Journal of The Electrochemical Society, 147 (10) 3637-3646 (2000)
3637
S0013-4651(99)12-003-2 CCC: $7.00 © The Electrochemical Society, Inc.
Corrosion Mechanism of Nickel in Hot, Concentrated H2SO4
J. R. Kish,a,z M. B. Ives,b,* and J. R. Roddab
aPulp and Paper Research
bW. W. Smeltzer Corrosion
Institute of Canada, Vancouver, British Columbia, Canada V6S 2L9
Laboratory, McMaster University, Hamilton, Ontario, Canada L8S 4L7
Electrochemical techniques, complemented by weight change and ex situ X-ray spectroscopic measurements, were employed to
characterize the corrosion of nickel in concentrated H2SO4 solutions. By use of a rotating cylinder electrode, it was found that corrosion is a mass-transport controlled process with the convective diffusion of nickel cations from a saturated NiSO4 layer as its
rate-determining step. The oxidizing nature of the acid solution leads to the formation of additional corrosion products including
metastable NiS, and elemental sulfur along with NiSO4, none of which is protective. When present on the surface, NiS establishes a galvanic interaction with the uncovered metal, significantly polarizing the anodic metal dissolution reaction. Since corrosion
is mass-transport controlled, the resultant corrosion rate of the metal is unaffected during the galvanic-induced polarization.
© 2000 The Electrochemical Society. S0013-4651(99)12-003-2. All rights reserved.
Manuscript submitted December 2, 1999; revised manuscript received June 21, 2000.
Sulfuric acid (H2SO4) is a highly corrosive, yet important, industrial inorganic chemical, being used at some stage in the manufacture
of a vast array of industrial products. The specification of appropriate construction materials for containment of this acid is a continuing challenge, especially for its production, where the nominal composition of the H2SO4-H2O solution is between 93 and 98.5 wt %
H2SO4. Nickel is a major alloying element in the austenitic stainless
steels used as a construction material for various critical components
required for the production of H2SO4.1-3 Although primarily alloyed
to retain the austenite phase, nickel has a significant beneficial influence on the resulting corrosion resistance of stainless steel to concentrated H2SO4.4,5 Alloyed nickel promotes a spontaneous cyclic
passivation and depassivation of the stainless steel alloy, which otherwise would not occur.4,5 A fundamental understanding of the role
of nickel in stabilizing passivity is essential in the development of
more corrosion-resistant stainless steel alloys required to improve
both the thermal efficiency and the reliability of the production
process.6,7 Elucidating the corrosion mechanism of nickel in concentrated H2SO4 should help considerably in this regard.
Corrosion in concentrated H2SO4 solutions is a more complex
process than in dilute H2SO4 solutions because of the change in
chemical structure. Molecular H2SO4 appears in solutions containing about 70 wt % H2SO4 and continues to increase in concentration
as the mixture approaches 100 wt % H2SO4.8,9 Above about 84 wt
% H2SO4, no free H2O molecules exist since in the presence of an
excess of molecular H2SO4, H2O acts as a strong base, reacting with
1 10-12 Therefore, concentrated
H2SO4 to produce HSO2
4 and H3O .
H2SO4 solutions (>84 wt % H2SO4) are essentially a solution consisting of molecular H2SO4 and HSO42 and H3O1 ions. The change
in chemical structure has a pronounced influence on the relationship
between the redox potential and acidity, which changes significantly at about 70 wt % H2SO4.13,14 This change coincides with the
appearance of undissociated H2SO4 molecules, which can behave as
oxidizing species and become reduced to various potential-dependent sulfur-containing species with a valence lower than 16.15-18
Concentrated H2SO4 solutions can act as either a reducing acid
or an oxidizing acid depending on the material exposed. For example, in contact with carbon steel, the cathode reaction involved in
corrosion is hydrogen evolution,19 which is typical for a reducing
acid. In contact with copper, the cathode reaction involved in corrosion is the reduction of undissociated H2SO4 molecules to various
reduced sulfur-containing species (SO2, S, H2S).20,21 Regardless of
the cathodic reaction, the formation of solid corrosion products
FeSO4?nH2O for carbon steel22 and CuSO4, Cu2S, and CuS for copper retard the corrosion rate to some degree.18 The corrosion mechanism of carbon steel in concentrated H2SO4 has been well charac* Electrochemical Society Active Member.
z E-mail: [email protected]
terized and involves a mass-transport-controlled process where the
rate-controlling step is the convective diffusion of ferrous ions from
a saturated layer of FeSO4?7H2O on the alloy surface.23-27 Several
works have been reported on the anodic behavior of nickel in concentrated H2SO4;28-30 all show anodic polarization results in the formation of a salt layer, which can impede the anodic current to a considerable extent, depending on the temperature and concentration.
X-ray diffraction analysis of the crystalline phase formed during
polarization in concentrated H2SO4 indicates that the film is most
likely b-NiSO4?6H2O,29 which forms because of the decreased nickel sulfate solubility in the more concentrated solutions.31
The objective of this study was to characterize the electrochemical behavior of nickel in 93.5 wt % H2SO4 in a attempt to elucidate
its corrosion mechanism. The procedures used involved mostly
potentiodynamic polarization and corrosion potential measurements
complemented by weight change determinations, all under the
hydrodynamic condition provided by a rotating cylinder electrode,
and ex situ X-ray spectroscopic measurements.
