STAnDI
TABLE OF CONTENTS Paragraph
Page I - Introduction
1 - 1
II - Procedure 2.1
Examination of Piling
2 - 1
2.2
Hemphis Test Rods
2 - 1
2.3
Pipe and Rod Specimens
2 - 2
2.4
Soil Tests
2 - 4 III - Discussion
3.1
r:eneral
3 - 1
3.2
Memphis Floodwall, Memphis, Tenn.
3 - 8
3,3
Sardis Dam, Sardis, Miss.
3 - 10
3.4
Grenada Dam, r.renada, Miss.
3.5
Bonnet Carre Spillway, Norco, La.
3 - 13
3.6
Stoplog Dam, Freeport, Tex.
3 - 13
3.7
Zia Pueblo Diversion Dam, San Ysidro, N.Hexico
3 - 14
3.8
Hartford Dike, Hartford, Conn.
3 - 14
3.9
Fairbanks Test Piles, Fairbanks, Alaska
3 - 15
3.10
Corrosion Specimens
3 - 16
' 3 - 11
IV - Conclusions 4.1
Piling
4 - 1
4.2
Corrosi·on Specimens
4 - 1 V - Bihliographv
1.
INTRODUCTION
Steel piling has been used underground for many years, and while there is no known instance of structural failure due to the corrosion of piling in soil, the increasing awareness of corrosion damage which can occur to buried steel structures has caused concern that piling may be similarly damaged. This study was undertaken jointly by the Corps of Engineers and the National Bureau of Standards to investigate the extent of underground corrosion of piling in various soil environments.
Steel piles,
in service for periods of six to fifty years, and exposed to a wide variety of soils, were examined for corrosion.
In addition, specimens
consisting of sections of steel pipe and rod were buried or driven in the soil at a few sites in order to compare their corrosion with that of the examined piling.
Polarization curves were run periodically on
a representative group of the corroding specimens. The report summarizes the results of the investigation.
A detailed
report (NBS Monograph 58) covering a portion of this study was published by the National Bureau of Standards in 1962. Standards has in preparation, and will
The National Bureau of
p~blish,
additional detailed
reports which will include the parts of the study not covered in the monograph.
1-1 •
2. 2.1
Examination of Piling.
PROCEDURE
Piling was pulled where possible; however,
pulling in most instances was not possible and it was necessary to dig inspection pits adjacent to the piling.
Pits were generally excavated
to a depth of 10 to 15 feet, but were limited to a lesser depth by the ground water at some sites. Exposed piling was cleaned of soil and corrosion products by means of wire brushes and scrapers. corrosion noted.
Pit depths were measured and the extent of
In most instances, sample sections were removed for
further cleaning, examination, and photographing.
They were generally cut
from areas of maximum corrosion at each location; however, two or more samples were removed at some sites where there was a marked difference in soil environment or in the severity of corrosion.
Each sample section was
die stamped with an identifying number. 2.2
Memphis Test Rods.
Eighteen 3/4-inch
hot-rolle~
mild steel rounds,
15 feet long were driven along the Memphis Floodwall in areas where corrosive conditions were throught to be severe.
The rods were cleaned
with a wire brush to remove loose mill scale and then washed in a solvent to remove grease and oil.
After cleaning, they were weighed and driven at
selected locations near the floodwall.
The rods remained in the ground for
periods of about one to three years, after which they were withdrawn, cleaned, weighed, examined, and photographed.
Two of the rods (Nos. 18 and 19) were
connected to 5-lb magnesium anodes for cathodic protection.
2-1
2.3
Pipe and Rod Specimens. 2.3.1
Pipe Specimens.
of four sites.
Sixteen test specimens were buried at each
The specimens which consisted of 14-inch lengths of
1-1/2-inch standard black, uncoated steel pipe were cleaned of loose mill scale, washed in a solvent, then weighed, and the inside surfaces coated with a heavy grease.
Rubber caps were placed over the open ends.
One-fourth of the specimens was equipped with insulated wire leads soldered to the inside pipe surface and extended through one of the rubber end caps.
Specimens were buried horizontally two to three feet deep in
trenches about twenty feet long.
They were arranged in groups of four
with one wired specimen in each group.
Lead wires were run to terminals
mounted in a junction box on a centrally located post and polarization curves were run periodically on the wired specimens.
A wire was run to
the junction box from a driven five-foot iron pipe, which was used as the remote electrode during tests.
Specimens were removed by groups after
exposures of one to seven years after which they were cleaned, photographed, and weighed. 2.3.2
Rod Specimens.
Two 3/4-inch by three-foot steel rods were
placed underground at each of the pipe specimen test sites.
The rods
were hot-rolled, mild steel rounds and were cleaned as described in paragraph 2.2 above.
One rod was buried vertically in an excavation; the
other was driven into the soil 18 inches from the buried rod.
-
All specimens
were installed with the tops of the rods six inches below grade.
2-2
The
soil in which the rods were placed was either undisturbed soil or compacted fill which had been in place for more than ten years.
Rods
were equipped with insulated wire leads which were run to terminals in the junction boxes mentioned in paragraph 2.3.1 above.
The top five
inches of each rod, including the lead connection, were coated with coal tar paint making the exposed (bare) surface area equal to same as the exposed surface area of the pipe specimens.
o.s
ft 2 ; the
Polarization
curves were run periodically on all rods during the five-year test period. 2.3.3
Polarization Curves.
Polarization curve data were obtained
using the "Holler" bridge circuit described in NBS Circular 579 {Bibliography, reference 1). the potential readings.
This circuit eliminates the lR drop from
After balancing the bridge, the specimen
cur~ent
was increased in equal steps with one minute allowed for polarization at each value of current.
At the end of each polarization period, the
specimen potential was measured with a potentiometer vqltmeter.
The
voltage measured by the potentiometer in this bridge circuit is one-half the actual potential between the specimen and the copper-copper sulfate reference electrode. The polarization data were used to plot specimen potential (volts) against current {milliamperes) on semi-logarithmic chart paper.
The
corrosion current in milliamperes (io) was calculated from the cathodic and anodic curves using the relation io
= Iplq Ip+Iq
where Ip and Iq are the · values of current at the "breaks" or "knees" in the cathodic and anodic
2-3 '
polarization curves, respectively. per day is equal to 2Sio. based on Faraday's law.
The corrosion rate in milligrams
The value of 25 used here is a constant From these corrosion rates, the weight loss
of the specimens was estimated by assuming that the average corrosion rate for the period between determinations to be equal to the average of the two determinations.
Polarization data were also plotted on
rectangular coordinate chart paper to verify the "breaks" obtained on the semi-logarithmic paper. 2.4
Soil Tests.
the field.
