Professional Documents
Culture Documents
J. Schwerdtfeger
(August 4, 1964)
Corrosion data disc ussed ill this paper arc based upon m eas ure ments mad e on abo ut
4500 specimens of commonly used plain wrou ght ferrous mat ri al wh ich had been buried
in back-filled Lre nches at 86 National Bureau of Standards sites for periods up to 17 years.
The soils ran ge d in resistivi ty from 50 to 54, 000 !1-cm a nd in p H f rom 2.6 to 10.2.
Maximum pit depths at 5 years of ex pos ure a re ta ke n from pi t depth- t ime cur ves and
the curves arc also extrapolated t o probable pit depths at 30 years for eac h of the 86 sites.
Furthermore, data O il t he s pecime ns are a dju st ed to m ax imum pi t depths that might be
expected on a lar ger area, equivale nt to t hat of the exterior surface of a 20 ft length of 8
in . ull coated wrought ferrou s pipe. Maximum pe netration ratcs a nd pit depths arc presen ted
with r es pect to soil resistivit.v a nd pH.
\Veight losses which rcs ulted from about 2 years of unde rg round ex posure a rc converted
to corrosion current densities a lld after this period of expos ure current densit ies arc calculated
from the slopes of weight loss-time curves for eac h site. On t he basis of r atios of protective
current t o corrosion curre nt obtai ned from polari zation curves on steel s pec imen s underground a nd in soils in the labo ratory, t he corrosio n current densities can be a dju sted to
approximate current densit ies necessary for cathod ic protection .
1. Introduction
Between the years 1922 and 1952 the Nationftl
Bureau of Standftrds exposed tbousttnds of metallic
specimens at numerous underground sites throughout the United States. Corrosion rates of O\Ter
300 vftrieties of protected and unprotected metals
and alloys were evaluated about every 2 year s for
exposw-e periods as long as 17 years. The accumulated corrosion data were compiled by Romanoff
[IV The engineering significance of early NBS
data pertaining to some co mmonly used wTought
ferrous materials from the original 47 test sites WftS
discussed in considerable detail by Logan [2]. The
present paper is similar but includes data from about
40 additional sites and, based on the total available
data, offers s uggestions 011 the current densit ies
probably required for cathodic protection.
In this paper, in addition to extrapolating the
average maximum pit depths to 30 years as Logan
[2] did, the weight losses of the specimens are converted to corrosion current densities. From these
data the approximate current densities required for
the cathodic protection of bare undergro und surfaces
can be estimated. Although, based on NBS data,
there is no precise relationship between the corro si \Tity and resistivity of soils, a general relationship is now revealed by the total ftccumulated data.
This is also indicated by the data of Scott [3].
Recently, Schaschl and .Marsh [4] discussed t he
effect of resisti \rity on the corrosion rates of steel
in aqueous environments and showed t h at for
resistivities above 400 Q-cm, the corrosion rate bears
an inverse relation to the resistivity of the electrolyte. Thus, a study is undertaken of the influence
of soil resistivity, a meaSLll'ement readily made in the
field. The co mbined relation of pH and soil resistivI
Figures in brackets indicate the literature refe rences at the end of this paper.
71
2. Corrosiveness of Soils
The corrosiveness of a soil might be defined as its
destructi \Te or deteriorating effect on a metallic
surface, ftS mea ured by weight loss and pit depth.
A soil Cftll be potentially cOl'l'osi \'e and yet have
a negligible effect on phin ferrous mftterials. This
was rece ntly shown by Romanoff [5] as a result of
his examination of steel pilings (pulled) exposed
from 7 to 40 years in a variety of soils some of which
hadresisti \'ities as low as 300 Q-cm. No appreciable
amount of corrosion was obselTed on pile surfaces
adjacent to soils into which the piles had been
dri\Ten below the water table, referred to as "undisturbed" soil. This type of exposLll'e is in contrast
to the soil exposure of metftls in back-filled trenches,
referred to as "disturbed" soil. On driven piling,
t he corrosion rate decreases rapidly as the initially.
available oxygen is depleted by the corrosion process
and the accessibili ty of more is limited. In distmbed soils oxygen is more readily replenished and
the soil is only very slowly restored to its natural
state. This gives rise to differential aeration and
then the corrosiveness of the soil depends largely on
its ph ysicftl and chemical properties. Properties
of the soils to be considered have already been
described [1].
The most severe corrosion usually takes place
at low elevations in poorly drained disturbed soils,
such as clays and tidal marshes, where, although
also poorly aerated, the differential-aeration effects
are large because of soil shrinkage. Then too, salts
accumulate in these areas and increase the conduc-
(1)
(2)
(3)
~here
0.4
V>
E 1000
.:!2100 0
vi
a::
>-
..
