2.

Apparatus

1)

Laboratory turbidimeter.

For highly colored water, a ratio-type

instrument is recommended to correct for absorbance. 2)

Sample cells for the appropriate turbidimeter.

Cells to be used should

be clean and free from scratches. 3)

Turbidity standards to calibrate the instrument.

4)

Lens paper and cleaning tissue, such as Kimwipes.

5)

Syringe or pipette for dispensing samples.

3.

Reagents

If the appropriate turbidity standards are not provided with the turbidi meter, it will be necessary to prepare them.

Proper calibration using such

standards is necessary if results are to be compared between past and

87

present jar test analyses.

Instructions for preparing these standards may

be found in Standard Methods for the Examination of Water and Wastewater. 4.

Procedure

Pipettes and syringes should be well rinsed between samples. The most accurate turbidity readings will be obtained if the same turbidity cell is used for all samples and it is marked to be oriented in the same direction each time it is placed in the turbidimeter. well cleaned between samples.

Be sure that the sample cell is

This insures accurate readings.

Lens paper

is recommended for the final cleaning. Collected samples should be carefully poured into the turbidity cells immediately before measuring the turbidity. There should be no visible bubbles in the turbidity cell since these interfere with light transmittance. If more than a minute passes between collecting the sample before pouring it into the cell, it will be necessary to gently swirl the sample before adding it to the turbidity cell. 5.

Data Interpretation

In most cases, interpretation of turbidity measurements is straightforward. However, if the readings appear to be lower than the sensitivity range of the instrument, it will be difficult to confidently evaluate coagulation results. The following changes may then be considered: 1)

Use a more sensitive turbidimeter.

2)

Increase the sampling depth in the jar test container.

3)

Decrease the settling period.

4) Use a particle size analyzer instead of a turbidimeter (refer to Module E which follows).

88

6. References Standard Methods for the Examination of Water and Wastewater, APHA, AWWA (Publication # 10035), and WPCF. Washington, D.C. (1985). Introduction to Water Quality Analyses. Vol. 4. American Water Works Association Publication # 1931 (1982). Methods for Chemical Analysis of Water and Wastes. U.S. EPA. Transfer Series. EPA-625-/6-74-003 (1974).

89

Technology

E.

MODULE E.

PARTICLE SIZE ANALYSIS FOR JAR TEST ASSESSMENT

This Module is concerned with general aspects of particle size analysis with respect to jar test assessment.

Refer also to the manual accompanying the

specific particle size analyzer being used. 1.

General Discussion

The particle size analyzer is a very sensitive instrument capable of characterizing water quality based on particle number and particle size in a given sample.

This information can be used to locate the optimal coagulant

or coagulant aid dose in cases where turbidity is not sufficiently sensi tive.

However, the equipment required is expensive and its purchase would

generally be economically justified only if it will be used for other pur poses as well.

In some cases, research laboratories or polymer suppliers

may provide access to such equipment. 2.

Apparatus

1)

Particle size analyzer with appropriate sensor.

2)

Glass sample bottles.

Although these are commercially available, nearly

any clean glass containers that will fit into the analyzer are acceptable (e.g. beakers or small juice bottles). 3)

Ultrasonic cleaner large enough to immerse sample containers.

4)

Small magnetic stir bars (1/2 inch or less).

5)

250 mL and 50 mL graduated cylinders.

6)

1-liter plastic reagent bottles to store filtered water.

7)

Filtering apparatus (e.g. Millipore) including vacuum pump and hose.

91

8)

0.45 micron filter papers.

3.

Reagents

1)

Distilled, filtered (0.45 micron) water or electrolyte solution.

4.

Procedure

Particle size analyzers vary substantially in their capabilities and operating procedures. Thus, the manufacturer's recommendations should be followed for the use of each specific instrument. Some general rules are provided here which should assure reliable results. 1)

All glassware should be thoroughly cleaned and rinsed prior to use. The sample containers can best be cleaned by immersion in an ultrasonic cleaner for several minutes. If possible, water filtered through a 0.45 micron filter should be used for the final rinse. Glassware should be stored upside down and rinsed again immediately prior to use. 2)

Samples should be taken from the jar test using a device that will not break up flocculated particles. The tip of a pipette can be cut off, or a piece of flexible plastic tubing fitted over a syringe, to prevent floe rupture by passage through a small orifice. The volume of sample to be taken will depend on the particular analyzer requirements and whether or not the sample is to be diluted. 3)

Samples may require dilution because particle size analyzers are usually limited in the particle concentrations they can handle. Consult the analyzer's manual for specific limitations. If samples must be diluted, dilution water should be prepared by repeated filtering of a portion of the sample through a 0.45 micron filter. Some types of particle size analyzers require concentrated electrolyte instead of water, and this should be prepared in a similar manner. This may then be used to dilute the original

92

sample.

Volumes of sample and dilution water to be combined should be

measured with graduated cylinders. 4)

Even repeated filtering of the dilution water (or electrolyte solution)

will not remove all particles. also be performed on this water.

Therefore, a particle size analysis should The water should have substantially lower

particle counts than do the diluted samples, e.g. less than 5% in any size category.

Otherwise, the sample should be combined with a lower proportion

of the dilution water.

The particle counts for the original, undiluted

sample can then be obtained simply by multiplying the diluted sample counts by the dilution factor.

The dilution factor is the volume of the diluted

sample divided by the original sample volume.

For example, if a 50 mL

sample is diluted with 150 mL of dilution water, the dilution factor equals 200/50 or 4.0.

The diluted sample counts would be multiplied by four.

Some

particle size analyzers are equipped to do these dilution calculations (background calculations) automatically. 5. References

Beard, J.D., and Tanaka, T.S., "A Comparison of Particle Counting and Nephelometry." Journal AWWA. 69 (1977). Tate, C.H. and Trussel, R.R., "The Use of Particle Counting in Developing Plant Design Criteria." Journal AWWA. 70:698 (1978).

93

F.

MODULE F.

PAPER FILTER TEST

This Module describes a paper filtration procedure which can be used for two purposes:

(1) determination of an optimal filtration aid dose, or (2)

optimizing coagulant and/or flocculant selection with regard to its effects on filtration.

These applications are described respectively in Sections 1

and 2 of this Module. The paper filter test described in this Module is fairly rapid and easy to perform.

It is intended to simulate the process of filtration in a water

treatment plant, just as a jar test is meant to simulate the flocculation and sedimentation processes. minimal.

The resources required for this test are

Test materials and apparatus, as described later, are readily

available in most water treatment plants.

Although the procedures are not

difficult, test precision will improve as the analyst becomes more familiar with them. More involved methods of assessing chemical dosage effects on filtration may also be considered.

Various bench-scale tests have been described in the

literature (Dentel et al., 1987) which, although generally more complex than the paper filter method, utilize the same media used in the actual filter and may therefore provide more exact results.

A bench-scale filter

procedure described by Brink et al. (1988) has been shown to adequately predict coagulant performance in full-scale filters, and may also be valid for assessing flocculant or filtration aid effects on filtration.

The

bench-scale filter procedure described in the original version of this Manual is not recommended for evaluation of flocculants or filtration aids since it was found to lack precision and did not adequately correlate with full-scale filter performance.

The user of this Manual may wish to screen a

number of polymers first with the paper filter method, and then further assess the more successful ones with bench- or pilot-scale filters.

The

Reference section at the end of this Module provides further sources of information on such evaluations.

95

1. Determining Optimum Filtration Aid Dose a. General Discussion A filter aid can be used to enhance filter performance either in direct filtration or in conventional water treatment (flocculation and sedimen tation followed by filtration).

Direct filtration has been defined by the

Committee on Coagulation-Filtration of the AWWA's Water Quality Division as being any water treatment scheme in which there is no in-plant sedimentation prior to filtration (McCormick and King, 1982). flocculation basin prior to filtration.

There may or may not be a

Direct filtration without the

flocculation step is called in-line filtration or contact flocculationfiltration (contact filtration), since the flocculation occurs during contact with the filter media as opposed to volume flocculation in a flocculation chamber (Adin and Rebhun, 1974). The paper filter test assesses the impact of a filtration aid on the filtration process and is used to qualitatively predict the variation of the plant effluent quality with filtration aid dosage. An indication of the head loss that would develop across the plant filter bed is provided by the filtration time measured with this test. In turn, this provides a rough estimate of the filter run length which will result. This method is applicable when the filter aid is to be used in a direct filtration process, or when it is to be added at the end of the sedi mentation basin in a conventional treatment plant. Approximately one and a half hours are required for running this test. b. Apparatus 1) Multiple stirring apparatus (bench stirrer), Phipps and Bird type, variable speed. 2) Illuminator base, for optional use with stirrer.

Enhances observation of

floe formation by providing effective glare-free illumination of floe

96

samples. plate.

Consists of fluorescent lamp mounted below translucent plastic Test jars and support posts of stirrer rest directly on diffuser top

plate. 3) Jars, 2-L capacity, square, acrylic plastic, with or without side outlet, for preparation of samples to be filtered. 4) Bucket, for collection of sufficient sample to fill all jars. 5) Pipettes or plastic syringes, for dispensing the coagulant solution, filtration aid solution, and any other chemicals.

If a sufficient number of

syringes are available, these may be pre-filled for quick dispensing. pipette filler (bulb or wheel type) will be needed if 6) Filtering flasks, 1-L, pyrex glass.

A

pipets are employed.

Side-arm accepts standard 1/4 in. ID

flexible hose for vacuum connection, neck fits # 8 perforated stopper. 7) Filter holders, 47 mm diameter filter size, 300-mL capacity funnel, borosilicate glass.

Funnel and base with coarse fritted glass support for

filter, anodized aluminum spring clamp, neoprene stopper. 8) Filter paper, 5.5 cm diameter, Whatman # 40. 9) Vacuum pump, capable of generating a vacuum of at least 15 in. Hg (7.3 psi) . 10) Vacuum hose, flexible, 1/4 in. ID, length 2-3 ft. 11) Miscellaneous: stopwatch (with buzzer) and pH meter, c. Reagents 1) Filtration aid working solution, made according to the instructions in Module A.

97

2) Coagulants and other chemicals (in the case of direct filtration). In general, all chemicals (primary coagulant, lime, etc.) should be identi cal to those used in the plant and preferably should be taken directly as supplied, since the actual concentrations are often not accurately known. For example, lime slurry should be obtained directly from the lime dosing line. Chemicals can also be prepared according to the procedure outlined in AWWA Introduction to Water Quality Analyses. However, alum or ferric solutions not should be diluted more often than necessary beforehand, since this may alter their effective dose. Alum stock solutions should be at least 30 g/L and ferric at least 200 g/L. If dilution is necessary, it should be done just prior (within seconds) to testing. The concentration of a diluted solution can be calculated as follows: C 2 = GI x Vi / V2 where

C2 = concentration of dilute solution (mg/L) GI = concentration of stock solution (mg/L) V^ = volume of concentrated solution used (mL) V 2 = total volume of dilute solution (mL)

and

1 mg/L = 1 ppm.

If coagulants are purchased as liquids, information concerning their concen trations may be obtained from the manufacturer. General concentration ranges are also given in AWWA Introduction to Water Treatment. d. Procedure i. Preparation of working solutions of filtration aids. followed is described in Module A.

The procedure to be

If the product data sheet on the polymer is available, try different dosage levels in the recommended range. If not, recommended values of filtration aid doses are: 0, 0.01, 0.03, 0.05, 0.07, and 0.1 mg/L. To calculate the

98

volume of the filtration aid solution needed for each jar, use the formula given below. Vi - C 2 x V2 / C]_ where

V]_ = volume required (L) 62 = desired concentration (mg/L or mL/L) V£ = volume of sample (should be 2 L) GI = concentration of working solution (mg/L or mL/L).

Record the volumes needed for the six doses.

ii. Preparation of samples for filtration. (1) Contact filtration. (a) Use the bucket to collect about 15 L (4 gal) of the plant influent raw water.

Assemble the filtration apparatus as shown in Figure 12.

(b) This step is required to determine the dose of lime or other pH-control chemicals in subsequent jar tests. are employed in the plant.

It is unnecessary if no such chemicals

Pour 2 L of the water into a jar, and place the

jar in the multiple stirrer unit.

Begin rapid mix.

If no other paddle

speed is otherwise specified, the maximum RPM is recommended (about 125 RPM on most bench stirrers). about 1/2 inch. coagulant.

A vortex should be achieved with a center depth of

Using a syringe or pipette, dose the jar with the primary

The coagulant dose should be the same as that used in the plant.

Insert the pH electrode into the jar and add small doses of lime or other chemical to adjust the pH to the level presently existing in the plant filter influent.

Allow 1 minute for good dispersion of chemicals, and note

the volume of lime or other chemical added.

This dosage is used in all the

following jar tests, and may be pre-measured to facilitate prompt chemical dosings during the tests. (c) Gently stir the sample water to ensure that a uniform sample will be obtained.