Determination of soil types was made visually in
Also, soil samples were classified and tested in the laboratory
by the Waterways Experiment Station or the National Bureau of Standards. Undisturbed samples were secured from pipe specimen trenches and initially from piling test pits for comprehensive tests, including chemical, physical, and Denison corrosion cell tests (Bibliography, reference 4).
In later
piling examinations, tests were limited to pH and resistivity measurements of disturbed samples.
In the case of extracted piles, samples were secured
from the soil adhering to the steel surface.
Laboratory resistivity
measurements were made using a l-inch diameter by l-inch soil box with a Model 274M VIBROGROUND.
Soil samples were saturated with distilled water
and all measurements were taken at a temperature of 78F.
The maximum
measurable resistivity with this equipment was 4000 ohm-em. pH
~eterminations
Laboratory
were made by the electrometric process with a Model 1115
PHOTOVOLT, which employs calomel and glass electrodes.
Measurements were
2-4 •
made without the addition of distilled water, except where it was necessary to moisten dry samples to obtain a reading.
Soils were
classified by means of the Unified Soil Classification System (Bibliography, reference 10).
.
Field resistivity measurements, using
Shepard Canes or the four-pin method, were made at most sites. measurable resistance with the Shepard Canes was 10,000 ohm-em.
2-5
Maximum
3. 3.1
General.
DISCUSSION
A total of 47 piling inspections was made at 33 sites
in this investigation.
The examined piling was exposed to soils which
varied from fill material (consisting of debris, cinders, shell, sand, silt, clay, and various combinations of these) to undisturbed natural soils ranging from well-drained sands to impervious clays.
The pH of
the soils varied from 2.9 to 8.6, and resistivities ranged from a low of 20 ohm-em to values exceeding 100,000 ohm-em. One of the characteristics of underground corrosion of steel is its irregular nature; often large areas of the surface are unaffected by corrosion while other areas may be deeply pitted or even perforated. It is evident from this investigation that piling corrodes in this manner.
It should be noted, however, that this type of corrosion,
which is very serious in a pipe line designed to contain fluids under pressure, may be of little consequence in the case of piling used as a load bearing structure.
Deep pits or perforations in piling are not
serious provided the amount of metal removed does not materially reduce the cross section. Initially, as mentioned in paragraph 2.4 above, comprehensive tests, including corrosion cell tests, were conducted on soil samples to determine if the observed corrosion could be predicted from test results.
It was apparent that there was little or no correlation.
With
few exceptions, the corrosion noted was much less severe than would be
anticipated on the basis of the tests.
For this reason. on subsequent
inspections. soil tests were limited to the determination of pH and resistivity. That part of piling exposed to fill soil in. and above. the zone of water table fluctuation was found to be the most susceptible to corrosion.
In soil below the water table. corrosion was negligible
in every instance.
In undisturbed soil. corrosion was negligible or
minor in all but a few instances.
The term "undisturbed" as used here
is defined as a consolidated naturally deposited soil composed of native materials.
The absence of corrosion in these areas is attributed to a
deficiency of oxygen.
It is thought that the observed corrosion occurred
initially. then slowed or ceased as the oxygen was depleted.
Laboratory
tests at the National Bureau of Standards (Bibliography. reference 2) . have conclusively demonstrated the marked effect of the presence. or absence. of oxygen on the corrosion of steel in soils. Table 1 is a summary of the results of the piling inspections.
In
reviewing the table. it should be noted that the laboratory soil data are for a sample of the soil to which the corresponding piling sample was exposed.
The field resistivity values indicate the range of readings
measured at different depths with Shepard Canes or the four-pin method. At a few sites. where indicated. field resistivity determinations were limited to surface measurements with Shepard Canes.
Some laboratory
resistivity determinations are outside the range of the field readings.
3-2 •
TABLE 1, INSPECTION SUMMARY SOI L DATA
PILING DATA
Laboratory SITE
Inspection or Sample No • .
.
Examined Length (Ft.)
Exposure (Years)
Sample Depth (Ft.)
%Mill Max. Pit Sca le Depth (c) (Inches) Removed
Classification
Field
Samp le Depth (Ft. )
pH
Corrosion Ce ll Wt. Loss (Oz/ Ft 2)
Resistivity Rang' Ohm-Cm
Resistivity Ohm- Cm
Memphis, Tenn.
AlOl* Al02 Al03*
6,5 8,0 7,8
7 7 8
8(a) 9(a) 9(a)
0.035 Nil 0.133
5 Nil 10
(f) Clayey Silt (ML), Tan & Gray (f) Clayey Silt (ML), Tan (f) Cinders, Glass, Decayed Wood, etc.
8(a) 9(a) 11 (a)
7. 6 7. 8 6 .8
3 .59 2.1 6
-
1420 1700 850
1000- 3200 1500- 3200 45-10000+
Vicksburg, Miss.
BlOl Bl02
8.5 8.5
7 7
lO(a) lO(a)
0.040 Nil
65 30
12 (a) 12 (a)
8 .2 8 .6
4.49 4.19
1760 3910
850-4000 720-1 700
ClOl(e)
3,5
20
5(b)
0.060
30
7(b)
2.9
10.71
1400
-
Cl02
5.0
20
9(b)
0,025
10
(f) Sandy Silt (ML), Gray (f) Silty Sand (SM), Gray, w/Brick & Glass (u) Sand (SP) w/Layers of Clayey Silt (ML), Gray (u) Silty Clay (CL), Tan
9(b)
6.0
5. 94
4000+
3000 - 7500
D101
5,5
12
9(a)
0.108
80
9(a)
4.0
5 .30
2290
1700- 3500
Dl02 Dl03A} (e) Dl03B
5,5
12
50 80 30
9(a) 6(a) 15 (a )
6 .4 4.9 3. 6
-
4000+ 4000+ 3850
5000-1 0 , 000
12
0.160 0.122 Nil
2.25
13.7
9(a) 8 (a) 17(a)
(f) Clayey Sand (SC), w/Pockets of Sandy Clay (CL), Brown (f) Silty Sand (SM), Yellow (f) Silty Sand (SM), Tan (u) Shale
Berwick, La.
FlOl Fl02
3.0 3,5
11 11
3(a) 3(a)
0,060 0,090
50 60
(u) Silty Clay (CL), Brown (u) Clay (CH), Brown
3(a) 3(a)
8 .5 8.1
3. 74 3.50
14 20 1050
800-1 000 750-1 300
Algiers, La.
Gl02
2.5
12
3 (a)
0,040
15
(f) Silty Clay (CL), w/Pockets of Clay (CL), Brown
3(a)
8.4
3. 37
1150
700-1 300
Enid, Miss.