It)
f-
<t
100
f<l.
I ena::
........
f<J1
W
<l.
W
W
0
'E
f-
<t
10
10 '
10'
f<J1
W
<l.
W
0
10 '
100
f<l.
10
-... ,.
to
....
>-
- C.
10
10
10'
10'
10'
105
10 '
RESISTIVITY, .a-em
FIG U RE
FIGUR E 1.
Avera(le of max'';mum pit depths on fTom 6 to 14
f errous specimens bltried in each of 86 underground soi l sites.
Pit de ptlls are take n from the pit depth versus time curves at 5 years.
10'
C!!
10 0
......
<J)
a::
>-
10
E
<t
- .,-'
.. - -
-. ,...
.-.
- .-.. - ..
It)
w 10'
>-
r-
a::
w
lL
<t
f-
r-
<t
a::
i.
<l.
X-
<t
:::2'
.1
10
a::
<J)
a.
w
w
10'
10'
100
vi
a::
o
10
>-
'b
co
.,0 ""
. .:,....
fu.
<t
w
'"a::
f-
<!
a::
:i.
SOIL RESISTIVIT Y
500 .Il.-em
.<
o
500 3000
3000- 60 0 00
a.
X
<t
::E
.1
10
12
pH
FIG U R E
3.
7 53 - J S O~65 --~G
'f'
10'
10 '
FIGURE
'>-
10'
RE SIS T IVI T Y, .a - em
"-
1
10 '
F I G UHE
10'
f-
10
10
<t
lO' -
. ...
ff-
.... .-........ -.
".,
~ e
f-
..
vi
a::
.:!2
E
73
TABLE
1.
Time- 5 yrs
Soil site
No.
p'
pH
Interna l
drainage d
fl-cm
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
:ll
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
51
53
55
56
58
59
60
61
62
63
64
65
66
1215
684
30000
6670
1345
45100
2120
350
2820
7460
11000
3190
290
3520
489
8290
5980
1410
) 970
2870
2370
5) 50
278
11400
1780
2980
570
408
1270
1300
20500
5700
800
4900
2060
11200
11200
38600
7440
970
1320
13700
60
1000
2(i3
1500
1770
190
17790
5210
406
712
1660
218
943
6920
84
62
148
232
7.0
7.3
5.2
5.6
7.0
5.9
4.4
7.6
6.8
6.6
5.3
7.1
9.5
6.2
7.5
4.4
4. 5
7. 3
4.6
7.5
6.2
4.9
9.4
4.5
7.2
7.3
6.6
6. 8
4.2
7.0
4. 7
7.3
6.8
6.7
7. 3
4.5
3.8
4.5
5.6
6.0
5.5
4.7
3.1
5.8
7.4
7. 0
7.6
6.2
4.8
5.8
7. 1
4.8
5. 6
2.6
6.8
4. 5
6.9
7.5
8.0
8.0
P
P
P
P
G
G
F
F
P
F
F
G
G
P
F
G
F
G
J<'
G
VP
VP
VP
P
G
G
VP
F
G
G
G
J<'
F
l'
VP
G
P
VP
G
G
VP
VP
VP
P
P
J<'
VP
J<'
G
l'
A ssumed
Mils
Mils
48
44
68
34
40
20
27
54
35
41
64
50
65
80
52
56
31
43
48
26
56
56
105
20
42
60
33
83
73
27
33
37
43
33
22
42
53
26
34
58
50
79
80
55
40
75
17
131
58
62
66
57
28
40
44
60
40
134
82
H2
96
82
134
72
79
39
47
100
79
98
124
97
161
169
83
98
56
72
77
63
103
83
202
45
63
114
74
137
169
48
75
81
234
8.8
VP
110
886
7. 0
P
26
Ie
261
7.3
104
]e
103
7.3
88
}'
81
7.3
97
8500
4.6
G
84
28000
4.8
G
51
25000
4. 8
G
63
54000
4.8
G
74
44400
4.8
G
68
497
8. 4
P
83
531
10.2
95
P
51
7. 3
P
102
149
7.4
P
185
Ie
102
7.4
177
320
7.1
VP
89
3450
6.9
G
70
320
9. 2
F
107
106
8.5
P
123
273
7. 3
F
160
10800
4.7
G
70
16400
4. 8
G
22
Ie
122
552
7.4
52
) 23
6840
4.1
P
45
124
1160
4.4
P
62
125
5770
3.9
P
52
Average of the deepest PIts on 6 to 14 speCImens (a rea, a pproxImately 0.4 fF)
for each soil.
b Romanoff [1] , table 6.
