1

Lead Corrosion Control Chemistry

Simoni TriantafyllidouUS EPArsquos Office of Research and Development Cincinnati

EPA ORD-Region 4 2020 Small Drinking Water Systems Meeting10142020

2

Who we are -Acknowledgments

Jennifer Tully

SimoniTriantafyllidou

Mike DeSantis

Christy Muhlen

DarrenLytle

Steve Harmon

Casey Formal

Mike Schock

Dan Williams

Regan Murray

Jonathan Burkhardt

Evelyne Doreacute

Mariah Caballero

33

What we do

ORD provides the data tools and information that form the sound scientific foundation the Agency relies on to fulfill its mission to protect the environment and safeguard public healthORD at a Glance httpswwwepagovaboutepaabout-office-research-and-development-ord Center for Environmental

Solutions amp Emergency Response (CESER) in CincinnatiWe conduct applied stakeholder-driven research and provide responsive technical support to help solve the Nationrsquos environmental challenges

CESER at a Glance httpswwwepagovaboutepaabout-center-environmental-solutions-and-emergency-response-ceser

2

Who we are -Acknowledgments

Jennifer Tully

SimoniTriantafyllidou

Mike DeSantis

Christy Muhlen

DarrenLytle

Steve Harmon

Casey Formal

Mike Schock

Dan Williams

Regan Murray

Jonathan Burkhardt

Evelyne Doreacute

Mariah Caballero

33

What we do

ORD provides the data tools and information that form the sound scientific foundation the Agency relies on to fulfill its mission to protect the environment and safeguard public healthORD at a Glance httpswwwepagovaboutepaabout-office-research-and-development-ord Center for Environmental

Solutions amp Emergency Response (CESER) in CincinnatiWe conduct applied stakeholder-driven research and provide responsive technical support to help solve the Nationrsquos environmental challenges

CESER at a Glance httpswwwepagovaboutepaabout-center-environmental-solutions-and-emergency-response-ceser

33

What we do

ORD provides the data tools and information that form the sound scientific foundation the Agency relies on to fulfill its mission to protect the environment and safeguard public healthORD at a Glance httpswwwepagovaboutepaabout-office-research-and-development-ord Center for Environmental

Solutions amp Emergency Response (CESER) in CincinnatiWe conduct applied stakeholder-driven research and provide responsive technical support to help solve the Nationrsquos environmental challenges

CESER at a Glance httpswwwepagovaboutepaabout-center-environmental-solutions-and-emergency-response-ceser

Pb sources

BUILDING

bull Lead Service Lines (LSLs)bull Lead Goosenecksbull Leaded Solder

bull Leaded Brass (valves fittings faucets fountains)bull Galvanized Iron Pipe downstream of leaded plumbing

Triantafyllidou amp Edwards 2012

Pb sources

Full Lead Service Line

Clark et al 2013

Brass unionvalve vs plastic

Partial Lead Service Line

Lead gooseneckTriantafyllidou et al 2020

What is in the distribution system Materials inventory is critical to understand where and what lead sources still exist

Pb sources

Leaded Solder

2 Pb

lt05Pb

3 Pb

Leaded Brass FaucetSelover 2005

Gal

vani

zed

Diverse legacy leaded materials may undergo different corrosion reactions and impact water quality differently

Corrosion is oxidation-reduction

Pb harr Pb2+ + 2e-

Pb harr Pb4+ + 4e-

OxidationLead metal losing electronsat anode

OCl- + H+ + 2e- harr Cl- + OH-ReductionOxidant gainingelectronsat cathode

Oxidant is dissolved oxygen free chlorine chloramine chlorine dioxide etc

2OCl- + 2H+ + 4e-harr 2Cl- + 2OH-

There are millions of anodecathode sites across interior fresh lead pipe surface

Pb sources

Full Lead Service Line

Clark et al 2013

Brass unionvalve vs plastic

Partial Lead Service Line

Lead gooseneckTriantafyllidou et al 2020

What is in the distribution system Materials inventory is critical to understand where and what lead sources still exist

Pb sources

Leaded Solder

2 Pb

lt05Pb

3 Pb

Leaded Brass FaucetSelover 2005

Gal

vani

zed

Diverse legacy leaded materials may undergo different corrosion reactions and impact water quality differently

Corrosion is oxidation-reduction

Pb harr Pb2+ + 2e-

Pb harr Pb4+ + 4e-

OxidationLead metal losing electronsat anode

OCl- + H+ + 2e- harr Cl- + OH-ReductionOxidant gainingelectronsat cathode

Oxidant is dissolved oxygen free chlorine chloramine chlorine dioxide etc

2OCl- + 2H+ + 4e-harr 2Cl- + 2OH-

There are millions of anodecathode sites across interior fresh lead pipe surface

Pb sources

Leaded Solder

2 Pb

lt05Pb

3 Pb

Leaded Brass FaucetSelover 2005

Gal

vani

zed

Diverse legacy leaded materials may undergo different corrosion reactions and impact water quality differently

Corrosion is oxidation-reduction

Pb harr Pb2+ + 2e-

Pb harr Pb4+ + 4e-

OxidationLead metal losing electronsat anode

OCl- + H+ + 2e- harr Cl- + OH-ReductionOxidant gainingelectronsat cathode

Oxidant is dissolved oxygen free chlorine chloramine chlorine dioxide etc

2OCl- + 2H+ + 4e-harr 2Cl- + 2OH-

There are millions of anodecathode sites across interior fresh lead pipe surface

Corrosion is oxidation-reduction

Pb harr Pb2+ + 2e-

Pb harr Pb4+ + 4e-

OxidationLead metal losing electronsat anode

OCl- + H+ + 2e- harr Cl- + OH-ReductionOxidant gainingelectronsat cathode

Oxidant is dissolved oxygen free chlorine chloramine chlorine dioxide etc

2OCl- + 2H+ + 4e-harr 2Cl- + 2OH-

There are millions of anodecathode sites across interior fresh lead pipe surface

Corrosion and scale formation

8

Corrosion

Pb(s)

OCl- OH-

2 e-

Pb+2 (or Pb+4)

2 e-

Scale Formation

water

Pb pipe

Pb pipe

water

Pb amp other solids

Pb2+

Pb(+

2) c

arbo

nate

sPb

(+2)

hyd

roxy

carb

onat

esPb

(+2)

orth

opho

spha

te

harr Pb4+

Pb(+

4) o

xide

harr

Idealized scenario of scale solids Scale is way more complex (heterogeneous several layers amorphous compounds) and it controls lead release

Tria

ntaf

yllid

ou e

t al

202

0

Corrosion and metal release

bull Pure corrosion (ie electrochemical oxidation-reduction reactions) becomes less important when the surface becomes covered with scale

bull Metal solubility or destabilization of the scale at the water contact will be more important (if the scale is porous corrosion may still occur at the base of the scale)

bull Metal solubility varies by factor of 5 to 10 or more across drinking water systems

CCT includes both pure corrosion and control of metal release from the pipe scale

Corrosion typesbull Uniform corrosion

- Materials degradation- Metal release (lead copper etc)

bull Non-uniform corrosion- Pinhole leaks (copper)- Dezincification (brass)- Tuberculation (iron galvanized brass)- Galvanic Corrosion

bull Galvanic corrosion- Soldered joints- Brass devices- Partial LSLs- Any coupling of different metals

Different leaded materials may undergo different corrosion mechanisms depending on composition configuration geometry

water quality and other factors

Pb pipe uniform

Pb pipe galvanic

DeSantis et al 2018

Corrosion and metal release

bull Pure corrosion (ie electrochemical oxidation-reduction reactions) becomes less important when the surface becomes covered with scale

bull Metal solubility or destabilization of the scale at the water contact will be more important (if the scale is porous corrosion may still occur at the base of the scale)

bull Metal solubility varies by factor of 5 to 10 or more across drinking water systems

CCT includes both pure corrosion and control of metal release from the pipe scale

Corrosion typesbull Uniform corrosion

- Materials degradation- Metal release (lead copper etc)

bull Non-uniform corrosion- Pinhole leaks (copper)- Dezincification (brass)- Tuberculation (iron galvanized brass)- Galvanic Corrosion

bull Galvanic corrosion- Soldered joints- Brass devices- Partial LSLs- Any coupling of different metals

Different leaded materials may undergo different corrosion mechanisms depending on composition configuration geometry

water quality and other factors

Pb pipe uniform

Pb pipe galvanic

DeSantis et al 2018

Corrosion typesbull Uniform corrosion

- Materials degradation- Metal release (lead copper etc)

bull Non-uniform corrosion- Pinhole leaks (copper)- Dezincification (brass)- Tuberculation (iron galvanized brass)- Galvanic Corrosion

bull Galvanic corrosion- Soldered joints- Brass devices- Partial LSLs- Any coupling of different metals

Different leaded materials may undergo different corrosion mechanisms depending on composition configuration geometry

water quality and other factors

Pb pipe uniform

Pb pipe galvanic

DeSantis et al 2018

Important factors affecting corrosion and metal release

bull pH and AlkalinityDissolved Inorganic Carbon

bull Type and Concentration of Oxidants (chlorine species dissolved oxygen etc)OxidationReduction Potential

bull Corrosion Inhibitors

bull Chloride to Sulfate Mass Ratio

bull Manganese

pH is master variable

12

-1

0

1

2

0 2 4 6 8 10 12 14

E (V

OLT

S)

pH

IMMUNE

Simplified Pourbaix diagram (EH-pH Diagram)

Desirable pH-EH combinations allow passivation (ie formation ofprotective pipe scales)

pH is master variable

12

PbO2 (plattnerite)

Pb ++

deg s)(3O 2 -2C H) 2)b O 3P O

00

( 2 -- -2) C 43 ( )

CO

b HP O( ( DIC = 18 mg CL

3 b

Pb P Pb = 0010 mgL

Pb metal

-10

13

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14

ndash8

ndash6

ndash4

ndash2

2

4

6

8

10

pH

Eh (v

olts

)

Dissolved Inorganic Carbon (DIC) and Total ALKalinity (TALK)

DIC = [CO32-] + [H2CO3

] + [HCO3- ]]

TALK = 2 [CO32-] + [HCO3

-] + [OH-] - [H+]

mg CL Dissolved Inorganic Carbon (DIC)0 10 20 30 40 50

mg

CaCO

3L

Tota

l Alka

linity

0255075

100125150175

200225250

pH 60pH 70pH 80pH 90pH 100

bull To understand corrosion it isimportant to keep up with thecarbonate system

bull DIC and TALK have linearrelationship but are not thesame thing

bullbullbullbullbull

Oxidants in drinking water

bull Disinfection (Free chlorine Chloramine Chlorine dioxide)Pre-(O3 H2O2 ClO2 KMnO4) Oxidative metal removal (eg As Fe Mn) Ammonia removalAeration (corrosion control VOCRn2S removal) Taste and odor control

bull Dissolved oxygen

Oxidants in drinking water

Oxidant Dosage (mgL)0 2 4 6 8 10

E H (V

olts

vs S

HE)

03

04

05

06

07

08

09

10

11

MCA Electrode 1 MCA Electrode 2 KMnO4 Electrode 1 KMnO4 Electrode 2 ClO2 Electrode 1 ClO2 Electrode 2 DO Electrode 1 DO Electrode 2 Cl2 Electrode 1Cl2 Electrode 2

ClO2 HOCldeg

KMnO4NH2Cl

DO

Different oxidants have different oxidizing power

Oxidation-Reduction Potential of several disinfectants in experimentJames et al 2004 (pH 7 10 mg CL 25degC)

pH is master variable

12

-1

0

1

2

0 2 4 6 8 10 12 14

E (V

OLT

S)

pH

IMMUNE

Simplified Pourbaix diagram (EH-pH Diagram)

Desirable pH-EH combinations allow passivation (ie formation ofprotective pipe scales)

pH is master variable

12

PbO2 (plattnerite)

Pb ++

deg s)(3O 2 -2C H) 2)b O 3P O

00

( 2 -- -2) C 43 ( )

CO

b HP O( ( DIC = 18 mg CL

3 b

Pb P Pb = 0010 mgL

Pb metal

-10

13

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14

ndash8

ndash6

ndash4

ndash2

2

4

6

8

10

pH

Eh (v

olts

)

Dissolved Inorganic Carbon (DIC) and Total ALKalinity (TALK)

DIC = [CO32-] + [H2CO3

] + [HCO3- ]]

TALK = 2 [CO32-] + [HCO3

-] + [OH-] - [H+]

mg CL Dissolved Inorganic Carbon (DIC)0 10 20 30 40 50

mg

CaCO

3L

Tota

l Alka

linity

0255075

100125150175

200225250

pH 60pH 70pH 80pH 90pH 100

bull To understand corrosion it isimportant to keep up with thecarbonate system

bull DIC and TALK have linearrelationship but are not thesame thing

bullbullbullbullbull

Oxidants in drinking water

bull Disinfection (Free chlorine Chloramine Chlorine dioxide)Pre-(O3 H2O2 ClO2 KMnO4) Oxidative metal removal (eg As Fe Mn) Ammonia removalAeration (corrosion control VOCRn2S removal) Taste and odor control

bull Dissolved oxygen

Oxidants in drinking water

Oxidant Dosage (mgL)0 2 4 6 8 10

E H (V

olts

vs S

HE)

03

04

05

06

07

08

09

10

11

MCA Electrode 1 MCA Electrode 2 KMnO4 Electrode 1 KMnO4 Electrode 2 ClO2 Electrode 1 ClO2 Electrode 2 DO Electrode 1 DO Electrode 2 Cl2 Electrode 1Cl2 Electrode 2

ClO2 HOCldeg

KMnO4NH2Cl

DO

Different oxidants have different oxidizing power

Oxidation-Reduction Potential of several disinfectants in experimentJames et al 2004 (pH 7 10 mg CL 25degC)

pH is master variable

12

PbO2 (plattnerite)

Pb ++

deg s)(3O 2 -2C H) 2)b O 3P O

00

( 2 -- -2) C 43 ( )

CO

b HP O( ( DIC = 18 mg CL

3 b

Pb P Pb = 0010 mgL

Pb metal

-10

13

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14

ndash8

ndash6

ndash4

ndash2

2

4

6

8

10

pH

Eh (v

olts

)

Dissolved Inorganic Carbon (DIC) and Total ALKalinity (TALK)

DIC = [CO32-] + [H2CO3

] + [HCO3- ]]

TALK = 2 [CO32-] + [HCO3

-] + [OH-] - [H+]

mg CL Dissolved Inorganic Carbon (DIC)0 10 20 30 40 50

mg

CaCO

3L

Tota

l Alka

linity

0255075

100125150175

200225250

pH 60pH 70pH 80pH 90pH 100

bull To understand corrosion it isimportant to keep up with thecarbonate system

bull DIC and TALK have linearrelationship but are not thesame thing

bullbullbullbullbull

Oxidants in drinking water

bull Disinfection (Free chlorine Chloramine Chlorine dioxide)Pre-(O3 H2O2 ClO2 KMnO4) Oxidative metal removal (eg As Fe Mn) Ammonia removalAeration (corrosion control VOCRn2S removal) Taste and odor control

bull Dissolved oxygen

Oxidants in drinking water

Oxidant Dosage (mgL)0 2 4 6 8 10

E H (V

olts

vs S

HE)

03

04

05

06

07

08

09

10

11

MCA Electrode 1 MCA Electrode 2 KMnO4 Electrode 1 KMnO4 Electrode 2 ClO2 Electrode 1 ClO2 Electrode 2 DO Electrode 1 DO Electrode 2 Cl2 Electrode 1Cl2 Electrode 2

ClO2 HOCldeg

KMnO4NH2Cl

DO

Different oxidants have different oxidizing power

Oxidation-Reduction Potential of several disinfectants in experimentJames et al 2004 (pH 7 10 mg CL 25degC)

Dissolved Inorganic Carbon (DIC) and Total ALKalinity (TALK)

DIC = [CO32-] + [H2CO3

] + [HCO3- ]]

TALK = 2 [CO32-] + [HCO3

-] + [OH-] - [H+]

mg CL Dissolved Inorganic Carbon (DIC)0 10 20 30 40 50

mg

CaCO

3L

Tota

l Alka

linity

0255075

100125150175

200225250

pH 60pH 70pH 80pH 90pH 100

bull To understand corrosion it isimportant to keep up with thecarbonate system

bull DIC and TALK have linearrelationship but are not thesame thing

bullbullbullbullbull

Oxidants in drinking water

bull Disinfection (Free chlorine Chloramine Chlorine dioxide)Pre-(O3 H2O2 ClO2 KMnO4) Oxidative metal removal (eg As Fe Mn) Ammonia removalAeration (corrosion control VOCRn2S removal) Taste and odor control

bull Dissolved oxygen

Oxidants in drinking water

Oxidant Dosage (mgL)0 2 4 6 8 10

E H (V

olts

vs S

HE)

03

04

05

06

07

08

09

10

11

MCA Electrode 1 MCA Electrode 2 KMnO4 Electrode 1 KMnO4 Electrode 2 ClO2 Electrode 1 ClO2 Electrode 2 DO Electrode 1 DO Electrode 2 Cl2 Electrode 1Cl2 Electrode 2

ClO2 HOCldeg

KMnO4NH2Cl

DO

Different oxidants have different oxidizing power

Oxidation-Reduction Potential of several disinfectants in experimentJames et al 2004 (pH 7 10 mg CL 25degC)

bullbullbullbullbull

Oxidants in drinking water

bull Disinfection (Free chlorine Chloramine Chlorine dioxide)Pre-(O3 H2O2 ClO2 KMnO4) Oxidative metal removal (eg As Fe Mn) Ammonia removalAeration (corrosion control VOCRn2S removal) Taste and odor control

bull Dissolved oxygen

Oxidants in drinking water

Oxidant Dosage (mgL)0 2 4 6 8 10

E H (V

olts

vs S

HE)

03

04

05

06

07

08

09

10

11

MCA Electrode 1 MCA Electrode 2 KMnO4 Electrode 1 KMnO4 Electrode 2 ClO2 Electrode 1 ClO2 Electrode 2 DO Electrode 1 DO Electrode 2 Cl2 Electrode 1Cl2 Electrode 2

ClO2 HOCldeg

KMnO4NH2Cl

DO

Different oxidants have different oxidizing power

Oxidation-Reduction Potential of several disinfectants in experimentJames et al 2004 (pH 7 10 mg CL 25degC)

Oxidants in drinking water

Oxidant Dosage (mgL)0 2 4 6 8 10

E H (V

olts

vs S

HE)