HlOl
5.5
12
9(a)
Nil
10
(u) Silty Sand (SM), Red
lO(a)
5.1
-
4000+
9000-10,000
El Dorado, Ark.
1101 (e) (g)
14.8
38
18 (a)
0.026
10
(u) Sandy Clay (CH), Blue
18(a)
6.2
-
1540
3000- 3500 (Surface)
440 300
--
1980
2800- 3000
-
2560
6500-1 0 , 000
4000+
5000-10,000
-
700
1100
Sardis, Miss.
Grenada, Miss.
.
- .
100 15
(u) Silty Sand (SM), Light Gray (u) Clay (CH), Tight, Gray
7(b) 14 (b)
7.8 6.9
8(b)
8.0
-
31.0
32
#a #b
KlOl
5.0
22
#b
Nil
10
(u) Silty Clay (CL), Tan, Soft
LlOl
6,0
13
9(a)
Nil
20
(u) Clay (CH), Brown, Stiff
l O(a)
6.4
MlOl
4.5
20
9(a)
Nil
30
(u) Clay (CH), Firm, Brown & Tan Organic Matter
lO(a)
4.5
Omaha, Neb,
Nl
2.3
14
16(a)
0.050
. 60
(u) Clayey Silt (ML), Gray
17 (a)
7.8
Matagorda, Tex.
Pl P2
15.7
16
7 (a) 14(a)
-Nil
100 Nil
6(a) (f) Oyster Shells, No Soil (u) S il ty Clay (CL), Brown (Lime in Soi 1) 14 (a)
7.3
Rl
14.0
16
31.5
0.040 0.040 Nil
10 70
(f) 0 yster She 1ls, No Soi 1
21
5(a) 4(a) 26(b)
(u) Silty Clay (CL), Firm Brown (u) Silty Clay (CL), Soft, Brown
5 (a) 1 (a) 27(b)
7.5
(n)
16 16 16
20(b) 3(a) 3(a)
Nil 0,040 0.015
Nil 10 5
(f) Gravelly Sand (SP), Tan (f) Sand w/Gravel (SP), Tan ~f) Sand w/Gravel (SP), Tan
20(b) 3(a) 3(a)
6.5 6.1 5,2
--
4000+ 4000+ 4000+
33000-260000
5(a)
sea)
0.015 Nil
70 Nil
(u) Clay (CH)' Gray (u) Clay (CH)' Gray
4(a) 10 (a)
7,4 7.4
--
950 1080
2900-8000
25
(u) Clay (CH), Brown
4 (a)
7.4
-
1890
1800-10,000+
Nil
(u) Clay (CH), Brown
8(a)
6.1
-
3870
1650-10,000+
New Orleans, La.
JlOl(e)
Harrisburg, Ill. Louisville, Ky.
0.040 0.104
•
3600-14, 000
Jeffersonville, Ind.
Freeport, Tex.
R2-l} R2-2 (e)
Pearl River, La.
Sl* 55* 58*
Detroit, Mich.
Tl T2
North Little Rock. Ark.
V1
Newport, Ark.
Wl*
10.0
..
11
10.0
24
4(a)
0.050
s.s
24
8(a)
Nil
75
/
-
7.3
.
-
-
150 80
600-10,000+ 10000 300- 350 (Surface)
.
•
TABLE l, I NSPECTION SU'tM.ARY - Continued SOIL DATA PILING DATA Laboratory
SITE Inspection or Sample No, Charleston,
s.
C,
St. Lucie Lock, Florida San Ysidro, N. H,
Xl I
'
Examined Length (ft.) 15. 0
Exposure (Years)
Sample Depth (Ft.)
24
8(a)
Yl*
11.0
24
8(a)
Zl
10. 0
27
6(a)
Max. Pit Depth (c) (Inches)
Classification
AAl(h)
Hartford, Conn. Nome, Alaska
Hinnesota City, Minn.
10.0
23
16(a)
0 .170
100
ACl
5. 0
30
3(a)
0 . 080
3S
(e)
35.3
O.l2S(P)
\
60
20- 10 , 000+
(u) Gravelly Sand (SP), Tan, to 1.0 Inch Max.
S(a)
8. 3
-
4000+
10000+
Sandy Silt (ML) w/Gravel & Cinders
l S(a)
7.8
-
2090
-
3(a)
7. 6
-
2090
-
(f)
40 Nil Nil
(u) Sand (SP) • Fine to Medium, Tan (u) Sand (SP), Fine to Medium, Red (u) Sandy Silt (1-1L) , Black
5(a) 23(b) 43(b)
7.7 4.8 7.3
Nil Nil
10 Nil
(u) Cinder-like sand (SP), Black (u) Silty Clay (CL), Soft, Tan & Gray
6(b) 28(b)
6.9 6. 7
(u)
~i} (e)
l O(b) 30(b)
Olcott, N. Y.
AH-l(e) (i) 2.5
sot
2(b)
0,045
100
Buffalo, N, Y.
AJ-l(g)
5.S
36
6(a)
o.oso
so
I
AK-1
3.2
32
3(a)
0 .04S
100
4(a)
0,020
100
Nil
Nil
30 22(b)
II
(u) Silty Sand (SC), Brown w/Lumps of Silty Clay (CL) & Small Gravel
.
AL-2
3600
0 .055 Nil Nil
27
22.3
-
Nil
30.3
Fairport, Ohio
7.8
Nil
Dresback, Minn.
AL-l} (e)
7(a)
32(b) 28
Presque Isle Peninsula, Pa.
(u) Sand (SP), Tan & ~ ray , Hedium to Coarse
so
40.0
AE3)
lS00-10 , 000+
Nil
(e)
~
1550
3(b)
6(b) 23(b) 43(b)
Winona, Minn.