Then pour 2 L of the water into a jar and place the jar in the

99

FUNNEL

SPHIHC CUHP

FILTER PAPEH BASE

7ACUUH PUMP FILTEHING FLASK

VACUUM GAUGE

AIH TO ATMOSPHERE

FIGURE 12.

Paper Filter Test Assembly

100

multiple stirrer unit.

Begin rapid mix.

Using a syringe or pipette, dose

the jar with the primary coagulant (identical to the plant dose). lime or any other chemical used for pH control. determined in step (b).

Also add

The dose to be added was

For the first run with zero polymer dose, proceed

directly to step (e). (d) Using the syringe or pipette, dispense the filtration aid solution into the contents of the jar, keeping the tip of the syringe or pipette half an The volumes required for different doses were

inch below the water level. calculated in step (i) above.

(e) Allow 1 minute for proper dispersion of added chemicals.

Switch off the

stirrer and remove the jar from the stirrer unit. (2) Direct filtration with flocculation (a) Use the bucket to collect about 15 L (4 gal) of the plant influent raw water. (b) This step is required to determine the dose of lime or other pH-control chemicals in subsequent jar tests. are employed in the plant.

It is unnecessary if no such chemicals

Pour 2 L of the water into a jar, and place the

jar in the multiple stirrer unit.

Begin rapid mix.

If no other paddle

speed is otherwise specified, the maximum RPM is recommended (about 125 RPM on most bench stirrers). about 1/2 inch.

A vortex should be achieved with a center depth of

Using a syringe or pipette, dose the jar with the primary

coagulant, at the same dose as used in the plant.

Insert the pH electrode

into the jar and add small doses of lime or other chemical to adjust the pH to the level presently existing in the plant flocculators.

Allow 1 minute

for proper dispersion of chemicals, and note the volume of lime or other chemical added.

This dosage is used in all the following jar tests, and may

be pre-measured to facilitate prompt chemical dosings during the tests. Reduce mixing speed to simulate mixing intensity during flocculation in the full-scale process (e.g. 20 RPM).

Visually observe floe size in the jar and

101

compare to the actual floe size observed in the plant.

If the floe size in

the jar is much different than in the plant, the mixing speed in the jar must be adjusted.

Refer to Module C (Jar Test), step 12, for details.

Make

a note of this mixing speed used. (c) Gently stir the sample water to ensure that a uniform sample will be obtained.

Then pour 2 L of the water into a jar and place the jar in the

multiple stirrer unit.

Begin rapid mix.

Using a syringe or pipette, dose

the jar with the primary coagulant (identical to the plant dose). lime or any other chemical used for pH control. determined in step (b).

Also add

The dose to be added was

For the first run with zero polymer dose, proceed

directly to step (e). (d) Using the pipet dispense the filtration aid solution into the contents of the jar, keeping the tip of the syringe or pipette a half-inch below the water level.

The volumes required for different doses were calculated in

step (i) above. (e) Allow 1 minute for proper dispersion of added chemicals. speed to that recorded in step (b).

Reduce mixing

Set the timer for the estimated floc-

culator detention time (e.g. 20 mins) and start the timer. (f) Assemble the filtration apparatus as shown in Figure 12.

Switch off the

stirrer when the buzzer goes off and remove the jar from the stirrer unit. (3) Conventional treatment (a) Use the bucket to collect about 15 L (4 gal) of unfiltered water from the plant.

If a filter aid is presently being used, withdraw the sample

from the end of the sedimentation basin prior to the injection point of the aid.

If no aid is being used, the sample can be taken from above the filter

media. (b) Assemble the filtration apparatus as in Figure 12. sample water to obtain a uniform sample.

102

Gently stir the

For the first run with zero poly-

mer dose, pour 2 L of the water into a jar and proceed directly to (iii) . (c) Gently stir the sample water in order to obtain a uniform sample. Then pour 2 L of the water into a jar and place the jar in the multiple stirrer unit.

Begin rapid mix.

If no other paddle speed is otherwise specified,

the maximum RPM is recommended (about 125 RPM on most bench stirrers). Using the syringe or pipet, dispense the filtration aid solution into the contents of the jar, keeping the tip half an inch below the water level. The volumes required for different doses were calculated in step (i). (d) Allow 1 minute for proper dispersion of added chemicals.

Switch off the

stirrer and remove the jar from the stirrer unit. iii. Filtration of samples. (1) Pour out 100 mL from the jar into the funnel.

The markings on the funnel may sometimes be inaccurate; it is

advisable to recalibrate them using a graduated cylinder. (2) Turn on the vacuum pump, and simultaneously start the timer (the pump should not be started before pouring in the sample, since filtration starts immediately and more than the required volume would then be poured in while bringing the liquid level up to the 100 mL mark). (3) When all of the sample has filtered through the paper, stop the timer, disconnect the vacuum hose, and turn off the pump.

Sometimes the time taken

for filtration of the entire sample is excessively long, especially at high er flocculant doses.

In such cases terminate the test after four minutes.

Make sure that the pump is vented to the atmosphere on shutdown.

This

prevents the oil in the pump from getting into parts where it can do damage. (4) Determine the quality of the filtered water by measuring its turbidity or particle count, according to the directions given Module G. Make any other measurements (color, UV absorbance, etc.) as desired. j.v. Testing for different doses.

Repeat the above procedure six times,

starting from step (c) in step ii., using a different dose for each run.

103

Maintain the same mixing time, mixing speed, and vacuum for all runs, e. Interpreting Results The example below illustrates the use of paper filter test data to predict an optimum dose of a filtration aid. Table 12 and Figure 13 show how experimental results could be tabulated and plotted. The graph of filtered water turbidity versus polymer dose suggests that a dose of 0.01 mg/L would give the best finished water quality. However, the filtration time, which is an indicator of the head loss developed across the filter media, is extremely high for this dose of the filtration aid. Increasing the dose to 0.03 mg/L would only marginally affect the quality of product water and substantially cut down on the rate of head loss development as compared to using a dose of 0.01 mg/L. Hence, in this case, a filtration aid dose of 0.03 mg/L could be chosen to make a significant improvement in finished water quality without drastically affecting the head loss developed across the filter bed. Still, the filtration time is more than double the value without any polymer dose, so if the filter run length is determined by terminal head loss, the run length might be cut in half. This would not be the case if turbidity breakthrough limits filter run length. Pilot or fullscale tests must be used for more accurate assessment of a polymer's effect on filter run length (See Section II.A for additional discussion). f. Data Tabulation and Analysis Table 13 and Figure 14 illustrate the recommended format for recording and plotting test results. These pages may be photocopied for laboratory use. 2. Optimizing Coagulant or Flocculant Usage With Regard To Filtration a. General Discussion Production of high-quality filtered water can sometimes be achieved with a coagulant or flocculant dose far less than that required for a floe that settles well (Hudson and Wagner, 1980). This lower dose would result in a

104

TABLE 12 Example Tabulation of Paper Filter Test Results Varying Filtration Aid Dose

DATE __ OPERATOR 20

MIXING TIME

40 "f~

MIXING SPEED POLIMER

SETTLED tfATEH CHARACTEaiSTICS TURBIDITT

130 *

TEMPERATURE

POLIMER COHCENTSATIOM PAPER FILTER USED:

~jj

o '

WHATMAN NO.

Jar no.

TRUE COLOR

ITC ____



Z

3

a

O-O 1

o -o 3

o

o • i.

1

2-n

H

5

6

O C'"'

0 ' I

Polymer dose

(ng/1)

Polymer volume for each jar (ml) Paper filter turbidity

0- 6

0-0$

1

'•4

2.

i- J3

0-4.1

9 -Si

o-SS

o-s-i

O-4.1

24-

111

51

36

(,1



Paper filter color Paper filtration time (aeca) COMMEOTS

105

n H •m

1-1

\

POLYMER

B.B5

0.BS

110 N

POLYMER DOSE (mg/L)

B.B3

B.B7

B.BB

B.B3

a. ta

FIGURE 13. Example of Graphed Results of Paper Filter Tests With Filtration Aid Dose Varied

0. 81

vA:

B.80

o-i

0-4

0-8

,.„

|-2

"4-

I-C

1-6

U.

ac ct h_j n

Z O n H

H

u n t-t

01

in u u

TABLE 13 Blank Data Sheet for Tabulation of Paper Filter Test Results When Varying Filtration Aid Dose

DATE _

OPERATOR

MIXING TIME

SETTLED WATER CHAflACTEHISTICS

KTXIHG SPEED

TURBIDITT

POLIMER

TEMPERATURE

____

TRUE COLOR

____

___

POLZMER CONCENTRATIOH _____ PAPER FILTER USED:

Jar no.

_____

VHATMAK HO.

1

2

Polymer dose (mg/1) Polymer volume for each jar (ml) Paper filter turbidity Paper filter color Paper filtration time (aecs) COMMENTS

107

3

4

5

- 6

o

00

D *-

m a:

O n

2

D h-

a.at 8.02

POLYMER

a.as

a.as

a.a?

POLYMER DOSE (mg'/L)

a. 03

B.BB

B.03

B.IB

FIGURE 14. Blank Graph for Plotting Paper Filter Test Results When Varying Filtration Aid Dose

a.ea

_J

H CC

n

o

U

O) O U

decrease in chemical costs, and the quantities of sludge generated, and in many instances, increased filter run lengths as well.

A floe that settles

well may not have good filtration characteristics, and a floe that has poor settleability may have good filterability; hence, a compromise will have to be made.

Selection of a coagulant and flocculant dose combination that will

optimize overall plant performance can be achieved by taking into account results from both the jar test and filterability studies (Ives, 1981). The paper filter test assesses the impact of coagulant and flocculant doses on the filtration process by evaluating the filterability of a suspension, and is performed subsequent to the jar test, which measures the settleabil ity of the suspension.

This procedure is used to qualitatively predict the

variation of the plant effluent quality with coagulant and/or flocculant dosage.

An indication of the head loss that would develop across the plant

filter bed is provided by the filtration time in this test. Along with the equipment needed to run the jar test, this procedure requires filtering flasks, filter holders, and a vacuum pump.

After the completion

of the jar test, an additional hour is required for running the paper filter test. b. Apparatus 1) Jar test apparatus: all apparatus required for conducting a jar test is needed, since the paper filter test is conducted using the jar test supernatant.

However, an additional requirement is that the jars must have

a side outlet for withdrawal of supernatant (lifting up the jars to pour the supernatant would resuspend sludge into the filtration samples).

See Figure

15 for dimensions. 2) Beakers, 2-L capacity, for collection of samples to be filtered.

This

allows sufficient excess volume to allow swirling of contents prior to filtration.

If sludge characteristics are not to be evaluated, or a 1-L jar

volume was used, beakers as small as 500 mL can be substituted.

109

4.5"

4.5

FIGURE 15. Jar Type Needed for Filtration Studies Subsequent to the Jar Test

110

3) Filtration aid apparatus: all the apparatus listed under filtration aid selection from 8) to 12). c. Reagents 1) Jar test reagents only; no additional reagents are required. d. Procedure 1) Since this test is run subsequent to the jar test, first follow the procedure for conducting a jar test, as outlined in Module C.

However, the

settled water quality is to be measured in a different manner as described in 2) below.

While the jar test is in progress, wash the beakers and

filtering apparatus with distilled water and have them ready to collect and filter jar test samples.

If distilled water is not available, then use the

effluent from the plant filters.

Keep all washed glassware upside down to

let wash water drain out, and to keep internal surfaces free from dust. 2) Collection of samples is done at the end of the settling period of the jar test.

Remove the pinch clamps on the side outlet tubings and let the

supernatant drain into a 2-L beaker.

Clamp the tubing again when the liquid

level is 1/2 in. above the side outlet.

Take a turbidity measurement (and

color or other measurements as desired) from this withdrawn volume. 3) Samples are filtered following the same procedure as that for filtration aid selection. 4) To test the different doses, repeat the filtration step for each beaker, since the contents of each beaker have been dosed with different amounts of flocculant. e) Interpreting Results To select the coagulant dose, flocculant dose, or combination of both that will optimize overall plant performance, results from both the jar tests and

111

the paper filter tests have to be taken into consideration.

The example

below illustrates the general procedure to be followed if evaluating various flocculant dosages only.

Table 14 and Figure 16 show how experimental

results could be tabulated and plotted. The optimum flocculant dose for settling, as suggested by the graph of jar test turbidity versus polymer dose, is 0.25 mg/L in this example.

However,

the plot of paper filter turbidity versus flocculant dose implies that decreasing the dose from 0.25 mg/L to 0.05 mg/L would only marginally increase the turbidity of the filtered water.

In addition, the effects of

the lower flocculant dose on filter run length can be estimated, since the filtration time in this test is a general indicator of the rate of head loss development across the filter bed.