03

04

05

06

07

08

09

10

11

MCA Electrode 1 MCA Electrode 2 KMnO4 Electrode 1 KMnO4 Electrode 2 ClO2 Electrode 1 ClO2 Electrode 2 DO Electrode 1 DO Electrode 2 Cl2 Electrode 1Cl2 Electrode 2

ClO2 HOCldeg

KMnO4NH2Cl

DO

Different oxidants have different oxidizing power

Oxidation-Reduction Potential of several disinfectants in experimentJames et al 2004 (pH 7 10 mg CL 25degC)

17

ldquoClassicrdquo divalent Pb+2 solubility

pH6 7 8 9 10 11

mg

PbL

001

01

1

10

100 1 mg CL 5 mg CL 10 mg CL 20 mg CL 35 mg CL 50 mg CL 75 mg CL100 mg CL

DIC

Optimum pHDIC rangefor LSLs

High pH is needed to minimize Pb solubility

Lead ldquocorrosion Inhibitorsrdquo

bull Phosphoric acidbull Alkalai-metal orthophosphatebull Zinc orthophosphatebull Blended orthopolyphosphatesbull Polyphosphatesbull Sodium Silicate

Polyphosphates are not corrosion inhibitors but rather sequestering agentsEfficiency depends on sufficient doseconcentration for thebackground pH and other reactive water quality constituentswhat if the dose is too low to actually work 18

Illustrative reactions of P in pipebull OrthophosphatePrecipitation (consumption of orthophosphate)3Pb+2 + 2PO4

-3 --gt Pb3(PO4)2 (solid)decreased Pb solubility

bull PolyphosphatesReversion partially (formation of orthophosphate)P3O10

-5 ---gt P2O7-3 + PO4

-3

polymer --gt monomersComplexation (soluble complexes)3Pb+2 + P3O10

-5 --gt PbP3O10-2

higher Pb solubility

Slid

e cr

edit

Mar

c Ed

war

ds

Conceptual possible reactions Actual reactants and products may bedifferent Polyphosphates cannot be easily quantified

20

Ortho-P Treatment for Pb+2

Effectiveness Depends on Dose DIC pH and ldquoCleanlinessrdquo of Pipe Surface

mg PO4L00 10 20 30 40 50

mg

PbL

000

005

010

015

020

025

030

035

040

48 mg CL

48 mg CL

pH = 70pH = 75pH = 80pH = 85

bull pH less critical at low DICbull pH less critical at high PO4bull Point of diminishing returns higher with high DIC

bull Faster Pb reduction at high PO4

Typical UK Dosages 4-6 mgL

Most PWSs with LSLs do not have corrosion control treatment that minimizes Pb release and exposure

Effect of pH and ortho-P on Pb release

21

DIC = 10 mg CL 1 mg PO4L

60 65 70 75 80 85 90 95

gL

b m

P

0001

001

01

1

10

US Action Level

Dissolved Pb with ortho-PTotal Pb with ortho-PDissolved Pb with CO3Total Pb with CO3

At low DIC orthophosphate improves lead release regardless of pH

pHSchock M R DeSantis M K Metz D H Welch M M Hyland R N Nadagouda M N Revisiting the pH Effect on the Orthophosphate Control of Plumbosolvency Proc AWWA Annual Conference and Exposition Atlanta GA 2008

Ortho-P at pH 90 (DIC 6 mgL)

22

Pb(μgL)

bull Low Pb at 1 mgL relative to higher doses needs to be further verified across water qualities

bull Must do dose optimization study for your own water quality especially at high pH

bull Ortho-P may precipitate with Ca

Miller S A Investigation of Lead Solubility and Orthophosphate Addition in High pH Low DIC Water Master of Science Department of Biomedical Chemical and Environmental Engineering College of Engineering and Applied Science University of Cincinnati Cincinnati OH 2014

Ortho-Ppoint of diminishing returns

bull Orthophosphate addition to where large increments result in small reductions in lead release

bull Key to cost-effective lead release control and exposure reduction

bull Becomes the rdquomaintenancerdquo dosage unless the scale ldquoagesrdquo

bull Background constituents such as aluminum hardness ions iron manganese and others interfere with optimum reduction of lead release

bull Varies with the background water chemistry from system to system

23

Sodium silicate

bull No systematic studies to look at pH carbonate silicate background chemistry relationships

bull Unclear mechanism pH increase Buffer intensity increase Protective film (thin amp amorphous)

bull Can sequester ironmanganese

bull Canrsquot be evaluated with fresh surfaces

24pH5 6 7 8 9 10 11

β ty

nsi

nte

Iuf

fer

B

00000

00005

00010TIC=48 mg CL20 mgL SiO250 mgL PO4β H2OTotal β

Silicate may contribute to Buffer Intensity

Carbonate orthophosphate silicate I=001 25ordmC

Chemical changes may reduce Pb+4 to Pb+2

10

8

6

4

)ts 2ol

h (v

E

ndash2

ndash4

ndash6

ndash8

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14

pH

12

PbO2 (plattnerite)

Pb++ )s

deg

(2

CO

3

(OH

) -2 2) 3

Pb

-2

0 (CO

-- 2

0

) 43(C

O

Pb

3 (OH

)

DIC = 18 mg CL

Pb Pb Pb = 0010 mgL

Pb metal

-10

Drop in ORP from treatment change or DS

oxidant demand

Drop in pH at surface from treatment change

chemical reactions nitrification etc

C D

ngto

nhisa

W

Newark NJ

Disinfectant demand in DS must be controlled amp enough free chlorine consistently maintained throughout LSL area to keep protective Pb+4 corrosion scales in place

Lead ldquocorrosion Inhibitorsrdquo

bull Phosphoric acidbull Alkalai-metal orthophosphatebull Zinc orthophosphatebull Blended orthopolyphosphatesbull Polyphosphatesbull Sodium Silicate

Polyphosphates are not corrosion inhibitors but rather sequestering agentsEfficiency depends on sufficient doseconcentration for thebackground pH and other reactive water quality constituentswhat if the dose is too low to actually work 18

Illustrative reactions of P in pipebull OrthophosphatePrecipitation (consumption of orthophosphate)3Pb+2 + 2PO4

-3 --gt Pb3(PO4)2 (solid)decreased Pb solubility

bull PolyphosphatesReversion partially (formation of orthophosphate)P3O10

-5 ---gt P2O7-3 + PO4

-3

polymer --gt monomersComplexation (soluble complexes)3Pb+2 + P3O10

-5 --gt PbP3O10-2

higher Pb solubility

Slid

e cr

edit

Mar

c Ed

war

ds

Conceptual possible reactions Actual reactants and products may bedifferent Polyphosphates cannot be easily quantified

20

Ortho-P Treatment for Pb+2

Effectiveness Depends on Dose DIC pH and ldquoCleanlinessrdquo of Pipe Surface

mg PO4L00 10 20 30 40 50

mg

PbL

000

005

010

015

020

025

030

035

040

48 mg CL

48 mg CL

pH = 70pH = 75pH = 80pH = 85

bull pH less critical at low DICbull pH less critical at high PO4bull Point of diminishing returns higher with high DIC

bull Faster Pb reduction at high PO4

Typical UK Dosages 4-6 mgL

Most PWSs with LSLs do not have corrosion control treatment that minimizes Pb release and exposure

Effect of pH and ortho-P on Pb release

21

DIC = 10 mg CL 1 mg PO4L

60 65 70 75 80 85 90 95

gL

b m

P

0001

001

01

1

10

US Action Level

Dissolved Pb with ortho-PTotal Pb with ortho-PDissolved Pb with CO3Total Pb with CO3

At low DIC orthophosphate improves lead release regardless of pH

pHSchock M R DeSantis M K Metz D H Welch M M Hyland R N Nadagouda M N Revisiting the pH Effect on the Orthophosphate Control of Plumbosolvency Proc AWWA Annual Conference and Exposition Atlanta GA 2008

Ortho-P at pH 90 (DIC 6 mgL)

22

Pb(μgL)

bull Low Pb at 1 mgL relative to higher doses needs to be further verified across water qualities

bull Must do dose optimization study for your own water quality especially at high pH

bull Ortho-P may precipitate with Ca

Miller S A Investigation of Lead Solubility and Orthophosphate Addition in High pH Low DIC Water Master of Science Department of Biomedical Chemical and Environmental Engineering College of Engineering and Applied Science University of Cincinnati Cincinnati OH 2014

Ortho-Ppoint of diminishing returns

bull Orthophosphate addition to where large increments result in small reductions in lead release

bull Key to cost-effective lead release control and exposure reduction

bull Becomes the rdquomaintenancerdquo dosage unless the scale ldquoagesrdquo

bull Background constituents such as aluminum hardness ions iron manganese and others interfere with optimum reduction of lead release

bull Varies with the background water chemistry from system to system

23

Sodium silicate

bull No systematic studies to look at pH carbonate silicate background chemistry relationships

bull Unclear mechanism pH increase Buffer intensity increase Protective film (thin amp amorphous)

bull Can sequester ironmanganese

bull Canrsquot be evaluated with fresh surfaces

24pH5 6 7 8 9 10 11

β ty

nsi

nte

Iuf

fer

B

00000

00005

00010TIC=48 mg CL20 mgL SiO250 mgL PO4β H2OTotal β

Silicate may contribute to Buffer Intensity

Carbonate orthophosphate silicate I=001 25ordmC

Chemical changes may reduce Pb+4 to Pb+2

10

8

6

4

)ts 2ol

h (v

E

ndash2

ndash4

ndash6

ndash8

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14

pH

12

PbO2 (plattnerite)

Pb++ )s

deg

(2

CO

3

(OH

) -2 2) 3

Pb

-2

0 (CO

-- 2

0

) 43(C

O

Pb

3 (OH

)

DIC = 18 mg CL

Pb Pb Pb = 0010 mgL

Pb metal

-10

Drop in ORP from treatment change or DS

oxidant demand

Drop in pH at surface from treatment change

chemical reactions nitrification etc

C D

ngto

nhisa

W

Newark NJ

Disinfectant demand in DS must be controlled amp enough free chlorine consistently maintained throughout LSL area to keep protective Pb+4 corrosion scales in place

Illustrative reactions of P in pipebull OrthophosphatePrecipitation (consumption of orthophosphate)3Pb+2 + 2PO4

-3 --gt Pb3(PO4)2 (solid)decreased Pb solubility

bull PolyphosphatesReversion partially (formation of orthophosphate)P3O10

-5 ---gt P2O7-3 + PO4

-3

polymer --gt monomersComplexation (soluble complexes)3Pb+2 + P3O10

-5 --gt PbP3O10-2

higher Pb solubility

Slid

e cr

edit

Mar

c Ed

war

ds

Conceptual possible reactions Actual reactants and products may bedifferent Polyphosphates cannot be easily quantified

20

Ortho-P Treatment for Pb+2

Effectiveness Depends on Dose DIC pH and ldquoCleanlinessrdquo of Pipe Surface

mg PO4L00 10 20 30 40 50

mg

PbL

000

005

010

015

020

025

030

035

040

48 mg CL

48 mg CL

pH = 70pH = 75pH = 80pH = 85

bull pH less critical at low DICbull pH less critical at high PO4bull Point of diminishing returns higher with high DIC

bull Faster Pb reduction at high PO4

Typical UK Dosages 4-6 mgL

Most PWSs with LSLs do not have corrosion control treatment that minimizes Pb release and exposure

Effect of pH and ortho-P on Pb release

21

DIC = 10 mg CL 1 mg PO4L

60 65 70 75 80 85 90 95

gL

b m

P

0001

001

01

1

10

US Action Level

Dissolved Pb with ortho-PTotal Pb with ortho-PDissolved Pb with CO3Total Pb with CO3

At low DIC orthophosphate improves lead release regardless of pH

pHSchock M R DeSantis M K Metz D H Welch M M Hyland R N Nadagouda M N Revisiting the pH Effect on the Orthophosphate Control of Plumbosolvency Proc AWWA Annual Conference and Exposition Atlanta GA 2008

Ortho-P at pH 90 (DIC 6 mgL)

22

Pb(μgL)

bull Low Pb at 1 mgL relative to higher doses needs to be further verified across water qualities

bull Must do dose optimization study for your own water quality especially at high pH

bull Ortho-P may precipitate with Ca

Miller S A Investigation of Lead Solubility and Orthophosphate Addition in High pH Low DIC Water Master of Science Department of Biomedical Chemical and Environmental Engineering College of Engineering and Applied Science University of Cincinnati Cincinnati OH 2014

Ortho-Ppoint of diminishing returns

bull Orthophosphate addition to where large increments result in small reductions in lead release

bull Key to cost-effective lead release control and exposure reduction

bull Becomes the rdquomaintenancerdquo dosage unless the scale ldquoagesrdquo

bull Background constituents such as aluminum hardness ions iron manganese and others interfere with optimum reduction of lead release

bull Varies with the background water chemistry from system to system

23

Sodium silicate

bull No systematic studies to look at pH carbonate silicate background chemistry relationships

bull Unclear mechanism pH increase Buffer intensity increase Protective film (thin amp amorphous)

bull Can sequester ironmanganese

bull Canrsquot be evaluated with fresh surfaces

24pH5 6 7 8 9 10 11

β ty

nsi

nte

Iuf

fer

B

00000

00005

00010TIC=48 mg CL20 mgL SiO250 mgL PO4β H2OTotal β

Silicate may contribute to Buffer Intensity

Carbonate orthophosphate silicate I=001 25ordmC

Chemical changes may reduce Pb+4 to Pb+2

10

8

6

4

)ts 2ol

h (v

E

ndash2

ndash4

ndash6

ndash8

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14

pH

12

PbO2 (plattnerite)

Pb++ )s

deg

(2

CO

3

(OH

) -2 2) 3

Pb

-2

0 (CO

-- 2

0

) 43(C

O

Pb

3 (OH

)

DIC = 18 mg CL

Pb Pb Pb = 0010 mgL

Pb metal

-10

Drop in ORP from treatment change or DS

oxidant demand

Drop in pH at surface from treatment change

chemical reactions nitrification etc

C D

ngto

nhisa

W

Newark NJ

Disinfectant demand in DS must be controlled amp enough free chlorine consistently maintained throughout LSL area to keep protective Pb+4 corrosion scales in place

20

Ortho-P Treatment for Pb+2

Effectiveness Depends on Dose DIC pH and ldquoCleanlinessrdquo of Pipe Surface

mg PO4L00 10 20 30 40 50

mg

PbL

000

005

010

015

020

025

030

035

040

48 mg CL

48 mg CL

pH = 70pH = 75pH = 80pH = 85

bull pH less critical at low DICbull pH less critical at high PO4bull Point of diminishing returns higher with high DIC

bull Faster Pb reduction at high PO4

Typical UK Dosages 4-6 mgL

Most PWSs with LSLs do not have corrosion control treatment that minimizes Pb release and exposure

Effect of pH and ortho-P on Pb release

21

DIC = 10 mg CL 1 mg PO4L

60 65 70 75 80 85 90 95

gL

b m

P

0001

001

01

1

10

US Action Level

Dissolved Pb with ortho-PTotal Pb with ortho-PDissolved Pb with CO3Total Pb with CO3

At low DIC orthophosphate improves lead release regardless of pH

pHSchock M R DeSantis M K Metz D H Welch M M Hyland R N Nadagouda M N Revisiting the pH Effect on the Orthophosphate Control of Plumbosolvency Proc AWWA Annual Conference and Exposition Atlanta GA 2008

Ortho-P at pH 90 (DIC 6 mgL)

22

Pb(μgL)

bull Low Pb at 1 mgL relative to higher doses needs to be further verified across water qualities

bull Must do dose optimization study for your own water quality especially at high pH

bull Ortho-P may precipitate with Ca

Miller S A Investigation of Lead Solubility and Orthophosphate Addition in High pH Low DIC Water Master of Science Department of Biomedical Chemical and Environmental Engineering College of Engineering and Applied Science University of Cincinnati Cincinnati OH 2014

Ortho-Ppoint of diminishing returns

bull Orthophosphate addition to where large increments result in small reductions in lead release

bull Key to cost-effective lead release control and exposure reduction

bull Becomes the rdquomaintenancerdquo dosage unless the scale ldquoagesrdquo

bull Background constituents such as aluminum hardness ions iron manganese and others interfere with optimum reduction of lead release

bull Varies with the background water chemistry from system to system

23

Sodium silicate

bull No systematic studies to look at pH carbonate silicate background chemistry relationships

bull Unclear mechanism pH increase Buffer intensity increase Protective film (thin amp amorphous)

bull Can sequester ironmanganese

bull Canrsquot be evaluated with fresh surfaces

24pH5 6 7 8 9 10 11

β ty

nsi

nte

Iuf

fer

B

00000

00005

00010TIC=48 mg CL20 mgL SiO250 mgL PO4β H2OTotal β

Silicate may contribute to Buffer Intensity

Carbonate orthophosphate silicate I=001 25ordmC

Chemical changes may reduce Pb+4 to Pb+2

10

8

6

4

)ts 2ol

h (v

E

ndash2

ndash4

ndash6

ndash8

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14

pH

12

PbO2 (plattnerite)

Pb++ )s

deg

(2

CO

3

(OH

) -2 2) 3

Pb

-2

0 (CO

-- 2

0

) 43(C

O

Pb

3 (OH

)

DIC = 18 mg CL

Pb Pb Pb = 0010 mgL

Pb metal

-10

Drop in ORP from treatment change or DS

oxidant demand

Drop in pH at surface from treatment change

chemical reactions nitrification etc

C D

ngto

nhisa

W

Newark NJ

Disinfectant demand in DS must be controlled amp enough free chlorine consistently maintained throughout LSL area to keep protective Pb+4 corrosion scales in place

Effect of pH and ortho-P on Pb release

21

DIC = 10 mg CL 1 mg PO4L

60 65 70 75 80 85 90 95

gL

b m

P

0001

001

01

1

10

US Action Level

Dissolved Pb with ortho-PTotal Pb with ortho-PDissolved Pb with CO3Total Pb with CO3

At low DIC orthophosphate improves lead release regardless of pH

pHSchock M R DeSantis M K Metz D H Welch M M Hyland R N Nadagouda M N Revisiting the pH Effect on the Orthophosphate Control of Plumbosolvency Proc AWWA Annual Conference and Exposition Atlanta GA 2008

Ortho-P at pH 90 (DIC 6 mgL)

22

Pb(μgL)

bull Low Pb at 1 mgL relative to higher doses needs to be further verified across water qualities

bull Must do dose optimization study for your own water quality especially at high pH

bull Ortho-P may precipitate with Ca

Miller S A Investigation of Lead Solubility and Orthophosphate Addition in High pH Low DIC Water Master of Science Department of Biomedical Chemical and Environmental Engineering College of Engineering and Applied Science University of Cincinnati Cincinnati OH 2014