Experimental
Corrosion of nickel in 93.5 wt % H2SO4 was studied under the
hydrodynamic condition provided by a rotating cylinder electrode
(RCE). Test cylinder electrodes were prepared from a Nickel 270
(UNS N02270) rod of stated purity >99.97%; all had areas of
2.5 cm2. A test electrode was mounted and rotated using a Pine
Instrument Company model AFMSRX rotator and MSRX speed
control unit. All electrodes used were wet-ground with emery paper
up to 400 grit, cleaned in soap solution, rinsed with methanol, dried
with absorbent paper, and then weighed immediately prior to immersion. On removal, the electrodes were rinsed with distilled water,
cleaned in soap solution, rinsed with methanol, dried with absorbent
paper, and then weighed.
Concentrated 93.5 wt % H2SO4 solutions were prepared from
British Drug House reagent grade 97-98 wt % H2SO4 and distilled
H2O. The concentration of the prepared solutions was confirmed by a
sound velocity measurement. No attempt was made to aerate or deaerate the acid solutions. A fresh acid solution was used for each test.
Test solutions were contained in a glass electrolysis cell, 80 mm diam
and 50 mm in high, capable of holding 250 mL of solution. The cell
was mounted in a water thermostat bath set at 608C, which maintained
the acid solution temperature to within 60.58C in all experiments.
The systems of electrodes employed in this study is described
schematically in Fig. 1. The counter electrode was a wire immersed
in its own glass compartment, 10 mm diam and 70 mm high, which
was connected to the electrolysis cell containing the test RCE
through the use of a porous glass plug. Isolation of the counter electrode was used as a precautionary measure to prevent the possible
contamination of the test RCE with reduced sulfur-containing
Downloaded on 2016-09-18 to IP 130.203.136.75 address. Redistribution subject to ECS terms of use (see ecsdl.org/site/terms_use) unless CC License in place (see abstract).
3638
Journal of The Electrochemical Society, 147 (10) 3637-3646 (2000)
S0013-4651(99)12-003-2 CCC: $7.00 © The Electrochemical Society, Inc.
Figure 1. Schematic of the electrolysis
cell employed in the study. WE 5 working
electrode (RCE), CE 5 counter electrode,
and RE 5 reference electrode.
species, in particular elemental sulfur, which can form during the
cathodic electrolysis of the sulfuric acid solution.15-18
The electrochemical behavior of nickel was measured using a
computer-controlled EG&G PARC model 273 potentiostat. All
potentials were measured against a platinum wire electrode, which
in 93.5 wt % H2SO4 at 608C has a measured potential of 10.604 6
0.006 V against a standard calomel electrode (SCE). A platinum reference electrode was chosen because it was used with success in the
oleum and concentrated H2SO4 research of Arvía et al.18,32-34 Furthermore, a platinum wire was been successfully employed in commercial anodic protection systems for containing concentrated
H2SO4.35 A platinum wire was employed as the counter electrode in
all polarization measurements. The scan rate was 20 mV/s for all
polarization measurements.
The procedure employed to verify a mass-transport-controlled
anodic dissolution involved comparing an estimated Sh, the Sherwood number (dimensionless mass-transfer rate), with that calculated from a correlation for the expected geometry as a function of the
Re, Reynolds number (dimensionless rotational velocity). See the
List of Symbols for definitions.
A correlation applicable to turbulent flow for a system of concentric electrodes with a central rotating electrode has been derived
by Eisenberg et al.36 and is expressed as
Sh 5 0.079Re0.7Sc0.356
[1]
where Sc, the Schmidt number, is a dimensionless fluid property characterizing the relative thickness of the momentum and mass-transfer
boundary layers. This correlation has been confirmed for corrosion
studies of mild steel in 68 wt % H2SO4 at 278C for a system of concentric electrodes.24 Using a system of nonconcentric electrodes,
Rahmani and Strutt37 confirmed the Eisenberg et al. correlation 1 for
the corrosion of mild steel in 68 wt % H2SO4 at 278C, but found a significant variation in the velocity exponent at higher acid concentrations and temperatures. Analysis of the data results in the following
correlation for the corrosion in 93 and 98 wt % H2SO4 at 408C 37
Sh 5 0.079Re0.81Sc0.356
[2]
Both correlations were used as a basis for this study. Use of the correlations is justified since the adopted RCE assembly of nonconcen-
tric electrodes (see Fig. 1), has successfully reproduced the masstransport-controlled corrosion of ferritic stainless steel in 93 wt %
H2SO4 at 608C38 previously reported for a system of concentric
electrodes.39
X-ray spectroscopic techniques, X-ray fluorescence (XRF) spectroscopy, and X-ray photoelectron spectroscopy (XPS) were employed to characterize the solid corrosion products. The XRF measurements were carried out on a Philips PW 1480 XRF spectrometer.
This technique was employed to characterize the elemental composition of insoluble nonadherent corrosion products. The procedure
to collect the insoluble nonadherent products involved removing the
deposit from the surface of the electrode by rinsing the covered electrode with distilled water and subsequent filtering. All deposits collected in this manner were dried in air for at least 24 h prior to analysis. The XPS measurements were carried out on a Perkin Elmer PHI
5500 system with a monochromated Al Ka source. This technique
was employed to characterize the composition of insoluble adherent
corrosion products. The system collected the data using a pass energy of 29.4 eV with a takeoff angle of 758. Sample preparation prior
to analysis involved rinsing the electrodes with methanol, cleaning
in an ultrasonic methanol bath, and drying using high purity nitrogen gas.
