United States
Environmental Protection Office of Water EPA 81 1-B-92-002
Agency (WH-550) September 1992
xvEPA LEAD AND COPPER RULE
GUIDANCE MANUAL
VOLUME II: Corrosion Control Treatment
:_ Pnnled on Recycled" '••
-------
LEAD AND COPPER RULE
GUIDANCE MANUAL
Volume II: Corrosion Control Treatment
for
Drinking Water Technology Branch
Drinking Water Standards Division
Office of Ground Water and Drinking Water
U.S. Environmental Protection Agency
Washington, D.C.
Contract No. 68-CO-0062
by
BLACK & VEATCH MALCOLM PIRNIE, INC.
8400 Ward Parkway 1 International Blvd.
Kansas City, Missouri Mahwah, New Jersey
September 1992
-------
Preface
On June 7, 1991, the U.S. Environ-
mental Protection Agency promulgated
NPDWRs for lead and copper. EPA is
developing a guidance manual in two
volumes to assist water systems and State
regulatory agencies in implementing the
technical requirements of the rule. The
first volume of the Lead and Copper Rule
Guidance Manual addressed the monitor-
ing requirements of the rule. The second
volume of the Lead and Copper Rule
Guidance Manual concentrates on corro-
sion control treatment and lead service
line replacement.
This volume focuses on the evaluation
of corrosion control treatment options and
optimization of the full-scale treatment.
The manual discusses the procedures that
can be used by water systems to determine
the appropriate corrosion control
treatment. The manual discusses the
available testing protocols for conducting
the demonstration studies that many large
systems will be required to perform prior
to making their treatment
recommendation to the State. For smaller
systems, the manual contains a summary
of case studies separated by the raw water
quality to assist these systems in making
their treatment recommendation to the
State. The manual also provides guidance
to assist State regulatory agencies in
reviewing data from corrosion control
studies and in specifying optimal water
quality parameters. An additional chapter
provides guidance on the lead service line
replacement requirements. The subject
matter discussed in this chapter includes
what constitutes a replacement of a lead
service line, replacement schedules, and
the criteria for discontinuing lead service
line replacements.
I hope that this volume of the manual
will be a practical tool for water systems
and State regulatory agencies in
implementing the corrosion control
treatment and lead service line
replacement requirements of the lead and
copper rule.
James R. Elder
-------
Table of Contents
Page No.
1.0 INTRODUCTION 1-1
1.1 Reference 1-2
2.0 REGULATORY REQUIREMENTS FOR CORROSION
CONTROL STUDIES 2-1
2.1 Large PWSs 2-1
2.1.1 Regulatory Requirements 2-1
2.1.1.1 Scope of testing activities 2-2
2.1.1.2 Source water treatment 2-2
2.1.2 State Actions and Decisions 2-5
2.2 Small and Medium-Size PWSs 2-6
2.2.1 Regulatory Requirements 2-6
2.2.2 State Actions and Decisions 2-11
2.2.2.1 Review of recommended treatment 2-11
2.2.2.2 Requirement for additional study 2-12
2.2.2.3 Designating alternative treatment 2-12
2.2.2.4 Notification requirements . . .- 2-13
2.3 References 2-13
3.0 SCREENING OF CORROSION CONTROL ALTERNATIVES ........ 3-1
3.1 Principles of Corrosion and Corrosion Control 3-1
3.2 Corrosion Control Treatment Alternatives 3-3
3.2.1 Alkalinity and pH Adjustment 3-6
3.2.2 Calcium Adjustment 3-10
3.2.3 Corrosion Inhibitors 3-13
3.2.3.1 Phosphate inhibitors 3-13
3.2.3.2 Silicate inhibitors 3-17
3.3 Evaluating Alternative Corrosion Control Approaches 3-18
3.3.1 Steps to Corrosion Control Assessments 3-19
3.3.2 Documenting Historical Evidence • 3-22
3.3.2.1 Water quality data 3-22
3.3.2.2 Corrosion activity 3-24
3.3.2.3 Review of the literature 3-25
3.3.2.4 Prior experience and studies 3-26
3.3.3 Identifying Constraints 3-27
3.3.4 Evaluating Source Water Contributions 3-33
3.3.5 Preparing Recommendations for Optimal Treatment 3-35
3.4 Case Studies 3-36
3.4.1 Softening Ground Water Supply (Single Source) 3-36
3.4.2 Low Alkalinity, pH, and Hardness Surface Water System 3-41
3.4.3 Multiple Sources of Supply 3-46
3.4.4 Consecutive Systems 3-48
3.5 References 3-50
ii
-------
4.0 CORROSION CONTROL DEMONSTRATION TESTING 4-1
4.1 Corrosion Study Organization 4-1
4.2 Overview of Demonstration Testing 4-3
4.3 Testing Protocols 4-4
4.3.1 Flow-Through Testing Protocols 4-6
4.3.1.1 General 4-6
4.3.1.2 Testing program elements 4-9
4.3.1.2.1 Pipe rig operation and fabrication 4-9
4.3.1.2.2 Test monitoring program 4-12
4.3.2 Static Testing Protocols . . 4-12
4.4 Alternative Measurement Techniques 4-14
4.4.1 Weight-Loss Measurement Techniques 4-14
4.4.1.1 Coupons 4-14
4.4.1.2 Pipe inserts 4-21
4.4.1.3 Calculation of corrosion rates 4-21
4.4.2 Corrosion Plates 4-22
4.4.3 Surface Inspection 4-22
4.5 Data Handling and Analysis 4-24
4.6 Testing of Secondary Impacts 4-25
4.7 Quality Assurance/Quality Control Programs 4-25
4.8 Selecting the Recommended Treatment Option 4-26
4.8.1 Example of Treatment Selection 4-26
4.9 Examples of Corrosion Studies . 4-27
4.9.1 Flow-Through Testing 4-27
4.9.2 Static Testing 4-33
4.10 References 4-43
5.0 FULL-SCALE OPERATION AND IMPLEMENTATION OF OPTIMAL
CORROSION CONTROL TREATMENT 5-1
5.1 Overview of Requirements 5-1
5.1.1 Installing Optimal Treatment 5-1
5.1.2 Schedule 5-2
5.2 Full-Scale Operation of Treatment Alternatives 5-4
5.2.1 Start-Up Operations 5-4
5.2.2 Operating Ranges 5-5
5.2.2.1 Historic operating ranges 5-5
5.2.2.2 Recommended operating ranges 5-7
5.2.3 Diagnostic Sampling 5-13
5.2.4 Operational Notes on Various Treatments 5-14
5.2.4.1 Calcium carbonate precipitation 5-14
5.2.4.2 Carbonate passivation 5-16
5.2.4.3 Inhibitors 5-16
5.2.5 Reliability 5-18
5.2.6 Instrumentation and Control 5-18
5.2.7 Troubleshooting 5-19
III
-------
5'.3 Optimization Techniques 5-20
5.3.1 Diagnosing the Need for Optimization 5-20
5.3.2 Methods for Evaluating Treatment 5-23
5.3.2.1 Water quality parameters 5-23
5.3.2.2 Lead and copper data 5-23
5.3.2.3 Coupons and pipe inserts 5-24
5.3.2.4 Corrosion indices 5-25
5.3.2.5 Corrosion monitors 5-25
5.3.2.5.1 Hydrogen probes 5-25
5.3.2.5.2 Electrical resistance 5-25
5.3.2.5.3 Linear polarization resistance 5-26
5.3.2.5.4 Electrochemical noise 5-26
5.3.2.5.5 Application suggestions 5-26
5.4 Optimizing Corrosion Control Treatment—Examples 5-26
5.4.1 Optimal Corrosion Control in a Consecutive System 5-27
5.4.2 Use of Corrosion Monitoring in a Large System 5-27
5.4.3 Use of Extra Monitoring 5-30
5.5 References 5-30
6.0 LEAD SERVICE LINE REPLACEMENT 6-1
6.1 Overview of LSL Replacement Requirements 6-1
6.2 LSL Control and Related Requirements 6-2
6.2.1 LSL Control Determination 6-2
6.2.2 Rebuttal of Control Presumption . . 6-2
6.2.3 Partial LSL Replacement 6-4
6.3 Materials Evaluation 6-5
6.4 LSL Replacement and Schedule Requirements 6-5
6.4.1 Rebuttal of Lead Contribution Requirements . . . . •. 6-5
6.4.2 Replacement/Elimination Rates 6-6
6.4.3 Size-Dependent LSLRP Schedules 6-7
6.4.4 LSLRP Discontinuation 6-7
6.5 Reporting Requirements v 6-7
6.6 Record-Keep ing Requirements 6-11
6.7 References 6-11
APPENDICES
Appendix A — Corrosion Indices for the Precipitation of Protective Coatings A-l
Appendix B — Summary of Corrosion Control Studies Available in the Literature . . B-l
Appendix C — Statistical Evaluation of Corrosion Performance Data C-l
IV
-------
List of Tables
Page No.
Table 1-1: Topical Locator by Subject Matter for Lead and Copper
Rule Guidance Manual 1-3
Table 2-la. Recommended Corrosion Control Study Components for
Large PWSs. Based on Lead Levels 2-3
Table 2-lb. Recommended Corrosion Control Study Components for
Large PWSs. Based on Copper Levels 2-4
Table 2-2. Timeline for Small PWSs to Comply with the Corrosion
Control and Source Water Treatment Requirements 2-7
Table 2-3. Timeline for Medium PWSs to Comply with the Corrosion
Control and Source Water Treatment Requirements 2-9
Table 2-4. Dates for State Notification 2-13
Table 3-1. Conceptual Framework for Corrosion Control Approaches 3-5
Table 3-2. Summary of Chemicals Typically Used in pH/Alkalinity
and Calcium Adjustment Corrosion Control Treatment 3-9
Table 3-3a. Constraints Worksheet for pH/Alkalinity or Calcium
Adjustment Treatment Alternatives 3-28
Table 3-3b. Constraints Worksheet for Inhibitor Treatment Alternatives 3-30
Table 3-4. Schedule of Drinking Water Regulatory Activity: 1990-2000 3-31
Table 3-5. Source Water Treatment Guidelines for Systems Exceeding an AL . . 3-34
Table 3-6. Checklist for Desk-Top Corrosion Control Evaluation 3-37
Table 3-7. Checklist for the Kashton County Water District (KCWD)
Desk-Top Evaluation 3-42
Table 4-1. Organization of the Major Components in Corrosion Control Studies . . 4-2
Table 4-2. Pipe Volume by Tubing Length and Diameter 4-11
Table 4-3. Desnsities for a Variety of Metals and Alloys 4-17
Table 4-4. Chemical Cleaning Procedures for Removal of Corrosion Products . . . 4-18
Table 4-5. Electrolytic Cleaning Procedures for Removal of Corrosion Products . 4-20
Table 4-6. Corrosion Control Treatment Performance Ranking Matrix 4-30
Table 4-7. Final Corrosion Control Treatment Selection Matrix 4-31
Table 4-8. Lead Concentrations from Pipe Loop Testing 4-32
Table 4-9. Calculated Student's t Values 4-33
Table 4-10. Average Raw, Treated, and Finished Water Quality for
the Static Demonstration Tests by the City of Starboard 4-34
-------
List of Tables (continued)
Page No.
Table 4-11. Average Chemical Feed Rates and Water Quality Characteristics
by Treatment Alternative for the Static Demonstration Tests
by the City of Starboard 4-34
Table 4-12. Testing Program Raw Data for Water Quality Parameters
and Metal Leaching Measurements for the Demonstration
Tests by the City of Starboard 4-38
Table 5-1. Key Compliance Dates for Large, Medium and Small Systems 5-3
Table 5-2. Operating Ranges for pH and Alkalinity for 10 Water
Treatment Plants 5-6
Table 5-3. Operating Guidelines for Final pH to Meet a CCPP Level
of 12 mg/L 5-13
Table 5-4. Relational Behavior of Changing Water Quality Conditions
for Corrosion Control Treatment and Other Water
Quality/Treatment Objectives 5-22
Table 6-1. LSLRP General Accounting Worksheet 6-8
Table 6-2. LSLRP General Accounting Worksheet (Example) 6-9
Table 6-3. Reporting Requirements Schedule 6-11
vi
-------
List of Figures
Page No.
Figure 3-1. Forms of Corrosion Activity Encountered in Potable
Water Distribution Systems 3-3
Figure 3-2. Contour Diagram of Lead (II) Solubility in the System
Lead (II)-Water-Carbonate at 25°C and an Ionic Strength
of 0.005 mol/L 3-7
Figure 3-3. Contour Diagram of Copper (II) Solubility in the
System Copper (II)-Water-Carbonate at 25°C and an
Ionic Strength of 0.005 mol/L 3-8
Figure 3-4. Solubility Diagram for Calcium Carbonate in a Closed
System at 25°C 3-12
Figure 3-5. Contour Diagram of Lead (II) Solubility in the Presence
of 0.5 mg/L PO4 at 25°C and an Ionic Strength
of 0.005 mol/L . 3-14
Figure 3-6. Logic Diagram for Evaluating Alternative Corrosion
Control Treatment Approaches 3-20
Figure 3-7. Suggested Corrosion Control Approaches Based on Water
Quality Characteristics 3-23
Figure 3-8. Lime Softening PWS: Treatment Schematic and Relevant Data .... 3-40
Figure 3-9. Surface Water PWS with Low Alkalinity, pH, and
Hardness: Treatment Schematic and Relevant Data 3-45
Figure 3-10. PWS with Multiple Sources of Supply 3-47
Figure 3-11. Configuration of Consecutive Systems .3-49
Figure 4-1. Logic Diagram for Corrosion Control Demonstration Testing . . 4-5
Figure 4-2. Conceptual Lay-out of Flow-Through Testing Schemes 4-7
Figure 4-3. Typical Coupon Testing Installation 4-15
Figure 4-4. Typical Pipe Coupon Insert Installation 4-16
Figure 4-5. Cross-Section of Polarization Flow Cell 4-23
Figure 4-6A. Reductions in Metal Concentrations by Treatment Alternatives .... 4-28
Figure 4-6B. Reduction in Coupon Weight-Loss by Treatment Alternatives 4-29
Figure 4-7. Immersion Testing Set-Up 4-36
Figure 4-8. Testing Program for the City of Starboard Static
Demonstration Tests 4-37
vii
-------
TERM
DESCRIPTION
Non-Parametric Statistics
Passivation
pH/Alkalinity Adjustment
Phosphate Inhibitor
Pipe Insert
Pipe Loop
Pipe Rig
Precipitation
Sample Plan
Sample Pool Category
Small Water System
Silicate Inhibitor
Source Water Sample
Source Water Treatment
Static Testing
Weight-Loss Measurement
Statistical measures of relative behavior between two
or more sets of data not predicated on the data being
normally distributed.
A corrosion control technique which incorporates the
pipe materials into metal/hydroxide/carbonate
compounds intended to protect the pipe.
The addition of chemicals to modify the pH and/or
alkalinity to produce a less corrosive water.
A phosphate based chemical intended to reduce corrosion
when added to water.
Pipe sections used to evaluate the rate of corrosion by
insertion into piping systems.
An experimental apparatus consisting of several feet
of pipe complete with joints, elbows, and connections
for flow through testing.
The overall apparatus used for flow through testing
which may consist of several individual pipe loops.
The shifting of chemical equilibria to cause the formation
of a solid protective coating, usually calcium carbonate,
on interior pipe surfaces.
A description of the sampling locations and criteria for
targeted sample sites for first-draw tap, distribution
system, and point of entry samples.
The sample pool category of a PWS reflects the relative
priority of targeted sample sites able to be identified
and included in the sample plan for first-draw tap
samples.
A water system that serves 3,300 persons or fewer.
A silicate based chemical intended to reduce corrosion
when added to water.
Samples collected at the entry point(s) to the distribution
system representative of each source of supply after
treatment.
Removal of lead and/or copper from the source of supply.
An experimental approach that retains the testing
surfaces within standing water.
An approved method of determining the amount of metal
lost to corrosion from a pipe insert or coupon.
-------
ACRONYM
DEFINITION
AC
AL
ASTM
AWWA
AWWARF
BAT
CCPP
CT
CTact
CT^
Cu
Cu-POE
DBPs
DBPR
CWS
GAC
GWDR
HPC
LCR
LSL
LSLRP
NTNCWSs
NSF
Pb
Pb/Cu-POE
Pb/Cu-TAP
Asbestos-Cement.
Action Level - the level of lead or copper in first-draw tap samples which
when exceeded triggers additional compliance actions on the part of PWSs.
The American Society for Testing and Materials.
The American Water Works Association.
The American Water Works Association Research Foundation.
Best Available Technology.
Calcium Carbonate Precipitation Potential.
The product of disinfectant concentration (C) in mg/L and the effective contact
time (T) in minutes.
Actual CT value achieved across a single disinfection segment.
Required CT value for a specific level of Giardia or virus inactivation as
a function of temperature, pH, and in the case of free chlorine, disinfectant
residual.
Copper
Copper concentration at Point of Entry.
Disinfection By-Products
Disinfection By-Products Rule
Community Water System
Granular Activated Carbon
Ground Water Disinfection Rule
Heterotrophic plate count.
Lead and Copper Rule.
Lead Service Line.
Lead Service Line Replacement Program.
Non-Transient, Non-Community Water Systems.
National Sanitation Foundation.
Lead
Lead and copper samples collected at the points of entry to the distribution
system representative of each source of supply after treatment.
Lead and copper samples collected as first-draw tap samples from targeted
sample sites.
xi
-------
ACRONYM
DEFINITION
POE
PQL
PWS
QA/QC
SDWA
SWTR
SOCs/IOCs
SDSTTHM
TCR
THAAs
THM
TTHMs
WQP
WQP-POE
WQP-DIS
WTP
90%Cu-Tap
90%Pb-Tap
[(90%Pb-Tap)
-(Pb-POE)]
Points of Entry to the distribution system representative of each source of
supply after treatment. Used to describe source water monitoring activity.
Practical Quantitation Level
Public Water System
Quality Assurance and Quality Control measures to ensure reliable data
are collected.
Safe Drinking Water Act of 1974 as amended in 1986.
Surface Water Treatment Rule.
Synthetic Organic Chemicals/Inorganic Chemicals - Classes of chemical
compounds.
Simulated Distribution System Total Trihalomethanes.
Total Coliform Rule.
Total HaloAcetic Acids.
Triha lometha ne.
Total Trihalomethanes.
Water Quality Parameters, defined in the Rule to include pH, temperature,
conductivity, alkalinity, calcium, orthophosphate, and silica.
Water Quality Parameters measured at the Points Of Entry to the distribution
system representative of each source of supply after treatment.
Water Quality Parameters measured at representative locations throughout
the Distribution system.
Water Treatment Plant.
The 90% copper level for first-draw tap samples collected at targeted sample
sites.
The 90% lead level for first-draw tap samples collected at targeted sample
sites.
The difference between the 90% lead level for first-draw tap
samples collected at targeted sample sites and the highest respective lead
level measured at the points of entry to the distribution system.
xii
-------
List of Figures (continued)
Page No.
Figure 4-9. Immersion Testing Data Recording and Documentation Sheets 4-39
Figure 4-10A. Reductions in Copper Corrosion for Treatment Alternatives 4-41
Figure 4-10B. Reduction in Lead Corrosion for Treatment Alternatives 4-42
Figure 5-1A. pH Cumulative Frequency Distribution, January - June, Plant D .... 5-8
Figure 5-IB. pH Cumulative Frequency Distribution, January - June, Plant H . . . . 5-9
Figure 5-2A. Finished Phosphates VS Time. Plant D 5-10
Figure 5-2B. Finished Phosphates VS Time. Plant H 5-11
Figure 5-3. Example 4.4.2 - Uses of Corrosion Monitors in the
Plimpton City Distribution System 5-29
Figure 6-1. Extent of LSL Control 6-3
Figure 6-2. LSLRP Schedules by System Size 6-10
viii
-------
Glossary of Terms'
TERM
DESCRIPTION
Calcium Adjustment
Consecutive System
Corrosion Inhibitor
Corrosion Control Study
Corrosion Control Treatment
Coupon
Demonstration Testing
Desk-Top Evaluation
Flow-Through Testing
First-Draw Tap Sample
Large Water System
LSL Sample
Materials Survey
Medium-Size Water System
The addition of calcium to shift chemical equilibria to
produce a less corrosive water.
A public water system (PWS) which receives treated
water from another PWS where the interconnection
of the systems justifies treating them as a single system
for monitoring purposes.
A chemical, usually phosphate or silicate based, that
can be used to reduce corrosion.
A desk-top evaluation, static testing, or flow through
testing designed to identify optimal corrosion treatment.
Treatment to minimize the dissolution of lead and/or
copper during water delivery to consumers.
Piece of metal used to evaluate the rate of corrosion
by insertion into piping systems.
Flow through or static testing methods used to illustrate
the effectiveness of a particular corrosion control
treatment.
An office study which compiles historical information
and literature to assist in determining appropriate
corrosion control treatment.
An experimental approach which uses a pipe loop(s)
or other apparatus that provides moving water to contact
the testing surfaces.
One-liter sample collected from the kitchen or bathroom
cold-water faucets of targeted sample sites representing
water standing in the interior piping for at least six
hours.
A water system that serves more than 50,000 persons.
One-liter samples collected from locations served by
lead service lines (LSLs) representing water standing
in the LSL for at least six hours.
An investigation of the materials used in home plumbing
and service lines to assist PWSs in located targeted
sample sites.
A water system that serves greater than 3,300 and less
than or equal to 50,000 persons.
'This glossary provides general descriptions of some of the technical terms used in this manual. Some
of these terms are also defined in the lead and copper rule (see 40 CFR section 141.2). The definitions in
this document, although worded somewhat differently, are intended to be consistent with the Agency's regulatory
definitions.
IX
-------
INTRODUCTION
Chapter 1.0 —
Introduction
The Lead and Copper Rule (LCR) was
promulgated by EPA on June 7, 1991 as
a treatment technique requirement with
major provisions to be implemented over
the following decade. The public water
systems (PWSs) that are subject to
compliance with the LCR are community
water systems and non-transient non-
community water systems. These PWSs
must either demonstrate that optimal
treatment has been installed to control
lead and copper or else that the existing
lead and copper levels in consumers' tap
water are below acceptable levels. In
addition to the water treatment require-
ments contained in the LCR, public
education and lead service line (LSL)
replacement provisions are part of the lead
and copper national primary drinking
water regulations.
In order to assist States in implement-
ing the requirements of the LCR, the EPA
has issued the LCR Guidance Manual.
Information regarding all components of
the Rule are discussed in the Guidance
Manual, along with supporting suggestions
and direction for State and PWS actions
which may be needed to fully implement
the Rule according to its intent.
The LCR Guidance Manual has been
issued in two volumes and is intended to
assist States and PWSs alike in furthering
their understanding of the LCR and its
implementation. The first volume was
released by EPA in September 1991 and
focuses on the monitoring portion of the
Rule. This second volume presents guid-
ance on implementing optimal corrosion
control treatment and the LSL replacement
aspects of the LCR. A separate document
has been prepared to assist PWSs in
developing and conducting an effective
public education program in response to
the LCR (USEPA, 1992).
The information presented in the LCR
Guidance Manual is not limited to the
strict terms of the LCR. Supplemental
information that may be useful to PWSs
is also provided regarding such topics as
performing corrosion studies, evaluating
material survey data for LSL replacement,
and formulating recommendations for
optimal treatment. Table 1-1 presents the
location of selected "topics" in which most
PWSs and/or State agencies would be
interested.
It is not the intent of the LCR Guidance
Manual to be an authoritative reference
on corrosion control - in theory or in
practice - but, rather, to (1) provide
direction about the implementation of the
corrosion control aspects of the LCR; (2)
indicate sources of additional information
regarding the application of theoretical
and practical aspects of corrosion control
treatment/evaluations; and (3) present a
logical and reasonable direction for
evaluating optimal corrosion control
treatment and performance for PWSs.
The Lead and Copper Guidance Manual
is intended to provide supporting direction
to States and public water suppliers so
that the requirements of the Lead and
Copper Rule may be achieved. The focus
of the manual is to supplement materials
readily available in the literature, referring
1-1
-------
INTRODUCTION
to these information sources for further
reading where appropriate, and to provide
practical suggestions and recommendations
for accomplishing the objectives of the
Rule. This document is designed to provide
technical guidance to primacy agencies
administering the SDWA as they exercise
their judgment in implementing the
national primary drinking water regula-
tions for lead and copper. This guidance
is a general statement of policy which does
not establish a binding norm on primacy
agencies or public water systems and is
not finally determinative of the issues
addressed. Decisions made in any particu-
lar case will be governed by the applicable
provisions of the SDWA and 40 CFR Parts
141 and 142.
1.1 Reference
USEPA. 1992. Lead in Drinking Water
Regulation: Public Education Guidance.
Office of Water (Washington, D.C.).
1-2
-------
Table 1-1. Topical Locator by Subject Matter for Lead and
Copper Rule Guidance Manual
Subject Matter
Requirements by System Size
Small
Medium
Large
Principles of Corrosion
Corrosion Control Treatment Alternatives
Regulatory Requirements
Intent of the LCR
Who Must Perform Studies
Corrosion Study Requirements
Recommending Treatment
Setting Operational Criteria
Replacing LSLs
State Actions
Monitoring Requirements
Public Education
Desk-Top Evaluations
Steps in Desk-Top Evaluations
Source Water Treatment Guidelines
Suggested Treatment by Water Quality
Criteria
Constraints in Defining Optimal
Treatment
Reporting Forms & Checklists
Case Studies
LCR Guidance Manual — Volume II
Chapters
1
•
2
•
i/
i/
•
•
3
•
•
•
•
•
•
•
•
•
•
•
4
•
•
•
5
•
i/
•
•
6
•
•/
•
•
Appendices
A
•
•
B
•
C
LCRGM
VOL.1
*:
•
•
•
•
LCRGM
Pub. Ed.
w*
•
•
•
•
-------
Table 1-1. Topical Locator by Subject Matter for Lead and
Copper Rule Guidance Manual (continued)
Subject Matter
Demonstration Testing
Scope of Studies
Organization
Flow-Through Testing
Static Testing
Measurement Techniques
Data Handling
Non-Parametric Statistics
Secondary Testing Programs
QA/QC Components
Examples
Recommending Optimal Treatment
Operating Full-Scale Treatment
Setting Operational Criteria
Start-Up Operations
Troubleshooting
Implementing Optimal Treatment
Examples
LCR Guidance Manual — Volume II
Chapters
1
•
2
•
3
•
4
•/
i/
•
•
•
•
•/
•
•
•
•
•
5
•
•
•
i/
•
6
-
Appendices
A
B
•
•
•
C
•
•
LCRGM
VOL.1
*
LCRQM
Pub. Ed.
«*
* USEPA. 1991. Lead and Copper Rule Guidance Manual — Volume 1. Office of Ground Water and Drinking
Water (Washington, D.C.).
** USEPA. 1992. Lead in Drinking Water — Public Education Guidance. Office of Ground Water and Drinking
Water (Washington, D.C.).
-------
REGULATORY REQUIREMENTS FOR STUDIES
Chapter 2.0 —
Regulatory Requirements for
Corrosion Control Studies
The regulatory requirements in the LCR for
corrosion control studies are presented below
with recommendations regarding the
implementation of these requirements by
primacy agents, namely state drinking water
authorities.
2.1 Large PWSs
Large PWSs subject to the provisions of the
LCR are any community water system (CWS)
or non-transient non-community water system
(NTNCWS) which serves populations over
50,000 people. All large PWSs are required
to define and maintain optimal corrosion control
treatment within their jurisdiction. This may
be the treatment currently in-place or an
alternative treatment recommended as a result
of performing a corrosion control study.
2.1.1 Regulatory Requirements.
The Rule (141.82(c), 56 FR 26550) specifies
six conditions to be met when performing a
corrosion control study as described below:
• Evaluate the effectiveness of each of the
following treatment and, if appropriate,
any combinations of these approaches:
- Alkalinity and pH Adjustment
- Calcium Hardness Adjustment, and
- Phosphate- or silicate-based corrosion
inhibitors.
Collect data from pipe rig/loop tests,
metal coupon tests, partial-system tests (full-
scale), or from documented, analogous
treatments used in or tested at other systems
of similar size, water chemistry, and
distribution system configuration.
Analyze the following water quality
parameters in the course of testing: lead,
copper, pH, alkalinity, calcium,
conductivity, water temperature, and
orthophosphate or silicate when an
inhibitor containing the respective
compound is used.
Identify constraints (chemical or
physical) which may limit the application
of a particular treatment option. The
existence of one of the following
conditions should be documented as part
of this process:
- A particular corrosion control
treatment has adversely affected other
water treatment processes when used
by another PWS with comparable
water quality characteristics; and/or,
- From the experience of the PWS, a
particular corrosion control treatment
was found to be ineffective and/or
to adversely affect other water
treatment processes.
Assess the secondary impacts due to
the effect of corrosion control treatment
on other water treatment processes.
2-1
-------
REGULATORY REQUIREMENTS FOR STUDIES
• Recommend to the State the optimal
corrosion control treatment as identi-
fied by the PWS based on an analysis
of the available data with supporting
documentation and rationale.
While each of the above elements
present important pieces of a corrosion
control study, the organization and
execution of a study are left to the PWS.
2.1.1.1 Scope of testing activities.
By requiring all systems conducting
studies to evaluate specific treatment
alternatives, EPA did not intend for each
PWS to construct pipe rigs or conduct
bench-scale tests to accommodate any and
all treatment options. EPA anticipated
that preliminary screening or "desk-top"
evaluations would be utilized as an initial
step to limit study comparisons and costs.
Alternatives would generally be screened
on the basis of available findings from:
(1) other corrosion control studies for
systems with comparable water quality;
(2) theoretical and applied research efforts;
and (3) the potential adverse impacts
associated with treatment modifications.
As a result of the desk-top evaluation, the
most feasible alternatives can be selected
(at most, two or three treatment options)
for additional evaluation through demon-
stration testing. EPA believes that, in
certain cases, the results of the desk-top
evaluation could suffice in the selection
of optimal treatment, and additional
testing may not be required. However, any
PWS that does not conduct a thorough
evaluation of its treatment recommenda-
tion must realize the risks involved. A
desk-top evaluation considers alternatives
based on the experience of other PWSs and
product manufacturers' recommendations.
As each PWS has a unique supply, treat-
ment, and distribution system, assurance
that the recommended treatment will be
effective is lacking without actual demon-
stration testing.
As discussed previously, demonstration
testing may not be necessary for some
large PWSs to identify optimal treatment.
Table 2-1 (a) presents a recommended
matrix of the minimum degree of testing
to be performed by large PWSs based on
the results of initial monitoring for lead.
The only provision of the Rule which
classifies the existing treatment of large
PWSs as optimized for corrosion control
is when the difference between the 90%Pb-
TAP and Pb-POE is less than the lead PQL
for each six-month period of the initial
monitoring program. By definition, the
PQL for lead is 0.005 mg/L; and the lead
value for the source water used in this
determination is the highest source water
lead concentration. If this condition is met,
then no study or testing is required.
However, States may consider the presence
of copper in tap samples when determining
whether the existing treatment is opti-
mized.
Large PWSs, while not experiencing
problems with lead corrosion (when [(90%
Pb-Tap)-(Pb-POE)] < PQL, may find ele-
vated levels of copper for which corrosion
control treatment would be warranted. The
recommended level of effort for corrosion
control studies by large PWSs based on
copper is presented in Table 2-1 (b).
2.1.1.2 Source water treatment.
PWSs are only required to monitor lead
and copper at the points of entry
(Pb/Cu-POE) if either AL is exceeded on
2-2
-------
Table 2-la. Minimum Recommended Corrosion Control Study Components
for Large PWSs. Based on Lead Levels
90th Percentile Tap
Lead Level, pg/L
Source Water (POE) Lead Level, pg/L
Pb-POE < PQL
POL < Pb-POE < 10
Pb-POE > 10
90% Pb-TAP < PQL
None Required
PQL < 90% Pb-TAP < 10
If [(90% Pb-TAP)-(Pb-POE)] 15
Desk-Top Evaluation and
Demonstration Testing
Desk-Top Evaluation and
Demonstration Testing
If [90% Pb-TAP)-Pb-POE)]
-------
Table 2-lb. Recommended Corrosion Control Study Components
for Large PWSs. Based on Copper Levels
90th Percentile Tap
Copper Level, jig/L
90% Cu-Tap > 1 .3 mg/L
90% Cu-Tap < 1.3 mg/L
Source Water (POE) Copper Level, ^g/L ; f 1
Cu-POE > AL
Desk-Top Evaluation,
Demonstration Testing* and
Source Water Treatment Required
Cu-POE
-------
REGULATORY REQUIREMENTS FOR STUDIES
the basis of first-draw tap samples.
Systems may choose to monitor the source
water contribution of these metals simulta-
neously with first-draw tap sampling in
order to determine whether the existing
treatment is optimal with regard to
corrosion control (90%Pb-Tap - Pb-POE
< PQL). Otherwise, this monitoring must
be completed within six months of exceed-
ing the lead or copper AL.
Source water treatment recommenda-
tions must be submitted to the State
within six months of exceeding an AL for
any system. Guidelines for source water
treatment needs are presented in chapter
3.0 (see Table 3-5). If the source water is
contributing more than the AL for either
lead or copper, then source water treat-
ment is required. In those cases where a
significant amount of lead or copper is
present, then treatment is recommended
in order to reduce the overall lead or
copper exposure and to assist PWSs in
meeting the ALs in future monitoring
events. Table 3-5 also shows that source
water treatment is optional when moder-
ate levels of metals are found, and unnec-
essary when very low levels of either lead
or copper are present.
In those cases where systems find
elevated levels of lead or. copper at the
points of entry, the sources of supply (raw
water) should be monitored prior to
treatment and at various stages within
the existing treatment facility (if currently
treating the supply) to determine the
source of the metals. This monitoring will
also assist in assessing the performance
of the existing treatment systems to
remove lead and copper.
Several types of treatment may be
appropriate for removal of lead and copper.
EPA specified the following techniques
within the LCR (USEPA, 1991):
• Ion Exchange
• Reverse Osmosis
• Lime Softening
• Coagulation/Filtration
If a PWS is currently providing conven-
tional treatment (whether alum or ferric
coagulation, iron/manganese removal, or
lime softening), optimizing these treatment
processes may improve lead and copper
removals. If treatment is not available,
package treatment units for any of the
above technologies may be installed at
individual wellheads (especially when the
elevated metals are contributed by a small
number of individual wells) or at a central-
ized treatment location. In the case of
elevated copper, better control or elimina-
tion of copper sulfate applications may
reduce the background level of copper for
some surface water supplies.
States must respond to the recommen-
dations for source water treatment within
six months. If required, PWSs have 24
months to install source water treatment
once approved by the State. For large
PWSs, the installation of source water
treatment could precede corrosion control
treatment by as much as 18 months.
Follow-up monitoring for Pb/Cu-POE and
first-draw lead and copper tap samples
will occur simultaneously after corrosion
control treatment has been installed.
2.1.2 State Actions and
Decisions.
Primacy Agencies, or States, are
responsible for the review of corrosion
study reports which support the PWS's
recommendation regarding optimal
2-5
-------
REGULATORY REQUIREMENTS FOR STUDIES
corrosion control treatment. State approval
for study design and implementation is
not required, although it would clearly
benefit most PWSs to involve States in the
planning of a corrosion control study so
that the decisions and criteria used in
selecting optimal treatment are acceptable
to all parties.
In cases where the lead or copper ALs
are exceeded during initial monitoring,
PWSs must submit source water monitor-
ing results and a source water treatment
recommendation to the State within six
months. After an additional six-month
period, States must determine whether
source water treatment is required. When
treatment is necessary, PWSs have
24 months to install the treatment
facilities and have them operational.
2.2 Small and
Medium Size PWSs
Small and medium-size PWSs are any
CWS or NTNCWS serving 3,300 people
or less and 3,300 - 50,000 people, respec-
tively. Corrosion control studies are not
required for these systems unless an AL
is exceeded.
2.2.1 Regulatory
Requirements.
The LCR requires small and medium-
size PWSs to perform initial first-draw tap
monitoring for lead and copper at targeted
sites located within their service area. If
either the lead or copper AL is exceeded
during a six-month monitoring period, the
PWS must submit recommendations for
optimal treatment to the State within six
months of exceeding the AL. For example,
a small PWS begins tap sampling for lead
and copper in July 1993 and by the end
of the first monitoring event (December
1993), the system discovers that the lead
AL was exceeded. The monitoring results
must be reported to the State by January
11, 1994 and recommendations for optimal
treatment are to be provided to the State
by July 1, 1994. The detailed time frames
for small and medium-size PWSs to comply
with the corrosion control and source water
treatment requirements of the LCR are
presented in Tables 2-2 and 2-3.
The treatment recommendations to be
generated include source water and
corrosion control treatment components.
Upon exceeding an AL during initial
monitoring, a small or medium-size PWSs
must also monitor lead and copper at each
point of entry (POE) to the distribution
system to determine whether excessive
metals are being contributed by the source
water. The POE lead and copper levels
must also be reported to the State in
conjunction with the system's recommenda-
tions for optimal treatment.
The recommendation for optimal
treatment (source water and/or corrosion
control) may be based on well-documented
desk-top evaluations, and need not be
determined by demonstration testing of
alternative treatment approaches. Howev-
er, states may require a system to perform
such testing, in which case an additional
18 months would be provided to complete
the corrosion control study. The require-
ment to include demonstration testing in
the determination of optimal treatment
for small and medium-size PWSs does not
have to rely on the PWS performing the
demonstration testing themselves if a
study is underway by another PWS with
comparable water quality characteristics.
2-6
-------
REGULATORY REQUIREMENTS FOR STUDIES
Table 2-2. Timeline for Small PWSs to Comply with the
Corrosion Control and Source Water Treatment Requirements41
PWS Action
Date
Submission to State
First Six-Month Initial Monitoring Period
Results**
• Exceed ALs
Desk-Top Treatment Evaluation Begins
Source Water Monitoring Results
Treatment Recommendation
State Requires Corrosion Studies
State Approves/Designates Treatment
(No Study)
Corrosion Study and Treatment
Recommendation (if Required by State)
State Approves/Designates Treatment
(with Treatment)
Certification that the State-approved
treatment has been installed
Without Study
With Study
First Six-Month Follow-Up Monitoring
Period Results ***
Without Study
With Study
Second Six-Month Fol low-Up Monitoring
Period Results
Without Study
With Study
State Specifies Optimal Water Quality
Parameters
Without Study
With Study
First Six-Month Monitoring Period Results
after State Specifies Optimal WQP —
Routine Monitoring
Without Study
With Study
Jan. 11, 1994
Jan. 1, 1994
July 1, 1994
July 1, 1994
Jan. 1, 1995
Jan. 1, 1996
July 1, 1996
Jan. 1, 1997
Jan. 1, 1998
Jan. 1, 1999
July 11, 1998
July 11, 1999
Jan. 11, 1999
Jan. 11, 2000
July 1, 1999
July 1, 2000
Jan. 11, 2000
Jan. 11, 2001
Form 141-A and Monitoring Results:
Pb/Cu-TAP; WQP-DIS; WQP-POE
Pb/Cu-POE
Treatment recommendations for corrosion
control and/or source water treatment
As necessary, State notifies PWSs required to
perform corrosion studies
Treatment Study Report and Results
Letter of Certification
Letter of Certification
Form 141-A and Monitoring Results:
Pb/Cu-TAP; WQP-DIS; WQP-POE
Pb/Cu-TAP; WQP-DIS; WQP-POE
Form 141-A and Monitoring Results:
Pb/Cu-TAP; WQP-DIS; WQP-POE
Pb/Cu-TAP; WQP-DIS; WQP-POE
Based on Follow-Up Monitoring Results
Form 141-A and Monitoring Results:
Pb/Cu-TAP; WQP-DIS; WQP-POE
Pb/Cu-TAP; WQP-DIS; WQP-POE
2-7
-------
REGULATORY REQUIREMENTS FOR STUDIES
Table 2-2. Timeline for Small PWSs to Comply with the
Corrosion Control and Source Water Treatment Requirements*
(continued)
PWS Action
Dale
Submission to Stale
Second Six-Month Monitoring Period
Results after State Specifies Optimal
WQP — Routine Monitoring
Without Study
With Study
Reduced Monitoring
Ultimate Reduced Monitoring
July 11, 2000
July 11, 2001
See
Appendix A
of Volume I
for Dates
See
Appendix A
of Volume I
for Dates
Form 141-A and Monitoring Results:
Pb/Cu-TAP; WQP-DIS; WQP-POE
Pb/Cu-TAP; WQP-DIS; WQP-POE
Form 141-B when State-specified WQPs have
been maintained for two consecutive six-month
monitoring periods
Form 141-A and Monitoring Results:
Pb/Cu-TAP; WQP-DIS; WQP-POE
Form 141-B when State-specified WQPs
maintained for three consecutive years under
reduced monitoring
Form 141 -A and Monitoring Results
Pb/Cu-TAP; WQP-DIS; WQP-POE
* Specifically for those small PWSs which exceed the ALs and are required to implement corrosion
control treatment and must meet State-specified WQPs.
If a small PWS does not exceed the ALs in the two consecutive monitoring periods, then they may
request reduced monitoring (Form 141-B) when submitting results of the second six-month monitoring
period. Those systems that meet the ALs are only required to submit Form 141-A and Pb/Cu-TAP
monitoring results under reduced monitoring.
** PWSs that meet the ALs in the first six-month round of initial monitoring and fail in the second six-month
monitoring period would submit Form 141 -A with Pb/Cu-TAP results on January 11,1993, and submit
Form 141-A with Pb/Cu-TAP, WQP-DIS, WQP-POE, Pb/Cu-POE results on July 11, 1993. All other
deadlines shown in Table 2-2 should be delayed by six months.
*** PWSs that meet the ALs in the first six-month period and fail to meet the ALs in the second six-month
period of the follow-up monitoring only need to submit Pb/Cu-TAP results for the first six-month period
of follow-up monitoring.
2-8
-------
REGULATORY REQUIREMENTS FOR STUDIES
Table 2-3. Timeline for Medium-Size PWSs to Comply with the
Corrosion Control and Source Water Treatment Requirements*
PWS Action
Submission to State
First Six-Month Initial Monitoring Period
Results**
• Exceed ALs
Desk-Top Treatment Evaluation Begins
Source Water Monitoring Results
Treatment Recommendation
State Requires Corrosion Studies
State Approves/Designates Treatment
(No Study)
Corrosion Study and Treatment
Recommendation (if Required by State)
State Approves/Designates Treatment
(with Treatment)
Certification that the State-designated
treatment has been installed
Without Study
With Study
First Six-Month Follow-Up Monitoring
Period Results ***
Without Study
With Study
Second Six-Month Follow-Up Monitoring
Period Results
Without Study
With Study
State Specifies Optimal Water Quality
Parameters
Without Study
With Study
First Six-Month Monitoring Period Results
after State Specifies Optimal WQP —
Routine Monitoring
Without Study
With Study
Jan. 11. 1993
Jan. 1, 1993
July 1, 1993
July 1, 1993
Jan. 1, 1994
July 1, 1994
July 1, 1995
Jan. 1, 1996
July 1, 1996
Jan. 1. 1998
Jan. 11, 1997
July 11, 1998
July 11, 1997
Jan. 11, 1999
Jan. 1. 1998
July 1, 1999
July 11, 1998
Jan. 11, 2000
Form 141-A and Monitoring Results:
Pb/Cu-TAP; WQP-DIS; WQP-POE
Pb/Cu-POE
Treatment recommendations for corrosion
control and/or source water treatment
As necessary, State notifies PWSs required to
perform corrosion studies
Treatment Study Report and Results as
Discussed in Volume II
Letter of Certification
Letter of Certification
Form 141-A and Monitoring Results:
Pb/Cu-TAP; WQP-DIS; WQP-POE
Pb/Cu-TAP; WQP-DIS; WQP-POE
Form 141-A and Monitoring Results:
Pb/Cu-TAP; WQP-DIS; WQP-POE
Pb/Cu-TAP; WQP-DIS; WQP-POE
Based on Follow-Up Monitoring Results
Form 141-A and Monitoring Results:
Pb/Cu-TAP; WQP-DIS; WQP-POE
Pb/Cu-TAP; WQP-DIS; WQP-POE
2-9
-------
REGULATORY REQUIREMENTS FOR STUDIES
Table 2-3. Timeline for Medium-Size PWSs to Comply with the
Corrosion Control and Source Water Treatment Requirements*
(continued)
PWS Action
Dale
Submission to Stale
Second Six-Month Monitoring Period
Results after State Specifies Optimal
WQP — Routine Monitoring
Without Study
With Study
Reduced Monitoring
Ultimate Reduced Monitoring
Jan. 11, 1999
July 11, 2000
See
Appendix A
of Volume I
for Dates
See
Appendix A
of Volume I
for Dates
Form 141-A and Monitoring Results:
Pb/Cu-TAP; WQP-DIS; WQP-POE
Pb/Cu-TAP; WQP-DIS; WQP-POE
Form 141 -B when State-specified WQPs have
been maintained for two consecutive six-month
monitoring periods
Form 141-A and Monitoring Results:
Pb/Cu-TAP; WQP-DIS; WQP-POE
Form" 141-B when State-specified WQPs
maintained for three consecutive years under
reduced monitoring
Form 141-A and Monitoring Results
Pb/Cu-TAP; WQP-DIS; WQP-POE
* Specifically for those small PWSs which exceed the ALs and are required to implement corrosion
control treatment and must meet State-specified WQPs.
If a small PWS does not exceed the ALs in the two consecutive monitoring periods, then they may
request reduced monitoring (Form 141-B) when submitting results of the second six-month monitoring
period. Those systems that meet the ALs are only required to submit Form 141-A and Pb/Cu-TAP
monitoring results under reduced monitoring.
** PWSs that meet the ALs in the first six-month round of initial monitoring and fail in the second six-month
monitoring period would submit Form 141-A with Pb/Cu-TAP results on January 11, 1993, and submit
Form 141-A with Pb/Cu-TAP, WQP-DIS, WQP-POE, Pb/Cu-POE results on July 11, 1993. All other
deadlines shown in Table 2-2 should be delayed by six months.
*** PWSs that meet the ALs in the first six-month period and fail to meet the ALs in the second six-month
period of the follow-up monitoring only need to submit Pb/Cu-TAP results for the first six-month period
of follow-up monitoring.
2-10
-------
REGULATORY REQUIREMENTS FOR STUDIES
Large PWSs performing demonstration
testing, for example, may provide the
States and small/medium-size PWSs with
relevant experiences and findings for
defining optimal corrosion control treat-
ment. Small and medium-size systems that
want to incorporate demonstration testing
results from another PWS must submit
recommendations to the State within six-
months of exceeding an AL that includes:
1) the rationale supporting the need for
additional information to make a final
recommendation for corrosion control
treatment;
2) the identity of the PWS performing
demonstration testing;
3) the comparability of the small or
medium-size PWS's water quality to
that of the system performing the
demonstration testing;
4) the feasibility for the small/medium-
size PWS to implement the alternative
treatments under investigation in the
demonstration testing program; and,
5) the small/medium-size PWS's willing-
ness to implement the recommenda-
tions resulting from the on-going
demonstration testing program.
For those systems performing their own
corrosion control demonstration testing
program, information is presented in
Chapter 4 of this Guidance Manual on how
to develop and conduct such a study.
States have six months to review the
recommendations of PWSs regarding
optimal treatment or the requirement for
additional testing, and either approve the
selected treatment option or else designate
an alternative treatment for installation.
PWSs have two years in which to install
and start up the approved treatment
alternative on a full-scale basis. At this
point, follow-up monitoring is to be
performed and compliance with the LCR
rests with the ability of the PWS to
properly operate the installed treatment.
2.2.2 State Actions and
Decisions.
State activity in implementing the LCR
requires decision-making, PWS notifica-
tion, monitoring and reporting of compli-
ance status, and oversight of PWS actions.
2.2.2.1 Review of recommended
treatment. Small and medium-size
PWSs which submit recommendations for
optimal treatment should provide the
checklist and Form 141-C for State review.
If insufficient information is made avail-
able by the PWS, the State may request
any additional data necessary to complete
the assessment of the recommendations.
Twelve months are provided for States to
review submittals from medium-size PWSs,
and 18 months are provided for small
system recommendation review. Accep-
tance of the recommended treatment may
be granted by the State or else optimal
treatment must be designated for systems
to install.
Small and medium-size systems are
not required to conduct demonstration
testing (static, flow-through, or full-scale)
before making their recommendations for
optimal corrosion treatment. However, any
PWS that does not conduct a thorough
evaluation of its treatment recommenda-
tion must realize the risks involved. A
desk-top evaluation considers alternatives
based on the experience of other PWSs and
product manufacturers' recommendations.
As each PWS has a unique supply,
2-11
-------
REGULATORY REQUIREMENTS FOR STUDIES
treatment, and distribution system,
assurance that the recommended
treatment will be effective is lacking
without actual demonstration testing.
Small and .medium-size PWSs may
recommend that the findings from a
comparable system performing
demonstration testing be incorporated into
the evaluation of their system; thereby
providing an opportunity for these systems
to utilize the results of relevant testing
programs in the selection of optimal
treatment. However, studies which utilize
static testing and flow-through testing
procedures do not automatically insure
that the selected process will provide
satisfactory results when implemented full
scale. Each PWS must carefully review
its individual situation before deciding
which approach is most appropriate for
its particular set of circumstances.