Ortho-Ppoint of diminishing returns

bull Orthophosphate addition to where large increments result in small reductions in lead release

bull Key to cost-effective lead release control and exposure reduction

bull Becomes the rdquomaintenancerdquo dosage unless the scale ldquoagesrdquo

bull Background constituents such as aluminum hardness ions iron manganese and others interfere with optimum reduction of lead release

bull Varies with the background water chemistry from system to system

23

Sodium silicate

bull No systematic studies to look at pH carbonate silicate background chemistry relationships

bull Unclear mechanism pH increase Buffer intensity increase Protective film (thin amp amorphous)

bull Can sequester ironmanganese

bull Canrsquot be evaluated with fresh surfaces

24pH5 6 7 8 9 10 11

β ty

nsi

nte

Iuf

fer

B

00000

00005

00010TIC=48 mg CL20 mgL SiO250 mgL PO4β H2OTotal β

Silicate may contribute to Buffer Intensity

Carbonate orthophosphate silicate I=001 25ordmC

Chemical changes may reduce Pb+4 to Pb+2

10

8

6

4

)ts 2ol

h (v

E

ndash2

ndash4

ndash6

ndash8

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14

pH

12

PbO2 (plattnerite)

Pb++ )s

deg

(2

CO

3

(OH

) -2 2) 3

Pb

-2

0 (CO

-- 2

0

) 43(C

O

Pb

3 (OH

)

DIC = 18 mg CL

Pb Pb Pb = 0010 mgL

Pb metal

-10

Drop in ORP from treatment change or DS

oxidant demand

Drop in pH at surface from treatment change

chemical reactions nitrification etc

C D

ngto

nhisa

W

Newark NJ

Disinfectant demand in DS must be controlled amp enough free chlorine consistently maintained throughout LSL area to keep protective Pb+4 corrosion scales in place

Ortho-P at pH 90 (DIC 6 mgL)

22

Pb(μgL)

bull Low Pb at 1 mgL relative to higher doses needs to be further verified across water qualities

bull Must do dose optimization study for your own water quality especially at high pH

bull Ortho-P may precipitate with Ca

Miller S A Investigation of Lead Solubility and Orthophosphate Addition in High pH Low DIC Water Master of Science Department of Biomedical Chemical and Environmental Engineering College of Engineering and Applied Science University of Cincinnati Cincinnati OH 2014

Ortho-Ppoint of diminishing returns

bull Orthophosphate addition to where large increments result in small reductions in lead release

bull Key to cost-effective lead release control and exposure reduction

bull Becomes the rdquomaintenancerdquo dosage unless the scale ldquoagesrdquo

bull Background constituents such as aluminum hardness ions iron manganese and others interfere with optimum reduction of lead release

bull Varies with the background water chemistry from system to system

23

Sodium silicate

bull No systematic studies to look at pH carbonate silicate background chemistry relationships

bull Unclear mechanism pH increase Buffer intensity increase Protective film (thin amp amorphous)

bull Can sequester ironmanganese

bull Canrsquot be evaluated with fresh surfaces

24pH5 6 7 8 9 10 11

β ty

nsi

nte

Iuf

fer

B

00000

00005

00010TIC=48 mg CL20 mgL SiO250 mgL PO4β H2OTotal β

Silicate may contribute to Buffer Intensity

Carbonate orthophosphate silicate I=001 25ordmC

Chemical changes may reduce Pb+4 to Pb+2

10

8

6

4

)ts 2ol

h (v

E

ndash2

ndash4

ndash6

ndash8

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14

pH

12

PbO2 (plattnerite)

Pb++ )s

deg

(2

CO

3

(OH

) -2 2) 3

Pb

-2

0 (CO

-- 2

0

) 43(C

O

Pb

3 (OH

)

DIC = 18 mg CL

Pb Pb Pb = 0010 mgL

Pb metal

-10

Drop in ORP from treatment change or DS

oxidant demand

Drop in pH at surface from treatment change

chemical reactions nitrification etc

C D

ngto

nhisa

W

Newark NJ

Disinfectant demand in DS must be controlled amp enough free chlorine consistently maintained throughout LSL area to keep protective Pb+4 corrosion scales in place

Ortho-Ppoint of diminishing returns

bull Orthophosphate addition to where large increments result in small reductions in lead release

bull Key to cost-effective lead release control and exposure reduction

bull Becomes the rdquomaintenancerdquo dosage unless the scale ldquoagesrdquo

bull Background constituents such as aluminum hardness ions iron manganese and others interfere with optimum reduction of lead release

bull Varies with the background water chemistry from system to system

23

Sodium silicate

bull No systematic studies to look at pH carbonate silicate background chemistry relationships

bull Unclear mechanism pH increase Buffer intensity increase Protective film (thin amp amorphous)

bull Can sequester ironmanganese

bull Canrsquot be evaluated with fresh surfaces

24pH5 6 7 8 9 10 11

β ty

nsi

nte

Iuf

fer

B

00000

00005

00010TIC=48 mg CL20 mgL SiO250 mgL PO4β H2OTotal β

Silicate may contribute to Buffer Intensity

Carbonate orthophosphate silicate I=001 25ordmC

Chemical changes may reduce Pb+4 to Pb+2

10

8

6

4

)ts 2ol

h (v

E

ndash2

ndash4

ndash6

ndash8

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14

pH

12

PbO2 (plattnerite)

Pb++ )s

deg

(2

CO

3

(OH

) -2 2) 3

Pb

-2

0 (CO

-- 2

0

) 43(C

O

Pb

3 (OH

)

DIC = 18 mg CL

Pb Pb Pb = 0010 mgL

Pb metal

-10

Drop in ORP from treatment change or DS

oxidant demand

Drop in pH at surface from treatment change

chemical reactions nitrification etc

C D

ngto

nhisa

W

Newark NJ

Disinfectant demand in DS must be controlled amp enough free chlorine consistently maintained throughout LSL area to keep protective Pb+4 corrosion scales in place

Sodium silicate

bull No systematic studies to look at pH carbonate silicate background chemistry relationships

bull Unclear mechanism pH increase Buffer intensity increase Protective film (thin amp amorphous)

bull Can sequester ironmanganese

bull Canrsquot be evaluated with fresh surfaces

24pH5 6 7 8 9 10 11

β ty

nsi

nte

Iuf

fer

B

00000

00005

00010TIC=48 mg CL20 mgL SiO250 mgL PO4β H2OTotal β

Silicate may contribute to Buffer Intensity

Carbonate orthophosphate silicate I=001 25ordmC

Chemical changes may reduce Pb+4 to Pb+2

10

8

6

4

)ts 2ol

h (v

E

ndash2

ndash4

ndash6

ndash8

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14

pH

12

PbO2 (plattnerite)

Pb++ )s

deg

(2

CO

3

(OH

) -2 2) 3

Pb

-2

0 (CO

-- 2

0

) 43(C

O

Pb

3 (OH

)

DIC = 18 mg CL

Pb Pb Pb = 0010 mgL

Pb metal

-10

Drop in ORP from treatment change or DS

oxidant demand

Drop in pH at surface from treatment change

chemical reactions nitrification etc

C D

ngto

nhisa

W

Newark NJ

Disinfectant demand in DS must be controlled amp enough free chlorine consistently maintained throughout LSL area to keep protective Pb+4 corrosion scales in place

Chemical changes may reduce Pb+4 to Pb+2

10

8

6

4

)ts 2ol

h (v

E

ndash2

ndash4

ndash6

ndash8

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14

pH

12

PbO2 (plattnerite)

Pb++ )s

deg

(2

CO

3

(OH

) -2 2) 3

Pb

-2

0 (CO

-- 2

0

) 43(C

O

Pb

3 (OH

)

DIC = 18 mg CL

Pb Pb Pb = 0010 mgL

Pb metal

-10

Drop in ORP from treatment change or DS

oxidant demand

Drop in pH at surface from treatment change

chemical reactions nitrification etc

C D

ngto

nhisa

W

Newark NJ

Disinfectant demand in DS must be controlled amp enough free chlorine consistently maintained throughout LSL area to keep protective Pb+4 corrosion scales in place

Switching disinfectants may reduce Pb+4 to Pb+2

Low

Switch from chlorine to chloramine disinfectant dissolved lead from pipe scales during the Washington DC ldquolead-in-water crisisrdquo in 2001-2004

Example profile of Pb(+4) (ie PbO2) scale house

27

20

L)

Cl1 Cincinnati OH

18

microg 16

n ( 14

o 12

tia 10

trn 8

e 6

nco 4

C 2

b P 0-1 0 1 2 3 4 5 6 7 8 9 10 11

Cumulative Water Volume (L)

CI-1 10 h 1209CI-1 10 h 0410

Plumbing sequence corresponding to water volumeFaucet LSLCu pipe Brass

Water MainMCu pipe 2

015

a

let

dou

illyaf

ant

irT

bull Pb release to water from PbO2 coated LSLs was found to be lowbull Be aware are there are no known cases of exhumed PbO2 coated pipes being able to

re-stabilize in pipe rigs

Chloride and sulfateCSMR a factor controlling galvanic corrosion (Oliphant 1983 Gregory 1986)- Chloride detrimental- Sulfate beneficial if sufficient to overcome chloride thus- CSMR ratio was created

[Cl- ] 12 mgL Cl-Chloride to Sulfate Mass Ratio (CSMR) = =[SO -2 2 = 06

4 ] 20 mgL SO-4

Confirmed empirically in case studies of water utilities (Dodrill and Edwards 1995 Edwards and Triantafyllidou 2007) - High CSMR gt 05-06 more corrosive- Low CSMR lt 05-06 less corrosive

28

Verified in laboratory experiments with fresh galvanic Pb junctions (eg Wang et al 2013 Triantafyllidou et al 2011 Nguyen et al 2010 2011)

Tria

ntaf

yllid

ou e

t al

201

1

Example profile of Pb(+4) (ie PbO2) scale house

27

20

L)

Cl1 Cincinnati OH

18

microg 16

n ( 14

o 12

tia 10

trn 8

e 6

nco 4

C 2

b P 0-1 0 1 2 3 4 5 6 7 8 9 10 11

Cumulative Water Volume (L)

CI-1 10 h 1209CI-1 10 h 0410

Plumbing sequence corresponding to water volumeFaucet LSLCu pipe Brass

Water MainMCu pipe 2

015

a

let

dou

illyaf

ant

irT

bull Pb release to water from PbO2 coated LSLs was found to be lowbull Be aware are there are no known cases of exhumed PbO2 coated pipes being able to

re-stabilize in pipe rigs

Chloride and sulfateCSMR a factor controlling galvanic corrosion (Oliphant 1983 Gregory 1986)- Chloride detrimental- Sulfate beneficial if sufficient to overcome chloride thus- CSMR ratio was created

[Cl- ] 12 mgL Cl-Chloride to Sulfate Mass Ratio (CSMR) = =[SO -2 2 = 06

4 ] 20 mgL SO-4

Confirmed empirically in case studies of water utilities (Dodrill and Edwards 1995 Edwards and Triantafyllidou 2007) - High CSMR gt 05-06 more corrosive- Low CSMR lt 05-06 less corrosive

28

Verified in laboratory experiments with fresh galvanic Pb junctions (eg Wang et al 2013 Triantafyllidou et al 2011 Nguyen et al 2010 2011)

Tria

ntaf

yllid

ou e

t al

201

1

Chloride and sulfateCSMR a factor controlling galvanic corrosion (Oliphant 1983 Gregory 1986)- Chloride detrimental- Sulfate beneficial if sufficient to overcome chloride thus- CSMR ratio was created

[Cl- ] 12 mgL Cl-Chloride to Sulfate Mass Ratio (CSMR) = =[SO -2 2 = 06

4 ] 20 mgL SO-4

Confirmed empirically in case studies of water utilities (Dodrill and Edwards 1995 Edwards and Triantafyllidou 2007) - High CSMR gt 05-06 more corrosive- Low CSMR lt 05-06 less corrosive

28

Verified in laboratory experiments with fresh galvanic Pb junctions (eg Wang et al 2013 Triantafyllidou et al 2011 Nguyen et al 2010 2011)

Tria

ntaf

yllid

ou e

t al

201

1

Chloride and sulfate

29

CSMR considered contributing factor to the Flint water Pb crisis (Pieper et al 2017)

30

Chloride and sulfate

bull Absolute concentrations of chloride and sulfate important chloride considered particularly detrimental to galvanic Pb corrosion

bull Increasing chloride and sulfate concentrations at a constant CSMR of 1 increased total lead concentrations in galvanic corrosion experiments (Ng amp Lin 2016)

bull High CSMR ratio proven empirically important in aggravating galvanic lead corrosion (verified by experiments and modeling)

bull Water source or treatment changes that affect the CSMR must be evaluated prior to change

bull Absolute concentrations also important particularly for chloride

31

Manganese

Two distinct pipe scale categories in 5 LSLs (Lead 1-5 Pipes) excavated from same city

31

Schock et al 2014 JAWWAbull The outer pipe scale is in direct contact with the flowing water affecting

water lead contamination the mostbull Plattnerite an insoluble Pb(+4) scale that protects water from lead

contamination was inhibited by Mn (and Fe) presence

Manganese

In this case study (Schock et al JAWWA 2014)bull Where Mn was present in more abundance

- Pb scales had less PbO2- Pb associated with Mn sorbing onto Mn+Fe-rich surface deposit

bull Erratic Pb persisted after full LSLR in areas rich with Mn deposits (and to lesser extent Fe)

bull Galvanized interior plumbing also accumulated dissolved Pb while LSL present

32

33

Manganese

Other studiesbull Dong et al 2000 Mn acted as vector for particulate Pb movementbull Trueman et al 2019 Increasing Mn increased Pb in experimentbull McNeill and Edwards 2004 Mn sequestration by polyphosphate destabilized

Pb scales and caused Pb release

bull Pan et al 2019 Possible benefit by promoting formation of Pb(+4)

Reference Analysis Mn Impact on PbSchock et al 2014 Excavated LSLs

amp historical water Pb data in a US city

bull Promoted amorphous Pb scalebull Inhibited insoluble Pb(+4) scalebull Erratic Pb persisted after full LSLR in areas rich

with Mn deposits

Tully et al 2019 Excavated LSLs amp water Pb data in several Midwest case studies

bull Mn was a component of amorphous pipe scales in several systems

bull Amorphous scales not amenable to determining their adherence to pipe walls their responses to changes in various water chemistry variables or quantitatively calculating their lead release

Itrsquos not just manganese

34

Broad variety of coatings on LSLs analyzed at EPA

Al Mn Fe P and Ca-rich coatings may interfere withorthophosphate pH adjustment or other CCT

35

More water parameters affect metal release

bull pH amp AlkalinityDICbull Concentration and type of

oxidants (disinfectantsdissolved oxygen)

bull ORPcorrosion potentialbull Corrosion Inhibitorsbull Chloridebull Sulfatebull Manganese

bull Temperaturebull Sorptive surfaces downstream of

LSLs (ie galvanized interior pipe)bull Iron (deposition and corrosion)bull Calciumbull Aluminumbull NOM (type amount)bull Mixing from water treatment plants

or sourcesbull Ammoniabull Hydrogen Sulfidebull Microbial activity (nitrification amp

other)

Corrosion Control Treatment is intertwined with all treatments affecting water chemistry

Water Quality Parameters(WQP)

bull pH alkalinity calcium orthophosphate or silica othersbull WQP measurements are important for

bull Process control (measurement frequency needs to be HIGH)bull Keeping treatment from being turned offbull Keeping inhibitor dosages amp pH control from being cut to save $$$bull Keeping required repairs to treatment on tight schedulebull Preventing frequent variations and alternations of water sources

bull Effectiveness of Pb treatment only determinable from tap sampling by combination of different assessment approaches for different lead plumbing sources to which people can be exposed

36

bull OWQP measurements important for process control look at yourdata to understand what they mean and avoid problems

bull OWQP do not predict lead release There are no substitutes fordirectly monitoring lead release

37

M+

M+

M+

M+

Corrosion barrier film hypothesis

Inert barrier film (eg CaCO3)

Pipe Wall

Barrier film would allow tobull Protect ALL pipe materialsbull Monitor corrosion based on one chemical reaction

Although barrier film would be convenient for important reasonsthe CaCO3 hypothesis has not been proven

38

In excavated pipe samples that the EPA analyzed

Inert barrier film (eg CaCO3) is a complex scale

+ M+M

M+

M+ Scale- Heterogeneous- Several layers- Amorphous amp crystalline

Pipe WallPb and non-Pb compounds

Although barrier film would be convenient for important reasonsthe CaCO3 hypothesis has not been proven and scale is overall

Are these ldquocorrosionrdquo indicesIndex Equation Range Water Condition

Langelier Saturation Index

LSI = pHactual - pHs LSIgt0

LSIlt0

bull Supersaturated tends to precipitate CaCO3

bull Undersaturated tends to dissolve CaCO3

Ryznar Stability Index RSI = 2pHs-pHactual RSIgt6

RSIlt6

bull Undersaturated tends to dissolve CaCO3

bull Supersaturated tends to precipitate CaCO3

Aggressive Index AI=pH+ log(AH) AIlt1010ltAIlt12AIgt12

bull Highly Aggressivebull Moderately Aggressivebull Non-aggressive

Larson Scold Index Ls=[Cl-+SO4-2] Alkalinity Lslt08

08ltLslt12

Lsgt12

bull Chloride amp Sulfate will not likely interfere with natural film formation

bull chloride amp sulfate may interfere with natural film formation

bull Tendency toward higher localized corrosion rates

39

bull Most indices were not developed or named for corrosion (eg saturation stability etc)bull The LSI and subsequent variations predict CaCO3 SCALING not corrosion or metal

releasebull Calcium carbonate was not the active inhibition mechanism in the hundreds lead and

copper pipes analyzed by EPA

ldquoCorrosionrdquo indices

bull When used outside their limitations these indices become unreliable

bull Different objectives for different materialsbull THERE IS NO UNIVERSAL ldquoINDEXrdquo FOR

MULTIPLE MATERIALS

40

Relying on the LSI as an indicator of a waterrsquos corrosivityshould be abandoned (AWWARF 1996 Internal Corrosion ofDistribution Systems)

