NCHRP 21-11 Update

Improved Test Methods & Practices

for Characterizing Steel Corrosion
Potential of Earthen Materials
PI : Ken Fishman, Earth Reinforcement Testing, Division of
McMahon & Mann Consulting Engineering and
Geology, P.C.
Co-PI : Soheil Nazarian, University or Texas, El Paso
2019 Midwest Geotechnical
Engineering Conference
Crown Plaza
Columbus, Ohio
September 18, 2019
1
 NCHRP 21-11 WORKPLAN
• PHASE I (Tasks 1-4) – Collect Existing Information
• Identify knowledge gaps
• Develop a detailed work plan to improve methods for sampling and testing
and characterization of corrosiveness of earthen materials.

• PHASE II (Tasks 5 & 6) – Implement Work Plan Developed in Phase I

• Study Laboratory and field tests for measurement of electrochemical
parameters, and characterizing steel corrosion
• Draft protocol for characterizing corrosiveness of earthen materials
• Formulate a detailed work plan to evaluate practical application of proposed
protocol
 NCHRP 21-11 WORKPLAN (Continued)
• Phase III (Tasks 7, 8 & 9) – Implement Work Plan
Developed in Phase II.

• Conduct trails in active construction projects

• Shadow specification to compare with current practice
• Demonstrate and evaluate recommendations and protocols
for sampling, testing and characterizing corrosiveness of
earthen materials.
• Initiate training with personnel from State DOTs
 QUESTIONS
• How do results obtained from different test methods compare?
• How fine does the material need to be before testing the fraction
passing the #10 sieve is appropriate?
• How can the test results be combined to characterize corrosion
potential?
• How well does the proposed characterization of corrosion potential
compare with performance?
• Is testing an aqueous extract (leachate) appropriate for coarse
materials?

4
 SUMMARY OF RESULTS
• Justify recommendations for draft protocol
• Comparisons between test results and identification of trends
• Comparisons between characterizations of corrosivity and
performance
• Identify alternatives for coarse open graded materials
• Identify needs for further study of unconventional materials

5
Site Information/Samples
 Composition, %

0%
20%
40%
60%
80%
100%
Florida

EL Paso MSE

Fine
M-U-D NY

South Carolina LWF

PIP @ 15ft

PIP @ 10ft
Gravel

South Carolina GB

PIP @ 5ft

Medium
Pharr TX

Lousiana LWF Crushed

Rochester NY

El Paso TX

Calagary AB
Coarse Sand

Prince George BC

Material Source
Ashdown AR

Temple TX

Sprain Brook NY

Raleigh NC

Garden City TX
Fine Sand

Coarse

Maple Rd NY
Material Composition

Wake Forest NC

Round Rock TX

Lousiana LWF Uncrushed

Waco TX
Fines

El Paso Coarse MSE

San Antonio
7

Bastrop
 CONSIDERATIONS FOR SELECTION OF TEST METHODS

1. Precision and repeatability of test methods.

2. Compatibility between parameters – e.g., salt
contents and resistivity
3. Correlations between resistivity
measurements, corrosivity and corrosion rates
4. Utility of test results

8
 Resistivity

Test Method Aggregate Size Preparation Box Size

Mix with 150 ml, let

AASHTO T288 (Standard) Passing #10 Small
stand for 12 hrs.

As is, remove
ASTM G187 particles greater Start as is Small
than ¼”

Small medium, or large

Saturate and wait
ASTM WK 2461 As is
24 hrs.
(depending on maximum
aggregate size)

TX-129-E (Small Box Method) Passing #8 Start dry Small

Small medium, or large

TX-129-M (Big Box Method) As is Start dry (depending on maximum
aggregate size)
 RESISTIVITY TESTS
TEST STANDARD SEPARATION MOISTURE CURE PERIOD BIAS w.r.t. PRECISION
CONDITIONS AASHTO T-288 µCOV (%)
AASHTO T-288 #10 increments 12 hours for 1st - 4.6
increment
TEX-129-E #8 increments None 1.07 3.2
ASTM G-187 ¼ inch As-is or None 1.41 5.3
saturated
TEX-129-M none increments, but
saturated for None 2.28 4.8
coarse materials
ASTM WK24621 none Drained from a Soak for 24 3.75 7.4
saturated hours before
condition draining

