Indi an Journal of C hemica l Technology

Vol. 9, March 2002, pp. 148- 153

Corrocal-A computer program for corrosion current density determinations

from polarisation experiments
R Rengesh Siva, R Prabhakar & V S Muralidharan*
Ce ntral Electrochemical Researc h Institute, Karaikudi , India
Received 8 March 200 1; revised 15 December 2001; accepted 23 JamWI)' 2002

One area in whi ch co mputers find use is in th e calculation of Corrosion curre nt density and rate calculations. To
calcul ate e lec trochemical parameters such as anodi c and cathodic Tafe l slopes, co rrosio n c urrent density and po lari sati on
resistance, a software has been developed using C++ language. The software wou ld process data obtained either from a
ga lvanostati c ex periment or potentiostatic experiment. The potentials are given in milli vo lts and the current as
microamperes as input. The developed software would find : a) Corrosion current s by anodic and cathodic Tafel line
ex trapo lations, b) corrosion rates expressed as mpy, c) Cathodic and Anodic Tafel Slopes, d) Corrosion currents by Stern -
Geary method, e) Corrosio n currents by Barnatt' s method, f) Corrosion currents by Oldham and Mansfe ld 's method. Details
of the Corrocal and its applications to various systems are presented.

Implementing computers is an organisational obtained either from galvanostatic or potentiostatic

necessi ty today , especially for those scientists who
have set their sights on global researches. Computers
in scenario hasn' t happened yet, the day isn't far
especially for the applications in the tedious and
repetitive field of Electrochemistry and Corrosion 1•
An earlier program calculated corrosion rates from
sets of data such as (i) resi stance data probe, (ii)
weight loss coupon, (iii) ion count and (iv) linear
polarization resistance method. Another similar short
program calculated corrosion rates and electro-
chemical parameters from polarisation data 2 from a
variety of corroding systems. The "BETACRUNCH"
program eliminated the cumbersome and often
inaccurate, linear polarisation and Tafel extrapolation
techniques. Another program analysed the
polarisation data in the vicinity of the corrosion
potential 3 and used the Gauss- Newton method to
generate a new set of parameter estimates and the
process was repeated until the nonlinear residual error
failed to change by more than a present value.
Mansfield 4 developed "CORFIT' which analysed the
polarisation resistance in the Pre-Tafel region. Since it
only requires the determination of a polarisation curve
and resulted in the values of all corrosion parameters
with their errors.
The above mentioned and similar programs were
written in "ancient" languages and not compatible to
the present day computers. Hence a software has been
developed to process current and potential data
*For Correspondence: (E mai l: varagur@ya hoo.com)
measurements to determine kineti c and mechanistic
parameters using C++ language.
Theory
Wagner and Traud 5 proposed the concept of mixed
potential theory which postulated that the net
corrosion reaction is the result of two or more partial
electrochemical reactions which are in principle
independent of each other. The interaction between
oxidation and reduction reaction s occurring on
corroding metals can be described o n the basis of the
mixed potential theory using the relationships
between current and potential 6 ·7 •
For the pair of electrochemical reactions,
. . . (1)
. .. (2)
which occur concurrently, the relationship between
current I and potential £ can be written as,
/1 = /I a +fi e
= /oJ[exp((E-EoJ)/b'J a)) - exp((E- €o t)lb'J c)] .. . (3)
/2 = ha + f 2c
= loz [exp((E-Eoz)/b'za)- exp(-(E- Eoz)lb'zc) ] .. . (4)
The above equations have a characteristic reversible
potential £ 0 and an exchange current density / 0 . The
constants depend on the reaction mechanism. The
 RENGESH SlY A eta/.: COMPUTER PROGRAM FOR CORROSION CURRENT DENSITY DETERMINATIONS 149

partial currents h, and 11c relate to the forward and The maximum error in potential was described as
backward reaction rates while / 2, and 12c relate to
Eq. (2) . e =bl(2.303* 0.48) [ fslleorr] ... (14)

At corrosion potential Ecorn the external current, 1=0, for b.= be = b.