Summary

41

bull Materials inventory is critical to know where and what lead sourcesstill exist

bull To understand corrosion and lead release it is important to keep upwith pH the carbonate system (ie dissolved inorganic carbon andalkalinity) and disinfectants (ie oxidation-reduction potential)

bull Corrosion inhibitor efficiency depends on type and sufficientdoseconcentration for the background pH and other reactive waterquality constituents

bull There are many types of scale on lead pipe with different solubilitiesPb(+4) solubility ltlt Pb(+2)

bull Optimum Water Quality Parameters (OWQP) important for processcontrol look at your data to understand what they mean and avoidproblems

bull Setting OWQP or calculating ldquocorrosionrdquo indices does not predictlead release

How we communicate our research (recent examples)

Peer-review journal papersbull Doreacute E Lytle DA Wasserstrom L Swertfeger J Triantafyllidou S Field Analyzers for

Lead Quantification in Drinking Water Samples Critical Reviews in Environmental Science and Technology 2020 httpsdoiorg1010801064338920201782654

bull Lytle D A Schock M R Wait K Cahalan K Bosscher V Porter A Del Toral M Sequential Drinking Water Sampling as a Tool for Evaluating Lead in Flint Michigan Water Res 2019 157 40-54

bull Tully J DeSantis M K Schock M R Water QualityndashPipe Deposit Relationships in Midwestern Lead Pipes AWWA Water Science 2019 1 (2) e1127

bull DeSantis M K Triantafyllidou S Schock M R Lytle D A Mineralogical Evidence of Galvanic Corrosion in Drinking Water Lead Pipe Joints Environ Sci Technol 2018 52 (6) 3365-3374

Conference presentationsbull Triantafyllidou S Hensley K Lytle D James R Lal V Tools to Identify Lead Service

Lines AWWA Water Quality Technology Conference Dallas 2019bull Schock M Triantafyllidou S Tully J DeSantis M and Lytle D Diagnostic Sampling

Tools for Lead in Drinking Water AWWA Annual Conference and Exposition Denver 2019bull Tully J DeSantis M and Schock M Actual vs Predicted Lead Scale Observations from

the Field AWWA WQTC Toronto 2018

How we communicate our research (recent examples)

Magazine Articlesbull Lytle D A Schock M R Triantafyllidou S Identify Lead Plumbing Sources to Protect

Public Health AWWA Opflow 2018 44 (3) 16-20EPA Science Matters Newslettersbull Revealing the Complicated Nature of Tap Water Lead Contamination A Madison Wisconsin

Case Study July 30 2018bull Identifying the Best Lead Sampling Techniques to Protect Public Health October 22 2018bull Scaling Back EPA Researchers Help Communities Protect Drinking Water Systems from Lead

April 8 2019bull Search httpswwwepagovsciencemattersFact Sheets on Leadbull httpswwwepagovwater-researchwater-research-fact-sheetsDWHow to Identify Lead Free Certification Marks for Drinking Water System and Plumbing ProductsConsumer Tool for Identifying POU Drinking Water Filters Certified to Reduce LeadStandard Developmentbull Participate on the workgroups for NSF Drinking Water Treatment Units (point-of-use filters)

NSF 6061 (Health Effects for Drinking Water Treatment Chemicals and Health Effects for Drinking Water System Components) and ASHRAE Standard 514 (Building water quality)

30

Chloride and sulfate

bull Absolute concentrations of chloride and sulfate important chloride considered particularly detrimental to galvanic Pb corrosion

bull Increasing chloride and sulfate concentrations at a constant CSMR of 1 increased total lead concentrations in galvanic corrosion experiments (Ng amp Lin 2016)

bull High CSMR ratio proven empirically important in aggravating galvanic lead corrosion (verified by experiments and modeling)

bull Water source or treatment changes that affect the CSMR must be evaluated prior to change

bull Absolute concentrations also important particularly for chloride

31

Manganese

Two distinct pipe scale categories in 5 LSLs (Lead 1-5 Pipes) excavated from same city

31

Schock et al 2014 JAWWAbull The outer pipe scale is in direct contact with the flowing water affecting

water lead contamination the mostbull Plattnerite an insoluble Pb(+4) scale that protects water from lead

contamination was inhibited by Mn (and Fe) presence

Manganese

In this case study (Schock et al JAWWA 2014)bull Where Mn was present in more abundance

- Pb scales had less PbO2- Pb associated with Mn sorbing onto Mn+Fe-rich surface deposit

bull Erratic Pb persisted after full LSLR in areas rich with Mn deposits (and to lesser extent Fe)

bull Galvanized interior plumbing also accumulated dissolved Pb while LSL present

32

33

Manganese

Other studiesbull Dong et al 2000 Mn acted as vector for particulate Pb movementbull Trueman et al 2019 Increasing Mn increased Pb in experimentbull McNeill and Edwards 2004 Mn sequestration by polyphosphate destabilized

Pb scales and caused Pb release

bull Pan et al 2019 Possible benefit by promoting formation of Pb(+4)

Reference Analysis Mn Impact on PbSchock et al 2014 Excavated LSLs

amp historical water Pb data in a US city

bull Promoted amorphous Pb scalebull Inhibited insoluble Pb(+4) scalebull Erratic Pb persisted after full LSLR in areas rich

with Mn deposits

Tully et al 2019 Excavated LSLs amp water Pb data in several Midwest case studies

bull Mn was a component of amorphous pipe scales in several systems

bull Amorphous scales not amenable to determining their adherence to pipe walls their responses to changes in various water chemistry variables or quantitatively calculating their lead release

Itrsquos not just manganese

34

Broad variety of coatings on LSLs analyzed at EPA

Al Mn Fe P and Ca-rich coatings may interfere withorthophosphate pH adjustment or other CCT

35

More water parameters affect metal release

bull pH amp AlkalinityDICbull Concentration and type of

oxidants (disinfectantsdissolved oxygen)

bull ORPcorrosion potentialbull Corrosion Inhibitorsbull Chloridebull Sulfatebull Manganese

bull Temperaturebull Sorptive surfaces downstream of

LSLs (ie galvanized interior pipe)bull Iron (deposition and corrosion)bull Calciumbull Aluminumbull NOM (type amount)bull Mixing from water treatment plants

or sourcesbull Ammoniabull Hydrogen Sulfidebull Microbial activity (nitrification amp

other)

Corrosion Control Treatment is intertwined with all treatments affecting water chemistry

Water Quality Parameters(WQP)

bull pH alkalinity calcium orthophosphate or silica othersbull WQP measurements are important for

bull Process control (measurement frequency needs to be HIGH)bull Keeping treatment from being turned offbull Keeping inhibitor dosages amp pH control from being cut to save $$$bull Keeping required repairs to treatment on tight schedulebull Preventing frequent variations and alternations of water sources

bull Effectiveness of Pb treatment only determinable from tap sampling by combination of different assessment approaches for different lead plumbing sources to which people can be exposed

36

bull OWQP measurements important for process control look at yourdata to understand what they mean and avoid problems

bull OWQP do not predict lead release There are no substitutes fordirectly monitoring lead release

37

M+

M+

M+

M+

Corrosion barrier film hypothesis

Inert barrier film (eg CaCO3)

Pipe Wall

Barrier film would allow tobull Protect ALL pipe materialsbull Monitor corrosion based on one chemical reaction

Although barrier film would be convenient for important reasonsthe CaCO3 hypothesis has not been proven

38

In excavated pipe samples that the EPA analyzed

Inert barrier film (eg CaCO3) is a complex scale

+ M+M

M+

M+ Scale- Heterogeneous- Several layers- Amorphous amp crystalline

Pipe WallPb and non-Pb compounds

Although barrier film would be convenient for important reasonsthe CaCO3 hypothesis has not been proven and scale is overall

Are these ldquocorrosionrdquo indicesIndex Equation Range Water Condition

Langelier Saturation Index

LSI = pHactual - pHs LSIgt0

LSIlt0

bull Supersaturated tends to precipitate CaCO3

bull Undersaturated tends to dissolve CaCO3

Ryznar Stability Index RSI = 2pHs-pHactual RSIgt6

RSIlt6

bull Undersaturated tends to dissolve CaCO3

bull Supersaturated tends to precipitate CaCO3

Aggressive Index AI=pH+ log(AH) AIlt1010ltAIlt12AIgt12

bull Highly Aggressivebull Moderately Aggressivebull Non-aggressive

Larson Scold Index Ls=[Cl-+SO4-2] Alkalinity Lslt08

08ltLslt12

Lsgt12

bull Chloride amp Sulfate will not likely interfere with natural film formation

bull chloride amp sulfate may interfere with natural film formation

bull Tendency toward higher localized corrosion rates

39

bull Most indices were not developed or named for corrosion (eg saturation stability etc)bull The LSI and subsequent variations predict CaCO3 SCALING not corrosion or metal

releasebull Calcium carbonate was not the active inhibition mechanism in the hundreds lead and

copper pipes analyzed by EPA

ldquoCorrosionrdquo indices

bull When used outside their limitations these indices become unreliable

bull Different objectives for different materialsbull THERE IS NO UNIVERSAL ldquoINDEXrdquo FOR

MULTIPLE MATERIALS

40

Relying on the LSI as an indicator of a waterrsquos corrosivityshould be abandoned (AWWARF 1996 Internal Corrosion ofDistribution Systems)

Summary

41

bull Materials inventory is critical to know where and what lead sourcesstill exist

bull To understand corrosion and lead release it is important to keep upwith pH the carbonate system (ie dissolved inorganic carbon andalkalinity) and disinfectants (ie oxidation-reduction potential)

bull Corrosion inhibitor efficiency depends on type and sufficientdoseconcentration for the background pH and other reactive waterquality constituents

bull There are many types of scale on lead pipe with different solubilitiesPb(+4) solubility ltlt Pb(+2)

bull Optimum Water Quality Parameters (OWQP) important for processcontrol look at your data to understand what they mean and avoidproblems

bull Setting OWQP or calculating ldquocorrosionrdquo indices does not predictlead release

How we communicate our research (recent examples)

Peer-review journal papersbull Doreacute E Lytle DA Wasserstrom L Swertfeger J Triantafyllidou S Field Analyzers for

Lead Quantification in Drinking Water Samples Critical Reviews in Environmental Science and Technology 2020 httpsdoiorg1010801064338920201782654

bull Lytle D A Schock M R Wait K Cahalan K Bosscher V Porter A Del Toral M Sequential Drinking Water Sampling as a Tool for Evaluating Lead in Flint Michigan Water Res 2019 157 40-54

bull Tully J DeSantis M K Schock M R Water QualityndashPipe Deposit Relationships in Midwestern Lead Pipes AWWA Water Science 2019 1 (2) e1127

bull DeSantis M K Triantafyllidou S Schock M R Lytle D A Mineralogical Evidence of Galvanic Corrosion in Drinking Water Lead Pipe Joints Environ Sci Technol 2018 52 (6) 3365-3374

Conference presentationsbull Triantafyllidou S Hensley K Lytle D James R Lal V Tools to Identify Lead Service

Lines AWWA Water Quality Technology Conference Dallas 2019bull Schock M Triantafyllidou S Tully J DeSantis M and Lytle D Diagnostic Sampling

Tools for Lead in Drinking Water AWWA Annual Conference and Exposition Denver 2019bull Tully J DeSantis M and Schock M Actual vs Predicted Lead Scale Observations from

the Field AWWA WQTC Toronto 2018

How we communicate our research (recent examples)

Magazine Articlesbull Lytle D A Schock M R Triantafyllidou S Identify Lead Plumbing Sources to Protect

Public Health AWWA Opflow 2018 44 (3) 16-20EPA Science Matters Newslettersbull Revealing the Complicated Nature of Tap Water Lead Contamination A Madison Wisconsin

Case Study July 30 2018bull Identifying the Best Lead Sampling Techniques to Protect Public Health October 22 2018bull Scaling Back EPA Researchers Help Communities Protect Drinking Water Systems from Lead

April 8 2019bull Search httpswwwepagovsciencemattersFact Sheets on Leadbull httpswwwepagovwater-researchwater-research-fact-sheetsDWHow to Identify Lead Free Certification Marks for Drinking Water System and Plumbing ProductsConsumer Tool for Identifying POU Drinking Water Filters Certified to Reduce LeadStandard Developmentbull Participate on the workgroups for NSF Drinking Water Treatment Units (point-of-use filters)

NSF 6061 (Health Effects for Drinking Water Treatment Chemicals and Health Effects for Drinking Water System Components) and ASHRAE Standard 514 (Building water quality)

31

Manganese

Two distinct pipe scale categories in 5 LSLs (Lead 1-5 Pipes) excavated from same city

31

Schock et al 2014 JAWWAbull The outer pipe scale is in direct contact with the flowing water affecting

water lead contamination the mostbull Plattnerite an insoluble Pb(+4) scale that protects water from lead

contamination was inhibited by Mn (and Fe) presence

Manganese

In this case study (Schock et al JAWWA 2014)bull Where Mn was present in more abundance

- Pb scales had less PbO2- Pb associated with Mn sorbing onto Mn+Fe-rich surface deposit

bull Erratic Pb persisted after full LSLR in areas rich with Mn deposits (and to lesser extent Fe)

bull Galvanized interior plumbing also accumulated dissolved Pb while LSL present

32

33

Manganese

Other studiesbull Dong et al 2000 Mn acted as vector for particulate Pb movementbull Trueman et al 2019 Increasing Mn increased Pb in experimentbull McNeill and Edwards 2004 Mn sequestration by polyphosphate destabilized

Pb scales and caused Pb release

bull Pan et al 2019 Possible benefit by promoting formation of Pb(+4)

Reference Analysis Mn Impact on PbSchock et al 2014 Excavated LSLs

amp historical water Pb data in a US city

bull Promoted amorphous Pb scalebull Inhibited insoluble Pb(+4) scalebull Erratic Pb persisted after full LSLR in areas rich

with Mn deposits

Tully et al 2019 Excavated LSLs amp water Pb data in several Midwest case studies

bull Mn was a component of amorphous pipe scales in several systems

bull Amorphous scales not amenable to determining their adherence to pipe walls their responses to changes in various water chemistry variables or quantitatively calculating their lead release

Itrsquos not just manganese

34

Broad variety of coatings on LSLs analyzed at EPA

Al Mn Fe P and Ca-rich coatings may interfere withorthophosphate pH adjustment or other CCT

35

More water parameters affect metal release

bull pH amp AlkalinityDICbull Concentration and type of

oxidants (disinfectantsdissolved oxygen)

bull ORPcorrosion potentialbull Corrosion Inhibitorsbull Chloridebull Sulfatebull Manganese

bull Temperaturebull Sorptive surfaces downstream of

LSLs (ie galvanized interior pipe)bull Iron (deposition and corrosion)bull Calciumbull Aluminumbull NOM (type amount)bull Mixing from water treatment plants

or sourcesbull Ammoniabull Hydrogen Sulfidebull Microbial activity (nitrification amp

other)

Corrosion Control Treatment is intertwined with all treatments affecting water chemistry

Water Quality Parameters(WQP)

bull pH alkalinity calcium orthophosphate or silica othersbull WQP measurements are important for

bull Process control (measurement frequency needs to be HIGH)bull Keeping treatment from being turned offbull Keeping inhibitor dosages amp pH control from being cut to save $$$bull Keeping required repairs to treatment on tight schedulebull Preventing frequent variations and alternations of water sources

bull Effectiveness of Pb treatment only determinable from tap sampling by combination of different assessment approaches for different lead plumbing sources to which people can be exposed

36

bull OWQP measurements important for process control look at yourdata to understand what they mean and avoid problems

bull OWQP do not predict lead release There are no substitutes fordirectly monitoring lead release

37

M+

M+

M+

M+

Corrosion barrier film hypothesis

Inert barrier film (eg CaCO3)

Pipe Wall

Barrier film would allow tobull Protect ALL pipe materialsbull Monitor corrosion based on one chemical reaction

Although barrier film would be convenient for important reasonsthe CaCO3 hypothesis has not been proven

38

In excavated pipe samples that the EPA analyzed

Inert barrier film (eg CaCO3) is a complex scale

+ M+M

M+

M+ Scale- Heterogeneous- Several layers- Amorphous amp crystalline

Pipe WallPb and non-Pb compounds

Although barrier film would be convenient for important reasonsthe CaCO3 hypothesis has not been proven and scale is overall

Are these ldquocorrosionrdquo indicesIndex Equation Range Water Condition

Langelier Saturation Index

LSI = pHactual - pHs LSIgt0

LSIlt0

bull Supersaturated tends to precipitate CaCO3

bull Undersaturated tends to dissolve CaCO3

Ryznar Stability Index RSI = 2pHs-pHactual RSIgt6

RSIlt6

bull Undersaturated tends to dissolve CaCO3

bull Supersaturated tends to precipitate CaCO3

Aggressive Index AI=pH+ log(AH) AIlt1010ltAIlt12AIgt12

bull Highly Aggressivebull Moderately Aggressivebull Non-aggressive

Larson Scold Index Ls=[Cl-+SO4-2] Alkalinity Lslt08

08ltLslt12

Lsgt12

bull Chloride amp Sulfate will not likely interfere with natural film formation

bull chloride amp sulfate may interfere with natural film formation

bull Tendency toward higher localized corrosion rates

39

bull Most indices were not developed or named for corrosion (eg saturation stability etc)bull The LSI and subsequent variations predict CaCO3 SCALING not corrosion or metal

releasebull Calcium carbonate was not the active inhibition mechanism in the hundreds lead and

copper pipes analyzed by EPA

ldquoCorrosionrdquo indices

bull When used outside their limitations these indices become unreliable

bull Different objectives for different materialsbull THERE IS NO UNIVERSAL ldquoINDEXrdquo FOR

MULTIPLE MATERIALS

40

Relying on the LSI as an indicator of a waterrsquos corrosivityshould be abandoned (AWWARF 1996 Internal Corrosion ofDistribution Systems)

Summary

41

bull Materials inventory is critical to know where and what lead sourcesstill exist

bull To understand corrosion and lead release it is important to keep upwith pH the carbonate system (ie dissolved inorganic carbon andalkalinity) and disinfectants (ie oxidation-reduction potential)

bull Corrosion inhibitor efficiency depends on type and sufficientdoseconcentration for the background pH and other reactive waterquality constituents

bull There are many types of scale on lead pipe with different solubilitiesPb(+4) solubility ltlt Pb(+2)

bull Optimum Water Quality Parameters (OWQP) important for processcontrol look at your data to understand what they mean and avoidproblems

bull Setting OWQP or calculating ldquocorrosionrdquo indices does not predictlead release

How we communicate our research (recent examples)

Peer-review journal papersbull Doreacute E Lytle DA Wasserstrom L Swertfeger J Triantafyllidou S Field Analyzers for

