CRITERIA OF SOIL AGGRESSIVENESS TOWARDS

BURIED METALS. 11*.ASSESSMENT OF VARIOUS SOILS

By G. H. BOOTH,** A. W. COOPER** and PAMELA M. COOPER**

The res.ultsof a five.-yearstudy of,the properties of 87 soils have shown that a reliable indication of the
aggressIvenessof s01,1towar~s bune? ferrous m~terials can,be obtained from ~easurements of resistivity
and of redo:,-potentIa1. SoIls consIdered,by th!s means to be on the borderlIne between aggressiveand
non-aggressIvemay be resolved by a consIderatIOnof the water content. Critical values for the various
parameters are suggested.

Introduction of these suggests that the aggressive sites may be characterised

In a previous paper,l we have described field and laboratory by the possession of either a mean resistivity of less than
tests of a simple practical nature to measure various soil 2000 Q-cm or a mean redox-potential of less than 0·400 V
properties (physical, chemical and microbiological) that might (N.H.E.) when corrected to pH = 7 (0'430 V if the soil is
be expected to have some bearing on the aggressiveness of the predominantly clay). Using these criteria, 38 of the 41
soil towards buried ferrous metal. In this paper, the results known aggressive sites would be predicted as aggressive and
on the application of these methods to a variety of soils over 16 of the 18 known non-aggressive sites would be predicted
a long period are reported, and discussed in the light of the as non-aggressive.
known corrosion history of the soils concerned. This leads Application of the same rules to the remaining 28 soils for
to a discussion of a scheme of testing for the prognosis of which advance knowledge of the corrosion behaviour of iron
probable corrosion risks to be expected in an unknown soil. and steel had not been obtained, gave a division into 6 sites
predicted as being aggressive and 22 as non-aggressive.' Sub-
sequent inquiry of authorities in the appropriate areas respon-
Methods sible for the maintenance of underground service using
Various public authorities with an extensive knowledge of ferrous materials (e.g. local water and gas undertakings)
the behaviour of various iron and steel structures (e.g. pipe- provided confirmation of these predictions in every case
lines) in the soil were approached and asked to provide except one, for which no information was forthcoming.
access to sites, the corrosion history of which was known. Thus application of the proposed rule gives the correct
87 soils from sites in which serious corrosion problems had prediction of aggressiveness in 81 cases out of the 87. The
arisen or from sites in which no trouble had ever been 6 'non-conforming' cases will be considered in detail later.
experienced were selected, the choice being largely on the Figs 1-3 show the results of the survey plotted as histo-
basis of accessibility, but with as wide a coverage of England grams using a division into aggressive and non-aggressive
and Wales as was possible. Some soils were included for sites based on our predictions. In Figs 1 and 2, the aggressive
which no corrosion history for iron and steel was ascertained sites have been subdivided according to whether the aggressive-
in advance. Each site was visited at approximately four- ness was attributed to low resistivity or to low redox-potential.
monthly intervals over a period of five years, field measure- These two histograms show the location of the critical limits.
ments were made in situ and soil samples were taken for It should be noted that the limiting value of 2000 Q-cm for
laboratory examination. The tests applied have already the resistivity is in very close agreement with the value sug-
been described in detail,l and were continued until each site gested by Schwerdtfeger2 and that the value of 0 ·430 V as a
had been surveyed once in each calendar month. critical one for the redox-potential approximates closely to the
A complete list of the sites, together with a brief description minimum value for non-aggressiveness put forward by
of the soils and their corrosion history, is given in the Starkey & Wight.3
Appendix. Fig. 3 shows histograms for water content, soluble iron
content and hydrogen uptake. The first of these has some
Results and discussion practical value in that it may be used to resolve a 'borderline
case', i.e. a case in which the resistivity is very close to
Typical results for the average taken of 12 individual
2000 Q-cm or in which the redox-potential is within the
measurements (one in each calendar month of the year) for
all the tests at two of the sites have already been given.l
o ·400-0·430 V range. In such a borderline case, the soil
can be considered aggressive if the water content is greater
Similar results are available for all the 87 sites.
than 20 % and non-aggressive if it is less than 20 %. This
On 59 of the sites, some knowledge was already available
critical value of 20 % for the water content of a soil has also
regarding the corrosivity towards iron and steel; 41 sites
been cited by Tomashov & Mikhailovsky4 as being the value
were designated 'aggressive' (i.e. where some maintenance of
at which 100 % of a buried metal surface becomes electro-
buried structures had been necessary) and 18 'non-aggressive' chemically active.
(i.e. where no trouble had been experienced). Consideration
Of the three 'borderline' soils encountered in this survey,
one was taken from a known non-aggressive site (No. 20)
* Part I: preceding paper and prediction as non-aggressive on the proposed rules was
** Metallurgy Division, National Physical Laboratory Tedding-
ton, Middlesex ' confirmed by the low water-content. A second (No. 10) was
Manuscript received 27 January, 1967 known to be non-aggressive and, although from resistivity it
Br. Corros. J., 1967, Vol. 2, May
110 Booth et al.: Criteria of Soil Aggressiveness towards Buried Metals. II.
25
(0) (a)