52
54
101
102
103
104
105
106
107
108
109
110
III
112
113
114
lL\
116
117
118
119
]21
e
d
Romanoff [1], fro m pi t depthtime c ur ves u sing data in tables 13, 14. 15. There
were from 6 to 14 specimens involved in each removal from a given soil and from
3 to 6 removals, the 3 r emovals being for t he 100 series of soils.
o
'rime- 3~
yrs
Specimen
Specimen
area Al e
areaA 2
9.)
69
99
67
131
54
69
115
78
125
201
HI
78
181
52
266
118
126
134
H6
57
81
89
122
81
272
166
228
S pecimen
rect.angular h
co-ord
Mils /vr
6.0
1.6
1.3
6.7
1.4
0.28
2.2
4.4
2. 1
1.0
1. 5
1.3
6. 0
3.2
0.84
2.3
1.3
1.1
2.7
4.4
4. 4
1.6
8.3
0.75
1.0
0.61
3.0
17.0
8. 1
3.2
1.3
2.7
8. 4
4.1
0.60
1.0
3.2
0.7
5.4
8.1
4. 7
4.0
12.2
2.5
7.3
4.2
2. 0
13.6
1.4
1.3
15.0
12.6
6.0
5.0
4.3
3.0
6.7
5.7
6. 9
6.0
l\lils
198
84
101
202
75
27
82
164
88
66
102
83
215
160
73
114
64
71
116
136
110
96
312
39
67
77
108
508
276
107
66
105
253
136
37
67
133
44
169
261
168
179
285
118
223
180
67
471
93
95
441
372
178
165
152
135
208
2i7
255
262
Specim en
log.co-o rd
lV il.
190
72
97
165
86
26
76
130
82
70
90
75
125
165
84
100
57
76
92
115
120
85
310
32
65
67
100
420
235
95
69
81
270
92
43
58
103
48
140
185
)30
135
270
95
240
140
60
550
87
90
390
305
260
145
160
HO
155
240
195
200
Assum ed
areaA2 j
Mils
378
134
192
351
170
52
131
244
185
167
174
]4.\
311
348
134
175
103
128
148
278
220
125
597
72
98
118
223
693
545
170
156
177
598
J93
J93
92
254
99
284
365
203
214
678
191
469
337
183
1117
177
183
792
619
528
294
325
223
314
487
396
406
873
223
18. 0
560
430
203
154
100
53
5.1
274
2H
3.0
179
135
376
185
4.
4
198
179
420
853
197
17.9
644
244
120
171
2.8
154
117
0.9
74
58
104
199
98
123
128
2. 4
294
182
150
4.3
145
203
2.2
123
100
138
355
218
175
169
5.4
609
300
193
H.7
388
650
11. 8
397
320
207
2530
1250
35.0
1060
375
2030
35.0
1052
1000
360
426
210
5.1
217
181
244
142
1.8
115
120
670
330
217
13.7
450
437
215
7.6
313
250
1240
795
610
25. 4
325
189
93
142
1.7
113
28
.\7
0.4
32
45
77
156
2. 4
112
106
78
158
1. 6
85
91
233
152
115
3.6
126
223
3. 1
110
106
130
r A 2= 45.16 ft', eq m va lent to a 20 ft length of 8 ill. pIpe . bOe eq (2) . Fo r soils
!through 47, u sed ex ponents a, from Logan. Ewing, and Denison [7]. For a!lother
soils, u sed exponent a=0. 15, from Schwerdtfeger (8).
g Based on best straig h t line after 5 years on pit depth-time curves on linear
coordinates.
h Extrapolated pi t depth-time curves.
i Extrapolated pit dept h-tim e curves.
j From log-coordina tes (i) and u se of eq (2).
~
74
I
1
are,Ls a nd is therefore indicative of a uniform environmen t. LaLer, Scott offered a theoret ical basis for
his empirical relationship [10]. Soil resistivity
m etts urements made by him along pipeline rig hts-ofway together with leak r ecords on th e pipelines show
that leaks occur in the places of low resis tivity and
alm ost invariably at resisti lTit ies below the value
shown by 50 percent probability on t he log r esistivity-prob ability curve.
TABLE
2.