Lead Quantification in Drinking Water Samples Critical Reviews in Environmental Science and Technology 2020 httpsdoiorg1010801064338920201782654

bull Lytle D A Schock M R Wait K Cahalan K Bosscher V Porter A Del Toral M Sequential Drinking Water Sampling as a Tool for Evaluating Lead in Flint Michigan Water Res 2019 157 40-54

bull Tully J DeSantis M K Schock M R Water QualityndashPipe Deposit Relationships in Midwestern Lead Pipes AWWA Water Science 2019 1 (2) e1127

bull DeSantis M K Triantafyllidou S Schock M R Lytle D A Mineralogical Evidence of Galvanic Corrosion in Drinking Water Lead Pipe Joints Environ Sci Technol 2018 52 (6) 3365-3374

Conference presentationsbull Triantafyllidou S Hensley K Lytle D James R Lal V Tools to Identify Lead Service

Lines AWWA Water Quality Technology Conference Dallas 2019bull Schock M Triantafyllidou S Tully J DeSantis M and Lytle D Diagnostic Sampling

Tools for Lead in Drinking Water AWWA Annual Conference and Exposition Denver 2019bull Tully J DeSantis M and Schock M Actual vs Predicted Lead Scale Observations from

the Field AWWA WQTC Toronto 2018

How we communicate our research (recent examples)

Magazine Articlesbull Lytle D A Schock M R Triantafyllidou S Identify Lead Plumbing Sources to Protect

Public Health AWWA Opflow 2018 44 (3) 16-20EPA Science Matters Newslettersbull Revealing the Complicated Nature of Tap Water Lead Contamination A Madison Wisconsin

Case Study July 30 2018bull Identifying the Best Lead Sampling Techniques to Protect Public Health October 22 2018bull Scaling Back EPA Researchers Help Communities Protect Drinking Water Systems from Lead

April 8 2019bull Search httpswwwepagovsciencemattersFact Sheets on Leadbull httpswwwepagovwater-researchwater-research-fact-sheetsDWHow to Identify Lead Free Certification Marks for Drinking Water System and Plumbing ProductsConsumer Tool for Identifying POU Drinking Water Filters Certified to Reduce LeadStandard Developmentbull Participate on the workgroups for NSF Drinking Water Treatment Units (point-of-use filters)

NSF 6061 (Health Effects for Drinking Water Treatment Chemicals and Health Effects for Drinking Water System Components) and ASHRAE Standard 514 (Building water quality)

Manganese

In this case study (Schock et al JAWWA 2014)bull Where Mn was present in more abundance

- Pb scales had less PbO2- Pb associated with Mn sorbing onto Mn+Fe-rich surface deposit

bull Erratic Pb persisted after full LSLR in areas rich with Mn deposits (and to lesser extent Fe)

bull Galvanized interior plumbing also accumulated dissolved Pb while LSL present

32

33

Manganese

Other studiesbull Dong et al 2000 Mn acted as vector for particulate Pb movementbull Trueman et al 2019 Increasing Mn increased Pb in experimentbull McNeill and Edwards 2004 Mn sequestration by polyphosphate destabilized

Pb scales and caused Pb release

bull Pan et al 2019 Possible benefit by promoting formation of Pb(+4)

Reference Analysis Mn Impact on PbSchock et al 2014 Excavated LSLs

amp historical water Pb data in a US city

bull Promoted amorphous Pb scalebull Inhibited insoluble Pb(+4) scalebull Erratic Pb persisted after full LSLR in areas rich

with Mn deposits

Tully et al 2019 Excavated LSLs amp water Pb data in several Midwest case studies

bull Mn was a component of amorphous pipe scales in several systems

bull Amorphous scales not amenable to determining their adherence to pipe walls their responses to changes in various water chemistry variables or quantitatively calculating their lead release

Itrsquos not just manganese

34

Broad variety of coatings on LSLs analyzed at EPA

Al Mn Fe P and Ca-rich coatings may interfere withorthophosphate pH adjustment or other CCT

35

More water parameters affect metal release

bull pH amp AlkalinityDICbull Concentration and type of

oxidants (disinfectantsdissolved oxygen)

bull ORPcorrosion potentialbull Corrosion Inhibitorsbull Chloridebull Sulfatebull Manganese

bull Temperaturebull Sorptive surfaces downstream of

LSLs (ie galvanized interior pipe)bull Iron (deposition and corrosion)bull Calciumbull Aluminumbull NOM (type amount)bull Mixing from water treatment plants

or sourcesbull Ammoniabull Hydrogen Sulfidebull Microbial activity (nitrification amp

other)

Corrosion Control Treatment is intertwined with all treatments affecting water chemistry

Water Quality Parameters(WQP)

bull pH alkalinity calcium orthophosphate or silica othersbull WQP measurements are important for

bull Process control (measurement frequency needs to be HIGH)bull Keeping treatment from being turned offbull Keeping inhibitor dosages amp pH control from being cut to save $$$bull Keeping required repairs to treatment on tight schedulebull Preventing frequent variations and alternations of water sources

bull Effectiveness of Pb treatment only determinable from tap sampling by combination of different assessment approaches for different lead plumbing sources to which people can be exposed

36

bull OWQP measurements important for process control look at yourdata to understand what they mean and avoid problems

bull OWQP do not predict lead release There are no substitutes fordirectly monitoring lead release

37

M+

M+

M+

M+

Corrosion barrier film hypothesis

Inert barrier film (eg CaCO3)

Pipe Wall

Barrier film would allow tobull Protect ALL pipe materialsbull Monitor corrosion based on one chemical reaction

Although barrier film would be convenient for important reasonsthe CaCO3 hypothesis has not been proven

38

In excavated pipe samples that the EPA analyzed

Inert barrier film (eg CaCO3) is a complex scale

+ M+M

M+

M+ Scale- Heterogeneous- Several layers- Amorphous amp crystalline

Pipe WallPb and non-Pb compounds

Although barrier film would be convenient for important reasonsthe CaCO3 hypothesis has not been proven and scale is overall

Are these ldquocorrosionrdquo indicesIndex Equation Range Water Condition

Langelier Saturation Index

LSI = pHactual - pHs LSIgt0

LSIlt0

bull Supersaturated tends to precipitate CaCO3

bull Undersaturated tends to dissolve CaCO3

Ryznar Stability Index RSI = 2pHs-pHactual RSIgt6

RSIlt6

bull Undersaturated tends to dissolve CaCO3

bull Supersaturated tends to precipitate CaCO3

Aggressive Index AI=pH+ log(AH) AIlt1010ltAIlt12AIgt12

bull Highly Aggressivebull Moderately Aggressivebull Non-aggressive

Larson Scold Index Ls=[Cl-+SO4-2] Alkalinity Lslt08

08ltLslt12

Lsgt12

bull Chloride amp Sulfate will not likely interfere with natural film formation

bull chloride amp sulfate may interfere with natural film formation

bull Tendency toward higher localized corrosion rates

39

bull Most indices were not developed or named for corrosion (eg saturation stability etc)bull The LSI and subsequent variations predict CaCO3 SCALING not corrosion or metal

releasebull Calcium carbonate was not the active inhibition mechanism in the hundreds lead and

copper pipes analyzed by EPA

ldquoCorrosionrdquo indices

bull When used outside their limitations these indices become unreliable

bull Different objectives for different materialsbull THERE IS NO UNIVERSAL ldquoINDEXrdquo FOR

MULTIPLE MATERIALS

40

Relying on the LSI as an indicator of a waterrsquos corrosivityshould be abandoned (AWWARF 1996 Internal Corrosion ofDistribution Systems)

Summary

41

bull Materials inventory is critical to know where and what lead sourcesstill exist

bull To understand corrosion and lead release it is important to keep upwith pH the carbonate system (ie dissolved inorganic carbon andalkalinity) and disinfectants (ie oxidation-reduction potential)

bull Corrosion inhibitor efficiency depends on type and sufficientdoseconcentration for the background pH and other reactive waterquality constituents

bull There are many types of scale on lead pipe with different solubilitiesPb(+4) solubility ltlt Pb(+2)

bull Optimum Water Quality Parameters (OWQP) important for processcontrol look at your data to understand what they mean and avoidproblems

bull Setting OWQP or calculating ldquocorrosionrdquo indices does not predictlead release

How we communicate our research (recent examples)

Peer-review journal papersbull Doreacute E Lytle DA Wasserstrom L Swertfeger J Triantafyllidou S Field Analyzers for

Lead Quantification in Drinking Water Samples Critical Reviews in Environmental Science and Technology 2020 httpsdoiorg1010801064338920201782654

bull Lytle D A Schock M R Wait K Cahalan K Bosscher V Porter A Del Toral M Sequential Drinking Water Sampling as a Tool for Evaluating Lead in Flint Michigan Water Res 2019 157 40-54

bull Tully J DeSantis M K Schock M R Water QualityndashPipe Deposit Relationships in Midwestern Lead Pipes AWWA Water Science 2019 1 (2) e1127

bull DeSantis M K Triantafyllidou S Schock M R Lytle D A Mineralogical Evidence of Galvanic Corrosion in Drinking Water Lead Pipe Joints Environ Sci Technol 2018 52 (6) 3365-3374

Conference presentationsbull Triantafyllidou S Hensley K Lytle D James R Lal V Tools to Identify Lead Service

Lines AWWA Water Quality Technology Conference Dallas 2019bull Schock M Triantafyllidou S Tully J DeSantis M and Lytle D Diagnostic Sampling

Tools for Lead in Drinking Water AWWA Annual Conference and Exposition Denver 2019bull Tully J DeSantis M and Schock M Actual vs Predicted Lead Scale Observations from

the Field AWWA WQTC Toronto 2018

How we communicate our research (recent examples)

Magazine Articlesbull Lytle D A Schock M R Triantafyllidou S Identify Lead Plumbing Sources to Protect

Public Health AWWA Opflow 2018 44 (3) 16-20EPA Science Matters Newslettersbull Revealing the Complicated Nature of Tap Water Lead Contamination A Madison Wisconsin

Case Study July 30 2018bull Identifying the Best Lead Sampling Techniques to Protect Public Health October 22 2018bull Scaling Back EPA Researchers Help Communities Protect Drinking Water Systems from Lead

April 8 2019bull Search httpswwwepagovsciencemattersFact Sheets on Leadbull httpswwwepagovwater-researchwater-research-fact-sheetsDWHow to Identify Lead Free Certification Marks for Drinking Water System and Plumbing ProductsConsumer Tool for Identifying POU Drinking Water Filters Certified to Reduce LeadStandard Developmentbull Participate on the workgroups for NSF Drinking Water Treatment Units (point-of-use filters)

NSF 6061 (Health Effects for Drinking Water Treatment Chemicals and Health Effects for Drinking Water System Components) and ASHRAE Standard 514 (Building water quality)

33

Manganese

Other studiesbull Dong et al 2000 Mn acted as vector for particulate Pb movementbull Trueman et al 2019 Increasing Mn increased Pb in experimentbull McNeill and Edwards 2004 Mn sequestration by polyphosphate destabilized

Pb scales and caused Pb release

bull Pan et al 2019 Possible benefit by promoting formation of Pb(+4)

Reference Analysis Mn Impact on PbSchock et al 2014 Excavated LSLs

amp historical water Pb data in a US city

bull Promoted amorphous Pb scalebull Inhibited insoluble Pb(+4) scalebull Erratic Pb persisted after full LSLR in areas rich

with Mn deposits

Tully et al 2019 Excavated LSLs amp water Pb data in several Midwest case studies

bull Mn was a component of amorphous pipe scales in several systems

bull Amorphous scales not amenable to determining their adherence to pipe walls their responses to changes in various water chemistry variables or quantitatively calculating their lead release

Itrsquos not just manganese

34

Broad variety of coatings on LSLs analyzed at EPA

Al Mn Fe P and Ca-rich coatings may interfere withorthophosphate pH adjustment or other CCT

35

More water parameters affect metal release

bull pH amp AlkalinityDICbull Concentration and type of

oxidants (disinfectantsdissolved oxygen)

bull ORPcorrosion potentialbull Corrosion Inhibitorsbull Chloridebull Sulfatebull Manganese

bull Temperaturebull Sorptive surfaces downstream of

LSLs (ie galvanized interior pipe)bull Iron (deposition and corrosion)bull Calciumbull Aluminumbull NOM (type amount)bull Mixing from water treatment plants

or sourcesbull Ammoniabull Hydrogen Sulfidebull Microbial activity (nitrification amp

other)

Corrosion Control Treatment is intertwined with all treatments affecting water chemistry

Water Quality Parameters(WQP)

bull pH alkalinity calcium orthophosphate or silica othersbull WQP measurements are important for

bull Process control (measurement frequency needs to be HIGH)bull Keeping treatment from being turned offbull Keeping inhibitor dosages amp pH control from being cut to save $$$bull Keeping required repairs to treatment on tight schedulebull Preventing frequent variations and alternations of water sources

bull Effectiveness of Pb treatment only determinable from tap sampling by combination of different assessment approaches for different lead plumbing sources to which people can be exposed

36

bull OWQP measurements important for process control look at yourdata to understand what they mean and avoid problems

bull OWQP do not predict lead release There are no substitutes fordirectly monitoring lead release

37

M+

M+

M+

M+

Corrosion barrier film hypothesis

Inert barrier film (eg CaCO3)

Pipe Wall

Barrier film would allow tobull Protect ALL pipe materialsbull Monitor corrosion based on one chemical reaction

Although barrier film would be convenient for important reasonsthe CaCO3 hypothesis has not been proven

38

In excavated pipe samples that the EPA analyzed

Inert barrier film (eg CaCO3) is a complex scale

+ M+M

M+

M+ Scale- Heterogeneous- Several layers- Amorphous amp crystalline

Pipe WallPb and non-Pb compounds

Although barrier film would be convenient for important reasonsthe CaCO3 hypothesis has not been proven and scale is overall

Are these ldquocorrosionrdquo indicesIndex Equation Range Water Condition

Langelier Saturation Index

LSI = pHactual - pHs LSIgt0

LSIlt0

bull Supersaturated tends to precipitate CaCO3

bull Undersaturated tends to dissolve CaCO3

Ryznar Stability Index RSI = 2pHs-pHactual RSIgt6

RSIlt6

bull Undersaturated tends to dissolve CaCO3

bull Supersaturated tends to precipitate CaCO3

Aggressive Index AI=pH+ log(AH) AIlt1010ltAIlt12AIgt12

bull Highly Aggressivebull Moderately Aggressivebull Non-aggressive

Larson Scold Index Ls=[Cl-+SO4-2] Alkalinity Lslt08

08ltLslt12

Lsgt12

bull Chloride amp Sulfate will not likely interfere with natural film formation

bull chloride amp sulfate may interfere with natural film formation

bull Tendency toward higher localized corrosion rates

39

bull Most indices were not developed or named for corrosion (eg saturation stability etc)bull The LSI and subsequent variations predict CaCO3 SCALING not corrosion or metal

releasebull Calcium carbonate was not the active inhibition mechanism in the hundreds lead and

copper pipes analyzed by EPA

ldquoCorrosionrdquo indices

bull When used outside their limitations these indices become unreliable

bull Different objectives for different materialsbull THERE IS NO UNIVERSAL ldquoINDEXrdquo FOR

MULTIPLE MATERIALS

40

Relying on the LSI as an indicator of a waterrsquos corrosivityshould be abandoned (AWWARF 1996 Internal Corrosion ofDistribution Systems)

Summary

41

bull Materials inventory is critical to know where and what lead sourcesstill exist

bull To understand corrosion and lead release it is important to keep upwith pH the carbonate system (ie dissolved inorganic carbon andalkalinity) and disinfectants (ie oxidation-reduction potential)

bull Corrosion inhibitor efficiency depends on type and sufficientdoseconcentration for the background pH and other reactive waterquality constituents

bull There are many types of scale on lead pipe with different solubilitiesPb(+4) solubility ltlt Pb(+2)

bull Optimum Water Quality Parameters (OWQP) important for processcontrol look at your data to understand what they mean and avoidproblems

bull Setting OWQP or calculating ldquocorrosionrdquo indices does not predictlead release

How we communicate our research (recent examples)

Peer-review journal papersbull Doreacute E Lytle DA Wasserstrom L Swertfeger J Triantafyllidou S Field Analyzers for

Lead Quantification in Drinking Water Samples Critical Reviews in Environmental Science and Technology 2020 httpsdoiorg1010801064338920201782654

bull Lytle D A Schock M R Wait K Cahalan K Bosscher V Porter A Del Toral M Sequential Drinking Water Sampling as a Tool for Evaluating Lead in Flint Michigan Water Res 2019 157 40-54

bull Tully J DeSantis M K Schock M R Water QualityndashPipe Deposit Relationships in Midwestern Lead Pipes AWWA Water Science 2019 1 (2) e1127

bull DeSantis M K Triantafyllidou S Schock M R Lytle D A Mineralogical Evidence of Galvanic Corrosion in Drinking Water Lead Pipe Joints Environ Sci Technol 2018 52 (6) 3365-3374

Conference presentationsbull Triantafyllidou S Hensley K Lytle D James R Lal V Tools to Identify Lead Service

Lines AWWA Water Quality Technology Conference Dallas 2019bull Schock M Triantafyllidou S Tully J DeSantis M and Lytle D Diagnostic Sampling

Tools for Lead in Drinking Water AWWA Annual Conference and Exposition Denver 2019bull Tully J DeSantis M and Schock M Actual vs Predicted Lead Scale Observations from

the Field AWWA WQTC Toronto 2018

How we communicate our research (recent examples)

Magazine Articlesbull Lytle D A Schock M R Triantafyllidou S Identify Lead Plumbing Sources to Protect

Public Health AWWA Opflow 2018 44 (3) 16-20EPA Science Matters Newslettersbull Revealing the Complicated Nature of Tap Water Lead Contamination A Madison Wisconsin

Case Study July 30 2018bull Identifying the Best Lead Sampling Techniques to Protect Public Health October 22 2018bull Scaling Back EPA Researchers Help Communities Protect Drinking Water Systems from Lead

April 8 2019bull Search httpswwwepagovsciencemattersFact Sheets on Leadbull httpswwwepagovwater-researchwater-research-fact-sheetsDWHow to Identify Lead Free Certification Marks for Drinking Water System and Plumbing ProductsConsumer Tool for Identifying POU Drinking Water Filters Certified to Reduce LeadStandard Developmentbull Participate on the workgroups for NSF Drinking Water Treatment Units (point-of-use filters)

NSF 6061 (Health Effects for Drinking Water Treatment Chemicals and Health Effects for Drinking Water System Components) and ASHRAE Standard 514 (Building water quality)