Results
Corrosion of nickel in stagnant 93.5 wt % H2SO4 at 608C.—Figure 2 shows the corrosion potential of a nickel electrode in stagnant
93.5 wt % H2SO4 at 608C as a function of time. The potential quickly reaches a maximum of ,20.30 VPt within the first 2 h of exposure, after which it begins a two-stage decay over a period of about
6 h before stabilizing at ,20.50 VPt. The two-stage decay consists
of an initial slow decay over a period of about 300 min (5 h) followed by a fast decay over a period of about 60 min. The noncontinuous nature of the slow decay is an artifact of the in situ SO2
analysis, which requires the periodic removal of a 10 mL solution
sample. The sensitivity of the potential during its decay to the disturbance induced by solution removal provided an early indication
of the strong dependence on solution agitation as reported later.
A solid black deposit formed immediately after immersion and
covered most of the electrode surface. The black deposit remained
on the surface until the potential transition, after which is subse-
Downloaded on 2016-09-18 to IP 130.203.136.75 address. Redistribution subject to ECS terms of use (see ecsdl.org/site/terms_use) unless CC License in place (see abstract).
Journal of The Electrochemical Society, 147 (10) 3637-3646 (2000)
3639
S0013-4651(99)12-003-2 CCC: $7.00 © The Electrochemical Society, Inc.
Figure 2. Corrosion potential of a nickel electrode and the dissolved SO2
concentration in stagnant 93.5 wt % H2SO4 at 608C as a function of time
immediately after immersion.
quently dissolved. After the potential transition, the nickel surface
has a film-free, dull gray appearance for about 120 min (2 h) until
the subsequent formation of a stable, yellow surface deposit. Although the acid solution remained clear, after the 30 h exposure, it
have developed a yellow color. No visible gas evolution occurred
during the corrosion of nickel at either potential plateau. Hydrogen
gas evolution is unlikely since both potential plateaus are more positive than the equilibrium potential of the hydrogen evolution reaction, which is 20.607 VPt in 93.5 wt % at 258C.13,14 Therefore, some
cathode reaction other than H2 evolution occurred during the corrosion of the nickel electrode.
Superimposed on the corrosion potential-time plot in Fig. 2 is the
dissolved SO2 concentration in the acid solution as a function of
time during the corrosion of the same nickel electrode in stagnant
93.5 wt % H2SO4 at 608C. The concentration of dissolved SO2 was
analytically measured in situ during corrosion using iodometry as
described by Hall.40 The SO2 concentration increased monotonically with time for 10 h before it became constant at 0.051 g/L (2.78 3
1023 wt %). Since this steady-state concentration is significantly
lower than the equilibrium solubility of ,18 g/L,41 it is unlikely that
the acid became saturated with respect to SO2. When comparing the
release of SO2 with the corrosion potential as a function of time, it
is evident that the release of SO2 occurred during the period prior to
the potential transition and not during the period after the transition.
Although the presence of SO2 in the acid supports a cathodic reaction involving the reduction of H2SO4 to SO2, it is not conclusive. It
is unclear whether SO2 was produced as a reaction product of nickel corrosion and/or the black deposit dissolution. For reasons discussed below, SO2 is believed to be produced as the main cathodic
product involved in a galvanic corrosion process taking place at this
potential plateau.
The weight loss of the single nickel electrode measured after the
30 h exposure is 6.72 mg/cm2. Such a measurement provides no
information regarding the protective nature of either the black or yellow surface. It is possible that the majority of weight loss occurred
during the period prior to the potential transition corrosion or during
the period after the transition. Making weight loss measurements
using multiple nickel coupons as a function of time provided more
information in this regard.
Figure 3 shows the weight loss of nickel in stagnant 93.5 wt %
H2SO4 at 608C as a function of time. The curve was constructed by
measuring the weight loss of nine nickel samples, all immersed at
approximately the same time in a glass cell containing 2000 mL
Figure 3. Weight change of nickel electrode in stagnant 93.5 wt % H2SO4 at
608C as a function of time. Superimposed is the corrosion potential transient
from Fig. 1.
acid, but each removed after different exposure times. The weight
loss-time dependence of nickel consisted of an initial, apparent parabolic kinetic stage followed by a linear kinetic stage. All six samples removed during the apparent parabolic stage (10 h) consisted of
an adhered noncontinuous black deposit; the adherence decreasing
as the exposure time approached 10 h. The remaining three samples
removed during the linear stage were covered with a loosely adherent yellow surface film. Therefore, the weight loss measurements of
samples containing the adhered black deposit are misleading since
they include the weight gained from the adherent black deposit.
Nonetheless, the weight loss data show that neither the black deposit
nor the yellow film is significantly protective.
When comparing the release of SO2 with the weight loss as a
function of time, it is evident that the corrosion of nickel continues
at an appreciable rate despite the absence of further SO2 generation.
Therefore, the cathode reaction involved in the linear stage of corrosion must involve the formation of products other than SO2.