Shorter filtration times imply slower

head loss build up and hence longer filter runs before terminal head loss is reached.

The plot of filtration time versus flocculant dose predicts that

filter run lengths may be drastically reduced if doses higher than 0.05 mg/L are employed and if run length is determined by terminal head loss rather than by turbidity breakthrough.

Pilot or full-scale tests would be required

for a more accurate prediction of filter run length. As compared to using only the primary coagulant, a dose of 0.05 mg/L of the flocculant together with the primary coagulant would improve filtered water quality and at the same time increase filter run length.

This dose would

cut down on chemical costs and quantities of sludge generated as compared to using a dose of 0.25 mg/L.

Hence, by using both the jar test and paper

filter test results, a dose of 0.05 mg/L could be chosen to optimize both coagulation and filtration. The same procedure would be used to assess the effect of various primary coagulant dosages on filtration, whether the coagulant was an inorganic salt or a polymer.

This includes lime as used in softening plants.

Results

would be interpreted in a similar manner as given in the example above.

112

TABLE 14 Example of Tabulation of Paper Filter Test Results When Varying Flocculant Dosage in Preceding Jar Test

DATE

ii/3/ys

OPERATOR

___

•ALUM/FERRIC DOSE LIME DOSE

PH

I I ~^/±J. *~

pH adjusted to POLIHER

RAW WATER CHARACTERISTIC:

2 "j 8 ' c>

TURBIDITr

l°i

THUE COLOR

/°*C ____

WHATMAH NO.

Jar no. Polymer doae («S/1)

' ^-

TEMPEHATURE

POLIHER CONCENTRATION PAPER FILTER USED:

7-r

1

z

*

3

O

o-oi"

o-io

O

2-

J

(•t

0-6

04.0

11

5

- *

o-&

| -OO

10

2.o

4*

9'f

9-2.

o 4.

o -r

O-Lf

o;s

o-\o

0- 1 '

o-iiT

4-S

163

714.0

^40

>l^o

O-iS"

Polymer voluoe for

•acn jar (ml) Jar teat

turbidity

Jar teat color Paper filter turbidity Paper filter color Paper filtration tine (sacs)

113

Of

ra

D H Z

0.8

o i-

ot

8.2 0.3

POLYMER

8.6

to i •

A^.OA-C

POLYMER DOSE (mg/l )

.S

8.7

B.8

*«*«*»

1.0

(.0

FIGURE 16. Example of Graphed Results of Paper Filter Tests Varying Flocculant Dosage in Preceding Jar Tests

fl.l

Pftft*

> J 1" i.

cc a:

z o

LJ O)

tn u

If combinations of coagulant and flocculant dosages are being evaluated, then the method of analysis is similar to that given in Module C and, specifically, shown in Figure 9.

The analysis is similar except filtered

turbidity is plotted instead of supernatant turbidity.

Log-log graph paper

should be used for such plots. f) Data Tabulation and Analysis Table 15 and Figure 17 are provided for recording and analyzing results, respectively.

These may be photocopied for convenient laboratory use.

3. References Adin, A. and Rebhun, M., "High-Rate Contact Flocculation-Filtration with Cationic Polyelectrolytes." Journal AWWA. 68:109 (1975). AWWA Seminar Proceedings-Coagulation and Filtration: Pilot to Full Scale. American Water Works Association Publication # 20017 (1987). AWWA Seminar Proceedings-Design of Pilot-Plant Studies. American Water Works Association Publication # 20164 (1982). AWWA Seminar Proceedings-Upgrading Existing Water Treatment Plants. American Water Works Association Publication # 20126 (1974). AWWA Seminar Proceedings-Upgrading Water Treatment Plants To Improve Water Quality. American Water Works Association Publication # 20153 (1980). Bratby, J.R., "Optimizing Coagulants and Flocculant Aids for Settling." Journal AWWA. 73 (1981). Brink, D.R., Choi, S., Al-Ani, M., and Hendricks, D.W., "Bench-Scale Evalua tion of Coagulants for Low Turbidity Waters." Journal AWWA. 80:199-204 (1988).

115

TABLE 15

Blank Data Sheet for Tabulation of Paper Filter Test Results When Varying Flocculant Dosage in Preceding Jar Test DATE _ OPERATOR ALUM/FERRIC DOSE

RAW VATER CHARACTERISTICS

LIKE DOSE

pH

pH adjusted to

TURBIDITY

POLrHER

TEMPERATURE ____

POUMEH CONCENTRATION _____

TRUE COLOR

PAPER FILTER USED:

Jar no.

_____ _____

____

WHATMAN NO.

1

2

Polymer doae (mg/1) Polymer volume for each Jar (ml)

Jar teat turbidity Jar teat color Paper filter turbidity Paper filter color Paper filtration time (sees) COMMENTS

116

3

1

5

.

6

D

m

I-H

1-1 Q

>-

2

Z)

a.i

a.a 0.3

POLYMER

POLYMER pOSE

P-5

(mg/l )

B.6

0.7

B.B

B.9

I. 0

FIGURE 17. Blank Graph for Plotting Paper Filter Test Results When Varying Flocculant Dosage in Preceding Jar Tests

B.B

O »—i hCE >_J

in

UJ

u

U)

Dentel, S.K., Resta, J.R., Shetty, P.V., and Bober, T.A., "Selecting Coagulant, Filtration, and Sludge Conditioning Aids." Journal AWWA. 79:72-84 (1987). Hudson, H.E. and Wagner, E.G., "Conduct and Use of Jar Tests."

AWWA Seminar

Proceedings-Upgrading Water Treatment Plants to Improve Water Quality.

AWWA

Publication # 20153 (1980). Introduction to Water Quality Analyses. Vol. 2.

American Water Works

Association Publication # 1931 (1982). Ives, K.J., "Deep Bed Filtration." ed.

(2nd edition).

Solid-Liquid Separation. L. Svarosky,

London: Butterworths (1981).

McCormick, R.F. and King, P.H., "Factors That Affect Use of Direct Filtration in Treating Surface Waters."

Journal AWWA. 74:234 (1982).

Standard Methods for the Examination of Water and Wastewater. (Publication # 10035), and WPCF.

Washington, D.C. (1985).

118

APHA, AWWA

G.

MODULE G.

ANALYTICAL METHODS FOR FILTER TESTS

This module describes two different means of filtrate quality assessment that can be performed in conjunction with the paper filter test (Module F). These two methods--turbidity measurement and particle size analysis--are outlined in Sections 1. and 2. respectively. 1. Turbidity Measurement a. General Discussion A turbidimeter is not as sensitive as a particle size analyzer to particles less than 5 microns in size; these are the first to break through the filter (Kavanaugh et al., 1978; Tunison, 1985).

Hence, it would have limited use

for accurately estimating filter run lengths.

However, in the filter test

only a fixed amount of sample is filtered; the test is not continued until break-through occurs.

Turbidity measurements correlate well with particle

counts for the filtrate from the paper filter test (Dentel et al., 1987) and thus turbidity measurement is a sufficiently accurate means of evaluating the effectiveness of a filtration aid.

Its low cost and simplicity of

operation also make it an attractive method. Detailed instructions on instrument operation can be found in the manual for the particular turbidimeter being used.

Additional information can be

obtained from publications listed at the end of this module under 3. Ref erences . as well as from Module D. b. Apparatus 1) Turbidimeter with sensitivity of 0.01 NTU or better. 2) Sample cells supplied with the instrument.

119

3) Cleaning tissue such as Kimwipes. 4) Lens paper, c. Reagents 1) Turbidity standards; if these are not supplied with the instrument, they can be made according to the instructions in Standard Methods. 2) Distilled water. d. Procedure 1) Turn on the turbidimeter and calibrate it using either the prepared or the supplied standard. 2) Gently swirl the filtered water, hold the sample cell at an angle, and carefully pour the filtrate into the cell up to the level specified in the turbidimeter manual. 3) Eliminate any air bubbles by gently tapping the sides of the cell. 4) Wipe the outside of the cell, first with cleaning tissue and then with lens paper. 5) Place the cell in the sample compartment of the turbidimeter and note the reading. 6) Before using the cell for the next sample, clean it thoroughly with dis tilled water and place it upside down to let the wash water drain out, and to keep internal surfaces free from dust. If the cell is to be reused immediately, rinse it with a small volume of the next sample to be used (after a gentle swirling) then return to step 2) above.

120

2. Particle Size Analysis a. General Discussion A particle size analyzer furnishes information regarding not only the number of particles in a given sample but also the categories these particles fit into with respect to size ranges.

It is more sensitive than a turbidimeter

to low particle concentrations commonly found in filter effluent (Tate and Trussel, 1978).

However, since a highly sensitive turbidimeter is suffi

ciently accurate for evaluating the clarity of filtrate from the paper filter test, the aquisition of this instrument solely for the purpose of evaluating a filtration aid is not justifiable due to its high cost. b. Apparatus 1) Particle size analyzer with appropriate sensor. 2) Sample bottles.

Any clean bottles that can fit into the sample compart

ment are acceptable. 3) Magnetic stir bars, 1 in. or less. 4) Stir bar retriever, c. Reagents 1) Demineralized water. 2) Distilled water.

Available in most supermarkets.

Also available in supermarkets,

d. Procedure Refer to the manual for the particular instrument being used for detailed instructions on operating the unit. information.

Refer also to Module E. for additional

A few general rules should always be followed:

121

1) Fill a sample bottle with demineralized water and place it in the sample compartment. When the unit is being restarted after an extended period during which it was not used, run the demineralized water through the instrument's sensor three or four times, since falsely high readings are obtained in the first few runs. 2) Gently swirl the contents of the container holding the filtered water, hold the sample bottle at an angle, and carefully pour the filtrate into the bottle. Drop a stir bar into the bottle and place it in the sample compartment. 3) Turn on the magnetic stirrer, run the sample through, and record the particle count readings. 4) After each run with a sample, perform one run with the demineralized water. In between samples, thoroughly clean the sample bottle with distilled water, use the demineralized water for a final rinse, and place the bottle upside down to let the wash water drain out and to keep internal surfaces free from dust. 3. References Beard, J.D. and Tanaka, T.S., "A Comparison of Particle Counting and Nephelometry." Journal AWWA. 69 (1977). Dentel, S.K., Bober, T.A., Shetty, P.V., and Resta, J.R. Procedures Manual for Selection of Coagulant. Filtration, and Sludge Conditioning Aids in Water Treatment: Supporting Documentation. AWWA Research Foundation Report (1987). Introduction to Water Quality Analyses. Publication # 1931 (1982).

122

American Water Works Association

Methods for Chemical Analysis of Water and Wastes.

U.S. EPA.

Technology

Standard Methods for the Examination of Water and Wastewater.

APHA, AWWA

Transfer Series.

EPA-625-/6-74-003 (1974).

(Publication # 10035), and WPCF.

Washington, D.C. (1985).

Tate, C.H. and Trussell, R.R., "The Use of Particle Counting in Developing Plant Design Criteria." Journal AWWA. 70, 698 (1978). Tunison, P.P., "Improving Water Quality Using Zeta Potential/Particle Counts." Paper presented at New York State Chapter AWWA Conference (1985).

123

H.

MODULE H.

SLUDGE VOLUME DETERMINATION

1. General Discussion This test is applicable to the determination of relative sludge volumes produced by differing doses of coagulants or flocculants. conjunction with a jar test.

It is used in

This information is required where the impact

of a coagulant or flocculant on subsequent sludge volume generation must be taken into consideration, for example, when sludge handling capacity is limited by the sludge storage volume or dewatering process capacity.

This

procedure may also be used to collect and concentrate sludge generated during jar tests for subsequent dewaterability testing using methods outlined in modules J or K. The resources (equipment, time analytical expertise, etc.) required for this test are minimal.

Test materials and apparatus, as outlined below, are

readily available and inexpensive.

Imhoff cones are suggested for volume

measurement as they are more accurate than beakers and can be modified to permit sludge removal for subsequent testing. 2. Apparatus 1) 3 to 6 1-L Imhoff cones (Figure 18) with a support rack or ring stands. Multiple Imhoff cones will allow the contents of all jars from a jar test (typically 6 jars) to undergo sludge volume determinations simultaneously. If the removal of the collected sludge for subsequent testing is desired, the Imhoff cones can be modified by replacing the drain plugs with sections of flexible tubing and squeeze clamps as shown in Figure 18.

One-L

graduated cylinders may be substituted for Imhoff cones if desired; however, these usually cannot be modified to permit sludge removal.

Do not compare

sludge volumes determined with different container types (i.e. Imhoff cones, graduated cylinders) as the container shape has an effect on the sludge compaction and final volume.