Itrsquos not just manganese

34

Broad variety of coatings on LSLs analyzed at EPA

Al Mn Fe P and Ca-rich coatings may interfere withorthophosphate pH adjustment or other CCT

35

More water parameters affect metal release

bull pH amp AlkalinityDICbull Concentration and type of

oxidants (disinfectantsdissolved oxygen)

bull ORPcorrosion potentialbull Corrosion Inhibitorsbull Chloridebull Sulfatebull Manganese

bull Temperaturebull Sorptive surfaces downstream of

LSLs (ie galvanized interior pipe)bull Iron (deposition and corrosion)bull Calciumbull Aluminumbull NOM (type amount)bull Mixing from water treatment plants

or sourcesbull Ammoniabull Hydrogen Sulfidebull Microbial activity (nitrification amp

other)

Corrosion Control Treatment is intertwined with all treatments affecting water chemistry

Water Quality Parameters(WQP)

bull pH alkalinity calcium orthophosphate or silica othersbull WQP measurements are important for

bull Process control (measurement frequency needs to be HIGH)bull Keeping treatment from being turned offbull Keeping inhibitor dosages amp pH control from being cut to save $$$bull Keeping required repairs to treatment on tight schedulebull Preventing frequent variations and alternations of water sources

bull Effectiveness of Pb treatment only determinable from tap sampling by combination of different assessment approaches for different lead plumbing sources to which people can be exposed

36

bull OWQP measurements important for process control look at yourdata to understand what they mean and avoid problems

bull OWQP do not predict lead release There are no substitutes fordirectly monitoring lead release

37

M+

M+

M+

M+

Corrosion barrier film hypothesis

Inert barrier film (eg CaCO3)

Pipe Wall

Barrier film would allow tobull Protect ALL pipe materialsbull Monitor corrosion based on one chemical reaction

Although barrier film would be convenient for important reasonsthe CaCO3 hypothesis has not been proven

38

In excavated pipe samples that the EPA analyzed

Inert barrier film (eg CaCO3) is a complex scale

+ M+M

M+

M+ Scale- Heterogeneous- Several layers- Amorphous amp crystalline

Pipe WallPb and non-Pb compounds

Although barrier film would be convenient for important reasonsthe CaCO3 hypothesis has not been proven and scale is overall

Are these ldquocorrosionrdquo indicesIndex Equation Range Water Condition

Langelier Saturation Index

LSI = pHactual - pHs LSIgt0

LSIlt0

bull Supersaturated tends to precipitate CaCO3

bull Undersaturated tends to dissolve CaCO3

Ryznar Stability Index RSI = 2pHs-pHactual RSIgt6

RSIlt6

bull Undersaturated tends to dissolve CaCO3

bull Supersaturated tends to precipitate CaCO3

Aggressive Index AI=pH+ log(AH) AIlt1010ltAIlt12AIgt12

bull Highly Aggressivebull Moderately Aggressivebull Non-aggressive

Larson Scold Index Ls=[Cl-+SO4-2] Alkalinity Lslt08

08ltLslt12

Lsgt12

bull Chloride amp Sulfate will not likely interfere with natural film formation

bull chloride amp sulfate may interfere with natural film formation

bull Tendency toward higher localized corrosion rates

39

bull Most indices were not developed or named for corrosion (eg saturation stability etc)bull The LSI and subsequent variations predict CaCO3 SCALING not corrosion or metal

releasebull Calcium carbonate was not the active inhibition mechanism in the hundreds lead and

copper pipes analyzed by EPA

ldquoCorrosionrdquo indices

bull When used outside their limitations these indices become unreliable

bull Different objectives for different materialsbull THERE IS NO UNIVERSAL ldquoINDEXrdquo FOR

MULTIPLE MATERIALS

40

Relying on the LSI as an indicator of a waterrsquos corrosivityshould be abandoned (AWWARF 1996 Internal Corrosion ofDistribution Systems)

Summary

41

bull Materials inventory is critical to know where and what lead sourcesstill exist

bull To understand corrosion and lead release it is important to keep upwith pH the carbonate system (ie dissolved inorganic carbon andalkalinity) and disinfectants (ie oxidation-reduction potential)

bull Corrosion inhibitor efficiency depends on type and sufficientdoseconcentration for the background pH and other reactive waterquality constituents

bull There are many types of scale on lead pipe with different solubilitiesPb(+4) solubility ltlt Pb(+2)

bull Optimum Water Quality Parameters (OWQP) important for processcontrol look at your data to understand what they mean and avoidproblems

bull Setting OWQP or calculating ldquocorrosionrdquo indices does not predictlead release

How we communicate our research (recent examples)

Peer-review journal papersbull Doreacute E Lytle DA Wasserstrom L Swertfeger J Triantafyllidou S Field Analyzers for

Lead Quantification in Drinking Water Samples Critical Reviews in Environmental Science and Technology 2020 httpsdoiorg1010801064338920201782654

bull Lytle D A Schock M R Wait K Cahalan K Bosscher V Porter A Del Toral M Sequential Drinking Water Sampling as a Tool for Evaluating Lead in Flint Michigan Water Res 2019 157 40-54

bull Tully J DeSantis M K Schock M R Water QualityndashPipe Deposit Relationships in Midwestern Lead Pipes AWWA Water Science 2019 1 (2) e1127

bull DeSantis M K Triantafyllidou S Schock M R Lytle D A Mineralogical Evidence of Galvanic Corrosion in Drinking Water Lead Pipe Joints Environ Sci Technol 2018 52 (6) 3365-3374

Conference presentationsbull Triantafyllidou S Hensley K Lytle D James R Lal V Tools to Identify Lead Service

Lines AWWA Water Quality Technology Conference Dallas 2019bull Schock M Triantafyllidou S Tully J DeSantis M and Lytle D Diagnostic Sampling

Tools for Lead in Drinking Water AWWA Annual Conference and Exposition Denver 2019bull Tully J DeSantis M and Schock M Actual vs Predicted Lead Scale Observations from

the Field AWWA WQTC Toronto 2018

How we communicate our research (recent examples)

Magazine Articlesbull Lytle D A Schock M R Triantafyllidou S Identify Lead Plumbing Sources to Protect

Public Health AWWA Opflow 2018 44 (3) 16-20EPA Science Matters Newslettersbull Revealing the Complicated Nature of Tap Water Lead Contamination A Madison Wisconsin

Case Study July 30 2018bull Identifying the Best Lead Sampling Techniques to Protect Public Health October 22 2018bull Scaling Back EPA Researchers Help Communities Protect Drinking Water Systems from Lead

April 8 2019bull Search httpswwwepagovsciencemattersFact Sheets on Leadbull httpswwwepagovwater-researchwater-research-fact-sheetsDWHow to Identify Lead Free Certification Marks for Drinking Water System and Plumbing ProductsConsumer Tool for Identifying POU Drinking Water Filters Certified to Reduce LeadStandard Developmentbull Participate on the workgroups for NSF Drinking Water Treatment Units (point-of-use filters)

NSF 6061 (Health Effects for Drinking Water Treatment Chemicals and Health Effects for Drinking Water System Components) and ASHRAE Standard 514 (Building water quality)

35

More water parameters affect metal release

bull pH amp AlkalinityDICbull Concentration and type of

oxidants (disinfectantsdissolved oxygen)

bull ORPcorrosion potentialbull Corrosion Inhibitorsbull Chloridebull Sulfatebull Manganese

bull Temperaturebull Sorptive surfaces downstream of

LSLs (ie galvanized interior pipe)bull Iron (deposition and corrosion)bull Calciumbull Aluminumbull NOM (type amount)bull Mixing from water treatment plants

or sourcesbull Ammoniabull Hydrogen Sulfidebull Microbial activity (nitrification amp

other)

Corrosion Control Treatment is intertwined with all treatments affecting water chemistry

Water Quality Parameters(WQP)

bull pH alkalinity calcium orthophosphate or silica othersbull WQP measurements are important for

bull Process control (measurement frequency needs to be HIGH)bull Keeping treatment from being turned offbull Keeping inhibitor dosages amp pH control from being cut to save $$$bull Keeping required repairs to treatment on tight schedulebull Preventing frequent variations and alternations of water sources

bull Effectiveness of Pb treatment only determinable from tap sampling by combination of different assessment approaches for different lead plumbing sources to which people can be exposed

36

bull OWQP measurements important for process control look at yourdata to understand what they mean and avoid problems

bull OWQP do not predict lead release There are no substitutes fordirectly monitoring lead release

37

M+

M+

M+

M+

Corrosion barrier film hypothesis

Inert barrier film (eg CaCO3)

Pipe Wall

Barrier film would allow tobull Protect ALL pipe materialsbull Monitor corrosion based on one chemical reaction

Although barrier film would be convenient for important reasonsthe CaCO3 hypothesis has not been proven

38

In excavated pipe samples that the EPA analyzed

Inert barrier film (eg CaCO3) is a complex scale

+ M+M

M+

M+ Scale- Heterogeneous- Several layers- Amorphous amp crystalline

Pipe WallPb and non-Pb compounds

Although barrier film would be convenient for important reasonsthe CaCO3 hypothesis has not been proven and scale is overall

Are these ldquocorrosionrdquo indicesIndex Equation Range Water Condition

Langelier Saturation Index

LSI = pHactual - pHs LSIgt0

LSIlt0

bull Supersaturated tends to precipitate CaCO3

bull Undersaturated tends to dissolve CaCO3

Ryznar Stability Index RSI = 2pHs-pHactual RSIgt6

RSIlt6

bull Undersaturated tends to dissolve CaCO3

bull Supersaturated tends to precipitate CaCO3

Aggressive Index AI=pH+ log(AH) AIlt1010ltAIlt12AIgt12

bull Highly Aggressivebull Moderately Aggressivebull Non-aggressive

Larson Scold Index Ls=[Cl-+SO4-2] Alkalinity Lslt08

08ltLslt12

Lsgt12

bull Chloride amp Sulfate will not likely interfere with natural film formation

bull chloride amp sulfate may interfere with natural film formation

bull Tendency toward higher localized corrosion rates

39

bull Most indices were not developed or named for corrosion (eg saturation stability etc)bull The LSI and subsequent variations predict CaCO3 SCALING not corrosion or metal

releasebull Calcium carbonate was not the active inhibition mechanism in the hundreds lead and

copper pipes analyzed by EPA

ldquoCorrosionrdquo indices

bull When used outside their limitations these indices become unreliable

bull Different objectives for different materialsbull THERE IS NO UNIVERSAL ldquoINDEXrdquo FOR

MULTIPLE MATERIALS

40

Relying on the LSI as an indicator of a waterrsquos corrosivityshould be abandoned (AWWARF 1996 Internal Corrosion ofDistribution Systems)

Summary

41

bull Materials inventory is critical to know where and what lead sourcesstill exist

bull To understand corrosion and lead release it is important to keep upwith pH the carbonate system (ie dissolved inorganic carbon andalkalinity) and disinfectants (ie oxidation-reduction potential)

bull Corrosion inhibitor efficiency depends on type and sufficientdoseconcentration for the background pH and other reactive waterquality constituents

bull There are many types of scale on lead pipe with different solubilitiesPb(+4) solubility ltlt Pb(+2)

bull Optimum Water Quality Parameters (OWQP) important for processcontrol look at your data to understand what they mean and avoidproblems

bull Setting OWQP or calculating ldquocorrosionrdquo indices does not predictlead release

How we communicate our research (recent examples)

Peer-review journal papersbull Doreacute E Lytle DA Wasserstrom L Swertfeger J Triantafyllidou S Field Analyzers for

Lead Quantification in Drinking Water Samples Critical Reviews in Environmental Science and Technology 2020 httpsdoiorg1010801064338920201782654

bull Lytle D A Schock M R Wait K Cahalan K Bosscher V Porter A Del Toral M Sequential Drinking Water Sampling as a Tool for Evaluating Lead in Flint Michigan Water Res 2019 157 40-54

bull Tully J DeSantis M K Schock M R Water QualityndashPipe Deposit Relationships in Midwestern Lead Pipes AWWA Water Science 2019 1 (2) e1127

bull DeSantis M K Triantafyllidou S Schock M R Lytle D A Mineralogical Evidence of Galvanic Corrosion in Drinking Water Lead Pipe Joints Environ Sci Technol 2018 52 (6) 3365-3374

Conference presentationsbull Triantafyllidou S Hensley K Lytle D James R Lal V Tools to Identify Lead Service

Lines AWWA Water Quality Technology Conference Dallas 2019bull Schock M Triantafyllidou S Tully J DeSantis M and Lytle D Diagnostic Sampling

Tools for Lead in Drinking Water AWWA Annual Conference and Exposition Denver 2019bull Tully J DeSantis M and Schock M Actual vs Predicted Lead Scale Observations from

the Field AWWA WQTC Toronto 2018

How we communicate our research (recent examples)

Magazine Articlesbull Lytle D A Schock M R Triantafyllidou S Identify Lead Plumbing Sources to Protect

Public Health AWWA Opflow 2018 44 (3) 16-20EPA Science Matters Newslettersbull Revealing the Complicated Nature of Tap Water Lead Contamination A Madison Wisconsin

Case Study July 30 2018bull Identifying the Best Lead Sampling Techniques to Protect Public Health October 22 2018bull Scaling Back EPA Researchers Help Communities Protect Drinking Water Systems from Lead

April 8 2019bull Search httpswwwepagovsciencemattersFact Sheets on Leadbull httpswwwepagovwater-researchwater-research-fact-sheetsDWHow to Identify Lead Free Certification Marks for Drinking Water System and Plumbing ProductsConsumer Tool for Identifying POU Drinking Water Filters Certified to Reduce LeadStandard Developmentbull Participate on the workgroups for NSF Drinking Water Treatment Units (point-of-use filters)

NSF 6061 (Health Effects for Drinking Water Treatment Chemicals and Health Effects for Drinking Water System Components) and ASHRAE Standard 514 (Building water quality)

Water Quality Parameters(WQP)

bull pH alkalinity calcium orthophosphate or silica othersbull WQP measurements are important for

bull Process control (measurement frequency needs to be HIGH)bull Keeping treatment from being turned offbull Keeping inhibitor dosages amp pH control from being cut to save $$$bull Keeping required repairs to treatment on tight schedulebull Preventing frequent variations and alternations of water sources

bull Effectiveness of Pb treatment only determinable from tap sampling by combination of different assessment approaches for different lead plumbing sources to which people can be exposed

36

bull OWQP measurements important for process control look at yourdata to understand what they mean and avoid problems

bull OWQP do not predict lead release There are no substitutes fordirectly monitoring lead release

37

M+

M+

M+

M+

Corrosion barrier film hypothesis

Inert barrier film (eg CaCO3)

Pipe Wall

Barrier film would allow tobull Protect ALL pipe materialsbull Monitor corrosion based on one chemical reaction

Although barrier film would be convenient for important reasonsthe CaCO3 hypothesis has not been proven

38

In excavated pipe samples that the EPA analyzed

Inert barrier film (eg CaCO3) is a complex scale

+ M+M

M+

M+ Scale- Heterogeneous- Several layers- Amorphous amp crystalline

Pipe WallPb and non-Pb compounds

Although barrier film would be convenient for important reasonsthe CaCO3 hypothesis has not been proven and scale is overall

Are these ldquocorrosionrdquo indicesIndex Equation Range Water Condition

Langelier Saturation Index

LSI = pHactual - pHs LSIgt0

LSIlt0

bull Supersaturated tends to precipitate CaCO3

bull Undersaturated tends to dissolve CaCO3

Ryznar Stability Index RSI = 2pHs-pHactual RSIgt6

RSIlt6

bull Undersaturated tends to dissolve CaCO3

bull Supersaturated tends to precipitate CaCO3

Aggressive Index AI=pH+ log(AH) AIlt1010ltAIlt12AIgt12

bull Highly Aggressivebull Moderately Aggressivebull Non-aggressive

Larson Scold Index Ls=[Cl-+SO4-2] Alkalinity Lslt08

08ltLslt12

Lsgt12

bull Chloride amp Sulfate will not likely interfere with natural film formation

bull chloride amp sulfate may interfere with natural film formation

bull Tendency toward higher localized corrosion rates

39

bull Most indices were not developed or named for corrosion (eg saturation stability etc)bull The LSI and subsequent variations predict CaCO3 SCALING not corrosion or metal

releasebull Calcium carbonate was not the active inhibition mechanism in the hundreds lead and

copper pipes analyzed by EPA

ldquoCorrosionrdquo indices

bull When used outside their limitations these indices become unreliable

bull Different objectives for different materialsbull THERE IS NO UNIVERSAL ldquoINDEXrdquo FOR

MULTIPLE MATERIALS

40

Relying on the LSI as an indicator of a waterrsquos corrosivityshould be abandoned (AWWARF 1996 Internal Corrosion ofDistribution Systems)

Summary

41

bull Materials inventory is critical to know where and what lead sourcesstill exist

bull To understand corrosion and lead release it is important to keep upwith pH the carbonate system (ie dissolved inorganic carbon andalkalinity) and disinfectants (ie oxidation-reduction potential)

bull Corrosion inhibitor efficiency depends on type and sufficientdoseconcentration for the background pH and other reactive waterquality constituents

bull There are many types of scale on lead pipe with different solubilitiesPb(+4) solubility ltlt Pb(+2)

bull Optimum Water Quality Parameters (OWQP) important for processcontrol look at your data to understand what they mean and avoidproblems

bull Setting OWQP or calculating ldquocorrosionrdquo indices does not predictlead release

How we communicate our research (recent examples)

Peer-review journal papersbull Doreacute E Lytle DA Wasserstrom L Swertfeger J Triantafyllidou S Field Analyzers for

Lead Quantification in Drinking Water Samples Critical Reviews in Environmental Science and Technology 2020 httpsdoiorg1010801064338920201782654

bull Lytle D A Schock M R Wait K Cahalan K Bosscher V Porter A Del Toral M Sequential Drinking Water Sampling as a Tool for Evaluating Lead in Flint Michigan Water Res 2019 157 40-54

bull Tully J DeSantis M K Schock M R Water QualityndashPipe Deposit Relationships in Midwestern Lead Pipes AWWA Water Science 2019 1 (2) e1127

bull DeSantis M K Triantafyllidou S Schock M R Lytle D A Mineralogical Evidence of Galvanic Corrosion in Drinking Water Lead Pipe Joints Environ Sci Technol 2018 52 (6) 3365-3374

Conference presentationsbull Triantafyllidou S Hensley K Lytle D James R Lal V Tools to Identify Lead Service