-
(f) Sand (SP), Medium, Black, w/cinders (u) Sand (SP), Fine to Medium, Brown w/Clay Balls (u) Sand (SP), Fine to -Medium, Brm.m
. AEl AE2
7.6
~
29
AD2,
8(a)
40
ABl
~
(u) Sandy Clay (CL), Tan & Gray
0 . 020
S(a)
ADl
4000+
20
14
pH
(u) Fine to Medium Sand (SP), Tan & Gray ' (f) Silty Sand (SC), Brown w/small Gravel (u) Siltv Clav (CL). Grav w/Gravel
1
l(a) 24(b)
7.4 5.4
l(a)
7.9
4(a)
8.0
2l(a)
-
-
4000+
-
-
4000+
--
7.5
-
1120
-
17 (bp)
6 .1
--
3000
--
-
2410
-
-
-
4000+ 4000+
-
4000+
-
1970 1870
-
11
3(ap) S(bp)
0,018 Nil
25 Nil
(f) Silty Clay (CL), Tan II
3(ap)
-
6.9
0.020 Nil
40 Nil
(f)
Silt (ML)' Gray (f) Silt (ML) • Gray
3(ap) 5{hp)
6.1 6.1
(u) Silty Clay w/High Peat Content (CL, PT), Gray (f) Silt (ML) • Gray (f) Clay (CL), Gray
S{hp)
5.6
3(ap) 5(bp)
6.9 6.7
0.027
50
6
3(ap) S(bp)
0.018 Nil
10 Nil
FC-8l(e) (k)
15.5
10 , 000+ (Surface)
1620
20.8
3(ap)
1750 3400
-
-
FA-25(e)(j)
8
10,000+ (Surface)
8.1
II (f) Silty Clay (CL), Brown
19.5
4000+ 4000+ 1890
7(a)
40 Nil
FBD-7(e)(m)
--
-
0.028 Nil
11
10,000+ (Surface)
-
3(ap) S(bp)
20.1
4000+ 1870
-
-
11
FA-32(e){j)
--
-
20.4
3(ap) S(bp)
-
•
-
not obtained. (a) Above or in the water table fluctuation zone. (b) Belm>7 the water table. (c) Haximum sample pit depth; where sample was the figure is the maximum pit depth measured above, or below, the water table , as indicated, (e) ~iling extracted by pulling. (f) Fill. Piling was 3/8-inch steel sheet type, e~cept: (g) 1/2-inch steel sJ1eet, (h) 1/8-:f.nch steel sheeting , (i) 12 X 4- inch beam "t-Yith separate interlocks, (j) 8Bl5 beam. (k) 6\7F2S beam, (m) 8-inch, 25 l h/f t. standard steel pipe. ~n) 4-inch square sa~ples cut from piling wall. (P) Perforation, (u) Undistrubed soil. (ap) Above permafro~t. (bp) Belmr permafro~t. *Coated piling. #Sample not obtained• .
400- 2500
-
FA-17 (e) (m)
Fairbanks, Alaska
Resistivi ty Ohm- Cm
Resistivitv Range Ohm- Cm
-
0,072
9.5
Corrosion Cell Wt, Loss(Oz/Ft 2)
8. 3
(f) Sand w/Small Clay Lumps (SP)
I
Lamar, Colo,
Sample Depth (ft.) 8(a)
100
Nil
'
% Mill Scale Removed
Field
-
.
This is explained by the fact that laboratory tests were made on small samples, some of which were not truly representative, or, because laboratory samples were saturated with water before measurement, whereas the moisture content of some of the soils in the field mav have been low. If it is assumed that the average reduction in thickness of the corroded areas is equal to one-half the maximum pit depth, the percentage reduction in cross section of sheet piling (R) can he using the relation: R
where P is the maximum pit depth, and T is the piling thickness.
l'H
=T ~f
rou~hly
estimated
(1)
is the percent of mill scale removed
Usi.ng this equation to evaluate the data
in Table 1, an arbitrary assumption was made that a maximum reduction of cross section in excess of 5 percent in 7 to 15 years, 10 percent in 15 to 25 years, or 15 percent for longer periods may be serious and should be further examined in more detail. 3/8 inch was assumed for the
In all calculations, a thickness of
pilin~.
In establishing these criteria, an extrapolated loss of approximately 35 percent in cross section over a 50-year period was assumed as the acceptable maximum.
l~ith
a constant corrosion rate, a reduction of
5 percent in 7 years will result in a loss of this magnitude.
It must
be borne in mind, however, that corrosion rates do not remain constant, but rather have a tendency to decrease, often very rapidly, with the passage of time.
Hence this method of estimating piling life is very
conservative, particularly when based on the shorter exposure times •
•
3 - 5
Application of equation (1) to piling exposed to soil below the water table indicates negligible corrosion in every instance. In undisturbed soils, piling samples FlOl, Fl02, Nl, and AAl have estimated reductions in cross section of 8 1 14, 8, and 20 percent, respectively, and warrant further study.
For each of these samples,
equation (1) gives an exaggerated result because all had a few deep pits, but on most of the corroded areas the metal attack and pitting were very shallow.
On
samples FlOl, Fl02, and Nl, the actual reductions
in cross section are believed to have been in the range of 3 to S percent which are not considered serious for the periods that these samples were exposed.
Sample AAl was removed from a 1/8-inch steel sheeting
retaining wall adjacent to an abutment for a highway bridge.
This soil
was classified as undisturbed because it was stratified; however, an automobile reflecting lens was found near the bottom of the excavation indicating that the soil had been deposited by the river, possibly only a short time before the sheeting was driven.
It is also possible that
some of the surrounding soil may have been removed by scour and later replaced by river deposits while the sheeting was in place.
In view
of the uncertainties regarding the history of this sheeting, it has been disregarded in the conclusions. 'Five of the samples exposed to fill soils (8101, 0101, 0102, Dl03A, and ABl) exceeded the above criteria.
Except for a few pits of
very small area, all of the metal attack and pitting on 8101 was less
3-6
than 0.010 inches in depth.
The actual reduction in cross section
is believed not to have exceeded a negligible 1 or 2 percent.
The
remainder of these samples is discussed in paragraphs 3.4 and 3.8. At some sites, only minor corrosion was noted even where the piling was exposed to potentially very corrosive fill soil.
The piling
in the Memphis (AlOl, Al02, and Al03) and Vicksburg (8101 and 8102) floodwalls, Berwick Lock (FlOl and Fl02) and Algiers (Gl02) Lock are good examples.
There is no apparent explanation for this.
It is
interesting to note, however, that at all of these sites, the major portion of the piling extended below the water table and the part above the water table was comparatively small.
It appears that the ratio
of the below-water length to the above-water length may be a factor in the severity of corrosion occurring above the water table.
The National
Bureau of Standards is continuing investigation of this to develop a possible explanation. Coated piling was examined at four sites:
Memphis, Tenn.; Pearl
River, La.; Newport, Ark.; and St. Lucie Lock, Fla.
All of the coated
piles had been painted with a coal-tar paint prior to driving.
Coatings
were in good to excellent condition, which was rather surprising since some of the piles had been driven through soils containing gravel, cinders, an~ebris.
While there is no way to evaluate the effectiveness of the
coatings, their condition attests to the fact that they did provide some measure of protection.
3-7
A representative group of some of the more interesting inspections and the corrosion specimen results are discussed below. 3.2
Memphis Floodwall, Memphis 1 Tenn.
Test rods were driven adjacent
to the Memphis Floodwall prior to the start of this investigation in an attempt to determine if the floodwall piling should be cathodically protected.
When the wall was constructed, provisions were made for
the future installation of a cathodic protection system.
The test rod
data are shown in Table 2. The first piling examinations (AlOl and Al02) indicated negligible corrosion; however, several of the test rods, subsequently removed, were badly corroded.