In reviewing the submittals, several
features of the checklist and Form 141-C
may assist the States in determining the
appropriateness of the recommended
treatment. Namely,
• Completeness of the information
provided;
• Supporting documentation regarding
the experiences of the PWS or other,
comparable PWSs with alternative
corrosion control treatment approaches;
• Consistency with the desk-top evalua-
tion procedures described in the Guid-
ance Manual; and,
• Evidence of the PWS's general under-
standing of the alternative treatment
methods and their application.
A primary concern for States will be
the appropriate use of treatment products
in order that successful corrosion control
programs may be implemented by small
and medium-size PWSs.
2.2.2.2 Requirement for additional
study. PWSs are to be notified within
six months of submitting recommendations
for optimal treatment that a corrosion
control study is required by the State.
Certain small or medium-size PWSs may
desire to perform corrosion control studies
in order to more fully evaluate the alterna-
tive treatment processes. If this is the case,
then these PWSs should submit recommen-
dations for the alternatives to be included
in the demonstration testing to the State
within six months of exceeding the AL in
lieu of recommendations for optimal
treatment. This will provide an additional
six-month period for performing the
demonstration study. Those systems
wishing to incorporate the findings of a
comparable system performing demonstra-
tion testing should include the five items
presented in Section 2.2.1 in their submit-
tal to the State. If the State approves this
recommendation, the PWS would have an
additional 18-months to present final
recommendations for optimal treatment,
documenting the incorporation of the
findings from the demonstration testing
performed by the relevant system.
2.2.2.3 Designating alternative
treatment. States have the authority to
designate treatment for small and medium-
size PWSs which have exceeded the ALs
and submitted recommendations for
optimal treatment. However, it is recom-
mended that States and PWSs mutually
determine optimal treatment in cases
where the recommended approach appears
to be questionable by the State. Additional-
ly, States could require demonstration
2-12
-------
REGULATORY REQUIREMENTS FOR STUDIES
testing when significant uncertainty
regarding the performance of alternative
treatments cannot be resolved through
other means.
2.2.2.4 Notification requirements.
States have several notification steps
relevant for small and medium-size PWSs
exceeding ALs during initial monitoring.
The dates and types of notification must
be issued by States as part of the treat-
ment requirements for the LCR are
presented in Table 2-4 for the case where
an AL is exceeded during the first six-
month period of initial monitoring.
2.3 References
USEPA. 1991. Technologies and Costs for
the Removal of Lead and Copper from
Potable Water Sources. Office of Ground
Water and Drinking Water. (Washington,
D.C.).
Table 2-4. Dates for State Notification*
Notification Action
Requirement for Corrosion
Control Studies
Source Water Treatment
Approval/Disapproval
Corrosion Control Treatment
Approval/Designation
Smaif PWSs
January 1995
January 1 995
July 1996
Medium-size PWSs
January 1994
January 1994
January 1995
These dates are based on the assumption that the water system exceeded an action level in the
first six-month period of the initial monitoring. For those small and medium-size systems that meet
the ALs in the first six-month period and fail in the second six-month period, the dates would be
delayed by six months.
2-13
-------
SCREENING OF ALTERNATIVES
Chapter 3.0 —
Screening of Corrosion
Control Alternatives
Many small and medium-size PWSs
will be required to evaluate, select and
implement optimal corrosion control
treatment to meet lead and copper action
levels (ALs). Additionally, most large
PWSs will be required to perform corrosion
control studies which includes desk-top
evaluations of alternative treatment
approaches. States will likewise be re-
quired to review the findings and recom-
mendations of corrosion control investiga-
tions, and, in some cases, designate
treatment for LCR compliance. To assist
each in these endeavors, this Chapter
provides:
• a discussion of the basic principles of
corrosion and the available corrosion
control treatment approaches;
• the steps necessary to develop treat-
ment recommendations for small and
medium systems exceeding an AL or
large systems required to perform desk-
top evaluations;
• a checklist for small and medium-size
PWSs and States to use in evaluating
the selected treatment; and
• several case studies illustrating the
procedure and rationale used to per-
form desk-top evaluations.
References are also provided for those
seeking more detailed and rigorous
presentations on this subject.
3.1 Principles of
Corrosion and Corrosion
Control
Corrosion causes the deterioration of
crystalline structures that form the pipe
materials, and can occur by one of three
principle mechanisms: abrasion, metabolic
activity, and dissolution. Abrasion is the
physical removal of pipe material due to
irregularities in the pipe surface which
may dislodge under high fluid velocities.
Metabolic activity refers to the utilization
of pipe materials as a nutrient supply by
microorganisms. The dissolution of pipe
materials occurs when favorable water
chemistry and physical conditions combine,
generating the following possible corrosion
scenarios:
• Uniform Corrosion - when the water
freely dissolves metals from the pipe
surface;
• Concentration Cell Corrosion - when
anodic and cathodic points are estab^
lished along the pipe surface, causing
the sacrifice of metals at the anode
(dissolved metal species) and the re-
precipitation of less soluble metal
compounds at the cathode.
• Galvanic Corrosion - when two dissimi-
lar metals are in contact with each
other, accelerating the dissolution of
the material with the greater tendency
to corrode.
3-1
-------
SCREENING OF ALTERNATIVES
Corrosion of drinking water distribution
systems can result from any of the above
mechanisms or combinations of the various
types of corrosion activity as illustrated
in Figure 3-1. Alteration of water quality
characteristics via treatment can exten-
sively reduce some forms of corrosion
activity, but may have a less significant
affect on others (AWWARF/DVGH, 1985).
Corrosion control treatment is princi-
pally intended to inhibit dissolution. The
objective is to alter the water quality such
that the chemical reactions between the
water supply and the pipe materials favor
the formation of a protective layer on the
interior of the pipe walls. Corrosion control
treatment attempts to reduce the contact
between the pipe and the water by creat-
ing a film that is: (1) present throughout
the distribution and home plumbing
systems; (2) relatively impermeable; (3)
resistant to abrupt changes in velocity
and/or flow direction; and (4) less soluble
than the pipe material (Neff, 1991).
Coincidental reductions of other
corrosion activity may be accomplished
when dissolution of lead and copper are
minimized. Abrasion of piping materials
is typically accelerated when corrosion
byproducts, such as tubercles, are present
in the distribution system. Abrasion
activity normally diminishes when tuber-
cles are reduced or if the tubercles can be
coated with a less permeable substance.
This effect has been noted by several full-
scale systems which have reported fewer
customer complaints about red or black
water events after corrosion control
treatment was implemented.
Most researchers agree that implement-
ing corrosion control will alter the finished
water chemistry which subsequently may
influence microbial growths within the
distribution system. Recent studies have
shown that biofilms are strongly associated
with corrosion byproducts within distribu-
tion systems (Allen, et al., 1980; Herson,
etal., 1991; AWWARF, 1990a). This
association makes the biofilms more
resistant to disinfection, and therefore,
more persistent when active corrosion
takes place in distribution system piping.
While biofilm formation may be promoted
by corrosion, it remains difficult to accu-
rately quantify the effects of microbial
activity on corrosion rates in distribution
systems and the effect of treatment on
such activity.
Some PWSs have also experienced
increases in distribution system microbial
growth when corrosion control treatment
was implemented due to the addition of
nutrients (phosphorus, inorganic carbon,
silica) to the finished water. In particular,
this may become a problem within distri-
bution systems where chloramines are
used for final disinfection and a phospho-
rus-based inhibitor is applied for corrosion
control. As chloramines are reduced during
oxidation, ammonia (a potential nitrogen
source) is released into the water. Thus
the presence of two major nutrients,
nitrogen and phosphorus, could increase
microbial growth. This is especially likely
in the extremes of the distribution system
where localized areas with inadequate
disinfectant may occur (Hoehn, 1991).
Algal growth may also occur in uncov-
ered distribution system reservoirs. The
primary nutrients necessary for algae to
proliferate are nitrogen and phosphorus.
Phosphorus tends to be the controlling
nutrient as some algal species are able to
obtain nitrogen from the atmosphere for
3-2
-------
SCREENING OF ALTERNATIVES
M« : SOLUBLE METAL
M«-C= LESS SOLUBLE METAL COMPOUND
(A)
CONCENTRATION CELL CORROSION
(8)
METABOUCAUY -ACTIVATED
CORROSION AND ABRASION
(C)
GALVANIC CORROSION
Figure 3-1. Forms of Corrosion Activity Encountered in
Potable Water Distribution Systems
3-3
-------
SCREENING OF ALTERNATIVES
their metabolic processes. Thus, the use
of a phosphate-based inhibitor may
promote unwanted algal growth in some
systems. In the early 1980s a state agency,
the Metropolitan District Commission
(MDC), was responsible for supplying
water to the Boston metropolitan area.
One reason that MDC chose to discontinue
feeding a zinc orthophosphate inhibitor
for corrosion control was the possibility
that the phosphate was responsible for
increased algal growth in the distribution
system reservoirs (Karalekas, et al., 1983).
3.2 Corrosion Control
Treatment Alternatives
As illustrated in Table 3-1, available
corrosion control technologies can be
characterized by two general approaches
to inhibiting lead and copper dissolution:
(1) forming a precipitate in the potable
supply which deposits onto the pipe wall
to create a protective coating; or (2)
causing the pipe material and the potable
supply to interact in such a way that metal
compounds are formed on the pipe surface,
creating a film of less soluble material.
The difference in these two approaches
is the mechanism by which the protective
film is formed. In the former method,
insoluble compounds are formed by
adjusting the water chemistry to cause the
precipitation of the compound onto the
pipe wall. The success of this method is
dependent on: (a) the ability to form
precipitates in the water column, and (b)
the characteristics of the deposit on pipe
walls, including its permeability, adher-
ence strength, and uniformity. In the latter
approach, the mechanism is the passiva-
tion of the pipe material itself through
the formation of less soluble metal com-
pounds (carbonates or phosphates) which
adhere to the pipe wall. In the case of non-
metallic pipe materials, such as asbestos-
cement (AC) pipe, passivation and precipi-
tation mechanisms are also operative. The
calcium present in the AC pipe acts as the
metallic component, being available to
react with the carbonate or phosphate
species under passivating conditions.
Various chemical treatment practices are
available to promote precipitation and/or
passivation in PWSs. The most effective
corrosion control treatment may actually
rely on some combination of these two
mechanisms (AWWARF/DVGM, 1985;
AWWARF, 1991; Kirmeyer and Logsdon,
1983; AWWARF, 1990b).
In general, the available corrosion
control treatment technologies are:
• Alkalinity and pH Adjustment,
which refers to the modification of pH
and/or alkalinity (as a surrogate for
dissolved inorganic carbonate) to induce
the formation of less soluble compounds
with the targeted pipe materials. This
method utilizes passivation as the
mechanism for corrosion control.
• Calcium Hardness Adjustment,
which refers to the adjustment of the
calcium-carbonate system equilibrium
such that a tendency for calcium
carbonate precipitation results. This
method of corrosion control depends
upon precipitation as the means of
protecting piping systems. The term
"calcium hardness adjustment", in
many cases, may be a misnomer since
calcium addition or reduction may not
be required. Instead, modifying the pH
and/or alkalinity through treatment
may be the mechanism for achieving
a tendency for calcium carbonate
precipitation.
3-4
-------
SCREENING OF ALTERNATIVES
Table 3-1. Conceptual Framework for Corrosion
Control Approaches
Control
mechanism
Treatment
Approach
Key Water
Quality
Parameters
Appropriate
Chemical Feed
Systems
Passivation
pH/Alkalinity
Adjustment
pH, Alkalinity,
TDS,
Temperature
Lime
Soda Ash
Sodium Bicarbonate
Caustic Soda
Carbon Dioxide
Corrosion
Inhibitor
pH, Alkalinity,
Metals, Hardness,
Temperature
Orthophosphate
Silicates
Polyphosphate
Ortho-Polyphosphate
Precipitation
Calcium
Adjustment
Calcium, pH,
Alkalinity, TDS,
Temperature
Lime
Soda Ash
Sodium Bicarbonate
Caustic Soda
Carbon Dioxide
3-5
-------
SCREENING OF ALTERNATIVES
• Corrosion Inhibitors, which refers
to the application of specially formulat-
ed chemicals characterized by their
ability to form metal complexes and
thereby reduce corrosion. This method
employs passivation of the metal
surface as the means of corrosion
control. The common corrosion inhibi-
tors generally available include ortho-
phosphate, polyphosphates, poly-ortho-
phosphate blends, and silicates.
Each of these treatment techniques is
discussed more extensively in the following
sections.
3.2.1 Alkalinity and pH
Adjustment.
The solubility of metals is dependent
on the specie in which that metal is found.
Elemental lead and copper will form
complexes with such chemical groups as
the hydroxyl (OH), carbonate (C03),
bicarbonate (HCOg), orthophosphate (POJ,
and silicate (SiO2). The pH/alkalinity
adjustment method relies upon the
formation of less soluble metal species
consisting of hydroxyl-carbonate com-
pounds.
Figures 3-2 and 3-3 present an example
of the family of solubility contour diagrams
for lead and copper, respectively, which
are derived for various temperature and
ionic strength conditions. These particular
contour diagrams are based on the theoret-
ical solubility of various metal hydroxy-
carbonate species for a water with moder-
ately low total dissolved solids (200 mg/L
TDS = 0.005 Ionic strength) and tempera-
ture of 25 °C. To read the chart, the x-axis
is the dissolved inorganic carbonate (DIG)
content, and the y-axis is the pH of the
treated water. A chart to convert total
alkalinity to DIG is provided in Table A-2
of Appendix A. For a particular pH and
DIG, the theoretical lead solubility, for
example at point A in Figure 3-2, would
be 1Q-07 = 0.20 mg/L lead. By increasing
the pH alone to pH = 9 (point B) the lead
solubility would decrease to 10"080 = 0.16
mg/L. If the DIG content were reduced as
well (moving from point B to point C on
Figure 3-2), the theoretical lead solubility
is further reduced to 10"0'90 =0.13 mg/L.
As Figure 3-2 illustrates, the minimum
lead solubility occurs at relatively high pH
conditions (pH 9.8) and low alkalinity (30-
50 mg/L as CaCO3 for DIG). Similar pH
and alkalinity conditions provide minimum
solubility for copper as shown in Figure
3-3. However, copper solubility appears
to be more strongly related to pH than
alkalinity.
These types of figures may be used to
assess the potential value of applying a
pH/alkalinity adjustment treatment
technique for particular supplies. Alterna-
tive water quality goals - consisting of
modified pH and alkalinity conditions -
may be evaluated by determining the
estimated reduction in theoretical lead and
copper solubility. The approach which
should be considered a candidate is able
to: (1) maximize the relative reduction
in lead and copper solubility with respect
to the existing treatment, and (2) meet all
other treatment objectives at the least cost.
The chemical feed systems which may
be installed to modify pH and alkalinity
conditions in the finished water are
summarized in Table 3-2. Many of the
chemicals shown in Table 3-2 will both
increase the pH and the alkalinity of the
finished water. In some cases,
3-6
-------
SCREENING OF ALTERNATIVES
Contour Interval = 0.05 units
90 100
ISO 200 290 300 390 400
. «« CaCO,A INORGANIC CO,
490 900
Note: Contour line.* represent theoretical concaurancos of soluble lead
expressed as a log,g(Pb-Coac) in mg/l_ The lead conrnmarion for example,
it point A is calculated as 10 *'• 0.20 mg/L.
Figure 3-2. Contour Diagram of Lead (II) Solubility in the
System Lead (Il)-Water-Carbonate at 25°C and an Ionic
Strength of 0.005 mol/L
3-7
-------
SCREENING OF ALTERNATIVES
Contour Interval = 0.10 units
5 SO 100 ISO ZOO 250 300 330 4OO 450 900
mg CaCOj/L INORGANIC COi
Note: Contour lines represent theoretical concentrations of soluble copper
expressed as a log,0 (Cu-Conc) in mg/I_ The copper concentration for example,
at point A is calculated as 10 '"» 0.0004 mg/l_
Sam: Scfaock. MJL 19U. EPA 6OV945-007
Figure 3-3. Contour Diagram of Copper (II) Solubility in the
System Copper (Il)-Water-Carbonate at 25°C and an Ionic
Strength of 0.005 mol/L
3-8
-------
Table 3-2. Summary of Chemicals Typically Used in pH/Alkalinity
and Calcium Adjustment Corrosion Control Treatment
VpChemical
Caustic Soda, NaOH
Lime. Ca(OH)2
Sodium Bicarbonate,
NaHCO3
Soda Ash, NajCO3
Carbon Dioxide, CO2
Use
Raise pH. Convert
excess CO2 to alkalinity
species
Raise pH. Increases
alkalinity and calcium
content
Increases alkalinity with
little increase in pH
Increases alkalinity with
moderate increase in
pH
Lowers pH. Converts
excess hydroxyls to
bicarbonate and
carbonate species
Composition
93% purity liquid bulk.
Colder climates, bulk
storage at <50% purity to
prevent freezing
95-98% purity as
Ca(OH)2. 74% Active
ingredient as CaO. Dry
storage with slurry feed
98% purity. Dry storage
with solution feed
95% purity. Dry storage
with solution feed
Pressurized gas storage.
Fed either through
eduction or directly
Alkalinity Change
1 .55 mg/L CaCO3
alkalinity per mg/L
as NaOH
1.21 mg/LCaCO3
alkalinity per mg/L
as Ca(OH)2
0.60 mg/L CaCO3
alkalinity per mg/L
as NaHCOj
0.90 mg/L CaCO3
alkalinity per mg/L
as NajHCO3
None
•--.;-.':'"t: :NOte8-,:. ;•,;. '"•;.'.
pH control is difficult
when applied to poorly
buffered water
pH control is difficult
when applied to poorly
buffered water. Slurry
feed can cause excess
turbidity. O&M intensive
Good alkalinity
adjustment choice, but
very expensive
More pH increase
caused as compared to
NaHCO3, but less costly
Can be used to
enhance NaOH or lime
feed systems
O
cb
-------
SCREENING OF ALTERNATIVES
combinations of the available chemical feed
systems are more appropriate to ensure
that pH and alkalinity goals may be met
simultaneously. This is especially impor-
tant in poorly buffered systems where pH
adjustment alone through the use of either
caustic soda or lime, for example, could
cause unacceptably elevated pH levels or
erratic pH levels in the treated water and
within the distribution system. In these
cases, the use of sodium bicarbonate or
carbon dioxide may be used in conjunction
with the lime or caustic soda system to
provide additional buffering capacity.
Apart from those chemical applications
shown in Table 3-2, other treatment
processes may affect the pH/alkalinity of
the finished water; namely, aeration, alum
coagulation, chlorination and fluoridation.
These additional sources of pH and
alkalinity impacts must be incorporated
into the comprehensive treatment design
in order to successfully achieve the
recommended finished water quality goals
for pH and alkalinity.
The operation of a full-scale facility
using the pH/alkalinity modification
approach should consider several factors
in the design of the corrosion control
program:
• the location of each chemical feed for
optimal utilization, including coagu-
lants, oxidants (such as chlorine),
fluoride, and pH/alkalinity modification
chemicals.
• monitoring locations for process control,
whether manual or automatic;
• sequencing the control of chemical feed
rates in order to reach all of the water
quality goals while minimizing chemi-
cal usage; and,
• the available contact time and mixing
conditions necessary to achieve a stable
finished water prior to entry to the
distribution system.
When determining the location of
chemical feed points, the pH adjustment
resulting from chemical additions must
be considered. This is especially relevant
for waters that are weakly buffered.
Chlorine addition in the gaseous form, for
example, will tend to lower the pH while
adding chlorine in the hypochlorite form
will tend to raise the pH. Likewise, both
sodium silicofluoride and hydrofluosilicic
acid which are commonly used in fluorida-
tion are acidic and will tend to lower the
pH. Adjustment of the finished water pH
for corrosion control cannot be permitted
to interfere with the objectives of other
water treatment operations. Disinfection
with free chlorine, for example, is more
effective at lower pH values because the
hypochlorous acid formed by the addition
of chlorine converts rapidly to the hypo-
chlorite ion above pH 7. Hypochlorite ion
has long been known to be less effective
as a biocide than hypochlorous acid. For
instance, under the SWTR, higher CT
values are required at higher pH levels
to accomplish equivalent microbial inacti-
vation.
3.2.2 Calcium Adjustment.
The formation of a calcium carbonate
precipitate may be used to coat the interior
walls of pipes and thereby reduce the
corrosion of the pipe surface. The success
of this treatment depends on delivering
a finished water slightly supersaturated
with calcium and carbonate (at a specified
pH condition) such that calcium carbonate
precipitation occurs. The availability of
3-10
-------
SCREENING OF ALTERNATIVES
the supersaturated conditions throughout
the distribution system and the reliability
of existing techniques to predict the
potential formation of calcium carbonate
precipitates are key factors to providing
corrosion control protection. Success also
depends on the ability to control the
formation of scale buildup to insure that
hydraulic capacity is not unduly sacrificed
in the course of providing corrosion
protection.
The calcium-carbonate equilibrium is
a dynamic system which will change
continuously from the point of entry to the
final service connection throughout the
distribution system. Achieving a continu-
ous coating of calcium carbonate precipi-
tate is difficult without causing excessive
precipitation in some portions of the
system. This can result in significant
reductions to the supply capacity of the
distribution system, especially in the
vicinity of the treatment plant, and require
those lines to be cleaned in order to
reestablish the necessary hydraulic
conditions.
The complications associated with
calcium adjustment are increased by the
difficulties in precisely determining the
degree of calcium carbonate precipitation
in the treated water. Several indices have
been proposed to describe the calcium-
carbonate equilibrium, and the tendency
of water to form precipitates. PWSs should
exercise caution, however, when using
traditional indices to predict performance
for lead and copper control. Such indices
may not be adequate to predict the
performance of the calcium adjustment
approach, although they may be useful to
initially estimate the water quality
conditions necessary to precipitate calcium
carbonate. The Calcium Carbonate Precipi-
tation Potential (CCPP) index may be the
most useful for this purpose. A more
detailed description of the CCPP and its
method of calculation is provided in
Appendix A.
To understand and effectively utilize
any of the indices discussed in Appendix
A, or to derive calcium carbonate satura-
tion conditions without the use of indices,
it is necessary to review the calcium-
carbonate equilibrium system. Figure 3-4
presents the solubility diagram for calcium
carbonate as a function of pH under "closed
system" conditions, i.e., no exchange of
carbonate species (COg) is permitted
between the water and air systems. Open
systems could involve the dissolving and
de-gassing of carbon dioxicte, which would
affect calcium carbonate solubility. As the
pH increases, the solubility of calcium
carbonate decreases such that more
calcium carbonate will precipitate rather
than stay in solution. However, these
reactions are not instantaneous, and
therefore, sufficient time must be provided
within the targeted pH range for precipita-
tion to occur. For example, lime softening
plants which have excess calcium carbon-
ate present after softening often re-carbon-
ate the clarified water (reduce the pH)
prior to filtration. This increases the
solubility of calcium and prevents the filter
media from becoming coated with calcium
carbonate precipitates which otherwise
would continue to form under the elevated
pH conditions.
The water treatment goals for this
approach should include the pH, carbonate
content (alkalinity) and calcium concentra-
tions necessary to achieve calcium carbon-
ate precipitation. The chemical feed
3-11
-------
0 —
ro
3
i* -
REGION 01
I
6
pll
I
in
I
12
I
14
Figure 3-4. Solubility Diagram for Calcium Carbonate in a
Closed System at 25°C
-------
SCREENING OF ALTERNATIVES
systems which may be used to implement
calcium adjustment treatment are
summarized in Table 3-2. Many of these
chemicals are applicable in the pH/alka-
linity adjustment approach, but the
finished water quality goals would differ.
3.2.3 Corrosion Inhibitors.
Two predominant forms of corrosion
inhibitors are available for potable water
treatment: phosphate and silicate-based
compounds. Somewhat different chemical
mechanisms of corrosion control and water
quality criteria are associated with the
effective use of phosphate and silicate-
based inhibitors. However, both utilize
passivation as the method of providing
corrosion protection.
A plethora of corrosion inhibitor
formulations are commercially available
to PWSs, and caution must be used in the
review and consideration of the alternative
products. As a direct additive to drinking
water supplies, corrosion inhibitors are
subject in most states to the American
National Standards Institute (ANSI)/-
National Sanitation Foundation (NSF)
Health Effects Standard 60 for direct
additives. Products must be certified or
approved by the primacy agent prior to
being used in treating potable supplies.
PWSs should contact their State agency
to determine: (1) whether the State has
adopted the ANSI/NSF Standard 60 for
direct additives, and (2) a list of the
certifying agencies or certified products
for corrosion control treatment.
3.2.3.1 Phosphate inhibitors. Lead
forms at least one orthophosphate solid
of low solubility under typical drinking
water conditions, which can serve as the
basis for corrosion control. Solubility
contour diagrams like those presented for
pH/alkalinity adjustment have been
developed for lead when 0.5 mg/L PO4 is
added to the finished water, as shown in
Figure 3-5. The minimum theoretical lead
solubility is reduced by approximately
0.5-logs with the addition of the orthophos-
phate, and the corresponding pH is much
lower than that associated with the
carbonate system alone.
Copper solubility does not appear to
be markedly reduced by the inclusion of
orthophosphate in solution until extremely
high dosages are applied. The results of
several corrosion studies using orthophos-
phate have found conflicting results with
respect to their contribution to copper
control (AWWARF, 1990b; Moser et al.,
1992). Until additional insight can be
garnered through additional research,
testing should be performed to evaluate
copper control by orthophosphate.
The pH range across which orthophos-
phate appears to be most effective for lead
is 7.4 to 7.8 (AWWARF, 1990b; Lee et al.;
1989; Lechner, 1991). At pH values much
above 7.8, metal phosphate precipitates
can form, causing scale buildup and
hydraulic capacity losses. Waters with low
hardness (calcium < 16 mg/L and a calcium
to magnesium ratio of 0.7) are well-suited
to the use of orthophosphate inhibitors.
The critical parameters to operating
an orthophosphate corrosion control
treatment program are: (1) maintaining
a stable pH in the inhibitor's effective
range throughout the distribution system;
(2) determining the inhibitor composition
best-suited for the specific water quality
objectives and conditions; and (3) applying
3-13
-------
SCREENING OF ALTERNATIVES
lid
Contour Interval = 5 units
100 - 190 ZOO Z90 300 390 400 . 450 9OO
•4 CaCOjA. INORGANIC CO,
Note: Contour lines represent theoretical concezunnans of soluble lead
expressed as 100 * log,0 (Fb-Cooc) in mg/I_ The lead concentration for example.
at poini A is calculated as 10 t-7lnao)» 0.20 mg/L.
•: Scteck. M.R. 19*5. EPA WXV9-SS-007
Figure 3-5. Contour Diagram of Lead (II) Solubility in the
Presence of 0.5 mg/L PO4 at 25°C and an Ionic Strength
of 0.005 mol/L
3-14
-------
SCREENING OF ALTERNATIVES
the appropriate dosage to accommodate
background orthophosphate demand as
well as the corrosion control protection
sought. Phosphate-based inhibitors are
acidic solutions, and the pH effect of their
addition to the finished water must be
considered in determining the suitability
of their application.
Since phosphates are effective over a
constrained pH range, maintaining that
range throughout the distribution system
is an important component of implement-
ing a successful corrosion control program.
For systems which are well buffered, and
whose pH is within the targeted range,
this may not be a critical issue. However,
for those PWSs with poorly buffered
supplies (low alkalinity levels), pH fluctua-
tions within the distribution system can
be significant. For example, with a finished
water alkalinity of less than 20 mg/L as
CaCO3 and pH of 7.5, a PWS found
distribution system pH values ranging
from 6.5 to 9.0, depending on whether the
water had passed through unlined ductile
iron pipe, lined cast iron pipe, or asbestos-
cement pipe. Such fluctuations in distribu-
tion system pH would adversely impact
the performance of the corrosion inhibitor.
Systems with poorly buffered water may
have to install treatment to stabilize pH
in addition to installing corrosion inhibitor
systems for reducing lead and copper
levels.
Thus, the use of inhibitors for corrosion
control within the distribution system is
analogous to maintaining a chlorine
residual within the system as a safeguard
against secondary contamination. Similar
to the chlorine residual, the orthophos-
phate concentration must be sustained to
be effective as a corrosion inhibitor
throughout the distribution system.
However, unlike the chlorine residual
which will inhibit biological functions at
trace concentrations, the inhibitor must
be carried above some minimum concentra-
tion to be useful. Because the composition
of inhibitors vary and in some cases it is
proprietary information, this minimum
concentration should be determined in
conjunction with the supplier.
Phosphate inhibitors are manufactured
in a variety of compositions, including
sodium orthophosphate, zinc orthophos-
phate, polyphosphates, and poly-ortho-
phosphate blends. Each of these groups
of compounds may have differing formula-
tions as to the percentage of effective PO4
present. The selection of a specific inhibitor
may require a preliminary evaluation of
the following: (a) effectiveness in control-
ling lead and/or copper, (b) effects of
depressing the final pH of the treated
water, and (c) impacts on wastewater
treatment facilities required to meet
effluent standards for phosphorus.
Polyphosphates revert (hydrolyze) with
time resulting in an increase in the
orthophosphate ion. This reversion is
affected by, among other parameters, pH,
and available metal ions such as calcium
and zinc. Because chemical suppliers
provide proprietary inhibitors with formu-
lations largely unknown to the user, it
becomes essential that polyphosphate
additives be tested under actual distribu-
tion system conditions. Testing for both
orthophosphate and polyphosphate (see
the hydrolyzable plus orthophosphate
pathway in Figure 4-2, Lead and Copper
Rule Guidance Manual, Volume I) should
be monitored at the point of entry and
throughout the distribution system. These
3-15
-------
SCREENING OF ALTERNATIVES
data will assist in determining the correct
inhibitor dose and in identifying and
understanding the predominant
mechanism of inhibition.
As Holm and Schock point out (Holm
and Schock, 199 la; and Holm and Schock,
1991b), water treatment measures can
sometimes unintentionally increase lead
solubility. Products that contain poly-
phosphates can fall into this category.
Holm and Schock refer to other research
to support their conclusions regarding
polyphosphates (Bailey, 1982; Sheiham
and Jackson, 1981; Neff, et al., 1987; and
Maas, et al., 1991). It is noteworthy that
some researchers disagree with Holm and
Schock, because some of this supporting
research has restrictions which narrow
their application. Nevertheless, EPA
believes that polyphosphates should be
used with caution because: "Applying
chemicals whose effects are not well
understood may be viewed in the extreme
sense as an uncontrolled toxicological
experiment on the general population. We
feel this is the true disservice to the water
utility industry" (Holm and Schock, 1991b).
Polyphosphates are not recommended
for corrosion control purposes in general,
although their application may be benefi-
cial, if not required, for other water
quality, operational, or treatment concerns.
The principle use of such chemicals is to
sequester dissolved metal or cationic
constituents - such as calcium, iron, or
manganese - and reduce their ability to
precipitate either in the distribution
system or within the water treatment
plant. In the case of calcium, polyphos-
phates are used in many softening plants
to minimize the encrustation of filter
media by post-precipitation of calcium
carbonate. For iron and manganese control,
polyphosphates can effectively reduce the
aesthetic discoloration caused by these
compounds. This is often a useful and
necessary benefit of their application,
particularly for groundwater systems
which are heavily mineralized and devoid
of oxygen, ideal conditions for iron and
manganese to solubilize. Seasonally high
levels of iron and manganese can also
occur with surface water supplies when
low dissolved oxygen and reducing condi-
tions in upstream reservoirs increase the
concentration of these minerals.
While polyphosphates have demonstrat-
ed limited direct success toward lead and
copper corrosion control, their use at water
treatment facilities will be necessary in
many instances. Ortho-polyphosphate
blends are being produced which may able
to offer some of the benefits of both uses
to PWSs. These should be considered when
orthophosphate inhibitors are a viable
corrosion control approach, but a poly-
phosphate is also required to meet other
treatment objectives.
Additionally, the proper application rate
for a specific inhibitor should be deter-
mined through testing. As a preliminary
assessment, the necessary dosage should
include the phosphate-demand exerted by
the water quality constituents present in
the finished water. Beyond the dosage
required for effective lead and/or copper
control, metals present in the supply will
combine with phosphates to differing
degrees, imposing an effective "phosphate-
demand" in the following order of preferen-
tial sequence (shown as: maximum >
minimum; or equivalent < > equiva-
lent) (Lechner, 1991).
3-16
-------
SCREENING OF ALTERNATIVES
I. Highest Demand
Manganese —
>lron
>Copper
> Aluminum
> Zinc/Lead
II. Moderate Demand
Calcium< >Magnesium< >Barium< >Radium
III. Lowest Demand
Sodium< >Potassium
The final dosage required should be
sufficient to accommodate the phosphate-
demand and provide the effective inhibitor
residual necessary to achieve lead and/or
copper corrosion control.
3.2.3.2 Silicate inhibitors. The
mechanism involved in controlling
corrosion is unclear for silicate applica-
tions. Silicates are manufactured by the
fusion of high-quality silica sands to
sodium or potassium salts. Sodium
silicates are generally most common with
sodium carbonate being used as the
bonding salt. Conventional sodium silicates
use silica to NagCOg molar ratios between
1.5 and 4 to 1.
The most common form of silicate in
water treatment is the 3.22 weight ratio
sodium silicates at 41 °Baume' solution
with 37-38 percent solids. This has been
used successfully for corrosion control
treatment when targeting reductions in
iron corrosion. For lower pH waters, a
more alkaline silicate product may be
appropriate, such as the weight ratio 2.00
SiaChNa/) with 50.5 °Baume' solution to
reduce acidity and increase the overall
buffering capacity of the water.
The method of controlling corrosion
attributed to silicates appears to be a
combination of adsorption and formation
of less soluble metal-silicate compounds.
Silicates are considered anodic inhibitors,
combining with the free metal released
at the anode site of corrosion activity and
forming an insoluble metal-silicate
compound. These corrosion products
crystallize to form a protective barrier on
the face of pipe walls. However, micro-
scopic and X-ray examinations have shown
two layers of film on iron pipes conveying
water treated with silicates. The majority
of the silicate appears in the uppermost
layer adjacent to the water. This film is
an amorphous silicate film adhered to the
underlying silicate-metal surface. A
slightly corroded surface may be necessary
to form the protective silicate film.
Simultaneously, the application of silicates
in a distribution system with extensive
corrosion byproduct buildup may result
in their release, causing red and turbid
water problems.
Like the use of phosphate inhibitors,
silicates can combine with other constitu-
ents in the delivered water besides the
materials targeted for protection. There-
fore, sufficient dosages must be applied
to compensate for the consumption of
silicate by other metals or cations. Specifi-
cally, calcium and magnesium will readily
react with silica over a large pH range.
Also, silicates are frequently used by small
water systems supplied by groundwater
for iron control. Silicates can sequester
3-17
-------
SCREENING OF ALTERNATIVES
soluble iron and manganese present in the
source water to reduce red and black water
events. Attention to the water quality
conditions prior to their application is
necessary depending on the intended use
and performance of the silicate. The
additional sodium contributed by sodium
silicate formulations should also be
considered by PWSs.
3.3 Evaluating
Alternative Corrosion
Control Approaches
The label "corrosion control" has
historically been applied to a variety of
water treatment techniques which are
frequently used to meet differing water
quality objectives. Until quite recently,
corrosion control practices by PWSs were
typically designed to improve aesthetics,
protect marginal hydraulic capacity, and/or
reduce long-term pipeline maintenance.
Although these objectives remain worth-
while, they have little to do with LCR
compliance, which essentially has rede-
fined corrosion control primarily on the
basis of public health impacts. The princi-
pal objective of the LCR is to minimize the
concentration of lead and copper in
drinking water without compromising
other health-related water quality goals.
This has created some confusion within
certain water supply utilities where long-
standing corrosion control procedures are
now being found "inefFective" with respect
to the new objectives.
A wide variety of proprietary chemicals
have evolved to control pipeline and valve
deterioration, eliminate "dirty water"
complaints, reduce laundry staining, etc.
Some of these "corrosion inhibitor" chemi-
cals can also help reduce lead and copper
levels in drinking water, although many
will not and some could even increase lead
concentrations. Comparisons of corrosion
inhibitors is often controversial because
of the proprietary nature of the specific
chemical formulations and varying water
chemistries. This issue is further compli-
cated by a lack of understanding by many
users about the differences between
chemical products (e.g., ortho and poly-
phosphates) and their relationship to the
formation of metallic precipitates and
protective films in potable water systems.
Beyond compliance with the LCR and
other drinking water standards, additional
benefits and detractions from the installa-
tion of corrosion control treatment may
also be considered when alternative
treatment approaches are reviewed and
assessed. Some examples of the secondary
issues which may be important to PWSs
include:
• Improve the aesthetic quality of the
potable supply (reducing customer
complaints).
• Provide cost savings on the operation
and maintenance of the distribution
system.
• Extend the sludge disposal options
available to wastewater treatment
plants (POTWs) by reducing the overall
metal content of the domestic
wastewater.
• Extend the usable life of customer
water systems, especially hot water
heaters or industrial applications.
• Minimize any unnecessary public
exposure to corrosion byproducts, such
as heavy metals or asbestos fibers.
• Reduce or, at least, not foster microbial
growth in the distribution system.
3-18
-------
SCREENING OF ALTERNATIVES
• Disturb existing coatings in distribution
system piping.
• Develop compatible treatment ap-
proaches for multiple sources of supply
to a distribution system.
• Improve or maintain the hydraulic
capacity of a distribution system.
PWSs must exercise caution in select-
ing technology which is consistent with
conflicting water quality objectives. While
it is not possible to devise a universal
approach for selecting the best corrosion
control scheme, the information provided
below is designed to identify interactions
between LCR treatment goals and those
associated with other SDWA regulations.
The use of chemical treatment to reduce
lead and copper in drinking water will be
dependent upon many site-specific chemi-
cal and physical interrelationships and
may require side-by-side demonstration
testing to assess performance.
Those small and medium-size PWSs
exceeding an AL during initial monitoring
must submit recommendations for optimal
treatment to the State. Large PWSs
required to perform corrosion control
studies will also have to submit either
recommendations for optimal treatment
or the alternative treatment approaches
to be evaluated further as a result of the
desk-top evaluation. To assist in the
development of these recommendations,
the following sections provide a step-by-
step procedure to be used to evaluate
alternative treatment approaches and a
basis for the selection of optimal
treatment.
3.3.1 Steps to Corrosion
Control Assessments.
In order to provide a treatment recom-
mendation to the State, those small and
medium-size PWSs required to install
optimal corrosion control treatment should
assess the three general approaches
discussed above by a desk-top evaluation.
The logic diagram shown in Figure 3-6
presents the process involved in performing
desk-top evaluations for selecting optimal
treatment. This procedure allows systems
to eliminate initially any treatment
approaches which are infeasible and to
then determine the water quality
conditions defining optimal corrosion
control treatment for the feasible
alternatives. Among the resultant alterna-
tives, optimal treatment is to be selected
on the basis of the following criteria:
• the results of lead and copper tap
sampling;
• corrosion control performance based
on either the reductions in metal
solubility or the likelihood of forming
a protective scale;
• the feasibility of implementing the
treatment alternative on the basis of
the constraints identified;
• the reliability of the alternative in
terms of operational consistency and
continuous corrosion control protection;
and,
• the estimated costs associated with
implementing the alternative treat-
ments.
The first step is to describe the existing
conditions of the PWS in terms of its water
quality parameters. As part of this first
3-19
-------
SCREENING OF ALTERNATIVES
ISiepl
1 Step 2
4
I
1
1
1 Step 4
i
T
J Historical Evidence
Define Exisung Conditions:
pH
Alkalinity
Calcium
Inhibitor
T *
i
Determine <
Treatme
-i i
Lead Solubility
Copper Solubility
Calcium Carbonate
rreopiiaQon
Potential
Ca-POE and
iource Water
in Needs
J Define Constraints:
> Other Water Quality Coals
t Distribution System Behavior
t Wastewater Considerations
• Commercial/Industrial User's Needs
-1 *
|
Identify Corrosion Control Priorities M
-, f
~i — Eliminate Unsuitable Approaches J
/ Based on Resolu of Steps 1-4 j
1 t
' T-' Evaluate Viable Alternative i
*
Carbonate Passivation
Alkalinity and pH Adjustment
1
Define Alternative Treatment
Goals for pH and Alkalinity
f
Find Lead and Copper
Solubility for each Alternative
t
Calculate Redactions in
Solubility: Exist- All „ ___
Exist Xl°°*
. f
( Reject \,
V Alternatives M
1
1
r
r
t
1 Inhibitor Passivation 1 C
(Inhibitor Addition) 1 (
jldom Carbonate Precipitation 1
Calcinm Hardness Adjustment) 1
» 1
1 Define Alternative Treatment 1
Goals for pH. Inhibitor Type 1
and Dose |
Goals for CCPP l*~j
» t ^
1 Find Lead and Copper 1
Solubility for each Altemanve 1
Calculate Resulting pH. I g_
Alkalinity, Calcium to Achieve 1 S
CCPP Goal I §
(Calculate Reductions in 1
Solubility: Existing- Alt „ _„ 1
Existing X100%|
i
— 1 i
Evaluate Feasibility of 1 |
Resultant Water Quality Coal] 1
f
— ' Evaluate Each Alternative Based On:
9 Performance
• Feasibility
• Reliability
• Cost
_J[ Selea Optimal \
~V Treatment M
Figure 3-6. Logic Diagram for Evaluating Alternative
Corrosion Control Approaches
3-20
-------
SCREENING OF ALTERNATIVES
step, PWSs can estimate the theoretical
lead and copper solubility as well as the
potential for calcium carbonate
precipitation based on the existing water
quality conditions. Changes in water
quality conditions for alternative treat-
ments can be compared to the existing
conditions to determine their relative
performance and potential to reduce
corrosion.
Each PWS operates under certain
constraints, such as specific water quality
goals, existing coatings in distribution
system piping, multiple sources of supply
of varying water quality, and wastewater
permit limits on metals or nutrient levels
which may be improved or compromised
by corrosion control treatment. Any
constraint which could impact the feasibili-
ty of implementing an alternative treat-
ment should be identified and documented.
This information will be important to the
selection of those treatment options which
are viable alternatives for the PWS to
consider further.
Based on the water chemistry of the
supply and site-specific constraints, the
PWS may eliminate corrosion control
treatment approaches which would be
infeasible to implement successfully. The
remaining options, deemed to be feasible,
should be evaluated on the basis of each
PWS's corrosion control treatment priori-
ties to properly judge the performance of
the alternative approaches. For example,
a system which experiences lead levels in
first-draw tap samples greater than the
AL for lead should set lead control as its
primary goal. A second system which finds
low lead levels, but has elevated copper
levels in first-draw tap samples should set
copper as the primary objective of
corrosion control treatment. However, in
the latter case, optimal treatment should
not worsen lead corrosion behavior and
therefore, the control of lead may be
considered as a constraint acting on the
decision-making process for selection of
optimal treatment.
Each of the three corrosion control
treatment approaches that are viable
options should be evaluated to determine
the water quality characteristics which
describes optimal treatment within each
option. For the passivation methods,
alternative treatments are evaluated by
comparing their relative reduction in the
solubility of each targeted metal (lead
and/or copper). The calcium carbonate
precipitation method is evaluated by the
ability of alternative treatments to produce
sufficient potential for scale-forming
conditions to exist in the distribution
system. The "rule of thumb" guidelines
presented in Appendix A may be used to
rank the alternatives evaluated within this
treatment approach.
The final selection of optimal treatment
will rest on the four factors discussed
above: performance, feasibility, reliability,
and costs. Direct comparison of corrosion
control performance for alternative
treatment approaches may be not possible.
Professional judgement and related
experiences will be necessary to provide
a basis for ranking alternatives on the
basis of performance.
The following sections provide more
detailed descriptions of the various steps
involved in performing a desk-top evalua-
tion of alternative treatments and the
development of final recommendations for
optimal treatment.
3-21
-------
SCREENING OF ALTERNATIVES
3.3.2 Documenting Historical
Evidence.
The first step of the desk-top evaluation
is to identify and document any existing
information pertinent to the evaluation
of corrosion control for the system. Four
categories of data should be compiled: (a)
water quality data; (b) evidence of corro-
sion activity; (c) available results of
corrosion studies performed by other PWSs
as reported in the literature that meet
LCR conditions, i.e. similar water chemis-
try, distribution system, etc.; and (d)
results from prior corrosion studies or
testing performed by the PWS. The most
pertinent information is the results of any
prior corrosion control testing performed
by the system. Beyond the direct testing
results, a comprehensive review of the
other sources of information should be con-
ducted by the PWS.
3.3.2.1 Water quality data. Current
and historical water quality data should
be compiled and analyzed. The key
parameters of interest include pH,
alkalinity, hardness, total dissolved solids
or conductivity, temperature, dissolved
oxygen, and metals (eg., aluminum,
manganese, iron, lead, and copper). These
basic water quality parameters only
represent those most commonly required.
Site-specific requirements should be
considered in the selection of water quality
parameters for review. The data collected
should pertain to raw and finished water
conditions, as well as the water quality
within the distribution system, if available.
Additionally, the results of the initial
monitoring program should be considered
when available.
Understanding the treatment processes
at a PWS facility and their respective
impacts on water chemistry is an impor-
tant aspect of interpreting the water
quality data and evaluating the appropri-
ateness of alternative corrosion control
treatment techniques. Figure 3-7 illus-
trates the relationship between water
quality and alternative corrosion control
treatment approaches. Three major regions
are shown on the basis of pH (low, moder-
ate, and high) with alternative treatment
approaches which may be viable on the
basis of water quality shown for each block
by its respective alkalinity and calcium
levels (low, moderate, or high). To demon-
strate the use of Figure 3-7, consider a
PWS with a pH 7.8, alkalinity of 40 mg
(CaCOg/L, and calcium content of 60 mg
CaCOg/L. The moderate pH (7.5-9.0) chart
is used with treatment alternatives
corresponding to the block for low alkalini-
ty (<50 mg CaCOg/L), and moderate
calcium (50-100 mg CaCO,/L). On the basis
of water quality alone, this PWS should
consider all four treatment alternatives
as viable.
In many cases, site-specific water
quality conditions will reduce the feasibili-
ty of an alternative treatment approach.
For example, it would be reasonable to
eliminate the calcium carbonate precipita-
tion option as a viable treatment approach
for those PWSs exhibiting low pH, alkalini-
ty, and hardness in the treated water due
to the excessive chemical modifications
which would be required to achieve
sufficient calcium carbonate precipitation
in the distribution system.
3-22
-------
SCREENING OF ALTERNATIVES
Low pH
Alkalinity
(mg/LCaCO-)
High
Low
(<50)
Moderate
(50-150)
Calcium
(mg/LCaC03)
Moderate pH
7.5-9.0 *
Alkalinity
(mg/LCaC03)
High
O150)
Moderate
(50-150)
Low
(<50)
Calcium
(mg/LaC03)
• Phosphate Inhibitor only appropriate
for pH conditions less than 8.
High pH
>9
High
0150)
Low
(<50)
Moderate
(50-150)
Calcium
(mg/LCaC03)
Alkalinity
(mg/LCaCO3)
Low Moderate High
(<50) (50-150) (>150)
I I = rjlriim Cvbonue Precpunoo
fjf m Cafeoette Puanhon
I I > Phoiphue Inhibitor
N.V1 ' Siliciir Inlttbilar
Figure 3-7. Suggested Corrosion Control Approaches
Based on Water Quality Characteristics
3-23
-------
SCREENING OF ALTERNATIVES
Conversely, a PWS exhibiting high pH
conditions with moderate to high alkalinity
and calcium contents might concentrate
their efforts on calcium carbonate
precipitation for the following reasons:
• While high pH conditions may be
optimal for lead control, these water
quality conditions are very aggressive
towards iron corrosion and would most
likely cause severe degradation in
distribution system water quality
should calcium carbonate precipitation
not be pursued; and
• High dosages of corrosion inhibitors
may be necessary to maintain an
effective residual throughout the
distribution system due to the presence
of calcium. Also, some inhibitors can
cause existing corrosion byproducts to
be released in the distribution system
causing water quality degradation.
Figure 3-7 is intended to provide
general guidelines on water quality
conditions versus alternative treatment
approaches; it is not intended to serve as
the sole basis for selection or elimination
of the available alternatives. Further,
caution must be raised any time a corro-
sion control approach requires a severe
modification in the existing water quality
entering the distribution system. Disrup-
tions and upset of existing corrosion
byproducts will impact the overall
effectiveness of any corrosion control treat-
ment approach.
3.3.2.2 Corrosion activity. Existing
records indicative of corrosion activity
within the distribution and home plumbing
systems should be identified and analyzed
to inform the PWS of the nature and
extent of corrosion activity anticipated
within the service area. Evidence of
corrosion activity may be obtained by:
(1) reviewing customer complaint records
for dirty water or metallic taste and odor
events, (2) performing an informal survey
of area plumbers regarding the frequency
and nature of plumbing repairs (especially,
for example, hot water heater replace-
ments), (3) reviewing records citing the
inspection of distribution system mains
and service line when being replaced or
repaired, (4) installing and evaluating
corrosion coupons placed within the
distribution system, and (5) water quality
monitoring for metals or other corrosion
byproducts within the distribution system
or home plumbing environments.