10 20

5 15

0 10
m.... (b)
in
II)
~ 5 •...
UJ
5
0:: V;
UJ
CD
::E ~
::> 0::
z UJ
0 '~ 0
(c) ::>
z (b)

10 5

5 o
(e)

5
50 75 100 125
RESISTIVITY, n - em x 10-2

Fig. 1. Electrical resistivity of soils o

(a) Non-aggressive (b) Aggressive-probably due to bacteria - 0·1 0 0·1 0·2 0·3 0'4 0·5 0'6
(c) Aggressive-not due to bacteria
~EDOX - POTENTIAL. V N.H.E. CORRECTED TO pH:: 7

Fig. 2. Redox-potential of soils

would be reckoned aggressive, the assessment would be
(a) Non-aggressive (b) Aggressive-probably due to low resistivity
reversed on consideration of the water content. The third (c) Aggressive-with high resistivity
case (No. 35) was one in which the North Thames Gas Board
had experienced corrosion but which we would class as non-
aggressive even when water content is taken into account.
There may be here some special consideration of which we per se, although they are required for the corrected com-
are not aware. putation of the redox-potential.
The other two histograms plotted in Fig. 3 give information Of the 6 sites which behaved in a manner other than that
of more limited application, but may be of some value in predicted, two (Nos. 10 and 35) have already been considered.
confirming a prediction of aggressiveness. All soils with a Two others were assessed by us as aggressive but the absence
mean soluble-iron content of greater than 120 pg/g or a of corrosion trouble at one (No.3) could be attributed to the
mean hydrogen uptake of greater than 20 ml/30 g soil/14 days pipeline being buried not in the natural soil but in a made-up
were aggressive. It is probable that both these parameters embankment on top of the soil. At the other (No. 62), no
are directly related to the redox-potential of the soil and not ferrous material was in the vicinity. The remaining two sites
at all to the resistivity. The values of these parameters are of (Nos. 57 and 59) were predicted to be non-aggressive, but
no value as confirmatory evidence for a prediction of non- corrosion had been reported by the West Midlands Gas
aggressiveness. Board and the Midlands Electricity Board, respectively.
The other tests (the 'bottle' test and the viable count of Both the sites were on housing estates with a great deal of
sulphate-reducing bacteria) are of no value at all in making a made-up -ground interspersed with the virgin soil and were
prediction. The mean value for the 'bottle' test over 5 years unlikely to be typical.
of testing for all the sites was 10-12 days for the onset of The results of the complete survey are summarised in
blackening in the absence of a redox-poising agent and Table I. Each site is listed with its corrosion history (with an
3-4 days in the presence of L-cysteine. The test merely indication of whether this was ascertained before or after the
indicates the presence of the organisms and it is significant assessment) and with the prediction of its aggressiveness
that the bacteria were present in everyone of the 87 soils toward ferrous materials determined by consideration of the
examined. Although the viable count of sulphate-reducers mean value of resistivity and redox-potential from twelve
was noticeably greater (in some cases by several orders) in measurements, one from each month in the year. The
soils of low redox-potential, this provides no additional table also indicates the number of times that a different
information of value. This also applies to the pH values prognosis (not necessarily a wrong prognosis) would have
Br. Corros. J., 1967, Vol. 2, May
 Booth et al.: Criteria of Soil Aggressiveness towards Buried Metals. II. 111
TABLE I
Corrosion history and prediction of aggressiveness of 87 soils

Number of times prediction Number of times prediction

would have been different would have been different
Number History Based if based on a single History Number History Based if based on a single History
of site Prediction Prediction
I on determination of II of site I on determination ,of II
--- --'- --- -- ------
R E R+E R E R+E