... (5) Barnatt 16 in his "linear corrosion kinetics" predicted

Where fcorr is the corrosion current.
lcorr = /o1 exp((Ecorr- Eo1)/b'1a))
= /02exp((-Ecorr- £02)/b'2a))
I =fcorr [exp((E-Ecorr)/b'la))--exp( -(E-Ec0 rr)/b'2c)]
Differentiation of the above equation w.r.t £ gives 8- 10
8118£ = /ol/b'la [exp((£-EoJ)/b'Ja))] + l02/b'2c
x[exp((Eor £)1b'2c)]
a range of linearity upto 60mV within the error. He
also proposed a three-point method for the study of
electrode reaction mechanisms. Three potentiostatic
pulses M:h 2fl£ 1 and -2fl£ 1 or -!:J.£1. -2M: 1 and 2fl£ are
applied. The resulting currents are measured and
... (6)
substituted in a system of simple equations from
which the corrosion current density and the Tafel
... (7) slopes are obtained. Oldham and Mansfield (loc. cit)
described a graphical method, which does not require
a prior knowledge of the Tafel slopes. It involved the
measurements in the vicinity of Ecorr. changes in the
test metal surface due to the measurement process are
... (8) avoided.

At the Ec0 rr, Program

(8/ I 8E)ecorr = feorr (llb'la + l/b'2J . . . (9) The program to process data is written In C++
17
language • Fig. l presents the flow chart.
The Eq. (9) shows that the lcorr can be calculated from The program would display
the measured slope of a /-£ plot at Ecorr provided the
constants b' 1a and b' 2c are known. Since these (a) E- I curves
constants are related to the b, and be by (b) E- log I curves
(c) Over Potential- log I curves
ba = 2.303 b'la = 8£ I 8 log /Ia . .. (10)
and also determine
be= 2.303 b'1 c = 8£ I 8 log he ... (II)
(a) Corrosion currents by Anodic and Cathodic Tafel
(8/ I 8E)ecorr = feorr (2.303 (ba + be) I ba be) ... (12) Line Extapolations
(b) Corrosion rates expressed as mpy .
This "polarisation conductance" is directly (c) Anodic and Cathodic Tafel Slopes
11 (d) Corrosion currents by Stern- Geary Method
proportional to lcorr· Bonhoeffer · and Jena and
12 13 (e) Corrosion currents by Barnatt's Three-point
Stern · preferred (8/18£) and christened the term
Method
polarisation resistance Rp.
(f) Corrosion currents by Oldham and Mansfeld
Method.
Therefore, fcorT = B IRP ... (13)
Example 1
The constant B has to be determined in an
independent experiment or by graphical technique or Polarisation Data on the corrosion of Carbon Steel in
computer analysis 14 . It was pointed in the original 1.0 N Sulphuric Acid 18
15
discussion of the Stern-Geary method , !:J.E/M = R'P
depends heavily on the judgement which an 21 (No. of Experimental Points)
investigator uses to determine the extent of the linear (Yoltage.mY) (Current,microAmps)
polarisation curve. In order to assess the potential -5 15 200
region within which a polarization curve is "linear -520 800
within a given error", the error was calculated. -525 . 1000
 150 INDIAN J. CHEM. TECHNOL., JANUARY 2002

-540 2000 -4zor-------------------------·-----------,

-560 3000
-570 4000
-600 8000 -491 ~·
+
-6 15 10000
-620 15000
E
-638 20000 -562 r-•
-5 10 200
-507 500 r-
-503 800 - 634 I-
-500 1000
-495 2000 -670 1--

-490 3000 1014 4805 8595 12386 t6176

-480 4000 E-1 curve
-475 8000
-470 10000 Fig. 2- Potentiai-Current curves for the corrosio n of Carbo n Steel
-468 15000 in IN Su lphuric Acid.
-450 20000
-42or-------------------------------------,

Figs 2,3 and 4 present the potenti al-current,potential-

Log Current and OverPotential-Log Current curves -491 1--
for Carbon Steel in 1.0 N sulphuric acid sol ution
respectively . E
- 562 r-