Comparing the measured concentration of dissolved SO2 with an
expected concentration calculated from the weight loss of the single
nickel electrode after 30 h exposure provided additional evidence
that the cathodic process involved the formation of additional product. Assuming that the corrosion of the single nickel electrode proceeds according to the reaction sequence
Anode:
Ni 5 Ni21 1 2e2
1 [3]
Cathode: 4H2SO4 1 2e2 5 SO422 1 SO2 1 2HSO2
4 1 2H3O
Total:
Ni 1 4H2SO4 5 NiSO4 1 SO2 1 2HSO42 1 2H3O1
where 1 mol of nickel reacted to produce 1 mol of SO2, with a 100%
current efficiency, then the SO2 concentration can be calculated from
the weight loss using Faraday’s law. According to Reaction 3, the
30 h weight loss of 6.72 mg/cm2 corresponds to a theoretical SO2
concentration of 0.160 g/L, which is significantly higher than the
measured 0.051 g/L and less than the equilibrium solubility of
18 g/L.41 Chemical analysis conducted to characterize the elemental
composition of the various solid corrosion products demonstrates
that other possible reduction products include sulfide ions and elemental sulfur.
An XPS analysis conducted on a nickel sample after a 15 min
exposure provided information regarding the strongly adherent noncontinuous black deposit. Figure 4 shows the sulfur 2p spectrum.
The sulfur region possessed a strong peak and shoulder at 161.12
Downloaded on 2016-09-18 to IP 130.203.136.75 address. Redistribution subject to ECS terms of use (see ecsdl.org/site/terms_use) unless CC License in place (see abstract).
3640
Journal of The Electrochemical Society, 147 (10) 3637-3646 (2000)
S0013-4651(99)12-003-2 CCC: $7.00 © The Electrochemical Society, Inc.
Table I. Binding energy (BE) of sulfur in selected compounds.
Compound
NiS
NiSO4
Figure 4. XPS sulfur 2p spectrum obtained from the surface of a nickel electrode after 15 min exposure in stagnant 93.5 wt % H2SO4 at 608C.
and 162.63 eV, respectively, and a small peak at 169.25 eV. Table I
presents a literature comparison42 of the binding energies of sulfur
in the most probable species that may have appeared in this study.
Based on the comparison, it is concluded that a NiS deposit covered
the majority of the nickel surface. The presence of NiSO4 was likely due to oxidation of the NiS deposit during exposure to the atmosphere prior to the XPS measurement. The most likely source of the
sulfide ion is the reduction of H2SO4 molecules.
Anticipating a subsequent chemical analysis, the loosely adherent yellow film that formed during the 30 h single nickel electrode
corrosion test was collected. The yellow precipitate consisted of a
water-soluble component and a water-insoluble component.
The soluble component produced a green solution when dissolved in water, consistent with the formation of a hydrated nickel
43 The yellow water-soluble component is consistent
ion (Ni 21
aq ).
with anhydrous NiSO4. If the solubility of NiSO4 in 93.5 wt %
H2SO4 at 608C is 0.12 wt %, extrapolated from the data reported by
Halstead and Lovey,44 then the time required for a nickel electrode,
with a dissolution rate of 0.24 mg/cm2/h, to saturate the solution
with NiSO4 is about 17 h. This is consistent with experimental
observation.
The chemical composition of the yellow water-insoluble corrosion product was analyzed by XRF. The analysis showed that the
deposit consisted of >99 wt % sulfur (S) and <1 wt % nickel (Ni),
indicating that the yellow water-insoluble corrosion product is elemental sulfur. Considering that the steady-state corrosion of nickel
did not involve the evolution of SO2, the reformation of NiS, or the
evolution of H2, it is reasonable to postulate that the cathodic process
involved the reduction of H2SO4 molecules to sulfur.
Corrosion of nickel in agitated 93.5 wt % H2SO4 at 608C.—Figure 5 shows the influence of the rotational speed on the corrosion
potential of a nickel RCE in 93.5 wt % H2SO4 at 608C as a function
of time. Superimposed on the plot is the corrosion potential of the
nickel electrode in otherwise stagnant 93.5 wt % H2SO4 at 608C
reported above. The curves show that the rotational speed significantly influences the potential transition time, the maximum potential attained prior to the transition, and the steady-state potential
attained after the transition. An increase in the rotational speed progressively decreased the potential transition time and progressively
Peak
Measured BE (eV)
Literature42 Be (eV)
2p3/2
2p
162.1
169.3
162.2
169.2
increased the maximum potential attained. The steady-state potential
after the transition has a more complex dependence of the rotational
speed. An increase in the rotational speed from 0 to 250 rpm increased the potential from ,20.50 to ,20.45 VPt, where it remained constant with a further increase to 500 rpm. The increase in
the rotational speed from 500 to 1000 rpm changed the stable potential after the transition from ,20.50 to ,20.62 VPt, where it remained constant with a further increase to 2000 rpm.
Several qualitative observations regarding the rate control of the
various processes can be made by associating the potential behavior
with the formation and subsequent dissolution of a NiS deposit.
Since the time required for the potential of the nickel RCE to reach
its maximum is independent of the rotational speed, the formation of
NiS is likely not a mass-transport-controlled process. In contrast,
since the potential transition time is clearly sensitive to the rotational speed, the dissolution of NiS is likely a mass-transport-controlled
process.