While the information listed above may,
in some instances, be incidental in nature -
i.e., causative relationships may not be
easily developed between the observed
effects of corrosion activity and the water
quality within the distribution system,
PWSs may gain a more complete sense
of the corrosion concerns facing their
system.
Example: After reviewing several
years of data, a PWS observed that
complaints from customers about red water
was the predominant source of
dissatisfaction with the water supply and
that the number of complaints was
increasing in recent years. The utility
manager interviewed City plumbing
inspectors, local plumbers, and the PWS's
maintenance department about corrosion
activity to learn more about the potential
problems. As a result of these inquiries,
it was discovered that (a) the average life
of household water heaters in the PWS's
service area is one half of that expected
normally; (b) copper plumbing in
3-24
-------
SCREENING OF ALTERNATIVES
residences often experienced pitting
corrosion resulting in pin-hole failures of
piping; and (c) the highest repair and
replacement rate for distribution system
mains and service lines was in the older
parts of the service area where unlined
cast iron mains and galvanized service
lines were still in-place. Based on these
findings, the utility manager initiated a
monitoring program to determine the
presence of corrosion byproducts and water
quality conditions in the distribution
system and at employees homes. The
incidental information indicated that
copper and iron corrosion were concerns
for the PWS, both in terms of material
failure and water quality. The monitoring
program confirmed these concerns, finding
pH and alkalinity shifts within the cast
iron distribution system and elevated
copper levels in home tap samples. While
the information gathered by the utility
manager did not determine the specific
cause of the distribution and home
plumbing system corrosion, it did further
the PWS's understanding of the potential
corrosion problems in its service area. It
also served as a basis for designing a
water quality monitoring program to the
corrosion activity experienced in the
distribution and home plumbing systems
after installation of treatment.
Several factors should be considered
in evaluating the usefulness of this
information; namely: (1) the frequency
of data collection; (2) the number of
coupons, if used, and their locations within
the distribution system; (3) the analytical
methods and their respective detection
limits; (4) the consistency of the data
temporally and spatially; and (5) the
reliability of the incidence reports.
Included in this pool of information should
be the results of the initial monitoring
program required by the Lead and Copper
Rule, if available.
This information may be used to
prioritize the corrosion control program
elements for the PWS in terms of the key
materials for protection and assess the
general effectiveness of the existing
treatment approach.
3.3.2.3 Review of the literature.
A search and review of the available
literature should be performed to ascertain:
(1) the findings of similar systems when
performing corrosion control testing; and
(2) the theoretical basis for alternative
corrosion control approaches to be
considered by the PWS - thereby, elimi-
nating those approaches which appear to
be infeasible.
Several corrosion control studies have
been performed and the results published
by several water suppliers in the United
States. Each study has site-specific goals
and objectives relevant to the testing
protocols as well as water treatment and
quality conditions. However, the experienc-
es of these systems provide a useful
resource to other PWSs investigating
corrosion control in terms of: (1) study
design and execution; (2) data handling
and interpretation; and (3) recommended
treatment given the goals and constraints
acting on the system. A summary of the
available literature on corrosion control
studies is provided in Appendix B. Note
that great care must be taken in evaluat-
ing studies reported in the literature so
that test protocol, water chemistry,
treatment processes, and so forth are
matched as closely as possible.
3-25
-------
SCREENING OF ALTERNATIVES
3.3.2.4 Prior experience and
studies. Corrosion control treatment is
not a new concern for water suppliers, and
many have performed studies in the past
to assist in the design and implementation
of corrosion control treatment. These past
experiences and studies should be revisited
by PWSs to incorporate their findings and
results in the present evaluation of
corrosion control for lead and copper.
Small systems could use the optimum
corrosion control treatment processes
which were recommended to the State by
the larger PWSs. In some cases, the prior
testing targeted lead and copper control,
and these findings would be directly
applicable to the corrosion control study
objectives for the Lead and Copper Rule.
Additional testing may not be necessary,
therefore, to formulate recommendations
for optimal corrosion control treatment (if
not already considered to be in place).
Example: The Town of Redfield, a
small PWS operating a groundwater well,
found lead levels above the action level
during initial monitoring. In order to
prepare recommendations for optimal
treatment, the PWS operator began
collecting information regarding the
condition of distribution system materials
and the experiences of nearby towns and
.communities. From previous pipe replace-
ment activities, the PWS operator had
noticed a thin, buff-colored deposit on the
walls of distribution system piping. The
groundwater source is well buffered with
an average pH 7.4, alkalinity of 160 mg
CaCOg/L, and calcium hardness of 110 mg
CaCOg/L. The CCPP calculated for the
system is -2.4 mg CaCOg/L.
Redfield needed to determine whether
they were successfully coating the pipes
of the distribution and home plumbing
systems with calcium carbonate deposits.
Plumbing materials from service lines,
distribution mains, and three homes in
the service area were extracting during
repair in order to chemically analyze the
constituents present in the scale. This
analysis confirms that the scale was
predominantly calcium carbonate. How-
ever, observation of the same showed that
it was not uniformly coating the pipe
materials, especially the home plumbing
piping.
The PWS considered the alternative
treatment approaches for corrosion control
and eliminated pH/alkalinity adjustment
(carbonate passivation) due to the excessive
alkalinity and calcium levels per Figure
3-7 presented in the LCR Guidance
Manual. The remaining alternatives were
calcium hardness adjustment and corrosion
inhibitors.
A nearby township having wells located
in the same aquifer as Redfield had
previously installed orthophosphate
inhibitor feed facilities for corrosion
control. After orthophosphate addition, the
treated water had a final pH of 7.35 and
PO4 concentration of 5 mg PO/L to account
for the phosphate demand exerted by the
calcium present in the well water and to
produce an effective residual throughout
the distribution system. Their experience
was not altogether positive, having a
significant number of turbid and dirty
water complaints occurring after the
addition of the orthophosphate. Additional-
ly, within three months of beginning the
phosphate treatment, it appeared that the
hydraulic capacity of the distribution
mains in the vicinity of the well heads was
being significantly reduced. They gave up
3-26
-------
SCREENING OF ALTERNATIVES
the use of the corrosion inhibitor in order
to restore the aesthetic quality of the
delivered water supply.
After learning of these experiences, the
Town of Redfield decided to eliminate the
use of orthophosphate from their alterna-
tive corrosion control treatment approach-
es. Redfield focused their evaluation on
the calcium carbonate precipitation
technique for the following reasons:
• The CCPP condition for the finished
water supply could be readily improved
to produce a more reliable calcium
carbonate deposit on the pipe walls.
This deposit can further be controlled
once treatment is in-place by dissolu-
tion and precipitation conditions in the
treated water to ensure that the
hydraulic capacity of the system is not
compromised.
• Little documentation exists to confirm
the corrosion control performance of
silicate inhibitors with respect to lead
and copper corrosion control for
supplies with high calcium contents.
• Difficulties may arise in controlling
silicate-based deposits to maintain the
hydraulic capacity of the distribution
system since they are not able to be
redissolved.
Based on a CCPP goal of 8.5 mg
CaCOg/L, Redfield determined that a pH
of 7.9 was needed for its finished well
water supply.
3.3.3 Identifying Constraints.
The Rule provides two conditions by
which constraints may be considered in
limiting the availability of alternative
corrosion control treatments. Namely,
options which have been shown either:
(1) to adversely impact other water
treatment processes and cause a violation
of a National Primary Drinking Water
Regulation; or (2) to otherwise be
ineffective for the PWS.
EPA recommends that all constraints
acting on PWSs be identified and consid-
ered in the selection of treatment ap-
proaches either for additional testing or
as the recommended treatment process.
Worksheets are provided in Table 3-3 for
each of the three treatment alternatives
(pH/alkalinity adjustment, calcium adjust-
ment, and corrosion inhibitors) to assist
PWSs in evaluating the constraints acting
on their systems. Constraints have been
extracted from an overview of corrosion
control literature (Swayze, 1983; AWWAR-
F, 1990c; Benjamin, 1990; AWWARF/-
DVGW, 1985; AWWA, 1986; AWWA, 1989).
PWSs should evaluate the impact of
alternative corrosion control treatment
options on regulatory compliance with
existing federal and state drinking water
standards in addition to those regulations
anticipated to be finalized within the time
frame for corrosion control installation by
small and medium PWSs. Table 3-4
presents the schedule for regulatory
actions during the next decade in conjunc-
tion with the compliance timeline for
medium-size and small system implemen-
tation steps for the Lead and Copper Rule.
The key regulatory actions which should
be fully evaluated by small and medium
PWSs for selecting optimal corrosion
control treatment are discussed at more
length below.
3-27
-------
SCREENING OF ALTERNATIVES
Table 3-3a. Constraints Worksheet for pH/Alkalinity
or Calcium Adjustment Treatment Alternatives
Adjusting pH/Alkalinity and/or calcium for corrosion control
typically consists of increasing their levels to generate
favorable conditions for lead and copper passivation or
calcium carbonate precipitation.
A. National Primary Drinking Water Regulations Constraints
Rule
Surface Water
Treatment Rule
Groundwater
Disinfection
Disinfection
Byproducts
Coliform Rule
Radionuclides
Constraint
Reduces inactivation effectiveness of free chlorine if pH adjusted
before disinfection.*
Potential for interference with- dissolved ozone measurements.
May increase turbidity from post-filtration precipitation of lime,
aluminum, iron, or manganese.
Reduces inactivation effectiveness of free chlorine if pH adjusted
before disinfection.*
Potential for interference with dissolved ozone measurements.
Higher THM concentrations from chlorination if pH adjusted
before disinfection.*
Reduced effectiveness of some coagulants for precursor removal if
pH adjusted before coagulation.*
Potential for higher total plate counts, confluent growth, or
presence of total coliforms when chlorination is practiced.
In-plant adjustments may affect removal of radioactive particles if
precipitation techniques are used for coagulation or softening.
Removal of radionuclides during softening may be linked to the
degree of softening. Modifying softening practices to achieve
corrosion control could interfere with removals.
3-28
-------
SCREENING OF ALTERNATIVES
Table 3-3a. Constraints Worksheet for pH/Alkalinity
or Calcium Adjustment Treatment Alternatives (continued)
B. Functional Constraints
Increased potential for post-filter precipitation may give undesirable levels of
aluminum, iron, or manganese.
Process optimization is essential. Additional controls, chemical feed equipment, and
operator attention may be required.
Multiple entry points will require pH/Alkalinity adjustment at each entry location.
Differing water qualities from multiple sources will require adjusting chemical doses
to match the source.
The use of sodium-based chemicals for alkalinity or pH adjustments should be
evaluated with regard to the total sodium levels acceptable in the finished water.
Users with specific water quality needs, such as health care facilities, should be
advised of any changes in treatment.
Excessive calcium carbonate precipitation may produce "white water" problems in
portions of the distribution system.
It may be difficult to produce an acceptable coating of calcium carbonate on interior
piping for large distribution systems. High CCPP levels may eventually lead to
reduced hydraulic capacities in transmission lines near the treatment facility while
low CCPP values may not provide adequate corrosion protection in the extremities of
the distribution system.
Unless operating restraints dictate otherwise, the optimum location for pH adjustment
is after disinfection and near the entrance to the distribution system. If quicklime is
used to adjust pH, for example, it needs to be added prior to filtration so inert
material does not accumulate in the clearwell or enter the distribution system.
3-29
-------
SCREENING OF ALTERNATIVES
Table 3-3b. Constraints Worksheet for
Inhibitor Treatment Alternatives
Corrosion inhibitors can cause passivation of lead and copper by the
interaction of the inhibitor and metal components of the piping system.
A. National Primary Drinking Water Regulations Constraints
Rule
Surface Water
Treatment Rule
Groundwater
Disinfection
Disinfection
Byproducts
Coliform Rule
Radionuclides
Constraint
The application of phosphate-based inhibitors to systems with
existing corrosion byproducts can result in the depletion of
disinfectant residuals within the distribution system. Additionally,
under certain conditions phosphate-based inhibitors may stimulate
biofilms in the distribution system.
Same as above.
No apparent effects.
If corrosion byproducts are released after the application of
inhibitors, coliforms may be detected more frequently and
confluent growth is more likely.
No apparent effects.
B. Functional Constraints
Potential post-filtration precipitation of aluminum.
Consumer complaints regarding red water, dirty water, color, and sediment may
result from the action of the inhibitor on existing corrosion byproducts within the
distribution system.
Multiple entry points will require multiple chemical feed systems.
The use of sodium-based inhibitors should be evaluated with regard to the total
sodium levels acceptable in the finished water.
The use of zinc orthophosphate may present problems for wastewater facilities with
zinc or phosphorus limits in their NPDES permits.
Users with specific water quality needs, such as health care facilities, should be
advised of any treatment changes.
NOTE: If pH adjustment is necessary to produce an effective pH range for the inhibitor,
then the constraints in Table 3-3a would also need to be evaluated.
3-30
-------
Table 3-4. Schedule of Drinking Water Regulatory Activity: 1990-2000
Regulatory Action
Phase 1 VOCs
Phase II SOCs & lOCs
Phase V SOCs & lOCs
Arsenic
Surface Water Treatment Rule
Total Coliform Rule
Radionuclides Rule
Groundwater Disinfection Rule
Disinfectants/Disinfection By-Products
Lead and Copper Rule
Proposal Date
11/85
05/89
07/90
11/92
11/87
11/87
07/91
06/93
06/93
08/88
Final Date
07/87
01/91 & 07/91
07/92
01/95
06/89
06/89
04/93
06/95
06/95
06/91
Effective Date
01/89
07/92 & 01/93
01/94
07/96
01/91
01/91
10/94
01/97
01/97
07/91 & 12/92
03
6
-------
SCREENING OF ALTERNATIVES
Surface and Groundwater Treatment
Rules (SWTR/GWTR) where PWSs will
be required to meet disinfection performance
criteria. Disinfection efficiency is pH depen-
dent for free chlorine where less effective
disinfection results under higher pH
conditions.
Total Coliiorm Rule (TCR) which requires
all PWSs to meet minimum occurrence
standards for the presence of total and fecal
conforms in distribution system samples.
Some PWSs have noted increases in microbi-
ological growth within the distribution
system with the installation of corrosion
control treatment. However, in most cases,
no adverse impact or reductions in heterotro-
phic plate count bacteria have been found
after implementing corrosion control
treatment.
Disinfectants/Disinfection Byproducts Rule
(D/DBPR), currently under development,
will be finalized within the same time frame
as PWSs are installing corrosion control
treatment as a result of the Lead and Copper
Rule. Adjusting pH conditions can affect
the level of certain DBPs, most notably, total
trihalomethanes (TTHMs) and total halo-
acetic acids (THAAs). These two contami-
nant groups are likely to be included in the
future DBPR, and they exhibit opposite
relationships to pH adjustment; TTHM
formation increases with increasing pH,
while THAA formation increases with
decreasing pH. An additional consideration
is the point of pH adjustment within treat-
ment plants since lower pH conditions favor
increased removal of DBP precursors during
coagulation by alum. Compliance with the
DBPR could be compromised by increasing
the pH of coagulation as part of the corro-
sion control treatment approach as it may
reduce the efficiency of conventional
treatment in removing precursor material.
Additional constraints should be considered
by PWSs beyond those required by the Rule.
As presented in Table 3-3b, a selected number
of such limiting conditions for alternative
corrosion control approaches include:
• Compatibility of a treatment approach with
multiple sources of supply.
• Compatibility of a treatment approach for
consecutive systems.
• Reliability features for the particular treat-
ment approach, including: (1) process
control; (2) operational redundancy require-
ments; and (3) chemical supply integrity and
availability.
• Adverse impacts on the service community,
including: (1) commercial users' water
quality criteria; (2) health-care facility water
quality criteria; and (3) wastewater opera-
tions - permit requirements for discharges
and solids handling programs.
The particular conditions which define the
constraints for each system will be site-specific,
and should be thoroughly investigated as part
of the desk-top evaluation aspect of the corrosion
study. Small and "medium systems exceeding
the ALs but not required to perform testing
should consider each of these items when
selecting the optimal treatment for recommenda-
tion to the State. For those large PWSs required
to perform only a desk-top evaluation, rigorous
documentation of any constraints must be
presented to support the recommended treatment
approach for the system. For any PWS perform-
ing corrosion testing, the availability of informa-
tion regarding system constraints will assist in
limiting the optional treatment approaches which
must be evaluated through the testing program.
Example: After exceeding the lead AL
during initial monitoring, the City of Dannyport
3-32
-------
SCREENING OF ALTERNATIVES
began investigating alternative corrosion control
treatment measures to provide the State with
recommendations for optimal treatment. The
City determined through its desk-top evaluation
that raising the pH of the treated water was a
viable treatment approach. Two alternative pH
levels were identified for further consideration.
As a medium-size surface water facility,
conderns were raised regarding compliance with
the SWTR and the ultimate feasibility of
implementing pH adjustment.
The existing treatment provided by Danny-
port is conventional coagulation/ flocculation
with rapid sand filtration. Under the SWTR, at
least 0.5-logs of inactivation of Giardia and
2.0-logs of virus inactivation were required.
The SWTR applied CT values - the product
of the disinfectant concentration at the end of
a disinfection segment and the effective contact
time available within the disinfection segment,
to determine the inactivation achieved during
treatment. The SWTR Guidance Manual
(USEPA, 1989) defined the CT^ as the CT
value required to achieve the desired level of
inactivation. The CTacl was defined as the CT
value actually achieved through treatment for
each disinfection segment within a water
treatment facility. Compliance with the disinfec-
tion requirements is achieved when the sum of
the CT^iCT^ ratios for all disinfection seg-
ments in a facility is greater .than or equal to
1.0.
For the Giardia requirements, the existing
plant's performance was determined to be
adequate to meet the CT required with the sum
of the CT^CT,^ ratios equal to 1.2. Virus
inactivation performance was satisfactory and
would not be affected by pH changes. However,
Giardia inactivation performance is a function
of pH. At the higher pH levels under consider-
ation for corrosion control, the sum of the
ratios would be 0.90 and 0.83,
respectively. Neither case would provide
adequate disinfection performance.
An additional concern is continued compli-
ance with the Total Trihalomethane (TTHM)
standard. Currently, an average of 60 /fg/L
TTHM is found in the distribution system with
seasonal peaks of nearly 100 ug/L TTHM. As
such, increasing the pH of the finished water
supply could only increase the probability of
Dannyport exceeding the future TTHM standard,
expected to be finalized concurrently with the
City's initiation of corrosion control treatment.
Given the above regulatory concerns, the
City of Dannyport determined that pH adjustment
would not be a feasible option.
3.3.4 Evaluating Source
Water Contributions.
When a small or medium PWS exceeds an
AL during initial monitoring, lead and copper
samples must be collected and analyzed at each
point of entry (POE) to the distribution system
within six months of exceeding the AL. It is
recommended that this monitoring be completed
as soon as possible after the AL is exceeded
in order to provide information regarding source
water lead and copper contributions to the desk-
top evaluation effort. The recommendations for
treatment which must be supplied to the States
within six months of exceeding the ALs must
contain source water treatment recommendations
in addition to corrosion control treatment recom-
mendations. Therefore, performing lead and
copper POE monitoring (Pb/Cu-POE) is critical
to the completion of desk-top evaluations.
Table 3-5 presents EPA's guidelines for
source water treatment requirements on the basis
of lead and copper POE monitoring results. If
the source water is contributing more than the
AL for either lead or copper, then source water
3-33
-------
Table 3-5. Source Water Treatment Guidelines for Systems
Exceeding an AL
Note: States have the discretion to set their own guidelines for Source Water Treatment.
Source Water Treatment Guidelines
Not Necessary
Optional
Recommended
Required
Point of Entry Monitoring Results
Lead, mg/L
s 0.005
0.005 — 0.010*
0.010 — 0.015
> 0.01 5
Copper, mg/L
sO.2
0.2 — 0.8
0.8 — 1.3**
>1.3
* Source water treatment is recommended if the corrosion treatment is at or near optimal and
the lead AL is still exceeded.
If the copper AL is exceeded, source water treatment may be required when corrosion control
treatment is unlikely to reduce copper levels below the AL.
-------
SCREENING OF ALTERNATIVES
treatment is required. In those cases where a
significant amount of lead or copper is present,
then treatment is recommended in order to
reduce the overall lead or copper exposure and
to assist PWSs in meeting the ALs. Table 3-5
also shows that the inclusion of source water
treatment is optional when moderate levels of
metals are found, and unnecessary when very
low levels of either lead or copper are present.
In those cases where systems find elevated
levels of lead or copper, the sources of supply
should be monitored in the raw water and at
various stages within the existing treatment
facilities (if providing treatment currently) to
determine the source of the metals. This
monitoring will also assist in determining
whether the existing treatment is already
generating any removal of lead and copper.
Several types of treatment may be appropri-
ate for removal of source water lead and copper.
EPA specified ion exchange, reverse osmosis,
lime softening, and coagulation/filtration as Best
Available Treatment (BAT) for removal of lead
and copper from source water (USEPA, 1991).
If a PWS is currently providing conventional
coagulation/filtration treatment (whether alum
or ferric coagulation, iron/manganese removal,
or lime softening), then modifying these existing
processes may produce the desired removals for
lead and/or copper. If treatment is not available,
then package treatment units for any of the
above technologies may be installed at individual
wellheads (especially when the elevated metals
are contributed by a small number individual
wells) or at a centralized treatment location. In
the case of elevated copper, elimination of
copper sulfate treatment for those surface water
systems employing it as an herbicide or algicide
may reduce the background levels of copper
without imposing treatment modifications.
States must respond to the recommendations
for source water treatment within six months
of receiving the submittals from PWSs. If
required, PWSs have 24 months to install source
water treatment once approved by the State.
Source water treatment would be installed, then,
six months in advance of corrosion control
treatment for medium PWSs and 12 months in
advance of corrosion control treatment for small
PWSs. Follow-up monitoring would not be
required until after all treatment is in place, i.e.,
after corrosion control treatment has been
installed.
3.3.5 Preparing Recommenda-
tions for Optimal Treatment.
Small and medium-size PWSs must submit
treatment recommendations to the State within
six months of exceeding an AL during initial
monitoring. To assist in preparing the recommen-
dations, a checklist (Table 3-6) has been
developed summarizing the steps of a desk-top
evaluation and key findings. More detailed data
and discussion regarding the findings of a desk-
top evaluation can be provided in the short form,
denoted as Form 141-C, at the end of this
chapter. Thus, the checklist (Table 3-6) provides
the State with a "map" of the evaluation process
and considerations involved in the desk-top
procedures employed by a PWS, while Form
141-C presents the State with the findings from
the desk-top evaluation. Small and medium
PWSs may choose to submit the completed
checklist and Form 141-C to the State for
purposes of recommending optimal treatment,
provided that sufficient documentation is
available should the State require additional
information during the recommendation review
period.
3-35
-------
SCREENING OF ALTERNATIVES
3.4 Case Studies
The following case studies illustrate the
assessment of source water and corrosion control
treatment for PWSs through a desk-top
evaluation. Special conditions and considerations
have also been shown to assist PWSs and States
in addressing the site-specific nature of corrosion
control treatment decisions.
3.4.1 Softening Groundivater
Supply (Single Source).
The Kashton County Water District
(KCWD), a medium-size system, found
excessive lead levels (90%Pb-TAP = 22 //g/L)
but low copper levels ((90% Cu-TAP = 0.6
mg/L) during the initial monitoring period for
the LCR. Using the checklist presented in Table
3-6, KCWD initiated a desk-top evaluation to
determine optimal treatment per the LCR
requirements. The first step taken was to monitor
each of the five wells servicing the lime
softening plant operated by KCWD. No lead
or copper was detected in the source water sam-
ples, ruling out the need for source water
treatment. The recommended treatment must
therefore focus on corrosion control alternatives.
Existing water quality data was reviewed,
generating average water quality parameter
values, estimates of lead and copper solubility,
and calculated values for CCPP. Figure 3-8
presents the treatment scheme and resultant
water quality data gathered by KCWD. The
water quality parameter monitoring conducted
within the distribution system showed no major
changes in water quality characteristics once
the finished water entered the distribution
system. Based on Figure 3-7, all corrosion
control treatment alternatives are possible for
KCWD except the use of orthophosphate since
the finished water pH is above 8.
KCWD has never investigated corrosion
control treatment in the past, but has noted
occasional red water complaints and some
tuberculation of unlined cast iron pipes when
replaced. The supervisor of the lime softening
plant had spoken with another PWS operator
also performing lime softening about their
experiences with polyphosphate inhibitors. The
other community successfully eliminated red
water complaints with the use of polyphosphates,
but also experienced elevated lead levels during
their initial monitoring period.
An evaluation of the constraints acting on
KCWD revealed only one known adverse impact:
disinfection byproducts. The current TTHM
levels are 75 yUg/L on average, and increasing
the final pH to 9.0 or above would cause this
level to increase even further.
Since phosphate inhibitors were eliminated
from further consideration, three treatment
alternatives remained: pH/alkalinity adjustment;
calcium adjustment; and silicate inhibitor
addition. Due to the solubility relationships, little
benefit or theoretical reductions in lead or copper
could be achieved by altering the pH and/or
alkalinity of the existing supply. It would require
either a pH greater than 9.0, which is not
feasible due to TTHM concerns, or increased
alkalinity removal during softening which would
be difficult to achieve. Therefore, pH/alkalinity
adjustment was eliminated as a feasible option.
To evaluate calcium adjustment, a CCPP
of 8.0 mg/L CaCO3 was selected as an initial
target value since it is higher than the existing
condition, but will most likely not plug the pipes
nearest the plant. To achieve the CCPP goal,
either the pH needs to be increased to 8.8
(keeping the alkalinity and calcium the same)
or the alkalinity must be increased to 102 mg/L
as CaCO3 (keeping the pH and calcium content
the same). Either method of achieving the CCPP
goal is feasible, and this option remains viable.
3-36
-------
SCREENING OF ALTERNATIVES
Table 3-6. Checklist for PWS Desk-Top Evaluations
I. Historical Evidence Review:
a. Determine Initial Water Quality
WQP-POE and WQP-DIS
Pb/Cu-POE
Lead Solubility
Copper Solubility
CCPP Index Value
b. Conduct Prior Corrosion Control Investigations
c. Assess Corrosion Activity in the Distribution System for:
Lead and Copper
Iron
A/C Pipe
Other Materials, please specify
Did your utility:
YES NO
d. Review the Literature
e. Identify Comparable PWS Experience with Corrosion
Control Treatment
(If YES, what was the overall performance
of the alternative treatment approaches)
Very Good Good
Poor
Adverse
pH/Alkalinity Adjustment
Calcium Adjustment
Corrosion Inhibitors
Phosphates
Silicates
f. Source Water Treatment Status
Required
Recommended
Optional
Not Necessary
3-37
-------
SCREENING OF ALTERNATIVES
Table 3-6. Checklist for PWS Desk-Top Evaluations (continued)
g. Based on your water quality characteristics, check
the suggested treatment approach (es) per
Figure 3-7 in Volume II of the Guidance Manual.
pH/Alkalinity Adjustment
Calcium Adjustment
Corrosion Inhibitors
Phosphates
Silicates
II. Constraint Definitions
Is the constraint Identified applicable to your system?
(Based on Rankings of 3 or 4 on Form 141-C)
Regulatory Constraints:
SOCs/IOCs
SWTR: Turbidity
Total Colitorms
SWTR/GWTR: Disinfection
D/DBPs
LCR
Radionuclides
Functional Constraints:
Taste and Odor
Wastewater Permit
Aesthetics
Operational
Other
III. Were any treatment approaches eliminated from further
consideration In the desk-top evaluation?
pH/Alkalinity Adjustment
Calcium Adjustment
Corrosion Inhibitors:
Phosphates
Zinc Orthophosphate
Sodium Orthophosphate
Orthophosphate
Poly-ortho-phosphates
Poly phosphates
Silicates
YES
NO
YES
NO
3-38
-------
SCREENING OF ALTERNATIVES
Table 3-6. Checklist for PWS Desk-Top Evaluations (continued)
IV. For each of the feasible treatment alternatives, did your
system evaluate the following In the desk-top evaluation?
Performance
Feasibility
Reliability
Costs
V. What Is the recommended treatment approach?
Source Water Treatment:
Method, specify:
YES
NO
YES
NO
Corrosion Control Treatment:
pH/Alkalinity Adjustment
Calcium Adjustment
Corrosion Inhibitors:
Phosphates
Specify type:
Silicates
Specify type:
3-39
-------
n
i \j
n
2 Vj
Well /^\
Field 3 W
n
4 W
P\
5 {J
CI2
I.
C
0
Aeraior
Ca
u!
Li me So
Ba.
Waier Quilily
pll
Alktlinily, mg/L CtCo 3
Cilcium, mg/L CaCo 3
CCPP.mg/LCaCo3
Lead Solubility, mg/L
Copper Solubility, mg/L
Chemical Feed Dosages
Chlorine. mg/L
Lime, mg/L
Carbon Dioxide, mg/L
O
^
(lening
tin
Raw Waier
7.3
230
240
36.4
0.224
0.08
7.0
CO 2
1
1 ,
Cl
|
2
Fillers
Settled Water
9.3
90
100
14.8
0.084
0.0004
200
^
\^
Storage
Finished Water
8.6
90
100
6.6
0.126
0.002
I.S
80
High
Service
Pumps
Figure 3-8. Lime Softening PWS: Treatment Schematic and Relevant Data
-------
SCREENING OF ALTERNATIVES
The use of silicates for corrosion control
presented some problems for KCWD in
terms of evaluating their usefulness. No
other lime softening plant that they knew
had any experience with silicates, and yet
some promising results had appeared in
the literature for different types of sup-
plies. Although they were not required by
the Lead and Copper Rule to conduct a
treatment study, KCWD decided to do
some experimental testing of silicates.
Both flow-through and static testing
procedures were considered; and after
evaluation of the advantages and
disadvantages of these methods (see
Chapter 4), KCWD decided that the static
testing approach was more suitable for
their personnel to manage.
The maintenance dosage recommended
(10 mg/L SiO2) was bench-tested with the
existing supply and found that it increased
the finished water pH to 8.9. However,
particles were observed in the containers
at the end of the static testing indicating
that calcium was probably with the silicate
and precipitating. Due to concerns with
turbidity problems in the distribution
system, the use of silicates were not
considered reliable.
Based on the above findings, the
recommended treatment, was calcium
adjustment achieved by increasing either
the pH or the alkalinity to meet the CCPP
goal of 8.0 mg/L as CaCO3. The KCWD
checklist for the desk-top evaluation as
presented in Table 3-7 was submitted to
the State for approval of the recommended
treatment in conjunction with a completed
short-form 141-C.
3.4.2 Low Alkalinity, pH, and
Hardness Surface Water
System.
The Town of Mulberry provides potable
water to its 1,200 residents and operates
a small package water treatment plant
(WTP) receiving water from the Lolla River
- a low alkalinity, pH, and hardness
surface water supply. The existing
treatment consists of in-line filtration
using polymer coagulation and final
disinfection with liquid chlorine. Figure 3-9
illustrates the treatment schematic of the
WTP and the relevant water quality
information for the system.
During the initial monitoring period
for lead and copper, excessive lead and
copper levels were found at the targeted
sites. Source water monitoring revealed
high copper concentrations in river
samples, such that source water treatment
was needed. Lead levels in the Lolla River,
however, were below detection and did not
require additional source water removal.
Corrosion control treatment, however, was
still required for Mulberry since the lead
levels exceeded the lead AL.
Reviewing the records of the Town, the
PWS operator discovered that the water
intake at the Lolla River was within a
reach of the river where the County
applied copper sulfate for algae control.
Since the source water monitoring coincid-
ed with the period of copper sulfate
applications, Mulberry requested that the
County use a substitute algicide to reduce
the copper levels. Meanwhile, additional
source water monitoring was performed
by the Town to determine the extent of
copper contamination with the river. After
three months of no copper sulfate
3-41
-------
SCREENING OF ALTERNATIVES
Table 3-7. Checklist for the Kashton County Water District
(KCWD) Desk-Top Evaluations
I. Historical Evidence Review:
a. Determine Initial Water Quality
WQP-POE and WQP-DIS
Pb/Cu-POE
Lead Solubility
Copper Solubility
CCPP Index Value
b. Conduct Prior Corrosion Control Investigations
c. Assess Corrosion Activity in the Distribution System for:
Lead and Copper
Iron
A/C Pipe
Other Materials, please specify
Did your utility:
YES NO
d. Review the Literature
e. Identify Comparable PWS Experience with Corrosion
Control Treatment
(If YES, what was the overall performance
of the alternative treatment approaches)
Very Good Good
Poor
Adverse
pH/Alkalinity Adjustment
Calcium Adjustment
Corrosion Inhibitors
Phosphates
Silicates
•
f. Source Water Treatment Status
Required
Recommended
Optional
Not Necessary
3-42
-------
SCREENING OF ALTERNATIVES
Table 3-7. Checklist for the Kashton County Water District
(KCWD) Desk-Top Evaluations (continued)
g. Based on your water quality characteristics, check
the suggested treatment approach (es) per
Figure 3-7 in Volume II of the Guidance Manual.
pH/Alkalinity Adjustment
Calcium Adjustment
Corrosion Inhibitors
Phosphates
Silicates
II. Constraint Definitions
Is the constraint Identified applicable to your system?
(Based on Rankings of 3 or 4 on Form 141 -C)
YES
NO
Regulatory Constraints:
SOCs/IOCs
SWTR: Turbidity
Total Coliforms
SWTR/GWTR: Disinfection
D/DBPs
LCR
Radionuclides
Functional Constraints:
Taste and Odor
Wastewater Permit
Aesthetics
Operational
Other
III. Were any treatment approaches eliminated from further
consideration In the desk-top evaluation?
pH/Alkalinity Adjustment
Calcium Adjustment
Corrosion Inhibitors:
Phosphates
Zinc Orthophosphate
Sodium Orthophosphate
Orthophosphate
Poly-ortho-phosphates
Poly phosphates
Silicates
YES
NO
3-43
-------
SCREENING OF ALTERNATIVES
Table 3-7. Checklist for the Kashton County Water District
(KCWD) Desk-Top Evaluations (continued)
IV. For each of the feasible treatment alternatives, did your
system evaluate the following In the desk-top evaluation?
Performance
Feasibility
Reliability
Costs
V. What Is the recommended treatment approach?
Source Water Treatment:
Method, specify:
YES
NO
YES
NO
Corrosion Control Treatment:
pH/Alkalinity Adjustment
Calcium Adjustment
Corrosion Inhibitors:
Phosphates
Specify type:
Silicates
Specify type:
3-44
-------
In-Line Sutic Mixer
High
Service
Pumps
Fillers
Storage
Wner Quality
pll
Alkalinity, mg/L CaCo 3
Calcium, mg/L CaCo 3
CCPP, mg/L CaCo 3
Lead Solubility, mg/L
Copper Solubility, mg/L
Chemical Feed Dosages
Alum, mg/L
Chlorine, mg/L
Lime, mg/L
Raw Water
1.2
24
18
-8.5
0.56
0025
IS
Finished Water
7.4
16
, 20
-14.3
0.40
0.013
1.5
4.5
Figure 3-9. Surface Water PWS with Low Alkalinity, pH, and Hardness:
Treatment Schematic and Relevant Data
-------
SCREENING OF ALTERNATIVES
applications, the source water copper levels
were less than 0.02 mg/L copper. The PWS
and the State agreed that additional
source water treatment would not be
necessary as. long as the County did not
apply copper sulfate in the reaches of the
river directly above Mulberry's intake.
Meanwhile, corrosion control treatment
investigations resulted in eliminating
pH/alkalinity adjustment and calcium
adjustment as viable treatment alterna-
tives. Limited storage is available at the
Mulberry package plant, and raising the
pH even slightly would jeopardize the
disinfection performance capability of the
plant. Additionally, the low alkalinity, pH,
and calcium content of the water indicated
that formation of calcium carbonate
deposits would require excessive chemical
treatment. The use of inhibitors was selec-
ted as the approach of choice for the Town.
Phosphate inhibitors were considered
preferable to the silicates given their
proven performance in the available
literature. Since the control of lead was
the targeted objective of corrosion control
treatment, zinc orthophosphate was
recommended as the optimal treatment
approach for Mulberry. Aware of the
possibility for initial disturbances within
the distribution system, Mulberry institut-
ed a flushing program simultaneously with
the startup of the phosphate feed. Higher
dosages were selected to initiate the
system (3.0 mg/L as PO4) with a mainte-
nance dose of 0.6 mg/L as PO4 based on
the experiences of two other communities
that had worked with Mulberry's chemical
supplier.
3.4.3 Multiple Sources of
Supply.
Chincee County, a medium-size system,
is in the process of building a new water
treatment plant which will receive surface
water from the Monohaggen Water Project.
Currently, the County operates several
groundwater wells (See Figure 3-10) which
have been experiencing increasing iron and
manganese levels over the last several
years. The objective of the County is to
provide the base-load of the distribution
system's water demand through the new
WTP and continue to use the well supply
during periods of high demand.
During the initial monitoring program,
the lead and copper ALs were met by the
County. The 90th percentile lead level was
0.012 mg/L and 0.010 mg/L for the first
and second monitoring periods, respective-
ly. The County applied to the State for
reduced monitoring.
While corrosion control treatment is
not required at present, concerns have
been raised about the corrosion control
performance of the distribution system
when the new WTP is brought on line as
the main supply source for the County.
The groundwater supply is well-buffered
and contains a moderate amount of
calcium hardness. The CCPP for the wells
averages 9.2 mg/L as CaCO3. However,
the surface water source is poorly buffered,
contains little hardness, and would have
a moderate to low pH after treatment. The
existing calcium carbonate films may not
be maintained within the distribution
system once supplied by the surface water.
Many residences in the county were
constructed in the early 1900s and still
have lead service lines in place. The
County is concerned that future exceed-
ances of the lead AL could invoke LSL
replacement requirements, an expense that
the County does not want to undertake.
Additionally, the design of the surface
water plant included provisions for
additional chemical feed systems if needed
3-46
-------
Source of Supply
\7
WTP
Distribution
System
Individual
Wells
Figure 3-10. PWS with Multiple Sources of Supply
-------
SCREENING OF ALTERNATIVES
in the future. Since calcium carbonate
films currently exist in the distribution
system, the corrosion control treatment
program for the surface water plant was
oriented toward maintaining the existing
film and providing lead corrosion control
protection in areas where no protective
film existed (such as some home plumbing
environments). The selected treatment was
pH/alkalinity adjustment for lead control
with supplemental calcium added to the
finished water to prevent dissolution of
the calcium carbonate film.
3.4.4 Consecutive Systems.
Fedarry Water Project 4 (the Project)
consists of four communities to which the
Project supplies potable water as shown
in Figure 3-11. Each member community
owns and operates their distribution
system. The Project initiated and had
approved a consolidation agreement
whereby the four communities and the
Project would be considered a single PWS
for purposes of compliance with the LCR.
In the consolidation agreement, corrosion
control treatment would be required if the
monitoring results for the comprehensive
service area exceeded an AL. During initial
monitoring, the lead AL was met but the
copper AL was exceeded with consistently
high copper levels found in Community B.
The corrosion problem appeared to be
limited to this community, since the copper
levels in A, C, and D were below the AL
in all cases.
The source of supply for the Project is
a low alkalinity, pH, and hardness surface
water with similar water quality condi-
tions to that presented in Section 3.4.2.
However, the Project had implemented pH
and alkalinity treatment five years prior
to the promulgation of the LCR to mini-
mize red water complaints occurring within
the comprehensive service area. Since that
time, the member communities had
experienced fewer problems with corrosion-
related complaints. Modification of the
existing corrosion control program was
determined to be needed since the source
water lead and copper levels were below
detection. The Project considered two
approaches to meet the LCR requirements:
modify the existing pH/alkalinity
adjustment treatment at the water treat-
ment plant (WTP) or implement modified
treatment at the master meter location
for Community B.
Based on a review of the water quality
conditions (using Figure 3-7), the most
promising alternative treatments were
pH/alkalinity adjustment or corrosion
inhibitors, either phosphates or silicates.
Since the literature contained mixed
results with the use of phosphates for the
control of copper corrosion, phosphate
inhibitors were eliminated from further
consideration. Based on Figure 3-3, further
pH/alkalinity adjustment does not appear
to present any additional benefit in copper
solubility reduction. For these reasons, the
use of silicates was determined to provide
optimal treatment for controlling copper
in Community B. Since silicate feed
systems can be easily installed and
operated at the storage reservoir located
at the master meter for Community B, the
Project decided to recommend to the State
that silicate inhibitor treatment be in-
stalled at this remote location initially.
If copper corrosion control was improved
and lead levels did not respond adversely,
the Project would consider installing the
silicate treatment system-wide.
3-48
-------
CO
^
(O
Comprehensive
Service
Area
Legend
TPOP • Total Population Served
- Mister Meters
Figure 3-11. Configuration of Consecutive Systems
-------
SCREENING OF ALTERNATIVES
3.5 References
Allen, M.J., et al. 1980. The Occurrence
of Microorganisms in Water Main Encrus-
tations. Journal AWWA. 72(ll):614-625.
AWWA. 1986. Corrosion Control for
Operators. AWWA (Denver, CO).
AWWARF. 1990a. Assessing and Control-
ling Bacterial Regrowth in Distribution
Systems. AWWA (Denver, CO).
AWWARF. 1990b. Chemistry of Corrosion
Inhibitors in Potable Water. AWWA
(Denver, CO).
AWWARF. 1990c. Lead Control Strategies.
AWWA (Denver, CO).
AWWARF/DVGW-Forschungsstelle
Cooperative. 1985. Internal Corrosion of
Water Distribution Systems. AWWARF
(Denver, CO).
Bailey, T.L. Corrosion Control Experiences
at Durham, NC. Proc. 1982 WQTC.
(Nashville, TN).
Benjamin, M.M., et al. 1990. Chemistry
of Corrosion Inhibitors in Potable Water.
AWWA and AWWARF (Denver, CO).
Corrosion Control for Operators. 1986.
AWWA, Manual No. 20232 (Denver, CO).
Donlan, R.M. and Pipes, W.O. 1988.
Selected Drinking Water Characteristics
and Attached Microbial Population Densi-
ty. Journal AWWA. 80(ll):70-76.
Faust, S.D. and Aly, O.M. 1983. Chemistry
of Water Treatment. Ann Arbor Science
Publishers (Stoneham, MA):397-452.
Faust, S.D. and Aly, O.M. 1981. Chemistry
of Natural Waters. Ann Arbor Science
Publishers (Stoneham, MA).
Hanson, H.F., et al. 1987. Deterioration
of Water Quality in Distribution System.
AWWARF and AWWA (Denver, CO).
Herson, D.S., et al. 1991. Association of
Microorganisms with Surfaces in Distribu-
tion Systems. Journal AWWA. 83(7): 103-
106.
Hoehn, R.C. 1991. Personal Communica-
tion. Virginia Polytechnic Institute & State
University (Blacksburg, VA).
Holm, T.R. and Schock, M.R. 1991a.
Potential Effects of Polyphosphate Prod-
ucts on Lead Solubility in Plumbing
Systems. Journal AWWA. 83(7):76-82.
Holm, T.R. and Schock, M.R. 1991b.
Polyphosphate Debate, in Letters. Journal
AWWA. 83(12):10-12.
Huck, P.M. 1990. Measurement of Biode-
gradable Organic Matter and Bacterial
Growth Potential in the Distribution
System. Journal AWWA. 82(7):78-86.
Karalekas, P.C., et al. 1983. Control of
Lead, Copper, and Iron Pipe Corrosion in
Boston. Journal AWWA. 75(2):92-95.
Katsanis, E.P., et al. 1986. Soluble Silicate
Corrosion Inhibitors in Water Systems.
Materials Performance. 25(5): 19-25.
Kirk-Othmer:Encyclopedia of Chemical
Technology - 3rd Ed. 1982. Silicon Com-
pounds. Vol 20:855-880.
Kirmeyer, G.J. and G.S. Logsdon. 1983.
Principles of Internal Corrosion and
Corrosion Monitoring. Journal AWWA.
75(2):78-83.
LeChevallier, M.W., et al. 1990. Disinfect-
ing Biofilms in a Model Distribution
System. Journal AWWA. 82(7):87-99.
3-50
-------
SCREENING OF ALTERNATIVES
LeChevallier, M.W., et al. 1988. Inactiva-
tion of Biofilms Bacteria. Applied and
Environmental Microbiology. 54(10):2492-
2499.
LeChevallier, M.W., et al. 1988. Factors
Promoting Survival of Bacteria in Chlori-
nated Water Supplies. Applied and
Environmental Microbiology. 54(3):649-654.
LeChevallier, M.W., et al. 1987. Examina-
tion and Characterization of Distribution
System Biofilms. Applied and Environmen-
tal Microbiology. 53(12):2714-2724.
Lechner, J.B. 1991. Personal Communi-
cation. Stiles-Kern Div., Met-Pro Corpo-
ration, (Zion, IL).
Lee, R.G. and R.H. Moser. 1988. Lead at
the Tap - Sources and Control: A Survey
of the American Water System. AWW
Service Company, Inc. Internal Report.
Maas, R.P., et al. 1991. A Multi-State
Study of the Effectiveness of Various
Corrosion Inhibitors in Reducing Residen-
tial Lead Levels. Proc. 1991 AWWA
Annual Con/1, (Philadelphia, PA).
Moser, R. H. 1992.
Neff, C.H. 1990. Corrosion and Metal
Leaching. Proc. 1991 AWWA Annual Conf.,
(Philadelphia, PA).
Neff, C.H., et al. 1987. Relationship
Between Water Quality and Corrosion of
Plumbing Materials in Buildings. EPA
Rept. No. EPA/600/S2-87/036. (Cincinnati,
OH).
Robinson, R. B., et al. 1992. Iron and
Manganese Sequestration Facilities Using
Sodium Silicate. Journal AWWA. 84(2):77-
82.
Schock, M. R. 1985. Treatment or Water
Quality Adjustment to Attain MCLs in
Metallic Potable Water Plumbing Systems.
Plumbing Materials and Drinking Water
Quality: Proceedings of a Seminar.
(Cincinnati, OH). EPA 600/9-85-007.
Schock, M.R. 1980. Response of Lead
Solubility to Dissolved Carbonate in
Drinking Water. Journal AWWA. 72(12):-
695-704.
Sheiham, I. and Jackson, P.J. 1981. The
Scientific Basis for Control of Lead in
Drinking Water by Water Treatment.
Journal Inst. Water Engrs. & Scientists.
35(6):491.
Stumm, W. and Morgan, JJ. 1981. Aquatic
Chemistry. John Wiley & Sons (New York,
NY).
Swayze, J. 1983. Corrosion Study at
Carbondale, Illinois. Journal AWWA.
75(2):101-102.
USEPA. 1991. Technologies and Costs for
the Removal of Lead and Copper from
Potable Water Sources. Office of Ground
Water and Drinking Water.
USEPA. 1989. SWTR Guidance Manual.
Office of Ground Water and Drinking
Water.
Wagner, I. Effect of Inhibitors on Corrosion
Rate and Metal Uptake in Drinking Water
Systems. Proc AWWA Seminar on Internal
Corrosion Control Developments and
Research Needs. Los Angeles. AWWA
(Denver, CO).
3-51
-------
Form 141-C Page 1 of 8
Desktop Evaluation Short Form for Small and Medium PWS
Treatment Recommendations
A. PWS General Information:
1. PWS Identification No..
2. Contact person:
Name
Mailing Address
Telephone Fax
3. Population served
4. Person responsible for preparing this form:
Name
Signature
Telephone
B. PWS Technical Information:
Monitoring Results:
Sampling dates: From To
First-Flush Tap Monitoring Results:
Lead:
Minimum concentration = mg/L
Maximum concentration = mg/L
90th percentile = mg/L
Copper:
Minimum concentration = mg/L
Maximum concentration = mg/L
90th percentile = mg/L
Polnt-of-Entry Tap Monitoring Results:
Points of Entry
1234
Lead Concentration in mg/L
Copper Concentration in mg/L
PH:
Temperature, °C:
Alkalinity, mg/L as CaCO3:
Calcium, mg/L as Ca:
Conductivity, ^.mho/cm @ 25°C:
Phosphate, mg/L as P:
Silicate, mg/L as SiO2:
3-52
-------
Form 141-C Page 2 of 8
1. Monitoring Results (continued):
Water Quality Parameter Distribution System Monitoring Results:
Indicate whether field or laboratory measurement.
