can resistivity/conductivity measurements really capture

36
Michael Thomas University of New Brunswick Chloride Specifications in Concrete Construction: Part 1 Issues & Ideas Anna Maria Workshop XVIII: Chemical Attack on Concrete, November 2017 Can Resistivity/Conductivity Measurements Really Capture Long-Term Chloride Resistance?

Upload: others

Post on 14-Apr-2022

4 views

Category:

Documents


0 download

TRANSCRIPT

Page 1: Can Resistivity/Conductivity Measurements Really Capture

Michael ThomasUniversity of New Brunswick

Chloride Specifications in Concrete Construction: Part 1 Issues & IdeasAnna Maria Workshop XVIII: Chemical Attack on Concrete, November 2017

Can Resistivity/Conductivity Measurements Really

Capture Long-Term Chloride Resistance?

Page 2: Can Resistivity/Conductivity Measurements Really Capture

1,000,000

100,000

10,000

1,000

100

10

Electrical Resistivity, ρ

(Ω∙m

)

Concrete (oven‐dry)

Concrete (dry‐indoors)

Concrete (dry‐outdoors)

Concrete (saturated)

Range of values forsaturated concrete

10‐8 Copper

1014 Neoprene Rubber

0.001

0.01

0.1

1

10

100

Electrical Conductivity, σ(m

S/m)

ASTM C 1760 (2012)“Bulk Conductivity”

Adapted from Whiting & Nagi, 2003

If ρ = 182 Ω∙m (σ = 5.5 mS/m) the charge passed through a standard‐size specimen (φ3.75‐in. x 2‐in.) in the “Rapid Chloride Permeability Test” (ASTM C 1202) would be approximately 1000 Coulombs (6h at 60V)

Page 3: Can Resistivity/Conductivity Measurements Really Capture

1,000,000

100,000

10,000

1,000

100

10

Electrical Resistivity, ρ

(Ω∙m

)

Concrete (oven‐dry)

Concrete (dry‐indoors)

Concrete (dry‐outdoors)

Concrete (saturated)

Range of values forsaturated concrete

10‐8 Copper

1014 Neoprene Rubber

0.001

0.01

0.1

1

10

100

Electrical Conductivity, σ(m

S/m)

ASTM C 1760 (2012)“Bulk Conductivity”

Adapted from Whiting & Nagi, 2003

1500 coulombs flowing through a 95.25 dia x 50.8 mm disc in 6 hours is equivalent to: a resistivity of 121 ohm‐m, conductivity of 0.008 S/m, or a formation factor of 137/ρo (where ρo is the conductivity of the pore solution). Alternatives:

RCP ≤ 1500 C; ρ ≥ 120 Ω‐m, σ ≤ 8 mS/m, or FF ≥ 140/ρo

Limits on fc’ and w/cm are not required

Page 4: Can Resistivity/Conductivity Measurements Really Capture

100

1000

10000

100000

1 10 100 1000

RC

P (C

oulo

mbs

)

Resistivity (k-cm)

RCPT vs. Electrical Resistivity

For tests where Tfinal > 50°C (122°)RCP = 12 x 30‐minute value

Page 5: Can Resistivity/Conductivity Measurements Really Capture

Electrical resistance of saturated concrete is primarily dependent on:

Ω

Page 6: Can Resistivity/Conductivity Measurements Really Capture

Electrical resistance of saturated concrete is primarily dependent on:

Pore structure• Volume, size & connectivity of

pores

Page 7: Can Resistivity/Conductivity Measurements Really Capture

Na+K+

OH-

OH- OH-

OH-Na+

K+

Electrical resistance of saturated concrete is primarily dependent on:

Pore structure• Volume, size & connectivity of

pores

Composition of pore solution• Concentration of ions

More ions in solution – increased electrical conductivity – i.e. reduced electrical resistance

Page 8: Can Resistivity/Conductivity Measurements Really Capture

Na+K+

OH-

OH- OH-

OH-Na+

K+

Electrical resistance of saturated concrete is primarily dependent on:

Pore structure• Volume, size & connectivity of

pores

Composition of pore solution• Concentration of ions

Chloride resistance of saturated concrete is primarily dependent on:

Cl-

Cl-

Cl-

Cl-

Cl-

Page 9: Can Resistivity/Conductivity Measurements Really Capture

Na+K+

OH-

OH- OH-

OH-Na+

K+

Electrical resistance of saturated concrete is primarily dependent on:

Pore structure• Volume, size & connectivity of

pores

Composition of pore solution• Concentration of ions

Chloride resistance of saturated concrete is primarily dependent on:

Pore structure• Volume, size & connectivity of

pores Cl- Cl-Cl-

Cl- Cl-

Cl-

Cl-

Cl-

Page 10: Can Resistivity/Conductivity Measurements Really Capture

Na+K+

OH-

OH- OH-

OH-Na+

K+

Electrical resistance of saturated concrete is primarily dependent on:

Pore structure• Volume, size & connectivity of

pores

Composition of pore solution• Concentration of ions

Chloride resistance of saturated concrete is primarily dependent on:

Pore structure• Volume, size & connectivity of

pores

Composition of cement hydrates• Ability of hydrates to bind

chlorides

Cl-

Cl-Cl-

Cl-

Cl-

Cl-

Cl-

Cl-

Page 11: Can Resistivity/Conductivity Measurements Really Capture

Cl-

Cl-Cl-

Cl-

Cl-

Cl-

Cl-

Cl-

Na+K+

OH-

OH- OH-

OH-Na+

K+

Electrical resistance of saturated concrete is primarily dependent on:

Pore structure• Volume, size & connectivity of

pores

Composition of pore solution• Concentration of ions

Chloride resistance of saturated concrete is primarily dependent on:

Pore structure• Volume, size & connectivity of

pores

Composition of cement hydrates• Ability of hydrates to bind

chloridesBound Cl-

Page 12: Can Resistivity/Conductivity Measurements Really Capture

Na+K+

OH-

OH- OH-

OH-Na+

K+

Electrical resistance of saturated concrete is primarily dependent on:

Pore structure• Volume, size & connectivity of

pores

Composition of pore solution• Concentration of ions

Hydraulic conductivity is primarily dependent on:

Pore structure• Volume, size & connectivity of

pores

Page 13: Can Resistivity/Conductivity Measurements Really Capture

Chl

orid

e (%

)

Depth (mm)

CC

erfxD t

x

a01

2

Co and Da found by fitting the equation

shown to the measured profile.

Concrete sample is immersed in NaCl solution for time t (minimum 35 days)Sample then ground in approx. 1-mm depth incrementsDust samples analyzed for chlorides To produce chloride profile ↓

ASTM C 1556 Test to Determine the Bulk Diffusion Coefficient of Concrete

Values of Da typically in the range:1 x 10-13 to 1 x 10-11 m2/s

Page 14: Can Resistivity/Conductivity Measurements Really Capture

100

1000

10000

100000

1.E-13 1.E-12 1.E-11 1.E-10

RC

PT (C

oulo

mbs

)

Diffusion Coefficient (m2/s)

RCPT vs. Bulk Diffusion

Page 15: Can Resistivity/Conductivity Measurements Really Capture

100

1000

10000

100000

1.E-13 1.E-12 1.E-11 1.E-10

RC

PT (C

oulo

mbs

)

Diffusion Coefficient (m2/s)

RCPT vs. Bulk Diffusion

1000Coulombs

10-12 to 10-11

Page 16: Can Resistivity/Conductivity Measurements Really Capture

0

1000

2000

3000

4000

PC PC‐CAC‐C$ C$A‐C2S

Charged Pa

ssed

(Cou

lombs)

Binder

28‐day

91‐day

Chloride Resistance: Rapid-Set Cements

ASTM C 1202“Rapid Chloride Permeability”