~ -634

I Input E,l
I -670

~ 5·35 6.36
Log I
1·37 8·38 9·39

Draw Graphs E-I , E-Log I

lj-Log [
Fig. 3-Potential-log Current curves for the co rrosion of Carbon
Steel in IN Sulphuric Ac id
l
Extrapolate the Tafel Region
To Find I corr
65
in the lj-Log I Graph Anodic & Cathodic
Tafel Slopes are
I 40
Calculated for Stem-
t Geary Method
15
From I corr, Corrosion Rate
0
is calculated for AnOdic & "1 -9
5·51 6·39 7·26 8.t4 9-02 9·46
Cathodic Tafel Regions
-35
! -60-
The Results are compared
with Three point & Four point
-85
method & Stored in a file

~
-110

Log I

Fig . 4--0ver Potenti al - log Current curves for th e corrosio n of

Fig. !- Flowchart for Corroca l Program. Carbon Steel in IN Sulphuric Ac id.
 RENGESH SIVA eta/.: COMPUTER PROGRAM FOR CORROS ION CURRENT DENS ITY DE I LI<M INATIONS 15 1

T he resul ts are displayed as shown below 90 590

IOO 8 10
Anodic Tafel Slope (40-80mv) =30.2982 12 110 1090
Anodic Tafel Slope (80- 125mv) =42.409309 120 1390
Cathodic Tafel Slope (40-80mv) =-41.703735 130 1670
Cathod ic Tafel Slope (80- 125mv) =-45.970024 140 19 10
I Corr(anod ic 40-80mv) =763.600952 micro Amps 150 2160
I corr(anodic 80- 125mv) = 1878.3125 micro Amps 160 2430
I Corr (cathodic 40-80mv) = 3 14.088806 micro Amps 170 27 10
I corr (cathodic 80- 125mv) =44 1.836975 micro Amps 180 30 10
190 3280
Cathodic corrosion rate(40-80mv) = 142.652988 mpy
Cathodic corrosion rate(80- 125mv) =200.673706 mpy 200 3620
Anodic corrosion rat e( 40-80mv) =346.8 12606 mpy 2 10 3910
Anodic corrosion rate(80-125mv) =853.560409 mpy
Figs 5,6 and 7 presen t the potenti al-current,potentiai-
STERN· GEARY METHOD Log C urrent and O verPotentiai-Log Cu rrent curves
for 90/1 0 cupronickel in 0.05 M sodium chloride
Charge Transfer Resistance = 20.58 1085 ohms
lcorr = 2.340349 micro amps solutions respectively .

BARNATT-THREE POINT-TECHNIQUE The resu lts are displayed as shown below:

= 30 14.159668 micro amps
l corr
Anodic Tafel Slope (40-80 mv) =62.435993
Anodic Tafel Slope = 163.33168 Anodic Tafel Slope (80- 125mv) =93.383423
Cathodic Tafel Slope =22.0883 14 Cathodic Tafel Slope (40-80mv) =-40.944374
OLDHAM MANSFELD TECHNIQUE Cathodic Tafel Slope (80-125 mv) = - I00.286446
I corr(anodic 40-80mv) =580.04 I443 micro Amps
I corr =2.903709 micro amps I corr(anodic 80- 125mv) =879.146606 micro Amps
I Corr(cathodi c 40-80mv) = 288.33667 micro Amps
Example 2
I corr(cathodic 80- 125mv) = 874.947388 micro Amps
Polarisation data on the corrosion of 90/1 0 Cathodic corrosion rate(40-80mv) = I66.697037 mpy
cupronickel alloy in 0.05M sodium chloride Anodic corros ion rate( 40-80mv) = 335.34 1278 mpy
19
solutions Cathodi c corros ion rate(80- 125 mv) =505.836243 mpy
Anodic corrosion rate(80-1 25 mv) =508 .263947 mpy
29 (No. of Experi mental Poi nts)
STERN- GEARY METHOD
(Voltage,mV) (Curre nt,microA mps)
-70 3330
C harge Transfe r Resistance =0.05008 1 ohm
I corr (A nodic) = 1032.658 mic ro amps.
-60 3040 (Cathodic) =288.366 micro amps.
-50 2760
-40 2490 THR EE-POINT TECHNIQUE
-30 2230
-20 1940 lcorr = 11 45 .549805 micro amps
-10 1640 Anodic Tafel Slope = 184.682404
Cathodic Tafel Slope = 33.649345
0 I400
10 11 20 FOUR-POI NT TECHNIQUE
20 860
30 600 l corr = 3.079888 micro amps
40 330
50 11 0 Discussions
60 30 Tables I and 2 presen t the comparison of the values
70 150 obtained from Corrocal and the reported values for
80 350 the corrosion of carbon steel in 1.0 N sulphuri c acid.
 !52 INDIAN J. CHEM. TECHNOL.. JANUARY 2002