Field Lab
pH: minimum = maximum =
alkalinity:
minimum = mg/L as CaCO3
maximum = mg/L as CaCO3
temperature:
minimum = °C
maximum = °C
calcium:
minimum = mg/L as Ca
maximum = mg/L as Ca
conductivity: •
minimum = ^mho/cm @ 25°C
maximum = jimho/cm @ 25°C
orthophosphate:
(if phosphate-based inhibitor is used)
minimum = mg/L as P
maximum = mg/L as P
silica:
(if silica-based inhibitor is used)
minimum = mg/L as SiO2
maximum = mg/L as SiO2
2. Existing Conditions:
Is treatment used? yes no
Identify water source(s):
Source No. 1
Source No. 2
Source No. 3
If treatment is used, is more than one source used at a time?
yes no
Identify treatment processes used for each source:
Process No. 1 No. 2 No. 3
Presedimentation
Aeration
Chemical mixing
Flocculation
Sedimentation
Recarbonation
3-53
-------
Form 141-C Page 3 of 8
2. Existing Conditions (continued):
Identify treatment processes used for each source:
Process- No. 1 No. 2 No. 3
2nd Stage mixing
2nd Stage flocculation
2nd Stage sedimentation
Filtration:
Single medium
Dual media
Multi-media
GAC cap on filters
Disinfection:
Chlorine
Chlorine dioxide
Chloramines
Ozone
Granular Activated Carbon
List chemicals normally fed:
List chemicals sometimes fed:
3. Present Corrosion Control Treatment:
None
Inhibitor
Date initiated
Present dose
Range in Residual in Distribution System:
Maximum mg/L Minimum mg/L
Brand name
Type : ,
Has it been effective? Please comment on your experience.
pH/alkalinity adjustment
pH Target
Alkalinity Target mg/L CaCO3
Calcium adjustment
Calcium Target mg/L CaCO3
3-54
-------
Form 141-C
Page 4 of 8
4. Water Quality:
Complete the-table below for typical untreated and treated water quality data.
Copy this form as necessary for additional sources. Include data for each
raw water source, if surface supplies are used, and finished water quality
information (point of entry) from each treatment plant. If wells are used,
water quality information from each well is acceptable but not necessary if
several wells have similar data. For groundwater supplies, include a water
quality summary from each wellfield or grouping of wells with similar quality.
Include available data for the following:
Parameter
pH, units
Alkalinity, mg/L as CaCO3
Conductivity, fimho/cm @ 25°C
Total dissolved solids, mg/L
Calcium, mg/L Ca
Hardness, mg/L as CaCO3
Temperature, °C
Chloride, mg/L
Sulfate, mg/L
Untreated Supply
-
Treated Water
(point of entry)
5. Distribution System:
Does the distribution system contain lead service lines?
yes no
If your system has lead service lines, mark below the approximate number of
lines which can be located from existing records.
None Some Most All
Is the distribution system flushed?
None Some Most All
-
3-55
-------
Form 141-C Page 5 of 8
6. Historical Information:
Is there a history of water quality complaints?
yes no
If yes, then answer the following:
Are the complaints documented? . yes no _
Mark the general category of complaints below. Use:
1 for some complaints in this category
2 for several complaints in this category
3 for severe complains in this category
Categories of complaints:
Taste and odor
Color
Sediment
Other (specify)
Have there been any corrosion control studies?
yes no
If yes, please indicate:
Date(s) of study From To
Study conducted by PWS personnel? yes no
Brief results of study were:
(optional) Study results attached yes no
Were treatment changes recommended? yes no
If yes:
Were treatment changes implemented? yes no
Have corrosion characteristics of the treated water changed? yes no
If yes, how has change been measured?
General observation
Coupons
Frequency of complaints
Other
Briefly indicate, if other:
3-56
-------
Form 141-C
Page 6 of 8
7. Treatment Constraints:
Optimal corrosion control treatment means the corrosion control treatment that
minimizes the lead and copper concentrations at users' taps while insuring that
the treatment does not cause the water system to violate any national primary
drinking water regulations. Please indicate below which constraints to treatment
will apply to your PWS. Use the following code:
1 Some constraint = Potential Impact but Extent Is Uncertain
2 Significant constraint = Other Treatment Modifications Required to
Operate Option
3 Severe constraint = Additional Capital Improvements Required to
Operate Option
4 Very severe constraint = Renders Option Infeasible
Constraint
A. Regulatory
SOCs/IOCs
SWTR: Turbidity
Total Coliforms
SWTR/GWDR: Disinfection
Disinfection Byproducts
Lead and Copper Rule
Radionuclides
B. Functional
Taste & Odor
Wastewater Permit
Aesthetics
Operational
Other
Treatments
pH/Alkalinity
Adjustment
Calcium
Adjustment
Inhibitor
P04
Si
3-57
-------
Form 141-C Page 7 of 8
8. Desktop Evaluation:
Briefly summarize the review of the corrosion control literature that pertains to your
PWS. A report or summary can be appended to this form if preferred.
Were other similar facilities located which are experiencing successful corrosion
control? yes no
If yes, identify their corrosion control treatment method.
None
pH/Alkalinity adjustment
Calcium adjustment
Inhibitor
Phosphate based
Silica based
9. Recommendations:
The corrosion control treatment method being proposed is:
pH/Alkalinity adjustment
Target pH is units
Target alkalinity is mg/L as CaCO3
Calcium adjustment
Target calcium concentration is mg/L Ca
Inhibitor
Phosphate based
Brand name
Target dose mg/L
Target residual mg/L orthophosphate as p
Silica based
Brand name
Target dose mg/L
Target residual mg/L as SiO2
Rationale for the proposed corrosion control treatment is:
Discussed in the enclosed report
Briefly explained below
3-58
-------
Form 141-C Page 8 of 8
List your proposed operating guidelines:
Parameter. Operating Range
Briefly explain why these guidelines were selected.
10. Please provide any additional comments that will assist in determining optimal
corrosion control treatment for your PWS.
3-59
-------
DEMONSTRATION TESTING
Chapter 4.0 —
Corrosion Control
Demonstration Testing
This chapter is intended primarily for
large systems and those small and medi-
um-size systems required by the State to
conduct corrosion control studies. Those
small and medium-size systems that are
not required by the State to conduct
corrosion control studies should proceed
to Chapter 5 after making their treatment
recommendation.
The Rule requires corrosion control
studies to be performed by large PWSs and
those small and medium-size PWSs
required by the State after exceeding the
lead or copper AL. Further, the Rule
defines certain conditions which must be
met by these studies, but it does not
specify the details of those studies. This
chapter provides guidance for and discuss-
es the following aspects of corrosion control
studies: (1) the components necessary to
accomplish the study; (2) the testing
protocols to be used; (3) the procedures
for evaluating data; and (4) the basis for
identifying "optimal" corrosion control
treatment.
The full scope of a corrosion study will
vary system-by-system, and the methods
and procedures used to reach a recommen-
dation will necessarily reflect this level
of site-specificity. Thus, States should
consider the following criteria in the
review of corrosion control studies and
subsequent recommendations:
• Reasonableness of the study design and
findings;
• Technical integrity of the data handling
and analysis procedures; and
• Best professional judgement of the
State regarding the decision-making
criteria used by the PWS in deter-
mining the recommended optimal
corrosion control treatment.
In the course of this chapter, examples
of corrosion control studies will be present-
ed to illustrate the approach and rationale
used in the design, implementation, and
interpretation of findings for corrosion
control studies. A summary of those
studies available in the literature is
provided in Appendix B for additional
resource material available to States,
PWSs, and engineers involved in perform-
ing corrosion control studies.
4.1 Corrosion Study
Organization
The suggested framework for a corro-
sion study as shown in Table 4-1 presents
a logical sequence of steps, organized to
satisfy the requirements and recommenda-
tions outlined. For completing steps 1-3,
a logic diagram was presented in Section
3.3.1 (Figure 3-6) and these steps refer to
the desk-top evaluation discussed at length
in Chapter 3. The result of the desk-top
evaluation for those systems performing
corrosion control studies is the selection
of alternative treatments to be tested.
Small and medium-size PWSs which are
4-1
-------
DEMONSTRATION TESTING
Table 4-1. Organization of the Major Components in
Corrosion Control Studies
Step 1
Step 2
Step 3
Step 4
Steps
Step6
Step 7
Step 8
Step9
Step 10
DOCUMENT HISTORICAL EVIDENCE
• Review PWS Water Quality and Distribution System Characteristics.
• Review PWS Evidence of Corrosion Activity.
• Identify Prior Corrosion Control Experiences and Studies Performed by PWS.
• Identify Prior Corrosion Control Experiences and Studies Performed by Other PWSs with Similar
Characteristics.
EVALUATE SOURCE WATER CONTRIBUTION
• Monitor Pb/Cu-POE.
• Determine Pb/Cu Contributed Due to Corrosion.
• Determine Source Water Treatment Needs.
IDENTIFY CONSTRAINTS
• Compatibility with Water Quality Characteristics (See Figure 3-7).
• Interferences with Other Water Treatment Processes.
• Compatibility of Multiple Sources of Supply.
• Compatibility for Consecutive PWSs.
• Reliability Features for Particular Treatment Approach, Including (1) Process Control; (2) Operational
Redundancy Requirements; and (3) Chemical Supply Integrity and Availability.
• Adverse Impacts on the Community: Commercial Users, Wastewater Operations, Hearth-Care Faalities.
IDENTIFY CORROSION CONTROL TREATMENT PRIORITIES
• Targeted Materials for Corrosion Control Protection.
• Competing Water Quality/Treatment Objectives.
• Secondary Benefits (i.e., Lowering Metal Content in POTW Sludges).
ELIMINATE UNSUITABLE APPROACHES BASED ON FINDINGS FROM STEPS 1-4.
EVALUATE VIABLE ALTERNATIVE TREATMENT APPROACHES:
• Apply Findings from Analogous System Experiences.
• Evaluate Alkalinity and pH Adjustment: Reductions in Theoretical Lead and Copper Solubility.
• Evaluate Inhibitor Addition: Reductions in Theoretical Lead and Copper Solubility.
• Evaluate Calcium Hardness Adjustment: Optimize Calcium Carbonate Precipitation Potential (CCPP).
DECISION:
For any PWS* NOT Required to Perform Testing to Evaluate Alternative
Treatments:
• Formulate Decision Criteria.
• Select Primary Treatment Alternatives.
• Go to Step 9. •
For any PWS Required to Perform Demonstration Testing to Evaluate Alternative
Treatments:
• Formulate Minimum Feasibility Criteria for Alternative Treatments.
• Select the Alternative Treatments to be Included in the Testing Program.
• Establish Decision Criteria to Select Optimal Corrosion Control Treatment
PERFORM CORROSION CONTROL DEMONSTRATION TESTING.
• Design Testing Apparatus.
• Develop Testing Protocols and Procedures.
• Perform Testing Program and Collect Data
• Analyze Data Generating Corrosion Control Performance Results.
• Rank Performance Results by Priority of Corrosion Control Program Goals.
PRELIMINARY COST ESTIMATES AND FACILITY MODIFICATIONS.
• Prepare Preliminary Facility Design.
• Prepare Preliminary Cost Estimate.
DECISION:
Based on the Decision Criteria Established at the Outset, Formulate RECOMMENDED CORROSION
CONTROL TREATMENT AND SUBMIT TO THE STATE.
4-2
-------
DEMONSTRATION TESTING
not required to perform demonstration
testing would select the recommended
treatment based on a desk-top evaluation
as shown in Figure 3-6.
A corrosion control demonstration
testing program is to be formulated and
implemented once alternative treatments
have been selected. This includes such
steps as:
• developing testing protocols, procedures
and frequency for data collection and
evaluation;
• analyzing the resultant data to
generate performance measurements;
and
• determining the performance ranking
of the alternative treatment approaches
on the basis of corrosion control,
secondary treatment impacts, and
process operations and control.
Preliminary design and cost estimates
are to be prepared for the alternative
treatments selected from the desk-top
evaluation. While cost is not directly a
factor in assigning optimal treatment, it
may be decisive when alternative treat-
ments have comparable performance.
Additionally, preliminary design will be
required for the State review process.
The final recommendation of optimal
corrosion control treatment may be based
on the results of a decision criteria matrix
and the ranking of the alternative treat-
ments. The selection process should be
documented and presented to the State.
4.2 Overview of
Demonstration Testing
The evaluation of corrosion control
treatment through demonstration testing
may be accomplished through a variety
of approaches and mechanisms. While
flexibility exists for the actual design of
a testing program, all such endeavors
should clearly define and document the
following elements of the study:
• Testing protocols, including sampling
program design which incorporates
sampling frequency, locations, volume,
parameters, and analytical methods;
and, methods of material exposure such
as flow-through or static environments
under predetermined operating
conditions.
• Materials used to simulate the targeted
piping environment whether lead,
copper, iron, lead soldered joints, brass,
etc;
• Measures of corrosion activity, such as
weight-loss, metal leaching, corrosion
rates, and surface condition inspections;
• Data handling and analysis techniques,
including statistical testing and
guidelines for interpreting the findings;
• Testing of secondary impacts to
determine the potential effects of
alternative treatments on existing PWS
operations and compliance with other
drinking water standards; and
• Quality assurance I quality control
program elements for each element of
the testing program.
The premise underlying demonstration
testing is that alternative treatment
approaches are to be evaluated in terms
of their relative reductions (or increases)
in corrosion activity for specific materials
of concern. Quite often, testing efforts are
used to predict the behavior of various
treatment components. In this respect,
corrosion studies differ. EPA does not
4-3
-------
DEMONSTRATION TESTING
consider the purpose of these studies to
either: (1) predict the levels of lead or
copper in first-draw tap samples from
targeted consumers' homes; or (2) predict
the actual reductions in corrosion activity
within the distribution or home plumbing
systems achievable through corrosion
control treatment. Instead, the purpose
of corrosion control testing is to demon-
strate the relative performance of alterna-
tive treatment approaches and identify
optimal treatment.
In order to determine the relative
performance of alternative treatment
approaches, a control condition must be
clearly defined throughout the testing
program. Some PWSs may find this
problematic due to changing source and
treated water conditions. Systems antici-
pating new sources of supply or new
treatment process for existing sources will
have to address the issue of which treated
water condition to use for its experimental
control. For example, a groundwater
system required to perform demonstration
testing currently provides water treated
only with chlorination prior to its delivery.
As a result of the S WTR, the well water
will be considered as under the influence
of a surface water and coagulation and
filtration treatment will be required. The
anticipated timeframe for completing
construction of the new filtration plant is
mid-1995. Meanwhile, the demonstration
testing program must be concluded by July
1994, prior to the new treated water being
available. In this instance, the PWS should
consider the water quality modifications
anticipated as a result of coagulation and
filtration (i.e., pH and alkalinity reductions
as a result of alum addition) to determine
whether the existing supply would be
adequately representative of future
conditions. For systems introducing new
sources of supply to the distribution
system, the control condition should be
the existing supply and the recommended
treatment should include provisions for
compatible treatment of the new supply
sources. The water delivered under normal
operating conditions should serve as the
control supply source for those PWSs that
experience fluctuations in water quality
either seasonally or due to the alternate
use of wells.
Each PWS will be responsible for the
design and execution of a testing program
which meets its specific overall goals and
objectives.
4.3 Testing Protocols
Testing protocols should be clearly
delineated prior to initiating the demon-
stration testing program. Some time will
need to be allocated for trouble-shooting
the methods and procedures to be used.
Quite often, a trial-and-error process is
required to fully "de-bug" the protocols and
establish a consistent monitoring, operat-
ing, and maintenance plan for the testing
program. Figure 4-1 is included to assist
in logically developing and successfully
completing a corrosion control study. As
can be seen from the diagram, several
different pathways are available enroute
to selecting optimum corrosion control
treatment. Some studies may be designed
to select more than one component, i.e.,
it would not be unusual for both coupons
and pipe inserts to be evaluated within
a single pipe loop, for example. Section
numbers have been added to the diagram
to assist the user in selecting which
4-4
-------
DEMONSTRATION TESTING
Desk-Top Evaluation I
Seoion3J |
i
Organize/Design Study 1
Identify Altematives/Fbrmulaie Criteria 1
Srmon 4.1 |
i
i ii
Flow-Througb Testing 1 Allem
Pipe Loops I Distr
Secnon 4J.1 I
1 . 1
anve Techniques 1
bunon Systems |
- | Coupons | | Inserts | ^^ \ Coupons | Loafs
, , xfaaa
1 1 4.4.3
Trial 1 i ta _ W
SS 1 Sa±oa4A2 1
1
(
Secondary Impacts I
Secaoo4.6 | •
Data Collection I
QA/QC 1
Seoioa4.7 |
|
^ S
\
1
Static Testing 1
Secnon 4 32 \
| Coupons 1 Inserts |
tanstical Analysis 1
AppencfixC 1
Rank Tieaonent Performance I
Select Treanneat Option I
\
Recommend Optimum 1
Tieaonent |
Figure 4-1. Logic Diagram for Corrosion Control
Demonstration Testing
4-5
-------
DEMONSTRATION TESTING
specific sections of the Guidance Manual
should be utilized. It is not necessary, for
instance, to read through Section 4.4.2 if
electrochemical techniques are not used.
4.3.1 Flow-Through Testing
Protocols.
The use of flow-through testing
methods to evaluate corrosion control
performance is preferred since it more
accurately simulates the home plumbing
which is the major source of lead and
copper. The following suggestions on flow-
through testing protocols and methods
should be considered by PWSs in the
design and execution of their
demonstration study.
4.3.1.1 General. Flow-through testing
refers to continuous or cycled flowing
conditions through a testing apparatus
where the solution is not recirculated.
Typically, flow-through testing is used to
describe pipe rig operations where pipe
loops or coupon/insert apparatus are
attached to a central manifold which
distributes the test water to one or more
corrosion testing units, as shown in
Figure 4-2. Detailed descriptions of
standardized pipe rig construction and
implementation may be found in either
the AWWARF Lead Control Strategies
Manual (1990a) or the Army Corps of
Engineers Pipe Loop (CERL, 1989),
including complete material and
fabrication specifications.
The following recommendations regard-
ing the design and implementation of a
flow-through testing program should be
considered when conducting such studies:
• Duration of testing should be between
9 and 15 months to ensure that steady-
state conditions have been achieved
and to capture seasonal effects; the
longer the testing period, the more
confidence a PWS may have in
distinguishing treatment performance.
• A standardized sampling program
should be established before initiating
the testing period to enhance the
analysis of results (See Section 3.3.3).
• Alternative locations for siting the
testing apparatus should be considered:
(a) laboratory or water treatment plant;
(b) remote within the distribution
system; or (c) distribution system in
situ .apparatus. PWSs should avoid
those sites where excess vibration or
humidity may be encountered as these
conditions can interfere with the perfor-;
mance of the testing apparatus.
• Evaluation of the test material surfaces
may be done at the conclusion of each
test run for each material in order to
assess the corrosion behavior of the
treatment alternative. However, this
would require the destruction of the
test materials, which may be
undesirable if future or on-going
operation of the testing equipment is
anticipated.
• When first-draw samples are being
collected, the samples should be drawn
slowly to minimize velocities and
turbulence within the test apparatus.
If air is entrained during sampling,
then the sampling velocity is most
likely to high.
• Water quality parameters, inhibitor
residuals (if appropriate), and metals
(lead and copper) should be sampled
at each pipe loop (first-draw samples)
and the water supply's entrance to each
pipe rig.
4-6
-------
SOURCE
FromWTP
Water Supply
CORROSION CONTROL
TREATMENT
-e-
feed
-e-
How liqualiulioa Chemical Treilmenl
Ruin Buin
/— Ctemiul
feed
-e-
Row liquaUudoa Chemical Tieaimeal
Buin Bum
CORROSION ACTIVITY
TESTING RIGS
Q
EH E
o o
o
Control Rig
Treatment
Allcnialive I
Treatment
Alternative 2
LEOEND
QT) • Coupon Flow Through Cell ® » Flow Measuring Device
O • Pipe Loop, Typically Tubing (&> • Water Quality Monitoring Location
-*- • Row Discharge Point and Monitoring Locations
Figure 4-2. Conceptual Layout of Flow-Through Testing Schemes
-------
DEMONSTRATION TESTING
• To the extent practical, the test condi-
tions should simulate the chemical feed
application points and finished water
quality conditions expected during full-
scale operations.
Flow-through testing methods provide
the following advantages and disadvan-
tages for determining corrosion control
treatment. Several of these have been
discussed by S chock (1990b):
Advantages:
• The corrosion can be measured on the
pipe instead of relying solely on
coupons inserted within the pipe.
• Loops can be placed at various locations
within the distribution system to assist
in determining differing corrosion rates
as water quality changes in the system.
• Multiple loops can be set up in a single
location to determine the corrosion
effects of dissimilar waters.
• The method allows corrosion rates and
treatment techniques to be evaluated
under controlled conditions. Chemical
feed rates can be refined to facilitate
defining optimal corrosion control
treatment.
• Using pipe loops is fairly economical.
• Pipe loop systems can include provi-
sions for intermittent flow which should
simulate "real-world" conditions more
appropriately than static testing
techniques.
Disadvantages:
• Pipe loops need to be operated for
several months before an accurate
comparison between difFering treatment
techniques can be obtained.
• Variations in corrosion rates that occur
during the testing period are not
measured.
• Dynamic testing systems may require
more attention than a static testing
apparatus.
An important feature of this method
of testing is the in-line corrosion control
treatment of the water. This requires some
degree of pretreatment components, such
as chemical feed pumps, flow equalization
basins, flowmeters, and water quality
sampling stations. In some cases, the
operation and control of the corrosion
control treatment component of the test
rig may'be more complicated than operat-
ing and monitoring the pipe rig itself.
Careful attention to the feasibility of
creating a "continuous" supply of treated
water should be addressed prior to any
final testing decisions.
PWSs may be able to utilize the flow-
through testing system on a long-term
basis to assist in understanding the
corrosion response of the distribution
system. Relationships between the flow-
through testing system and the metal
levels found in first-draw tap samples may
be developed in terms of trends in respons-
es to treatment conditions. Calibration of
the flow-through testing system to first-
draw tap samples necessitates concurrent
flow-through testing and first-draw
sampling activity beyond the initial
monitoring period. The AWWARF Pipe
Loop Study (Kawczynski, 1992) [Note:
Expected publication date is early 1993. Available
from AWWARF, Denver, CO] presents a testing
program designed to evaluate the predic-
tive capability of pipe loop systems in
simulating first-draw lead and copper
levels in targeted homes. Continued
4-8
-------
DEMONSTRATION TESTING
utilization of the flow-through testing
systems could provide PWSs with an
additional mechanism to determine the
potential effects of treatment changes on
the full-scale level.
43.1.2 Testing program elements.
The design and operation of a flow-through
testing program requires special consider-
ation of several study components which
are briefly discussed below to assist in
directing PWSs and others performing
such studies. Conducting successful testing
programs is dependent on systems making
the commitment to sufficiently staff the
testing effort, including apparatus design,
fabrication, and operation for the duration
of the testing program. This resource
commitment will be significant. For
example, a one-year testing program could
require allocation of a full-time operator
responsible for fabrication, maintenance,
operation, and sampling; as well as
analytical support for metals and water
quality parameter analyses.
4.3.1.2.1 Pipe rig operation and
fabrication. The required flow rate
through a pipe rig depends on the number
of connections it is supplying. Typically,
between 0.5 and 2 gallons per minute
(gpm) of flow through a single pipe loop
is adequate. If a pipe rig consists of two
or three loops, then at least 1.5 to 6 gpm
of flow is required. Operating a rig at
much higher flow rates could compromise
its feasibility depending on the complexity
of the pretreatment component. For
example, a system feeding 20 mg/L soda
ash (NaaCOg) for alkalinity and pH
adjustment and operating a 6 gpm testing
rig for 16 hours of continuous flow with
8 hours of standing time each day would
require 29-gallons of stock solution (20 mg
NagCCy mL). Daily stock solution require-
ments much beyond 30-gallons becomes
difficult to handle, especially when ex-
tremely concentrated solutions are used.
Additional attention must be given to
the limitations of a slurry feed, such as
lime. Analytical grade hydrated lime with
a purity exceeding 98 percent is recom-
mended for the preparation of stock
solutions (the use of quick lime for testing
rigs is not practical due to the large
amount of impurities and the inability to
properly slake the lime). To avoid plugging
pump heads and tubing, solutions more
concentrated than 10 mg/mL should not
be used. These solutions also require
continuous, rigorous mixing during their
application in order to ensure a consistent
suspension of the slurry solids.
Feeding a corrosion inhibitor with its
typically much lower dosages and feed
rates is less limiting on the design and
operation of the pipe rig system. More
flexibility may exist for systems testing
corrosion inhibitors in terms of the number
of loops, coupons, and inserts a single pipe
rig can accommodate. When evaluating
silicate inhibitors as a treatment alterna-
tive, consideration should be given to
providing ample time and dilution for the
silicates to depolymerize prior to introduc-
tion in a pipe loop system. Silicates in
concentrated solutions primarily exist as
polymers and break down with time to a
monomeric form, which is analogous to
the reversion of polyphosphates to ortho-
phosphates. Therefore, if a silicate is
injected directly into a pipe loop system,
the form of silica present in the pipe loop
would be different from the form of silica
present in the full-scale distribution
4-9
-------
DEMONSTRATION TESTING
system. A design of a pipe loop system
should include some sort of holding tank
to provide adequate detention time and
dilution.
The pipe, loops attached to the rig
should be of sufficient length to permit a
sample to be collected without getting
water from the central pipe. Pipe loops
should be sized to provide at least 15-20
percent additional sample volume to
ensure that interferences from other
materials in the pipe rig are avoided.
Table 4-2 presents the volume of water
contained in various lengths of piping by
interior diameter dimension. The shaded
lines correspond to the minimum length
of pipe of the corresponding diameter (the
last column shaded) to provide at least
15 percent additional volume in the pipe
loop for a one-liter sample. Standard
plumbing materials should be used for the
pipe loop tubing, and all materials used
for each rig should be obtained from the
same lot of piping. For example, if copper
piping loops are to be used in three
different pipe rigs, evaluating three
different treatments, then all of the copper
used in each rig should be purchased at
the same time from the same lot of the
manufacturer. Variability in the testing
results due to differences in materials can
be minimized in this fashion.
For copper loops with lead soldered
joints, fabrication of all of the loops should
be done by the same person and at one
time (do not fabricate one set of loops and
then wait several weeks or months before
fabricating the next set). Additionally, the
solder should come from the same spool.
After soldering, the piping should be
flushed prior to starting the testing
program to remove any loose debris.
In constructing the pipe rig, plastic
materials are recommended for all parts
that would be in contact with the water
except the pipe loops. The use of brass
materials should be avoided due to their
ability to leach lead and copper into the
test water, thereby cross-contaminating
the samples and invalidating the test
results.
During the startup of the testing
program, all pipe loops and the pipe
manifold should be flushed to remove any
material debris attached to the interior
walls of the piping. Flushing should be
performed using the control water. The
pipe loops should be flushed after fabrica-
tion but prior to attachment to the mani-
fold. The complete pipe rig (manifold plus
loops) may then be flushed while trouble-
shooting the apparatus for leaks and the
performance of equipment such as flow-
meters, timers, valves, and pumps.
Some PWSs may want to incorporate
pre-conditioning of the pipe loops into the
testing program. Pre-conditioning consists
of using control water for all pipe rigs until
all pipe loops achieve steady-state corro-
sion activity. The alternative test waters
would then be introduced into the pipe
loops for their respective pipe rig system.
The relative performance of the control
and alternative test conditions would be
assessed in the same manner as those
testing programs which did not pre-
condition the test loops with control water.
It is not known whether this step would
provide PWSs with any greater accuracy
in the evaluation of corrosion control
performance, or whether it would reduce
or increase the required testing duration.
4-10
-------
DEMONSTRATION TESTING
Table 4-2. Pipe Volume by Tubing Length and Diameter
Pipe Volume Table (Volumes
Pipe
Length
(Feet)
1
2
3
4
5
6
7
a
9
10
11
12
13
14
15
16
17
18
19
20
25
30
35
40
45
50
55
60
Listed in
Liters)
Pipe Diameter (Inches)
3/8
0.02
0.04
0.07
0.09
0.11
0.13
0.15
0.17
0.20
0.22
0.24
0.26
0.28
0.30
0.33
0.35
0.37
0.39
0.41
0.43
0.54
0.65
0.76
0.87
0.98
1.09
1.19
1.30
1/2
0.04
0.08
0.12
0.15
0.19
0.23
0.27
0.31
0.35
0.39
0.42
0.46
0.50
0.54
0.58
0.62
0.66
0.69
0.73
0,77
0.97
1.16
1.35
1.54
1.74
1.93
2.12
2.32
5/8
0.06
0.12
0.18
0.24
0.30
0.36
0.42
0.48
0.54
0.60
0.66
0.72
0.78
0.84
0.90
0.97
1.03
1.09
1.15
1.21
1.51
1.81
2.11
2.41
2.71
3.02
3.32
3.62
3/4
0.09
0.17
0.26
0.35
0.43
0.52
, 0.61
0,69
0.78
0.87
0.96
1.04
1.13
1.22
1.30
1.39
1.48
1.56
1.65
1.74
2.17
2.61
3.04
3.47
3.91
4.34
4.78
5.21
1
0.15
0.31
0.46
0.62
0.77
0.93
1 .08
1.24
1.39
1.54
1.70
1.85
2.01
2.16
2.32
2.47
2.63
2.78
2.93
3.09
3.86
4.63
5.40
6.18
6.95
7.72
8.49
9.27
1-1/4
0.24
0.48
0.72
0.97
1.21
1.45
1.69
1.93
2.17
2.41
2.65
2.90
3.14
3.38
3.62
3.86
4.10
4.34
4.58
4.83
6.03
7.24
8.44
9.65
10.86
12.06
13.27
14.48
Notes: 1. Volumes can be added together for pipe lengths not listed.
2. Liters can be converted to gallons by dividing by 3.785.
4-11
-------
DEMONSTRATION TESTING
In order to collect first-draw sample,
the pipe rigs must be operated in a cyclical
fashion with water running off and on,
permitting a standing time of six-eight
hours for the sampling program. The on/off
cycles used by a PWS should be consistent
throughout the testing program's duration
and for each pipe rig under evaluation.
Timers may be installed to control the
operating cycle of the testing program, or
manual operations may be used.
The water entering the pipe rigs should
be treated per the operation of the PWS
facility. The presence of a disinfectant
residual, however, entering the pipe rig
may not ensure the absence of biological
growth within the testing system. Partici-
pants in the AWWARF Pipe Loop Study
(Kawczynski, 1992) noted significant
growth of heterotrophic plate count (HPC)
bacteria at the sample taps in the pipe
rigs. To reduce the biological growth, the
taps were removed, soaked in a concen-
trated chlorine solution, and then rinsed
prior to being re-attached. Even though
the pipe loops and/or manifold may become
seeded with bacteria, they should not be
superchlorinated or receive excessive
dosages of disinfectant as this could affect
the steady-state corrosion behavior of the
pipe loops.
4.3.1.2.2 Test monitoring
programs. The sampling program for
testing rigs should include: (1) the metals
being investigated; (2) water quality
parameters denning the treatment process;
(3) chemical feed rates and stock solution
strengths; (4) water flow rate through each
testing apparatus; and (5) sample identifi-
cation criteria such as test run, date,
analyst, time of sampling, sample handling
steps, and location of sample. The
frequency of monitoring for specific
parameters and the method of sample
collection should be defined prior to the
initiation of the testing program. Lead and
copper samples from the pipe loops should
be first-draw sample representing a
standing time between six and eight hours.
For example, first-draw samples may be
collected every two weeks over a 12-month
period for metals and water quality
parameters representative of tap samples.
Daily water quality parameter sampling
and recording of the appropriate chemical
feed and flow rate measurements may be
performed when operating the pipe rig,
even though tap samples are not collected,
in order to document the water quality
conditions to which the test loops are
exposed during the study.
4.3.2 Static Testing Protocols.
Static tests may offer an alternative
to flow-through pipe loops to ascertain the
performance of various treatments with
different piping materials (Frey and Segal,
1991). Static testing generally refers to
"no flow-through" conditions, or batch
testing (for example, jar testing to evaluate
coagulant dosages represents a batch
testing protocol). The most common form
of static testing is immersion testing where
a pipe material, typically a flat coupon,
is immersed into a test solution for a
specified period of time. Corrosion can then
be described by weight-loss, metal
leaching, or electrochemical measurement
techniques. Other static testing methods
include: (1) using a pipe segment of the
desired material, filling it with test water
and measuring the metal pick-up at the
conclusion of a specified holding time; and
4-12
-------
DEMONSTRATION TESTING
(2) recirculation testing where a reservoir
of test water is circulated through pipe
segments or pipe inserts over a period of
time (Note that while water is flowing
through the piping segments, it is the
same "batch" of water which is being
recirculated during the holding time; in
this sense, it represents a static test.).
These methods have not been widely used,
and appropriate test design would be a
function of the overall goals and objectives
of the testing program.
Static testing procedures do not directly
simulate distribution systems. Further-
more, substantial time-savings over flow-
through testing methods are not realized
with this approach. Single short-term
exposure of the metal specimens does not
adequately give results about long-term
corrosion. Data must be collected for at
least nine months before equilibrium
conditions are approached and metal
leaching has stabilized. Several other
critical limitations of static testing are:
• Static testing conditions do not
represent the conditions to which
piping systems are subject during
normal operations. Containers are
typically not pressurized and experi-
mental procedures allow the inter-
mittent exposure of containers and
coupons/inserts to atmospheric drying.
Household plumbing environments
experience on-and-off cycles of flow and
the distribution system piping network
experiences continuous flow-through
conditions.
• Exposing coupons and containers to
atmospheric conditions, disturbing films
on coupons/specimens and containers
during replenishing of the containers,
evaporation, and other bench-scale
limitations will affect the system water
chemistry. Subsequent film formation
and metal leaching may not accurately
reflect the relative effectiveness of
various treatment techniques.
• The variation of test results may
confound a PWS's ability to differen-
tiate treatment performance among the
alternatives tested. Replicate testing
and measurements are important
components to the test design in order
to provide adequate precision and
accuracy.
• Comparability of the test results with
full-scale performance is uncertain
based on existing information. PWSs
may want to place coupons or pipe
inserts within the service area and at
the POE during the testing program.
This would provide a basis of
comparison between the static tests
(control conditions only) and the full-
scale system.
In spite of these disadvantages, some
utilities may find static testing useful to
screen various potential treatments prior
to flow-through testing or full-scale
implementation. Static tests may be used
to evaluate a greater number of treatment
alternatives for a PWS. Time permitting,
this procedure could allow a PWS to
narrow the treatment approaches to a
more limited number for additional flow-
through testing, if required. Since flow-
through testing programs tend to be more
complex and costly, satisfying the
demonstration testing needs of a PWS or
else eliminating inappropriate treatment
alternatives prior to performing flow-
through testing would be advantageous.
To the extent that static testing may
4-13
-------
DEMONSTRATION TESTING
provide such capabilities, it should be
included in the demonstration testing.
As discussed in the previous section
on flow-through testing protocols, the
testing of silicates as a treatment alterna-
tive poses special difficulties. The initial
silicate mixture will likely contain poly-
meric forms which will change over time.
This reversion may be partially mitigated
by pre-mixing the silicate in a separate
container and letting the diluted mixture
age for a day or two prior to using.
4.4 Alternative
Measurement Techniques
The amount of corrosion may be
determined by measuring a number of
physical parameters, including weight-loss,
metal leaching, corrosion rates, or inspec-
tion of surface films and corrosion byprod-
ucts. A summary of each of these methods
is presented below.
4.4.1 Weight-Loss
Measurement Techniques.
Gravimetric analysis, or weight-loss,
is the traditional method of measuring
corrosion in the drinking water industry.
Many PWSs have placed rectangular
coupons or pipe inserts into distribution
system mains and service lines to assess
corrosion within their system. Figures 4-3
and 4-4 illustrate a typical coupon and
pipe insert installation, respectively.
4.4.1.1 Coupons. Rectangular coupons
can be obtained directly from the
manufacturer prepared for installation.
Once installed, they are typically exposed
for a period of no less than 30-days, and
more commonly, for a period of 90 to 180
days. The coupons are then removed,
cleaned, and reweighed using specific
procedures. In many cases, the coupons
can be shipped back to the manufacturer
for final preparation and weighing.
Coupon geometry and materials have
been standardized by ASTM. Flat coupons
typically are made from sheet metal;
however, cast iron and cast bronze coupons
can be prepared from castings. Coupon
sizes should be 13 by 102 by 0.8 milli-
meters (0.5 by 4.0 by 0.032 inches) for all
sheet metals, and 13 by 102 by 4 mm for
cast metals. Other sizes may be used
provided the total surface area is approxi-
mately 258 cm2 (or 4 in2). A 7-mm hole is
punched through the coupon such that its
center is approximately 8-mm from one
end of the coupon. The coupons are then
smoothed and stamped with an identifica-
tion number between the edge and the
mounting hole in order to track the results.
Table 4-3 lists the ASTM material
specifications for coupons by the metal
alloy and its reference number (ASTM
1990, G-l). ASTM has standard protocols
for coupon preparation for weight-loss
experiments with water (ASTM 1990, D-
2688). These protocols can be obtained
directly from ASTM or at most technical
libraries. ASTM references are used
throughout the industry regarding the
application and handling of mild steel,
copper, and galvanized coupons. Tables
4-4 and 4-5 summarize the cleaning
procedures for the coupons after they have
been exposed to the test environment for
the required period of time.
4-14
-------
DEMONSTRATION TESTING
Discharge
2 - 6" Typical
(31-93mm)
4X1/2"
(102 XI 3mm)
Typical Coupon
Pipe Plug
(1.315-or
33.40mm)
Globe Valve
or Gate Valve
Figure 4-3. Typical Coupon Testing Installation
4-15
-------
Pipe Coupon
Hose Barb
Sleeve Headpiece
Acrylic Sleeve
111!
5
o>
] [
) [
.
Ten Coupon Sleeve (20 la)
Wall
Thickness
0.11 cm
4cm
Q
Standard Copper Coupon
lcm 3cm
Wall
Thickness
O.I I cm
Galvanically Coupled Coupon
Figure 4-4. Typical Pipe Coupon Insert Installation
From Wysock et al. 1991
-------
DEMONSTRATION TESTING
Table 4-3. Densities for a Variety of Metals and Alloys
UNS Number
520100
520200
530200
530400
530403
530900
531000
531100
531600
531603
531700
532100
532900
N08330
534700
541000
543000
544600
550200
Stainless Steels
Type 201
Type 202
Type 302
Type 304
Type 304L
Type 309
Type 310
Type 311
Type 316
Type 316L
Type 317
Type 321
Type 329
Type 330
Type 347
Type 410
Type 430
Type 446
Type 502
Density
(g/enn
7.94
7.94
7.94
7.94
7.94
7.98
7.98
7.98
7.98
7.98
7.98
7.94
7.98
7.98
8.03
7.70
7.72
7.65
7.82
Copper and Copper Alloys - Brass & Bronze
C38600
C23000
C26000
C44300, 44400, 44500
Copper
Red brass 230
Cartridge brass 260
Admiralty 443, 444, 445
8.94
8.75
8.52
8.52
Aluminum Alloys
C68700
C22000
C60800
*
*
*
L53305-53405
L5XXXX
Aluminum brass 687
Commericial bronze 220
Aluminum bronze, 5% 608
Aluminum bronze, 8% 612
Composition M
Composition G
8.33
8.20
8.16
7.78
8.45
8.77
Lead
Antimonial
Chemical
10.80
11.33
Note X1.1 All UNS numbers that include the letter X indicate a series of numbers
under one category.
Note X1 .2 An asterisk indicates that a UNS number not available.
4-17
-------
DEMONSTRATION TESTING
Table 4-4. Chemical Cleaning Procedures for
Removal of Corrosion Products
De»lg-
:: nation
C.2.1
C.2.2
C.2.3
C.2.4
C.2.5
C.3.1
C.3.2
C.3.3
Material
Copper and
Copper Alloys
Iron and Steel
Sotutton
500 mL hydrochloric acid
(HCI, spgr 1.19)
Reagent water to make 1000 mL
4.9 g sodium cyanide (NaCN)
Reagent water to make 1000 mL
100 mL sulfuric acid (H2SO«,
spgr 1.84)
Reagent water to make 1000mL
120 mL surfuric acid (HjSO«,
spgr 1.84)
Reagent water to make 1 000 mL
54 mL sulfuric acid (H2SO4,
spgr 1.84)
Reagent water to make 1000 mL
1000 mL hydrochloric acid (HCI,
sp gr 1.19)
20 g antimony trioxide (Sb2Oj)
50 g stannous chloride (SnCLj)
50 g sodium hydroxide (NaOH)
200 g granulated zinc or zinc chips
Reagent water to make 1000 mL
200 g sodium hydroxide (NaOH)
20 g granulated zinc or zinc chips
Reagent water to make 1000 mL
Tim*
1 to 3 min
1 to 3 min
1 to 3 min
5 to 10s
30 to 60 min
1 to 25 min
30 to 40 min
30 to 40 min
T«mp*r«tur*
20 to 25° C
20 to 25° C
20 to 25° C
20 to 25° C
40 to 50° C
20 to 25° C
80 to 90°C
80 to 90°C
Remark*
Deaeration of solution
with purified nitrogen
will minimize base
metal removal.
Removes copper
suHide corrosion
products that may not
be removed by
hydrochloric acid
treatment (C.2.1).
Remove bulky
corrosion products
before treatment to
minimize copper
redeposftion on
specimen surface.
Removes redepostted
copper resulting from
sulfuric acid
treatment.
Oeaerate solution with
nitrogen. Brushing of
test specimens to
remove corrosion
products followed by
re-immersion for 3 to
4 s is recommended.
Solution should be
vigorously stirred or
specimen should be
brushed. Longer
times may be
required in certain
instances.
Caution should be
exercised in the use
of any zinc dust since
spontaneous ignition
upon exposure to air
can occur.
Caution should be
exercised in the use
of any zinc dust since
spontaneous ignition
upon exposure to air
can occur.
4-18
-------
DEMONSTRATION TESTING
Table 4-4. Chemical Cleaning Procedures for
Removal of Corrosion Products (continued)
Desig-
nation
C.3.4
C.3.5
C.3.6
C.4.1
C.4.2
C.4.3
Material
Iron and Steel
(continued)
Lead and
Lead Alloys
Solution
200 g diammonium citrate
(NH«),HC5He07)
Reagent water to make 1 000 ml
500 mL hydrochloric acid (HCI, sp
gM.19)
3.5 g hexamethylene tetramine
Reagent water to make 1 000 mL
Molten caustic soda (NaOH) with
1.5-2.0 % sodium hydride (NaH)
1 0 mL acetic acid (CH3COOH)
Reagent water to make 1 000 mL
50 g ammonium acetate
(CHjCOONHJ
Reagent water to make 1 000 mL
50 g ammonium acetate
(Cr^COONHJ
Reagent water to make 1000 mL
Time
20 min
10 min
1 to 20 min
5 min
10 min
5 min
T«mp«r«ture
75 to 90° C
20 to 25° C
370° C
Boiling
60 to 70" C
60 to 70° C
Rwnwfcs
Depending upon the
composition of the
corrosion product,
attack of base metal
may occur.
Longer times may be
required in certain
instances.
For details refer
to Technical Informa-
tion Bulleton SP29-
370. "DuPont Sodium
Hydride Descaling
Process Operating
Instructions."
4-19
-------
DEMONSTRATION TESTING
Table 4-5. Electrolytic Cleaning Procedures for
Removal of Corrosion Products
Desig-
nation
E.1.1
E.I. 2
E.2.1
E.3.1
Material
Iron,
Cast Iron,
Steel
Lead and
Lead Alloys
Copper and
Copper Alloys
Solution
75 g sodium hydroxide (NaOH)
25 g sodium sulfate (Na^OJ
75 g sodium carbonate (Na^COJ
Reagent water to make 1000 mL
28 mL sulfuric acid (H2SO4,
sp gr 1.84)
0.5 g inhibitor (diorthotolyl thiourea
or quinoline ethyliodide or
bethanaphthol quinoline)
Reagent water to make 1000 mL
28 mL sulfuric acid (H2SO4,
sp gr 1.84)
0.5 g inhibitor (diorthotolyl thiourea
or quinoline ethyliodide or
betanaphtol quinoline)
Reagent water to make 1000 mL
7.5 g potassium chloride (KCI)
Reagent water to make 1000 mL
Time
20 to 40 min
3 min
3 min
1 to 3 min
Temperature
20 to 25* C
75"C
75" C
20 to 25° C
Remark*
Cathodic treatment
with 100 to 200 A/m2
current density. Use
carbon, platinum or
stainless steel anode.
Cathodic treatment
with 2000 A/m2
current density. Use
carbon, platinum or
lead anode.
Cathodic treatment
with 2000 A/m2
current density. Use
carbon, platinum or
lead anode.
Cathodic treatment
with 100 A/m2 current
density. Use carbon
or platinum anode.
4-20
-------
DEMONSTRATION TESTING
In general, ASTM recommendations
are that coupons should be similar in
composition to the piping within the
system being evaluated. Materials com-
monly found within water distribution
systems include cast iron, ductile iron,
galvanized iron, copper, lead, lead/tin
solder, mild steel, brass, bronze, asbestos-
cement, and plastic. Some of these
materials, such as brass and bronze, may
be present in household plumbing fixtures
and may contain metal impurities such
as lead and zinc.
Several advantages and disadvantages
of coupon testing are summarized below
(Schock, 1990b):
Advantages:
• Provides information on the amount
of material undergoing corrosion for
a specific set of conditions.
• Coupons can be placed within actual
distribution systems.
• The method is relatively inexpensive.
Disadvantages:
• Coupons are generally in the system
for 90 to 120 days before data are
obtained.
• Variations in corrosion rates within the
testing period are not identified.
• Standard coupons may not be
representative of the actual material
within the system undergoing
corrosion.
• The coupon is located within the pipe
section. Thus, it may not accurately
indicate the corrosion occurring at the
pipe wall because the weight-loss may
be due to abrasion not corrosion.
• It is difficult to remove corrosion
products during analysis without
disturbing some of the attached metal.
4.4.1.2 Pipe inserts. The first use of
piping inserts in lieu of rectangular
coupons was developed by T.E. Larson at
the laboratories of the Illinois State Water
Survey (1975), corresponding to ASTM
standard 2688-82 method C. Pipe inserts
consist of a short piece of 1-inch diameter
tubing of the desired material, inserted
into a PVC sleeve and plumbed into a
convenient delivery line or laboratory
testing equipment.
A modified approach using pipe inserts
was presented by Reiber et al. (1988)
which permitted multiple inserts within
a single assembly and allowed replicate
results to be gathered. Additionally, the
methods used by Reiber et al. (1988) use
only mechanical means of insert prepara-
tion and cleaning after exposure which
eliminates chemical treatment and acid
rinses.
4.4.1.3 Calculation of corrosion
rates. The difference between the initial
and final weights of the coupon or pipe
inserts reflects the corrosion activity within
the system. This measurement is in mils
per year of material-loss or gain.
For most applications, the following
equation is sufficiently accurate to esti-
mate the corrosion rate based on coupon
testing results:
P = [H^ - W^A^D] x 1.825 x 106
where, P = corrosion rate, mils per year;
H = original thickness of the coupon,
inches; Wj = original weight of the coupon,
milligrams; W2 = final weight of the
coupon, milligrams; and D = exposure time,
4-21
-------
DEMONSTRATION TESTING
days. In those cases where more precise
control is exerted over all variables
defining the test conditions, the corrosion
rate for a rectangular coupon may be
calculated as follows:
P = 1/[1/H + 1/X + 1/Y]] x [(Wj - W2)/W,D] x 1.825 x 10s
where P = corrosion rate, mils per year;
H * original thickness of the coupon,
inches; X = original length of the coupon,
inches; Y = original width of the coupon,
inches; Wt = original weight of the coupon,
milligrams; W2 = final weight of the
coupon, milligrams; and D = exposure
time, days.
Rates of corrosion using pipe inserts
may be calculated as either milligrams per
square decimeter per day (mdd) or as mils
per year of loss/gain. The method for
calculating corrosion rates in mdd is as
follows:
For Steel and Galvanized Specimens:
mdd = 1180 W/T,
For Copper specimens:
mdd = 1230 W/T,
where W = actual weight loss of the insert,
milligrams; and T = installation time,
days. To convert mdd to mpy, use the
following equation:
mpy = (1.437 mdd/d)
where d = density of the coupon material,
grams/cubic centimeter.
4.4.2 Corrosion Rates.
Electrochemical methods of determining
corrosion rates may also be applied to
drinking water systems. The difference
in electrostatic potential between a test
and reference electrode under applied
current densities can be related to the rate
of corrosion reactions. Linear polarization
techniques have produced good correlation
with weight-loss measurement techniques
(Reiber and Benjamin, 1990).
Figure 4-5 illustrates the polarization
cell utilized by Reiber and Benjamin
(1990). The test electrodes are actual pipe
inserts, and can be of materials of interest
to the PWS. The cell and its instrumenta-
tion can be easily reproduced by PWSs.
The investigators felt that their cell design
simulated pipe flow conditions which
allowed turbulence and scour effects on
the corrosion control to be investigated.
4.4.3 Surface Inspection.
Visual inspection of piping or coupon
surfaces should be performed when
possible in all testing programs. The type
of corrosion action should be noted, i.e.,
pitting, uniform corrosion, scale
characteristics (continuous, patchy, non-
existent), and coloration. Additionally, the
scale, if present, may be scraped from the
surface of the pipe material and chemically
analyzed to determine the key components
contained in the scale. This process does
not identify the specific chemical
compounds composing the scale, but it does
indicate the elements which are part of
the chemical matrix.
Beyond visual inspection and chemical
analyses of scale material, X-ray diffraction
techniques may be employed to further
identify the scale composition and crystalli-
zation characteristics. However, these
methods are extremely expensive, and only
a few laboratories are capable of perform-
ing such tests.