An additional inspection excavation (Al03) was
then made to a depth of about 12-1/2 feet at the site of rod No. one of the severely corroded rods.
s, .
The soil here was fill consisting
of cinders, glass, brick, decomposed wood and silt.
The resistivity varied
from 45 ohm-em to more than 10,000 ohm-em. That this was an extremely corrosive environment is confirmed by the fact that the test rod lost in excess of 7.5 percent of its weight in a 26-month period.
Corrosion was severe over most of the rod.
was extremely severe at depths of 7 to 10 feet and 13 to 16 feet.
It The
latter was in the zone of normal water table fluctuation. The severity ~f
the metal attack increased progressively from the 13-foot depth to
the end of the rod, producing a noticeable taper.
The diameter of the
3/4-inch rod near the tip was reduced to 0.612 inches.
3-8
.
•
TABLE 2 TEST RODS - MEMPHIS FLOODWALL
SIDE ROD NO.
~
I (0
STATION
OF WALL
DISTANCE FROM WALL
1 2 3 4 18* 17 6
34+00 34+00 37+40 37+40 37+40 41+91.5 42+00
L.S. R.S. L.s. R.S. R.S. L.s. R.S.
18 15 18 19 23 17
7 8 9 10 19* 11 12 13 14 15 16q
51+50 51+50 56+00 56+00 56+00 64+00 64+00 70+00 70+00 83+00 88+00
L.s. R.S. L.S. R.S. R.S. L.S. R.S. L.S. R.S. L.S. L.S.
5 5 5 5
17
9
5 5 5 5 41.5 41.7
DATE INSTALLED
DATE WITHDRAWN
MONTHS EXPOSED
SURF LENGTH AREA SQ. FT. FT.
1/21/59 1/22/59 1/22/59 1/22/59 3/3/59 1/21/59 1/22/59
3/31/61 12/5/61 12/5/61 12/5/61 12/5/61 12/5/61 3/20/60
26 34 34 34 33 34 14
2.95 2.95 2.95 2.95 2.95 2.95 2.95
1/20/59 1/20/59 1/20/59 1/20/59 3/3/59 1/20/59 1/20/59 1/20/59 1/20/59 1/16/59 1/16/59
12/5/61 3/31/61 3/20/60 12/5/61 12/5/61 3/31/61 12/6/61 12/5/61 3/30/61 3/30/61 12/6/61
35 26 14 35 33 26 35 35 26 26 35
2.95 2.95 2,95 2.95 2.95 2.95 2,95 2.95 2.95 2.95 2,95
FINAL wt
wt
WT LO~S LOSS OZ/Ff I YR OZ/FT2
WT
WT
GMS
GMS
LOSS GMS
15.02 15.02 15.01 15.02 15.02 15.01 15.02
10099 9986 10003 9874 9971 9821 9955
9495 9413 9485 9425 9808 9595 9919
604 573 518 449 163 226 36
7.22 6.85 6,19 5,37 1.95 2.70 0,43
3.34 1.94 1.76 1.52 0.54 0.77 0.37
15.01 15.02 15.01 15.02 15.00 15.02 15.02 15.01 15.02 15.01 15.01
9986 9962 9906 10016 10160 9962 9938 9923 9989 9914 9907
9743 243 9210 752 9815 91 9879 137 10151 9 9732 230 9701 237 9521 402 9893 96 9874 40 7737 2170
2.91 9.00 1.09 1.64 0.11 2,75 2.84 4.81 1.15 0,48 26.0
0.85 4.15 0.93 0.48 0.03 1.27 0.82 1.40 0.530.22 7.59
*Rod cathodically protected by means of a 5-lb. magnesium anode. NOTE: All rods are 3/4" 0 with conical points, and have 1/4" 0 holes drilled approximately 2" from top. Rods were driven to such a depth that the tops of the rods were about 1 ft. below grade. Side of wall is designated as landside (L.S.) or riverside (R.s.).
I
The piling, which had been exposed for approximately 8 years, was painted with two coats of cold-applied, coal-tar-base paint before driving.
On initial inspection, the coating appeared to be intact;
however, on removal of the paint, scattered areas of pitti~g and metal attack were noted.
It is believed that these were the result of small
holidays in the coating.
In this type of environment, accelerated and
very rapid corrosion would be anticipated at any holidays; however, this was not the case. was present on 90
The corroded areas were not large (mill scale
p~rcent
of the piling surface) nor were the pits
unusually deep (maximum pit depth was 0.133 inches). The marked difference in the corrosion attack on the rod and piling lends credence to the theory that the below-water, above-water ratio may be significant.
It should be noted that the rod, as installed, was
almost entirely above the normal water table, whereas, about two-thirds of the piling was below the water table. 3.3
Sardis Dam, Sardis, Miss.
Two piling inspections, described below,
were made at this site, which was also selected for the location of test specimens.
From available information (Bibliography references 8 and
9), it was known that the ground water was very corrosive.
The age of the
piling was about 20 years. 3.3.1
Outlet Channel Piling (ClOl).
This was a 3-1/2-foot length
of piling installed as part of a sump box in the dam outlet channel during construction of the project.
The entire section was belowthe normal
3-10
water table.
The top foot of the pile was in a gravel bed below the
channel riprap while the remainder was exposed to a black lignitic clay soil having a resistivity of about 1400 ohm-em and pH of 2.9.
This was
the lowest pH of all the soils encountered in the investigation, and the corrosion cell tests indicated this soil to be the most aggressive of the soils tested.
Corrosion of the piling, however, was negligible, generally
occurring as shallow uniform metal attack and pitting over most of the surface in the section exposed to gravel. inches was measured in this area.
A maximum pit depth of 0.060
Mill scale over the remainder of the
piling which was exposed to soil was about 70 percent intact and pit depths did not exceed 0.030 inches. 3.3.2
Wingwall Piling (Cl02).
piling wingwall of the dam spillway.
A test pit was excavated to expose a The upper two feet of the piling
were exposed to a red clayey sand fill having a resistivity in excess of 10,000 ohm-em.
The remaining 3 feet of piling were in undisturbed soil
consisting of a tan silty clay having a resistivity of 3000 to 7500 ohm-em and a pH of 6.0. the steel surface.
Mill scale was present over about 90 percent of
Most of the corrosion was confined to the top 8 inches
of the piling, where shallow metal attack and pitting were noted. pitting was in evidence over the rest of the surface.
Isolated
A few pits had depths
of about 0.025 inches, but most were less than 0.010 inches. 3.4
Grenada Dam, Grenada, Miss.
this site.
Three piling examinations were made at
The first two (0101 and 0102) were made by excavating test
3-11
pits adjacent to the 12 year old dam spillway wingwalls.