125

1-Liter Imhoff Cone

.Flexible Tubing

FIGURE 18. Imhoff Cone

126

2) Stopwatch 3) Stirring rod or laboratory spatula 4) Thermometer 3. Reagents Contents of jar test containers after performing jar tests (Module C). 4. Procedure 1) Complete Module C procedure for jar tests, preferably using 2-L con tainers . 2) If the jar test supernatant has not been removed for filtration testing, decant 1 L from the top of each jar, or from the sampling ports if the jars are equipped with ports. 3) Transfer the remaining contents of each jar test container to an Imhoff cone. 4) At 15 minute intervals, gently stir the cone contents for 10 to 15 seconds to dislodge the solids adhering to the cone sides.

Do this by

scraping the cone sides with a stirring rod or spatula. 5) After 60 minutes, record the volume of sludge in mL from the cone graduations.

See Table 16 for a suggested data sheet.

If a 2-L jar test

was used, multiply the sludge volume by 0.5 to give mL of sludge per 1 L of water.

If a 1 L jar test was used, do not multiply the sludge volume by 0.5

since this is already the volume of sludge per 1 L of water.

The resulting

sludge volume value can also be reported as gal sludge/1000 gal water treated (L sludge/1000 L water treated).

If the settled sludge contains

large pockets of water entrained in the sludge zone, estimate the volumes of these pockets and subtract from the total volume.

127

The practical

TABLE 16 Suggested Sludge Volume Test Data Recording Sheet Date: Analyst:. Coagulant:. Flocculant: Cone No. Coagulant Dose (mg/L) Flocculant Dose (mg/L) Sludge Volume (mL/L) Sludge Volume x 0.5 (gal/1000 gal)

Remarks:

128

lower limit of measurement is approximately ImL/L. 6) Plot sludge volume vs. coagulant or flocculant dose.

If both coagulant

and flocculant doses were varied, a plot similar to that shown in Figure 9 (Module C, Jar Test) can be constructed, but developing lines of equal final sludge volume rather than final turbidity.

Use log-log paper if con

structing such a plot. 5. Data Analysis From the sludge volume/coagulant dose plot, jar test results, and filterability results (if these procedures were performed), determine the coagulant or flocculant dose which provides acceptable settled water quality while minimizing the volume of sludge produced.

If graphs of turbidity vs.

dose and sludge volume vs. dose are prepared using the same scale for the dose axis, analysis of results will be simplified. 6. Results Interpretation Test results should be interpreted cautiously, as only relative sludge volumes are estimated by this method.

Correlations between the estimated

sludge volumes and actual sludge volumes can only be determined through operational experience. 7. Variables Variations in water temperature as well as agitation while transferring the jar test contents can affect results.

Therefore, ensure that all tests are

run under similar conditions. 8. Precision And Accuracy Triplicate analyses of 10 sample sets resulted in an average method precision of 2.3 mL/L.

Method precision refers to the standard deviation of

the results of a series of replicate samples.

129

Method accuracy, which refers

to the agreement between the value determined by the test method and the real value, could not be determined as there is no independent means of determining sludge volume.

If an optimum coagulant aid dose is to be chosen

based on the results of this method, it is recommended that replicate tests (3 or more) be conducted because of the relatively high inherent imprecision. 9. References Standard Methods for the Examination of Water and Wastewater. (Publication # 10035), and WPCF.

Washington, D.C. (1985).

130

APHA, AWWA

I.

MODULE I.

PREPARATION OF SLUDGE SAMPLES

1. General Discussion This method is applicable to the preparation of sludge samples for the subsequent evaluation of sludge conditioner (or flocculant) effectiveness. It is used in conjunction with the methods outlined in Modules J or K. The resources required for this test are minimal.

Test materials and

apparatus, as outlined below, are readily available and are located in most water treatment plants. 2. Apparatus 1) Jar Test stirrer (Phipps and Bird type with maximum speed of 130 RPM). Substitution of a different type of mixer (i.e. magnetic stirrer or high speed mixer) will require modification of recommended mixing durations due to differences in mixer velocity gradients. the recommended mixing times. mixing times.

For a magnetic stirrer, double

For a high speed mixer, halve the recommended

See the Werle, et al. reference for modification methods for

other stirrer types. 2) Large bore pipet.

This can be constructed by breaking off the tip of a

glass pipet after scoring with a file and then polishing the broken end with a flame to remove any sharp edges. 3. Reagents 1) Sludge samples 2) Miscellaneous sludge conditioners (See Module A for polymer preparation methods).

131

4. Procedure 1) In order to insure uniformity of sludge samples, obtain one large sample (5-10 gal for several tests) and gently stir it to insure a homogenous consistency throughout. Then immediately measure equal volumes (typically 1 L) of sludge into the jars (typically 2-L jars). Since subsequent tests are influenced by such variables as sludge solids concentrations and temperature differences, it is imperative that these differences be minimized by ensuring uniformity between samples. 2) Start mixing the jar contents at maximum speed (minimum 100 rpm). For magnetic stirrers or high speed mixers operate the mixer at the highest speed possible without splashing the sludge out of the containers. 3) Add the chosen conditioning polymer at various dosages to the jars. jar should be a control (no conditioner added).

One

4) Mixing duration is dependent upon the type of dewatering process to be used. For low-stress dewatering processes such as drying beds, the minimum recommended mixing time is 2 minutes (4 minutes for magnetic stirrers, 1 minute for high speed mixers). For medium-stress processes such as vacuum filters, the minimum recommended mixing time is 15 minutes (30 minutes for magnetic stirrers, 7 minutes for high speed mixers). For high-stress dewatering processes such as centrifuges or belt-filter presses, the minimum recommended mixing time is 1 hour (2 hours for magnetic stirrers, 30 minutes for high speed mixers). 5) Following rapid mix completion, reduce the mixer speed to approximately 30 rpm and allow the sludge to flocculate for 5 minutes. 6) Remove the sludge from the mixer and proceed with the dewaterability evaluations outlined in Module J or K. For a rapidly settling sludge, gently stir at slow speed (10-20 RPM) while awaiting testing.

132

7) Sludge conditioner doses are dependent upon the type of polymer used (i.e. dry, emulsion, liquid).

Any comparison between different products

will have to be based on the amount of the polymer and its cost as supplied by the manufacturer.

Consequently, all sludge conditioner doses should be

reported in terms of the form that the polymer is available (i.e. pounds of dry products, gallons of liquid or emulsion products). 5. Variables The variables in this procedure that affect the subsequent sample properties are mixing intensity and duration.

Low intensity, low duration mixing has

been shown to underpredict the optimum dose for high stress dewatering processes such as centrifuges.

Consequently, ensure that all samples

undergo the same preparation procedures prior to comparison.

Particular

attention should be paid to ensuring homogenity among samples (i.e. same solids concentration and temperature). 6. References Werle, C.P., Novak, J.T., Knocke, W.R., and Sherrard, J.H., "Mixing Intensity and Polymer Sludge Conditioning." Engineering. 110:5:919 (1984).

133

ASCE-Journal of Environmental

J.

MODULE J.

TIME TO FILTER TEST

1. General Discussion This test is applicable to the determination of the filterability or dewaterability of sludges.

It can be used:

- To evaluate various sludge conditioning polymers and dosages (large-volume test). - To evaluate the impact of coagulants and/or flocculants on sludge dewaterability in conjunction with a jar test and the sludge volume determination procedure (small-volume test). - To assist in the daily operation of sludge dewatering processes (large-volume test). The resources required for this test are moderate. apparatus, as outlined below, are widely available.

Test materials and Test duration and

required analytical expertise are minimal. The test consists of placing a sludge sample in a Buchner funnel with a paper support filter, applying a vacuum, and measuring the time required for a fixed volume of filtrate (usually 50% of the sample volume) to collect in the graduated cylinder.

While similar to the better known Specific

Resistance to Filtration test (SRF), the Time to Filter test (TTF) is superior for these purposes because of its ease and simplicity of use. 2. Apparatus i. Large volume TTF (see Figure 19 for apparatus setup) 1) Vacuum pump or other vacuum source.

135

9 cm diameter 3uchner Funnel Whatman Mo. 1 or 2 Pressure Cere Pinch C1-3T.3 Side Arm

^Flexible Tubing

V 250 mL Graduated Cylinder W/2U.-UQ neck.

FIGURE 19.

Large Volvune TTF Equipment

136

2) 5-9 cm diameter Buchner funnel. 3) 24-40 fritted glass side arm. 4) 250 mL graduated cylinder with a 24-40 fritted glass neck. 5) Filter paper (Whatman No. 1 or 2 or equivalent). 6) Assorted corks, vacuum tubing, vacuum flasks, and glass tubing.

The

number of each of these required depends on how many Buchner funnels are in the assembly. 7) Stopwatch. b. Small-Volume TTF (see Figure 20 for apparatus setup). 1) Vacuum pump or other vacuum source.

2) 2.5 cm diameter Buchner funnel. 3) 10 mL graduated cylinder equipped with side arm (Figure 20) or 10 mL graduated cylinder with a funnel adaptor. 4) Filter paper (Whatman No. 1 or 2 or equivalent). 5) Stopwatch. 3. Reagents 1) Sludge samples. 2) Miscellaneous sludge conditioning products.

If this method is to be used

to evaluate polymeric conditioners, see Module A for polymer preparation methods.

137

2.5 err.. Diameter Buchner Funnel _

°lnch Clamo Side Arm Adaptor'

Flexible Tubing,

10 mL Graduated Cylinder

FIGURE 20.

Small Volume TTF Equipment With Side Arm Adaptor

138

Pressure Gare

4. Procedure a. Large-Volume TTF 1) Assemble apparatus as shown in Figure 19. 2) Place paper support filter into funnel. 3) To seal filter, pre-wet the filter with a small volume of water.

Drain

any excess water. 4) Accurately measure 100-200 mL of conditioned sludge into a graduated cylinder or beaker. dures.

See Module I for polymer addition and mixing proce

The sample volume is dependent upon Buchner funnel diameter.

volume should be used that will almost fill funnel completely. 200 mL for a 9 cm diameter funnel.

A

This will be

Ensure that all tests are conducted at

the same initial sample volume. 5) Start vacuum pump. 6) Pour conditioned sludge sample into the funnel. 7) Start stopwatch or timer. 8) Record the time required for 50% of the sample volume to collect in the graduated cylinder.

A suggested data sheet is shown as Table 17.

9) If possible, repeat for a minimum of three determinations per conditioner dose. 10) Repeat for different sludge conditioning products or doses, b. Small-Volume TTF

139

TABLE 17 Suggested Time To Filter Test Data Recording Sheet

Date:

Analyst:.

Chemical Assessed:

Sample Volume:.

Filtrate Volume:

Chemical Dose (mg/L)

Time to Filter (sec) #1

Remarks:

140

#2

#3

Avg TTF

1) Assemble apparatus as shown in Figure 20. 2) Place paper support filter into funnel. 3) To seal filter, pre-wet the filter with a small volume of water.

Drain

any excess water. 4) Following sludge volume determination as outlined in Module H, remove the collected sludge from the bottom of the Imhoff Cones.

Minimize the dilution

of the sludge with the supernatant by decanting as much supernatant as possible from the Imhoff Cones prior to sludge withdrawal. 5) Accurately measure 7-10 mL of sludge into a graduated cylinder or beaker (7 mL is the minimum sample volume required by this method).

Ensure that

all tests are conducted at the same initial sample volume. 6) Start vacuum pump. 7) Pour sludge sample into the funnel. 8) Start stopwatch or timer. 9) Record the time required for 50% of the sample volume to collect in the graduated cylinder.

A suggested data sheet is shown as Table 17.

10) If possible, repeat for a minimum of three determinations per coagulant or flocculant dose. 11) Repeat for different coagulants, flocculants, or combinations, varying dosages. 5. Data Analysis From the data sheet, plot TTF vs. chemical dose for the products evaluated (See Figure 21).

Determine which chemical combination and dose give the

141

u.

o o T™

u 0) in

D

O

3OO -

4OO -

500 -

900

FIGURE 21

Polymer A

40

160

Polymer C

200

Sample Time to Filter (TTF100) Data Plot

Polymer Dose (mg/L or mL/L) + Polymer B o

12O

24O

optimum results.

Optimum conditioning occurs at the dose which results in

the minimum TTF for the least cost (i.e. lowest dose of chemical). 6. Results Interpretation Test results can be interpreted fairly confidently as the TTF test is a variation of the more familiar Specific Resistance to Filtration test which has been used to model most full-scale dewatering processes. 7. Variables Variations in the vacuum pressure, support filter type, sludge temperature and sample volume can affect test results. 1) Sludge suspended solids concentration has a significant effect on the test results.

When using the TTF procedure to evaluate sludge conditioning

products or assist in the operation of a dewatering process, this can be avoided by adhering to the sample preparation procedures outlined in Module I, particulary those ensuring homogeneity between samples.

Comparison of

TTF data between different original samples (especially if taken on different days), cannot be made with confidence unless suspended solids concentrations are comparable.