PC‐CAC‐C$ is a ternary blend of:• Portland cement• Calcium‐aluminate cement• Gypsum

CSA‐C2S is a blend of:• Calcium‐alumino‐sulfate

(Klein’s compound)• Belite

Page 17: Can Resistivity/Conductivity Measurements Really Capture

0

1000

2000

3000

4000

PC PC‐CAC‐C$ C$A‐C2S

Charged Pa

ssed

(Cou

lombs)

Binder

28‐day

91‐day

ASTM C 1202“Rapid Chloride Permeability”

0.0

0.2

0.4

0.6

0.8

1.0

1.2

0 10 20 30 40

Chlorid

e Co

nten

t (%)

Depth (mm)

PC

PC‐CAC‐C$

C$A‐C2S

ASTM C 1567“Chloride Diffusion”

Chloride Resistance: Rapid-Set Cements

Page 18: Can Resistivity/Conductivity Measurements Really Capture

F = formation factorρ = resistivity of concreteρ0 = resistivity of concrete pore solutionD = chloride diffusion coefficientDi = self diffusion coefficient for ion (= 2.032 x 10-9 m2/s for Cl-)Φ = Porosityβ = Tortuosity of pore system

Formation Factor

Formation Factor is a fundamental material property

Courtesy Jason Weiss

Page 19: Can Resistivity/Conductivity Measurements Really Capture

Resistivity of the pore solution can be

(i) assumed (e.g. ρ0 = 0.1 Ω-m [σ0 = 10 S/m] for PC mixes),

(ii) estimated (e.g. “NIST model”) – mill certs and mix proportions

(iii) measured (by pore-solution extraction)

then:

1. F could be used to specify concrete

(correcting resistivity for pore solution effects)

2. F could be used to determine D from measurement of ρ

Formation Factor

Really?

Page 20: Can Resistivity/Conductivity Measurements Really Capture

0.0

0.5

1.0

1.5

2.0

0.00 0.05 0.10 0.15 0.20 0.25 0.30(Na2Oe x CaO)/(SiO2)2 of CM

OH

at 9

0 da

ys (M

ol/L

) .

Shehata, 2001 Unpublished Bleszynski, 2002 Ramlochan, 2000

R2 = 0.913

79 blends of:• Portland cement• Fly ash• Slag• Silica fume• Natural pozzolan

Cement Composition & Pore Solution Alkalinity

Results for cement pastes:

w/cm = 0.50

OH- by titration

Na+ & K+ by flame photometry, atomic absorption or ICP

OH- ~ Na+ + K+

Page 21: Can Resistivity/Conductivity Measurements Really Capture

0.0

0.5

1.0

1.5

2.0

0.00 0.05 0.10 0.15 0.20 0.25 0.30(Na2Oe x CaO)/(SiO2)2 of CM

OH

at 9

0 da

ys (M

ol/L

) .

Shehata, 2001 Unpublished Bleszynski, 2002 Ramlochan, 2000

R2 = 0.913

Cement Composition & Pore Solution Alkalinity

18.8 S/m

2.4 S/m

In most cases0.1 < [OH‐] < 1.0

31.4 S/m

Page 22: Can Resistivity/Conductivity Measurements Really Capture

https://ciks.cbt.nist.gov/poresolncalc.html

Page 23: Can Resistivity/Conductivity Measurements Really Capture

Comparison of Predicted vs Measured Values• Measurements made on pore solution pressed from hardened paste

samples at 90 days

• Predictions made using NIST ‐ https://ciks.cbt.nist.gov/poresolncalc.html

Page 24: Can Resistivity/Conductivity Measurements Really Capture

Comparison of Predicted vs Measured Values• Measurements made on pore solution pressed from hardened paste

samples at 90 days

• Predictions made using NIST ‐ https://ciks.cbt.nist.gov/poresolncalc.html

Page 25: Can Resistivity/Conductivity Measurements Really Capture

40 days

1 year

RCPTSF 1202MK 1342

RCPTSF 868MK 901

8% MK

8% SF

PC

Friedel’s Salt

Metakaolin vs. Silica Fume Bulk Diffusion Tests (C 1556)