The piling,
which was exposed to fill soils, indicated the possibility of serious corrosion.
A third inspection (0103) was then made by pulling the
14-foot end pile of the north upstream wingwall. above the normal water table.
The entire piling was
The top 5 feet were exposed to a fill soil
of tan silty sand having a resistivity of 7,000 to 14,000 ohm-em and a pH of 4.9.
The remainder of the piling was in undisturbed soil consisting
primarily of blue-grey shale with a resistivity of about 3,900 ohm-em, and a pH of 3.6.
On the port.ion which was exposed to fill, the corrosion
was similar to that of the previous two inspections.
About 20 percent
of the mill scale remained and there were many pits with depths ranging up to a maximum of 0.122 inches.
The portion of the pile exposed to the
undisturbed but potentially more corrosive soil had practically no corrosion. the mill scale being intact over about 70 percent of the surface and the pits having depths of less than 0.010 inches.
Using equation
(1), the estimated reduction of cross section of the samples exposed to fill (0101, 0102, and Dl03A) is 23, 21, and 26 percent, respectively. Although the actual reduction in cross section is believed to have been in the 10 to 15 percent range, the possibility of a serious reduction of cross section in the fill soil over a period of 50 or more years cannot be disregarded.
It should be noted that none of these examined pilings
extended down to the water table.
3-12
3.5
Bonnet Carre Spillway, Norco, La.
No piling was accessible for
inspection at this site, and data are not listed in Table 1; however, information on a previous inspection was available.
A 122-foot, 12-inch .
65-pound test H pile was extracted at the site in 1950 after an exposure of 17 years. silt.
The soil consisted primarily of clays, organic clays, and
On examination, this piling was found to be free of corrosion,
with mill scale intact over the entire surface, except for the 3-foot section in the zone of the water table fluctuation.
In this area, between
elevation -1.5 and +1.5 feet, the metal was coated with a hard, lightcolored material, under which there was slight metal attack. 3.6
Stoplog Dam, Freeport, Texas (R2-l and R2-2).
This inspection is
of special interest because of the low soil resistivity which varied from about 300 ohm-em at the surface to 80 ohm-em at a 27-foot depth. Two sample sections were cut from an extracted piling; R2-l at a 4-foot depth, and R2-2 at a 26-foot depth.
The pH was about 7.3.
consisted of clays and silty clays.
The water table was about 5 feet below
the ground surface.
The soil
On this piling, which had been installed about
eight years, most of the corrosion was concentrated in the interlock area at the ground line.
Here, the interlock fingers were reduced in
thickness about 50 percent by
unifo~
metal attack, and there was
localized pitting with about 50 percent of the mill scale intact over the remainder of the surface.
In the area of very low soil resistivity
(25-to 28-foot depth), about 70 percent of the mill scale had been removed,
•
but the metal attack was limited to the mill scale thickness and there were no measurable pits. piling was negligible.
Corrosion over the remainder of the
About 85 percent of this piling was below the
water table. 3.7
Zia Pueblo Diversion Dam, San Ysidro, N. Mex. (Zl).
This site is
of interest because of the extremely wide range of soil resistivity which varied from 20 (the lowest value encountered in this investigation) to more than 10,000 ohm-em. with free lime dispersed
The soil consisted of sand and silty sand
thro~ghout.
The pH of the soil was 7.8.
The
top 3 feet were fill. The mill scale on the piling was 100 percent intact except in the 3- to 8-foot depth area (normal water table fluctuation zone) where about 40 percent of the mill scale had been removed.
There were only a few
localized areas where pit depths exceeded the mill scale thickness; however, all pits were less than 0.020 inches in depth. 3.8
Hartford Dike, · Hartford, Conn. (AB-1).
corroded of all the piling examined.
This was the most severely
The installation was exposed for
23 years to soil consisting of a conglomeration of clay, cinders, and shale.
The cinder content was as high as 90 percent at the 20-foot
depth.
The resistivity ranged upward from a low of about 2,000 ohm-em.
The pH varied from 6.2 to 7.8.
In the area where cinders were present,
the mill scale was entirely removed and deep pits covered approximately 15 percent of the surface.
The 3/8-inch piling had one
3-14
perfo~ation
-
a hole of approximately one square inch in area.
The major portion
of this piling was above the normal water table (about 25 feet below the surface); however, at high stages of the Connecticut River, the water level rises above the top of the piling, With equation (1), the estimated reduction in cross section is 45 percent and although the actual reduction is believed to have been about 20 to 30 percent, this amount of corrosion over a 23-year period must be considered as serious. It is believed that the severe corrosion experienced at this site can be attributed to the following: 1.
The extremely aggressive nature of the fill soil.
2.
The fact that the major portion of the piling exposed to this
soil was above the normal water table, 3.
The fact that the soil, in the area of severe corrosion, was
saturated annually for one to three months during high stages of the Connecticut River. 3,9
Fairbanks Test Piles, Fairbanks, Alaska (FA, FB, and FC).
The
Fairbanks, Alaska, piling deserves special mention because these piles extended into the permafrost region.
They were installed by the
u. s.
Army Cold Regions Research and Engineering Laboratory at the Fairbanks, Alaska, Field Station for investigating the corrosion of the pilings in the permafrost and also the possible loss of adhesion between the steel and ice as the result of slow chemical reactions •
•
3-15
Nine piles were extracted from three locations.
The results of
the examinations of five representative pilings are included in Table 1. The portions of the piles which were exposed to the permafrost region were entirely free of corrosion with mill scale 100 percent intact.
In
the freeze-thaw-zone, above the permafrost, corrosion was negligible. Mill scale was 50 to 90 percent intact and except for a few pits which had depths up to 0.028
inches, none exceeded the thickness of the mill
scale (about 0.020 inches). 3.10
~s
Corrosion Specimens.
described in para 2.3.1, above, sixteen
pipe specimens were buried at Sardis Dam, Grenada Dam, Bonnet Carre Spillway, and Berwick Lock at the start of the investigation.
About two
years later, two rod specimens, one buried and one driven, were placed at each of these sites to detect, if possible, a difference in the rate of corrosion between the rods.
Results of the tests are indicated in Tables
3 through 6. 3.10.1
Pipe Specimens.
Corrosion of the Sardis Dam specimens,
which were exposed to soil having characteristics (type. pH, and resistivity) similar to that which piling sample Cl02 was exposed, appeared to be of the same order of magnitude as that of the piling sample.
Likewise,
corrosion of the Grenada Dam specimens appeared to be approximately the same as that of piling samples DlOl, Dl02, and Dl03A.
However, at Bonnet
Carre Spillway and Berwick Lock, the specimens were much more severely corroded than the piling.