Table II shows the influence of the rotational speed on the weight
loss of a nickel RCE during corrosion in 93.5 wt % H2SO4 at 608C.
An increase in the rotational speed from 250 to 2000 rpm increases
the weight loss from 0.44 to 12.12 mg/cm2. Although a direct correlation between weight loss and the rotational speed cannot be concluded from this data alone since the increase in weight loss may be
attributed to a higher rate of weight loss for a NiS-free surface, it can
be concluded when considering the anodic polarization behavior.
Figure 6 superimposes the potentiodynamic anodic polarization
behavior of a nickel RCE (1000 rpm) with and without a NiS surface
deposit in 93.5 wt % H2SO4 at 608C. Polarizing the potential in the
anodic direction produced a limiting current density (,1 mA/cm2)
Figure 5. Influence of the rotational speed on the corrosion potential of a
nickel RCE in 93.5 wt % H2SO4 at 608C as a function of time immediately
after immersion. represents steady-state potential in stagnant (0 rpm) acid
measured after 30 h exposure.
Downloaded on 2016-09-18 to IP 130.203.136.75 address. Redistribution subject to ECS terms of use (see ecsdl.org/site/terms_use) unless CC License in place (see abstract).
Journal of The Electrochemical Society, 147 (10) 3637-3646 (2000)
3641
S0013-4651(99)12-003-2 CCC: $7.00 © The Electrochemical Society, Inc.
Table II. Influence of rotational speed on the weight loss of nickel RCE in 93.5 wt % H2SO4 at 608C.
Rotational speed
(rpm)
Exposure time
(min)
Weight loss
(mg/cm2)
1250
1500
1000
2000
150
150
150
150
10.44
15.44
17.84
12.12
over the potential range examined, independent of the starting potential and, thus the surface state. The limiting anodic current density
shows that anodic dissolution is the mass-transport-controlled process involved in the corrosion of nickel conjectured above. Furthermore, this comparison shows that the NiS deposit has no protective
capability, consistent with the weight change measurements reported
above (see Fig. 3 and Table II).
The similar magnitude of the limiting current indicates that the
rate-determining step in mass-transport-controlled anodic dissolution of nickel is identical, regardless of whether a NiS surface deposit is present on the surface. Therefore, the dissolution rate of NiS
is controlled by the rate-determining step in the anodic dissolution of
nickel. Since nickel has a similar response to anodic polarization in
concentrated H2SO4 as does iron in its well-known sulfation
stage,45,46 and forms a similar sulfate salt film, it is reasonable to
postulate that nickel has a similar rate-determining step as does iron;
the convective mass transfer of the metal cation from the saturated
metal sulfate layer.
Mass transfer correlation.—To confirm whether the NiS dissolution process is controlled by the convective mass transfer of nickel
cations (Ni21) from a saturated NiSO4 layer, an estimated Sherwood
number was compared to that calculated using the Eisenberg et al.
correlation (Eq. 1) and the Rahmani and Strutt correlation (Eq. 2)
and as a function of the Reynolds number (see Fig. 7). Table III lists
the magnitudes of all parameters used to estimate the Sherwood,
Schmidt, and Reynolds numbers for the comparison. The Sherwood
number was estimated from the limiting anodic current density made
Figure 6. Influence of NiS on the potentiodynamic anodic polarization behavior of nickel RCE (1000 rpm) in 93.5 wt % H2SO4 at 608C.
on a nickel RCE with an NiS deposit on its surface through the following relation
 d 
i 
1
Sh* 5  l  
 nF   Cs 2 Cb   D 
[4]
The nickel ion diffusion coefficient was calculated from the Wilketype extrapolation formulas47 as proposed by Ellison and Schmeal.24
Figure 7 indicates that the anodic dissolution process of the nickel RCE with a NiS deposit on its surface is mass transport controlled
with the diffusion of nickel cations from a saturated NiSO4 layer as
its rate-determining step; the limiting anodic currents increased with
flow. An analysis of the data results in the following expression for
the mass-transfer process
Sh 5 0.079Re0.76Sc0.356
[5]
The best fit correlation lies between the Eisenberg et al. correlation
(Eq. 1) and the Rahmani and Strutt correlation (Eq. 2).
As discussed by Rahmani and Strutt,37 a possible explanation of
the observed increase in mass-transfer rate in highly concentrated
H2SO4, as compared to the Eisenberg et al. correlation (Eq. 1), involves the extrapolation formula suggested by Ellison and
Schmeal24 to calculate the diffusion coefficient. The formula was
arrived at through measurements of the diffusion coefficient for ferrous ion in 68 wt % H2SO4 using a rotating platinum disk electrode.
Although an agreement between the model (Eq. 1) and the experimental results was found in 68 wt % H2SO4 using the formula,24 it
is possible that the formula underpredicts the diffusion coefficient of
ferrous ion (and nickel ion) acids of higher concentration. A possible explanation for the lack of agreement of the correlation of Eq. 5,
as compared to the Eisenberg et al. correlation (Eq. 1) and the Rahmani and Strutt correlation (Eq. 2), involves the solubility of NiSO4.
The NiSO4 solubility used to estimate the Sherwood number was
Figure 7. Influence of rotational speed on the limiting anodic current density of a nickel RCE in 93.5 wt % H2SO4 at 608C, expressed as a Sh/Sc0.356Re relationship.