10
 PUBLISHED PRECISION & BIAS
FOR ASTM G-187
Soil #1 Soil #2 Soil #3
Average Resistivity (Ω-cm) 2296.95 450.10 19577.14
Repeatability Standard Deviation, s, (Ω-cm) 105.78 40.82 1194.95
Repeatability Coefficient of Variation, COV, % 4.6 9.1 6.1
Reproducibility Standard Deviation, s, (Ω-cm) 318.4 40.82 1721.30
Reproducibility Coefficient of Variation, COV, % 13.9 9.1 8.8
From ILS in Tampa Florida on November 18, 2003. Triplicate soil resistivity measurements by seven participants.

11
 AASHTO T-288 vs. Tex-129-E
20000.0
Tex-129-E (Ω-cm)

15000.0

y = 0.9614x
10000.0
R² = 0.9543
5000.0
AASHTO T-288 vs. Tex-129-M
40000.0

Tex-129-M (Ω-cm)
0.0 35000.0
0.0 5000.0 10000.0 15000.0 20000.0 30000.0
AASHTO T-288 (Ω−cm) 25000.0
20000.0
15000.0
AASHTO T-288 vs. ASTM G 187 10000.0
ASTM G187 (Ω-cm)

5000.0
20000.0
y = 1.1484x 0.0
15000.0 0.0 10000.0 20000.0 30000.0 40000.0
R² = 0.8835
AASHTO T-288 (Ω−cm)
10000.0

5000.0

0.0
0.0 5000.0 10000.0 15000.0 20000.0
AASHTO T-288 (Ω−cm)

12
 BIAS OF RESISTIVTY MEASUREMENTS AS A
FUNCTION OF TEXTURE
7

4
BIAS

0
Fine Sand Coarse Sand Gravel
MATERIALS
Tex-129-E ASTM G187 Tex-129-M ASTM WK24621

13
 50000.0
Effect of Texture on Resistivity
Florida
EL Paso MSE
45000.0
M-U-D NY
South Carolina LWF
40000.0
PIP @ 15ft
PIP @ 10ft
35000.0
RESISTIVITY (Ω-cm)

South Carolina GB
PIP @ 5'
30000.0
Pharr TX
Rochester NY
25000.0
El Paso TX
Calagary AB
20000.0
Raleigh NC
Prince George BC
15000.0
Temple TX
Sprain Brook NY
10000.0
Maple Rd NY
Wake Forest NC
5000.0
Round Rock TX
El Paso Coarse MSE
0.0
AASHTO T-288 Tex-129E G-187 Tex-129M ASTM WK-2461
TEST METHOD

The resistivity measurements vary with respect to coarseness of the sample, especially for the TX-
620-M, TX-129-M and ASTM WK2461 test methods. In general, for the same materials, the
measurements of resistivity vary with respect to test methods, however, results from AASHTO T 288
and TX-129-E method are similar for each material.
 TRENDING OF BIAS FROM TEX-129-M
9.00

8.00
Bias Tex-129-M/AASHTO T-288

7.00
Fine Sand
6.00
Coarse
5.00 Sand
Gravel
4.00

3.00

2.00

1.00

0.00
0.001 0.010 0.100 1.000
d2#10*nT288*100/(n129-M*(wg*d2g+wcs*d2cs+wfs*d2fs+wf*d2f))

15
GN &PP #10 CORRESPONDING TO BIAS TRENDS
Sample Bias < 1.5 1.5 < Bias < 3.0 Bias > 3
Types GN PP #10 GN PP #10 GN PP #10
Gravel - - 1.99 – 3.00 6- 40 3.00 – 3.56 24 - 40
Coarse 4.48- 4.83 60 - 70 3.85 - 4.48 50 - 60 - -
Sand
Fine Sand 5.00 - 6.65 > 80 - - - -