Specimens at these sites were placed in soils
3-16
•
with characteristics similar to that which the piling was exposed. At Berwick, all four specimens removed after 2 years of exposure had pits with depths exceeding the 0.090-inch maximum pit depth measured on the piling with eleven years of exposure.
At both of these sites,
the piling, except for the top two to three feet, was entirely below the water table. 3.10.2
Rod Specimens.
The two rods at Sardis Dam were equally
corroded, but at the other sites, the buried rod was the more severely corroded.
The
rela~ively
low rate of corrosion of the Berwick rods as
compared with the pipe specimens may be due to the fact that approximately 65 percent of the exposed portion of the rods was below the water table
most of the time.
At this site, the water table, except during infrequent
very dry periods, remains practically constant at a depth of about
2
feet.
The buried rod at Bonnet Carre Spillway corroded at about the same rate as the pipe specimens; however, the driven rod corroded at a rate of about 55 percent of this.
While no general conclusions can be drawn
from these limited tests, they do tend to substantiate the fact that steel corrodes. more rapidly in disturbed soils. 3.10.3
Polarization Curves.
The polarization curves quite accurately
indicated the corrosion weight losses for the pipe and rod specimens at Grenada Dam and for the pipe specimens at Sardis Darn.
While indicated
weight losses of the other specimens varied considerably from actual losses, the .results are probably within the accuracy to be expected from
3-17 •
TABLE 3
PIPE AND ROD CORROSION TEST
SPECI~ffiNS
SARDIS DAM
Specimen No.
Date Buried
Date Removed
Months Exposed
Weight (Gms)
Final Wt. (Gms)
Cl C2 C3 C4*
3/29/60
5/1/61
13
" " "
" "n
1435.0 1450.6 1429.5 1407.8
1430.9 1447.5 1426.0 1403.5
4.1 3.1 3.5 4.3
1433.9 1399.5 1411.5 1410.5
1427.9 1395.5 1408.7 1406.8
6.0 4.0 2.8 3.7
1401.4 1412.4 1410.8 1417.5
1396.8 1405.9 1404.5 1410.4
4.6 6.5 6.3 7.1
1379.6 1416.1 1421.3 1433.0
1371.9 1410.2 1415.5 1428.2
7.7 5.9 5.8 4.8
1983.0 31.1 1983.7 30.8
•
t.N I
.....
00
Cl3* C14 Cl5 C16
" " "
C9* ClO Cll C12
" "
C5 C6 C7 C8*
"
"
" " "
" "
4/18/62
" " "
"
25 "
"
"
"
"
4/21/64
" " " 7/12/67 " "
"
49 "
" "
87 "
" "
Wt. Loss (Gms)
Est. wt. Max. Pit Wt. Loss Loss Depth (Gms)** (oz/Ft2/Yr) (inches) 0.27 0.20 0.23 0.28
0.008 0.002 0.007 0.002
7.0
0.20 0.14 0.10 0.13
0.004 0.008 0.006 0.003
5.9
0.08 0.11 0.11 0.12
0.003 0.006 0.010 0.006
6.9
0.07 0.06 0.06 0.05
0.012 0.003 0.003 0.006
13.0 15.0
0.43 0.42
3.1
Rod Specimens (d) C21 * C22*
5/16/62
"
7/12/67
"
62
"
2014.1 2014.5
*Insulated lead wire attached to specimen **Estimated from polarization curves NOTE:
•
· 0.020 (a) 0.022 (a)
(a) Deepest pits at middle of rod (d) Driven rod
Effective exposed area of all specimens was 0.5 ft.2
•
TABLE 4 PIPE AND ROD CORROSION TEST SPECIMENS GRENADA DAM Spec1.men Date No. Buried
Date Removed •
Months Exposed
We1.ght (Gms)
Final 'Wt. (Gms)
Wt. Loss (Gms)
Est. Wt. Loss (Gms)**
Wt. L'os~
(oz/Ft /Yr)
Max. P1.t Depth (inches)
P.IPE SPECIMENS 01 D2 D3 D4*
w I
..... t.O
3/30/60
5/3/61
13
" " "
" " "
" " "
D13* D14 D15 · D16
" " " "
4/19/62
D9* D10 Dll D12
" " " "
4/22/64
D5 D6 D7
" " " "
7/13/67
DB*
" " " " " " " " "
25
"
" " 49
" " " 87
" "
"
1384.4 1425.6 1424.1 1379.2
1363.7 14-02.5 14-04.3 1362.5
20.7 23.1 19.8 16.7
1423.9 1421.9 1427.1 1423.5
1383.8 1385.8 1390.2 1383.8
40.1 36.1 36.9 39.7
1377.0 1417.1 1399.8 1373.7
1313.5 1359.8 1348.6 1325.1
63.5 57.3 51.2 48.6
1401.1 1381.1 1426.9 1406.6
1348.7 1328.3 1374.6 1351.2
52.4 52.8 52.3 55.4
59.1 64.4
1.35 1.51 1.29 1.09
0.039 0.034 0.041 0.032
48.1
1.36 1.23 1.25 1.35
0.059 0.069 0.061 0.061
59.0
1.10 0.99 0.89 0.84
0.058 0.078 0.044 0.070
60.1
0.51 0.52 o.51 0.54
0.082 0.067 0.060 0.102
52.3 55.8
o.s1 0.88
0. 065 (a)
14.3
Rod Specimens (d)D21* D221:
5/15/62 "
7/13/67
"
62
"
2015.8 2014.1
1956.7 1949.7
· * Insulated lead wire attached to specimen ** NOTE:
Estimated from polarization curves
(a) (d)
0.095 (a)
Deepest pits on bottom tip Driven rod
Effective exposed area of all specimens was 0.5 ft. 2
TABLE 5 PIPE AND ROD CORROSION TEST SPECIMENS BONNET CARRE SPILLWAY Specimen Date Buried No.