Lines AWWA Water Quality Technology Conference Dallas 2019bull Schock M Triantafyllidou S Tully J DeSantis M and Lytle D Diagnostic Sampling

Tools for Lead in Drinking Water AWWA Annual Conference and Exposition Denver 2019bull Tully J DeSantis M and Schock M Actual vs Predicted Lead Scale Observations from

the Field AWWA WQTC Toronto 2018

How we communicate our research (recent examples)

Magazine Articlesbull Lytle D A Schock M R Triantafyllidou S Identify Lead Plumbing Sources to Protect

Public Health AWWA Opflow 2018 44 (3) 16-20EPA Science Matters Newslettersbull Revealing the Complicated Nature of Tap Water Lead Contamination A Madison Wisconsin

Case Study July 30 2018bull Identifying the Best Lead Sampling Techniques to Protect Public Health October 22 2018bull Scaling Back EPA Researchers Help Communities Protect Drinking Water Systems from Lead

April 8 2019bull Search httpswwwepagovsciencemattersFact Sheets on Leadbull httpswwwepagovwater-researchwater-research-fact-sheetsDWHow to Identify Lead Free Certification Marks for Drinking Water System and Plumbing ProductsConsumer Tool for Identifying POU Drinking Water Filters Certified to Reduce LeadStandard Developmentbull Participate on the workgroups for NSF Drinking Water Treatment Units (point-of-use filters)

NSF 6061 (Health Effects for Drinking Water Treatment Chemicals and Health Effects for Drinking Water System Components) and ASHRAE Standard 514 (Building water quality)

37

M+

M+

M+

M+

Corrosion barrier film hypothesis

Inert barrier film (eg CaCO3)

Pipe Wall

Barrier film would allow tobull Protect ALL pipe materialsbull Monitor corrosion based on one chemical reaction

Although barrier film would be convenient for important reasonsthe CaCO3 hypothesis has not been proven

38

In excavated pipe samples that the EPA analyzed

Inert barrier film (eg CaCO3) is a complex scale

+ M+M

M+

M+ Scale- Heterogeneous- Several layers- Amorphous amp crystalline

Pipe WallPb and non-Pb compounds

Although barrier film would be convenient for important reasonsthe CaCO3 hypothesis has not been proven and scale is overall

Are these ldquocorrosionrdquo indicesIndex Equation Range Water Condition

Langelier Saturation Index

LSI = pHactual - pHs LSIgt0

LSIlt0

bull Supersaturated tends to precipitate CaCO3

bull Undersaturated tends to dissolve CaCO3

Ryznar Stability Index RSI = 2pHs-pHactual RSIgt6

RSIlt6

bull Undersaturated tends to dissolve CaCO3

bull Supersaturated tends to precipitate CaCO3

Aggressive Index AI=pH+ log(AH) AIlt1010ltAIlt12AIgt12

bull Highly Aggressivebull Moderately Aggressivebull Non-aggressive

Larson Scold Index Ls=[Cl-+SO4-2] Alkalinity Lslt08

08ltLslt12

Lsgt12

bull Chloride amp Sulfate will not likely interfere with natural film formation

bull chloride amp sulfate may interfere with natural film formation

bull Tendency toward higher localized corrosion rates

39

bull Most indices were not developed or named for corrosion (eg saturation stability etc)bull The LSI and subsequent variations predict CaCO3 SCALING not corrosion or metal

releasebull Calcium carbonate was not the active inhibition mechanism in the hundreds lead and

copper pipes analyzed by EPA

ldquoCorrosionrdquo indices

bull When used outside their limitations these indices become unreliable

bull Different objectives for different materialsbull THERE IS NO UNIVERSAL ldquoINDEXrdquo FOR

MULTIPLE MATERIALS

40

Relying on the LSI as an indicator of a waterrsquos corrosivityshould be abandoned (AWWARF 1996 Internal Corrosion ofDistribution Systems)

Summary

41

bull Materials inventory is critical to know where and what lead sourcesstill exist

bull To understand corrosion and lead release it is important to keep upwith pH the carbonate system (ie dissolved inorganic carbon andalkalinity) and disinfectants (ie oxidation-reduction potential)

bull Corrosion inhibitor efficiency depends on type and sufficientdoseconcentration for the background pH and other reactive waterquality constituents

bull There are many types of scale on lead pipe with different solubilitiesPb(+4) solubility ltlt Pb(+2)

bull Optimum Water Quality Parameters (OWQP) important for processcontrol look at your data to understand what they mean and avoidproblems

bull Setting OWQP or calculating ldquocorrosionrdquo indices does not predictlead release

How we communicate our research (recent examples)

Peer-review journal papersbull Doreacute E Lytle DA Wasserstrom L Swertfeger J Triantafyllidou S Field Analyzers for

Lead Quantification in Drinking Water Samples Critical Reviews in Environmental Science and Technology 2020 httpsdoiorg1010801064338920201782654

bull Lytle D A Schock M R Wait K Cahalan K Bosscher V Porter A Del Toral M Sequential Drinking Water Sampling as a Tool for Evaluating Lead in Flint Michigan Water Res 2019 157 40-54

bull Tully J DeSantis M K Schock M R Water QualityndashPipe Deposit Relationships in Midwestern Lead Pipes AWWA Water Science 2019 1 (2) e1127

bull DeSantis M K Triantafyllidou S Schock M R Lytle D A Mineralogical Evidence of Galvanic Corrosion in Drinking Water Lead Pipe Joints Environ Sci Technol 2018 52 (6) 3365-3374

Conference presentationsbull Triantafyllidou S Hensley K Lytle D James R Lal V Tools to Identify Lead Service

Lines AWWA Water Quality Technology Conference Dallas 2019bull Schock M Triantafyllidou S Tully J DeSantis M and Lytle D Diagnostic Sampling

Tools for Lead in Drinking Water AWWA Annual Conference and Exposition Denver 2019bull Tully J DeSantis M and Schock M Actual vs Predicted Lead Scale Observations from

the Field AWWA WQTC Toronto 2018

How we communicate our research (recent examples)

Magazine Articlesbull Lytle D A Schock M R Triantafyllidou S Identify Lead Plumbing Sources to Protect

Public Health AWWA Opflow 2018 44 (3) 16-20EPA Science Matters Newslettersbull Revealing the Complicated Nature of Tap Water Lead Contamination A Madison Wisconsin

Case Study July 30 2018bull Identifying the Best Lead Sampling Techniques to Protect Public Health October 22 2018bull Scaling Back EPA Researchers Help Communities Protect Drinking Water Systems from Lead

April 8 2019bull Search httpswwwepagovsciencemattersFact Sheets on Leadbull httpswwwepagovwater-researchwater-research-fact-sheetsDWHow to Identify Lead Free Certification Marks for Drinking Water System and Plumbing ProductsConsumer Tool for Identifying POU Drinking Water Filters Certified to Reduce LeadStandard Developmentbull Participate on the workgroups for NSF Drinking Water Treatment Units (point-of-use filters)

NSF 6061 (Health Effects for Drinking Water Treatment Chemicals and Health Effects for Drinking Water System Components) and ASHRAE Standard 514 (Building water quality)

38

In excavated pipe samples that the EPA analyzed

Inert barrier film (eg CaCO3) is a complex scale

+ M+M

M+

M+ Scale- Heterogeneous- Several layers- Amorphous amp crystalline

Pipe WallPb and non-Pb compounds

Although barrier film would be convenient for important reasonsthe CaCO3 hypothesis has not been proven and scale is overall

Are these ldquocorrosionrdquo indicesIndex Equation Range Water Condition

Langelier Saturation Index

LSI = pHactual - pHs LSIgt0

LSIlt0

bull Supersaturated tends to precipitate CaCO3

bull Undersaturated tends to dissolve CaCO3

Ryznar Stability Index RSI = 2pHs-pHactual RSIgt6

RSIlt6

bull Undersaturated tends to dissolve CaCO3

bull Supersaturated tends to precipitate CaCO3

Aggressive Index AI=pH+ log(AH) AIlt1010ltAIlt12AIgt12

bull Highly Aggressivebull Moderately Aggressivebull Non-aggressive

Larson Scold Index Ls=[Cl-+SO4-2] Alkalinity Lslt08

08ltLslt12

Lsgt12

bull Chloride amp Sulfate will not likely interfere with natural film formation

bull chloride amp sulfate may interfere with natural film formation

bull Tendency toward higher localized corrosion rates

39

bull Most indices were not developed or named for corrosion (eg saturation stability etc)bull The LSI and subsequent variations predict CaCO3 SCALING not corrosion or metal

releasebull Calcium carbonate was not the active inhibition mechanism in the hundreds lead and

copper pipes analyzed by EPA

ldquoCorrosionrdquo indices

bull When used outside their limitations these indices become unreliable

bull Different objectives for different materialsbull THERE IS NO UNIVERSAL ldquoINDEXrdquo FOR

MULTIPLE MATERIALS

40

Relying on the LSI as an indicator of a waterrsquos corrosivityshould be abandoned (AWWARF 1996 Internal Corrosion ofDistribution Systems)

Summary

41

bull Materials inventory is critical to know where and what lead sourcesstill exist

bull To understand corrosion and lead release it is important to keep upwith pH the carbonate system (ie dissolved inorganic carbon andalkalinity) and disinfectants (ie oxidation-reduction potential)

bull Corrosion inhibitor efficiency depends on type and sufficientdoseconcentration for the background pH and other reactive waterquality constituents

bull There are many types of scale on lead pipe with different solubilitiesPb(+4) solubility ltlt Pb(+2)

bull Optimum Water Quality Parameters (OWQP) important for processcontrol look at your data to understand what they mean and avoidproblems

bull Setting OWQP or calculating ldquocorrosionrdquo indices does not predictlead release

How we communicate our research (recent examples)

Peer-review journal papersbull Doreacute E Lytle DA Wasserstrom L Swertfeger J Triantafyllidou S Field Analyzers for

Lead Quantification in Drinking Water Samples Critical Reviews in Environmental Science and Technology 2020 httpsdoiorg1010801064338920201782654

bull Lytle D A Schock M R Wait K Cahalan K Bosscher V Porter A Del Toral M Sequential Drinking Water Sampling as a Tool for Evaluating Lead in Flint Michigan Water Res 2019 157 40-54

bull Tully J DeSantis M K Schock M R Water QualityndashPipe Deposit Relationships in Midwestern Lead Pipes AWWA Water Science 2019 1 (2) e1127

bull DeSantis M K Triantafyllidou S Schock M R Lytle D A Mineralogical Evidence of Galvanic Corrosion in Drinking Water Lead Pipe Joints Environ Sci Technol 2018 52 (6) 3365-3374

Conference presentationsbull Triantafyllidou S Hensley K Lytle D James R Lal V Tools to Identify Lead Service

Lines AWWA Water Quality Technology Conference Dallas 2019bull Schock M Triantafyllidou S Tully J DeSantis M and Lytle D Diagnostic Sampling

Tools for Lead in Drinking Water AWWA Annual Conference and Exposition Denver 2019bull Tully J DeSantis M and Schock M Actual vs Predicted Lead Scale Observations from

the Field AWWA WQTC Toronto 2018

How we communicate our research (recent examples)

Magazine Articlesbull Lytle D A Schock M R Triantafyllidou S Identify Lead Plumbing Sources to Protect

Public Health AWWA Opflow 2018 44 (3) 16-20EPA Science Matters Newslettersbull Revealing the Complicated Nature of Tap Water Lead Contamination A Madison Wisconsin

Case Study July 30 2018bull Identifying the Best Lead Sampling Techniques to Protect Public Health October 22 2018bull Scaling Back EPA Researchers Help Communities Protect Drinking Water Systems from Lead

April 8 2019bull Search httpswwwepagovsciencemattersFact Sheets on Leadbull httpswwwepagovwater-researchwater-research-fact-sheetsDWHow to Identify Lead Free Certification Marks for Drinking Water System and Plumbing ProductsConsumer Tool for Identifying POU Drinking Water Filters Certified to Reduce LeadStandard Developmentbull Participate on the workgroups for NSF Drinking Water Treatment Units (point-of-use filters)

NSF 6061 (Health Effects for Drinking Water Treatment Chemicals and Health Effects for Drinking Water System Components) and ASHRAE Standard 514 (Building water quality)

Are these ldquocorrosionrdquo indicesIndex Equation Range Water Condition

Langelier Saturation Index

LSI = pHactual - pHs LSIgt0

LSIlt0

bull Supersaturated tends to precipitate CaCO3

bull Undersaturated tends to dissolve CaCO3

Ryznar Stability Index RSI = 2pHs-pHactual RSIgt6

RSIlt6

bull Undersaturated tends to dissolve CaCO3

bull Supersaturated tends to precipitate CaCO3

Aggressive Index AI=pH+ log(AH) AIlt1010ltAIlt12AIgt12

bull Highly Aggressivebull Moderately Aggressivebull Non-aggressive

Larson Scold Index Ls=[Cl-+SO4-2] Alkalinity Lslt08

08ltLslt12

Lsgt12

bull Chloride amp Sulfate will not likely interfere with natural film formation

bull chloride amp sulfate may interfere with natural film formation

bull Tendency toward higher localized corrosion rates

39

bull Most indices were not developed or named for corrosion (eg saturation stability etc)bull The LSI and subsequent variations predict CaCO3 SCALING not corrosion or metal

releasebull Calcium carbonate was not the active inhibition mechanism in the hundreds lead and

copper pipes analyzed by EPA

ldquoCorrosionrdquo indices

bull When used outside their limitations these indices become unreliable

bull Different objectives for different materialsbull THERE IS NO UNIVERSAL ldquoINDEXrdquo FOR

MULTIPLE MATERIALS

40

Relying on the LSI as an indicator of a waterrsquos corrosivityshould be abandoned (AWWARF 1996 Internal Corrosion ofDistribution Systems)

Summary

41

bull Materials inventory is critical to know where and what lead sourcesstill exist

bull To understand corrosion and lead release it is important to keep upwith pH the carbonate system (ie dissolved inorganic carbon andalkalinity) and disinfectants (ie oxidation-reduction potential)

bull Corrosion inhibitor efficiency depends on type and sufficientdoseconcentration for the background pH and other reactive waterquality constituents

bull There are many types of scale on lead pipe with different solubilitiesPb(+4) solubility ltlt Pb(+2)

bull Optimum Water Quality Parameters (OWQP) important for processcontrol look at your data to understand what they mean and avoidproblems

bull Setting OWQP or calculating ldquocorrosionrdquo indices does not predictlead release

How we communicate our research (recent examples)

Peer-review journal papersbull Doreacute E Lytle DA Wasserstrom L Swertfeger J Triantafyllidou S Field Analyzers for

Lead Quantification in Drinking Water Samples Critical Reviews in Environmental Science and Technology 2020 httpsdoiorg1010801064338920201782654

bull Lytle D A Schock M R Wait K Cahalan K Bosscher V Porter A Del Toral M Sequential Drinking Water Sampling as a Tool for Evaluating Lead in Flint Michigan Water Res 2019 157 40-54

bull Tully J DeSantis M K Schock M R Water QualityndashPipe Deposit Relationships in Midwestern Lead Pipes AWWA Water Science 2019 1 (2) e1127

bull DeSantis M K Triantafyllidou S Schock M R Lytle D A Mineralogical Evidence of Galvanic Corrosion in Drinking Water Lead Pipe Joints Environ Sci Technol 2018 52 (6) 3365-3374

Conference presentationsbull Triantafyllidou S Hensley K Lytle D James R Lal V Tools to Identify Lead Service

Lines AWWA Water Quality Technology Conference Dallas 2019bull Schock M Triantafyllidou S Tully J DeSantis M and Lytle D Diagnostic Sampling

Tools for Lead in Drinking Water AWWA Annual Conference and Exposition Denver 2019bull Tully J DeSantis M and Schock M Actual vs Predicted Lead Scale Observations from

the Field AWWA WQTC Toronto 2018

How we communicate our research (recent examples)

Magazine Articlesbull Lytle D A Schock M R Triantafyllidou S Identify Lead Plumbing Sources to Protect

Public Health AWWA Opflow 2018 44 (3) 16-20EPA Science Matters Newslettersbull Revealing the Complicated Nature of Tap Water Lead Contamination A Madison Wisconsin

Case Study July 30 2018bull Identifying the Best Lead Sampling Techniques to Protect Public Health October 22 2018bull Scaling Back EPA Researchers Help Communities Protect Drinking Water Systems from Lead

April 8 2019bull Search httpswwwepagovsciencemattersFact Sheets on Leadbull httpswwwepagovwater-researchwater-research-fact-sheetsDWHow to Identify Lead Free Certification Marks for Drinking Water System and Plumbing ProductsConsumer Tool for Identifying POU Drinking Water Filters Certified to Reduce LeadStandard Developmentbull Participate on the workgroups for NSF Drinking Water Treatment Units (point-of-use filters)

NSF 6061 (Health Effects for Drinking Water Treatment Chemicals and Health Effects for Drinking Water System Components) and ASHRAE Standard 514 (Building water quality)

ldquoCorrosionrdquo indices

bull When used outside their limitations these indices become unreliable

bull Different objectives for different materialsbull THERE IS NO UNIVERSAL ldquoINDEXrdquo FOR

MULTIPLE MATERIALS

40

Relying on the LSI as an indicator of a waterrsquos corrosivityshould be abandoned (AWWARF 1996 Internal Corrosion ofDistribution Systems)

Summary

41

bull Materials inventory is critical to know where and what lead sourcesstill exist

bull To understand corrosion and lead release it is important to keep upwith pH the carbonate system (ie dissolved inorganic carbon andalkalinity) and disinfectants (ie oxidation-reduction potential)

bull Corrosion inhibitor efficiency depends on type and sufficientdoseconcentration for the background pH and other reactive waterquality constituents

bull There are many types of scale on lead pipe with different solubilitiesPb(+4) solubility ltlt Pb(+2)

bull Optimum Water Quality Parameters (OWQP) important for processcontrol look at your data to understand what they mean and avoidproblems

bull Setting OWQP or calculating ldquocorrosionrdquo indices does not predictlead release

How we communicate our research (recent examples)

Peer-review journal papersbull Doreacute E Lytle DA Wasserstrom L Swertfeger J Triantafyllidou S Field Analyzers for