Downloaded on 2016-09-18 to IP 130.203.136.75 address. Redistribution subject to ECS terms of use (see ecsdl.org/site/terms_use) unless CC License in place (see abstract).
3642
Journal of The Electrochemical Society, 147 (10) 3637-3646 (2000)
S0013-4651(99)12-003-2 CCC: $7.00 © The Electrochemical Society, Inc.
Table III. Hydrodynamic data for nickel mass-transfer correlation in 93.5 wt % H2SO4 at 608C.
r 5 1.785 g/cm3 53
m 5 7.935 3 1022 g/cm s 54
n52
D 5 7.05 3 1027 cm2/s
d 5 1.27 cm
F 5 96,485 C/mol
Cs 5 1.39 3 1025 mol/cm3 44
Cb 5 0 mol/cm3
v
(rpm)
v
(cm/s)
il
(A/cm2)
Re
Sh*
Sc
1500
1000
1500
2000
133.23
166.46
199.70
132.93
1.01 3 1023
1.57 3 1023
2.34 3 1023
2.86 3 1023
1952
1904
2855
3807
1680
1056
1571
1919
62,900
62,900
62,900
62,900
extrapolated from existing data, and it is possible that this value
under- or overpredicts the real solubility.
Further support of the proposed mass-transfer correlation (Eq. 5)
was provided by measuring the corrosion potential transient behavior as a function of the bulk nickel cation concentration (see Fig. 8).
An increase in the bulk concentration from 0 to 500 ppm progressively increased the potential transition time from ,40 to ,150 min.
This is the expected response for a mass-transport-controlled NiS
dissolution process with the diffusion of nickel cations as the ratelimiting step. An increase in the bulk concentration from 0 to
250 ppm by weight had no effect on the stable potential after the
plateau. A further increase in the bulk concentration to 500 ppm by
weight, however, increased the stable potential after the transition
from ,20.60 to ,20.45 VPt. Again, this is consistent with a masstransport-controlled corrosion process with the diffusion of nickel
cations as the rate-limiting step. It is unclear whether the bulk concentration has an effect on the maximum potential attained since the
variance may be experimental uncertainty. The time required to
reach the maximum potential is essentially independent of the bulk
nickel ion concentration.
oxidative dissolution of the deposit and an appreciable bulk electronic conductivity. Metal sulfides such as galena (PbS), pyrite
(FeS2), and covellite (CuS) oxidatively dissolve in the presence of
an oxidant with the formation of a metal cation and elemental sulfur.48-50 The aforementioned metal sulfides are semiconductors, and
their respective resistivity is sufficiently low to permit participation
in galvanic dissolution couples. Peters et al.51 demonstrated that a
galvanic-induced enhanced leaching of galena (PbS) occurs when in
electrical contact with pyrite (FeS2). Another excellent example of a
galvanic interaction involving mineral sulfides is the enhanced
leaching of a chalcopyrite (CuFeS2) bearing material when in electrical contact with pyrite (FeS2).52
Considering the difference in the equilibrium dissolution potentials, E o 5 10.35 VSHE for the anodic dissolution of NiS
NiS 5 Ni21 1 S 1 2e2
[6]
and E o 5 20.25 VSHE, of the anodic dissolution of nickel
Ni 5 Ni21 1 2e2
[7]
Influence of NiS on resultant corrosion potential.—A possible
mechanism through which the NiS deposit influences the corrosion
potential of nickel involves a galvanic interaction between the deposit and the uncovered metal. A galvanic interaction requires an
it is reasonable to assume that the NiS-H2SO4 corrosion potential is
more noble than the Ni-H2SO4 corrosion potential. Therefore, during a galvanic interaction, NiS would be the cathodic phase and
Figure 8. Influence of the bulk solution nickel cation concentration on the
corrosion potential of a nickel RCE (1000 rpm) in 93.5 wt % H2SO4 at 608C
as a function of time immediately after immersion.
Figure 9. Anodic polarization behavior of a nickel RCE (1000 rpm) without
NiS on its surface superimposed with cathodic polarization of nickel RCE
(1000 rpm) with NiS on its surface in 93.5 wt % H2SO4 at 608C.
Downloaded on 2016-09-18 to IP 130.203.136.75 address. Redistribution subject to ECS terms of use (see ecsdl.org/site/terms_use) unless CC License in place (see abstract).
Journal of The Electrochemical Society, 147 (10) 3637-3646 (2000)
3643
S0013-4651(99)12-003-2 CCC: $7.00 © The Electrochemical Society, Inc.
nickel the anodic phase. From an electronic conduction point of
view, NiS should be capable of sustaining a galvanic interaction
since its bulk electrical resistivity is 2-4 3 1027 V m.45
Support for a galvanic interaction is provided by superimposing
the anodic polarization of a nickel RCE without NiS on its surface
with the cathodic polarization of 93.5 wt % H2SO4 on a nickel RCE
with NiS on its surface for various magnitudes of the cathode/anode
surface area ratio, u (see Fig. 9). As u is increased from 0.01 to 1, the
galvanic potential, Eg, the potential at which the anodic current
equals the cathodic current, increases from ,20.62 to 20.28 VPt.
The series of curves demonstrates how the potential of nickel could
change in a galvanic interaction with NiS during the growth and dissolution of the NiS deposit.