4-22
-------
DEMONSTRATION TESTING
Rotator Cap
Ft Counter Electrode
Fused Glass Seal
To Potennostat
Ag-AgCI Reference Hecirode
Test Electrode Pipe Coupon
Row Out
Row In
To Potentiostat
From: Reiber, RS. and M. Benjamin, 1990.
Figure 4-5. Cross-Section of Polarization Flow Cell
4-23
-------
DEMONSTRATION TESTING
4.5 Data Handling and
Analysis
Data needs are an important consider-
ation in the design of the testing program.
Analytical procedures should be clearly
defined as part of the testing program
development to: (1) describe the behavior
of the testing data; and (2) generate
performance rankings for the alternative
treatments. The most useful approach to
statistically evaluating corrosion control
data involves the application of non-
parametric statistics.
Underlying all statistical measures are
certain, fundamental assumptions regard-
ing the "true" behavior of the data. Those
statistical tests which are most commonly
applied (such as the Student-t Test, chi-
square distribution, difference of means,
analysis of variance) are based on popula-
tions of data that are normally distributed.
A normally-distributed population will
form a bell-shaped curve which is symmet-
rical about the mean, or average, of the
data. Although standard statistical tests
developed for a normal population are
often used for sets of water quality data,
most water quality data do not follow a
normally-distributed curve. The reader is
referred to Appendix C and statistical
reference books for further discussions of
this topic.
Corrosion control testing data tend to
be non-normal, and therefore, conventional
statistical measures may not accurately
describe the behavior of the data, or
reliably generate results which could be
used to rank alternative treatments
without modification. The example pre-
sented in Section 4.9.1 demonstrates the
use of traditional statistical tests using
the skewness coefficient and Student's t
test to compare the performance of alter-
native treatments.
Alternatively, non-parametric analyses
accommodate non-normal conditions, and
can be applied to develop relative perfor-
mance measures for numerous treatments.
The non-parametric tests of importance
are: (1) the Wilcoxon test or U-test which
can compare the results of two conditions
to determine whether they behave simi-
larly (i.e., no difference in corrosion per-
formance can be ascertained) or whether
they behave differently (i.e., one treatment
method produces better corrosion protec-
tion); and (2) the Kruskal-Wallis test, or
H-test, which is the more general case and
can evaluate more than two test condi-
tions. Additional information on the
application of non-parametric statistics
in evaluating demonstration testing data
is provided in Appendix C.
The information to be collected for each
testing run include descriptions of: (1) the
test conditions (run number, treatment
dosages of applied chemicals, water quality
parameters, and date); (2) sampling event
(control versus test apparatus, location
of sampling point, time, and type of
material); and (3) the analytical results
(water quality parameters such as pH,
temperature, alkalinity, hardness, inhibitor
residual, disinfectant residual, lead,
copper, iron, etc. and/or coupon weight
conditions).
The use of spreadsheets or database
management skills with personal comput-
ers will be satisfactory for the analysis of
data from most corrosion studies. Comput-
er software, including statistical analysis
programs, is generally locally available.
4-24
-------
DEMONSTRATION TESTING
4.6 Testing of Secondary
Impacts
Testing of secondary impacts is vital
to the overall study design for optimal
treatment. A primary area of concern for
secondary impacts is how the alternative
corrosion control treatment may be
successfully installed and operated so as
to meet future State-mandated operating
conditions that define compliance with the
Lead and Copper Rule. When pH, alkalini-
ty or calcium adjustment are components
of a treatment alternative, the stability
of these parameters between the point of
adjustment, the POE, and throughout the
distribution system should be ascertained.
Additionally, the likelihood of inhibitors
and key water quality parameters to
remain within acceptable limits in the
distribution system should be investigated.
Compliance with existing and future
drinking water standards must be
achieved after the installation of corrosion
control treatment. Testing to evaluate
these conditions should be included in the
design of the corrosion control study. Of
particular concern may be changes in: (1)
the impact on compliance with the
disinfection performance requirements of
the SWTR and the up-coming Ground
Water Disinfection Rule (GWDR); (2) the
levels and types of disinfection byproducts
(DBFs) that may occur; (3) the occurrence
of positive total coliform events or
inducement of confluent growth in total
analyses due to increases in heterotrophic
plate count bacteria; or (4) disinfectant
residual concentrations.
The impact of alternative treatment
on compliance capability of current and
future regulatory requirements should be
fully explored. Disinfection performance
may be determined by applying the CT
values and calculation procedures pre-
sented in the SWTR Guidance Manual
(USEPA, 1989) and briefly discussed in
Section 3.3.3 of this manual. The regula-
tion of disinfection byproducts will affect
all PWSs regardless of the population
served. Evaluating the effect of corrosion
control treatment alternatives on the
formation of total trihalomethanes
(TTHMs) and other DBFs can be accom-
plished during the testing program by
generating either rate of formation curves
for the key DBFs or simulated distribution
system levels of DBFs. PWSs may refer-
ence the AWWA Standard Methods, 17th
Edition (AWWA, 1989) for an analytical
method to determine the simulated
distribution system total trihalomethane
concentration (SDSTTHM).
4.7 Quality Assurance/
Quality Control Programs
The interpretation of data is founded
upon the assurances that proper quality
assurance and quality control (QA/QC)
procedures were followed during the
testing program. A well-designed QA/QC
program permits the investigator to more
accurately describe the variability intro-
duced into the data by the response of
testing materials to the corrosion control
treatment processes being evaluated alone.
Elements to be included in a QA/QC
program include:
• Sufficient sampling frequency for water
quality parameters during the period
of time when water is flowing to
adequately describe the test conditions
to which the materials where subject
4-25
-------
DEMONSTRATION TESTING
between first-draw samples. For example,
if standing samples are collected each week,
then at least daily sampling for water quality
parameters should be performed for the
treated water supplied to the pipe rig.
• Split samples for metal analyses, especially
when metal test kits are being used. EPA
recommends that at least five percent of the
samples collected be split samples.
• Sample blanks and spikes should be prepared
by someone other than the chemical analyst
to verify routine measurements. A sample
blank and spike should be performed during
each testing period for metals.
• Proper calibration of all analytical
instruments should be performed at the
beginning of each testing period. Chemical
feed and flow rate meters should be fully
calibrated prior to the initiation of testing
and periodically checked during the testing
program.
• Sample handling procedures should follow
those required in the Rule for metals and
water quality parameters. Special care should
be given to the cleaning procedures utilized
for metals sample containers to minimize
cross-contamination between samples.
Each testing program will need to address
its specific QA/QC requirements, and should
delineate these elements at the beginning in
order to prevent the collection of data which
cannot be adequately verified.
4.8 Selecting the Recommended
Treatment Option
The factors affecting the selection of a
treatment technique include:
• Performance of alternative treatments
evaluated during demonstration testing
for mitigating corrosion based on the
prioritization of (a) the targeted materials;
(b) the measurement technique used to
describe corrosion activity (metal
solubility, weight-loss, corrosion rate,
etc); and (c) confidence in the testing
program results (QA/QC and statistical
analysis validity).
• Feasibility of implementing the
alternative corrosion control treatment.
• Reliability features of the alternative
treatment, approaches based on treated
water quality and full-scale operational
characteristics.
• Costs associated with installation and
operation, where alternative treatments
have comparable performance.
A decision matrix including each of the
above factors may be developed and applied
as the basis for selecting the 'optimal'
corrosion control treatment. Weighting
factors which assign relative priorities should
be related to site-specific criteria. In most
cases, however, the performance of the
alternative treatments in reducing lead and/or
copper should receive the greatest priority.
4.8.1 Example of Treatment
Selection.
A large PWS performed a desk-top
evaluation of their system and identified two
alternative treatments for further study by
4-26
-------
DEMONSTRATION TESTING
corrosion testing. Flow-through testing was
performed using pipe rigs with: (1) iron tubing
and copper tubing with lead solder, and (2)
copper, lead, and iron coupon flow-through
cells. Figures 4-6A and 4-6B present results
of corrosion testing in terms of reductions in
metal concentrations for standing samples and
average weight-loss for treatment alternatives
A and B as compared to the existing treatment.
The first-step in developing the final
treatment selection decision matrix is defining
the performance ranking of each treatment
evaluated. The score for the best treatment
option used in this analysis is 7, for second 4,
and for the worst option 0. Given the priorities
of the PWS, the weighting factors used for each
metal were 0.45, 0.40 and 0.15 for lead,
copper, and iron, respectively. Due to the
increased importance in controlling lead and
copper solubility, the weighting factors for
measurement technique were 0.7 and 0.3 for
metal concentration and weight-loss results,
respectively, for lead and copper. For iron,
however, the measurement weighting factor was
0.3 and 0.7 for metal concentration and weight-
loss results, respectively, due to more concerns
about maintenance and repair of iron piping.
Table 4-6 presents the corrosion control
performance matrix with the appropriate
weighting factors. The resultant score indicates
that treatment A provided the best corrosion
control protection, while treatment B provided
the second best and the existing treatment
provided the worst performance. These results
are used in the final treatment selection matrix.
Table 4-7 presents the final treatment
selection matrix for the PWS. A desk-top
evaluation of treatments A and B prior to testing
revealed that these treatment options were
equally feasible. As a result, feasibility of
treatments A and B is not a part of the decision
matrix. By far, the most important factor for
identifying optimal treatment in this case is
treatment performance, shown by setting its
weighting factor at 0.75. The reliability and cost
weighting factors were set at 0.15 and 0.10,
respectively. The reliability of the treatment
options is considered more important than the
costs since compliance will eventually be
determined by the ability of the PWS to
consistently produce finished water which meets
its treatment objectives. The costs of treatment
should be assigned a low weighting factor (here
0.1) to reflect the fact that costs are not directly
relevant to selecting die optimal treatment, exept
in helping to decide between alternative
treatments with comparable performance. Based
on the results of the final treatment selection
decision matrix, Treatment A would be
recommended as optimal corrosion control
treatment.
4.9 Examples of Corrosion
Studies
4.9.1 Flow-Through Testing.
Utility A exceeded the action level for lead
during its first 6-month period of diagnostic
monitoring and initiated a corrosion control
study. The Utility treats water from a surface
supply to provide a treated water with the
following general characteristics:
pH = 7.8 Total hardness = 85 mg/L as CaCO3
SO4 = 4O mg/L Ca hardness = 52 mg/L as CaCO3
Cl =5 mg/L Total alkalinity = 6O mg/L as CaCO3
Na =10 mg/L Total solids = 275 mg/L
As illustrated on Figure 3-7, several avenues
for treatment exist. After conducting a desk top
study and visiting with some other utilities using
similar water sources, Utility A decided to
utilize pipe loops to further define optimal
corrosion control treatment.
4-27
-------
rb
oo
80 —
70
£ 50 -
O
1,0
30 -
20 -
10 —
0 —
47
TRT - A
75
• Copper
• Lttd
I HI -1)
• Iron
Figure 4-6A. Reduction in Metal Concentrations by Treatment Alternatives
-------
DEMONSTRATION TESTING
Table 4-6. Corrosion Control Treatment Performance
Ranking Matrix
Performance Criteria
Treatment Alternative
Cop|
Metal Solubility
>er Lead Iron
Weighting Factors 0.40 0.45 0.15
Treatment A 4
Treatment B 7
Existing 0
Interim Performance Scores
7 5.5
4 5.5
0 0
Treatment A 1.6 3.2 0.8
Treatments 2.8 1.8 0.8
Existing 0.0 0.0 0.0
Measurement Technique
Weighting Factors 0.7 0,7 0.3
Measurement Scores
Treatment A 1.1
2.2 0.2
Treatment B 2.0 1.3 0.2
Existing 0.0 0.0 0.0
Total Score
Weight-Loss
Copper
0.40
7
4
0
2.8
1.6
0.0
0.3
0.8
0.5
0.0
Lead
0.45
7
0
4
3.2
0.0
1.8
0.3
0.9
0.0
0.5
Iron
0.15
4
7
0
0.6
1.1
0.0
0.7
0.4
0.7
0.0
Treatment A 5.8
Treatment B 4.7
Existing 0.5
4-30
-------
(O
TRT-A
TRT - B
-Copper
-Leid
• Iron
Figure 4-6B. Reduction in Coupon Weight-Loss by Treatment Alternatives
-------
DEMONSTRATION TESTING
Table 4-7. Final Corrosion Control Treatment
Selection Matrix
Treatment
Alternative
Weighting Factors
Treatment A
Treatment B
Existing
Corrosion Control
Performance
0.75
7
4
0
Treatment
Reliability
0.15
7
0
4
Estimated
Costs
0.1
0
4
7
Total
1
6.3
3.4
1.3
Three identical pipe loops were con-
structed of copper pipe with lead/tin
soldered connections. Loop 1 represented
a control loop without treatment, Loop 2
used finished water treated with lime
addition, and Loop 3 used finished plant
water with the addition of a phosphate
inhibitor. The target pH for Loop 2 was
8.3 and the alkalinity and final hardness
were allowed to fluctuate to satisfy the
final pH goal. Loop 3 water was treated
by the addition of a proprietary phosphate
inhibitor at a dose calculated to yield
1 mg/L as PO4.
The three loops were run for a period
of 35 weeks until they appeared to stabi-
lize and testing was terminated. Water
flowed through the loops for 16 hours
followed by an 8 hour standing period.
Standing water samples were collected for
lead analysis once per week for the
35-week period. Data from the tests are
given in Table 4-8.
Unless conditioned for an extended
period, new piping materials are likely to
yield higher metals concentrations than
actual household plumbing systems. Yet,
it is extremely difficult to construct pipe
loops with materials removed from house-
hold plumbing systems without disturbing
films and scales present on piping interi-
ors. Results from testing programs,
therefore, are used to select treatment
techniques; and final action levels after
installation of full scale treatment can only
be estimated. In the testing program being
discussed here, finished water from the
treatment facility flowed continuously
through all three loops for four weeks in
order to partially acclimate the pipe rig
before the initiation of the weekly sampling
program.
Parametric statistics were selected to
compare the two treatments with the
control. The data were found to be skewed
and were transformed into the log normal
mode for analysis. This type of transfor-
mation is frequently made when analyzing
water quality data and the procedure is
4-31
-------
DEMONSTRATION TESTING
Table 4-8. Lead Concentrations from Pipe Loop Testing
Week
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
Loopl
Pb, mg/L
0.062
0.078
0.125
0.110
0.175
0.205 .
0.190
0.162
0.078
0.112
0.095
0.132
0.126
0.103
0.115
0.138
0.092
0.100
0.118
0.107
0.068
0.082
0.097
0.112
0.085
0.078
0.060
0.092
0.075
0.087
0.063
0.072
0.068
0.080
0.091
Loop 2
Pb, mg/L
0.130
0.100
0.080
0.095
0.110
0.135
0.108
0.092
0.079
0.085
0.090
0.076
0.079
0.108
0.087
0.072
0.068
0.052
0.097
0.075
0.048
0.072
0.103
0.096
0.072
0.080
0.052
0.058
0.045
0.053
0.060
0.055
0.052
0.048
0.057
Loop 3
Pb, mg/L
0.078
0.102
0.115
0.109
0.126
0.102
0.098
0.075
0.082
0.070
0.068
0.065
0.081
0.073
0.065
0.068
0.072
0.038
0.055
0.062
0.050
0.068
0.076
0.072
0.075
0.080
0.062
0.054
0.058
0.045
0.052
0.068
0.030
0.051
0.042
4-32
-------
DEMONSTRATION TESTING
explained more fully in Appendix C for this
example. The Student's t statistic was used
to compare paired data among the three
loops and the results from these analyses
are reproduced in Table 4-9 from
Appendix C.
Using the entire data set for 35 weeks,
the data in Table 4-9 seem to indicate that
either treatment would be beneficial for
reducing lead concentrations. However,
after reviewing the data, it was noted that
the data had fewer fluctuations during~the
later weeks. These results are reasonable
as the pipes become more acclimated and
the system stabilizes as the testing
program proceeds. Using a data set from
week 25 on, the data were examined once
again. This analysis showed that each
treatment was significantly different when
compared to the control, but there was no
apparent statistical difference between
treatments. Thus, Utility A will examine
other factors such as initial cost, operating
costs, and operating philosophy before
deciding which treatment to implement
for full-scale treatment.
4.9.2 Static Testing.
The City of Starboard, a large PWS,
has a surface water supply with low pH,
alkalinity and hardness levels as shown
in Table 4-10. Based on the desk-top
evaluation, the optimal corrosion control
treatment recommended for further
evaluation was pH/alkalinity adjustment.
The use of inhibitors was eliminated on
the basis of the desk-top evaluation. The
water quality goals selected on the basis
of lead and copper passivation were: pH
7.6-7.8; total alkalinity = 40-45 mg/L
CaCO3; and total hardness ^ 30 mg/L
CaCO3.
Three treatment alternatives were
selected for demonstration testing using
static tests: (1) lime and carbon dioxide;
(2) soda ash and carbon dioxide; and (3)
lime and sodium bicarbonate. The average
chemical feed rates and water quality
characteristics for testing are presented
in Table 4-11.
The demonstration tests used to
evaluate corrosion control performance
consisted of immersion tests with flat
Table 4-9. Calculated Student's t Values
Comparison
Loop 1 and Loop 2
Loop 1 and Loop 3
Loop 2 and Loop 3
t
5.46***
6.98***
2.87**
Notes: All test data transformed to logarithmic values
** Highly significant difference at the 0.01 level
*** Extremely significant difference at the 0.001 level
4-33
-------
DEMONSTRATION TESTING
Table 4-10. Average Raw, Treated, and Finished Water Quality
for the Static Demonstration Tests by the City of Starboard
Water Quality
Parameter
pH
Alkalinity, mg/L
CaCO3
Calcium Hardness,
mg/L CaCO3
Chlorine Residual,
mg/L
Raw Water
7.2
24
18
N/A
Treated
Water
6.7
10
16
0.4
Finished Water
7-4
16
20
1.2
Table 4-11. Average Chemical Feed Rates and Water Quality
Characteristics by Treatment Alternative for the Static
Demonstration Testing Program by the City of Starboard
Treatment
Alternative
CaO/CO2
Na2CO3/CO2
NaHCOg/CaO
Water Duality
Characteristics,
mg/L
THd
42
22
22
Talk
32
27
29
pH
8.0
8.0
8.0
Chemical Feed Rates,
mg/L
CaO
17.4
0
1.4
Na,COs
0
16.8
0
NaHCOj
0
0
31.5
CO,
15.3
7.8
0
4-34
-------
DEMONSTRATION TESTING
metal coupons of iron, lead, and copper.
Figure 4-7 illustrates the experimental set
up for the immersion tests. The testing
program was conducted by suspending four
metal coupons in each of three test jars
and the one control jar for each metal
included in the investigation. The solutions
were maintained for one-week testing
periods, then sampled, drained, and
replaced with fresh solutions. Water
quality parameters were measured daily
in each jar to ensure their relative consis-
tency throughout the testing period. The
pH was adjusted with carbon dioxide or
sodium hydroxide, as needed. Alkalinity
and hardness contents remained very
stable during the week holding period, and
did not require adjustment.
The testing schedule, as presented in
Figure 4-8, included: iron coupon testing
for 4.5 months; lead coupons for 7 months;
and copper coupons for 13 months in order
to achieve stable conditions by the end of
the testing period. Metal leaching data
were collected by sampling the test and
control solutions prior to draining the jars
at the conclusion of each week. The control
and test jars were all treated the same in
terms of the monitoring frequency. This
ensured the integrity of the relative metal
leaching data between control and test
conditions.
Table 4-12 presents the raw data
generated during the testing program in
terms of water quality parameter monitor-
ing and metal leaching. A sample log sheet
for the testing program is presented in
Figure 4-9 to illustrate the data recording
and documentation requirements.
Figures 4-10A and 4-1 OB present the
metal leaching results for copper and lead
in terms of the reduction in total metal
between the test and control jars. A high
degree of variability is evident from the
copper results, while more consistent data
was found for lead. The lime and carbon
dioxide treatment provided the greatest
reduction in copper levels consistently
throughout the testing period. The differ-
ence in the performance between the other
two treatments for copper control is
minimal, and, throughout the majority of
the testing period, both indicated increased
copper corrosion over the existing condi-
tions (i.e., negative reductions as presented
in Figure 4-10A).
Each of the three alternative treat-
ments provided positive reductions in lead
corrosion as shown in Figure 4-10B. Large
variability was observed in the perfor-
mance of soda ash plus carbon dioxide
while lime plus carbon dioxide and sodium
bicarbonate plus carbon dioxide provided
very consistent results. The lime and
carbon dioxide treatment, however,
resulted in lower lead levels with respect
to the control throughout the entire
evaluation period.
Based on these results, the lime and
carbon dioxide treatment was selected as
optimal treatment since it provided the
greatest and most consistent reduction in
corrosion for lead and copper.
4-35
-------
Sel of 4 Lead Coupons
Sampling
Port(iyp)
^
- 3-Oallon
Containers
dyp)
I^S$$S$S$S$$$S^S$$S$S^^^
CONTROL
TREAT-1
LIME + CO 2
TREAT - 2
TREAT - 3
Figure 4-7. Immersion Testing Set-Up
-------
Parameters
Metal Pick-Up Data
Lead
Copper
Iron
Coupon Weight-Loss
Water Quality Parameters
pll
Alkalinity
Calcium
Time - Weeks
8 12 16 20 24 28 32
*
Figure 4-8. Testing Program for City of Starboard Static Demonstration Tests
-------
Table 4-12. Testing Program Haw Data for Water Quality Parameters3 and Metal
Leaching4 Measurements for the Static Demonstration Tests
by the City of Starboard
oo
Test
Week
2
4
6
8
10
12
14
16
18
20
22
24
26
28
30
Alternative
Treatment
Lime + CO2
Lime + C02
Lime + C02
Lime + CO2
Lime + CO2
Lime + CO2
Lime + CO2
Lime + CO2
Lime + CO2
Lime + CO2
Lime + CO2
Lime + CO2
Lime + CO2
Lime + CO2
Lime + CO2
Inital Conditions, mg/L
PH'
6.48
6.52
.6.22
6.38
6.51
6.47
6.44
6.52
6.66
6.56
6.31
6.43
6.54
6.63
6.48
Alk2
5
6
5
5
4
4
6
5
4
4
5
6
4
5
5
Ca*
9
8
9
7
12
10
15
9
8
9
9
7
6
8
8
Cu
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
Pb
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
Final Conditions, mg/L
PH'
7.88
7.98
8.25
8.12
8.05
8.16
8.23
7.92
7.78
8.16
8.08
8.01
8.11
8.04
7.94
Alk2
32
33
35
37
30
28
34
27
29
30
34
32
30
35
29
Caa
44
42
43
44
50
48
'48
42
37
40
42
44
38
39
40
Cu
0.248
0.254
0.361
0.182
0.268
0.177
0.198
0.241
0.220
0.154
0.146
0.132
0.148
0.162
0.127
Pb
0.124
0.135
0.122
0.146
0.138
0.166
0.153
0.142
0.121
0.118
0.092
0.062
0.056
0.078
0.063
The measurement units are standard pH units.
Alkalinity and Calcium measurements are in mg/L CaCO3.
Water Quality Parameter results are the average of measurements taken even/ other day
within the two week test period.
Metal results are the average of three aliquots taken at each sampling event.
-------
DEMONSTRATION TESTING
Figure 4-9. Immersion Testing Data Recording and
Documentation Sheets
Date:
Time:
Analyst:
Test Week:
Test Day:
Water Quality Parameters
Treatment
Control
TrtAlt 1
TrtAlt2
TrtAlt 3
Water Quality Parameters
PH
Rep 1
Rep 2
T
Rep 1
Rep 2
Alk
Rep 1
Rep 2
Ca
Rep 1
Rep 2
Metal teaching Analysis
Immersion Testing
Treatment
Control
TrtAlt 1
TrtAlt 2
TrtAlt 3
Metal Content in mg/L
Lead
Rep 1
Rep 2
Copper
Rep 1
Rep 2
Iron
Rep 1
Rep 2
QA/QC Testing Result*
Treatment
Control
Trt Alt 1
TrtAlt 2
Trt Alt 3
Lead
Blank
Spike
Copper
Blank
Spike
Iron
Blank
Spike
Weight-Loss Mctasurments, mpy
Treatment
Control
Trt Alt 1
Trt Alt 2
Trt Alt 3
Lead
Rep 1
Rep 2
Copper
Rep 1
Rep 2
Iron
Rep 1
Rep 2
4-39
-------
DEMONSTRATION TESTING
Figure 4-9. Immersion Testing Data Recording and
Documentation Sheets (continued)
Date:
Time:
Analyst:
Test Week:
Test Day:
NOTES:
Visual Inspection of Coupons:
Testing/Analytical Procedures:
QA/QC Program:
4-40
-------
60 -
8
z
o
-40
10 12
TIME, WEEKS
Na,CO,
16
18
NaHCOj + CnO
FIGURE 4-1OA. Reductions in Copper Corrosion for Treatment Alternatives
-------
fi
0
14
TEST WEEK
16
Figure 4-1 OB. Reductions in Lead Corrosion for Treatment Alternatives
-------
DEMONSTRATION TESTING
4.10 References
AWWA. 1989. Standard Methods for the Examination
of Water and Wastewater, 17th Edition. AWWA
(Denver, CO).
AWWARF. 1990a. Lead Control Strategies. AWWA
i(Denver, CO).
AWWARF. 1990b. Chemistry of Corrosion Inhibitors
in Potable Water. AWWA (Denver, CO).
AWWARF. 1987. Deterioration of Water Quality in
Distribution Systems. AWWA (Denver, CO).
ASTM. 1987. Std. G 5-87: Standard Reference Test
Method for Making Potentiostatic and
Potentiodynamic Anodic Polarization Measurements.
ASTM (Philadelphia, PA).
ASTM. 1989. Std. G 3-89: Standard Practice for
Conventions Applicable to Electrochemical
Measurements in Corrosion Testing. ASTM
(Philadelphia, PA).
ASTM. 1989. Std. G 16-88. Standard Guide for
Applying Statistics to Analysis of Corrosion Data.
ASTM (Philadelphia, PA).
ASTM. 1990. Std. D 2688-90: Corrosivity of Water
in the Absence of Heat Transfer (Weight Loss
Methods). ASTM (Philadelphia, PA).
ASTM. 1990. Std. G1-90. Recommended Practice
for Preparing, Cleaning, and Evaluating Corrosion
test Specimens. ASTM (Philadelphia, PA).
ASTM. 1991. Std. G 15-90a: Standard Terminology
Relating to Corrosion and Corrosion Testing. ASTM
(Philadelphia, PA).
Breach, R.A., et al. 1991. A Systematic Approach
to Minimizing Lead Levels at Consumers Taps. Proc.
Annual Conf. AWWA (Denver, CO).
Encyclopedia of Chemical Technology. 1982. Silicon
Compounds, 3rd ed., Vol. 20. Wiley (New York).
EPA. 1984. Corrosion Manual for Internal Corrosion
of Water Distribution Systems. EPA 570/9-84-001.
EPA (Washington, D.C.).
EPA. 1987. Evaluation of Silicate and Phosphate
Compounds for Corrosion Control. NT1S (Springfield,
VA).
Frey, M.M. and Segal, B. 1991. "Corrosion Control
Studies: What Can Utilities Do?". Proc. Annual Conf.
AWWA (Denver, CO).
Goold, R.R., et al. 1991. "Enhancing Distribution
System Water Quality at Water District No. 1". Proc.
Annual Conf. AWWA (Denver, CO).
Holm, T.R. and Schock, M.R. 1991. "Potential Effects
of Pdyphosphate Water Treatment Products on Lead
Solubility in Plumbing Systems". Journal AWWA.
83(7):76.
Holm, T.R. and Smothers, S.H. 1990. "Characterizing
the Lead-Complexing Properties of Polyphosphate
Water Treatment Products by Competing-LJgand
Spectrophotometry Using 4-(2-Pyridylazo)Resorcinol".
Intern. J. Environ. Anal. Chem. 41:71.
llges, A. 1991. "Control of Lead and Copper in
Drinking Water Champlain Water District Presentation
Outline". Trans. EPA/AWWA National Workshop on
Corrosion Control. AWWA (Denver, CO).
Kawczynski, E. 1992. Personal Communication.
AWWARF (Denver, CO).
Needeman, H.L, et al. 1990. "The Long-Term Effects
of Exposure to Low Doses of Lead in Childhood:
An 11-Year Follow-up Report". The New England
Jour, of Medicine. 322(2) :83.
Reiber, S.H., and Benjamin, M. 1990. "A Multiple
Set-Point Polarization Technique for Electrochemical
Corrosion Rate Measurement in Potable Water
Distribution Systems". Proc. Annual Conf. AWWA
(Denver, CO).
Reiber, S. et al. 1988. "An Improved Method for
Corrosion-Rate Measurement by Weight Loss".
Journal AWWA 80(11):41.
Reiber, S. 1991. "Galvanic Stimulation of Corrosion
of Lead-Tin Solder-Sweated Joints". Journal AWWA
83(7):83.
Schock, M.R. 1990a. "Causes of Temporal Variability
of Lead in Domestic Plumbing Systems".
Environmental Monitoring and Assessment. 15:59.
Schock, M.R., 1990b. "International Corrosion and
Deposition Control". In, Water Quality and Treatment,
4th Ed. McGraw-Hill (New York, NY).
Wysock, B.M., et al. 1991. "A Study of the Effect of
Municipal Ion Exchange Softening on the Corrosion
of Lead, Copper and Iron in Water Systems". Proc.
Annual Conf. AWWA (Denver, CO).
4-43
-------
OPERATION AND IMPLEMENTATION
Chapter 5.0 —
Full-Scale Operation and
Implementation of Optimal
Corrosion Control Treatment
The purpose of this chapter is to
provide guidance on several aspects of full-
scale implementation of corrosion control
treatment, including the following steps
which PWSs and States may encounter:
• Developing the operating ranges for
optimal treatment after follow-up
. monitoring has been completed.
• Diagnosing problems associated with
the startup of full-scale treatment.
• Identifying the need to modify the
installed treatment to improve corro-
sion control protection.
• Implementing changes in treatment
which may improve the overall perfor-
mance of corrosion control treatment.
5.1 Overview of
Requirements
5.1.1 Installing Optimal
Treatment.
The Rule requires that once treatment
is installed, follow-up monitoring must be
performed by PWSs. At the conclusion of
this monitoring effort, States will review
the results and establish operational
conditions which must be met during all
routine monitoring events for all large
PWSs and those small and medium-size
PWSs that exceed an AL in the follow-up
monitoring. The operating conditions will
consist of minimum, maximum, or ranges
of water quality parameter values which
must be achieved in the potable water
entering and residing in the distribution
system at all times. Additionally, PWSs
will be required to at least report the
chemical dosages applied during the
reporting period.
States will be facing the challenges of:
(1) determining whether the recommended
treatments provided by PWSs are accept-
able, or whether additional action on their
part is required; (2) establishing operating
conditions which adequately define optimal
treatment for each PWS; and (3) determin-
ing compliance for each PWS on the basis
of continual achievement of the site-specific
operating conditions.
PWSs, on the other hand, will be facing
challenges regarding the identification and
execution of optimizing corrosion control
treatment. Many factors act on distribution
and home plumbing systems beyond
treated water quality to cause corrosion
activity increases and decreases. It will
be difficult for many PWSs to properly
assess the ability of treatment changes
to optimize or improve the corrosion
control protection afforded due to the
complex nature of corrosion activity and
the variety of materials targeted for
protection.
A two-year installation and startup
period follows State designation of optimal
corrosion control treatment. Systems
5-1
-------
OPERATION AND IMPLEMENTATION
should conduct additional sampling and
monitoring during this period in order to
optimize their operations prior to conduct-
ing the required follow-up monitoring.
Information needs to be gathered
regarding changes in the active chemical
forms used for corrosion control regardless
of whether precipitation or passivation
techniques are employed. For example,
if a polyphosphate inhibitor is added to
the treated water, it is important to know
the concentrations of both orthophospliate
and polyphosphate within the distribution
system. Important water quality
parameters should be measured at the
entry points to the distribution system and
at various locations within the system
including the extremities. Some of the
locations monitored during the initial
monitoring period should be included.
The primary goal of corrosion control
optimization is to achieve and maintain
compliance with the lead and copper ALs.
However, optimized treatment may exist
even though the ALs are exceeded. Fur-
ther, corrosion control treatment programs
must be coordinated with the requirement
that all other drinking water standards
be met. As noted previously, variations
in pH, calcium, and alkalinity that may
have positive impacts on compliance with
the lead and copper ALs may be detrimen-
tal to meeting other criteria. These
interrelationships are site-specific and
must be defined in each corrosion study.
Additionally, optimization may include
economic factors so that the most cost-
effective means of implementing optimal
corrosion control treatment may be
achieved.
After corrosion control facilities are
operational, optimization should be viewed
as a dynamic, rather than static process,
where ongoing efforts are made to mini-
mize lead and copper concentrations over
time. In addition, future follow-up monitor-
ing, even at reduced frequencies, will
require PWSs to formally review the
effectiveness of their programs on a
periodic basis. As such, corrosion control
is an essential and permanent component
of an overall water treatment program and
PWSs should audit their programs on a
routine basis.
5.1.2 Schedule.
Optimal corrosion control treatment,
if required, must be installed and opera-
tional by the dates presented in Table 5-1.
Large systems will have 30 months from
the time the corrosion control study is
complete until the optimal facilities are
on line. This includes a six-month period
for the State to review the study and
approve the optimal corrosion control
approach. Small and medium systems will
be required to submit recommendations
for optimal corrosion control treatment to
the State within six months of exceeding
an AL. The State may then take one of
the following actions:
• Approve the recommended treatment
approach.
• Disapprove or modify the recommended
treatment approach.
• Require the installation of an alternate
treatment approach.
• Require the purveyor to prepare a
study that identifies the optimal
corrosion control approach for that
system.
Should a small or medium PWS be
required to conduct a corrosion control
5-2
-------
Table 5-1. Key Compliance Dates for Large, Medium, and Small Systems
Ul
6
System Designation
Large
Medium
Medium
(State Designates Treatment)
Small
Small
(State Designates Treatment)
System
Population
>50,000
> 3,300 & a 50,000
> 3,300 & * 50,000
s 3,300
* 3,300
Initiate
Tap
Sampling
Program
01/92
07/92
07/92
07/93
07/93
Complete
Tap
Sampling
Program
01/93
01/93*
01/93*
01/94*
01/94*
Complete
Corrosion
Control
Study
07/94
07/95
N/A
07/96
N/A
Complete
Installation
of Optimal
Facilities
01/97
01/98
07/96
01/99
01/98
* The deadlines for those small and medium-size PWSs that meet the ALs in the first six-month round of initial monitoring
and fail in the second six-month monitoring period would be delayed by six months.
-------
OPERATION AND IMPLEMENTATION
study by the State, then 18 months are
provided for performing the study and an
additional 30 months until treatment must
be installed and on line.
5.2 Full-Scale Operation
of Treatment Alternatives
The development of reasonable operat-
ing criteria by which optimal treatment
may be described is a compliance step
which PWSs and States will be required
to implement. At the completion of follow-
up monitoring, States are expected to
establish the operating ranges or condi-
tions by which PWSs will be judged to be
operating optimal corrosion control
treatment. These conditions establish the
compliance requirements for large PWSs
and those small and medium-size PWSs
that exceed an AL in the follow-up
monitoring. Therefore, it is extremely
important that a balance be achieved
between: (1) accurately defining optimal
treatment goals, and, (2) realistically
setting conditions which are feasible to
be met by full-scale treatment facilities
AT ALL TIMES.
5.2.1 Startup Operations.
The transition between bench scale or
pipe loop studies and full scale operation
is a major one and some difficulties are
to be anticipated. The purchase, installa-
tion, and trouble-shooting of new equip-
ment are considered to be a normal part
of operating a treatment facility and are
not discussed here. These functions are
extremely necessary, however, and should
be performed accordingly.
Startup procedures will vary from
facility to facility depending upon the
chemicals fed, whether chemicals are dry
or liquid, and the type of metering equip-
ment used. In general, more attention
needs to be given the entire system during
the startup period to ascertain that the
proper results are being achieved. Cumula-
tive feed rates for chemicals should
initially be recorded at least once per hour
and never should be recorded less than
once each shift. Metering equipment
should be checked for initial accuracy and
occasionally thereafter. Some new equip-
ment has a tendency to "drift" at first and
it may take a few weeks of operation before
the feed rate is consistent. Routine process
control and monitoring the product in the
treated water are essential elements of
any startup program.
Unfortunately many chemicals can be
expressed in different units and this can
lead to confusion for the unsuspecting
operator. The operator needs to determine
the amount of active ingredient that is to
be fed for corrosion control and then
monitor for that ingredient. For instance,
the results of a corrosion control study may
have determined that a certain inhibitor
should be fed at a dose rate of approxi-
mately 0.2 mg/L as phosphorus, P, while
the supplier identifies his product in terms
of phosphate, PO4. Calcium is sometimes
expressed as calcium (Ca), or as lime
(CaO), or as hydrated lime [CaCOHDJ, or as
calcium carbonate [CaCOJ. The following
information may help to avoid confusion.
• 1 mg/L Ca = 1.40 mg/L Ca as CaO
= 1.85 mg/L Ca as Ca(OH)2
= 2.50 mg/L Ca as CaCO3
• 1 mg/L P = 3.1 mg/L P as PO4
Although the startup of chemical feed
systems for pH, alkalinity, and calcium
5-4
-------
OPERATION AND IMPLEMENTATION
adjustment may require more in-plant
attention to regulate dose rates and final
adjusted water quality; the addition of an
inhibitor is likely to cause more customer
concerns. It is not unusual for inhibitors
to loosen existing corrosion byproducts
when introduced into a distribution system
for the first time. These corrosion materi-
als can then be transported to the user's
tap and water quality complaints regard-
ing red water, dirty water, sediment, color,
or taste and odor may result. Initial doses
may be substantially higher than the
recommended maintenance dose (three to
ten times) in order to acclimate the
distribution system to the inhibitor.
Alternatively, some systems may need to
gradually increase the initial dosages to
the maintenance level to minimize the
adverse effects that may result from
loosening existing corrosion scale or
byproducts in the distribution system.
These doses may be necessary from a few
days to several months in order to accom-
plish the objectives of the corrosion control
program. Additionally, a flushing program
during this time can assist in removing
corrosion byproducts from dead end
locations within the distribution system
and also in ensuring that the inhibitor
moves throughout the entire system.
Because most inhibitors are proprietary
products with unknown formulations, it
can be difficult to chemically monitor the
residual of the inhibitor in the system.
Therefore, physical inspections along with
maintaining customer comment logs are
recommended. It is important to work
closely with a reputable supplier to
minimize customer complaints while
installing full-scale corrosion control
treatment.
5.2.2 Operating Ranges.
This section will discuss some of the
factors that impact an operator's ability
to control chemical feed rates and the
concentration of calcium, carbonate, and
corrosion inhibitors within the distribution
system.
5.2.2.1 Historic operating ranges.
Technically, pH is an exponential function
of the hydrogen ion concentration and
calculating the mean hydrogen ion concen-
tration is the appropriate procedure for
determining average pH levels. Practically,
however, the finished water pH is normally
stable enough to allow the arithmetic mean
to be used without introducing significant
error into operating guidelines. Table 5-2
presents a statistical summary of pH and
alkalinity data taken from a variety of
water treatment plants across the country.
This table shows the annual average
(mean) values and the operating range in
which 90 percent of the daily values fell.
The table also indicates how far the
90 percent range varied from the mean.
A key finding of this work is that site-
specific water quality considerations
influence the operating ranges that can
be achieved. For example, the 90 percent
operating range for pH varied from the
mean by ±0.2 units at several sites to
±1.0 units at another. For alkalinity, the
variance from the mean was about ±10
mg/L CaCO3 for alkalinities below 50 mg/L
CaCO3; ±20 mg/L CaCO3 for alkalinities
between 50 to 100 mg/L CaCO3; and
±30 mg/L CaCO3 for alkalinities over
100 mg/L CaCO3.
The type of chemical used to adjust pH
may also influence daily variations from
the mean. Figure 5-LA shows the pH
5-5
-------
OPERATION AND IMPLEMENTATION
Table 5-2. Operating Ranges for pH and Alkalinity for
10 Water Treatment Plants
Facfltty
Plant A
(Texas)
Plant B
(Texas)
Plant C
(Illinois)
Plant D
(North
Carolina)
Plant E
(North
Carolina)
Plant F
(Minnesota)
Plant G
(Georgia)
Plant H
(North
Carolina)
Plant 1
(Missouri)
Plant J
(Colorado)
PH
(units)
Annual
Average
8.6
8.4
9.0
8.0
7.2
7.1
7.1
7.6
7.8
7.5
90%
Range
8.2-9.0
8.0-8.8
8.6-9.4
7.4-8.6
6.8-7.6
6.9-7.3
6.1-8.1
7.4-7.8
7.6-8.0
6.9-8.1
Variance
±0.4
±0.4
±0.4
±0.6
±0.4
±0.2
±1.0
±0.2
±0.2
±0.6
Alkalinity
(rrtft/l as CaCOS)
Annual
Average
36
29
58
29
25
74
N/A
31
132
37
90%
Range
26-46
19-39
30-86
23-35
19-31
56-92
N/A
23-39
102-162
29-45
Variance
±10
±10
±28
±6
±6
±18
N/A
±8
±30
±8
5-6
-------
OPERATION AND IMPLEMENTATION
frequency distribution for Plant D, where
moderate amounts of hydrated lime, 10 to
40 mg/L as Ca(OH)2, are added to neutral-
ize the acidity of the raw water. As shown,
the finished pH can vary substantially.
In this example, 90 percent of the average
daily pH values were within 0.6 units of
the mean. In contrast, Figure 5-IB shows
the pH frequency distribution for Plant
H treating the same type of water using
sodium hydroxide. At this location, 90
percent of the values were within 0.2 units
of the mean, a tighter range.
In this comparison, sodium hydroxide
seemed to provide for tighter pH control.
This is not always the case, however. For
instance, if the raw water did not have
sufficient alkalinity and buffering capacity,
the use of sodium hydroxide can result in
the same or wider pH variations than for
lime. It is also of interest that for both
plants noted in this example, a phosphate
corrosion inhibitor is used and the pH
variations are not a major concern with
respect to corrosion control. For those
systems, one of the parameters that must
be monitored is the active chemical agent
within the inhibitor, i.e., orthophosphate
or silica. It is not appropriate, for example,
to monitor zinc levels when using a zinc
orthophosphate inhibitor and presume the
orthophosphate concentrations correspond
directly.
Figures 5-2A and 5-2B show the
finished phosphate content of the water
prior to entry into the distribution system.
These data indicate that the daily residual
may vary from 0.2 to 0.5 mg-P/L from the
long-term average. The distribution system
residual will, of course, experience even
greater variations. Because some of the
active chemical ingredient in an inhibitor
will be consumed or deposited within the
system, the inhibitor dose will need to be
larger than the minimum level which
should be maintained throughout the
distribution system.
States should consider the average,
minimum and maximum values for such
water quality parameters based on several
years of operating data, if possible, in
determining the minimum or range to be
established for the water quality parame-
ters. Variations in water quality conditions
entering the distribution system will affect
the effectiveness of corrosion control
treatment. Calcium carbonate can be
somewhat resistant to interruptions in
effective treatment once the deposits have
hardened. However, the protection provid-
ed by carbonate passivation and inhibitor
systems is more vulnerable to disruptions
in treatment or water quality variability
(Elmund, 1992; Lechner, 1991). In these
cases, minimum values rather than
average distribution system conditions are
preferable for pH, alkalinity, orthophos-
phate or silica (whichever parameters
apply).
5.2.2.2 Recommended operating
ranges. Based on the above discussion,
site-specific conditions contribute to the
achievable operating ranges for finished
water quality parameters at each facility.
States are required by the Rule to set
operating conditions which best describe
the "optimal" corrosion control treatment
installed at each facility. Additionally, the
Rule requires that the results from water
quality parameter monitoring at distribu-
tion system points of entry—minimally
required to be performed every two
weeks—and at representative locations
5-7
-------
Ol
100-
7.4 7.6 7.8 8.0 8.2 8.4 8.6 8.8 9.0 9.2
0
6.6 6.8 7.0 7.2
Figure 5-1A. pH Cumulative Frequency Distribution, January to June
Plant D
-------
T T T I i i I
7.1 I 7.3 | 7.5 | 7.7
6.6 6.8 7.0 7.2 7.4 7.6 7.8 8.0 8.2 8.4 8.6 8.8 9.0 9.2
PH .
Figure 5-1B. pH Cumulative Frequency Distribution, January to June
Plant H
-------
Ol
_A
O
07
(\f.t
^ 0>60 ~
a
s0-55-
,§,
ft 0.50 -
^Jj
& 0.45 -
O
33
0.45 -
0.40 -
030
1
•
1
1
1
•
. . • 1. "
• I "
•
|
• *
B
I _ I
•
• I
.
• •
i |
I
; i
i I
• I
n. •
U •' '
• _
• ^
• :
• I " 1
f .1 [
I i
i
14-Scp 23-Dec 02-Apr 11-Jul 19-Oct 27-Jan 07-May
DATE
Figure 5-2A. Finished Phosphates vs. Time
Plant D
-------
Ul
0.90 -
0.80 -
0.70 -
£
a 0.60 -
^
>§ 0.50 -
B
< 0.40 -
CO
O 0.30 -
5
CL,
0.20 -
0.10 -
•
• •
' IB " ^
•
••
?• a 11
• •
•
. ' .'
•
•
i •
• :
\
i . '
•
• •:
•'
. ••
•• • ;i ••
• i
• I
1 1
14-Sep 23-Dec 02-Apr It-Jul 19-Oct 27-Jan 07-May
DATE
Figure 5-2B. Finished Phosphates vs. Time
Plant H
-------
OPERATION AND IMPLEMENTATION
throughout the distribution system — twice
every six months — be reported to the
States to demonstrate compliance with the
established operating conditions. Real-
world variability likely to be encountered
at water treatment plants should be
considered by States in setting these
parameters, To accommodate this feature
of the Rule and still provide reasonable
operating criteria, States may want to
consider establishing one-sided conditions
such as minimum values, for the water
quality parameters used to describe the
optimal treatment process.
For example, assume that a PWS using
carbonate passivation as optimal corro-sion
control treatment initially set water
quality goals for its finished water as
follows:
pH
Alkalinity
= 7.8 - 8.0
= 40 - 50
CaC0
Total Hardness = * 30 mg/L CaCO3
During the year of follow-up monitor-
ing, the PWS monitored the finished water
for each of these parameters four times
a day. The results of this monitoring
resulted in the following range of values
for each water quality parameter:
pH = 7.65 - 8.21
Alkalinity = 37 - 46 mgL CaCO3
Total Hardness =32-56 mgL CaCO3
In this case, the State set as operating
criteria for the PWS a minimum pH value
of 7.6; minimum alkalinity of 35 mg/L
CaCO3; and a minimum hardness of
30 mg/L CaCO3.
For a PWS which practices softening
and tries to establish a calcium carbonate
precipitation, setting water quality criteria
by the pH, alkalinity, and calcium levels
independently may be irrelevant to the
successful formation of calcium carbonate
deposits. States should set minimum water
quality parameters such that the PWS can
achieve finished water pH, alkalinity, and
calcium levels which produce a targeted
range in calcium carbonate precipitation
potential (CCPP) values. Those systems
should include in their reporting data the
calculated CCPP value for each monitoring
event during the reporting period.
For example, a lime softening system
experiences large variability in raw water
calcium hardness and alkalinity at differ-
ent times of the year. The softened water
quality also reflects this variability.
However, the system can control the
finished water pH leaving the plant
through its recarbonation process. After
reviewing the historical treated water
quality, the PWS found that the range for
final calcium hardness and alkalinity each
was 80 - 160 mg CaCOg/L. The PWS
determined that the CCPP target value
was 12 mg CaCOg/L for optimal corrosion
control treatment. Based on these observa-
tions, the PWS calculated the final pH
needed to reach the targeted CCPP level
based on variable calcium and alkalinity
contents as shown below.
The State set the operating guidelines
the system based on the above information
as follows:
Minimum Alkalinity and
Calcium Hardness = 80 mg CaCOg/L
Minimum pH =7.8 units
Average CCPP Value =
When inhibitors are applied as the
method of corrosion control treatment,
minimum finished water inhibitor levels
should be included in the operating criteria
set by the State. In addition, the finished
5-12
-------
OPERATION AND IMPLEMENTATION
Table 5-3. Operating Guidelines for Final pH to
Meet a CCPP Level of 12 mg/L
Treated Water
AfkaBntty
mg CaCOj/L
80
100
120
140
160
Treated Water Calcium Hardness, mg CaCOyt
80
9.2
9.0
8.9
8.7 ^
8.5
120
9.1
8.9
8.6
8.3
8.0
160
9.0
8.7
8.3
8.0
7.8
water pH is often important to the perfor-
mance of the specific corrosion inhibitor,
and thus an operating criteria for pH is
also required. The results of the follow-up
monitoring should be evaluated to deter-
mine the minimum inhibitor dosage
needed to provide an effective residual
inhibitor level throughout the distribution
system. States should recognize that the
introduction of inhibitors into distribution
systems can cause initial disturbances in
the existing corrosion byproducts and
thereby reduce the aesthetic quality of the
delivered water to the consumer. There-
fore, many PWSs may begin inhibitor
treatment with elevated or reduced
inhibitor dosages (as compared to that
recommended for optimal treatment) in
order to cause the least distribution
system upset during this initial
conditioning period.