Soil
site
N o. b
p '
-----0- 2 y r
After 5
yr e
-------- ---
0- 2 y r d
N O, b
After 5
yr e
- - - - - - - - - --
mA /ft '
2.0
1.1
0.41
1. 3
1.1
51
53
55
56
58
190
17i90
5210
406
712
mA/fl'
mA/fl '
45100
2120
:350
2820
74(;()
0.28
1.1
1.8
23
2.9
0. 26
0. 59
1. 8
1. 0
1.1
59
(i3
J660
218
9] 3
6920
84
2.5
8. 4
3. 4
5.3
4. 7
1.6
5.6
1.6
1.1
2.5
12
13
14
15
11000
3 190
290
3520
489
0.90
l. 3
5. ()
l. 7
3. G
0.34
0.8 1
:l.6
l.1
1. 8
64
65
6G
52
54
62
148
232
Zl4
886
20.3
9.4
12.5
(i. 2
1. 5
12.7
1.4
2.8
6.2
1.1
16
L7
J8
I!!
20
8290
5890
141 0
1970
2870
3 G
:l. 1
1.2
2.0
2. 9
1. 7
1.1
0.56
0.5:1
l. 5
101
102
103
I(}I
105
261
103
81
8500
28000
7.0
7.2
6.9
4.6
.5.3
f) . Ii
21
22
2:3
24
25
2370
5 L50
278
11400
li80
4. I
2. 9
J I. 2
062
1. :3
1. 5
0.72
4.7
0. 22
0. 65
106
107
108
J09
11 0
25000
54000
44400
497
53 t
:1. 9
:3. G
5.2
8.7
6. 9
26
27
28
2n
:l0
2980
408
1270
L300
2. 2
2.0
5.9
5.9
2.8
O. (lS
l. (i
5.5
4.2
1. 2
III
LI 2
11 3
J 14
115
51
149
102
:320
:1450
9.2
11. 9
J3.5
4.8
2.8
5. !)
7.5
9. !I
:1.8
0.411
31
:12
33
34
35
20500
5700
SOO
4900
2OGO
1. 6
1. :3
:1. 4
2.9
1. :3
0.50
0.84
3. 4
1.1
0.1 2
11 (;
LL 7
LL S
119
J2 L
320
106
273
10SOO
1&100
10. I
J2.8
8.7
2. S
1. 4
4. (l
:16
:17
1. 7
3.7
O. G2
2.7
3. 4
0. 40
1. 9
0.53
1.2
I.G
122
IZ1
124
125
552
6840
II (;()
5770
3. I
4.8
9.0
5. ()
1.4
2. 0
I. -I
1. :1
40
11 200
11 200
:)S(iOO
7410
970
41
42
43
44
45
1:320
13700
GO
1000
263
2.0
4.5
4.8
I. L
3. I
0. 91
3.0
:1.9
1. 0
2. S
46
47
1500
L?iO
2. :1
1.0
20
o 97
5
6
7
75
p '
mA /fl'
site
f!-cm
12 15
(lS4
30000
6670
1345
1
2
3
Corrosion rate
Soil
9
10
11
as
:19
570
3. i
3. 1
2.8
3. I
1. 8
fl -cm
(;()
61
62
14.2
5.0
4. 4
6.5
5.2
3.6
0.50
041
6.5
3. 4
2.2
f), li
I. .)
o.:n
1.!)
l. 0
O. .I!!
2. (;
n.2
5. :~
5. :{
1. 8
o. IU
!~ Calculat ed from we i ght lossLimc curves on data b y ilom an ofr II I t1 s ill ~ I i'ara~
d a.y's law, assumin g 100 pcrcent co rro sion eni ciency a nd Lha. t th e iron is cti va lent.
For weight losses, sec Roma ll oO' 11] , t a.b les J3, 14, Hnd 15.
b Rom an off [11 . t able 6.
e Soil reSistivit y, sa turated soil at. GO OF .
d Based on first relll ova l from soil , u su a ll y 1 to 2 years.
e Ba sed on the best stra ight line on the wrig ht loss-tim e cur ves afiel' 5 years.
100
......
N_
::::<l
.
.. ..........
"I.... .
.
..,'
.::..... . ...
r-:
10
3 rd
a::
a::
.......
::>
0
Z
0
e...... :
a::
a::
, I
10 '
, I
la'
10 '
10 '
FIG U RE
100
N
"<l
10
w
a::
a::
::>
0
.....-._:
I .. :
e. .. .,_
., .. . .- ...
....
. e.!.
ifi
0
a::
a::
0
0
_:.~
.1
10'
10
10'
10 '
10'
RESISTIVITY. D-e m
FIGU R E
SOIL RESISTIVIT Y
:: 12
<l
"-
E 10 ~:
500 3000
3000- 6 0 0 00
w 8
a:
a:
(3 6
.'t
z
Q 4
en
o
a:
a: 2
o
"
10
10
12
14
16
~
FIG U R E
f--'
:r--o--
MONTH
3 2 nd MONTH
C.R.'S.l mA / ft 2
IP
10
C.R.'6.S m A / ft 2
.1
8t h
..