11.5
210

108
..
. 135 70

33

E 60
1) -4 J .6J 4 · ~5 5·48 6-41 7·34 7· 80

-41
-13
-78
- 51

208 947 1686 2425 3164 3534 - 116

E-1 curve

Fig. 5-Potentiai-Current curves for the corrosion of 90/10

Cupronickel Alloy in 0.05M Sodium chloride solutions. Fig. 7- 0ver Potential-log Current curves for the corrosion of
90/10 Cupronickel alloy in 0.05M Sodiu m ch loride solutions.

210 1-
Table !-Comparison of the values obtained by the Tafel
extrapolation obtained- Corrocal and the earlier reported values
Parameter Corrocal Reported value/
135 1-
calculated value

Corrosion current density 763 .6 600

E 60 1- (anodic)

23 1- Corrosion current density 314.08 600

(cathodic)
-13

-51 Anodic Tafel slope 30&40 30

3-63 4 -55 5·48 6 41 7·34 7·80
Cathodic Tafel slope 41 &45 110
Log I

Fig. 6-Potential-log Current curves for the corrosion of 90110

Cupronickel Alloy in 0.05M Sodium chloride solutions.
Tables 3 and 4 present the comparison of the values
obtained from Corrocal and the reported values for
the corrosion of 90110 cupronickel in 0.5M sodium
chloride solutions .
Sources of Errors 20 -27
It was found that in deriving Eq. (3):
(a) The Charge Transfer kinetics can be adequately
described with the Butler-Volmer equation.
(b) The adsorption of any species on the surface is
potential dependent.
(c) The reactions are under kinetic control and the
effect of mass transport is negligible.
(d) The double layer effect of electrode kinetics is
negligible.
(e) The uncompensated solution resistance between
the metal and the reference electrode is
negligible.
(f) The rate of the reverse reactions of the anodic
and cathodic partial reactions is negligible.
(g) Only one cathodic reaction and only one anodic
reaction take place during the corrosion process.
(h) The corrosion is "Uniform" over the whole
surface of the specimen, i.e., the anodic and
cathodic current densities are uniform and equal
and can be calcu lated on the basis of the fu ll
surface area of the specimen .
(i) The products of the anodic reaction are soluble
in the solution at least to the extent that no
product film forms on the surface.
U) The system is time invariant i.e, all system
parameters are constant with time of
measurement.
Error sources and the calculation methods
Error of the measurement occurs whenever one of
the simplifying assumptions is not justified. The
errors defined as I OOUcorncalc - icOIT>lrue llicorr,true are
 RENGESH SIVA eta/ .: COMPUTER PROGRAM FOR CORROSION CURRENT DENSITY DETERMINATIONS 153