CONCLUSIONS
1. IF PP #10 > 60% then BIAS ≈ 1
2. IF GN > 3 and PP #10 < 40% then BIAS > 3
• corresponds to gravels with a coarse sand component ≈ 30%

16
 Tests on
Leachate
Test Aggregate Size Soil to Water Ratio Set up

Large plastic
SCDOT T143 Passing 1 ½” ~ 1:4
container with lid
Stirring plate and
TX-620-J Passing #40 1:10
stirring bar
Roller and 2-Liter
TX-620-M Passing 1 ¾” 1:10
plastic bottle
 MEASUREMENTS OF SULFATE CONTENT
TEST SEPARATION DILUTION HEATED MIXING FILTRATION PRECISION
STANDARD RATIO µCOV (%)
AASHTO T-290 #10 1:3 NO SHAKEN CENTRIFUGE 10.1
& FILTER
STIR EVERY
TEX-620-J #40 1:10 140°F HOUR FOR 12 FILTERED 11.8
HOURS
TEX-620-M NONE 1:10 NO MIXED FOR FILTERED 10.7
60 MINUTES

18
MEASUREMENTS OF CHLORIDE CONTENT
TEST SEPARATION DILUTION HEATED MIXING FILTRATION PRECISION
STANDARD RATIO µCOV (%)
SHAKE
AASHTO T-291 #10 1:3 NO FOR 20 SEC, CENTRIFUGE 7.5
STAND FOR 1 & FILTER
HOUR, SHAKE
STIR EVERY
TEX-620-J #40 1:10 140°F HOUR FOR 12 FILTERED 3.7
HOURS
TEX-620-M NONE 1:10 NO MIXED FOR FILTERED 12.9
60 MINUTES

19
CORRELATIONS BETWEEN SALT CONTENT MEASURMENTS
1000
900
Sulfate
Chloride
800
Linear (Sulfate)
Tex-620-M (mg/kg)

700 Linear (Chloride)

600
500
400 620-M (mg/kg) = 0.71(T-290)
300 R² = 0.79

200 620-M (mg/kg) = 0.51(T-291)

R² = 0.76
100
0
0 100 200 300 400 500 600 700 800 900 1000
AASHTO T-290 & 291 (mg/kg)

20
CORRELATION BETWEEN BIAS FROM SALT CONTENT MEASUREMENTS AND PP #10
1.40
620-Mcl +0.65*620-MSO4/(0.65*T290+T291)

>60%
1.20 CONCLUSIONS

1.00 1. PP #10 < 25%

= lowest bias
0.80
Bias = 0.06(PP#10)0.64
R² = 0.76 2. PP #10 > 60%
0.60 = bias > 1

0.40

0.20
<25%
0.00
0 10 20 30 40 50 60 70 80 90 100

PP#10 (%)

21
 OBSERVATIONS FROM MEASUREMENTS OF SALT
CONTENTS
• Precision/repeatability of test methods evaluated in
this study for measurements of salt contents are
similar.
• Bias statistics describing how salt contents measured
via other test methods compare to those measured
from the current the AASHTO tests are dependent
upon material characteristics including texture as
gravel, coarse sand, or fine sand.
22
 Equivalent Salt Content vs. Resistivity
with
AASHTO (PP #10 > 22%) or Texas Modified (PP#10 < 22%)
Wake Forest, NC
100000 Ocala, FL
South Carolina LWF
Prince George, BC
Ashdown, AR
Wake Round Rock, TX
M-U-D, NY
Raleigh, NC
RR
Resistivity (Ω-cm)

10000 South Carolina GB

Maple Road, NY
Pharr, TX
Sprain Brook NY
El Paso – MSE Coarse GC Garden City, TX
El Paso MSE Fine
El Paso MSE Coarse
El Paso, TX
1000
Rochester, NY
Calgary, AB
Temple, TX
PIP, NY @ 15 feet
Louisiana LWF Uncrushed
Bastrop, TX
Louisiana LWF Crushed
100 Waco, TX
1 10 100 1000 10000 Chloride