I -
ifi
DA Y
G. R., 2.5 m A /f t 2
II
12
o anodic
. cathodi c
pH
FI GURE
8.
5. Conclusions
76
TABLE
Maximum penetration
Soil resistivity
n-cm
50 to 500 .........
500 to 3000 .......
3000 to 54,000.....
N umber
of soil
sites
ra.te
Min
Corrosion
current b
Avg
Max
Min
Avg
,"lax
mils/yr
26
28
32
60
0.. 84
.28
4. 2 1
1 11.4
2.2
35.0
12. G
G.7
1.. 12
4
. 19
1m1./t
I. 6 1
12.7
6.2
0.99
3. 0
Calcu lated from t he specimen pittiug rates after 5 years (table 1).
b Calculated from t he corrosion rates (currents) after 5 yea rs (table 2).
6 . References
[I] Melvin Romanoff, Underground corrosion, N BS circ.
579, U.S. Government Printing Office, Washington ,
D. C. (1957).
[2] Kirk H. Loga n, E ngin ee ring significance of National
Bureau of Standards soil-corrosion data , J . R es.
NBS 22, 109 (1939) RP1l71.
[3] Go rdon N. Scott, The di stribu tion of soil conductivities
and so me consequences, Corrosion 140, 396t (August
1955) .
[4] Edwa rd Schaschl an d Glenn A. Marsh, Some New
Views on So il Corrosiotl, :Materials Protection 2,
8 (Nove mber 1963) .
[5] Mcl vin Romanoff, Co rrosio n of steel pilings in soils, J.
Res. N BS 66C (Eng. and In str. ) No. 3,223 (July-Sep t .
1962).
[6] G. N . Scott, Adjustment of so il-co rrosion pit d ep th
meas ure ments for s ize of sa mpl e, Proc. American
Petroleum Instit ute 140, No.4, 204 (1933).
[7] Ie H . Loga n, S. P . Ewing, and 1. A. D enison, Soil
co rros ion test in g, Sy mp osium on Corrosion Testin g
Procedures, AST1VI (J 937).
[S] ' V. J. Sc hwerd tfeger, Laboratory meas ure men t of the
co rros ion of fel'l'ous metals in SOils, J. Res. N BS 50,
329 (June 1953) RP2422.
[9] Gordon N. Scott, Dist ribu t ion of soil conductivity an d
its relation to underg round corros ion , J ournal A W'V A
52, 37S (Ma rch 1960) .
[10] Gordon N . Scott, A t heory for t he probability distribution of soil cond uctivi ty, COl'l'osion 18, 25lt (Jul y
1962).
[11] W. J . Schwerd tfeger and O. N. McDor rnan, Po tential
and current req uirements for t he cathodic protection
of steel in so ils, J . R es. NBS 407, 104 (Au gust 195 1)
RP2233 ; Co rrosion 8, 391 (1952).
[12] W . J . Sc hwe rd tfeger and O. N. Mc Do rman, :Meas urement of t he cor rosion ra te of a metal from its polarizin g characteristics, J . Electro che m. Soc. 99, 407
(1952) .
[l 3] 1. A. D enison, Electrolytic bc' hav ior of ferrous and
non ferrous metals in so il-co rros ion circui ts, Trans.
Electrochem. Soc. 81,4:35 (1942).
[14] Robe rt Pope, Cell curren ts and poten tials, Cor rosion 11,
IS9t (April 1955).
[15] E. Schasch l a nd G. A. Ma rs h, The effect of dissol ved
oxygen on co rrosion of steel and on c urren t req uired
for catho d ic protection, Co rrosion 13, 243t (Apl'iI1957).
[16] 'vV. J . Schwerdtfeger, A Study by polarization techn iques
of the co rrosion rates of alumin um and steel und ergro und for sixteen mont hs, J . R es . NBS 65C (Eng.
and In str.) No.4, 271 (Oct.- D ec. 196 1).
[17] E. II . Ardahl, Corps of Enginee rs, U.S. Army, Lower
Mississ ippi Vall ey Di vis ion , Vicksb urg, M iss. (private
co rrespondence) .
[IS] 'V . J . Schwerd tfege r, Current and potential relations for
t he cathodic protection of steel in a high resistivity
environment, J. R es. NBS 63C, 37 (Jul y- Sept. 1959);
Corrosion 16, 209t (May 1960).
77