Date Removed
Months Exposed
Wei~ht
(Gms)
Final Wt. (Gms)
Wt. Loss (Gms)
Est. Wt. Loss (Gms)**
Max. Pit Wt. Loss Depth (oz/Ft2/Yr) (inches)
Pipe Specimens El E2 E3 E4*
(A
• N
0
4/6/60
" "
5/11/61 "
13
" "
"
" "
El3* El4 E15 ' El6
"
4/26/62
25
" "
",,
"
"
" " "
E9* ElO Ell E12
"
4/29/64
49
" " "
"
"
"
" "
E5 E6 E7 E8*
"
7/26/67 "
88 "
"
" "
"
" "
"
"
"
1415.7 1412.2 1407.4 1394.9
1403.1 1401.2 1395.7 1382.6
12.6 11.0 11.7 12.3
1412.6 1372.6 1387.5 1417.6
1391.7 1356.0 1370.6 1395.5
20.9 16.6 16.9 22.1
1372.0 1416.6 1413.8 1409.2
1331.5 1377.2 1373.7 1365.2
40.5 39.4 40.1 44.0
1392.6 1374.7 1418.4 1362.6
1324.2 1321.2 1356.3 1291.2
68.4 53.5 62.1 71.4
0.82 0.72 0.76 0.80
0.006 0.003 0.003 0.004
52.2
0.71 0.56 0.57 0.75
0.008 0.006 0.008 0.008
85.3
0.70 0.68 0.69 0.76
0.036 0.042 0.041 0.032
0.66 0.52 0.60 0.69
0.056 0.084 0.080 0.086
23.3
105.9
Rod SEecimens . (d)E21* E22*
0.008 (a) 0.37 53.2 26.8 1989.7 2016.5 0.068 (b) 0.67 48.4 51.9 1965.0 2013.4 " " " (a) Deepest pits at upper end * Insulated lead wire attached to specimen (b) Deepest pits at bottom tip ** Estimated from polarization curves . (d) Driven rod
6/21/62
7/26/67
61
NOTE: Effective exposed area of all specimens was 0. 5 ft. 2
•
•
•
PIPE AND ROD Specimen No.
Date Buried
Date Removed
Months Exposed
TABLE 6 CO~ROSIO~
TEST SPECIMENS
BER\\'! ~K LO~K
Weight (Gms)
Final Wt. (Gms)
wt. Loss (Gms)
wt.
Est. Loss (Gms)**
Wt. Loss ( oz/Ft 2/Yr)
Max. Pit Depth (inches)
20.1
1.25 1.02 1.23 1.25
0.034 0.027 0.038 0.017
Pipe Specimens Fl F2 F3 F4*
(,.a
• N
4/5/60 " " "
5/10/61 "
13 "
" "
" "
Fl3* F14 Fl5 Fl6
"
4/25/62
25
"
"
"
"
"
"
"
"
"
F9* FlO F11 F12
"
4/28/64
" "
" " "
.....
F5 F6 F7 F8*
" "
" "
"
7/25/67 "
49 "
" " 88
"
" "
"
"
1417.9 1401.4 1415.4 1429.5
1398.8 1385.8 1396.5 1410.4
19.1 15.6 18.9 19.1
1429.4 1414.8 1437.7 1410.9
1380.8 1371.5 1384.3 1363.4
48.6 43.3 53.4 47.5
79.1
1.65 1.47 1.81 1.61
0.094 0.145 (P) 0.110 0.095
1417.9 1400.2 1423.2 1433.8
1350.8 1328.4 1352.2 1364.0
67.1 71.8 71.0 69.8
128.2
1.16 1.24 1.23 1.21
0.116 0.145 (P) 0.145 (P) 0.145 (P)
1379.3 1419.8 1406.0 1409.6
1271.0 1314.6 1310.0 1284.4 \
0.081 0.145 (P) 0.082 0.145 (P)
0.008 (a) 0.088 (b)
108.3 105.2 96.0 125.2
164.2
1.06 1.03 0.94 1.23
20.4 28.5
29.7 46.6
0.28 0.40
Rod Specimens (d)F21 * F22*
6/20/62
"
7/25/67
61
"
"
2014.1 2013.0
1993.7 1984.5
(a) Rod uniformly corroded * Insulated lead wire attached to specimen (b) Deepest pits at bottom tip ** Estimated from polarization curves (d) Driven rod (P) Perforation Ncr.tE: Effective exposed area of all specimens was 0.5 ft.2 .
the tests conducted at the time intervals at which they were performed in this investigation.
The curves did indicate the relative
corrosivity of the soils and also correctly indicated the change in corrosion rate with time.
No correlation was noted between specimen
half-cell potentials and soil corrosivity; although, the potentials of the driven rods were. with a few exceptions, consistently more negative than the buried rod potentials.
3-22 •
4. 4.1
Piling.
CONCLUSIONS
The conclusions in regard to the corrosion of piling
may be summarized as follows: 1.
Little or no corrosion occurred on that part of steel piling
exposed to soil below the water table.
This was true irrespective of
the type of soil. 2.
Corrosion of piling exposed to undisturbed soils was negligible
or minor in most instances, and in all instances, it was much less than would be expected in similar disturbed soils. 3.
Except in unusual circumstances, such as heavy cinder concentrations
or unusually aggressive soils, piling exposed to fill in, or above, the water table fluctuation zone evidenced moderate, but not serious, 4.
corr6sion~
The area of maximum corrosion usually occurred in the zone of
water table fluctuation.
s.
In permafrost soils, corrosion was confined to the part of the
piling exposed to soil subject to annual freezing and thawing. The above conclusions apply only to piling in an underground environment. 4,2
Corrosion Specimens,
Corrosion specimen test results, while too
limited in scope to permit general conclusions, tended to confirm conclusion 2 above.
4-1 •
5.
BIBLIOGRAPHY
1.
Underground Corrosion, National Bureau of Standards Circular 579, Melvin Romanoff.
2.
Corrosion of Steel Piling in Soils, National Bureau of Standards Monograph 58, Melvin Romanoff.
3.
Current and Potential Relations for the Cathodic Protection of Steel in a High Resistivity Environment, Journal of Research of the NBS, Vol. 63C, No. 1, Jul-Sep 59, w. J. Schwerdtfeger.
4.
Laboratory Measurement of the Corrosion of Ferrous Metals in Soils, Journal of Research of the NBS, Vol. SO, No. 6, June 1953, RP 2422, w. J, Schwerdtfeger.
s.
Soil Resistivity as Related to Underground Corrosion and Cathodic Protection, Journal of Research of the NBS, Vol. 69C, No. 1, JanMar 65, w. J. Schwerdtfeger.
6.
A Study by Polarization Techniques of the Corrosion Rates of Aluminum and Steel Underground for Sixteen Months, Journal of Research . of the NBS, Vol. 65C, No. 4, Oct-Dec 61, w. J. Schwerdtfeger.
7.
Corrosion and Its Control, reprinted from the Oil and Gas Journal, Marshall E. Parker,
B.
Corrosion of Drainage Wells at Sardis Dam, Mississippi, u. s. Army Engineer Waterways Experiment Station Technical Memorandum No. 3-287, June 1949.
9.
Corrosion Tests of Metals, Sardis Dam, Mississippi, u. s. Army Engineer Waterways Experiment Station Miscellaneous Paper No. 3-86, April 1954.
10. The Unified Soil Classification System, u. s. Army Engineer Waterways Experiment Station Technical Memorandum No. 3-357, April 1960 (reprinted May 1967).
•
•