1 N N R,E 1 1 2 45 A A E 0 7 7
2 N N R,E 0 0 0 46 A A E 0 0 0
3 N A E 2 2 2 47 A A R,E 0 0 0
4 A A R 0 3 0 48 A A E 0 0 0
5 A A R 0 4 0 49 A A R 2 0 2
6 A 'A R 1 0 1 50 N N R,E 0 0 0
7 N N R,E 6 1 7 51 A A E 3 0 0
8 N N R,E 0 1 1 52 A A R 0 4 0
9 A A R,E 1 7 1 53 N N R,E 0 1 1
10* N A(N) R 4 2 3 54 A A E 0 3 3
11 N N R,E 0 0 0 55 A A E 2 2 1
12 N N R,E 1 4 5 56 A A R,E 2 0 0
13 N N R,E 0 1 1 57* A N(N) R,E 0 2 2
14 N N R,E 1 2 2 58 N N R,E 0 2 2
15 N N R,E 2 1 3 59* A N(N) R,E 0 0 0
16 N N R,E 0 1 1 60 X N R,E 0 0 0 N
'17 A A R,E 4 2 2 61 X N R,E 0 0 0 N
18 N N R,E 0 2 2 62 X A E 0 0 0 X
19 N N R,E 0 2 2 63 X A E 0 6 6 A
20* N N(N) R,E 5 2 6 64 X N R,E 0 1 1 N
21 A A R 2 2 2 65 X A E 0 2 2 A
22 A A R 3 4 3 66 X N R,E 0 0 0 N
23 A A R,E 0 4 0 67 X A E 0 2 2 A
24 A A R,E 1 6 1 68 X N R,E 0 0 0 N
25 A A E 3 0 0 69 X N R,E 0 0 0 N
26 A A R 2 2 1 70 X N R,E 2 2 4 N
27 A A R 1 0 1 71 X A R 1 1 1 A
28 A A R,E 2 5 1 72 X N R,E 0 0 0 N
29 A A R,E 3 5 2 73 X N R,E 0 0 0 N
30 A A R 0 3 0 74 X N R,E 0 0 0. N
31 A A R,E 0 6 0 75 X N R,E 0 0 0 N
32 A A R,E 0 2 0 76 X N R,E 0 0 0 N
33 A A R 6 1 6 77 X N R,E 0 0 0 N
34 A A R 0 5 0 78 X N R,E 0 I
0 0 N
35* A N(N) R,E 2 2 3 79 X N R,E 0 0 0 N
36 A A R 1 4 1 80 X N R,E 0 0 0 N
37 A A R 0 3 0 81 X N R,E 0 4 4 N
38 A A R 0 4 0 82 X N R,E 0 0 0 N
39 A A R 1 2 0 83 X A R 7 3 7 A
40 A A R 1 3 1 84 X N R,E 0 0 0 N
41 A A E 0 4 4 85 X N R,E 0 0 0 N
42 A A R,E 6 3 3 86 X N R,E 0 0 0 N
43 A A R,E 0 5 0 87 X N R,E 0 0 0 N
44 A A R,E 3 1 0

A-aggressive to iron and steel History I-ascertained before survey

N-non-aggressive to iron and steel History II-ascertained after survey
X-unknown *-'borderline' cases. Alternative prediction in brackets indicates that
R-resistivity deduced before consideration of water content.
E-redox-potential

Br. Corros. J., 1967, Vol. 2, May

112 Booth et al.: Criteria of Soil Aggressiveness towards Buried Metals. II.
occurred had the assessment been based on a single measure- Conclusions and Recommendations
ment of resistivity or of redox-potential or of both. Although It is concluded that a reliable estimate of the aggressiveness
such a procedure would not be recommended, cases may of a soil towards ferrous metal buried therein can be made
occur where such a preliminary prediction is desirable and it from a study of the resistivity and the redox-potential. Soils
is of interest to consider these figures. Table II summarises of low resistivity may be expected to be aggressive by virtue
the errors to be expected if a single measurement of the two of the ability to allow a substantial flow of corrosion currents.
parameters was taken from those available, and gives some Soils of low redox-potential may be expected to provide a
indication of the degree of confidence that may be placed on I suitable environment for the proliferation of sulphate-
a prediction based on these. reducing bacteria and hence either to support microbiological
There did not appear to be, any regularity in the variation
of any measurement at anyone site from month to month.
Fig. 4 shows examples of the way the redox-potential varied 0·6
with time of testing for a range of the soils. The significance
of the O·400 V-value is very marked.
0'4

30 0·2
(Q) (It)