In reviewing the follow-up data for
inhibitor applications, States should
evaluate the inhibitor-demand exerted
throughout the distribution system. The
inhibitor-demand is the depletion of
inhibitor concentration from the points of
entry to the distribution system to the
locations where water quality parameters
monitoring occurs (or the dose minus the
residual concentration). Since water
systems may either over-dose or under-
dose initially, the minimum dosage
required by the State should be equivalent
to the average inhibitor demand found
during follow-up monitoring plus the
concentration of an effective inhibitor
residual. For example, if a PWS found that
the average orthophosphate demand within
its distribution system during follow-up
sampling was 0.5 mg PO/L, the State may
require a minimum dosage of 0.8 mg POJL
to produce an residual orthophosphate
residual of 0.3 mg POJL throughout the
distribution system.
5.2.3 Diagnostic Sampling.
The LCR has specific monitoring
requirements for initial monitoring, follow-
up monitoring, and reduced monitoring.
Specified periods in which to monitor as
well as certain sampling and testing
procedures all must be followed for tap
5-13
-------
OPERATION AND IMPLEMENTATION
samples and water quality parameters at
points of entry and within the distribution
system. Additional sampling, however,
should be considered by the PWSs. Termed
diagnostic sampling, the purpose of
additional sampling and monitoring would
be to assist in defining problems so that
proper corrective action could be taken.
Gathering additional information early in
the process can be the key to successfully
meeting the lead and copper ALs during
the follow-up monitoring period. These
additional data do not need to be reported
as part of the compliance monitoring.
Sampling procedures do not need to
follow the protocol outlined in the LCR,
but instead can be designed to evaluate
a specific situation. For example, perhaps
a certain sampling location gives abnor-
mally high lead values during the initial
monitoring period. The initial first-draw
tap sample was a one-liter sample as
required by the Rule and was collected at
the kitchen sink. Additional sequential 100
mL samples collected at the same tap
might show an extremely high lead
concentration in the first 100 mL while
subsequent samples had low lead levels.
Data such as these would tend to indicate
a problem with the immediate water
fixture which the homeowner could be
encouraged to replace.
Additional sampling within the
distribution system will almost surely be
necessary if the water chemistry is
changed or an inhibitor is added. In this
case, diagnostic monitoring will help
stabilize treated water quality by
indicating if chemical feed systems are
properly adjusted or if inhibitor
concentrations are penetrating throughout
the distribution system, for example. Using
diagnostic monitoring to assist in
optimizing corrosion control treatment to
meet the ALs can aid a PWS by reducing
future monitoring, eliminating the need
to replace lead service lines, and allowing
the public education program to be
discontinued.
5.2.4 Operational Notes on
Various Treatments.
Achievement of operating goals is
dependent on the raw water quality
variability, process control capabilities,
chemical feed systems employed, and the
equipment used at each PWS. This section
discusses aspects of operating a corrosion
control treatment program successfully
and the various problems which may be
encountered based on the chemical feed
system used for each treatment approach.
5.2.4.1 Calcium carbonate precipi-
tation. With this technique, the
concentration of calcium and carbonate
ions is such that their solubility is
exceeded and calcium carbonate solids
precipitate to form a protective coating on
the interior pipe walls. In essence, the use
of cement-lined metal pipes is an effort
to provide a protective lining even if the
water quality conditions do not favor
calcium carbonate precipitation.
Depending upon water chemistry, it
may be necessary to adjust the calcium,
carbonate, or hydrogen ion content of the
water to form a calcium carbonate film.
Calcium supplementation is usually
achieved by adding hydrated or quick lime;
and carbonate adjustment can be accom-
plished by adding soda ash, sodium
bicarbonate, or carbon dioxide. Hydrogen
ion, or pH, adjustment may be accom-
5-14
-------
OPERATION AND IMPLEMENTATION
plished by adding any of these chemicals
in addition to other bases or acids such
as caustic soda, hydrochloric acid, or
sulfuric acid. It is difficult to establish a
uniform thickness of calcium carbonate
on interior pipe walls throughout the
distribution system. If excessive precipita-
tion occurs in some portions of the system,
a significant reduction in hydraulic
capacity may be experienced.
For large systems, storage silos are
provided for solid chemicals, such as lime
(CaO), and the bins are equipped with
vibrators and compressed air agitators to
reduce clumping and promote the flow of
chemical into gravimetric or volumetric
feeders. The feeders discharge the dry
chemical into solution tanks where it is
dissolved into water. Lime can have a
significant amount of impurities and
solution tanks have provisions for collect-
ing and purging inert particles. In some
situations, it is more economical to use
quicklime, and slake it onsite, rather than
purchase hydrated lime directly from the
supplier. To use quicklime requires
additional equipment to slake the calcium
oxide and remove impurities that are
contained within the material. As a result,
slaking operations are generally used in
larger facilities or lime softening plants
which use more lime where the investment
and operation of such equipment can be
justified.
For smaller plants or those requiring
low to moderate dosages, hydrated lime
is normally used and continuous or semi-
continuous solution tanks are more
appropriate and economical for large
operations. Removal of impurities is still
important with hydrated lime, particularly
when it is fed downstream of filters or in
situations where no filters are used.
Depending upon site-specific design factors,
the solutions can be fed to the water by
gravity, using weirs or rotodip feeders, or
chemical feed pumps.
Lime and soda ash systems require a
high degree of operator attention due to
calcium carbonate plugging of bins, tanks,
pumps, and piping. To reduce the amount
of downtime due to such plugging prob-
lems, bins must be kept dry and provisions
should be made for acid cleaning of the
feed systems. Because of this concern, the
reliability of dry chemical feed equipment
is less than that for liquid systems.
Redundancy of solution tanks and chemical
feed pumps can reduce the likelihood of
extensive treatment interruptions, provid-
ing further assurances of continuous
corrosion control operations.
Where the natural concentrations of
calcium and inorganic carbonate contents
are sufficient but the pH is too low for a
precipitate to form, sodium hydroxide can
be used to increase the pH to the point
that calcium carbonate precipitation will
occur. Sodium hydroxide is normally
delivered as a 50 percent solution and
diluted at the time of delivery to 20 to 30
percent. Dilution is helpful in reducing
crystallization that can occur at tempera-
tures below 50°F. Indoor storage facilities
are normally used even for the diluted
sodium hydroxide solution. If dilution is
practiced, consideration should be given
to ion exchange softening of the dilution
water to prevent calcium carbonate
plugging of the sodium hydroxide feed
system. Employee safety and spill contain-
ment are important design concerns with
sodium hydroxide systems.
5-15
-------
OPERATION AND IMPLEMENTATION
For lime and lime/soda softening
systems, lime and soda ash are added at
various points throughout the treatment
train to remove carbonate and non-
carbonate hardness. The final pH is
adjusted through carbon dioxide or acid
addition to prevent excessive encrustation
of filters and still provide the long-term
accumulation of a calcium carbonate film
within the distribution system pipes. For
turbidity removal plants using aluminum
or iron salts, the location of lime, soda ash,
or bicarbonate feed points should be
carefully considered, balancing the water
quality requirements for coagulation,
filtration, and disinfection performance
with that of corrosion control.
In summary, the choice regarding
which chemical(s) to use and where to
apply them must integrate water chemis-
try, customer acceptance, cost, reliability,
safety, and operator preference issues.
Therefore, there is no chemical that is
right for all locations and representatives
of management, operations, and engineer-
ing should be included in the decision-
making process.
5.2.4.2 Carbonate passivation.
Carbonate passivation is a corrosion
control technique where the pipe materials
are incorporated into a metal/hydroxide/
carbonate film that protects the pipe. This
technique is most suitable for low hardness
and alkalinity waters where the PWS does
not want to drastically alter the water
chemistry, and historic customer accep-
tance of the water, to the point that
calcium carbonate precipitation will occur.
Passivation may be achieved by alkalin-
ity and pH modification using such chemi-
cals as lime, soda ash, sodium bicarbonate,
sodium hydroxide, potassium hydroxide,
and/or carbon dioxide. Consequently, the
same chemicals and feed systems are used
for carbonate passivation as for the
calcium carbonate film technique noted
above.
5.2.4.3 Inhibitors. A wide variety of
specialty chemicals, most of them phos-
phate or silicate based, can be added to
the finished water to reduce corrosion
within the distribution system. One
phosphate-based inhibitor is zinc ortho-
phosphate. With this chemical, it is
suspected that the zinc and phosphate
components are involved in forming a
protective film and providing corrosion
protection. With polyphosphate chemicals,
direct corrosion protection appears to be
minimal except that which may be afforded
by the formation of orthophosphate
constituents within the distribution system
as the polyphosphate reverts to orthophos-
phate. Sodium silicate is another corrosion
inhibitor for which limited performance
data is available regarding the reduction
of lead and copper corrosion activity.
A potential problem with orthophos-
phates is that when they are first added
to a distribution system, previously
corroded material may be released and
cause aesthetic problems with the water.
This is especially true if the treated water
pH is lowered concurrently with the
addition of the inhibitor. Additionally,
polyphosphates generally will exhibit more
of a tendency to remove corrosion
byproducts than orthophosphate formula-
tions. Polyphosphates are sometimes used
to chemically remove tuberculation and
scale. To minimize the potentially negative
customer reaction to such a situation, low
dosages may be used initially and then
5-16
-------
OPERATION AND IMPLEMENTATION
slowly increased to the desired full
strength dose. Alternatively, systems may
initially feed inhibitor doses much higher
than the maintenance level to acclimate
the distribution system when concerns
about releasing excess corrosion products
in the delivered water are not significant.
Properly timed public education and
flushing programs can also be used to
minimize the temporary aesthetic
problems that may occur with ortho-
phosphate addition.
No direct evidence is available indicat-
ing that the introduction of phosphate-
based corrosion inhibitors would foster or
encourage the growth of bacteria in the
distribution system. Instead, the findings
of a number of studies indicate the positive
response of distribution system water
quality to the implementation of effective
corrosion control programs (LeChevalier
et al., 1987, 1988a, 1988b, 1990). Microbio-
logical growth within piping systems
appears to be more strongly linked to their
tendency to grow in conjunction with
corrosion byproducts, such as tubercles,
than the supplementation of nutrients in
the form of phosphates or inorganic
carbonate species (AWWARF, 1990).
Corrosion control programs which reduce
corrosion scale buildup has appeared to
reduce the occurrence of bacteria in
distribution system water samples.
However, the potential impact of any
treatment method on the microbiological
behavior of the distribution system is an
important consideration. Some testing
methods are available for evaluating this
impact.
Although not to be interpreted as
corrosion control, a primary use of poly-
phosphates is to sequester dissolved metal
or cationic constituent—such as calcium,
iron, or manganese—and reduce their abil-
ity to precipitate either in the distribution
system or within the water treatment
plant. In the case of calcium, polyphos-
phates are used in many softening plants
to minimize the encrustation of filter
media by post-precipitation of calcium
carbonate. For iron and manganese control,
polyphosphates can effectively reduce the
aesthetic discoloration caused by these
compounds. This is often a useful and
necessary benefit of their application,
particularly for groundwater systems
which are heavily mineralized and devoid
of oxygen, ideal conditions for iron and
manganese to solubilize. Seasonally high
levels of iron and manganese can also
occur with surface water supplies when
low dissolved oxygen and reducing condi-
tions in upstream reservoirs increase the
concentration of these minerals.
While polyphosphates have demon-
strated limited direct success toward lead
and copper corrosion control, their use at
water treatment facilities will be necessary
in many instances. New orthopolyphos-
phate blends are being produced which
can offer some of the benefits of both uses
to PWSs. These should be considered when
orthophosphate inhibitors are a viable
corrosion control approach, but a polyphos-
phate is also required for other treatment
objectives.
With respect to chemical feed systems,
silicate and phosphate compounds are not
inherently dangerous or corrosive, have
a long shelf life, and are highly soluble in
water. These features allow the use of
relatively small batch tanks and feed
pumps. If sodium silicates are being used
after dilution, the day tanks should be
5-17
-------
OPERATION AND IMPLEMENTATION
sized to fully utilize the solubilized silicate
within 24 hours to ensure effective applica-
tion. Since aging times are not needed and
short disruptions of service can be
tolerated, single tank systems are feasible.
For more reliable service, multiple batch
tanks with automatic switch-over can be
used. Multiple tanks also facilitate
cleaning and maintenance of the system.
Depending upon which chemical is
selected, the batch tanks may contain a
phosphate-rich solution. Since biological
growths can occur in the tanks, provisions
should be made to routinely clean them.
It may also be desirable to provide for
supplemental chlorination of the water in
the batch tanks to reduce biological
growths. Chlorine addition to polyphos-
phate or ortho-polyphosphate solution
water may not be advisable for those
situations where sequestering of iron and
manganese is important. The chlorine will
tend to oxidize these metals, causing some
of them to precipitate before they can be
sequestered by the polyphosphates.
Silicates are not a nutrient and feed
system design and maintenance require-
ments are less than for phosphate-based
inhibitors. In fact, some plants have been
known to mix silicates and fluoride in the
same tank and feed them concurrently
using the same pumps. The sodium
concentration is also a consideration when
sodium silicate is used.
If the PWS desires the inhibitor to also
act as a sequestering agent for iron and
manganese, chemical addition should occur
upstream of the first point where chlorine
is added. If iron and manganese are not
a problem, the inhibitor can be added to
the finished water downstream of chlorina-
tion.
5.2.5 Reliability.
The reliability of the various treatment
approaches to continuously provide
corrosion control protection is not clearly
understood with regard to the home
plumbing environment. Some limited
evidence has indicated that copper
corrosion can recur when carbonate
passivation treatment is interrupted for
very brief periods of time (on the order of
a couple of days) (Elmund, 1992). It has
also been reported that phosphate-based
inhibitors can support longer interruptions
prior to reversion of corrosion activity
(30 or more days) (Lechner, 1991). How-
ever, the Lead and Copper Rule requires
that corrosion control treatment be
operated at all times. The water quality
monitoring is required every two weeks
to demonstrate that treatment is continu-
ously provided. Therefore, treatment
interruptions due to maintenance, chemical
inventory problems, and/or equipment and
instrumentation failures must be
minimized regardless of the treatment
approach selected.
While the goal for system reliability
is operational functioning 100 percent of
the time, realistic performance may be less
than this goal. The design of the full-scale
system, however, can incorporate redun-
dancy and/or alarm features which can
assist PWSs in maintaining continuous
operations.
5.2.6 Instrumentation and
Control.
For calcium carbonate precipitation and
carbonate passivation, pH is typically used
as a real-time instrumentation and control
parameter. While pH is an indirect
5-18
-------
OPERATION AND IMPLEMENTATION
measurement of calcium carbonate precipi-
tation or carbonate passivation, it is a
proven, direct loop control parameter and
with experience, the operators can reliably
produce finished water with the pH within
the range that works best for a particular
system. Although specific ion probes are
available for calcium, carbonate is most
commonly measured by a titration process.
While calcium content or pH can be used
for on-line monitoring and chemical feed
control, they may be correlated to such
corrosion monitors as lead and copper
levels in tap sampling programs, or test
results from coupons or pipe inserts from
the distribution system.
For inhibitors, the treatment goals
typically concern only the applied dosage.
Therefore, pacing chemical feed according
to flow and routinely checking for system
residuals of the inhibitor may be sufficient
operational control. Through the corrosion
control study, PWSs may determine the
residual concentration that minimizes
corrosion, cost, and undesirable side
effects. If a higher level of operational
control is desired, it may be possible to
tie the inhibitor feed pump to a corrosion
activity monitor utilizing electronic
measurement techniques with settings
predetermined for optimal corrosion
control conditions.
5.2.7 Troubleshooting.
The purity of the chemicals used for
corrosion control treatment can vary,
especially for hydrated lime and quicklime.
These chemicals will contain inert material
which must be removed through a
de-gritting process, and an allowance for
inert material must be made when
establishing chemical feed rates. For
example, the amount of impurities in
quicklime can vary from 4 to 30 percent,
with a typical value of 10 percent for
municipal grade lime. If a PWS determines
that the quicklime they are using contains
90 percent calcium oxide (i.e. 10 percent
impurities), an additional 11 percent of
the bulk chemical must be added to
achieve the desired lime dose (100/90 =
1.11, i.e., 11 percent additional). The purity
factors are better for hydrated lime which
may only contain 1 to 5 percent impurities,
and better still for soda ash which may
only contain 1 to 2 percent impurities.
Purity factors are less of a concern with
sodium hydroxide and corrosion inhibitors.
There are many proprietary corrosion
inhibitors, particularly for the phosphate
group of chemicals. While these chemicals
may be effective, PWSs may not always
know the exact amount and type of com-
pounds contained in the product. Suppliers
should submit documentation that their
products are safe to use in a potable water
application. In some situations, Food and
Drug Administration approvals are appropv
riate. In other cases, the general type of
chemical will be listed by the Code of
Federal Regulations as a "...substance
generally recognized as safe."
Each State has a drinking water direct
additives program which follows either the
National Sanitation Foundation Health
Effects Standard 60 or its own standards
for judging the suitability of direct addi-
tives for potable use. Any corrosion control
chemical used at a PWS must comply with
the State's direct additives program
requirements.
Staffing requirements may increase
with the implementation of corrosion
control treatment and additional testing
5-19
-------
OPERATION AND IMPLEMENTATION
required by the Lead and Copper Rule.
Due to the factors noted above, labor and
maintenance requirements may be higher
for calcium carbonate precipitation and
carbonate passivation than for the addition
of a corrosion inhibitor. Regardless, PWSs
that are required to provide optimal
corrosion control treatment need to
schedule and budget for the additional
staff that may be needed. The schedule
presented in Table 5-1 can be used to
coordinate hiring and training require-
ments in advance of the dates when
corrosion control treatment is required to
be on line.
Recognizing that problems may occur
with the startup of any new treatment
component, PWSs should collect appropri-
ate data and analyze the trends that occur.
One of the unique issues about corrosion
control is that a substantial amount of
time may elapse between the time treat-
ment changes are made and their effects
are detected through the analysis of tap
samples or corrosion monitors. For this
reason, detailed recordkeeping procedures
should be developed and followed to
correlate proper control of the treatment
processes with the desired effects in the
distribution system.
Important records will include customer
complaints such as colored water, stained
fixtures or laundry, taste and odor prob-
lems, and the lack of water pressure.
These records should also include data
regarding the age of the house, type of
interior and exterior plumbing, and the
utilization of onsite point of use (POU)
equipment such as softeners or carbon
filtration systems. The PWS's follow-up
action to the customer complaint should
also be noted in the records.
5.3 Optimization
Techniques
Optimization of corrosion control treat-
ment encumbers two overall phases:
(1) diagnosis of the need for optimization;
and (2) methods for implementing optimi-
zation techniques and addressing the
possible outcomes from such actions.
5.3.1 Diagnosing the Need for
Optimization.
Many PWSs may install optimal
corrosion control treatment and still
experience excessive lead, copper, or other
corrosion byproducts in the delivered
water. Determining when treatment has
been optimized—i.e., providing the maxi-
mum corrosion protection possible through
water treatment—is the first step. Addi-
tional sampling and monitoring (see
Diagnostic Monitoring, Section 5.2.3)
should be used to assist in optimizing
corrosion control treatment. Monitoring
during the two-year installation period can
be an important key in meeting ALs during
subsequent compliance monitoring periods,
and data collected during this two-year
period do not have to be reported to the
State. PWSs should consider monitoring
for the appropriate water quality parame-
ters at the entry point(s) to the distribution
system as well as within the distribution
system. Collecting tap samples for lead
and copper determinations will assist in
maintaining contact with homeowners who
assisted during the initial monitoring
phase as well as providing important
information regarding improvements to
water quality resulting from the corrosion
control treatment.
5-20
-------
OPERATION AND IMPLEMENTATION
Since corrosion control practices to
minimize lead and copper levels in first-
draw tap samples has not been generally
practiced by the drinking water communi-
rty, little information is available demon-
strating the performance of treatment
optimization techniques. Therefore, PWSs
must approach optimization with caution,
allowing sufficient time for treatment to
become effective and stabilize before
implementing any changes in an attempt
to improve system performance.
Data collection efforts regarding the
corrosion behavior of the distribution
system and home plumbing environments
should be used to develop long-term trends
in system behavior. Given the variability
in corrosion activity, observations of
improvements in corrosion protection
should be confirmed by at least one year
of monitoring data before any changes are
considered. However, distribution system
upsets by the installation of corrosion
control treatment—such as the release of
existing corrosion byproducts when
inhibitors are first applied—may be
realized very quickly after startup of
treatment. When degradation of water
quality in the distribution system and at
consumers' taps occurs, a timely response
should be made by PWSs to address these
problems.
The following steps should be addressed
in the priority shown to logically progress
through optimization of the installed
treatment process:
• Step 1 - Select treatment chemicals
which enable the WTP to meet its
optimal corrosion control treatment
objectives;
• Step 2 - Select chemical application
points within the WTP to provide
optimal utilization of each chemical
additive;
• Step 3 - Reduce water quality
parameter variability at the points of
entry to the distribution system;
• Step 4 - Reduce water quality
parameter variability within the
distribution system;
• Step 5 - Modify the water quality
parameter goals that define optimal
corrosion control treatment and thereby
the chemical feed requirements.
Steps 1 and 2 should be addressed
initially during the corrosion control study;
however, changes in other water treatment
processes or the need to improve corrosion
control performance may cause their
reevaluation. Steps 3 and 4 focus on the
ability of the WTP and distribution system
to be operated in accordance with the
corrosion control treatment goals. In many
cases, optimization for PWSs will consist
of addressing these conditions. Maintaining
consistent water quality leaving plants and
within distribution systems can be difficult,
and optimizing treatment without such
control may not be possible. Step 5 relies
on changing the actual goals defining that
treatment for optimization, and should
only be pursued as the last option by PWSs
and must be approved by the State.
Any change in treatment or plant
operations can impart adverse effects on
other water treatment or water quality
goals. General relationships may be
described to illustrate the effect of water
quality changes on treatment and finished
water quality objectives. Table 5-4 identi-
fies the major water quality characteristics
of concern and provides a general indica-
tion of their influence and effects.
Decisions related to corrosion control
5-21
-------
OPERATION AND IMPLEMENTATION
Table 5-4. Relational Behavior of Changing Water Quality
Conditions for Corrosion Control Treatment and Other
Water Quality/Treatment Objectives
^iiiEIiiiiyV^torQuallty'Change
Non-Softening WTPs:
pH Increase - After Filtration
Softening and Non-Softening
WTPs:
pH Increase - Before Filtration
Softening WTPs:
pH Decrease - Before Filtration
Alkalinity Increase
Alkalinity Decrease
Calcium Increase
Calcium Decrease (Softening
WTPs)
Phosphate Increase
Silicate Increase
frnpsct
• Increase in TTHM formation.
• Decrease in haloacetic acid formation.
• Increase in final turbidity when lime is used.
• Reduced disinfection efficacy.
• Post-filtration precipitation of manganese.
• Reduced disinfection by-product precursor removal when alum
coagulation is practiced.
• Increase in TTHM formation.
• Decrease in haloacetic acid formation.
• Reduced disinfection efficacy unless at pH levels above 9.0.
• Increased soluble aluminum levels when alum coagulation is
practiced.
• Increased removal of manganese.
• Increased encrustation of filter media when excess calcium
carbonate available.
• Excess precipitation of calcium carbonate when available in pipe
network near WTP.
• Decrease in TTHM formation.
• Increase in haloacetic acid formation.
• Reduced encrustation of filter media.
• Reduced soluble aluminum levels when alum is added during
softening.
• Increase ozone demand for disinfection.
• At very low levels, reduced coagulation performance when using
alum.
• Increased encrustation of filter media when excess calcium
carbonate available.
• Excess precipitation of calcium carbonate when available in pipe
network near WTP.
• Increase scavenging of phosphate inhibitors used for either
corrosion control or chelation.
• If after filtration, finished water turbidity increases.
• Prevent excess precipitation of calcium carbonate in pipe network
near WTP.
• Stripping of existing corrosion by-products in the distribution
system causing aesthetic quality degradation and increasing HPC
levels initially due to biofilm disturbances.
• May reduce useful life of domestic hot water heaters due to
"glassification"; silicates precipitate rapidly at higher temperatures.
5-22
-------
OPERATION AND IMPLEMENTATION
optimization should not be based solely
on the limited information presented in
Table 5-4. This table is more appropriately
used as a screening tool and a vehicle for
focusing current and future investigative
:efforts in developing optimization
approaches.
5.3.2 Methods for Evaluating
Treatment.
Identifying the corrosion behavior of
the distribution system and home plumb-
ing environments after the implementation
of optimal corrosion control treatment is
necessary to determine whether potential
improvements may be made by optimizing
corrosion control treatment. The methods
discussed in this section may assist PWSs
in developing long-term trends in corrosion
control performance. Not all data collection
efforts are necessary, but PWSs should
consider more than one method for evalu-
ating the actual performance of treatment
since no single technique can completely
describe the variety of corrosion activity
and its possible causes.
5.3.2.1 Water quality parameters.
After implementing an optimal corrosion
control program, follow-up monitoring is
required for pH, alkalinity, and calcium
for all large PWSs and those small and
medium-size PWSs that exceed an AL. In
addition, orthophosphate or silica
monitoring is also mandatory if one of
these corrosion inhibitors is used. To assist
in the optimization process, PWSs are
encouraged to measure these water quality
parameters in the tap samples collected
from consumers' homes. This would be
useful in tracking both the success of the
corrosion control program and the
alteration of water quality within
consumers' plumbing systems.
Increases in THM formation are
observed with increasing pH and this may
be a concern for those systems where the
pH is increased as part of the corrosion
control program. Therefore, additional
THM testing within the distribution
system may give insight about whether
further adjustments can or should be made
in the finished water pH.
Microbiological activity within the
distribution system should be closely
monitored after installing corrosion control
treatment. Several studies investigating
the impact of corrosion control on the
behavior of biofilms have generally
concluded that reductions in corrosion
activity significantly reduces: (1) the
likelihood of biofilm growth; and (2) the
resistance of microorganisms to disinfec-
tants. The additional nutrients which may
be added as a result of corrosion control
treatment has not been shown to increase
the biological activity of the distribution
system. Total coliform monitoring as
required by the Total Coliform Rule and
regular testing for heterotrophic plate
count bacteria would assist PWSs in
understanding the response of the distribu-
tion system to corrosion control treatment.
5.3.2.2 Lead and copper data. All
PWSs required to install corrosion control
treatment must perform routine monitor-
ing of first-draw tap samples for lead and
copper. These data may be used to deter-
mine the long-term effectiveness of
corrosion control treatment and the
ongoing actions required by the Rule, such
as public education or lead service line
replacement programs.
5-23
-------
OPERATION AND IMPLEMENTATION
As a check on the consistency of the
treatment process and the impact of
varying hydraulic conditions and water
blends within the distribution system, the
PWS may find it useful to collect first-
draw samples for five or more consecutive
days at a representative number of sites
prior to the start of the follow-up
monitoring period. The additional tests
should be considered diagnostic monitoring
(Section 5.2.3) rather than compliance
sampling. If the lead and copper results
from a particular tap vary significantly
from day-to-day, it indicates that the
corrosion control program is not achieving
consistent results in that location.
Depending upon how widespread the
inconsistencies are, the PWS should
investigate whether chemical feed prob-
lems, variations in raw water source,
hydraulic changes in the distribution
system, or site-specific conditions are
contributing to the daily variation in lead
and copper values. The goal of such addi-
tional testing is to ensure that corrosion
control objectives are consistently met at
all times.
5.3.2.3 Coupons and pipe inserts.
Coupons are available in a variety of
metals, such as lead, copper, cast iron,
bronze, and mild steel. Mild steel and
copper coupons are most frequently used.
Typically, coupons are placed in 8-inch or
larger pipes and in locations that have
moderate flow velocities (2-6 fps). Coupon
locations should avoid both stagnant and
high velocity flow conditions that are not
representative of the system as a whole.
When properly placed within the distri-
bution system, coupons provide a direct
indication of corrosion rates within the
pipe network. Some of their limitations,
however, include the fact that it takes a
long time to obtain accurate values and
coupons cannot be used to indicate short-
term changes in water quality characteris-
tics. For example, multiple coupons should
be used at each site so that corrosion rates
over varying lengths of exposure time may
be measured. This also provides informa-
tion regarding the impact of seasonal
variations on corrosion activity. In addi-
tion, while the coupon insertion and
removal equipment is moderately priced,
additional costs may be incurred to
construct access vaults at the locations
where coupons should be placed. Finally,
coupons are typically located within the
main pipe network and this is not neces-
sarily representative of the home plumbing
environment which lead and copper
monitoring reflects.
In summary, coupons can provide
meaningful information regarding the rate
at which exposed metal will corrode or
become encrusted with scale-forming
deposits within distribution system piping
networks. As such, PWSs should consider
their use as part of a comprehensive
corrosion monitoring program but should
not rely solely on these measures to assess
corrosion control performance.
Pipe inserts are small segments of un-
coated metal pipe that are part of the
distribution system. Inserts can be placed
in a vault that includes a bypass line so
the insert can be removed from service and
inspected for corrosion or deposition.
Inserts provide the opportunity to see what
is happening to the pipe wall itself, rather
than pieces of metal inserted into the pipe.
By coring or cutting sections from the pipe,
the thickness of the remaining metal or
deposit can be directly measured. Ultra-
5-24
-------
OPERATION AND IMPLEMENTATION
sonic instruments are also available to
indirectly measure the thickness of the
metal and/or deposit.
Pipe inserts suffer many of the same
limitations as coupons since they must
remain in place for long periods of time,
may be expensive to install and remove,
and are not be representative of home
plumbing conditions. As with coupons,
however, pipe inserts can provide relative
information on the effectiveness of a
corrosion control program.
5.3.2.4 Corrosion indices. Corrosion
indices have been used within the drinking
water community to assess the likelihood
of forming calcium carbonate scales on
pipes, and are derived from calcium
carbonate equilibrium relationships.
Limitations of the usefulness of these
indices needs to be recognized. When
optimal treatment consists of calcium
carbonate precipitation, indices may
properly describe the mechanisms of
corrosion control desired. However, the
equilibrium relationships upon which most
indices are based do not hold true when
any inhibitor is present, including poly-
phosphates which are typically used to
prevent metals and/or calcium from
precipitating (in the case of many softening
plants, polyphosphates are applied before
filtration to keep the filters fi-om becoming
encrusted by calcium carbonate). Corrosion
indices have little merit for those PWSs
applying carbonate or inhibitor passivation
as corrosion control treatment, and should
not be used to describe treatment goals.
For calcium carbonate precipitation
treatment, the CCPP index is recommend-
ed and Appendix A provides a detailed
description of its calculation methods.
5.3.2.5 Corrosion monitors. There
are several means for making discrete
observation or measurement of corrosion.
These include X-ray, ultrasonic, visual,
and destructive testing. While each of
these measurement techniques may be
useful in a particular situation, this section
will focus on electronic monitoring systems
which can be used while the distribution
system is in operation. Some of the
electronic monitoring devices measure the
byproducts of galvanic corrosion and others
will detect the loss of metal whether it is
due to galvanic action, leaching, or some
other corrosion mechanism.
5.3.2.5.1 Hydrogen probes. As
part of the oxidation/reduction reaction
in acidic solutions, hydrogen atoms will
migrate to cathode sites on the inside
surface of metal pipes. One type of probe
allows these hydrogen atoms to penetrate,
combine, and form hydrogen gas. The gas
will exert pressure which is proportional
to the amount of galvanic corrosion that
is occurring within pipeline. Another type
of probe uses palladium foil to create an
electrical output which is directly propor-
tional to the hydrogen evolution rate. By
recording pressure reading or electrical
output trends, changes in the corrosion
rate can be detected.
5.3.2.5.2 Electrical resistance.
This type of instrument measures the
electrical resistance of a thin metal probe
inserted into the pipeline. Compared to
conventional coupons, electrical resistance
probes provide results with a minimal
amount of effort. Continuous readings can
be made and the data analyzed to identify
corrosion trends. For example, an increase
5-25
-------
OPERATION AND IMPLEMENTATION
in electrical resistance would likely
indicate an increase in corrosion rates.
5.3.2.5.3 Linear polarization
resistance. In a galvanic corrosion cell,
the pipe metal is oxidized, or corroded, at
the anode and cations in the solution are
reduced at the cathode. In this process,
electrons are transferred between anodic
and cathodic areas on the corroding metal.
When a small voltage potential is applied
across the electrolyte fluid, the electrical
resistance is linear and the corrosion
current flow (corrosion rate) is directly
proportional to the measured current flow.
In dilute solutions such as drinking water,
the resistance of the electrolyte can be
significant compared to the polarization
resistance of the anode and cathode sites.
In these situations, the probe must be of
the type that will measure and compensate
for the resistance of the solution.
Compared to metal coupon and electri-
cal resistance monitoring, linear polariza-
tion probes provide a direct reading of the
corrosion current and rate. This allows for
instantaneous measurement of the changes
that occur with the type and amount of
corrosion control chemicals that are added
to the water. Linear polarization measure-
ments, however, cannot be made in non-
conductive fluids or those which coat the
electrodes. Therefore, they may not be
appropriate in those situations where a
calcium carbonate film is used to coat the
distribution system.
5.3.2.5.4 Electrochemical noise.
This a monitoring technique which
measures the electrochemical disturbances
created by corrosion activity. Potential
limitations include the fact that other
sources of electrical disturbance, such as
those from an impressed current system,
can result in overestimates of the corrosion
rate. This technique, however, is used by
some equipment manufacturers to indicate
the "pitting index" for a particular pipe-
line/electrolyte combination.
5&2.5£ Application suggestions.
Electronic corrosion monitoring equipment
can provide a rapid evaluation of corrosion
control treatment alternatives and chemi-
cal feed rates. These probes are not
infallible, however, and the electronic
measurements need to be correlated with
the results from other indicators such as
lead and copper data, conventional cou-
pons, pipe inserts, and water quality
indices. Once the relationship between
electronic measurements and actual field
corrosion conditions is established, cor-
rosion monitors can be a useful tool for
monitoring plant performance and main-
taining the finished water within the
operating parameters discussed in Section
5.2.2.
5.4 Optimizing Corrosion
Control Treatment—
Examples
Each PWS will experience unique
circumstances surrounding the optimiza-
tion of corrosion control treatment based
on site-specific conditions, treatment
objectives, and other considerations
affecting the performance and operation
of the distribution system. The following
examples illustrate the types of problems
which PWSs may encounter and approach-
es to solve treatment and operational
concerns.
5-26
-------
OPERATION AND IMPLEMENTATION
5.4.1 Optimal Corrosion
Control in a Consecutive
System.
East Bumford County runs a surface
water filtration plant that delivers water
;to a number of small and medium-size
towns. To more efficiently implement the
rule, the towns petitioned and received
approval to have the county water
treatment plant and the entire distribution
system be considered as a consolidated
large water system.
As part of the consolidation, East
Bumford County agreed to be responsible
for implementing the provisions of the
Lead and Copper Rule. All water treat-
ment, sampling, and monitoring costs were
to be paid for by the County with reim-
bursement by the towns on a population-
weighted basis. Additionally, any lead
service line replacements were to be paid
for by the town in which they were being
replaced.
Addition of a phosphate inhibitor to
the water treatment plant effluent was
approved by the state as optimal corrosion
control. While conducting follow-up
monitoring to determine their optimal
corrosion control parameters, the consoli-
dated system was found to have met both
the lead and copper action levels. However,
East Bumford County found that the water
in a remote section of Wakuska Township
was not maintaining a phosphate residual.
Apparently, the long residence time
between the treatment plant and the
remote section of Wakuska caused its
depletion.
At a subsequent meeting of East
Bumford County^ member communities,
an agreement was reached whereby
Wakuska Township would pay for a
chemical feed station to supplement the
phosphate inhibitor in their distribution
system. It was located downstream of
Wakuska's storage reservoir after the
Town's master flowmeter. Wakuska Town-
ship agreed to pay all costs associated with
the chemical feed station and East
Bumford County agreed to provide opera-
tional and maintenance support.
During the subsequent round of routine
monitoring, effective residual phosphate
concentrations were achieved throughout
the entire consolidated distribution system.
This permitted East Bumford County and
its member communities to comply with
the State-specified operating ranges under
the Lead and Copper Rule.
5.4.2 Use of Corrosion
Monitors in a Large System.
As a large system with historical
corrosion problems, Plimpton City had
been experimenting with control strategies
for a number of years. The City found that
it could reduce the number of red water
complaints if the lime-softening plant was
operated to achieve a CCPP index of 6
mg/L CaCO3 in the plant effluent. Even
with this operation, though, the distribu-
tion system still experienced some red
water, and early initial monitoring results
showed that the system would not meet
the lead action levels.
Realizing that their attempts at
corrosion control to date would not satisfy
the provisions of the LCR, the City
supplemented their existing program by
installing linear polarization resistance
corrosion control monitors in different
areas of the distribution system before the
beginning of the second round of initial
5-27
-------
OPERATION AND IMPLEMENTATION
monitoring for lead and copper. Figure 5-3
shows where the monitors were located.
The control monitors used iron electrodes
to simulate the material in the actual
system piping.
Results from the corrosion monitors
indicated that the highest corrosion rates
were found in the southeast section of the
City. However, those rates were not
consistently higher, but often fell to the
levels found in other areas of the city. The
city realized that the southeast comer
contained a large industrial sector, causing
the distribution system to experience wide
fluctuations in localized demand. This
resulted in significantly higher velocities
passing through the pipes in that area of
the city. Since water quality characteristics
remained fairly consistent throughout the
entire distribution system, the elevated
corrosion rates in the southeast zone was
attributed to the intermittently high
velocities experienced in that area. This
effect appeared to be causing disturbances
to the coating on the pipes by either
physically stripping the precipitated layer,
or preventing the water in the pipe from
attaining an equilibrium condition under
which a calcium carbonate film could be
maintained.
To address this concern, the City
installed a 5 million gallon storage tank
to service the southeast portion of the
distribution system. The storage tank
allowed the industries to satisfy their peak
demands without causing wide velocity
fluctuations in adjoining areas. After the
tank's installation, corrosion rates in the
southeast section corresponded more
closely with rates in the other areas of the
City.
While the storage tank was being
installed, the City began their required
corrosion control study. Pipe-loop systems
were set up in the treatment plant's
existing filter gallery. The same type of
linear polarization resistance monitors
used in the distribution system were also
used in the study. Corrosion rates were
monitored not only in iron electrodes, but
lead electrodes as well. The series of runs
which were conducted allowed for the
comparison of corrosion rates of waters
with different CCPP indices. As the
treatment plant had discovered years
earlier, iron corrosion was not much
further reduced as the CCPP index rose
above 6.0 mg/L CaC03. However, lead
corrosion reached a minimum at 9.5 mg/L
CaCO3.
Before the City began full-scale treat-
ment changes to reflect the new CCPP
index, they replaced selected iron elec-
trodes in the distribution system with lead
electrodes. When they then changed the
main plant treatment, they were able to
verify that lead corrosion was reduced
when the distribution system was receiving
water with a higher CCPP index, and that
iron corrosion remained at the same levels
as when the system received water with
the lower CCPP index. Although the lead
corrosion rates were lower than they had
been previously, they were not quite as
low as the pipe-loop study indicated they
could be. The pipe-loop system is still being
used to determine whether another
parameter might more clearly define
optimal treatment, and periodic modifica-
tions are being made to the pipe-loop in
a continuing attempt to model actual
distribution system conditions.
5-28
-------
en
*>
Corrosion Control Monitor Locations
Figure 5-3. Example 5.4.2. — Use of Corrosion Monitors in the Plimpton City
Distribution System
-------
OPERATION AND IMPLEMENTATION
5.4.3 Use of Extra Monitoring.
Corrosion studies in the small town of
Gechlik Mills showed that raising the pH
of their direct groundwater supply from
7.2 to 7.8 caused a dramatic decrease in
distribution system lead and copper levels.
Direct in-line injection of caustic soda at
their two wellheads to achieve a pH of 7.8
was designated as the town's optimal
corrosion control. Since facilities to house
chemical feeds already existed for sodium
hypochlorite, installing the caustic soda
system was able to be completed quickly.
The Town made arrangements for a
chemical company to fill the chemical feed
tanks biweekly with a 50 percent caustic
soda solution that was directly injected
at the wellheads. The Town began using
the new system six months ahead of the
mandated schedule for treatment instal-
lation by the LCR.
The early installation allowed Gechlik
Mills six months before they had to collect
any lead and copper tap samples. However,
the Town began monitoring not only the
lead and copper levels; but also water
quality parameters at selective distribution
system sites to verify that the treatment
was working correctly. Many sites were
found to experience wide fluctuations in
pH over time. Lead and copper levels also
fluctuated, and a large number of samples
continued to exceed the action levels.
By running a series of bench-scale
tests, the caustic soda was found to be
working as desired. However, very precise
amounts had to be used in order to achieve
the 7.8 pH. Although the amounts of the
50 percent caustic soda which had to be
added were not very large, substantial pH
fluctuations often resulted due to slight
variations in feeding. By using larger
amounts of a more dilute caustic solution,
similar misfeedings were not found to have
as pronounced an effect on pH.
Before the follow-up monitoring period
was to begin, Gechlik Mills had already
installed larger caustic holding tanks, and
had begun injecting a 25 percent caustic
soda solution. In the subsequent monitor-
ing round, pH monitoring was conducted
along with the lead and copper monitoring.
The pH monitoring was found to be
unnecessary for Lead and Copper Rule
compliance, since the lead and copper
monitoring showed that both action levels
had been met. Although Gechlik Mills is
not always required to, they routinely
monitor their distribution system pH in
order to troubleshoot any potential increas-
es in corrosion activity.
5.5 References
Ainsworth, RG., et al. 1980. "The Introduc-
tion of New Water Supplies into Old
Distribution Systems, Tech Rep. TR-143".
Water Research Centre (Medmenham,
England).
AWWA. 1990. Water Quality and Treat-
ment, 4th ed., Chapter 17. McGraw-Hill
(New York).
AWWA. 1985. Corrosion Control for
Operators. AWWA (Denver, CO).
AWWA. 1971. Water Quality and Treat-
ment, Srded., Chapter 17. McGraw-Hill
(New York).
AWWARF. 1990. Assessing and Controlling
Bacterial Regrowth in Distribution Sys-
tems. AWWA (Denver, CO).
AWWARF. 1989. Economics of Internal
Corrosion Control. AWWA (Denver, CO).
5-30
-------
OPERATION AND IMPLEMENTATION
AWWARF/DVGW-Forschungsstelle
Cooperative. 1985. Internal Corrosion of
Water Distribution Systems. AWWARF
(Denver, CO).
Donlan, R.M. and Pipes, W.O. 1988.
"Selected Drinking Water Characteristics
and Attached Microbial Population
Density". Journal AWWA. 80(ll):70-76.
Elmund, K. 1992. Corrosion Control
Experiences of Fort Collins, Colorado.
AWWA Rocky Mountain Section Corrosion
Control Seminar, Denver, CO (April).
Hasson, D. and Karmon, M. 1984. "Novel
Process for Lining Water Mains by Con-
trolled Calcite Deposition". Corros. Prev.
Control. Vol. 9.
Huck, P.M. 1990. "Measurement of
Biodegradable Organic Matter and Bacteri-
al Growth Potential in Drinking Water".
Journal AWWA. 82(7):78-86.
Koudelka, M., et al. 1982. "On the Nature
of Surface Films Formed on Iron in
Aggressive and Inhibiting Polyphosphate
Solution". J. Electrochem, Soc. 129:1186.
LeChevallier, M.W., et al. 1990. "Disinfect-
ing Biofilms in a Model Distribution
System". Journal AWWA. 82(7):87-99.
LeChevallier, M.W., et al. 1988a. "Factors
Promoting Survival or Bacteria in Chlori-
nated Water Supplies". Applied and
Environmental Microbiology. 54(3):649-654.
LeChevallier, M.W., et al. 1988b. "Inactiva-
tion of Biofilm Bacteria". Applied and
Environmental Microbiology. 54(10):2492-
2499.
LeChevallier, M.W., et al. 1987.
"Examination and Characterization of
Distribution System Biofilms". Applied and
Environmental Microbiology. 53(12):2714-
2724.
Loewenthal, R.E. and Marais, G.v.R. 1976.
Carbonate Chemistry of Aquatic Systems:
Theory and Applications. Ann Arbor
Science. (Ann Arbor, MI).
McCauley, R.F. 1960. "Controlled
Deposition of Protective Calcite Coatings
in Water Mains". Journal AWWA.
52(11):1386.
Merrill, D.T. and Sanks, R.L. 1977.
"Corrosion Control by Deposition of CaC03
Films: A Practical Approach for Plant
Operators-!". Journal AWWA. 69(11):592.
Merrill, D.T. and Sanks, R.L. 1977.
"Corrosion Control by Deposition of CaCO3
Films: A Practical Approach for Plant
Operators-II". Journal AWWA. 69(12):634.
Merrill, D.T. and Sanks, R.L. 1977.
"Corrosion Control by Deposition of CaCO3
Films: A Practical Approach for Plant
Operators-II". Journal AWWA. 70(1):12.
Obrecht, M.F. and Pourbaix, M. 1967.
"Corrosion of Metals in Potable Water
Systems". Journal AWWA. 59(8):977.
Reiber, S., et al. 1989. "An Improved
Method for Corrosion Rate Measurement
by Weight Loss". Journal AWWA.
Rossum, J.R. 1987. "Dead Ends, Red
Water, and Scrap Piles". Journal AWWA.
79(7):113.
Rossum, J.R. and Merrill, D.T. 1983. "An
Evaluation of the Calcium Carbonate
Saturation Idices". Journal AWWA,
75(2):95.
5-31
-------
OPERATION AND IMPLEMENTATION
Schock, M.R. 1989. "Understanding
Corrosion Control Strategies for Lead".
Journal AWWA. 81(7):88.
Schock, M.R, et al. 1988. "The Significance
of Sources of Temporal Variability of Lead
in Corrosion Evaluation and Monitoring
Program Design". Proc. AWWA Water
Qual. Tech. Conf. (St. Louis, MO).
Schock, M.R. 1985. "Treatment or Water
Quality Adjustment to Attain MCLs in
Metallic Potable Water Plumbing System".
Proc. Plumbing Materials and Drinking
Water Quality. (Cincinnati, OH).
Schock, M.R. 1985. U.S. Environmental
Protection Agency, Water Engineering
Research Laboratory, Rep. 600/9-85/007.
Schock, M.R 1984. "Temperature and Ionic
Strength Corrections to the Langelier
Index-Revisited". Journal AWWA.
76(8):72.
Schock, M.R and Buelow, RW. 1981. "The
Behavior of Asbestos-Cement Pipe Under
Various Water Quality Conditions: Part 2.
Theoretical Considerations". Journal
AWWA. 73(12):636.
Snoeyink, V.L. and Jenkins, D. 1980.
Water Chemistry. Wiley (New York).
Sontheimer, H., et al. 1981. "The Siderite
Model of the Formation of Corrosion-
Resistant Scales". Journal AWWA.
73(11):572.
Stumm, W. 1956. "Calcium Carbonate
Deposition at Iron Surfaces". Journal
AWWA. 48(3) :300.
Temkar, P.M., et al. 1987. "Pipe Loop
System for Evaluating Effects of Water
Quality Control Chemicals in Water
Distribution System". Proc. AWWA Water
Qual. Tech. Conf. (Baltimore, MD).
5-32
-------
LEAD SERVICE LINE REPLACEMENT
Chapter 6.0 —
Lead Service Line Replacement
6.1 Overview of LSL
Replacement
Requirements
Lead Service Lines (LSLs) have been
shown to contribute significant amounts
of lead to drinking water at the consumer's
tap. Corrosion control techniques are often
effective in minimizing lead levels associat-
ed with LSLs by establishing a protective
coating on the interior pipe surface.
Although EPA believes that corrosion
control treatment will be the primary
means of lead level reduction for the
majority of water systems, the establish-
ment of such protection can vary from
house to house. In many instances,
corrosion control and/or source water
treatment alone will not be sufficient to
reduce lead levels below the lead AL. In
such cases a PWS must replace its LSLs
in accordance with the LSLRP require-
ments (§141.84). EPA believes that the
progressive replacement of LSLs which
contribute to lead levels above 0.015 mg/L
will reduce adverse health risks imposed
by lead exposure.
The Lead Service Line Replacement
Program (LSLRP) in the June 7, 1991 rule
is premised on five principles: (1) corro-
sion control can reduce lead levels from
LSLs in some instances, but high levels
may persist after treatment; (2) a system
is triggered into a LSLRP if the system
exceeds the lead AL after installing
optimal corrosion control and source water
treatment (follow-up monitoring); (3) water
systems should only be responsible for
removing that portion of each LSL they
control; (4) a system is not required to
physically replace individual LSLs if direct
sample lead concentrations are 0.015 mg/L
or less and (5) water systems must
annually replace at least 7 percent of the
total number of LSLs in place at the
beginning of the LSLRP.