A rough correction for different solids

contents can be made by dividing the sludge's TTF value by its corresponding solids concentration. 2) When using the small-volume TTF test to evaluate the impact of coagulants or flocculants on sludge dewaterability, variations in sludge solids content are to be expected, since different chemical doses will have different effects on sludge settling and compaction in the Imhoff cones.

Because

these differences occur in full-scale applications as well, the TTF results should be interpreted as reflecting the overall impact of the different chemical doses, including the different solids concentrations of the settled sludge.

143

8. Precision and Accuracy Triplicate analyses of 18 sample sets of conditioned and unconditioned alum sludge resulted in an average method precision of 19 seconds (approximately 4% of the average value) for the large-volume TTF test.

Triplicate analyses

of 9 sample sets of conditioned and unconditioned alum sludge resulted in a method precision of 9 seconds (approximately 6% of the average value) for the small-volume TTF test.

Method precision refers to the standard

deviation of the results of a series of replicate samples.

Method accuracy,

which refers to the agreement between the value determined by the test method and the real value could not be determined as there is no independent means of determining Time to Filter. 9. References Knocke, W.R and Wakeland, D.L., "Fundamental Characteristics of Water Treatment Plant Sludges."

Journal AWWA. 113:10:516 (1983).

144

K.

MODULE K.

CAPILLARY SUCTION TIME TEST

1. General Discussion This test is applicable to the determination of the filterability or dewaterability of sludges.

It can be used:

- To evaluate various sludge conditioning aids and dosages. - To evaluate the impact of coagulant aids on sludge dewaterability in conjunction with a jar test and the sludge volume determination. - To assist in the daily operation of sludge dewatering processes. The resources required for this test are moderate.

Test materials and

apparatus, as outlined below (and shown in Figure 22), are available from at least two manufacturers, although one is located in Great Britain.

Test

duration and required analytical skill are quite minimal, offsetting the cost of the test apparatus. The test consists of placing a sludge sample in a small cylinder on a sheet of chromatography paper, which extracts the liquid from the sludge by capillary action.

The time required for the liquid to travel one centimeter

is recorded. 2. Apparatus 1) Capillary Suction Time Apparatus (available from Venture Innovations, P.O. Box 80277, Lafayette, LA; or Triton Electronics Ltd., Bigods Hall, Dunmow, Essex, England, CM63BE). 2) CST paper available from CST suppliers above, or use Whatman No. 17, chromatography grade paper cut into 7 cm x 9 cm sections, with the grain

145

SLUDGE RESERVOIR

^ 7

SLUDGE —————————————— BLOCK HOLDING PROBES —|

n—*-

FILTER PAPER —J-U— PA«5P —————————-^

—ft ————d|

DIGITAL TIMER

(PROFILE)

TO TIMER

REFERENCE MARKS ON BLOCK

(PLAN)

FIGURE 22. England)

Capillary Suction Time Apparatus (Triton Electronics Ltd.,

146

parallel to the 9 cm side.

The Whatman No. 17 paper is available through

Fisher, Thomas, or other scientific supply firms. 3) Large bore pipet. 3. Reagents 1) Sludge samples. 2) Miscellaneous sludge conditioning products.

If this method is to be used

to evaluate polymeric conditioners, see Module A for polymer preparation methods. 4. Procedure 1) Turn CST meter on. 2) Reset meter. 3) Place the CST paper into the test cell with the rougher side up and the grain parallel to the 9 cm length side. 4) If the test is to be used to evaluate sludge conditioners, refer to Module I for polymer addition and mixing procedures. 5) If the test is to be used to evaluate the effect of coagulants and/or flocculants on sludge dewaterability following sludge volume determination as outlined in Module H, remove the collected sludge from the bottom of the Imhoff Cones.

Minimize the dilution of the sludge with the supernatant by

decanting as much supernatant from the Imhoff Cones as possible prior to sludge withdrawal. 6) If the test is to be used to assist in the operation of a sludge dewatering process by optimizing the sludge conditioner dose, refer to the polymer preparation procedures outlined in Module A.

147

However, since the

dosage range should be known from past operation, restrict the evaluation to smaller intervals of doses (e.g. 30, 35, 40, 45, 50 mg/L instead of 0, 10, 25, 50, 100 mg/L). 7) Remove 5-7 mL of sludge from the sample jar and pour into the test cell reservoir.

This can be done easily with a large bore pipet of the type used

to add the conditioning chemical.

Record CST shown on digital display.

A

suggested data recording sheet is shown as Table 18, and an example set of data is presented in Table 19. 8) Repeat for a minimum of three determinations per sample. 9) Repeat for different sludge conditioners or doses. 5. Data Analysis From the data sheet, plot CST vs. conditioning polymer or flocculant dose for the aids evaluated.

An example plot is shown in Figure 23.

Determine

which coagulant, flocculant, or sludge conditioner and dose combination give the optimum results.

Optimum conditioning occurs at the dose which results

in the minimum CST for the least cost. 6. Results Interpretation If Module I has been used to prepare the sludges used, test results can be interpreted fairly confidently.

The CST test has been used successfully to

predict the performance of most sludge dewatering processes, and corre lations between CST and full-scale dewatered solids content can be developed for each individual dewatering process.

If Module I is not employed (such

as when using small volumes of sludge from the Imhoff cone procedure) then CST results may not provide adequate predictions for high stress dewatering processes as discussed in Module I.

148

TABLE 18 Suggested Capillary Suction Time Test Data Recording Sheet Date:

Analyst:.

Chemical Assessed:

Chemical Dose (mg/L)

Capillary Suction Time (sec) #1

#2

Remarks:

149

#3

Avg CST

TABLE 19 Example Capillary Suction Time Test Data Recording Sheet Date:

Analys t:

Chemical Assessed:

Chemical Dose (mg/L)

Capillary Suction Time (sec) #1

o 10

#2

#3

Avg CST

73-1 li.o

7/.$-

lo. C

"7f.3

62. So .^

Sb . { .3

'LOO

So.-if

"9J.O

Remarks:

150

21-7

u

O

0) M

Q

FIGURE 23

Polymer A

16O

Polymer C

200

Sample Capillary Suction Time Data Plot

Polymer Dose (mg/L or mL/L) + Polymer B o

12O

2 9) through the addition of NaOH.

See Table 1 for

summarized descriptions of polymeric charge pH constraints. In either instance it is recommended that the positive reagent employed for the charge density determinations be DDPM, since it retains a constant cationic charge density at any solution pH. In order to observe a distinct color change at the titration endpoint (and thereby achieve a reproducible anionic charge density value), it is necessary in this case to perform a "back-titration" analysis.

The

procedural modifications for such an analysis are as follows. It is desirable to perform this titration at several known solution pHs, so the experimental set-up employs a 150 mL beaker and a pH detection system consisting of a pH probe (or probes) and a calibrated pH meter.

At the

outset of the titration, the beaker contains: - pH probe(s) - 1.00 mL of weak stock (0.05%) anionic polymer solution - 10.00 mL of DDPM positive standard - 2 drops of TBO indicator - a magnetic stirring bar - sufficient deionized, distilled water to raise the solution level up to that of the pH probes - and NaOH or HC1 as required for initial pH adjustment.

174

Since, as the titration proceeds, it will be necessary to continually monitor and adjust the solution pH, the pH probe(s) should remain in the beaker over the entire course of the titration. The contents of the beaker are titrated with the standardized PVSK solution until the purple endpoint color is observed.

The average amount of PVSK

required to titrate 10.00 mL of DDPM is known from the previous standard ization data; a smaller titrant volume than this amount will be required to cause the desired color change because some of the positive DDPM has been consumed by the anionic polymer.

A trend should be observed such that

decreasing amounts of titrant are required for equivalence point conditions to be met as pH increases.

At low pH, (approximately pH 5), a maximum

titrant volume is approached, which may be equal to or slightly less than the average titrant volume required to titrate the DDPM alone.

The

normality of the unknown anionic polymer is calculated as illustrated in Table 25.

Table 24 is provided for recording experimental results, and may

be photocopied for laboratory use. Notes:

Unlike the quaternized positive charge on many cationic polymers,

any negative charge observed will be influenced by solution pH.

Anionicity

and solution pH are directly related, and maximum anionicity occurs at high solution pH. If nonionic polymers are being sought, adjust solutions to near neutral or even acidic pHs.

Since only very small amounts of titrant will be required

for equivalence point conditions to be reached, the use of a more dilute titrant solution may be beneficial in terms of avoiding titration errors. When ordering polymer samples, be sure to request Material Safety Data Sheets and any other available information.

Use caution according to

manufacturers' instructions, and do not exceed any maximum dosage limitations set by the National Sanitation Foundation (Refer to Section II. C of this Manual).

175

f. Charge Density Calculation and Use Tables 22 and 24 supply equations for the normality calculation of cationic and anionic polymers, respectively.

From these normalities and from quant

itative information concerning polymer solution strength, it is possible to obtain the charge density values of interest.

Tables 23 and 25 provide

examples of the manner in which Tables 22 and 24 are to be used. Example Charge Density Determination In Table 23 a cationic polymer normality of 0.002250 N is observed.

From

this value and other values recorded on Table 23 we may obtain a polymeric charge density value for Catfloc C, in units of g/eq.

(One equivalent is

the same as one mole of positive or negative charge.)

In order to perform

this calculation, the following steps should be completed. 1) Obtain the weight of polymer (in grams) present in 1000 mL or 1 L of the titrated polymer sample.

Use the polymer density and volume as follows;

(polymer density, g/mL) x (volume neat polymer, mL)/(volume solution, mL) = (weight neat polymer, g)/(200 mL solution) 1.0624 g/mL X (0.1883 mL/200 mL solution) = 0.2000 g Catfloc C/200 mL soln.

(weight neat polymer, g)/(200 mL solution) x (proportionality factor) = (weight neat polymer, g)/(l L solution) 0.2000 g/200 mL soln. X 5/5 = 1.0000 g Catfloc C/1000 mL solution 2) Divide this weight by the polymer normality to obtain the desired charge density result.

(weight neat polymer, g/L polymer solution)/(polymer normality, eq/L) = charge density, g/eq (1.0000 g/L)/(0.002250 eq/L) = 444.4 g/eq

176

Date Analyst

Data Sheet #___ TABLE 24

Data Sheet for Back-Titration of Anionic ___ Polyelectrolyte With Standardized PVSK Titrant Density of polymer _______ (g/mL) (NA* if solid) Primary polymer make-up _______ (mL or g neat/L water) - stock Secondary polymer make-up _______ mL stock/_______ mL water - working Volume of _____________ polymer being titrated _______(mL) (stock or working) Normality of PVSK titrant ________ (eq/L) Normality of DDPM solution ________ (eq/L) • see data sheet #_ (1) Average volume of PVSK required to titrate 10 mL DDPM _______ (mL) Duration of titration _______ (min) Run # 1 2 3 4 5 6 7 8 9 10 11 12 (2)

Initial buret reading (mL) __________ ________ __________ __________ __________ __________ __________ __________ __________ __________ __________ __________

Final buret reading (mL) __________ ________ __________ __________ __________ __________ __________ •______ __________ __________ __________ __________

Net Volume PVSK delivered (mL) __________ __________ __________ __________ __________ ________ __________ __________ __________ __________ __________ __________

Average Volume of PVSK required to titrate polymer and 10 mL DDPM - _____________(mL)

(1) - (2) - ________ mL - (3) - Amount of PVSK carrying negative charge equivalent to polymer's. Polymer Normality Calculation: (_______mL (3)) (______N PVSK)

-

N polymer

. „ - _______ N polymer

polymer) (_______mL * J ——————— NOTES:

*NA - not applicable 177

Data Sheet

Date Analyst

6-19-88 deBarbadillo

TABLE 25

Sample Data Sheet for Back-Titration of Anionic Percol LT26 Polyelectrolyte With Standardized PVSK Titrant Density of polymer NA __ (g/mL) (NA* if solid) 0.6285 Primary polymer make-up (mt-er g neat/L water) - stock Secondary polymer make-up NA mL stock/ NA mL water - working Volume of ____stock_____ polymer being titrated 0.50 (mL) (stock or working) Normality of PVSK titrant 0.000549 (eq/L) Normality of DDPM solution 0.000206 (eq/L) see data sheet # 14 (1) Average volume of PVSK required to titrate 10 mL DDPM 3.75 (mL) Duration of titration 5.0 (min) Run

pH

2 3 i 4 5 67 '

9 7

5

8 9 10 11 12 (2)

Initial buret reading (mL) 1.55 4.60 7.82 10.85 13.90 16.98 20.20

11.40 15.12

Final buret reading (mL) 4.60 7.70 10.85 13.90 16.98 20.07 23.98 15.12 18.89

Average Volume of PVSK required to titrate polymer and 10 mL DDPM -

(1) - (2)

0.68

Net Volume PVSK delivered (mL) 3.05 3.10 3.03 3.05 3.08 3.09 3.78 3.72 3.77

3.07 ± 0.03

(mL)

mL - (3) - Amount of PVSK carrying negative charge equivalent to polymer's.