0
20
r--
II -0'2
:r:
a.
10 0
.- 0'"
c
.-
UJ
U
UJ
0:: 0·2
0::
50 0 25 50 75 0
U
WATER CONTENT, %
cJ)
~20 ui
x
iiS :i
LL. > -0'2
0
a: .1" (t)
~1O
<
i=
~ z 0·6
~ .-
UJ

0
a..
I
x 0'4
0 0
C
0 100 200 300 0 100 200 300 itOO UJ
0::
SOLUBLE IRON, I-'g/g
0·2
20
(a) (b)

0'6

10
0'4

0'2
20 it0 60 0 20 itO 60 80 J F M A MJ J AS 0 N D
HYDROGEN UPTAKE, cm3 Fig. 4. Variation of redox-potential of soils
Fig. 3. Water content, soluble iron content and hydrogen up-take (a) Aggressive soils (clays) x- x Fortis Green .-. Morriston
(b) Aggressive soils (not clays) x- x Yateley .-. Haverfordwest
of soils (c) Non-aggressive soils (clays) x- x Malden .-. Rockhill
(a) Non-aggressive (b) Aggressive (d) Non-aggressive soils (not clays) x- x Hurley .-. Passfield

TABLE II
Errors to be expected if a single measurement of the two parameters was taken

Wrong diagnoses (out of twelve) from a single measurement of both Rand E

Nature of site Total number
0 1 2 3 4 5 6 7

Aggressive 46 20 10 7 4 1 0 2 2

Non-aggressive 41 23 5 6 2 2 1 1 1

Br. Corros. J., 1967, Vol. 2, May

 Booth et aI.: Criteria of Soil Aggressiveness towards Buried Metals. II. 113
corrosion or to provide a localised anodic zone for corrosion are rapid and the results require very little computation from
arising from differential aeration. the direct measurements made.
It is recommended that measurements should be made as
often as possible over as long a period as is practicable, and Acknowledgments
that the aggressiveness of the soil be assessed from the mean We are greatly indebted to a very large number of people
values obtained according to the following scheme: who so willingly gave their time and assistance in the
Aggressive Non-aggressive provision of.sites and in actual manual labour in sampling,
Resistivity, ,Q':'cm <2000 >2000 in particular members of the staff of the Metropolitan Water
and/or Board, North Thames Gas Board, West Midlands Gas Board,
Redox-potential at Wales Gas Board and the General Post Office.
pH=7, V (N.H.E.) <0·400 > 0·400 The work described above has been carried out as part of
or < 0·430 if clay or > 0 ·430 if clay the General Research Programme of the National Physical
Borderline cases to Laboratory.
be resolved by: References
water content % w/w > 20 < 20 1. Booth, G. H., Cooper, A. W., Cooper, P. M., & Wakerley, D. S.,
Br. Corros. J., 1967, 2, 104
The recommended testing procedure has the advantage that 2. Schwerdtfeger, W. J., J. Res. natn Bur. Stand., 1965, 69C, 71
it requires a minimum of laboratory work, the greater part 3. Starkey, R. L., & Wight, K. M., Am. Gas Ass. Monogr., 1945,
'Anaerobic Corrosion of Iron in Soils'
being based on field measurements using readily available 4. Tomashov, N. D., & Mikhailovsky, Y. N., Trudy Inst. jiz.
and simply constructed robust equipment. The methods Khim., 1960, (8), 190