Table 2-Compari son of the values obtained by the linear point technique are generally somewhat larger than
po lari sation obtained- Corroca/ and th e earli er repo rted values those of the polarization technique. The errors in
Method Corrocal Reported va lue/ curve fitting technique have been reported to be
calculated value approximately the same as those of the three-point
Stern- Gea ry 2. 34 3 technique. In contrast to the polarisation resistance
Barnan 's method 3014 technique numerical values for the appropriate
Oldham & Mansfe ld 2.9 correction terms are not needed, since they can be
treated as unknowns in the curve fitting .
Table 3-Compari so n of the va lues obtained by the Tafel
extrapolation obtained - Corroca/ and the earlier reported values References
Parameter Co rrocal Reported va lue/ I John D G & Hpadk y K, Mater Pe1jonn, 1911 ( 1980) 42 .
calculated value 2 Greene N D & Ghandi R H, Mat er Perform. 217 ( 1982) 34.
3 Gerchakov S M, Udey L.R & Mansfe ld F, Corrosion. 371 2
Corrosion current density 580 500 (1981) 36.
(anodi c) 4 Mansfeld F, Corrosion, 29 (1973) 397 .
5 Wagner C W & Traud W, Z Electrochem. 44 (1938) 391.
Corrosion current density 879 500
6 Kaeshe K Z, Metallkunde, 61 ( 1970) 94.
(cathodi c)
7 Yetter K J, Electrochemical kinetics (Academi c press,
An odi c Tafel slope 62 & 93 60 Newyork), 1967.
Cathodi c Tafel slope 42 & 100 40 8 Oldham K B & Mansfeld, F Corr Sci, 13 ( 1973) 813 .
9 Oldham K B &Mansfeld F Corrosion. 27 ( 1971) 434.
I0 Mansfeld F & Oldham K B, Corr Sci, II ( 1971) 787 .
Table 4-Compari son of the values obtained by the linear
II Bonhoeffer K F & Ze na W, Z Electrochem. 55 ( 195 I ) I 5 I.
polarisation obtained Corrocal and the earlier reported values
12 Stern M & Geary A L, J Electrochem Soc. I 04 ( 1957) 56.
Method Corrocal Reported value/ 13 Stern M & Roth R M. J E/ectrochem Soc, 104 (1957) 390.
calcul ated value 14 MansfeldF, J.EiectrochemSoc,l20(1973)397.
IS Stern M, Corrosion, 14 ( 1958) 440t.
Stern - Geary 1032 500 16 Barnatt S, CorrSci, 9 (1969) 145.
Barnatt' s method 1145 17 Herbert Schildt, C++ The Complete Reference (Tata
Oldham & Mansfe ld 3.08 McGraw Hill , New Delhi), 3rd Edition, 1998 .
18 Bandy R & Jones D A, Corrosion, 32 (1976) 126.
19 .Mathiarasu J, Palanisamy N & Muralidharan V S, Proc
generally a function of several parameters Indian Acad Science [Chem Sci}, Il l 2 (1999) 377 .
characterizing both the corroding system and the 20 Periasamy M & Krishnaswamy P R, J Electroanal Chem
measuring technique. The simulation is typically Interfacial Electrochem, 61 (1975) 349.
21 Roy S K & Sircar S C , 8 Corros J, 13 ( 1978) 193.
carried out for a practical range of all parameters. The
22 Meszaros L, Lengyel B & Janaszik F Korroz Figyelo, 21 2
three-point technique is typically evaluated with (1981) 30.
several data sets in the polarisation range of± 30 mY. 23 Gill J S, Callow L M & Scantlebury J D, Corrosion, 39
The curve fitting technique is typically evaluated with (1983)61.
24 Drazic D M & Vascic V, Corros Sci, 25 (1985) 483 .
61 points in the range of ±30m V.
25 Drazic D M & Vascic V, J Electroanal Chem Interfacial
Of the four electrochemical techniques discussed, Electrochem, 185 ( 1985) 229 .
the errors of the Tafel Plot technique have been the 26 Nagy Z, Land R H, Leaf G K & Minkoff M, J Electrochem
least evaluated. Consequently, it is difficult to Soc, 112 (1985) 2626.
compare this technique to the others 28 . The errors of 27 Nagy Z , in Modem Aspects of Electrochemistry, Vol 21 ,
the polarization resistance technique have been very edited by White R E, Bockris J 0 M & Conway B E (Plen um
Press, NY), 1990, 237 .
thoroughly and quantitatively evaluated and the 28 Nagy Z , in Mordern Aspects of Electrochemistry, Vol 25,
reported errors are the smallest among the four edi ted by White R E, Bockris J 0 M & Conway B E (Plenum
techniques for all categories . The errors of the three- Press, NY), 1993, 237 .