Polymer Normality Calculation: (

0.68 ( NOTES:

mL (3)) ( 0.000549 N PVSK) 0.50 mL polymer)

-

N polymer

-

0.000747

N polymer

(2) for pH 7 - 3.07. (2) for pH 9 - 3.07. (2) for pH 5 - 3.76:

i.e.. polymer is essentially nonionic at pH 5._________________________ *NA - not applicable 178

This value may now be used for cost comparison and rough dosage calcula tions.

Using the polymer cost per pound and a factor for converting pounds

to grams (454 g per pound), a 'cost per equivalent' value may be obtained for the comparison of prospective products.

Once jar test experience yields

knowledge of the approximate number of equivalents of positive charge necessary for optimum coagulation and flocculation, dosage ranges for initial testing may be more easily selected. The advanced analyst may wish to determine how many charges are present per monomer unit in the polymer chain.

For such a calculation, two more pieces

of information are required; the monomer molecular weight, and the product's percent purity.

This information is calculated using Table 2 and supplied

by the manufacturer, respectively.

Continuing with the example above, these

values are 161.676 g/mole monomer PDADMAC, and 40% pure.

The calculation is

as follows; [(weight neat polymer, g) x (percent purity/100)/(total diluted volume, L)]/ (polymer normality, eq/L) = polymer charge density, g/eq [(1 g)(40/100)/(l L)]/(0.002250 eq/L)

=

177.8 g/eq

(molecular weight monomer, g/mole monomer)/(polymer charge density, g/eq) = polymer charge density, eq/mole monomer (161.676 g/mole monomer)/(177.8 g/eq)

=

0.9 eq/mole monomer

Keeping in mind the approximate nature of the purity value, as well as analyst error, this polymer is considered to display one positive charge per monomer unit. polymer.

This is as expected for a fully quaternized PDADMAC type

A value less than one equivalent per mole monomer would indicate

that not all monomer units possess charge.

A number greater than one may

indicate that some or all of the monomer units possess more than one charge. This quantity is most useful if the analyst can predict the outcome of such calculations and thereby check the accuracy of his or her work.

179

4. References Cox, C., et al., "Particle Aggregation in Water Treatment with Polymers," Master's Thesis, Univ. of Missouri-Colombia, 1984. Introduction to Water Quality Analyses. Vol. 4.. AWWA Publication #1931 (1982). Kawamura, S., et al., "Application of Colloid Titration Technique to Flocculation Control," Journal AWWA. Aug. 1967, p. 1003-1013. i Kawamura, S., et al., "Applying Colloid Titration Techniques to Coagulant Dosage Control," Water and Sewage Works. 113:348 (1966). Reference Handbook: Basic Science Concepts and Application. AWWA Publication #1940 (1980). Simplified Procedures for Water Examination Including Supplement on Instrumental Methods. Manual 12. AWWA Publication #30012 (1978). Standard Methods for the Examination of Water and Wastewater. APHA, AWWA (Publication # 10035), and WPCF. Washington, B.C. (1985). Wang, L.K., et al., "Application and Determination of Organic Polymers," Water. Air and Soil Pollution. 9:337 (1978). Yeh, H. H., et al., "Selection and Use of Polymers in Direct Filtration," Doctoral Dissertation, Univ. of Missouri-Colombia, 1980.

180

N.

MODULE N.

1.

Materials

a.

Apparatus

MOLECULAR WEIGHT DETERMINATION FOR ORGANIC POLYELECTROLYTES

1)

Module A equipment for polymer make-up

2)

Viscometer (either type #1, Kinematic or Ostwald-Fenske, Fisher catalog #13-616A; or type #2, Ubbelohde, Fisher catalog #13-614A)

3)

Timer, accurate to ±0.05 seconds

4)

Ring stand support

5)

2 Three-pronged clamps, rubber tipped (1 small--Fisher catalog #05-738, 1 large--Fisher catalog #05-740)

6)

2 perpendicular clamp holders (Fisher catalog #05-755)

7)

Plumb line and weight

8)

Rubber pipette bulb

9)

Glass pipette, 3 or 5 mL volume

10)

Millipore filtration apparatus, or equivalent

11)

Whatman 41 filter paper, or equivalent

12)

Removable plug or other device for sealing center glass tube when necessary (for type #2 viscometer only)

13)

Constant temperature water bath (optional, but improves accuracy)

14)

Thermometer

b.

Reagents

1)

Neat (undiluted) polymer products

2)

Sodium thiosulfate; Na2S04«5H20

3)

Sodium nitrate; NaN03

5)

Deionized, distilled, filtered water

6)

Acetone; C3HgO

7)

Hydrochloric acid solution, 1 N; HC1

181

2.

Procedure

a.

Apparatus Preparation

The glass viscometer must be thoroughly cleaned if good results are to be obtained. Initially, wash the apparatus with hot soapy tap water. Then rinse with large amounts of hot water by drawing liquid up into the capil lary tube with a rubber pipette bulb. (Never force air or water down the capillary tube with the rubber bulb, since small fragments of rubber from • the bulb's interior can dislodge and then become firmly planted in the capillary passage.) Once the viscometer is free of soap, rinse the apparatus several times with 1 N HC1. Each time the viscometer is rinsed, draw the dilute acid solution up into the capillary passage with the rubber bulb, and then allow it to flow back down; repeat two or three times per rinse.

Be very careful not to

allow any of the HC1 solution to enter the rubber pipette bulb.

Flush the

instrument with copious amounts of distilled, deionized, filtered water to remove all traces of the acid, and those impurities which it dissolved. Draw the rinse water up the passage and allow it to fall several times per rinse. At this time no smudges or other evidence of dirt should be visible on the inside or outside of the viscometer.

If such dirt is present,

particularly if in the capillary passage, perform the chromic acid rinse procedure described in the Appendix to this module. This initial rinse with dilute HC1 need not be repeated if the instrument is properly cleaned and stored between uses. Finally, do two rinses with plenty of acetone, followed by several rinses with deionized, distilled, filtered water. During each rinse, be sure to clean the capillary passage by drawing fluid up into it and then allowing the fluid to fall several times. These last steps (acetone and water rinses), preceded by the hot soapy water wash, will be the general method of This cleaning procedure applies if a new or very clean viscometer is to be used. Previously used viscometers may require the more extensive "Chromic Acid Rinse" cleaning described in the Appendix to this module.

182

cleaning between polymer types.

(Be sure to remove all traces of polymer

with the hot soapy water before introducing acetone into the viscometer, since otherwise the acetone will cause polymer to precipitate onto the inner glass surfaces of the viscometer.

The same type of reaction takes place

between acetone and soap, so be sure to rinse out all soap as well.) Note:

For overnight storage or between uses, fill the viscometer completely

with acetone or deionized, distilled, filtered water.

Inverting the visco

meter over a sink or beaker while it is still supported by the ring stand can be a convenient way to empty the bulb reservoirs while still maintaining the necessary vertical alignment of the apparatus.

Just be sure to allow

time for the capillary passage to become empty. b.

Setting up the Apparatus

Use a vertical ring stand with a weighted base to support the glass viscometer on the two three-pronged clamps.

For a type #2 viscometer the

capillary passage must be in exact vertical alignment.

For a type #1 visco

meter, the double-bulb portion with the fiducial marks must be in exact vertical alignment.

When this condition is met, both topmost portions of

the parallel glass tubing will also be in vertical alignment.

These

conditions can be achieved through the use of a reference plumb line for comparative purposes.

Since the apparatus can become skewed during usage,

it is best to affix a permanent plumb line which hangs behind the viscometer within the user's field of vision.

In this manner the technician can check

for proper vertical orientation of the appropriate portions of the specific viscometer several times and from several angles during the course of a viscosity test.

See Figure 25 for illustrations of these setups.

A thermometer should be located near the viscometer. used, the thermometer should be in the water.

If a water bath is

This will allow the user to

easily make temperature readings during the viscosity test.

Also close by

should be a timer accurate to the tenth of a second, for timing the duration of fluid flow between the two fiducial marks of the viscometer.

183

Type #1

Type #2

Kinematic or Ostwald-Fenske

FIGURE 25.

Ubbelohde

Proper Vertical Orientation of Viscometers.

184

c.

Preparing the Polymer and Solvent Solutions

Because ionic strength (or salt concentration) can have a significant effect on results, it should be kept constant for all experiments. that results can be compared between experiments .

This insures

Therefore , all polymer

solutions are to be uniformly prepared using the high ionic strength solvent described in this section.

This is the only difference from the polymer

solution preparation in Module A; once these solvent solutions have been prepared, the technician should refer to the appropriate section of Module A for further preparation instructions.

Again, the only difference in

preparation methods is that the pure polymer products are to be diluted only with the following high ionic strength solvents, not distilled, deionized, filtered water. Two distinct types of high ionic strength solvent solutions are used; one for the preparation of cationic polymers, and another for the preparation of either anionic or nonionic polymers.

(If the type of charge is not known

for a polymer, use the "two-drop TBO test" from Module M.) The solvent solution for the preparation of cationic polyelectrolyte solutions consists of: 5 g sodium thiosulfate 42.5 g sodium nitrate (NaNC>3) sufficient deionized, distilled, filtered water to volumetrically dilute the above salts to 1L. The solvent mixture used to prepare solutions of anionic or nonionic polymers is made from: 5 g sodium thiosulfate ( 85.0 g sodium nitrate (NaN03) sufficient deionized, distilled, filtered water to volumetrically dilute these solids to 1L.

185

After dilution, each type of solvent solution should mix on a magnetic stir plate for one hour.

Upon completion of mixing, filter each solution to

remove any undissolved solids using Whatman 41 filter paper and a Millipore filtration system (or equivalents).

These liquids are then used to dilute

the neat polymer products to the five following concentrations:

5 g/L (or

0.5%), 3 g/L (0.3%), 2 g/L (0.2%), 1 g/L (0.1%), and 0.5 g/L (0.05%). Notes:

Exercise extreme care when preparing these solutions because even

slight dilution errors at the outset of this type of experiment lead to very large errors in the final results.

Therefore, follow Module A carefully,

and be sure to update old polymer density values.

Once polymer samples have

been unsealed, density values may change due to hygroscopic or evaporative interactions with the atmosphere.

For this reason do not use a polymer

density value which is over two weeks old. d.

Determining Background Flow-through Time (to )

Before viscosity measurements can be obtained for the polymer-solvent systems, it is first necessary to determine the viscosities of the two types of solvents in their pure states at known temperatures. flow-through times will be used as to in Section 3

These "background"

(Data Reduction).

When measuring the viscosity of any solution with the glass viscometer, there are several procedural steps which must always be followed: 1)

Measure and record the temperature before and after each viscosity run.

2)

Rinse the viscometer by drawing the solution to be tested up the capillary tube and then allowing it to fall three times.

Discard this

aliquot of liquid and completely empty the viscometer before refilling with the same liquid for measurement.

(Always introduce fluid into the

opening of the widest glass tube.) 3)

If a type #1 viscometer is being used, an exact and constant volume of fluid must be used for each measurement. convenient volume for constant use.

Either 3 mL or 5 mL is a

If a type #2 viscometer is used,

the test volume does not need to be known or constant.

186

4)

The viscometer should never be left soaking in anything other than distilled, deionized, filtered water; acetone; or less preferably, solvent.

For overnight or more extended storage, leave the apparatus

soaking in distilled, deionized, filtered water or acetone only. Given these general guidelines, specific directions may now be detailed. Using the appropriate solvent (appropriate for the type of polymer to be tested), proceed as follows: 1)

Rinse the apparatus thoroughly and fill as described in (2) and (3) on the previous page.

2)

If a constant temperature water bath is available, set it to 30°C ±0.2 and allow it to equilibrate, then immerse the viscometer for about 15 minutes for the sample temperature to stabilize.

If no water bath is

used, record the initial ambient temperature. 3)

Using the rubber pipette bulb, draw the solvent solution into the upper most viscometer bulb until the fluid surface is well above the higher fiducial mark.

Release the pipette bulb vacuum and remove the rubber

bulb from the viscometer.

(For a type #2 viscometer, the narrowest

glass tube must be plugged while the fluid is being drawn into the uppermost bulb; no vacuum will be achieved otherwise.

Remove this plug

and the pipette bulb prior to timing the fluid flow.) 4)

As the upper meniscus of the liquid passes the higher fiducial mark, activate the timer.

5)

As the lower fiducial mark of the viscometer is passed by the meniscus of the liquid surface, deactivate the stopwatch.

Record this time

period, accurate to ± 0.05 seconds, on Table 26. 6)

Repeat this measurement on the same aliquot of fluid until the standard deviation of the average of 5 or 10 trials is within ±0.5 seconds.