Page 26: Can Resistivity/Conductivity Measurements Really Capture

MixBulk ρ

(KΩ.cm)RCPT

(Coulomb)

100PC 8.4 3050

25FA 23.2 1230

10SF 16.9 1630

50SG 29.7 905

Concrete Blocks (1 x 1 x 3 ft) at Treat Island in Tidal Zone for 25 years

w/cm = 0.40

Page 27: Can Resistivity/Conductivity Measurements Really Capture

No binding:

Linear isotherm:

Langmuirisotherm:

Freundlichisotherm:

Page 28: Can Resistivity/Conductivity Measurements Really Capture
Page 29: Can Resistivity/Conductivity Measurements Really Capture
Page 30: Can Resistivity/Conductivity Measurements Really Capture

100

1000

10000

10 100 1000Age (days)

RC

PT

(Cou

lom

bs)

100 PC8 SF25 FA4 SF & 20 FA

W/CM = 0.40

Time‐Dependent Changes in Electrical and Transport Properties

Page 31: Can Resistivity/Conductivity Measurements Really Capture

Diffu

sion Co

efficient (m

2 /s)

10‐10

10‐11

10‐12

10‐13

10‐14

OPC

Fly Ash

Lab ConcreteW/CM = 0.50

10 100 1000 10000 100000

Age (days)

Harmon G.S.W/CM ~ 0.58

15X reductionwith fly ash !

Bulk Diffusion (ASTM C 1556) Results for Lab & Field Concretes

M.D.A. Thomas, P.B. Bamforth/Cement and Concrete Research 29 (1999) 487–495

Page 32: Can Resistivity/Conductivity Measurements Really Capture

Taywood Blocksnr. Folkestone

Thomas & Bamforth, 1999

Page 33: Can Resistivity/Conductivity Measurements Really Capture

6 Months

0

0.1

0.2

0.3

0.4

0.5

0 20 40 60Depth (mm)

Chlo

ride

(%)

1 Year

0

0.1

0.2

0.3

0.4

0.5

0 20 40 60Depth (mm)

Chlo

ride

(%)

2 Years

0

0.1

0.2

0.3

0.4

0.5

0 20 40 60Depth (mm)

Chlo

ride

(%)

3 Years

0

0.1

0.2

0.3

0.4

0.5

0 20 40 60Depth (mm)

Chlo

ride

(%)

6 Years

0

0.1

0.2

0.3

0.4

0.5

0 20 40 60Depth (mm)

Chlo

ride

(%)

8 Years

0

0.1

0.2

0.3

0.4

0.5

0 20 40 60Depth (mm)

Chlo

ride

(%)

OPC70% Slag30% Fly Ash

Thomas & Bamforth, 1999

Page 34: Can Resistivity/Conductivity Measurements Really Capture

M.D.A. Thomas, P.B. Bamforth/Cement and Concrete Research 29 (1999) 487–495

D v t for Folkestone Blocks D v t for Lab & Field Concrete

m = 0.1

m = 0.7

m = 1.2

m = 0.12

m = 0.62

Page 35: Can Resistivity/Conductivity Measurements Really Capture

Conclusions

• Electrical resistivity/conductivity provides a good indication of the mass‐transport properties of Portland‐cement concrete

• The relationship between electrical and mass‐transport properties of concrete is strongly influenced by a number of parameters including pore‐solution conductivity

• Predicting or measuring the pore‐solution conductivity is nottrivial

• Electrical properties do not account for the interaction between chlorides and the hydrates (binding) and this can be a verysignificant factor for some binders

• Early‐age measurements of either electrical or mass‐transport properties do not capture the substantial changes that may occur with high replacement levels of pozzolans or slag

Page 36: Can Resistivity/Conductivity Measurements Really Capture

Questions?