Any water system that continues to
exceed the lead AL after implementing
optimal corrosion control treatment and/or
source water treatment (whichever is
installed later), or during any subsequent
monitoring period, must begin replacing
LSLs identified within the distribution
system. The LSLRP begins on the date the
system exceeds the lead AL as referenced
above (i.e., January 1 or July 1 of a given
year). The State also has the authority to
require LSLRP commencement immediate-
ly for systems who have failed to install
source water or corrosion control treatment
by the deadline for follow-up monitoring
as provided in §141.86(d)(2).
A water system which is triggered into
the LSLRP is required to take three steps:
(1) conduct a comprehensive materials
evaluation (if not already completed) to
identify all homes or buildings served by
LSLs; (2) establish a schedule for replacing
LSLs; and (3) physically replace all LSLs
controlled by the system. Water systems
can avoid replacing individual LSLs that
are shown to contribute 0.015 mg/L or less
to tap water lead levels as measured in
LSL samples. Water systems can discon-
tinue the LSLRP if they demonstrate that
the lead levels in first-draw water collected
6-1
-------
LEAD SERVICE LINE REPLACEMENT
at targeted taps are below the lead AL for
two consecutive six-month monitoring
periods. If a system subsequently exceeds
the lead AL during any monitoring period,
the LSLRP must be recommenced.
The following sections discuss the
rationale of the LSLRP requirements, LSL
control and related requirements, materi-
als evaluation, LSL replacement schedules,
and reporting/record-keeping requirements.
6.2 LSL Control and
Related Requirements
EPA believes its authority to impose
regulatory requirements on PWSs extends
only to those distribution facilities under
the control of the PWS. Under the Rule,
systems replacing LSLs are required to
replace the portions of LSLs under their
control, presuming that the system
controls the entire LSL (up to the building
inlet). PWSs may rebut the presumption
that they control the entire lead service
line and replace only that portion which
an appropriate legal authority (i.e., State
statute, municipal ordinance, public
service contract, etc.) defines as controlled
by the PWS. The definition of control is
discussed in the following subsection and
is followed by explanations of the require-
ments for control presumption rebuttal
and partial LSL replacement.
6.2.1 LSL Control
Determination.
Control is defined in §141.84(e) as one
of the following forms of authority:
• Authority to set standards for
construction, repair, or maintenance
of the line;
• Authority to replace, repair, or
maintain the service line; and,
• Ownership of the line.
EPA acknowledges that ownership
and/or control of LSLs is often split
between the PWS and the property owner.
Depending upon State laws or municipal
ordinances, some public water systems
control and/or own connections up to the
property line, others control and/or own
the LSL and other connections up to the
building, and still others control and/or
own the service connections only up to the
curb (see Figure 6-1). It should be noted
that a lead gooseneck is part of the LSL
only when it is attached to the LSL. Where
LSL ownership is split between the utility
and the user, utilities sometime retain
authority to prescribe the standards for
construction, repair, and maintenance of
service lines, and a right of entry to
perform work deemed necessary.
6.2.2 Rebuttal of Control
Presumption.
Water systems are required to replace
the entire LSL (up to the building inlet)
unless they can successfully demonstrate
to the State that part of the LSL is beyond
their control. A water system can rebut
the control presumption by citing local
ordinances or State statutes, or in the case
of private systems, the contract between
the systems and their customers that limit
the extent of control.
Systems that do not intend to replace
the entire LSL are required to submit a
letter to the State, within the first year
of their replacement schedule, demonstrat-
ing that their control is limited. This letter
must be accompanied by a copy of the legal
6-2
-------
o>
6
I
L
OmeNeck.
SftUm-Owiutl
Slop Box
tSL(l)rp.) -f
D
-------
LEAD SERVICE LINE REPLACEMENT
authority the water system is relying upon
to justify its position that the extent of its
control is limited On the basis of this legal
authority, the State will determine
whether the system correctly interprets
its relevant legal authority.
EPA believes that requiring State
review of PWS control presumption
rebuttals is important to ensure that
systems correctly apply the regulatory
definition of control to their system. To
expedite implementation of LSL replace-
ment, States are not required to approve
a system's interpretation of its legal
authority prior to the system beginning
LSL replacement. However, the State may
determine that a system has incorrectly
interpreted the extent of its control over
LSLs. In all such instances, the State is
required to explain the basis for its
decision in writing and notify the system
of that decision. The system must then
replace the portion of the LSL under its
control as determined by the State. Where
a system's control does not extend over
the entire LSL, the system is required to
offer to replace the portion of the LSL
controlled by the homeowner but is not
required to bear the cost of replacing the
building owner's portion of the line.
6.2.3 Partial LSL
Replacement.
Systems replacing LSLs are required
to replace the portions of the LSLs that
are under their "control" as defined in
§141.84(e) of the LCR and Section 6.2.1
of this manual. Control is often split
between the PWS and the property owner.
This potentially limits the PWS's ability
to remove the entire LSL. The Rule
requires that the system offer to replace
any portion of the LSL controlled by the
homeowner but is not required to bear the
cost of replacing the homeowner's portion.
Partial replacement of LSLs has been
observed, in some cases, to result in short-
term but significant increases in tap water
lead levels (Hulsmann, 1990; Schock,
1990). EPA believes that such increases
and associated health impacts will be
minimized since effective corrosion control
should be in place by that time, and also
because customers will be informed of how
they can minimize their exposure. The
primary concern regarding lead in drinking
water is not acute toxicity, but rather
lead's capacity to accumulate in the body
and result in chronic health effects. Thus,
EPA believes that the potential risks posed
by such temporary increases are out-
weighed by the importance of having lead
levels reduced over the long term.
In those locations where only a portion
of the LSL is replaced, PWSs must notify
affected customers and offer them the
option of having a follow-up tap sample
collected and analyzed to determine
whether there has been an increase in tap
water lead levels. The PWS will not be
required to pay for the collection or
analysis of such samples nor will the
system be required to collect and analyze
the sample itself. However, if a customer
accepts the offer, the PWS must report the
sampling results to the customer within
14 days of partial replacement. The
purpose of collecting the follow-up samples
is to identify those locations where tran-
sient increases in water lead levels could
occur and inform residents of the precau-
tionary steps they should take (i.e.,
flushing water at the taps). Methods for
collecting LSL samples at consumers' taps
6-4
-------
LEAD SERVICE LINE REPLACEMENT
are described in the LCR Guidance Manual,
Volume 1 (Section 4.3).
6.3 Materials Evaluation
A complete determination of all LSL sites
may not have been accomplished during the
Material Survey for the Sample Plan
Development. This survey effort must be
completed if a system is triggered into the
LSLRP, because the initial number of LSLs
determined serves as the basis for
replacement rate determination. Methods
for determining the locations of LSLs were
presented in Section 3.0 of the Lead and
Copper Rule Guidance Manual, Volume 1.
Twelve months after a water system is
triggered into the LSLRP, it is required to
submit to the State a revised materials
evaluation identifying the initial number of
LSLs in its distribution system. The initial
number of LSLs is the number of LSLs in
place at the time the LSLRP begins.
EPA believes 12 months is an adequate
period of time because water systems should
have obtained such information either when
they were required to determine whether
their distribution system contained lead or
copper pipes [§141.42(d)]? or when they
established their sampling pool for tap
monitoring under this Rule. While some
municipalities will undoubtedly have
inadequate records documenting the location
of its LSLs, most systems are not required
to submit a complete material evaluation of
LSLs to the State until 8 to 10 years after
promulgation of the Rule. EPA believes this
provides water systems with sufficient time
to locate all LSLs and recommends that
systems with monitoring data indicating LSLs
may be a problem begin identifying the
location of LSLs now.
6.4 LSL Replacement and
Schedule Requirements
Systems which become subject to the
LSLRP must physically replace all LSLs,
except those for which the lead concentration
in all lead service line samples is less than
or equal to 0.015 mg/L. Thus, systems have
a choice between replacing LSLs or
conducting monitoring of the line to
determine if the lead levels are less than or
equal to 0.015 mg/L. LSLs may be consid-
ered to be "physically replaced", via die lead
contribution presumption rebuttal and when
excavation reveals that a presumed LSL is
in fact not a LSL. Regardless of how LSLs
are replaced/monitored, the process must
proceed at the annual rate specified by the
State and in accordance with size dependant
LSLRP schedules. The following subsections
discuss rebuttal of the lead contribution
presumption, replacement/elimination rates,
size-dependent LSLRP schedules, and
LSLRP discontinuation.
6.4.1 Rebuttal of Lead Contribution
Presumption.
The "lead contribution presumption"
essentially presumes that each LSL scheduled
for replacement significantly contributes to
lead concentrations of more man 0.015 mg/L
at the tap. Systems may rebut this
presumption for individual LSLs, via
sampling and analysis, if LSL samples (not
6-5
-------
LEAD SERVICE LINE REPLACEMENT
first-draw) reveal that die lead concentrations
are no more than 0.015 mg/L.
Detailed sampling procedures for LSL
monitoring are provided in Chapter 4.0 of
Volume I of the Lead and Copper Rule
Guidance Manual; a general description is
provided below. Each LSL sample must be
one liter in volume and stand motionless in
the LSL for at least six hours. LSL samples
must be collected in one of the following
ways:
1. Calculating the interior diameter and
length of the pipe between the tap and
the LSL, flushing the calculated volume
of water, and collecting the next one liter
of water; (Table 3-3 provides volumes
of standing water for various pipe lengths
and diameters);
2. Tapping directly into the LSL and
collecting one liter of water from the
line; or
3. Allowing the water to run until there is
a change in temperature and collecting
one liter of water immediately after the
change takes place. This method may
be used only when the sampb'ng site is
constructed as a single family residence.
If the concentration in the LSL sample
is less than or equal to 0.015 mg/L, then
the system need not replace the individual
LSL. Furthermore, each of these LSLs may
be counted as "replaced" in the LSLRP
accounting system. LSL monitoring by
PWSs is strictly optional. A water system
may choose to replace LSLs without
conducting any monitoring, regardless of
actual lead contribution, or if lead levels in
LSLs are expected to exceed 0.015 mg/L.
6.4.2 Replacement/Elimination
Rates.
It is difficult to establish a replacement
rate that can be applied nationwide because
the number of LSLs in each system varies
tremendously. EPA estimates mat LSLs may
comprise anywhere from 10 to 50 percent
of the service lines in those systems that have
LSLs. Replacement of all LSLs via normal
maintenance schedules could take as long
as 50 years for some systems. EPA believes
that it is necessary to accelerate the rate at
which LSLRP systems replace LSLs in order
to ensure that public health will be adequate-
ly protected. States will be in the best
position to assess the factual circumstances
of each individual system and the schedule
the system can feasibly meet.
EPA decided that in no case can a LSLRP
system take more than 15 years to replace
all its LSLs; where LSL "replacement"
consists of the summation of the following:
• LSLs physically replaced;
• LSLs for which the "lead contribution
presumption" is successfully rebutted via
sampling and analysis; and
• Lines identified as LSLs in the materials
evaluation which are found not to be
LSLs upon excavation.
Therefore water systems subject to the
LSLRP are required to annually "replace"
at least 7 percent of their initial number of
LSLs as identified in the materials evaluation
(see Section 6.3). For example, a system
that identifies 10,000 LSLs in its materials
evaluation would be required to cumulatively
account for replacement of at least 700
individual/additional LSLs each year via
6-6
-------
LEAD SERVICE LINE REPLACEMENT
physical LSL replacement, LSL lead
contribution rebuttals, and/or when initially
identified LSLs are found not to be LSLs
upon excavation. The system's LSL
replacement pool for a given year, for
example, could potentially consist of
690 LSLs which are physically replaced,
8 LSL lead contribution rebuttals, and 2 via
excavation. An LSL replacement work sheet,
which could be used to assist systems in their
LSLRP accounting process, and a completed
example have been included as Tables 6-1
and 6-2, respectively.
The Rule also requires that water systems
replace LSLs at a greater rate than 7 percent
annually where the State finds that an
accelerated schedule is feasible. The State
must make such determinations in writing
and must notify the system of its findings
within six months after the system is
triggered into the LSLRP.
6.4.3 Size-Dependent
LSLRP Schedules.
The timing of LSLRP requirements is
dependent upon when systems complete
corrosion control and/or source water treatment,
which in turn varies based upon system size.
This is particularly true for small and medium
size systems based upon whether or not a
corrosion control study is conducted. Schedules
for small and medium-size systems, as well as
large systems, are presented in Figure 6-2.
6.4.4 LSLRP Discontinuation.
It is conceivable that systems can meet the
lead AL which they had previously exceeded
through improved treatment—corrosion control
or source water treatment—or because they
obtain an alternative source of water. Thus,
water systems can discontinue the LSLRP if
they can demonstrate that first-draw tap water
lead levels are below the lead AL for two
consecutive six-month monitoring periods. EPA
decided to require lead AL compliance over
the course of an entire year to ensure that the
lower levels genuinely reflect a lowering of lead
levels, and not normal variability in lead levels
at the tap. Recommencement of the replacement
program is required if a system subsequently
exceeds the lead AL during any single
monitoring period.
6.5 Reporting Requirements
Once the LSLRP is initiated, a system must
meet reporting requirements in accordance with
the standardized schedule presented in Table
6-3 and outlined below.
Within three months of being required to
begin the LSLRP, a system seeking to rebut
the control presumption (presumes the system
controls the entire LSL) must submit a letter
to the State describing the legal authority which
limits the system's control over the LSL and
explain the extent of the system's control. The
letter must include copies of the State statute,
municipal ordinance, public service contract,
or any other legal authority the system contends
limits control.
Within 12 months a system must submit to
the State a schedule for the replacement of all
its LSLs at the annual rate approved by the
State. The schedule must also state the initial
number of LSLs. The schedule could include
the location of the LSLs within the distribution
system, and identify the LSLs scheduled for
replacement during each year of the replacement
schedule.
Every 12 months a system must
demonstrate that 7 percent (or more as
specified by the State) of its LSLs have
6-7
-------
Table 6-1. LSLRP General Accounting Worksheet
Initial No. of LSLs:
Required Annual Replacement (No.):
Required Annual Replacement (%):
o>
Year
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
Annual Numbers
PHS
RBT
EXC
No.REP
•'
\
%REP
Cumulative Numbers
CNo.REP
A
C%REP
.,
PHS = Number of LSLs physically replaced in the given year.
RBT = Number of LSLs eliminated via Pb contribution rebuttals.
EXC = Number of initially identified LSLs which are found not to be LSLs upon excavation for given year.
No.REP = Number of LSLs "replaced" in the given year (PHS + RBT + EXC).
%REP = Percent of LSLs replaced for the given year (No.Rep/Initial No.)(100)
CNo.REP = Cumulative number of LSLs replaced through given year.
C%REP = Cumulative percent of LSLs replaced through given year.
-------
Table 6-2. LSLRP General Accounting Worksheet
Initial No. of LSLs:
Required Annual Replacement (No.):
Required Annual Replacement (%):
Year
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
Annual Numbers
PHS
693
694
690
690
690
690
688
689
670
664
665
645
638
630
186
RBT
7
6
10
10
10
10
12
11
29
34
34
55
60
68
12
EXC
0
0
0
0
0
0
0
0
1
2
1
0
2
2
2
No.REP
700
700
700
700
700
700
700
700
700
700
700
700
700
700
200
%REP
7
7
7
7
7
7
7
7
7
7
7
7
7
7
2
Cumulative Numbers
CNo.REP
700
1400
2100
2800
> 3500
4200
4900
5600
6300
7000
7700
8400
9100
9800
10000
C%REP
7
14
21
28
35
42
49
56
63
70
77
84
91
98
100
-------
LEAD SERVICE LINE REPLACEMENT
NOTE: Figure 6-2 is a large fold-out sheet. Use this page number.
6-10
-------
LEAD SERVICE LINE REPLACEMENT
Table 6-3. Reporting Requirements Schedule
Within a Mo of
LSLRP Triggering
Within 12 Mo of
LSLRP Triggering
Within Each 12 Mo Period
Following LSLRP Triggering
Rebuttal of "control
presumption"
Materials evaluation
and replacement
schedule
Certification of "replacement" of
LSLs equaling at least 7 percent of
the initial number of LSLs as
specified by the State
been replaced. Sample documentation
demonstrating rebuttal of the lead
contribution presumption must also be
submitted in accordance with this
schedule. The annual letter must contain
the following information:
1. The number of LSLs scheduled to be
replaced during the previous year of
the replacement schedule.
2. The number and location of each LSL
replaced during the previous year of
the schedule.
3. If measured, the water lead concentra-
tion and location of each LSL sampled,
the sampling method, and the date of
sampling.
EPA believes that this information
must be submitted annually to insure that
the system is properly completing the
LSLRP.
6.6 Record-Keeping
Requirements
Any system subject to the record-
keeping requirements of §141.91 must
retain on its premises original records of
all sampling data and analyses, reports,
surveys, letters, evaluations, schedules,
State determinations, and any other
information required by §141.81 through
§141.88. Each water system shall retain
these records for no fewer than 12 years.
6.7 References
Hulsmann, A.D. 1990. Particulate Lead
in Water Supplies. Jour. Institution of
Water and Envir. Mang. 4(l):19-25.
Schock, M.R. 1990. Causes of Temporal
Variability of Lead in Domestic Plumbing
Supplies. Envir. Monitoring and Assess.
15:59-82.
6-11
-------
Appendix A
Corrosion Indices for the
Precipitation of Protective
Coatings
Many corrosion control programs for
water utilities have targeted the pro-
tection of metal pipes through precipi-
tation of calcium carbonate (CaCO3).
This process depends on the equilibrium
reactions involving the calcium ion
(Ca2+), hydrogen ion (H+), hydroxide'ion
(OH"), carbonic acid (H2CO3*), carbon
dioxide (CO2), bicarbonate ion (HCO3~),
and carbonate ion (CO32'). The objective
of the process is to produce a finished
water which will evenly precipitate
calcium carbonate on the pipe walls
within the distribution system. This
means that the finished water should be
supersaturated with respect to calcium
carbonate to the extent that precip-
itation occurs.
A multitude of corrosion indices have
been developed over the years to
describe the precipitation of calcium
carbonate. The recommended index is
the Calcium Carbonate Precipitation
Potential (CCPP) for use in evaluating
the water quality goals necessary to
successfully provide corrosion control
protection through the formation of
calcium carbonate films. The Langelier
Saturation Index (LI) may also be used
by PWSs due to its long history of
application and the ability of some
systems to develop reliable relationships
between LI and corrosion control
protection. Other corrosion indices are
not recommended for determining
water quality goals generating calcium
carbonate precipitation in distribution
and home plumbing systems.
Calcium Carbonate
Precipitation Potential.
The term calcium carbonate precipi-
tation potential (CCPP) refers to the
theoretical quantity of calcium carbonate
that can be precipitated from waters
that are super-saturated. A treated
water CCPP of 4-10 mg/1 (as CaCO3) is
typically required to promote formation
of protective calcium carbonate deposits.
For large systems, higher CCPPs may be
required to ensure maintenance of
calcium carbonate deposits throughout
the distribution system.
CCPP has also been shown to relate
directly to reaction kinetics as found by
Nancollas and Reddy (1976) and pre-
sented by Rossum and Merrill (1983):
d[Ca2+]/dt = -ID'5 KS(CCPP)2
where K is the rate constant for crys-
talline growth and S is the surface area
available for precipitation of a given
particle size. When applying corrosion
indices as a surrogate measure of cor-
rosion control performance, it is impor-
tant that the application be supported
by additional information, such as
distribution system monitoring, in-situ
coupon testing, bench-scale corrosion
testing, and inspection of pipe materials
removed from the distribution system
during maintenance and repair.
A-1
-------
Determining the Calcium
Carbonate Precipitation
Potential
CCPP can be determined graphically
through use of Caldwell-Lawrence
diagrams, analytically through equilib-
rium equations, or by computer analysis.
CCPP = 50,000 * ([Alk]s - [Alk]^)
Theoretical basis for determining the
amount of CaCO3 precipitated or
dissolved by waters depending on their
saturation condition as presented by
Merrill and Sanks (1977a, 1977b, 1978).
CCPP = 0: CaCO3 saturated solution.
CCPP > 0: CaCO3 supersaturaed solu-
tion, and the CCPP value denotes the
milligrams per liter of CaCO3 which
will be precipitated.
CCPP < 0: CaCO3 undersaturated
solution, and the CCPP value denotes
the milligrams per liter of CaCO3
needed to be dissolved into solution
to bring to saturation.
Rule of Thumb Goal: 4-10 mg/L CaCO3
CCPP Calculation Procedures:
A. Definition of Terms and Values
of Constants
[Alk]j Measured value of alkalinity in
the finished water,
representing the alkalinity of
solution prior to precipitation of
calcium carbonate.
[Alk]^ Equilibrium alkalinity resulting
after precipitation of the
calcium carbonate content
beyond saturation. Calculation
of this term requires an itera-
tive solution for the hydrogen
ion concentration at equilibri-
um. Once this is done, [Alk]^
can be calculated as follows:
[Alk]., = WPi * (Acyi - Seq) - Seq
where t^, = (2K,,' +
Peq = (2[H+]eq
Sec, = [HO, -
and,
KH = Henry's law constant for CO2
K,, = dissociation constant for water
Ka' = first dissociation constant of
carbonic acid.
Ka' = second dissociation constant of
carbonic acid
[Acy]j = Acidity of the finished water.
= CT*(ai + 2*a0) + [H+] - [OKI
= (1 + KjYlH*] +
The equilibrium constants used in
the above equations are given in Table
A-l for various temperature conditions.
A-2
-------
Table A-l. Equilibrium Constants for Carbonate-Water System
Temp °C
0
5
10
15
20
25
30
pKsp1
8.03
8.09
8.15
8.21
8.27
8.33
8.38
pK*
14.93
14.73
14.53
14.35
14.17
14.00
13.83
pKH
1.11
1.19
1.27
1.32
1.41
1.47
1.53
PK»'
6.579
6.517
6.464
6.419
6.381
6.352
6.327
PK2'
10.625
10.557
10.490
10.430
10.377
10.329
10.290
1 Derived from equation, pKsp =~0.01183*(Temp) + 8.03, Larson and Boswell,
1942.
B. Algorithm for Iterative Solution
The CCPP represents in mg/L as
CaCO3 the saturation state of calcium
carbonate with respect to existing
conditions (AlkJ and the equilibrium
conditions which would exist after the
water's potential to precipitate or dissolve
calcium carbonate had occurred (Alk^.
During this process, the equivalents of
calcium precipitated (or dissolved) must be
equal to the equivalents of alkalinity
precipitated (or dissolved). However, the
acidity of the water remains constant and
therefore can be used to determine the
equilibrium alkalinity conditions as
described below.
Acyi = Acye, = [(Allq + sj/tj'pi + ^
where s;, tj, and tj are defined as follows:
s, = [H+]
t, - (2*K2* +
Pi = (2*[in
Since acidity remains conservative
through the precipitation/dissolution of
calcium carbonate, the actual acidity of the
water (AcyJ may be used to define the
equilibrium alkalinity (Alk,^ as shown
below:
Alk.,, = teq/peq*(Acyi - Seq) - Seq
The equilibrium alkalinity condition may
also be related to the inital calcium and
alkalinity through the following equation:
2*[Ca21 - Alkj =
2*V%/(Alkeq + Se,) - Alk^,
with req = ([H+],, + 2*K2')/K2'
Substituting Alk,, = f(AcyJ into the
above equation yields the relationship
below:
2*[Ca2+]; -
[(2*KJp'
[V(Acyi -
i - s,,)] -
If we let TERMO equal the left side of
the above equation, and TERM1 and
TERM2 equal the first two terms on the
right side of the above equation, then this
reduces to:
TERMO = TERM1 - TERM2 + s^
To solve for the equilibrium terms, H,,,
is assigned a value initially. The above
equation is tested to determine whether the
A-3
-------
assigned value satisfies the conditions (i.e.,
does TERMO = TERM1 - TERM2 + s,,?).
If not, then iterate the process by assigning
a new value for H^, until an adequate degree
of accuracy is reached. In the following
examples, this method of solving for CCPP
was used with a tolerance of 0.001 for the
above equation.
Spreadsheet formats are provided to
assist in the design and development of a
CCPP calculation tool.
C. Finding the CCPP Value for a Specific
Water Quality Condition
A PWS performing lime softening has a
finished water with the following
characteristics: pH = 8.6; alkalinity = 90
mg/L as CaCO3; and calcium hardness =
100 mg/L as CaCO3. The worksheet
presented on the following page (Exhibit A-
1) calculates the CCPP (6.6 mg/L as
CaCO3) for this supply using the iterative
solution discussed above.
D. Finding the Water Quality Conditions
for a Desired CCPP.
To achieve a desired CCPP, any one
or more of the three key water quality
parameters may be modified. Exhibits A-
2 and A-3 demonstrate this by modifying
pH and alkalinity, respectively, to
achieve a desired CCPP of 8.0 mg/L as
CaCO3 for the same water described
above (Part B). When pH and calcium
held constant, the required alkalinity is
101.8 mg/L as CaCO3 for the targeted
CCPP; with alkalinity and calcium
contents are held constant, the resultant
pH of 8.8 is required to achieve the
desired CCPP.
Langelier Saturation Index. A
commonly used measure of a water's
ability to deposit calcium carbonate is
Langelier's Saturation Index (LI). This
value is determined by subtracting the
pH of saturation (known as pH,, and
dependent upon the calcium ion
concentration, alkalinity, temperature,
and dissolved solids concentration of the
water) from the actual pH (pHa).
A negative LI value indicates under-
saturation and a tendency for the water
to dissolve calcium carbonate. A positive
value indicates supersaturation and a
tendency for the water to deposit calcium
carbonate. A value of zero indicates that
the water is in chemical balance with
respect to calcium carbonate.
While the LI is widely used, it has
several notable shortcomings. Due to its
qualitative nature, it indicates only the
tendency or direction of calcium
carbonate precipitation. It cannot predict
the actual precipitation potential, or the
amount of excess calcium carbonate
available for precipitation.
For example, it has often been found
that although a positive LI was
maintained, severe corrosion had
occurred in the distribution system, and
inspections of pipe and fittings revealed
no evidence of a coating of calcium
carbonate. In other situations, however,
PWSs have had limited corrosion
problems with slightly negative LIs. In
these instances, the amount of alkalinity
may have been sufficient for carbonate
passivation to reduce corrosion activity.
In practice, the appropriate LI for a
given system is highly site-specific, and
is dependent upon treated water
composition and distribution system size
and complexity.
A-4
-------
Langelier Index (LI) = pH - pH.
Developed by W.F. Langelier
(1936)
LI = 0: CaCO3 saturation
LI > 0: CaCO3 supersaturation
LI < 0: CaCO3 undersaturation
Rule of Thumb Goal: +0.8 - +1.0
Calculation Procedure
A. Definition of Terms
pHg Saturation pH for calcium
carbonate calculated as follows:
pHB = -Iog10 [H+] - loglo fm
[H+] = (-B +/- B2 - 4AO/2A
where:
A = 1 -
B =
C = Kw'K2'[Ca2+]/KB'
and,
K,,' = Dissociation constant for water
Kg' = Solubility product constant for
calcium carbonate.
Kg' = Second dissociation constant
for carbonic acid.
fm = activity coefficient for the
monovalent ions. This term is
normally neglected in
calculating LI.
Conversion Between Total
Alkalinity and Dissolved Inorganic
Carbonate. To more easily utilize the
solubility contour diagrams presented
in Chapter 2.0 of this volume, Table A-
2 provides a conversion chart for total
alkalinity (Talk) and dissolved
inorganic carbonate (DIG) by water
temperature and pH. To use Table A-2,
a PWS with a known Talk (expressed
as mg CaCOg/L), pH, and water
temperature, find the factors A and B
corresponding to their conditions. The
equivalent DIG level for that water
supply can be calculated as follows:
DIG (mg CaCCyL) = [(Talk/50,000) + A] * B
The resulting DIG can be used in
finding the lead or copper solubility for
the defined condition per Figures 2-2,
2-3, and 2-5 in this volume.
References
Dye, J.F. 1952. Calculations of the
Effect of Temperature on pH, Free
Carbon Dioxide, and the Three Forms
of Alkalinity. J AWWA. 44(4):356.
Langlier, W.F. 1936. The Analytical
Control of Anti-Corrosion Water
Treatment. J AWWA. 28(10):1500.
Merrill, D.T. and R.L. Sanks. 1977a.
Corrosion Control by Deposition of
CaCOg Films: A Practical Approach for
Plant Operators - Part 1. J AWWA.
69(11):592.
Merrill, D.T. and R.L. Sanks. 1977b.
Corrosion Control by Deposition of
CaCO3 Films: A Practical Approach for
Plant Operators - Part 2. J AWWA.
69(12):634.
Merrill, D.T. and R.L. Sanks. 1978.
Corrosion Control by Deposition of
CaCOg Films: A Practical Approach for
Plant Operators - Part 3. J AWWA.
Nancollas, G.H. and M.M. Reddy. 1976.
Crystal Growth Kinetics of Minerals
Encountered in Water Treatment
Processes. Aqueous-Environmental
Chemistry of Metals. A. J. Rubin,
editor.(Ann Arbor Science Publ., Ann
Arbor, Michigan).
Schock, M.R. 1990. Internal Corrosion
and Deposition Control. Water Quality
and Treatment, 4th Ed. AWWA
(Denver, Colorado 80235).
A-5
-------
Exhibit A-l. CCCP Calculation Procedures
Example 1 - Spreadsheet for Calculating CCPP
Determine CCPP
Given:
pH=8.6
Alk=90 mg/l as CaCOS
Cal=100 mg/l as CaC03
Temp=20 C
NO Variable Definition
Comments Input Output
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
Ca = Calcium, moles/I (1)
Alki = Alkalinity, equiv/l . (1)
Hi = Hydrogen Ion, moles/I (1)
K'sp = Solubility Constant, CaCOS
K'w = Dissociation Constant for Water
K'1 = 1st Carbonic Dissociation Constant
K'2 = 2nd Carbonic Dissociation Constant
Req = (Heq - 2*K'2)/K'2
Peq = (2*Heq + K'1)/K'1
Teq = (2*K'2 + Heq)/Heq
Seq = Heq - K'w/Heq
Pi = (2*Hi + KMJ/K'I
Si = Hi - KWHi
Ti = (2*K'2 + HO/Hi
Acyi = ((Alki + Si)/Ti)*Pi + Si
Alkeq = Teq/Peq*(Acyi-Seq) - Seq, mg/l(3)
Terml = 2*K'sp*Req*Peq/T/(Acyi - Seq)
Term2 = (Acyi - Seq)*Teq/Peq
Heq = Equilibrium H. moles/1 (2)
TermO = 2*Ca - Alki
Right = Terml - Term2 + Seq
CCPP = Alki - Alkeq. mg/l as CaCO3 (3)
(1) Convert given information into proper
given
given
given
Table 2
Table 2
Table 2
Table 2
calculated
calculated
calculated
calculated
calculated
calculated
calculated
calculated
calculated
calculated
calculated
iterate
calculated
calculated
calculated
1.0E-03
1.8E-03
2.5E-09
5.4E-09
6.8E-15
4.2E-07
4.2E-11
3.0E+02
1.1E+00
1.0E+00
-5.2E-07
1.0E+00
-2.7E-06
1.0E+00
1.8E-03
83.4
1.9E-03
1.7E-03
1.3E-08
.2.0E-04
2.0E-04
6.6
units of moles/1 and equiv/l.
(2) Iteration can be accomplished by several procedures.
Manual iteration
requires the user to input various values of Heq until rows 20 and 21
converge. Another option in spreadsheets such as Lotus 123 and Micrsoft
Excel, allow cells to be "dependant1 on one another. In this case rows
20 and 21 could be equated and the recalculate key used to iterate.
Macros could also be written that would iterate using a loop command.
(3) Covert to mg/l by multiplying by 50,000.
A-6
-------
Exhibit A-3. CCCP Calculation Procedures
Example 3 - Spreadsheet for Calculating pH for Given CCPP
Determine pH Given:
(Based on information from Example 1)
NO Variable Definition
CCPP=8 mg/l
Alk=90 mg/l as CaC03
Cal=100 mg/l as CaCO3
Temp=20 C
Comments Input
Output
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
Hi = Hydrogen Ion, moles/I
Ca = Calcium, moles/I
Alki = Alkalinity, equiv/l
0) (2)
(1)
(D
K'sp = Solubility Constant CaCO3
K'w = Dissociation Constant for
K'1 = 1st Carbonic Dissociation
Water
Constant
K'2 = 2nd Carbonic Dissociation Constant
Req = (Heq - 2*K'2)/K'2
Peq.= (2*Heq + K'1)/K'1
Teq = (2*K'2 + Heq)/Heq
Seq = Heq - K'w/Heq
Pi = (2*Hi + K'1)/K*1
Si = Hi - ICw/Hi
Ti = (2*JC2 + Hi)/Hi
Acyi = ((Alki + Si)/Ti)*Pi + Si
Alkeq = Teq/Peq*(Acyi-Seq) - Seq, mg/I(4)
Terml = 2*K'sp*Req*Peq/T/(Acyi
Term2 = (Acyi - Seq)*Teq/Peq
Heq = Equilibrium H, moles/I
TermO = 2*Ca - Alki
Right = Terml — Temn2 + Seq
CCPP = Alki - Alkeq. mg/l
pH = pHi, - log Hi
-Seq)
(3)
(4)
vary
given
given
Table 2
Table 2
Table 2
Table 2
calculated
calculated
calculated
calculated
calculated
calculated
calculated
calculated
calculated
calculated
calculated
iterate
calculated
calculated
calculated
calculated
1.8E-09
1.0E-03
1.8E-03
5.4E-09
6.8E-15
4.2E-07
4.2E-11
1.2E-08
2.9E+02
1.1E+00
1.0E+00
-5.3E-07
1.0E+00
-3.8E-06
1.0E+00
1.7E-03
82.0
1.8E-03
1.6E-03
2.0E-04
2.0E-04
8.0
8.8
(1) Convert given information into proper units of moles/I and equiv/l.
(2) Enter in values for Hi (moles/l) and then iterate Heq as in
Example 1. Continue this process until CCPP converges to the targeted
goal value (8 mg/l for Example 3).
(3) Iteration can be accomplished by several procedures. Manual iteration
requires the user to input various values of Heq until rows 20 and 21
converge. Another option in spreadsheets such as Lotus 123 and Micrsoft
Excel, allow cells to be "dependant1 on one another. In this case rows
20 and 21 could be equated and the recalculate key used to iterate.
Macros could also be written that would iterate using a loop command.
(4) Covert to mg/l by multiplying by 50,000.
A-8
-------
Exhibit A-2. CCCP Calculation Procedures
Example 2 — Spreadsheet for Calculating Alkalinity for Given CCPP
Determine Alkalinity Given:
(Based on information from Example 1)
NO Variable Definition
CCPP=8 mg/l
pH=8.6
Cal=100 mg/l as CaCOS
Temp=20 C
Comments Input
Output
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
Alki = Alkalinity, equiv/l (1) (2)
Ca = Calcium, moles/l (1)
Hi = Hydrogen Ion, moles/l (1)
K'sp = Solubility Constant, CaCOS
K'w = Dissociation Constant for Water
K'1 = 1 st Carbonic Dissociation Constant
K'2 = 2nd Carbonic Dissociation Constant
Req = (Heq - 2*K'2)/K'2
Peq= (2*Heq + IC1)/K'1
Teq = (2*K'2 + Heq)/Heq
Seq = Heq - K'w/Heq
Pi= (2*Hi + K'1)/K'1
Si = Hi - K'w/Hi
Ti = (2*K'2 + Hi)/Hi
Acyi = ((Alki + Si)/Ti)*Pi -i- Si
Terml = 2*K'sp*Req*Peq/T/(Acyi - Seq)
Term2 = (Acyi - Seq)*Teq/Peq
Alkeq = Teq/Peq*(Acyi-Seq) - Seq, mg/l(4)
Heq = Equilibrium H, moles/I (3)
TermO = 2*Ca - AJki
Right = Terml - Term2 + Seq
CCPP = Alk - Alkf. mg/I (4)
Alki = Alkalinity, mg/l as CaCO3 (4)
vary
given
given
Table 2
Table 2
Table 2
Table 2
calculated.
calculated
calculated
calculated
calculated
calculated
calculated
calculated
calculated
calculated
calculated
iterate
calculated
calculated
calculated
calculated
2.0E-03
1.0E-03
2.5E-09
5.4E-09
6.8E-15
4.2E-07
4.2E-11
3.3E-J-02
1.1E+00
1.0E+00
-4.7E-07
1.0E+00
-Z7E-06
LOE+OO
2.0E-03
1.8E-03-
1.9E-03'
93.7
1.4E-08
-3.5E-05
-3.5E-05
8.0
101.8
(1 ) Convert given information into proper units of moles/I and equiv/l.
(2) Enter in values for alkalinity (moles/0 and ^ert iterate Heq as in
Example 1 . Continue this process until CCPP converges to the targeted
goal value (8 mg/l for Example 2).
(3) Iteration can be accomplished by several procedures. Manual iteration
requires the user to input various values of Heq until rows 20 and 21
converge. Another option in spreadsheets such as Lotus 123 and Micrsoft
Excel, allow cells to be "dependant" on one another. In this case rows
20 and 21 could be equated and the recalculate key used to iterate.
Macros could also be written that would iterate using a loop command.
(4) Covert to mg/l by multiplying by 50,000.
A-7
-------
Table A-2
Total Alkalinity (TALK) to Dissolved Inorganic Carbonate (DIG) Conversion Variables
Determine A & B via the below table. Then compute DIC as follows:
DIG as ppm CaCO3 = ((TALK in ppm CaCO3/50,000) + A] * B
PH
6.0
6.1
6.2
6.3
6.4
6.5
6.6
6.7
6.8
6.9
7.0
7.1
7.2
7.3
7.4
7.5
7.6
7.7
7.8
7.9
8.0
8.1
8.2
8.3
8.4
8.5
8.6
8.7
8.8
8.9
T=0 (deg. C)
A B
9.99E-07 2.40E+05
7.93E-07 2.01 E+05
6.29E-07 1.70E+05
4.99E-07 1.45E+05
3.95E-07 1.25E+05
3.13E-07 1.10E+05
2.47E-07 9.76E+04
1.94E-07 8.78E+04
1.51E-07 8.00E+04
1.17E-07 7.39E+04
8.83E-08 6.89E+04
6.46E-08 6.50E+04
4.45E-08 6.19E+04
2.67E-08 5.95E+04
1.03E-08 5.75E+04
-5.53E-09 5.60E+04
-2.17E-08 5.47E+04
-3.89E-08 5.37E+04
-5.83E-08 5.29E+04
-8.07E-08 5.23E+04
-1.07E-07 5.18E+04
-1.40E-07 5.13E+04
-1.80E-07 5.10E+04
-2.29E-07 5.07E+04
-2.91 E-07 5.05E+04
-3.68E-07 5.02E+04
-4.65E-07 5.00E+04
-5.87E-07 4.98E+04
-7.40E-07 4.96E+04
-9.32E-07 4.93E+04
T=5 (deg. C)
A B
9.98E-07 2.14E+05
7.92E-07 1.81 E+05
6.28E-07 1.54E+05
4.97E-07 1.32E+05
3.93E-07 1.15E+05
3.10E-07 1.02E+05
2.44E-07 9.13E+04
1.90E-07 8.28E+04
1.47E-07 7.60E+04
1.11 E-07 7.07E+04
8.14E-08 6.64E+04
5.60E-08 6.30E+04
3.36E-08 6.03E+04
1.30E-08 5.82E+04
-6.96E-09 5.65E+04
-2.73E-08 5.51 E+04
-4.90E-08 5.41 E+04
-7.34E-08 5.32E+04
-1.02E-07 5.25E+04
-1.35E-07 5.20E+04
-1.76E-07 5.15E+04
-2.26E-07 5.11E+04
-2.89E-07 5.08E+04
-3.67E-07 5.05E+04
-4.64E-07 5.03E+04
-5.86E-07 5.01 E+04
-7.39E-07 4.99E+04
-9.31 E-07 4.96E+04
-1.17E-06 4.94E+04
-1.48E-06 4.91 E+04
T«10(deg. C)
A B
9.97E-07 1.96E+05
7.91 E-07 1.66E+05
6.26E-07 1.42E+05
4.95E-07 1.23E-t-05
3.91 E-07 1.08E+05
3.07E-07 9.60E+04
2.39E-07 8.65E+04
1.85E-07 7.90E+04
1.40E-07 7.30E+04
1.02E-07 6.83E+04
7.05E-08 6.45E+04
4.23E-08 6.15E+04
1.63E-08 5.91 E+04
-8.77E-09 5.73E+04
-3.43E-08 5.57E+04
-6.17E-08 5.45E+04
-9.24E-08 5.36E+04
-1.28E-07 5.28E+04
-1.70E-07 5.22E+04
-2.22E-07 5.17E+04
-2.85E-07 5.13E+04
-3.64E-07 5.09E+04
-4.61 E-07 5.07E+04
-5.84E-07 5.04E+04
-7.37E-07 5.02E+04
-9.30E-07 4.99E+04
-1.17E-06 4.97E+04
-1.48E-06 4.95E+04
-1.86E-06 4.92E+04
-2.34E-06 4.90E+04
T=15(deg. C)
A B
9.96E-07 1.81E+05
7.89E-07 1.54E+05
6.24E-07 1.33E+05
4.92E-07 1.16E+05
3.87E-07 1.02E+05
3.02E-07 9.15E+04
2.33E-07 8.29E+04
1.77E-07 7.62E+04
1.30E-07 7.08E+04
9.04E-08 6.65E+04
5.53E-08 6.31E+04
2.32E-08 6.04E+04
-7.70E-09 5.82E+04
-3.90E-08 5.65E+04
-7.24E-08 5.52E+04
-1.10E-07 5.41 E+04
-1.53E-07 5.32E+04
-2.04E-07 5.25E+04
-2.66E-07 5.20E+04
-3.42E-07 5.15E+04
-4.37E-07 5.11 E+04
-5.54E-07 5.08E+04
-7.02E-07 5.05E+04
-8.86E-07 5.03E+04
-1.12E-06 5.01 E+04
-1.41E-06 4.98E+04
-1.78E-06 4.96E+04
-2.24E-06 4.94E+04
-2.82E-06 4.91 E+04
-3.55E-06 4.88E+04
T=20 (deg. C)
A B
9.93E-07 1.70E+05
7.86E-07 1.45E+05
6.20E-07 1.26E+05
4.88E-07 1.10E+05
3.81 E-07 9.78E+04
2.95E-07 8.80E+04
2.24E-07 8.02E+04
1.66E-07 7.40E+04
1.16E-07 6.90E+04
7.22E-08 6.51 E+04
3.24E-08 6.20E+04
-5.68E-09 5.95E+04
-4.41 E-08 5.75E+04
-8.48E-08 5.60E+04
-1.30E-07 5.47E+04
-1.82E-07 5.37E+04
-2.44E-07 5.29E+04
-3.19E-07 5.23E+04
-4.11 E-07 5.18E+04
-5.24E-07 5.13E+04
-6.66E-07 5.10E+04
-8.43E-07 5.07E+04
-1.07E-06 5.04E+04
-1.34E-06 5.02E+04
-1.69E-06 5.00E+04
-2.13E-06 4.97E+04
-2.69E-06 4.95E+04
-3.39E-06 4.92E+04
-4.26E-06 4.89E+04
-5.37E-06 4.86E+04
T=25 (deg. C)
A B
9.90E-07 1.62E+05
7.82E-07 .1.39E+05
6.15E-07 1.21E+05
4. 81 E-07 1.06E+05
3.73E-07 9.48E+04
2.85E-07 8.55E+04
2. 11 E-07 7.82E+04
1.49E-07 7.24E+04
9.54E-08 6.78E+04
4.65E-08 6.41 E+04
1.53E-21 6.12E+04
-4. 65 E-08 5.89E+04
-9.54E-08 5.70E+04
-1.49E-07 5.56E+04
-2.11 E-07 5.44E+04
-2.85E-07 5.35E+04
-3.73E-07 5.27E+04
-4.81 E-07 5.21 E+04
-6.15E-07 5.16E+04
-7.82E-07 5.12E+04
-9.90E-07 5.09E+04
-1.25E-06 5.06E+04
-1.58E-06 5.03E+04
-1.99E-06 5.01 E+04
-2.51 E-06 4.99E+04
-3.16E-06 4.96E+04
-3.98E-06 4.94E+04
-5.01 E-06 4.91 E+04
-6.31 E-06 4.88E+04
-7.94E-06 4.84E+04
-------
Table A-2
Total Alkalinity (TALK) to Dissolved Inorganic Carbonate (DIG) Conversion Variables
Determine A & B via the below table. Then compute DIC as follows:
DIC as ppm CaCO3 - [(TALK In ppm CaCO3/50,000) + A] • B
PH
9.0
9.1
9.2
9.3
9.4
9.5
9.6
9.7
9.8
9.9
10.0
10.1
10.2
10.3
10.4
10.5
10.6
10.7
10.8
10.9
11.0
11.1
11.2
11.3
11.4
11.5
11.6
11.7
11.8
11.9
12.0
T=0 (deg. C)
A B
-1.17E-06 4.90E+04
-1.48E-06 4.87E+04
-1.86E-06 4.84E+04
-2.34E-06 4.79E+04
-2.95E-06 4.74E+04
-3.72E-06 4.68E+04
-4.68E-06 4.61 E+04
-5.89E-06 4.52E+04
-7.41 E-06 4.43E+04
-9.33E-06 4.32E+04
-1.17E-05 4.20E+04
-1.48E-05 4.07E+04
-1.86E-05 3.93E+04
-2.34E-05 3.78E+04
-2.95E-05 3.64E+04
-3.72E-05 3.50E+04
-4.68E-05 3.37E+04
-5.89E-05 3.24E+04
-7.41 E-05 3.13E+04
-9.33E-05 3.02E+04
-1.17E-04 2.94E+04
-1.48E-04 2.86E+04
-1.86E-04 2.79E+04
-2.34E-04 2.74E+04
-2.95E-04 2.69E+04
-3.72E-04 2.66E+04
-4.68E-04 2.63E+04
-5.89E-04 2.60E+04
-7.41 E-04 2.58E+04
-9.2^-04 2.56E+04
-1.^B03 2.55E+04
T=5 (deg. C)
A B
-1.86E-06 4.88E+04
-2.34E-06 4.85E+04
-2.95E-06 4.81 E+04
-3.71 E-06 4.76E+04
-4.68E-06 4.70E+04
-5.89E-06 4.63E+04
-7.41 E-06 4.55E+04
-9.33E-06 4.46E+04
-1.17E-05 4.35E+04
-1.48E-05 4.24E+04
-1.86E-05 4.11 E+04
-2.34E-05 3.97E+04
-2.95E-05 3.83E+04
-3.72E-05 3.69E+04
-4. 68 E-05 3.54E+04
-5.89E-05 3.41 E+04
-7.41 E-05 3.28E+04
-9.33E-05 3.16E+04
-1.17E-04 3.06E+04
-1.48E-04 2.96E404
-1.86E-04 2.88E+04
-2.34E-04 2.81 E+04
-2.95E-04 2.76E+04
-3.72E-04 2.71 E+04
-4.68E-04 2.67E+04
-5.89E-04 2.63E+04
-7.41 E-04 2.61 E+04
-9.33E-04 2.59E+04
-1.17E-03 2.57E+04
-1.48E-03 2.56E+04
-1.86E-03 2.54E+04
T«10(deg. C)
A B
-2.95E-06 4.86E+04
-3.71 E-06 4.82E+04
-4.68E-06 4.78E+04
-5.89E-06 4.72E+04
-7.41 E-06 4.66E+04
-9.33E-06 4.58E+04
-1.17E-05 4.49E+04
-1.48E-05 4.39E+04
-1.86E-05 4.28E+04
-2.34E-05 4.15E+04
-2.95E-05 4.02E+04
-3.72E-05 3.88E+04
-4.68E-05 3.73E+04
-5.89E-05 3.59E+04
-7.41 E-05 3.45E+04
-9.33E-05 3.32E+04
-1.17E-04 3.20E+04
-1.48E-04 3.09E+04
-1.86E-04 2.99E+04
-2.34E-04 2.91 E+04
-2.95E-04 2.83E+04
-3.72E-04 2.77E+04
-4.68E-04 2.72E+04
-5.89E-04 2.68E+04
-7.41 E-04 2.64E+04
-9.33E-04 2.62E+04
-1.17E-03 2.59E+04
-1.48E-03 2.57E+04
-1.86E-03 2.56E+04
-2.34E-03 2.55E+flfe
-2.95E-03 2.54E-M
T=15(deg. C)
A B
-4.47E-06 4.84E+04
-5.62E-06 4.80E+04
-7.08E-06 4.74E+04
-8.91 E-06 4.68E+04
-1.1 2 E-05 4.61 E+04
-1.41 E-05 4.53E+04
-1.78E-05 4.43E+04
-2.24E-05 4.32E+04
-2.82E-05 4.20E+04
-3.55E-05 4.07E+04
-4.47E-05 3.94E+04
-5.62E-05 3.79E+04
-7.08E-05 3.65E+04
-8.91 E-05 3.51 E+04
-1.12E-04 3.37E+04
-1.41 E-04 3.25E+04
-1.78E-04 3.13E+04
-2.24E-04 3.03E+04
-2.82E-04 2.94E+04
-3.55E-04 2.86E+04
-4.47E-04 2.80E+04
-5.62E-04 2.74E+04
-7.08E-04 2.70E+04
-8.91 E-04 2.66E+04
-1.12E-03 2.63E+04
-1.41E-03 2.60E+04
-1.78E-03 2.58E+04
-2.24E-03 2.57E+04
-2.82E-03 2.55E+04
-3.55E-03 2.54E+04
F-4.47E-03 2.53E+04
T=20 (deg. C)
A B
-6.76E-06 4.82E+04
-8.51 E-06 4.77E+04
-1.07E-05 4.71 E+04
-1.35E-05 4.65E+04
-1.70E-05 4.57E+04
-2.14E-05 4.48E+04
-2.69E-05 4.38E+04
-3.39E-05 4.26E+04
-4.27E-05 4. 14 E+04
-5.37E-05 4.00E+04
-6.76E-05 3.86E+04
-8.51 E-05 3.72E+04
-1.07E-04 3.57E+04
-1.35E-04 3.43E+04
-1.70E-04 3.30E+04
-2.14E-04 3.18E+04
-2.69E-04 3.08E+04
-3.39E-04 2.98E+04
-4.27E-04 2.90E+04
-5.37E-04 2.83E+04
-6.76E-04 2.77E+04
-8.51 E-04 2.72E+04
-1.07E-03 2.67E+04
-1.35E-03 2.64E+04
-1.70E-03 2.61 E+04
-2.14E-03 2.59E+04
-2.69E-03 2.57E+04
-3.39E-03 2.56E+04
-4.27E-03 2.55E+04
-5.37E-03 2.54E+04
-6.76E-03 2.53E+04
T«25 (deg. C)
A B
-1.00E-05 4.80E+04
-1.26E-05'4.74E+04
-1.58E-05 4.68E+04
-2.00E-05 4.61 E+04
-2.51 E-05 4.53E+04
-3.16E-05 4.43E+04
-3.98E-05 4.32E+04
-5.01 E-05 4.20E+04
-6.31 E-05 4.07E+04
-7.94E-05 3.93E+04
-1.00E-04 3.79E+04
-1.26E-04 3.65E+04
-1.58E-04 3.51 E+04
-2.00E-04 3.37E+04
-2.51 E-04 3.25E+04
-3.16E-04 3.13E+04
-3.98E-04 3.03E+04
-5.01 E-04 2.94E+04
-6.31 E-04 2.86E+04
-7.94E-04 2.80E+04
-1.00E-03 2.74E+04
-1.26E-03 2.70E+04
-1.58E-03 2.66E+04
-2.00E-03 2.63E+04
-2.51 E-03 2.60E+04
-3.16E-03 2.58E+04
-3.98E-03 2.57E+04
-5.01 E-03 2.55E+04
-6.31 E-03 2.54E+04
-7.94E-03 2JK+04
-1.00E-02 2V+04
-------
Appendix B
-------
PWS Characterization: Low pH, Alkalinity, & Calcium
PWS Name & Location: Bennington, Vermont
^ Item
i
ii
HI
IV
Case Study Number
Description
1
PWS Description
a. Raw Water Supply
b. Water Quality
1. Raw
2. Treated
c. Treatment
Surface water
Low pH, alkalinity and calcium content.