7)

The entire procedure 1) to 6) is repeated for the remaining solvent.

e.

Determining Flow-through Time for a Polymer-Solvent Solution

The procedure to be followed is much as described above, but steps (1) through (6) are carried out on each of the five polymer-solvent solutions

187

Data Sheet #

Date Analyst TABLE 26

Data Sheet for Determining the Average Pure Solvent Flow Time, to

1.

Solvent make-up: weight Na2S04 • 5^0 ___________ g/L weight NaNC>3 __________________ g/L Charge type: _______ cationic or _______ anionic/nonionic Ambient temperature (°C): ________________ before, ____________ after, ________________ average

2.

Flow times for solvent only (seconds).

Standard deviation (SD) of ten

trials should be < ± 0.5 seconds. 1.

5.

2.

6.

3.

7.

8.

10.

4.

3.

Average solvent flow time and standard deviation. t0

=

±

seconds

188

which were specified in Section c.

Again, be very careful in preparing

these solutions since any errors made now will be magnified greatly in the final results. A good way to avoid errors while working very efficiently is to perform the viscosity measurements on the five serial dilutions working from most dilute to most concentrated.

The solution of least polymer concentration is

measured first, followed by increasingly concentrated solutions of the same product.

In this way, the viscometer need only be analytically cleaned (see

soap-acetone-water rinses as described in Section 2a. Apparatus Preparation) before the most dilute polymer solution is run; careful rinsing with the next, more concentrated solutions between uses will then suffice. judgment with this policy.

(Use

If a long length of time--3/4 to 1 hour--will

elapse between runs of more concentrated polymer solution, then empty, clean, and fill the viscometer with either distilled, deionized, filtered water; acetone; or the appropriate solvent solution between uses.) Another important source of error is temperature fluctuation.

The tempera

ture at which the polymer-solvent solution is run must be the same as that at which the pure solvent "background" flow-through time (t0 ) was measured to within ± 0.5°C.

If these values are different, obtain a pure solvent

measurement at the appropriate temperature for use in data treatment, or repeat the polymer-solvent measurements at the proper temperature.

Use of a

water bath helps to alleviate temperature variations. Times of flow for each of the five dilutions of the neat polymer can now be obtained and recorded onto Table 27.

The average time for each polymer

concentration may then be calculated. Be sure to perform a final cleaning of the viscometer when all readings have been completed.

If it is to be reused in the near future, fill the visco

meter with either distilled, deionized, filtered water; acetone; or the appropriate solvent solution.

If it is to be stored empty, clean thoroughly

so that no residues will be left on the inner glass surfaces.

189

Date

Data Sheet #

Analyst TABLE 27 Data Sheet for Reduced Viscosity Determination 1.

, supplier

Polymer name:

Expiration Date

Lot/batch # cationic,

Charge type:

anionic, before,

Ambient temperature (°C): 2.

type

nonionic after,

Flow times for polymer/solvent solutions (seconds). (SD) of each ten trials should be < Polymer

average

Standard deviation

±0.5 seconds.

Time (seconds)

Average and

Concentration, g/L

SD of Times (sec.)

1.

2.

3.

4.

6.

7.

8.

9.

10.

1.

2.

3.

4.

5.

6.

7.

9.

10.

C2 =

190

TABLE 27 (Continued)

t3 1.

2.

3.

4.

5.

6.

7.

8.

9.

10.

2.

3.

4.

7.

8.

9.

1.

___

6.

10.

±

3.

1.

2.

3.

4.

5.

6.

7.

8.

9.

10.

Calculation of specific viscosity (SV) and reduced viscosity (RV) Specific Viscosity Formula

Reduced Viscosity

Result

Formula

SV1 = ti/t0 - I = _____

RV1 = SV1/C1

SV2 = t 2 /t0 - 1 = _____

RV2 = SV2/C2

SV3 = t 3 /t0 - 1 = _____

RV3 = SV3/C3

SV4 = t4/t0 - 1 = _____

RV4 = SV4/C4

SV5 = t5/t0 - 1 = _____

RV5 = SV5/C5

(See data sheet #_____ for to value.)

191

Result

TABLE 27 (Continued) 4.

Plot the reduced viscosity values (RV1 through RV5) against the corresponding concentration values (Cl through C5) on Figure 26.

5.

Draw a straight line on the graph passing as closely as possible to all points.

Extend this line to the left until it intersects the vertical

axis, and read off the reduced viscosity value at this intersection. This is the INTERCEPT in L/g.

INTERCEPT

=

____________

L/g.

(Alternatively, find the equation of the best fit line through the points using linear regression on a calculator or computer. Solve this equation for the y intercept at x = 0. 6.

This value is the INTERCEPT.)

Determine the percent solids of the product, P s . may be assumed equal to 100.

For dry products, P s

For liquid or emulsion polymers, the value

may be given in the product data sheet.

Otherwise, the Standard

Methods procedure for Total Residue may be used; the percent solids is

"c

s

7.

____Total Residue (mg/L) [product density (g/cnr)] x

The intrinsic viscosity (IV) is then IV

=

INTERCEPT x 100 / P s

=

___________ L/g

*For PDADMAC and EpiDMA polymers, P s may be estimated from the EPA maximum dose for use in potable water treatment, using the equation Ps

=

1000 / (maximum dose in mg/L).

192

Date __

Data Sheet #

Analyst

Ul O U U)

LJ u ID a ui a: POLYMER CONCENTRRTION (g/L)

Polymer name Lot/batch //_ Charge type:

_, Supplier Expiration date cationic,

anionic,

nonionic

For original data, see data sheet # Intercept = _ __________ L/g FIGURE 26.

Graph Used for Determining Intrinsic Viscosity (IV).

193

3.

Data Reduction

Having obtained to (flow time for the appropriate pure solvent) and t^ through t5 (flow times for the increasingly concentrated polymer-solvent The

solutions), the user can arrive at the polymer's intrinsic viscosity.

specific viscosity, SV, must first be calculated for each concentration of polymer.

Then the reduced viscosity, RV, can also be calculated for each.

Calculate and record these values using Table 27.

Now the RV values may be

graphed against the corresponding polymer concentrations using Figure 26. (Directions for making this plot are found at the end of Table 27.)

The

intersection of a "best fit" line drawn through these points and the verticle (or y) axis gives the intrinsic viscosity, IV, in L/g.

Table 28

and Figure 27 provide examples of how this is done. 4.

Results Interpretation

The molecular weight of a polymer (M, in g/mole) is directly related to its intrinsic viscosity (IV, in L/g) by the formula IV = kMa where k and a are constants for a given type of polymer.

Therefore, if the

constants k and a are known, the molecular weight M can be calculated: I/a These values for aqueous polymer solutions are difficult to establish for most polymer types.

This is due to difficulties in measuring molecular

weight by other means (in order to calibrate the above equation), and also to effects of other polymer characteristics which are often only known to the polymer manufacturer.

Fortunately, the above equations indicate that a

*i^

Values have been determined for polyacrylamide and for polyacrylic acid (a = 0.66 and k = 3.7 x 1(T 5 L/g) and Epi/DMA and PDADMAC (a - 0.66 and k= 1.35 10- 5 ) (Molyneux, 1983).

194

d-

Data Sheet #

Date Analyst TABLE 28

Sample Data Sheet for Reduced Viscosity Determination 1.

Polymer name:

supplier

Lot/batch #

, type

Expiration Date

Charge type:

\/

cationic,

Ambient temperature (°C):

anionic ,

'Z 3 •*? before,

nonionic after,

fcq-H- j>i

2.

(O [ ^

Flow times for polymer/solvent solutions (seconds). (SD) of each ten trials should be
s. vrr- 3*

10. t4

3.Q

5.

2.

10.

?. y^y. ay-

t5

2.

5.

?. 3.

Calculation of specific viscosity (SV) and reduced viscosity (RV) Specific Viscosity

Formula

Reduced Viscosity Formula

Result

Result

SV1 =

1

RV1

SV1/C1

SV2 = t 2 /t0 SV3 = t3/t0

1

RV2

SV2/C2

1

RV3

SV3/C3

SV4 = t4/t0

1

RV4

SV4/C4

SV5 = t 5 /t0

1

RV5

SV5/C5

(See data sheet # /4

for to value.) 'S- sec. 196

TABLE 28 (Continued) 4.

Plot the reduced viscosity values (RV1 through RV5) against the corresponding concentration values (Cl through C5) on Figure 26.

5.

Draw a straight line on the graph passing as closely as possible to all points. Extend this line to the left until it intersects the vertical axis, and read off the reduced viscosity value at this intersection. This is the INTERCEPT in L/g. INTERCEPT

=

L/g.

C) • OOfe

(Alternatively, find the equation of the best fit line through the points using linear regression ffn fl f^lculatpr or computer. Solve this equation for the y intercept at x = 0. This value is the INTERCEPT.) 6.

Determine the percent solids of the product, Ps . For dry products, Ps may be assumed equal to 100. For liquid or emulsion polymers, the value Otherwise, the Standard may be given in the product data sheet. Methods procedure for Total Residue may be used; the percent solids is JL.

•L C

's

____Total Residue (mg/L) [product density (g/cnr)] x

PS 7.

The intrinsic viscosity (IV) is then IV

=

INTERCEPT x 100 / P s

=

O -0/5"

L/g

*For PDADMAC and EpiDMA polymers, P s may be estimated from the EPA maximum dose for use in potable water treatment, using the equation Ps

=

1000 / (maximum dose in mg/L) .

197

_,

Date __

Data Sheet #

Analyst

>

e>.6i

in o u tn LJ u 13 a u o:

o.oo? 6.001

POLYMER CONCENTRRTION

Polymer name

, Supplier /W>?ct>n

Lot/batch //

Expiration date

Charge type:

cationic,

anionic,

For original data, see data sheet # Intercept = 0, O®T&

Figure 27.

L/g

Sample Graph Used for Determining Intrinsic Viscosity (IV),

198

Cg/L)

nonionic

rough idea of the relative molecular weights of polymers can be obtained using comparisons of intrinsic viscosities.

Other uses for intrinsic

viscosity values are: Comparing different batches of the same polymer type for quality control.

If the values are significantly different, potential product

substitution or deterioration may be an explanation.

(For such

comparisons, only specific viscosities need be computed as long as the same polymer concentration is used). - Comparing two or more different products for polymer type elucidation. If two products provided by different suppliers are believed, or claimed to be the same, a comparison of intrinsic viscosities and charge densities could confirm or deny such suspicions.

Also, potential

dilution of the same initial liquid product is also indicated since the charge densities and reduced viscosities will be altered by the same dilution factor, while the intrinsic viscosity will remain unchanged. - Monitoring the shelf life of a particular batch of polymer in terms of degradation.

Increases or decreases in intrinsic viscosity over time

(if greater than possible experimental error) indicate either further polymerization or product breakdown, respectively.

Usage experience

will dictate the point at which the product must be discarded.

Also, a

more rational application of polymers is generally made available in terms of selecting potential products to enhance specific water treatment processes. 5.

References

Billmeyer, F.W., Textbook of Polymer Science. 2nd ed., John Wiley and Sons, Inc., NY, 1971. Bowen, P.T., "Sludge Conditioning:

Effects of Polymer and Sludge

Properties," Doctoral dissertation, Clemson University, 1982.

199

Molyneux, P., Water-Soluble Synthetic Polymers:

Properties and Behavior.

Vol.1, CRC Press, Boca Raton, FL, 1983. Shoemaker, D.P., et al., Experiments in Physical Chemistry. 4th ed., McGrawHill, Inc., NY, 1981. 6. Appendix to Module N: Chromic Acid Rinse This procedure is to be used for cleaning the internal passages of a viscometer if residue or blockage is visible in the instrument, particularly within the capillary passage.

It should not be required for cleaning a new

viscometer or one that has been properly cleaned during previous use.

Read

the entire procedure before beginning. 1. Materials a. Apparatus 1)

Glass-stoppered bottle

2)

Glass stirring rod

b. Reagents 1)

Concentrated sulfuric acid;

2)

Sodium dichromate;

2. Procedure a. Reagent Preparation To prepare the chromic acid cleaning solution, dissolve approximately 10 grams Na2Cr20y in the least amount of hot water necessary.

After the

solution cools, add approximately 200 mL concentrated ^804 slowly while stirring.

Store in a glass-stoppered bottle labelled "CHROMIC ACID.

Caution: Strong Acid and Oxidant".

Perform this reagent preparation

procedure under a fume hood with caution.