Equivalent Salt Content (mg/kg) minus 25 % error

plus 25% error
23
 CORRELATION BETWEEN TOTAL SALT CONTENT
and RESISTIVITY MEASUREMENTS
100000.00
100000.0
Resistivity Tex-620-M (Ω-cm)

Resistivity AASHTO T-288

y = 142105x-1.1
10000.00 R² = 0.88 10000.0

(Ω-cm)
1000.00 1000.0 y = 13488x-0.82
R² = 0.64

100.00 100.0
0.01 0.10 1.00 10.00 100.00 1.00 10.00 100.00
Total Salt Content Tex-620-M (mEq) Total Salt Content AASHTO (mEq)

Total Salt (mEq) = ppm ÷ equiv. atomic weight = Cl-(kg/mg) ÷ 35.5 + SO4 (kg/mg) ÷ 48. + HCO3 (kg/mg)÷ 61

24
 pH MEASURMENTS
TEST AIR SEPARATION DILUTION HEATED MIXING STAND PRECISION
STANDARD DRY RATIO TIME µCOV (%)
AASHTO T-289 Y #10 1:1 N Stir every 10-15 min NONE 1.0
for 1 hour
ASTM D 4972 Y #10 1:1 N Mix thoroughly 1 HR 1.1
NCHRP 21-06 N 3/8 IN 1:1 N Stirred 30 MIN 0.7
TEX-129-E Y #40 1:5 Y Stir every 15 min. for NONE 0.9
1 hour
TEX-620-M Y NONE 1:10 N Agitate for 1 hour NONE 1.2

25
 COMPARISON of pH MEASURMENTS RELATIVE TO AASHTO T-289

10.0 10.0

ASTM D 4972 (pH)

9.0 9.0

8.0 8.0
NCHRP 21-06 (pH)

7.0 7.0 D 4972 = 0.99(T-289)

21-06 = 0.98 (T-289)
6.0 R² = 0.70 6.0 R² = 0.69

5.0 5.0
5.0 6.0 7.0 8.0 9.0 10.0 5.0 6.0 7.0 8.0 9.0 10.0
AASHTO T-289 (pH) AASHTO T-289 (pH)

10.0 10.0

9.0 9.0

Tex-620-M (pH)
Tex-128-E (pH)

8.0 8.0

7.0 7.0
Tex-620-M = 1.08(T-289)
128-E = 1.03(T-289) R² = 0.33
6.0 R² = 0.63 6.0

5.0 5.0
5.0 6.0 7.0 8.0 9.0 10.0 5.0 6.0 7.0 8.0 9.0 10.0
AASHTO T-289 (pH) AASHTO T-289 (pH)
26
 DISTIBUTION of BIAS FROM pH MEASURMENTS
10

Frequency
6

0
1.03 1.05 1.07 1.10 1.12 More
Bias - Tex-620-M/AASHTO T-289

CONCLUSIONS
• Measurements of pH from Tex-620-M are less repeatable compared to measurements from
other test standards.
• In general, Tex-620-M renders pH that are higher compared to the pH values from the other
test standards included in the study.
• Results from NCHRP 21-06 are more repeatable compared to AASHTO T-289 and do not
have a significant bias with respect to results from AASHTO T-289.
27
 CORROSION RATES & RESISTIVITY MEASUREMENTS FROM
AASHTO T-288 WITH > 22% PASSING #10

40
45 Ocala, FL
35
Corrosion Rate (µm/yr)

40 South Carolina LWF

Measured CR (µm/yr)
35 Ashdown, AR 30
30 M-U-D, NY 25
25 Raleigh, NC
20 South Carolina GB
20
15 CR = 2877ρ-0.73 El Paso, TX 15
10 R² = 0.50 Rochester, NY
CR = 5267ρ-0.84
10 R² = 0.62
5 Sprain Brook NY
0 El Paso MSE Fine 5
0 2000 4000 6000 8000 10000 12000 14000 16000 18000 20000 PIP, NY @ 15 feet
0
Power (Trendline)
0 10000 20000 30000 40000 50000
Resistivity (Ω-cm)
Resistivity (Ω-cm)

Data from NCHRP 21-11

Worldwide Data

28
 PROPOSED PROTOCOL
1. IF GN > 3 or PP #10 > 25%

• AASHTO T-288, T-289, T-290 & T-291

2. IF GN < 3 and PP # 10 < 25%

• Tex-129-M, Tex-620-M
• ??? Considering using NCHRP 21-06 for pH ???