Lead Quantification in Drinking Water Samples Critical Reviews in Environmental Science and Technology 2020 httpsdoiorg1010801064338920201782654

bull Lytle D A Schock M R Wait K Cahalan K Bosscher V Porter A Del Toral M Sequential Drinking Water Sampling as a Tool for Evaluating Lead in Flint Michigan Water Res 2019 157 40-54

bull Tully J DeSantis M K Schock M R Water QualityndashPipe Deposit Relationships in Midwestern Lead Pipes AWWA Water Science 2019 1 (2) e1127

bull DeSantis M K Triantafyllidou S Schock M R Lytle D A Mineralogical Evidence of Galvanic Corrosion in Drinking Water Lead Pipe Joints Environ Sci Technol 2018 52 (6) 3365-3374

Conference presentationsbull Triantafyllidou S Hensley K Lytle D James R Lal V Tools to Identify Lead Service

Lines AWWA Water Quality Technology Conference Dallas 2019bull Schock M Triantafyllidou S Tully J DeSantis M and Lytle D Diagnostic Sampling

Tools for Lead in Drinking Water AWWA Annual Conference and Exposition Denver 2019bull Tully J DeSantis M and Schock M Actual vs Predicted Lead Scale Observations from

the Field AWWA WQTC Toronto 2018

How we communicate our research (recent examples)

Magazine Articlesbull Lytle D A Schock M R Triantafyllidou S Identify Lead Plumbing Sources to Protect

Public Health AWWA Opflow 2018 44 (3) 16-20EPA Science Matters Newslettersbull Revealing the Complicated Nature of Tap Water Lead Contamination A Madison Wisconsin

Case Study July 30 2018bull Identifying the Best Lead Sampling Techniques to Protect Public Health October 22 2018bull Scaling Back EPA Researchers Help Communities Protect Drinking Water Systems from Lead

April 8 2019bull Search httpswwwepagovsciencemattersFact Sheets on Leadbull httpswwwepagovwater-researchwater-research-fact-sheetsDWHow to Identify Lead Free Certification Marks for Drinking Water System and Plumbing ProductsConsumer Tool for Identifying POU Drinking Water Filters Certified to Reduce LeadStandard Developmentbull Participate on the workgroups for NSF Drinking Water Treatment Units (point-of-use filters)

NSF 6061 (Health Effects for Drinking Water Treatment Chemicals and Health Effects for Drinking Water System Components) and ASHRAE Standard 514 (Building water quality)

Summary

41

bull Materials inventory is critical to know where and what lead sourcesstill exist

bull To understand corrosion and lead release it is important to keep upwith pH the carbonate system (ie dissolved inorganic carbon andalkalinity) and disinfectants (ie oxidation-reduction potential)

bull Corrosion inhibitor efficiency depends on type and sufficientdoseconcentration for the background pH and other reactive waterquality constituents

bull There are many types of scale on lead pipe with different solubilitiesPb(+4) solubility ltlt Pb(+2)

bull Optimum Water Quality Parameters (OWQP) important for processcontrol look at your data to understand what they mean and avoidproblems

bull Setting OWQP or calculating ldquocorrosionrdquo indices does not predictlead release

How we communicate our research (recent examples)

Peer-review journal papersbull Doreacute E Lytle DA Wasserstrom L Swertfeger J Triantafyllidou S Field Analyzers for

Lead Quantification in Drinking Water Samples Critical Reviews in Environmental Science and Technology 2020 httpsdoiorg1010801064338920201782654

bull Lytle D A Schock M R Wait K Cahalan K Bosscher V Porter A Del Toral M Sequential Drinking Water Sampling as a Tool for Evaluating Lead in Flint Michigan Water Res 2019 157 40-54

bull Tully J DeSantis M K Schock M R Water QualityndashPipe Deposit Relationships in Midwestern Lead Pipes AWWA Water Science 2019 1 (2) e1127

bull DeSantis M K Triantafyllidou S Schock M R Lytle D A Mineralogical Evidence of Galvanic Corrosion in Drinking Water Lead Pipe Joints Environ Sci Technol 2018 52 (6) 3365-3374

Conference presentationsbull Triantafyllidou S Hensley K Lytle D James R Lal V Tools to Identify Lead Service

Lines AWWA Water Quality Technology Conference Dallas 2019bull Schock M Triantafyllidou S Tully J DeSantis M and Lytle D Diagnostic Sampling

Tools for Lead in Drinking Water AWWA Annual Conference and Exposition Denver 2019bull Tully J DeSantis M and Schock M Actual vs Predicted Lead Scale Observations from

the Field AWWA WQTC Toronto 2018

How we communicate our research (recent examples)

Magazine Articlesbull Lytle D A Schock M R Triantafyllidou S Identify Lead Plumbing Sources to Protect

Public Health AWWA Opflow 2018 44 (3) 16-20EPA Science Matters Newslettersbull Revealing the Complicated Nature of Tap Water Lead Contamination A Madison Wisconsin

Case Study July 30 2018bull Identifying the Best Lead Sampling Techniques to Protect Public Health October 22 2018bull Scaling Back EPA Researchers Help Communities Protect Drinking Water Systems from Lead

April 8 2019bull Search httpswwwepagovsciencemattersFact Sheets on Leadbull httpswwwepagovwater-researchwater-research-fact-sheetsDWHow to Identify Lead Free Certification Marks for Drinking Water System and Plumbing ProductsConsumer Tool for Identifying POU Drinking Water Filters Certified to Reduce LeadStandard Developmentbull Participate on the workgroups for NSF Drinking Water Treatment Units (point-of-use filters)

NSF 6061 (Health Effects for Drinking Water Treatment Chemicals and Health Effects for Drinking Water System Components) and ASHRAE Standard 514 (Building water quality)

How we communicate our research (recent examples)

Peer-review journal papersbull Doreacute E Lytle DA Wasserstrom L Swertfeger J Triantafyllidou S Field Analyzers for

Lead Quantification in Drinking Water Samples Critical Reviews in Environmental Science and Technology 2020 httpsdoiorg1010801064338920201782654

bull Lytle D A Schock M R Wait K Cahalan K Bosscher V Porter A Del Toral M Sequential Drinking Water Sampling as a Tool for Evaluating Lead in Flint Michigan Water Res 2019 157 40-54

bull Tully J DeSantis M K Schock M R Water QualityndashPipe Deposit Relationships in Midwestern Lead Pipes AWWA Water Science 2019 1 (2) e1127

bull DeSantis M K Triantafyllidou S Schock M R Lytle D A Mineralogical Evidence of Galvanic Corrosion in Drinking Water Lead Pipe Joints Environ Sci Technol 2018 52 (6) 3365-3374

Conference presentationsbull Triantafyllidou S Hensley K Lytle D James R Lal V Tools to Identify Lead Service

Lines AWWA Water Quality Technology Conference Dallas 2019bull Schock M Triantafyllidou S Tully J DeSantis M and Lytle D Diagnostic Sampling

Tools for Lead in Drinking Water AWWA Annual Conference and Exposition Denver 2019bull Tully J DeSantis M and Schock M Actual vs Predicted Lead Scale Observations from

the Field AWWA WQTC Toronto 2018

How we communicate our research (recent examples)

Magazine Articlesbull Lytle D A Schock M R Triantafyllidou S Identify Lead Plumbing Sources to Protect

Public Health AWWA Opflow 2018 44 (3) 16-20EPA Science Matters Newslettersbull Revealing the Complicated Nature of Tap Water Lead Contamination A Madison Wisconsin

Case Study July 30 2018bull Identifying the Best Lead Sampling Techniques to Protect Public Health October 22 2018bull Scaling Back EPA Researchers Help Communities Protect Drinking Water Systems from Lead

April 8 2019bull Search httpswwwepagovsciencemattersFact Sheets on Leadbull httpswwwepagovwater-researchwater-research-fact-sheetsDWHow to Identify Lead Free Certification Marks for Drinking Water System and Plumbing ProductsConsumer Tool for Identifying POU Drinking Water Filters Certified to Reduce LeadStandard Developmentbull Participate on the workgroups for NSF Drinking Water Treatment Units (point-of-use filters)

NSF 6061 (Health Effects for Drinking Water Treatment Chemicals and Health Effects for Drinking Water System Components) and ASHRAE Standard 514 (Building water quality)

How we communicate our research (recent examples)

Magazine Articlesbull Lytle D A Schock M R Triantafyllidou S Identify Lead Plumbing Sources to Protect

Public Health AWWA Opflow 2018 44 (3) 16-20EPA Science Matters Newslettersbull Revealing the Complicated Nature of Tap Water Lead Contamination A Madison Wisconsin

Case Study July 30 2018bull Identifying the Best Lead Sampling Techniques to Protect Public Health October 22 2018bull Scaling Back EPA Researchers Help Communities Protect Drinking Water Systems from Lead

April 8 2019bull Search httpswwwepagovsciencemattersFact Sheets on Leadbull httpswwwepagovwater-researchwater-research-fact-sheetsDWHow to Identify Lead Free Certification Marks for Drinking Water System and Plumbing ProductsConsumer Tool for Identifying POU Drinking Water Filters Certified to Reduce LeadStandard Developmentbull Participate on the workgroups for NSF Drinking Water Treatment Units (point-of-use filters)

NSF 6061 (Health Effects for Drinking Water Treatment Chemicals and Health Effects for Drinking Water System Components) and ASHRAE Standard 514 (Building water quality)

How we communicate our research (recent examples)

Workshopsbull ORD-SSWROWASDWA 17th Small Drinking Water Systems Annual Workshop in Cincinnati next

one September 1-3 2020 Information at httpswwwepagovwater-research17th-annual-epa-drinking-water-workshop

bull Contributions to the 16th Annual EPA Drinking Water Workshop Cincinnati OH September 24-26 2019

Presentation Pipe Scale Analysis for Solving Lead Problems DeSantis et alBreak-out sessions on Corrosion Premise Plumbing and Lead in Schools (3Ts)Training Session Setting up Bench and Pilot Scale Studies Cost -Effectively

Webinarsbull ORDOW Small Systems Monthly Webinar Series schedule and recordings at

httpswwwepagovwater-researchsmall-systems-monthly-webinar-seriesLead Management in Homes and Buildings DeSantis Tully and Latham March 26 2019Lead and Copper Sampling and Water Quality Challenges Viveiros and Lytle July 26 2016Corrosion Control for Drinking Water Systems Schock and DrsquoAmico June 30 2015Reduction of Lead in Drinking Water Schock and Latham December 15 2015

bull 2018 AWWA Corrosion 101 WebinarSchock M Lead Corrosion Control 101 New Understanding of Best PracticesLytle D Copper Corrosion 101 Principles and Guidance

DisclaimerThe information in this presentation has been reviewed andapproved for public dissemination in accordance with USEnvironmental Protection Agency (EPA) The views expressedin this presentation are those of the author(s) and do notnecessarily represent the views or policies of the AgencyAny mention of trade names or commercial products does notconstitute EPA endorsement or recommendation for use

45

Corrosion is Oxidation-Reduction

AWWARF 1990 Lead Control Strategies AWWA Research Foundation and AWWA Denver CO

Millions of anodecathode sites across interior fresh lead pipe surface

What is Buffer Intensity (β)

Buffer Intensity is a measure of the resistance of a water to change pH upon addition of [H+] or [OH-] at constant DIC concentration

β = (partCApartpH)DIC for added acid concentration CA

Alkalinity is a measure of the capacity of a water to neutralize [H+] to a fixed point normally [HCO3

-] = [H2CO3]

Difference between Alkalinity and BufferIntensity for Different pH

mg CaCO3LAlkalinity

20202020

pH mg CL Buffer IntensityDIC (β10000)

60 1483 62670 580 16080 486 02790 443 072

Computations for 25ordmC I=0005

• Slide Number 1
• Slide Number 2
• Slide Number 3
• Slide Number 4
• Slide Number 5
• Slide Number 6
• Slide Number 7
• Corrosion and scale formation
• Slide Number 9
• Slide Number 10
• Slide Number 11
• pH is master variable
• pH is master variable
• Slide Number 14
• Slide Number 15
• Slide Number 16
• ldquoClassicrdquo divalent Pb+2 solubility
• Lead ldquocorrosion Inhibitorsrdquo
• Slide Number 19
• Ortho-P Treatment for Pb+2
• Effect of pH and ortho-P on Pb release
• Ortho-P at pH 90 (DIC 6 mgL)
• Ortho-Ppoint of diminishing returns
• Sodium silicate
• Slide Number 25
• Slide Number 26
• Example profile of Pb(+4) (ie PbO2) scale house
• Chloride and sulfate
• Chloride and sulfate
• Chloride and sulfate
• Manganese
• Manganese
• Manganese
• Itrsquos not just manganese
• More water parameters affect metal release
• Water Quality Parameters (WQP)
• Corrosion barrier film hypothesis
• In excavated pipe samples that the EPA analyzed
• Are these ldquocorrosionrdquo indices
• ldquoCorrosionrdquo indices
• Summary
• Slide Number 42
• Slide Number 43
• Slide Number 44
• Slide Number 45
• Slide Number 46
• What is Buffer Intensity (b)
• Slide Number 48

Corrosion is Oxidation-Reduction

AWWARF 1990 Lead Control Strategies AWWA Research Foundation and AWWA Denver CO

Millions of anodecathode sites across interior fresh lead pipe surface

What is Buffer Intensity (β)

Buffer Intensity is a measure of the resistance of a water to change pH upon addition of [H+] or [OH-] at constant DIC concentration

β = (partCApartpH)DIC for added acid concentration CA

Alkalinity is a measure of the capacity of a water to neutralize [H+] to a fixed point normally [HCO3

-] = [H2CO3]

Difference between Alkalinity and BufferIntensity for Different pH

mg CaCO3LAlkalinity

20202020

pH mg CL Buffer IntensityDIC (β10000)

60 1483 62670 580 16080 486 02790 443 072

Computations for 25ordmC I=0005

• Slide Number 1
• Slide Number 2
• Slide Number 3
• Slide Number 4
• Slide Number 5
• Slide Number 6
• Slide Number 7
• Corrosion and scale formation
• Slide Number 9
• Slide Number 10
• Slide Number 11
• pH is master variable
• pH is master variable
• Slide Number 14
• Slide Number 15
• Slide Number 16
• ldquoClassicrdquo divalent Pb+2 solubility
• Lead ldquocorrosion Inhibitorsrdquo
• Slide Number 19
• Ortho-P Treatment for Pb+2
• Effect of pH and ortho-P on Pb release
• Ortho-P at pH 90 (DIC 6 mgL)
• Ortho-Ppoint of diminishing returns
• Sodium silicate
• Slide Number 25
• Slide Number 26
• Example profile of Pb(+4) (ie PbO2) scale house
• Chloride and sulfate
• Chloride and sulfate
• Chloride and sulfate
• Manganese
• Manganese
• Manganese
• Itrsquos not just manganese
• More water parameters affect metal release
• Water Quality Parameters (WQP)
• Corrosion barrier film hypothesis
• In excavated pipe samples that the EPA analyzed
• Are these ldquocorrosionrdquo indices
• ldquoCorrosionrdquo indices
• Summary
• Slide Number 42
• Slide Number 43
• Slide Number 44
• Slide Number 45
• Slide Number 46
• What is Buffer Intensity (b)
• Slide Number 48

What is Buffer Intensity (β)

Buffer Intensity is a measure of the resistance of a water to change pH upon addition of [H+] or [OH-] at constant DIC concentration

β = (partCApartpH)DIC for added acid concentration CA

Alkalinity is a measure of the capacity of a water to neutralize [H+] to a fixed point normally [HCO3

-] = [H2CO3]

Difference between Alkalinity and BufferIntensity for Different pH

mg CaCO3LAlkalinity

20202020

pH mg CL Buffer IntensityDIC (β10000)

60 1483 62670 580 16080 486 02790 443 072

Computations for 25ordmC I=0005

• Slide Number 1
• Slide Number 2
• Slide Number 3
• Slide Number 4
• Slide Number 5
• Slide Number 6
• Slide Number 7
• Corrosion and scale formation
• Slide Number 9
• Slide Number 10
• Slide Number 11
• pH is master variable
• pH is master variable
• Slide Number 14
• Slide Number 15
• Slide Number 16
• ldquoClassicrdquo divalent Pb+2 solubility
• Lead ldquocorrosion Inhibitorsrdquo
• Slide Number 19
• Ortho-P Treatment for Pb+2
• Effect of pH and ortho-P on Pb release
• Ortho-P at pH 90 (DIC 6 mgL)
• Ortho-Ppoint of diminishing returns
• Sodium silicate
• Slide Number 25
• Slide Number 26
• Example profile of Pb(+4) (ie PbO2) scale house
• Chloride and sulfate
• Chloride and sulfate
• Chloride and sulfate
• Manganese
• Manganese
• Manganese
• Itrsquos not just manganese
• More water parameters affect metal release
• Water Quality Parameters (WQP)
• Corrosion barrier film hypothesis
• In excavated pipe samples that the EPA analyzed
• Are these ldquocorrosionrdquo indices
• ldquoCorrosionrdquo indices
• Summary
• Slide Number 42
• Slide Number 43
• Slide Number 44
• Slide Number 45
• Slide Number 46
• What is Buffer Intensity (b)
• Slide Number 48

Difference between Alkalinity and BufferIntensity for Different pH

mg CaCO3LAlkalinity

20202020

pH mg CL Buffer IntensityDIC (β10000)

60 1483 62670 580 16080 486 02790 443 072

Computations for 25ordmC I=0005

• Slide Number 1
• Slide Number 2
• Slide Number 3
• Slide Number 4
• Slide Number 5
• Slide Number 6
• Slide Number 7
• Corrosion and scale formation
• Slide Number 9
• Slide Number 10
• Slide Number 11
• pH is master variable
• pH is master variable
• Slide Number 14
• Slide Number 15
• Slide Number 16
• ldquoClassicrdquo divalent Pb+2 solubility
• Lead ldquocorrosion Inhibitorsrdquo
• Slide Number 19
• Ortho-P Treatment for Pb+2
• Effect of pH and ortho-P on Pb release
• Ortho-P at pH 90 (DIC 6 mgL)
• Ortho-Ppoint of diminishing returns
• Sodium silicate
• Slide Number 25
• Slide Number 26
• Example profile of Pb(+4) (ie PbO2) scale house
• Chloride and sulfate
• Chloride and sulfate
• Chloride and sulfate
• Manganese
• Manganese
• Manganese
• Itrsquos not just manganese
• More water parameters affect metal release
• Water Quality Parameters (WQP)
• Corrosion barrier film hypothesis
• In excavated pipe samples that the EPA analyzed
• Are these ldquocorrosionrdquo indices
• ldquoCorrosionrdquo indices
• Summary
• Slide Number 42
• Slide Number 43
• Slide Number 44
• Slide Number 45
• Slide Number 46
• What is Buffer Intensity (b)
• Slide Number 48