Coagulated water pH 4.5 - 5.0, alkalinity < 5 mg CaCO3/L.
Filtration, chlorination.
Corrosion Control Study Elements
a. Desk-top
Evaluation
b. Demonstration
Testing
1. Row-Through
2. Static
3. Full-Scale
c. Full-Scale
Confirmation
1 . Tap sampling
2. WQP-DIS
3. WQP-POE
• Community blood lead monitoring program revealed elevated levels in 1 977.
• Material survey of the distribution system found that one-third of the system was
served by lead service lines.
• Tap samples were collected, finding lead levels as high as 0.86 mg/L.
• Theoretical performance of carbonate passivation.
NA
NA
Implemented pH and alkalinity adjustment treatment
Monthly first-draw and running tap samples from 1 0 targeted sites.
pH, alkalinity, and scale analysis using X-ray diffraction.
pH and alkalinity
Testing Program Desrlption
a. Constraints
b. Priorities
1. Primary
2. Secondary
c. Treatment
Alternatives
d. Monitoring
programs
e. QA/QC Elements
Due to the elevated blood lead levels in the population, treatment installation had to be
easily and quickly installed and operable.
Reduce lead levels in consumers' tap water.
pH and alkalinity adjustment: Increase pH to 8.0-6.5 and increase alkalinity to above
25 mg CaCO3/L.
• Tap monitoring for lead in first-draw and running samples: 1977-1991.
pH and alkalinity monitoring in distribution system and at the POEs: 1977-1991.
• Evaluate scale on lead service line pipes using X-ray diffraction: 1977.
See Reference
B-1
-------
PWS Characterization:
PWS Name & Location:
Low pH, Alkalinity, & Calcium
Bennington, Vermont
Item
V
VI
VII
Description
Testing Results
a. Corrosion Control
Performance
b. Secondary Impacts
c. Treatment Issues
Notes/Qual If (cations
Reference (s)
Average monthly lead levels in first-draw tap samples were reduced from a high
of 250 ug/L to approximately 20 ug/L lead within six-months of operations.
Ongoing monitoring has showed a continual decline in lead levels in first-draw tap
samples with early 1991 data indicating lead levels less than 10 ug/L.
X-ray diffraction analysis confirmed scale formation consisting of cerrusite
(PbCO3) and hydrocerrusite (Pb3(CO3)2(OH)2).
NA
Initial operation of the sodium bicarbonate and sodium hydroxide feed systems was
manual, and targeted values for pH and alkalinity were not always achieved. To
improve treatment consistency, automated operational controls were installed in 1990.
NA
Vinci, A. 1 992. Bennington, Vermont Corrosion Control Studies with the
Bicarbonate/pH System. Technical comments submitted to USEPA in response to the
proposed Lead and Copper Rule from Church & Dwight, Company, 469 N. Harrison
St, Princeton, NJ, 08543-5297.
B-2
-------
PWS Characterization: Low pH, Alkalinity, & Calcium
PWS Name & Location: MWRA, Boston, Massachussetts
Item
i
ii
in
IV
Case Study Number
Description
2
PWS Description
a. Raw Water Supply
b. Water Quality
1. Raw
2. Treated
c. Treatment
Unfiltered surface water.
Low turbidity, low pH-Alkalinity-Calcium.
Finished water quality: pH 6.5-6.7; Total hardness 12 mg CaCO3/L; Alkalinity 12 mg
CaCO3/L; Total dissolved solids 37 mg/L.
Chlorination, ammoniation, fluoridation.
Corrosion Control Study Elements
a. Desk-top
Evaluation
b. Demonstration
Testing
1. Row-Through
2. Static
3. Full-Scale
c. Full-Scale
Confirmation
1 . Tap sampling
2. WQP-DIS
3. WQP-POE
NA
NA
NA
Alternative treatments were implemented system-wide.
Diagnostic and verification tap sampling was performed.
pM, alkalinity, and inhibitor residual
pH, alkalinity, and inhibitor residual
Testing Program Desription
a. Constraints
b. Priorities
1. Primary
2. Secondary
c. Treatment
Alternatives
d. Monitoring
programs
e. QA/QC Elements
Treatment alternative had to be compatible with the facilities of an unfiltered supply.
Reduce lead levels at consumers' taps.
Reduce copper levels at consumers' taps.
Orthophosphate inhibitor and pH adjustment.
First-draw tap sampling for lead and copper; WQP-POE; and WQP-DIS.
NA '
B-3
-------
PWS Characterization:
PWS Name & Location:
Low pH, Alkalinity, & Calcium
MWRA, Boston, Massachussetts
item
V
VI
VII
Description
Testing Results
a. Corrosion Control
Performance
b. Secondary
Impacts
c. Treatment Issues
Notes/Qualifications
Reference(s)
• Six-month trial full-scale treatment using zinc orthophosphate:
1 . Initial passivation dose of 1 3 mg/L used for several weeks, then reduced the
dosage to between 3.2 and 4.5 mg/L.
2. Initial increase in tap lead levels observed, then slow decline in lead noted
toward the end of the six month period.
3. Algal growth appeared to be stimulated in the open, finished water storage
reservoirs due to the additional phosphate content.
pH Adjustment using sodium hydroxide was subsequently installed.
1 . Lead and copper levels in first-draw samples were reduced by increased pH.
2. Researchers noted that when the pH dropped from pH 9 to below pH 8, the
lead levels increased.
• Algal growth was stimulated in the open, treated water reservoirs when zinc
orthophosphate was used.
• Dirty water complaints arose with the introduction of the phosphate inhibitor.
A high pH at the POE was necessary to maintain targeted pH values throughout
the distribution system.
The poor performance of the zinc orthophosphate inhibitor is most likely the result of
an excessively low pH for its effectiveness. Had the treated water pH been increased
to above 7, it is likely that the performance results would have been improved.
Karalekas, P.C. et al. 1983. Control of Lead, Copper, and Iron Pipe Corrosion in
Boston. Journal AWWA 75(2):92-95.
B-4
-------
PWS Characterization: Low pH, Alkalinity, & Calcium
PWS Name & Location: FCWD, Fort Collins, Colorado
Item
i
ii
in
IV
Case Study Number
Description
3
PWS Description
a. Raw Water Supply
b. Water Quality
1. Raw
2. Treated
c. Treatment
Poudre River & Horsetooth Reservoir
Cold, low turbidity, moderate pH, low alkalinity and low calcium
Coagulated Water: pH = 5.8 - 7.2, Alkalinity = 5-25 mg CaCOJL, and Calcium =
20-30 mg CaCOa/L
Alum coagulation, fluoridation, and chlorination
Corrosion Control Study Elements
a. Desk-top
Evaluation
b. Demonstration
Testing
1. Row-Through
2. Static
3. Full-Scale
c. Full-Scale
Confirmation
1 . Tap sampling
2. WQP-DIS
3. WQP-POE
Analogous Systems: Seattle, WA and Bennington, VA
Theoretical: Evaluation of carbonate passivation
Process Testing: Marble Chip Testing
NA
NA
Implemented pH/alkalinity adjustment full-scale in two stages to optimize treatment.
First-draw samples from public taps: 1981 - 1992.
WQP-DIS for pH, alkalinity and calcium: 1981 - 1992.
WQP-POE for pH, alkalinity and calcium: 1981 - 1992.
Testing Program Desription
a. Constraints
b. Priorities
1. Primary
2. Secondary
c. Treatment
Alternatives
d. Monitoring
programs
e. QA/QC Elements
• Brewery which required consistent, moderate pH.
• Land application of sewage sludge limited by copper.
• Use of phosphate inhibitor questionable due to wastewater treatment requirements
and public acceptance.
Reduce copper levels in tap water and sewage sludge.
No adverse effects on other water treatment goals or aesthetic quality of the treated
water.
pH and alkalinity adjustment using lime and sodium bicarbonate.
Stage 1 : pH Goal = 7.6-7.8 & Alk Goal > 30 mg CaCOjfl.-
Stage 2: pH Goal = 7.8-8.0 & Alk Goal = 35-45 mg CaCOj/L
• In-line pH monitors located after stabilization chemical feed points.
Alkalinity measured at POE every 4-hours.
• 8-10 sampling stations monitoring monthly for first-draw copper and WQP-DIS.
Process controls.
B-5
-------
PWS Characterization:
PWS Name & Location:
Low pH, Alkalinity, & Calcium
FCWD, Fort Collins, Colorado
Item
Description
Testing Results
a. Corrosion Control
Performance
Tap copper levels were reduced from high levels ranging between 0.8-1.0 mg/L to
maximum values between 0.2-0.4 mg/L
Sludge metal content reduced: Copper 20%; Lead 3O-50%.
b. Secondary
Impacts
Post-filtration turbidity spikes with lime addition
Elevating pH caused post-precipitation of manganese during period of reservoir
stratification. This caused brown water complaints. FCWD installed potassium
permanganate pretreatment to control soluble manganese present after filtraiton.
c. Treatment Issues
Process control for stable and consistent final pH took between one and two years
to debug.
FCWD has been able to achieve the pH and alkalinity goals over 90 percent of the
time.
For more cost-effective treatment FCWD is installing carbon dioxide in lieu of
sodium bicarbonate.
Redundant feed systems are being installed for lime and carbon dioxide to ensure
continuous operation.
VI
Notes/Qualifications
During the first incident of manganese post-precipitation, FCWD stopped the pH
adjustment portion of their corrosion control program. Within days of this, copper levels
began to increase in first-draw tap samples, and the copper and lead content of the
sewage sludge increased during the period when pH adjustment was not being
practiced. This indicated to FCWD that (1) effective corrosion control could only be
assured if continuously practiced; and (2) while the loss of corrosion protection became
apparent in a matter of days, it took several weeks to months to regain the control
conditions experienced prior to the treatment interruption.
VII
Reference (s)
Smith, M. et al. 1992. Corrosion Control Studies and Strategies - Fort Collins, Colorado.
AWWA Corrosion Control Seminar (Denver, CO).
Kuchenrither, R.D. et al. 1988. Sludge Quality Benefits Realized from Drinking Water
Stabilization. Proc. Annual WPCF Conference.
Elmund, G.K. et al. 1986. Stablilization of a Finished Water: Fort Collins, Colorado.
Proc Joint Regional AWWA-WPCA Conference.
B-6
-------
PWS Characterization: Low pH, Alkalinity, & Calcium
PWS Name & Location: Bureau of Water, Portland, Oregon
• Hem
i
ii
in
IV
Case Study Number
Description
4
PWS Description
a. Raw Water Supply
b. Water Quality
1. Raw
2. Treated
c. Treatment
Unfiltered surface water supply
Low pH, alkalinity, and calcium
Finished water quality: pH 6.9; Total hardness 14 mg CaCO3/L; Alkalinity 10 mg
CaCO3/L; Total dissolved solids 24 mg/L
Chlori nation/chloramination .
Corrosion Control Study Elements
a. Desk-top
Evaluation
b. Demonstration
Testing
1. Row-Through
2. Static
3. Full-Scale
c. Full-Scale
Confirmation
1 . Tap sampling
2. WQP-DIS
3. WQP-POE
Theoretical: Carbonate passivation
Analogous Systems: Seattle, Washington -
Coupons and copper tubing
NA
NA
NA
NA
NA
Testing Program Desription
a. Constraints
b. Priorities
1. Primary
2. Secondary
c. Treatment
Alternatives
d. Monitoring
programs
e. QA/QC Elements
Compatibility with unfiltered surface water treatment facilities.
Assess the corrosion rates of domestic plumbing materials.
Determine the most cost effective approach to treatment
Existing treatment at two locations in the distribution system:
1 . Pipe rig consisted of coupon (6) flow-through units with black iron, galvanized
steel, copper, lead, lead:tin solder-coated copper, and asbestos-cement
2. A single loop (220 feet) of lead tin soldered copper tubing was also included
in the pipe rig. Soldered joints were placed every 20-feet.
3. Pipe rig 1 was located at the source of supply with a free chlorine residual of
1 mg/L
4. Pipe rig 2 was located several miles from the source and chloramines were
added.
Metal leaching and water quality parameters enterring the pipe loops systems.
NA
B-7
-------
PWS Characterization:
PWS Name & Location:
Low pH, Alkalinity, & Calcium
Bureau of Water, Portland, Oregon
Hem
V
VI
VII
Description
Testing Results
a. Corrosion Control
Performance
b. Secondary Impacts
c. Treatment Issues
Notes/Qualifications
Reference(s)
Equilibrium corrosion rates appeared to result after six months of operation for all
of the materials.
• Based on the test coupons, lead corrosion rates increased with free chlorine as
compared to chloramines.
• All other materials experienced comparable corrosion rates (on the basis of
coupons) regardless of the disinfectant present.
• The copper tubing with leadrtin solder showed increased corrosion activity with
chloramines as compared to the free chlorine loop.
NA
The source of lead in first-draw tap samples appeared to be the lead-based solder as
confirmed by the pipe loop testing program. No treatment was recommended since
the City of Portland had instituted a lead ban on plumbing materials for domestic
supply systems.
Portland has participated in the AWWARF Pipe Loop Study, and more information on
the corrosion behavior of it system, especially as it relates to first-draw tap samples
will be available in the final report
Treweek, G.P. et al. 1 985. Pilot-plant Simulation of Corrosion in
Domestic Pipe Materials. Journal AWWA 77(10):74-82.
B-8
-------
PWS Characterization:
PWS Name & Location:
Low pH, Alkalinity, & Calicum
SWD, Seattle, Washington
Item
i
ii
in
IV
Case Study Number
Description
5
PWS Description
a. Raw Water Supply
b. Water Quality
1. Raw
2. Treated
c. Treatment
Two unfiltered surface water supplies - Tolt & Cedar Rivers
Low pH, alkalinity, calcium and mineral content.
Finished water quality: pH 5.7-6.2; Alkalinity 3-5 mg CaCO3/L; Chlorine residual 0.2-
0.4 mg/L.
Chlori nation and fluoridation.
Corrosion Control Study Elements
a. Desk-top
Evaluation
b. Demonstration
Testing
1. Row-Through
2. Static
3. Full-Scale
c. Full-Scale
Confirmation
1 . Tap sampling
2. WQP-DIS
3. WQP-POE
Theoretical: Carbonate passivation
Series of pilot loop tests conducted prior to full-scale treatment installation.
NA
Corrosion rate and metal leaching studies conducted after installing treatment.
First-draw tap samples collected from 300 homes in service area.
pH, alkalinity, chlorine residual, dissolved oxygen, conductivity
pH, alkalinity, chlorine residual
Testing Program Desription
a. Constraints
b. Priorities
1. Primary
2. Secondary
c. Treatment
Alternatives
d. Monitoring
programs
e. QA/QC Elements
Compatibility with unfiltered, surface water facilities.
Reducing the corrosion activity in the distribution system toward lead, copper, zinc
(galvanized piping) and iron.
Reliability and operational feasibility of the selected treatment process.
• Pipe loop testing evaluated pH/alkalinity alternatives and orthophosphate
corrosion inhibitors:
1 Recirculating pipe loops were constructed using 1/2-inch copper tubing in
which a bead of 50:50 lead: tin solder was attached longitudinally in the
piping.
2. Copper tubing lengths were 6-8 inches individually; then several were
connected using plastic tubing.
3. Treated water was circulated through a test loop using a peristaltic pump
cyclically.
• Corrosion rate testing was performed using linear polarization techniques once
full-scale treatment was installed.
Metal leaching, corrosion rates, and water quality parameters were monitored in the
flow-through testing apparatus and in the full-scale systems once the recommended
treatment was installed.
See Reference Materials.
B-9
-------
PWS Characterization:
PWS Name & Location:
Low pH, Alkalinity, & Calicum
SWD, Seattle, Washington
Item
V
VI
VII
Description
Testing Results
a. Corrosion Control
Performance
b. Secondary Impacts
c. Treatment Issues
Notes/Qualifications
Reference (s)
• pH and alkalinity adjustment was the recommended treatment on the basis of the
flow-through testing program.
• After installation, reductions in tap lead and copper levels (as first-draw samples)
of 12 and 60 percent were found within the first year of operation.
• Electrochemical testing results showed a 50% decrease in corrosion rates on
new copper plumbing with greater decreases in aged materials.
• Short-term variations in copper corrosion rates were found to be strongly
correlated to free chlorine (direct relationship), and to a lesser degree with pH
(inverse relationship).
NA
pH and alkalinity adjustment took place gradually over the first year of operations,
increasing the average pH to 7.8-8.3 and the average alkalinity to 15-17 mg
CaCO3/L.
Reiber, S.H. et al. 1987. Corrosion Monitoring and Control in the Pacific Northwest.
Journal AWWA 79(2):71-74.
Stone, A. et al. 1987. The Effects of Short-term Changes in Water Quality on Copper
and Zinc Corrosion Rates. Journal AWWA. 79(2):75-82.
Reiber, S.H. et al. 1987. Corrosion in Water Distribution Systems of the Pacific
Northwest. EPA 600/S2-87-042.
Herrara, C.E. et al. 1984. Seattle Distribution System Corrosion Control Study -
Volume 2: Tolt River Water Pilot Plant Study. EPA 600/2-84-065.
Hoyt, B.P. et al. 1982. Seattle Distribution System Corrosion Control Study - Volume
1: Cedar River Water Pilot Plant Study. EPA 600/2-82-026. .
B-10
-------
PWS Characterization: Low pH, High Alkalinity & Calcium
PWS Name & Location: Oakwood, Ohio
J item
i
ii
in
Case Study Number
Description
6
PWS Description
a. Raw Water Supply
b. Water Quality
1. Raw
2. Treated
c. Treatment
Two well-fields
Lower pH, High alkalinity and calcium, elevated iron.
Finished water quality: pH 7.1; Total hardness 200 mg CaCO3/L; Alkalinity
CaCO3/L
Water from one well-field removes iron through green-sand filtration and is
using zeolite softening. The treated water is then blended with water from
well-field and chlorinated prior to distribution.
370 mg
then softened
the other
Corrosion Control Study Elements
a. Desk-top
Evaluation
b. Demonstration
Testing
1. Row-Through
2. Static
3. Full-Scale
c. Full-Scale
Confirmation
1 . Tap sampling
2. WQP-DIS
3. WQP-POE
NA
Pipe loop study to evaluate corrosion rates before and after softening.
NA
Tap sampling before and after softening to determine difference in corrosion activity.
First-draw tap samples at sites receiving water only from Well-field 1 prior to being
blended with water from the second well-field.
pH, alkalinity, hardness
pH, alkalinity, hardness
B-11
-------
PWS Characterization:
PWS Name & Location:
Low pH, High Alkalinity & Calcium
Oakwood, Ohio
item
IV
Description
Testing Program DesrlptJon
a. Constraints
b. Priorities
1. Primary
2. Secondary
c: Treatment
Alternatives
d. Monitoring
programs
e. QA/QC Elements
NA
Determine if softening would impact corrosion rates.
• Row-through testing program:
1 . Two pipe rigs were constructed with one receiving unsoftened water and the
other receiving softened water.
2. Each rig had three test loops: black iron, lead, and pipe sleeves consisting of
black iron, copper, leadrtin solder-coated copper, and galvanically coupled
(copper and lead solder) coupons (4 coupons of each).
3. Each pipe loop (black iron and lead) was pre-conditioned by receving only hard
water for 7-months.
4. The pipe sleeves received hard and soft water throughout the entire testing
period.
5. Rowrate conditions for the pipe sleeves were: 0.5 gpm at 5 psi for six days, with
one day of standing tme.
6. Rowrate conditions for the pipe loops were: Iron pipe, recirculating rate of 1
gpm with an effluent rate of 0.0172 gpm; Lead pipe flow-through rate of 0.5
gpm for 1 6 hours with an 8-hour standing time cyclically operated.
7. Each pipe rig operated two lead loops: one was of new material and the second
was excavated (old material) from the distribution system.
Full-scale evaluation testing program consistent of sampling consumers' homes for
metals and water quality parameters while receiving hard and softened water.
• Flow-through testing program:
In addition to the WQP monitoring performed were the following:
Pipe loops: Metal leaching
Pipe Sleeves: Metal leaching and coupon weight-loss
Dissolved oxygen depletion was used to calculate the corrosion rates in the iron
pipe loops.
• Full-scale evaluation:
First-draw samples for lead, copper, pH, alkalinity, and hardness.
See Reference
B-12
-------
PWS Characterization:
PWS Name & Location:
Low pH, High Alkalinity & Calcium
Oakwood, Ohio
item
V
VI
VII
Description
Testing Results
a. Corrosion Control
Performance
b. Secondary Impacts
c. Treatment Issues
Notes/Qualifications
Reference(s)
• Flow-through testing program:
1 . Coupon results: no difference was observed on the basis of weight-loss between
hard and softened water, except in the case with galvanically-coupled coupons
where a modest increase in corrosion rate was noted for the softened condition.
Generally, corrosion rates did decrease over time.
2. Iron pipe loop results: At the end of the 7-month pre-conditioning period, the iron
loops produced similar corrosion rate results.
3. Lead pipe loop results: At the end of the 7-month pre-conditioning period,
significant variability in the performance of loops were observed as follows:
New STOId did not perform alike;
Old & Old did not perform alike;
While the two new loops behaved statistically comparable, the variability was
high with 90% confidence intervals ranging between 32 and 72 percent of the
mean values of lead.
• Full-scale evaluation - No significant difference in metal concentrations were found
before and after the installation of softening treatment.
NA
NA
• Row-through testing program:
1 . Replicate performance of pipe loops showed a large variability in metal leaching.
2. Results from the evaluation of hard and softened water conditions have yet to
be published.
• Full-scale evaluation - 23 homes were included in the tap monitoring program. Of
these, 12 had point of entry softeners which were to be bypassed during the
sampling day. However, this means that about 50 percent of the sites had already
been exposed to softened water prior to the utility installing its ion exchange
treatment unit
Wysock, B.M. et al. 1991. A Study of the Effect of Municipal Ion Exchange Softening on
the Corrosion of Lead, Copper, and Iron in Water Systems. Proc. Annual AWWA
Conference (Philadelphia, PA).
B-13
-------
PWS Characterization: Moderate pH, High Alkalinity & Calcium
PWS Name A Location: Fort Shawnee, Ohio
Item
i
it
in
Case Study Number
^ Description
7
PWS Description
a. Raw Water Supply
b. Water Quality
1. Raw
2. Treated
c. Treatment
Qroundwater
Moderate pH, high alkalinity and hardness, and elevated iron and hydrogen sulfide
levels.
Final water quality: pH 7.3-8.0; Total hardness 250-300 mg CaCO3/L; Alkalinity 290-
350 mg CaCO3/L; sulfate 206-330 mg SO4/L; chloride 1 6-45 mg CI/L; and carbon
dioxide 18-28 CO2 mg/L
Well water is aerated and filtered for iron and hydrogen sulfide removal; split treatment
for zeolite softening.
Corrosion Control Study Elements
a. Desk-top
Evaluation
b. Demonstration
Testing
1. Row-Through
2. Static
3. Full-Scale
c. Full-Scale
Confirmation
1 . Tap sampling
2. WQP-DIS
3. WQP-POE
• Copper pitting failures occurring in domestic cold water piping in relatively new
condominums.
• Investigators determined that excessive carbon dioxide and oxygen primarily
responsible for corrosive behavior of the treated water. This was thought to be
exacerbated by the higher sulfate and chloride content of the water supply.
• Raising the pH to approximately 8.3 would reduce the carbon dioxide content of
the finished water.
Pipe loop testing was performed.
NA
NA
NA
NA '
NA
B-14
-------
PWS Characterization:
PWS Name & Location:
Moderate pH, High Alkalinity & Calcium
Fort Shawnee, Ohio
. Hern
IV
V
VI
VII
Description
Testing Program Desription
a. Constraints
b. Priorities
1. Primary
2. Secondary
c. Treatment
Alternatives
d. Monitoring
programs
e. QA/QC Elements
• Treatment facility was not continuously staffed.
• Chemical storage was located in an unheated building, so that freezing was a
concern for the winter months.
Reduce the copper pitting failures in home-owners' plumbing systems.
Operational feasibility.
• Two pipe loops constructed of 50-by-1 meter lengths of type L copper tubing - 3/4
inch diameter, soldered with 50:50 leaditin solder.
• Pipe loop 1 received existing finished water; Pipe loop 2 received finished water
treated with soda ash to raise the pH to approximately 8.3.
• Loops were operated cyclically: running for 10-minutes; standing for 1 10-minutes
for sixteen hours; then standing for 8-hours.
• Water quality parameters were measured enterring the pipe loops, but no metal
leaching data was collected.
• After one-years's operation, the tubes were removed and physically inspected for
corrosion activity.
• Energy dispersive spectroscopy (EOS) and microchemical analysis were used to
confirm the corrosion byproducts present on the interior walls of the copper piping.
NA
Testing Results
a. Corrosion Control
Performance
b. Secondary
Impacts
c. Treatment Issues
Notes/Qualifications
Reference (s)
• EDS and microchemical analyses confirmed that pipe loops treated with soda ash
showed cuprous oxide and calcium carbonate coatings on the surface of the pipe
walls. No pitting corrosion was evident in any of the pipes extracted from the soda
ash-treated loop.
• Pitting corrosion was evident in pipes exposed to water not treated with soda ash
based on visual inspection. EDS and microchemical analysis confirmed the
following:
1) Green tubercles overlaying pits consisted principally of copper carbonate
(malachite) mixed with copper sulfate;
2) Pits examined by scanning electron microscopy were found to contain cubic
crystalline byproducts. EDS findings indicated that these byproducts contained
major quantities of copper with semiminor quantities of sulfate, and trace
amounts of chloride.
• Tap sampling was not performed, but following installation of soda ash treatment,
the pitting failures of copper, dometic piping ceased within 6-months.
NA
Average dosage of soda ash was 35 mg/L for a pH goal of 8.1-8.3.
Cohen, A. and J.R. Meyers. 1987. Mitigating Copper Pitting through Water Treatment.
Journal AWWA 79(2):58-61.
B-15
-------
PWS Characterization:
PWS Name & Location:
Moderate pH, High Alkalinity & Calcium
Pinellas County, Florida
item
i
it
in
IV
Case Study Number
Description
8
PWS Description
a. Raw Water Supply
b. Water Quality
1. Raw
2. Treated
c. Treatment
Deep well water
High hardness and alkalinity with elevated hydrogen sulfide
Finished water quality: pH 7.6-7.85; Total hardness 200-214 mg CaCO3/L;
Alkalinity 200-21 1 mg CaCO3/L; chlorine residual 2.5 mg/L
Reduced draft aeration of well water for hydrogen sulfide removal; pM
adjustment using caustic soda, and chlorination.
Corrosion Control Study Elements
a. Desk-top
Evaluation
b. Demonstration
Testing
1. Row-Through
2. Static
3. Full-Scale
c. Full-Scale
Confirmation
1 . Tap sampling
2. WQP-DIS
3. WQP-POE
NA
Screening of options for full-scale evaluation and pipe rig operation using
various water quality conditions to determine the cause and effect of corrosion
problems as a function of dissolved oxygen, pH, and corrosion inhibitor.
NA
In-situ testing using pipe rig systems after full-scale treatment installation.
First-draw tap samples for copper.
pH, dissolved oxygen, carbon dioxide, inhibitor residual
pH, dissolved oxygen, inhibitor dose
Testing Program Desrlption
a. Constraints
b. Priorities
1. Primary
2. Secondary
c. Treatment
Alternatives
NA
Minimize copper corrosion.
Not adversely affect lead corrosion.
• Screening testing evaluated four corrosion inhibitors, combinations of
ortho- and polyphosphates for four months.
• Full-scale evaluation:
1. Coupons and flow-through test racks were installed at 1 1 locations in
the distribution system.
2. The test racks were operated at a flow of 2 gpm for a month at a time.
3. The recommended orthophosphate inhibitor treatment was installed
full-scale.
4. Corrosion rates were evaluated for 6-months prior to installing
phosphate treatment and 6-months after treatment was in-place.
• Demonstration flow-through testing:
1 . Four pipe rigs were constructed with lead, copper, and mild steel flat
coupons (4 each), and galvanized steel and copper pipe inserts (2
each).
2. Row-through testing was conducted for 6-months under varying
combinations of dissolved oxygen, pH, and inhibitor dosages.
B-16
-------
PWS Characterization:
PWS Name & Location:
Moderate pH, High Alkalinity & Calcium
Pinellas County, Florida
item
V
VI
VII
d. Monitoring
programs
e. QA/QC Elements
Description
• Screening tests utilized Virchem 2902 corrosion test units which measure
corrosion rates through electrochemical measurements.
• Full-scale evaluation: corrosion rates were measured via weight loss
measurements for the coupons and pipe inserts. First-draw tap samples
were also collected from 25 sites in the affected area of the distribution
system.
• Demonstration flow-through testing: corrosion rates were measured using
weight loss.
Water quality parameters were measured throughout all phases of the testing
program.
NA
Testing Results
a. Corrosion Control
Performance
b. Secondary Impacts
c. Treatment Issues
Notes/Qualifications
Reference(s)
Screening tests indicated that orthophosphate was most effective in controlling
copper and lead corrosion.
• Full-scale evaluation:
1 . Reductions in copper corrosion rate of 30% after adding 1 mg/L
orthophosphate.
2. Lead corrosion reduction was nominal, approximately 10%.
3. Tap sampling results showed copper and lead reductions in first-draw
samples of 47% and 40%, respectively.
4. Wastewater influent copper levels were noted to be reduced by 57%.
• Demonstration flow-through testing
1 . Optimum orthophosphate dose was 1 .0 mg/L as PO4.
2. Lead corrosion rates increased slightly when the pH was increased
from 7.5 to 8.1.
3. A pH value of 7.7 was found to be optimum with dissolved oxygen
concentrations of 0.4-6.0 mg/L
4. Dissolved oxygen concentrations between 0.4 and 6 mg/L has little
effect on corrosion rates in the presence of 1 mg/L orthophosphate.
NA
NA
Powell, P.M. et al. 1991. Corrosion in Water Distribution Systems. AWWA/EPA
Corrosion Control Seminar (Chicago, IL).
B-17
-------
PWS Characterization: High pH, Alkalinity & Calcium
PWS Name & Location: Water District No. 1, Johnson County, Kansas
item
i
II
in
Case Study Number
Description
9
PWS Description
a. Raw Water Supply
b. Water Quality
1. Raw
2. Treated
c. Treatment
Kansas River and Missouri River
High turbidity, hardness and alkalinity; Moderate pH.
Rnished water quality: pM 9.1, Total hardness 122 - 130 mg CaCO3/L, Alkalinity
mg CaCO3/L, TDS 300 mg/L, chloride 35-70 mg/L, sulfate 135-200 mg/L.
53
Lime softening, chloramination, and polyphosphate addition.
Corrosion Control Study Elements
a. Desk-top
Evaluation
b. Demonstration
Testing
1. Row-Through
2. Static
3. Full-Scale
c. Full-Scale
Confirmation
1 . Tap sampling
2. WQP-DIS
3. WQP-POE
• Reviewed historical water quality data: raw, treated, and within the distribution
system.
• Removed several unlined fittings from the distribution system for visual inspection.
• Reviewed technical literature about corrosion behavior for high pH waters.
• Surveyed other lime softening utilities about corrosion problems encountered in
their distribution systems.
• Recommendation was to try to achieve calcium carbonate deposits for corrosion
protection.
NA
NA
Coupon inserts in distribution system during three phases of treatment modification.
NA
NA
pH, alkalinity, calcium, temperature
B-18
-------
PWS Characterization: High pH, Alkalinity & Calcium
PWS Name & Location: Water District No. 1, Johnson County, Kansas
item
IV
V
VI
VII
Description
Testing Program Desrlption
a. Constraints
b. Priorities
1. Primary
2. Secondary
c. Treatment
Alternatives.
d. Monitoring
programs
e. QA/QC Elements
Compatibility of treatment approach for lime softening facility. Polyphosphate addition
was practiced for calcium sequestering prior to filtration to prevent excessive
deposition on filter media.
Based on customer complaints, reduce iron corrosion.
Determine if optimal corrosion for iron control would also benefit the District for lead
and copper corrosion control.
Three phases of softening modification:
1 . Phase 1 - Increase pH to 9.3;
2. Phase 2 - Increase pH to 9.5 and increase final alkalinity and calcium content
to 140-160 mg CaCO/L and 90-1 10 mg CaCO3/L, respectively.
3. Phase 3 - Increase pH to 9.8-10.0
Removal and replacement of mild steel coupons placed in-situ in the distribution
system every 3-months. Rnished water quality monitored to achieve goals of each
treatment phase.
NA
Testing Results
a. Corrosion Control
Performance
b. Secondary Impacts
c. Treatment Issues
Notes/Qualifications
Reference(s)
• Phase 2 treatment corrosion rates were 35% lower than Phase 1 corrosion rates
based on the mild steel coupon results.
• A 40% decrease in customer complaints was observed overall (across Phases 1
and 2); however, complaints in areas of the service area composed of older,
unlined case iron mains showed little improvement
NA
NA
• The corrosion behavior of the distribution system is still be evaluated through the
Phase 3 treatment modifications.
• Elimination of the polyphosphate feed is being considered in order to improve the
deposition of calcium carbonate.
Goold, R.R. et al. 1991. Enhancing Distribution System Water Quality at Water District
No. 1. Proc. Annual AWWA Conference (Philadelphia, PA).
B-19
-------
Appendix C
Statistical Evaluation of
Corrosion Control
Performance Data
Non-parametric Statistics
The use of non-parametric statistical
measures may assist PWSs in interpret-
ing the findings of demonstration tests.
These methods of data analysis are
independent of the normality of data
.distributions, and provide measures of
the relationship between distinct data
populations. In applying the non-
parametric methods to lead and copper
testing results, the type of question
which they may answer is: Is the popu-
lation of lead levels from experimental
condition 'A' higher, lower, or the same
as those from experimental conditions
The Wilcoxon Test (also known as the
Mann-Whitney Test, the U-Test, or the
Rank Sums Test) is a non-parametric
alternative to the two sample Student's
t-Test. Using the lead concentration data
presented in the flow-through example
in Section 4.9.1 (Table 4-8), the Wilcoxon
Test may be used to select the treatment
method which minimized lead levels in
simulated first-flush samples. The prob-
lem may be stated as follows:
• Are the lead levels from Pipe Loop 1
larger than those from Pipe Loop 2?
• Are the lead levels from Pipe Loop 1
larger than those from Pipe Loop 3?
• Are the lead levels from Pipe Loop 2
larger than those from Pipe Loop 3?
By applying the Wilcoxon Test to
answer the three questions above we
will be able to determine whether lime
addition (Loop 2) or phosphate inhibitors
(Loop 3) provide improvements in lead
corrosion control, and whether either
method is superior to the other.
The first step is to rank the two
populations of data under evaluation as
one set of data, from the smallest to the
largest value. Table C-l presents the
results of ranking the lead concentration
data for our three comparison
conditions. Note that when a value
occurs multiple times in the data base,
the mean rank is assigned to each
occurrence.
The next step is to sum the ranking
for the data each of the populations. For
example, Loop 1 and Loop 2 rankings
were summed under the first comparison
condition as shown in Table C-l, result-
ing in 1,504 and 979, respectively. The
U-value may be calculated based on the
sum of the ranks, Wj, and the number of
observations, r^, as follows:
Us = W, - ni*(n,+ l)/2
with the statistic U being the smaller of
D! and U2 for any comparison condition.
The mean and variance for any popula-
tion of U values may be calculated as:
Mean U = ni
Var U = n *
* n2 / 2
n * (n
+ 1) / 12
The U statistic approximates a normal
distribution when both nt and n% are
greater than 8.
To test the null hypothesis that the
two data groups come from the same
population, the z-statistic is calculated
as a function of U, Mean U, and
C-1
-------
Table C-l. Wilcoxon Test for Comparing Flow-Through
Testing Results
Comparison 1 — Loop 1:2
Pb,ppb
45
48
48
52
52
52
53
55
57
58
60
60
62
63
68
68
68
72
72
72
72
75
75
76
78
78
78
79
79
80
80
80
82
85
85
87
87
90
91
92
92
92
Rank
1
2
3
4
5
6
7
8
9
10
11.5.
11.5
13
14
16
16
16
19.5
19.5
19.5
19.5
22.5
22.5
24
26
26
26
27.5
27.5
31
31
31
33
34.5
34.5
36.5
36.5
38
39
41
41
41
Loop
2
2
2
2
2
2
2
2
2
2
1
2
1
1
1
1
2
1
2
2
2
1
2
2
1
1
1
2
2
1
2
2
1
1
2
1
2
2
1
1
1
2
Comparison 2 — Loop 1:3
Pb,ppb
30
38
42
45
50
51
52
54
55
58
60
62
62
62
63
65
65
68
68
68
68
68
68
70
72
72
72
73
75
75
75
76
78
78
78
78
80
80
81
82
82
85
Rank
1
2
3
4
5
6
7
8
9
10
11
13
13
13
15
16.5
16.5
20.5
20.5
20.5
20.5
20.5
20.5
24
26
26
26
28
30
31
31
32
34.5
34.5
34.5
34.5
37.5
37.5
39
40.5
40.5
42
Loop
3
3
3
3
3
3
3
3
3
3
1
1
3
3
1
3
3
1
1
3
3
3
3
3
1
3
3
3
3
1
3
3
1
1
1
3
1
3
3
1
3
1
Comparison 3 — Loop 2:3
Pb.ppb
30
38
42
45
45
48
48
50
51
52
52
52
52
53
54
55
55
57
58
58
60
62
62
65
65
68
68
68
68"
68
70
72
72
72
72
72
73
75
75
75
76
76
Rank
1
2
3
4.5
4.5
6.5
6.5
8
9
11.5
11.5
11.5
11.5
14
15
16.5
16.5
18
19
20.5
20.5
22.5
22.5
24.5
24.5
28
28
28
28
28
31
34
34
34
34
34
37
39
39
39
41.5
41.5
Loop
3
3
3
2
3
2
2
3
3
2
2
2
3
2
3
2
3
2
3
2
2
3
3
3
3
2
3
3
3
3
3
2
2
2
3
3
3
2
3
3
2
3
C-2
-------
Table C-l. Wilcoxon Test for Comparing Flow-Through
Testing Results (continued)
Comparison 1 — Loop 1:2
P»,ppb
95
95
96
97
97
100
100
103
103
107
108
108
110
110
112
112
115
118
125
126
130
132
135
138
162
175
190
205
Rank
43.5
43.5
45
46.5
46.5
48.5
48.5
50.5
50.5
52
53.5
53.5
55.5
55.5
57.5
57.5
59
60
61
62
63
64
65
66
67
68
69
70
Loop
1
2
2
1
2
1
2
1
2
1
2
2
1
2
1
1
1
1
1
1
2
1
2
1
1
1
1
1
Comparison 2 — Loop 1:3
Pb.ppb
87
91
92
92
95
97
98
100
102
102
103
107
109
110
112
112
115
115
118
125
126
126
132
138
162
175
190
205
Rank
43
44
45.5
45.5
47
48
49
50
51.5
51.5
53
54
55
56
57.5
57.5
59.5
59.5
61
62
63.5
63.5
65
66
67
68
69
70
Loop
1
1
1
1
1
1
3
1
3
3
1
1
3
1
1
1
1
3
1
1
1
3
1
1
1
1
1
1
Comparison 3 — Loop 2:3
Pb,ppb
78
79
79
80
80
80
81
82
85
87
90
92
95
96
97
98
100
102
102
103
108
108
109
110
115
126
130
135
Rank
43
44.5
44.5
47
47
47
49
50
51
52
53
54
55
56
57
58
59
60.5
60.5
62
63.5
63.5
65
66
67
68
69
70
Loop
3
2
2
2
2
3
3
3
2
2
2
2
2
2
2
3
2
3
3
2
2
2
3
2
3
3
2
2
C-3
-------
StdDev U (STDDEV U = square root of
Var U) as follows:
z = (U - Mean Tj) / StdDev U
Table C-2 presents sum of ranks,
U values, and z values for the three
comparison conditions for the lead
concentration data from the flow
through testing results using the above
calculations.
To test the comparison conditions,
the z values are evaluated with respect
to za values for the alpha (i.e.,a) level of
significance desired. In the lead testing
example, Z001 = 2.575 was used to evalu-
ate the three comparisons. When z < -za,
then the distribution of the data with
the larger U value is said to be
stochastically higher than the other
population's distribution. For the first
comparison condition, for example, the
larger U value corresponded to Loop 1
data and the z value was less than -za,
then the lead levels found in the control
loop (Loop 1) are higher than those
found from the lime addition loop (Loop
2). Conversely, when z > za, then the
distribution of the data with the larger
U value is said to be stochastically lower
than the other population's distribution.
The results shown in Table C-2 indi-
cate that Loop 2 and Loop 3 lead levels
were significantly different from Loop 1
lead levels, but not significantly differ-
ent from each other. Additionally, the z
values show that Loop 1 (control) lead
levels were higher than both Loop 2 and
Loop 3 lead concentrations.
Parametric Statistics
Water quality measurements
obtained during corrosion control studies
will seldom represent the one true value
present at the time of sampling. Errors
will be associated with both sampling
techniques and analytical measure-
ments. It is generally assumed that the
errors indigenous to these measured
values are random errors. Therefore, the
mean of several values should be a
better indicator of the true value than a
single measurement.
The configuration in which the data
are arranged is called its distribution,
and many statistical procedures utilize a
normal distribution in which the data
are symmetrical and form a bell shaped
curve. Parametric statistics make use of
these procedures.
Most sample sets of water quality
data do not exactly form a bell shaped
curve, and they are sometimes "trans-
formed" by the application of some
mathematical function into another form
which more closely follows a normal
distribution. As an example of this
procedure, the lead data used for the
example of Section 4.9.1 (See Table 4-8)
will be transformed into the log normal
mode by using the log of the individual
determinations.
Parametric statistics were used to
compare the two treatments with the
control. The data were investigated for
skewness recognizing that as the mo-
ment coefficient of skewness approaches
zero that the data approach a more
normal distribution. If the distribution is
normal, or can be made more normal by
a transformation, the statistical
techniques based on a normal distri-
bution are appropriate; otherwise they
are only approximations and the use of
non-parametric statistics as presented
previously in Appendix C may be more
appropriate.
C-4
-------
Table C-2. Summary of WilcoxoirTest Measures for Comparing
the Performance of Lead Corrosion Control Alternatives
Condition
Sum of Ranks, W
Comparison 1
Comparison 2
Comparison 3
U — Values
Comparison 1
Comparison 2
Comparison 3
Key Statistical Measures
Comparison 1
Comparison 2
Comparison 3
Measurements
Loop 1
1,504.0
1,612.5
Loop 1
874.0
982.5
MeanJJ
612.5
612.5
612.5
Loop 2
979.0
1,366.0
Loop 2
349.0
736.0
Var_U
7,248.0
7,248.0
7,248.0
Loop 3
874.5
1,199.0
Loop 3
244.5
569.0
z
-3.1
-4.3
-0.5
Resultant
349.0
244.5
569.0
Finding
Loop 1 > Loop 2
Loop 1 > Loop 3
No Difference
The skewness coefficient, Y, is defined
as:
where:
Table C-3 gives the calculated means,
moments, and skewness coefficients for
the lead data of Table 4-8 for both nor-
mal and log normal distributions. The
smaller coefficients for the log normal
distribution were used as indicators that
the data would more appropriately adapt
to parametric statistics using a logarith-
mic transformation.
fl73 = -
n
= individual samples, i = 1 to n
C-5
-------
Table C-3. Skewness Coefficients
Normal:
mean
m2
m3
Y
Loopl
0.1038
0.0012
0.0001
1.21
Loop 2
0.0791
0.0005
5.83 x 10*
0.47
LoopS
0.0711
0.0005
5.76 X10"6
0.60
Log Normal:
mean
m2
m3
Y
-1 .0058
0.0182
0.0013
0.53
-1.1204
0.0163
-0.0001
-0.04
-1.1683
0.0178
-0.0007
-0.32
The student's t statistic was used to
compare paired data among the three
loops. These results are shown in both
Table C-4 and Table 4-9. Student's t can
be defined as:
where the numerator represents the
mean difference between paired sample
data and the denominator represents the
standard deviation appropriate to the
difference between the sample means.
These values are then compared to stan-
dard statistical tables to determine if
there is a statistical difference in treat-
ments.
Table C-4. Student's t Values
Comparison
Loop 1 and Loop 2
Loop 1 and Loop 3
Loop 2 and Loop 3
Notes:
t
5.46***
6.98***
2.87**
All test data transformed to logarithmic values
** Highly significant difference at the 0.01 level
*** Extremely significant difference at the 0.001 level
C-6
-------
As indicated in the text of Section
4.9.1, the last 10 weeks of data were
independently examined. These data are
shown in Table C-5. Again, prior to
conducting an examination of the data
using the Student's t statistic, a log
transformation was made, i.e.:
0.078 was used as log 0.078 = -1.1079
0.060 was used as log 0.060 = -1.2218;
etc.
Using Student's t and examining the
paired data between loops for week
26 through week 35 gave the results
shown in Table C-6. Standard statistical
tables were used to compare the t value
against with the sign ignored, i.e., either
a positive or negative value was accept-
able. For 9 degrees of freedom (10 sets of
data minus 1), the t statistics are:
t @ 0.05 level = 2.262
Significant difference
t @ 0.01 level = 3.250
Highly significant difference
t @ 0.001 level = 4.781
Extremely significant difference
Table C-5. Lead Data from Final 10 Weeks of Testing
Week No.
26
27
28
29
30
31
32
33
34
35
Loopt
0.078
0.060
0.092
0.075
0.087
0.063
0.072
0.068
0.080
0.091
Loop 2
0.080
0.052
0.058
0.045
0.053
0.060
0.055
0.052
0.048
0.057
Loops
0.080
0.062
0.054
0.058
0.045
0.052
0.068
0.030
0.051
0.042
Table C-6. Calculate Student's t Values for Final 10 Weeks
Comparison
Loop 1 and Loop 2
Loop 1 and Loop 3
Loop 2 and Loop 3
Notes:
t
4.88**
3.60**
0.67, not
significant
All test data transformed to logarithmic values
** t value > 3.25, highly significant difference
Thus the analysis shows that each treatment is significantly different from the control, but
there is no apparent statistical difference between treatments.
C-7
------- |