3. Consider alternatives for very coarse, open graded material

29
 Summary of Screening Techniques and
Characterizations
UNIVARIATE – Binary Systems
1. AASHTO (1992) - Galvanized Steel
2. PTI – Prestressing Steel (High Strength)
3. Burec (2009) Resistivity - 10th Percentile – DIP
and CIP
4. FHWA (2003) – Solid Bar Soil Nails – Carbon
Steel
5. European Standard – EN 12501-2 (2003)
MULTIVARIATE
1. German DVGW GW 9 –Pipelines
2. AWWA (DIP) – 10 Point Method
3. Jones (1985) – steel soil reinforcements
4. Clouterre (1993) – Soil Nails
5. Brady and McMahon (1994), UK – Galvanized
steel structures/Culverts
6. Beavers and Durr (1998), NACE (2001) – Steel
Piles
7. AGA (1983) – Hot-dipped Galvanized Steel
8. Demisse (2015) - Bayes Network - waterlines 30
 Parameter AASHTO Test Requirement
AASHTO Electrochemical
Method Requirements for Mechanically
ρmin T 288 >3000 Ω-cm Stabilized Earth Fill Used with
pH T 289 5 to 10 Galvanized Steel
Sulfates T 290 <200 ppm Reinforcements

Chlorides T 291 <100 ppm

Criteria Used in
the US for
Assessing Ground
Corrosion
Potential Relative
to SBSN’s (after
FHWA, 2003)
31
German Gas and Water Works Engineers’
Association Standard (DVGW GW9)

32
ITEM MEASURED VALUE MARKS

Calcareous, marly limestone, sandy marl, +2

not stratified sand
Soil Loam, sandy loam (loam content 75% or 0
Composition less), marly loam, sandy clay soil (silt
content 75% or less)
Clay, marly clay, humus -2
Peat, thick loam, marshy soil -4
Ground water None 0
level at buried Exist -1
position
Vary -2
> 10,000 Ω-cm 0
5000 Ω-cm – 10,000 Ω-cm -1
Resistivity
2300 Ω-cm – 5000 Ω-cm -2
1000 Ω-cm – 2300 Ω-cm -3
> 10000 Ω-cm -4 33
 ITEM MEASURED MARKS ITEM MEASURED VALUE MARKS
VALUE < 200 ppm 0
Moisture Content 20% or less 0 Sulfate 500 ppm – 200 ppm -1
20% or more -1 1000 ppm – 500 ppm -2
pH 6 or more 0 > 1000 ppm -3
6 or less -2 None 0
None 0 Cinder exist -4
Sulfide and Trace -2 & Coke
Hydrogen Sulfide
Exist -4
5% or more +2
Carbonate 1% to 5% +1
< 1% 0
< 100 ppm 0
Chloride > 100 ppm -1
34
 DVGW GW9 –
Characterization of Corrosivity
SCORE CHARACTERIZATION
>0 Noncorrosive
0 to -4 Slightly Corrosive
-5 to -10 Corrosive
< -10 Very Corrosive

35
 Soil Corrosivity/Aggressiveness
(Carbon Steel) DIN 50 929 Part 3
General Range Localized Range
Corrosion (Pitting)
Total Rate Corrosion
Score Rate
µm/yr µm/yr
≥0 Ia 5 2.5 – 10 30 15 – 60
-1 to -4 Ib 10 5 – 20 60 30 – 120
-5 to -10 II 20 10 – 40 200 100 – 400
< -10 III 60 30 - 120 400 200 - 800
36
Sample GN #10 Test Corros. CR Cluster
Method RankA (µm/yr)
(Proposed
Protocol) (I) Data Clustering Relating
Galv. Plain
San Antonio, TX
Wake Forest, NC
0.18
2.21 8
2 Tex-129-M
Tex-129-M 2
1.0
0.3
NAB
< 0.1 Corrosivity Rankings to
Bastrop, TX 0.15 2 Tex-129-M 2 0.4 NA
1.8C
Observed Rates of