200

b. Reagent Use Wash the viscometer with hot soapy water, rinse well, and drain. Then introduce approximately 10 mL of the chromic acid cleaning solution into the viscometer using a glass pipette. Rinse all inner surfaces of the visco meter thoroughly by drawing the cleaning solution up into the uppermost bulb-type reservoir using a rubber pipette bulb, and then allowing the fluid to flow back down; repeat three times per aliquot of cleaning solution. Use two or three fresh aliquots of cleaning solution as needed per cleaning. Then flush the viscometer with distilled, deionized, filtered water for a prolonged period of time to remove all traces of the acid, as well as those impurities which it either dissolved or dislodged. At this time the visco meter should appear very clean.

If smudges or other visible dirt are still present, especially if in the capillary passage, repeat the chromic acid rinse procedure, allowing the acid to remain in contact with the areas of concern for up to 30 minutes. Once the viscometer is clean, drain and rinse thoroughly as described above. Finally, before use can begin, do two rinses with plenty of acetone, followed by several rinses with distilled, deionized, filtered water. During each rinse, be sure to clean the capillary passage well by drawing the liquid up into it and then allowing the fluid to flow back down several times. Drain the instrument well just prior to use. If use will not begin for a period of time (one hour to several days), fill the viscometer with distilled, deionized, filtered water or acetone to protect it for storage. c. Precautions

When using the chromic acid cleaning solution extreme caution should be exercised. The use of safety goggles, a laboratory coat, and a fume hood is strongly recommended. This dangerous liquid will dissolve calibrations and other markings off of the outer surfaces of glassware. Rubber is also soluble in this solution,

201

as are most compounds.

Consequently, this liquid is very dangerous to the

analyst, and skin contact should be strongly avoided.

(If such contact

occurs, rinse the affected area with copious quantities of cold water.) Also, do not allow the cleaning solution to come in contact with the inner surfaces of the rubber pipette bulb as this exposure will not only contam inate the cleaning solution, but also damage the pipette bulb. A period of 15 to 30 minutes contact with the inner glass surfaces of the viscometer should be sufficient for the chromic acid cleaning solution to perform its function.

Rinse the apparatus with plenty of distilled,

deionized, filtered water after cleaning. If the cleaning solution turns a greenish color after prolonged use or storage, replace the solution with a fresh batch.

202

IV.

USING MODULE RESULTS FOR CHEMICAL AID SELECTION

A. RESULTS EVALUATION

1. Baseline Data The most important step in evaluating test results is the collection of accurate baseline operating data prior to chemical aid selection.

While

the current coagulant dose, filter run duration, backwash frequency, sludge generation rates or operating costs may appear to be readily available, most treatment plants are probably deficient in this baseline information. Common causes of these deficiencies include poorly calibrated flow meters and chemical metering pumps, inaccurate and inadequate operational analy ses, changing chemical consistency, insufficient operation and maintenance records, inadequate operational cost information, and too much reliance on "operator experience" in place of detailed historical records and analyses. Without first obtaining this information, it is improbable that a chemical aid could be selected rationally. 2. Simple Approach The simplest approach to evaluating test results is to concentrate on a single treatment process without regard to any other processes that may be affected.

An example would be using jar tests to determine the optimum

coagulant aid without evaluating its impact on filtration or sludge generation rates.

Choosing the optimum chemical aid in this case is fairly

easy since the optimum aid/dose is the one that gives the best settled water quality at the lowest cost. are outlined below in Section IV.B. obvious, given its simplicity.

Methods for determining chemical costs Advantages of this approach are fairly

Disadvantages can include possible

detrimental impacts on subsequent treatment processes, such as decreased filter run duration or increased sludge volume. generally vary from plant to plant.

These disadvantages will

This approach is most suitable for

small to medium size water treatment plants where the costs of a

203

comprehensive process evaluation could not be justified by the savings yielded from such an evaluation.

Where detailed accurate baseline

information is unavailable, this approach is the only method that should be used. 3. Comprehensive Approach A more complicated approach to evaluating test results is to evaluate the effect of a chemical aid on the entire treatment process.

An example of

this approach would be flocculant evaluation using jar tests, paper filtration tests, sludge volume tests, and time to filter tests to evaluate the subsequent impact of the polymer on coagulation, filtration, sludge volume, and sludge characteristics respectively.

The advantages of this

comprehensive approach are obvious, given the sometimes uncertain effects of chemical additives on subsequent processes like filtration or sludge management.

The disadvantages are also obvious given the large number of

tests that need to be conducted, as well as the required existence of detailed, accurate baseline data.

As a result, this approach is most

likely restricted to use in larger water treatment plants where even small reductions in coagulant doses can lead to significant savings. B. COST ESTIMATING

Estimating the costs when choosing a chemical aid are critical, because performance is rarely the sole criteria in the selection process.

A number

of factors must be considered in ultimately determining whether using a chemical aid to improve treatment will be cost effective.

These include

the cost of the chemical aid, any savings due to decreases in other process chemicals or reduced operational costs, and other miscellaneous costs.

The

following procedure outlines a method for estimating the chemical costs associated with using a chemical aid such as an organic polymer.

It is

assumed that the required physical facilities (mixing tanks, pumps, piping, etc.) are already in place.

The procedure does not include any estimates

for increases or decreases in labor or energy costs.

For estimating those

costs and the construction costs of the required facilities, consult the

204

Environmental Protection Agency's Estimating Water Treatment Costs. EPA 600/2-79-162c, August 1979. 1. The First Step In estimating the costs of using a chemical aid, the first step is to determine the current chemical costs prior to selection of a chemical aid. This is outlined in Section 1 of Table 29. obtained from the purchase orders. treatment plant records.

Chemical unit costs can be

Chemical doses should be known from the

A minimum of 1 year's records should be used so

that seasonal variations in chemical usage will be considered. 2. The Second Step Next, determine the costs of the chemical aids selected from the lab tests It is very important that lab test

(jar tests, filtration tests, etc). doses be calculated accurately.

All chemical doses should be expressed in

terms of product as supplied (prior to any dilution) per volume of water treated.

For a powdered polymer, the dose in mg/L should then be converted

to pounds of polymer per MG (millon gallons) of water treated. factors are included in Table 29.

Conversion

For a liquid or emulsion polymer the

dose in mL/L should be converted to gallons of polymer as supplied per MG of water treated.

For sludge conditioners, doses should be reported as

pounds or gallons of product supplied per ton of dry solids conditioned. This requires that the solids concentration of the sludge be known.

If

sludge solids concentration is assumed to be fairly consistent, then these doses can be reported as pounds or gallons of product per 1000 gallons of sludge conditioned or MG of water treated.

An example of these

calculations is shown in Table 30. 3. The Third Step Now determine the savings from any reduction in chemical requirements allowed by the use of a chemical aid.

For example, if a flocculant dose

of 1 mg/L allows the alum or ferric chloride dose to be reduced by 5 mg/L

205

and the lime dose by 10 mg/L, then these cost savings should be included as well. 4. The Fourth Step Finally, determine any savings in operation and maintenance costs attri butable to use of the chemical aid.

For example, if the use of a floccu-

lant reduces the sludge volume and hence disposal costs by 20 percent, or a filtration aid increases the filter run duration by 24 hours, the savings associated with reduced sludge disposal costs or backwash costs should be included.

These final savings can sometimes be the most significant

benefit of using a chemical aid.

Without detailed baseline data, however,

it is difficult to quantify any realized savings.

Nonetheless, estimates

of these costs may at least permit relative cost-effectiveness to be computed when several different chemical additives are being considered. These estimates also require that the appropriate Modules were used to generate the needed values. 5. References Estimating Water Treatment Costs.

EPA 600/2-79-162c, August 1979.

206

TABLE 29. Chemical Cost Estimation Worksheet (Photocopy and revise as necessary: at least one copy per chemical aid evaluated) 1. Current Chemical Costs Average Treated Water Flow Rate (MGD) : (MGD = millon gallons per day) Chemical Unit Cost

Average Chemical Usage Alum (Ib/day):

_____

Alum ($/lb) :

Ferric Chloride (gal/day) :

_____

Ferric Chloride ($/gal) :

Lime (Ib/day) :

_____

Lime ($/lb) :

Polymer (Ib/day or gal/day): _____

Polymer ($/lb or $/gal) :

Other* (Ib/day or gal/day): _____

Other* ($/lb or $/gal) :

Average Chemical Cost (Chemical usage x unit cost/treated water flow rate) Alum ($/MG):

_______

Ferric Chloride ($/MG):

_______

Lime ($/MG):

_______

Polymer ($/MG):

_______

Other ($/MG):

_______

TOTAL CURRENT CHEMICAL COSTS:

_____

* Include only chemicals with dosage changes under new conditions.

207

2 . Projected Chemical Costs: (Requires results from Modules C,F,J,K, and/or L) Conversion Factors: mg/L x 8.34 -OO

Ferric Chloride ($/MG) Lime ($/MG): Polymer ($/MG): Other ( $ /MG): TOTAL CURRENT CHEMICAL COSTS:

—— ————— ? '•

>

\

* Include only chemicals with dosage changes under new conditions.

212

2. Projected Chemical Costs: (Requires results from Modules C,F,J,K, and/or L) Conversion Factors: mg/L x 8.34 = Ib/MG mL/L x 1000 - gal/MG a.

Projected Chemical Aid Doses:

Alum Dose (mg/L): Ferric Chloride Dose (mL/L) : Lime Dose (mg/L) : 1C ^ X

="

(Ib/MG) :

*~~ ~"

(gal/MG):

lQ

(Ib/MG):

e;

Polymer Dose (mg/L oK-miL/fr) : Other* (mg/L or mL/L): b.

1.O

w • a r* w

~ /



O'^

(Ib/MG or gal/MG) :

~"

(Ib/MG or gal/MG):

..

. fc

" M T~

Projected Chemical Costs Ib/MG or gal/MG x unit cost):

Alum ($/MG):



^IL

Ferric Chloride ($/MG): Lime ($/MG): ff3.f Polymer ($/MG> :*/* ^ **£?

' ^Z • $"&

^^^



(from Module H results) (mL/L): Projected Sludge Volume to Current Sludge Volume Ratio (PSV/CSV) : Projected Unit Cost of Sludge Management and Disposal (Unit Cost x PSV/CSV) ($/MG):

^.^ XO-£» *»

PROJECTED PROCESS SAVINGS (Current Unit Costs Projected Unit Costs) ($/MG):

214

•£ ° Z.

,,,.. _,,,.,.,, w b. Filtration Processes Unit Cost of Filtered Water ($/MG): ^worr/Zed

Mru=( r

~:

Average Filtration Rate for a given filter (MGD) : Current Filter Run Duration for a given filter (hours) : ___/ 8 i» (T a

Average Filtered Volume for a given filter (MG) : (Filtration rate x filter run duration/24)

/. OOP

Backwash Flow Rate (gpm) : Backwash Duration for a given filter (min) :

&'' ' /

Current Backwash Volume for a given filter (MG) : (Backwash flow rate (gpm) x backwash duration (min)/l,000,000) Backwash Volume to Filtered Volume Ratio:

^s.

Current Backwash Cost for a given filter ($/MG): (BV/FV x unit cost of treated water) Projected Filter Run Duration (hours): (from Module F results) Current Filter Run Duration to Projected Filter Run Duration Ratio (CFR/PFR): Projected Backwash Cost ($/MG) :

S'/S'.S'** O.040 ^

(Current Backwash Cost x CFR/PFR) PROJECTED PROCESS SAVINGS ($/MG):

^/$". 3 1 - *?• /?

(Current Backwash Cost - Projected Backwash Cost)

215

"^

6.

c. Sludge Dewatering and Disposal Processes (Requires results from Modules L and J or K) Average Daily Treated Water Flow Rate (MGD) : JtooQ

Average Annual Treated Water Flow Rate (MG) : (Average Daily Flow Rate x 365)

"37

Total Annual Cost of Sludge Dewatering and Disposal: Unit Cost of Sludge Dewatering and Disposal ($/MG) :

?* e

(Total Cost/Average Annual Treated Water Flow Rate) ____ (

Current Sludge Depth (from Module L results) (in) : Projected Sludge Depth (from Module L results) (in) :

^ '3

Projected Sludge Depth to Current Sludge Depth 2 '*/t{

Ratio (PSD/CSD):

O«fe3

~

Projected Unit Cost of Sludge Dewatering and Disposal ~ (Unit Cost x PSD/CSD) ($/MG): ^B.f'^ * PROJECTED PROCESS SAVINGS (Current Unit Cost ^S.?^ — £"• ^fr Projected Unit Cost) ($/MG) :

TOTAL PROJECTED PROCESS SAVINGS ($/MG): (Coagulation + Filtration + Sludge Savings) TOTAL PROJECTED CHEMICAL SAVINGS ($/MG): TOTAL PROJECTED SAVINGS ($/MG): (Projected Process Savings + Projected Chemical Savings)

216

~~

3.

American Water Works Association

6666 WQuincy Avenue, Denver, CO 80235 (303) 794-7711

1P-1M-90553-10/89-SE

ISBN 0-89867-481-6

Suggest Documents