Discussion
Corrosion mechanism of nickel in 93.5 wt % H2SO4.—A possible mechanism through which nickel corrodes in 93.5 wt % H2SO4
is depicted in Fig. 10. The mechanism consists of the following four
stages.
NiS formation.—Immediately after immersion, nickel corrodes at a
relatively high rate with the formation of H2S according to the following reaction sequence
Anode:
4Ni 5 4Ni21 1 8e2
Cathode: 9H2SO4 1 8e2 5 4SO422 1 H2S 1 4H3O1 1 4HSO2
4 [8]
Total:
4Ni 1 9H2SO4 5 4NiSO4 1 H2S 1 4H3O1 1 4HSO42
The cathodic reaction is strongly polarized to achieve the corrosion
current, as indicated by the negative corrosion potential, albeit unstable, measured immediately upon immersion. The degree of polarization is sufficient to promote the formation of H2S, which is the primary cathode reaction (H2SO4 reduction) at such a degree of polarization.17 The relatively high reaction rate and solution viscosity
result in a rapid buildup of H2S at the metal-solution interface. The
cathode formation of H2S from the reduction of H2SO4 molecules
provides a source of the required sulfide anions.
Once formed, H2S provides an alternative reaction path for surface nickel atoms. Thus, during corrosion, nickel atoms can either
Figure 10. Schematic of nickel corrosion mechanism in 93.5 wt % H2SO4 at 608C.
Downloaded on 2016-09-18 to IP 130.203.136.75 address. Redistribution subject to ECS terms of use (see ecsdl.org/site/terms_use) unless CC License in place (see abstract).
3644
Journal of The Electrochemical Society, 147 (10) 3637-3646 (2000)
S0013-4651(99)12-003-2 CCC: $7.00 © The Electrochemical Society, Inc.
anodically dissolve with the formation of a soluble cation according
to Reaction 7 or they can anodically dissolve with the formation of
insoluble NiS according to
Ni 1 H2S 5 NiS 1 2H1 1 2e2
[9]
Therefore, after the initial formation of H2S, the corrosion of nickel
occurs according to the following reaction sequence, which makes
use of the relation 2H1 1 SO422 5 H2SO4
3Ni 5 3Ni21 1 6e2
Anode:
Ni 1 H2S 5 NiS 1 2H1 1 2e2
[10]
1 1 4HSO2
Cathode: 9H2SO4 1 8e2 5 4SO22
1
H
S
1
4H
O
4
2
3
4
Total:
4Ni 1 8H2SO4 5 3NiSO4 1 NiS 1 4H3O1 1 4HSO2
4
resulting in the formation of a NiS surface deposit. The solid-state
mechanism is consistent with the formation of an initial strongly
adhered surface deposit and an initial corrosion potential transient
that is essentially independent of the electrode’s rotational speed.
It is rather difficult to comment on the nucleation process because heterogeneities may catalyze nucleation, and because of the
complexity of the process when heterogeneities are not involved. It
is reasonable to consider, however, that nucleation involves a number of nuclei one monolayer thick on the nickel surface from which
the growth stage commences.
NiS growth-galvanic polarization stage.—The growth of NiS following nucleation involves the formation of a thick, noncontinuous
film on the surface. A thick film is a likely result of the reasonable
good electronic and ionic conductivity of the NiS phase. A noncontinuous film is a likely result of the excess of nickel equivalents dissolving a soluble cations (Ni21) coupled with the limited supply of
the reactant H2S at the metal-solution interface (discussed below).
Once its growth stage commences, NiS establishes a galvanic
interaction with the nickel substrate. During the interaction, the cathode reaction involved in the corrosion of NiS
4H2SO4 1 2e2 5 SO422 1 SO2 1 2H3O1 1 HSO2
4
[11]
is enhanced on NiS sites, whereas the anode reactions involved in
the dissolution of nickel (Reactions 7 and 9) are enhanced on nickel
sites. The growth of NiS continues despite the discontinued cathode
production of H2S due to the presence of excess H2S at the metalsolution interface and terminates once Reaction 9 consumes all the
excess H2S. The measured apparent corrosion potential of nickel is
now the galvanic potential of the NiS-Ni couple and is a function of
both the cathode and anode kinetic parameters and u.
NiS dissolution-galvanic depolarization stage.—The magnitude of u
at the instant the growth process terminates is relatively large since
the noncontinuous NiS film covers the majority of the nickel surface.
The large u coupled with the high cathodic current for the reduction
of H2SO4 molecules on NiS produce a galvanic situation where the
galvanic potential is controlled essentially at the corrosion potential
for the oxidative dissolution of NiS. Therefore, the anode reaction
NiS 5 Ni21 1 S 1 2e2
[12]
takes place with a significant rate along with the cathode reaction
(Reaction 11) on NiS. The significance of this result is that the dissolution of the Ni-NiS galvanic interaction proceeds according to the
following reaction sequence
Anode (Ni):
Anode (NiS):
Ni 5 Ni21 1 2e2
NiS 5 Ni21 1 S 1 2e2
Cathode (NiS): 8H2SO4 1 4e2 5 2SO22
4 1 2SO2
[13]
1 4H3O1 1 4HSO2
4 00 0
Total:
Ni 1 NiS 1 8H2SO4 5 2NiSO4 1 S 1 2SO2
1 4H3O1 1 HSO2
4
As NiS dissolves during galvanic corrosion, u decreases and thus
the galvanic potential decreases. Figure 9 demonstrates how the galvanic potential changes with u during the dissolution stage. As u decreased from 1 to 0.01, the galvanic potential decreased from
,20.28 to 20.62 VPt. Considering the potential transients measured
during the corrosion of nickel, the finite residence time of NiS results from its mass-transport-controlled dissolution rate being significantly slower than its activation-controlled growth rate.