Dr. Corros. J., 1967, Vol. 2, May

114 Booth et al.: Criteria of Soil Aggressiveness towards Buried Metals. II.
Appendix
Details of soil sampling sites

Number Name Description of soil at 3 ft depth '. Authority History

1 Hounslow Heath Heterogenous clay M.W.B. N

2 Heston, Bath Road Light-brown loanl N
"
3 Norwood Green Heavy brown clay N
"
4* Greenford Mixed clay and loam A
"
5 Perivale Sticky clay-loam mixture A
"
6 Stonebridge Park Heavy mixed soil A
"
7* Cricklewood Heterogenous clay N
"
8 Kidderpore Mixed mottled clay and topsoil N
"
9 Malden (Marsh Lane) Heavy clay A
"
10 Malden (Elm Road) Clay N
"
11 Raynes Park Fine dark loam N
"
12 Putney Heath Grey clay and ballast N
"
13* Brixton Brown clay N
"
14 Crystal Palace Mixed clay and gravel N
"
15 Rock Hill Light brown homogenous clay N
"
16 Nunhead Brown homogenous clay N
"
17 Fortis Green Light brown clay and sand A
"
18 Southgate Mixed clay, sand and gravel N
"
19 Botany Bay Light brown clay an~ ballast N
"
20* Chigwell Heavy brown clay N
"
21 Canvey Island-Eastern Esplanade Homogenous brown clay N.T.G.B. A
22 Canvey Island-Westman Road Grey-brown clay A
"
23 Canvey Island- Yadsoe Road Grey clay A
"
24 Canvey Island-Central Wall Road Homogenous brown clay A
"
25 Canvey Island-Haven Road Mixed clay and decayed wood A
"
26* Canvey Island-Gas Holder Clay A
"
27 Pitsea-Pound Lane Homogenous brown clay A
"
28 Pitsea-Burnt Mills Road Heavy loam A
"
29 Pitsea-Rectory Road Grey-brown clay A
"
30 Pitsea-Northlands Homogenous brown clay A
"
31 Pit sea-Station Lane Homogenous clay A
"
32 Benfleet-Brook Road Black clay A
"
33 Benfleet- Thundersley Park Road Clay-loam mixture A
"
34 Benfleet -Jotmans Lane Homogenous brown clay A
"
35 Benfleet-Hill Road Brown clay A
"
36 Benfleet-Yilla Road Clay-loam mixture A
"
37 Thundersley-Rushbottom Lane Homogenous brown clay A
"
38 Thundersley- Woodside Avenue Homogenous brown clay A
"
39 Thundersley-Eversley Road Brown clay and small stones A
"
40 Thundersley-Church Road Brown clay-loam mixture A
"
41 Thundersley-Catherine Road Homogenous brown clay A
"
42* Basildon Clay-sand mixture A
"
43* Rayleigh Brown clay A
"
44 Abergavenny Brown loam W.G.B. A

Br. Corros. J., 1967, Vol. 2, May

 Booth et al.: Criteria of Soil Aggressiveness towards Buried Metals. II. 115

Appendix (continued)

Number Name Description of soil at 3 ft depth Authority History

45 Aberdare Clay-sand mixture W.G.B. A

46 Moriston Dark muddy clay A
"
47 Hendy Peat A
"
48 Trostre Soft grey clay A
"
49 Llanelly Black mud and gravel A
"
50 Trimsaran Red-brown clay N
"
51 Haverfordwest Heavy black loam A
"
52 Llandudno Clay-clinker mixture A
"
53 Llandudno Junction Red homogenous clay N
"
54 Rhylj Abergele Blue clay A
"
55 Bartley Green Clay and stones W.M.G.B. A
56 Sheldon Peat A
"
57 Shard End Clay, loam and stones mixture A
"
58 Falcon Lodge Red stony clay N
"
59 Willenhall Dark loam A
"
60* Ripley Fine sandy loam G.P.O. X
61 Clandon West Fine brown loam X
"
62 Shere Dark loam X
"
63 Ewhurst Brown clay X
"
64 North Chapel Red/yellow clay X
"
65 Haslemere Heavy loam, some clay X
"
66 Passfie1d Dark loam X
"
67 Yateley Clay, sand and loam mixture X
"
68 Crowthorne Fine loam X
"
69* Eversley Fine loam X
"
70 Arborfield Soft brown clay X
"
71 Spencers Wood Brown and grey clay X
"
72 Hurley Brown loam X
"
73 Littlewick Green Brown loam X
"
74 Bourne End Brown clay-sand mixture X
"
75 Farnham Common Fine loam and stones X
"
76 Colnbrook Dark fine loam X
"
77 Shephall Loam and clay n1ixture X
"
78 Walkern Fine loam X
"
79 Stevenage Heterogenous clay X
"
80 Knebworth Red loam X
"
81 Codicote Clay with stones X
"
82 Kinsbourne Green Sandy- gravel X
"
83* Broughton Clay X
"
84 Great Brickhill Homogenous brown clay X
"
85 Tring Flint and chalk X
"
86* Turville Fine loam, stones and chalk J.C. X
87* Teddington Fine loam with fine gravel N.P.L. X

l
M.W.B. = Metropolitan Water Board G.p.a. = General Post dffice
N.T.G.B. = North Thames Gas Board J.C. = the late Rev. J ..Charlesworth
W.G.B. = Wales Gas Board N.P.L. = National Physical Laboratory
W.M.G.B. = West Midlands Gas Board
A = aggressive, corrosion of iron and/or steel reported x = unknown, no information obtained until end of survey
N = non-aggressive, no corrosion of iron and/or steel reported * = sites also used for burial tests

Br. Corros. J., 1967, Vol. 2, May