Not Corrosive
Ashdown, AR 2.88 36 AASHTO T-288 2 NA
TTC, NC 3.51 24 AASHTO T-288 1 5.8 1.6

(I) ≥0
Ocala, FL 5.65 91 AASHTO T-288 0 1.8 3.8
LWF, South
Carolina
El Paso
4.83

0.22
68

2
ASTM
WK 24261
Tex-129-M
0

0
1.2

0.2
8.4

NA
Corrosion
Coarse/MSE
Waco, TX 1.26 7 Tex-129-M 0 0.3 NA
M-U-D, NY 5.24 82 AASHTO T-288 -1 4.8 39

Slightly Corrosive
Garden City, TX 2.52 22 Tex-129-M -1 4.3 NA
GB, South Carolina 4.48 56 AASHTO T-288 -1 3.2D 5.8 -3 ≤ (I) <0 Corrosivity Clusters Observed Corrosion Rates, CR
Galvanized Plain Steel
Maple Rd., NY 2.50 22 Tex-129-M -2 3.7 16 (I) ≥ 0 CR < 2 µm/yr CR < 5 µm/yr
El Paso Fine/MSE 5.52 87 AASHTO T-288 -2 21E NA -3 ≤ (I) < 0 2 µm/yr < CR < 5 µm/yr 5 µm/yr < CR < 20 µm/yr
Prince George, BC 2.89 32 AASHTO T-288 -3 NA 20 -5 ≤ (I) < -3 10 µm/yr < CR < 35 µm/yr 20 µm/yr < CR < 40 µm/yr
PIP, NY 4.62 61 AASHTO T-288 -4 37 30
-5 ≤ (I) <-3

Sprain Brook Pkwy, 2.54 27 AASHTO T-288 -4 33 NA

Corrosive

NY
Quarry; El Paso, TX 3.64 41 AASHTO T-288 -4 14.8 NA

Rochester, NY 3.85 49 AASHTO T-288 -5 9.6 20

37
 NCHRP 21-11 – PHASE III
• Phase III (Tasks 7, 8 & 9) – Implement Work Plan
Developed in Phase II.

• Conduct trials in active construction projects

• Shadow specification to compare with current practice
• Demonstrate and evaluate recommendations and protocols
for sampling, testing and characterizing corrosiveness of
earthen materials.
• Initiate training with personnel from State DOTs

38
Conduct Field Tests to Support Recommended Specifications for New
Methodology

• Identify fill source prior to construction

• Relevant activities include:
• Sampling the materials before construction
• Sampling the materials during compaction
• Conducting field resistivity for comparison to the lab resistivity
• Conducting moisture-density tests with help of DOT

• Conduct index and electrochemical tests on:

• Materials collected before construction
• Materials collected during construction

39
 Insitu Soil Testing

• Stevens Hydraprobe uses a five-tine probe

that can measure the subsurface moisture
content and conductivity which can be
converted into resistivity
• The Wenner 4-electrode method (ASTM G-57)
measures the resistivity of the subsurface by Calgary, AB
inserting 4 equally spaced electrodes into
the ground in a line, applying an electric
current between two outer electrodes, and
measuring the corresponding voltage drop
between the inner electrodes.

Marcy – Utica – Deerfield, NY

COARSE GRAINED SAMPLE AT SATURATION

41
 REMARKS
• Consistent trends are observed between results obtained with
different test methods and between materials with different textures
• Salt contents are a good check on measured resistivity
• AASHTO Test Methods apply well to materials with more than 25%
passing the #200 sieve.
• TX-129-M and TEX-620-M apply well to coarse materials
• Results from TX-129-M and TEX-620-M can be correlated
• Good correlation between observed performance (CR’s) and
characterization of corrosion potential based on resistivity
measurements.