Steady-state stage.—The galvanic interaction no longer exists once
all the NiS dissolves. At this point, the steady-state corrosion of nickel occurs, proceeding according to the following reaction sequence
Anode:
3Ni 5 3Ni21 1 6e2
Cathode: 8H2SO4 1 6e2 5 3SO422 1 S 1 4H3O1 1 4HSO2
4 [14]
Total:
3Ni 1 8H2SO4 5 3NiSO4 1 S 1 4H3O1 1 4HSO2
4
Without NiS on the surface, the cathode reaction occurs once
again on the nickel surface. Since the anode reaction is now mass
transport controlled, not activation controlled as in the NiS formation
stage, it is the kinetically slowest reaction and controls the overall
rate of corrosion. The cathode reaction in this stage depolarizes, as
compared to the NiS formation stage, since it does not need to be driven as hard to balance the current. The extent of the depolarization is
sufficient such that the primary cathode reaction product at the corrosion potential now becomes elemental sulfur and not H2S, as with
the NiS formation stage. Therefore, without the generation of H2S as
the primary cathode reaction product, NiS does not reform.
Modeling the galvanic potential behavior.—According to the galvanic corrosion model, Eg is a function of u, which is a function of
time. Therefore, by modeling the NiS growth and dissolution kinetics
and determining the corresponding Eg transient, the galvanic coupling model can be validated. As pointed out above, the exact NiS
nucleation and growth mechanism is unknown and is rather difficult
to determine experimentally. Nonetheless, an attempt based on the
physical situation described below for nickel corrosion in 93.5 wt %
H2SO4 at 608C successfully reproduced the general form of the Eg-t
dependence and thus supported the concept of a galvanic interaction.
As a first approximation, it is reasonable to consider that the
growth of NiS is controlled by the rate at which nickel atoms anodically dissolve to form NiS according to Reaction 9. The corresponding growth rate of the anodic deposit can be expressed mathematically as follows
dV
IM
5
dt
nFr
[15]
Assuming that NiS grows as a disk with radius r and constant height
h on a nickel surface with radius ro (constant), the change in volume
at a constant height is governed by the change in radius according to
dV 5 phrdr
[16]
When metal dissolution controls the anodic growth rate, the current
is given by
I 5 i(Ao 2 A)
[17]
Substituting Eq. 16 and 17 into Eq. 15, and integrating and rearranging, gives
 t 
r 2 5 ro2 2 exp 2 
 tg 
[18]
where tg 5 nFrh/2iM.
Considering now the NiS dissolution kinetics, it is reasonable to
assume that the rate is controlled by the diffusion of nickel ions
(Ni21) into solution from the NiSO4 film-solution interface. The
anodic polarization behavior of the nickel RCE and the effect of the
electrode’s rotational speed on the weight loss of a nickel RCE support the assumption of mass-transport-controlled dissolution. When
Downloaded on 2016-09-18 to IP 130.203.136.75 address. Redistribution subject to ECS terms of use (see ecsdl.org/site/terms_use) unless CC License in place (see abstract).
Journal of The Electrochemical Society, 147 (10) 3637-3646 (2000)
3645
S0013-4651(99)12-003-2 CCC: $7.00 © The Electrochemical Society, Inc.
mass transport controls the dissolution rate, the change in volume
with time is given by
r dV
DAo
5
(Cs 2 Cb )
M dt
d
[19]
Assuming NiS dissolves as a disk of initial radius rm and constant
height h on a nickel surface of radius ro and Cb 5 0, then substituting Eq. 16 and 19, integrating, and rearranging gives
r 2 5 rm2 2
1 t
ro2 t d
[20]
where td 5 rdh/DMCs.
The corrosion potential of a nickel RCE, when rotated at
1000 rpm is 93.5 wt % H2SO4 at 608C, reached a maximum after
about 12 min exposure and then proceeded to decay, first slowly and
then more rapidly, for about 36 min before attaining its steady state
(Fig. 5). It is reasonable to postulate that the growth of NiS occurred
during the initial 12 min period, whereas the dissolution of NiS
occurred during the next 36 min period. By assuming that the growth
rate is significantly higher than the dissolution rate at all times during growth such that the dissolution rate is neglected, then Eq. 18
determines the Eg-t dependence during the initial 12 min exposure
(NiS growth) and Eq. 20 determines the Eg-t during the next 36 min
(NiS dissolution). The Eg-t dependence corresponding to either
Eq. 18 or 20 can be determined graphically using the nickel and
NiS(Ni) polarization curves as shown in Fig. 9.
It is believed that the NiS deposit forms a noncontinuous film on
the surface of nickel, and, therefore, u attains a maximum value um.
Based on the aforementioned assumptions, u reaches um at the point
at which the growth process terminates; i.e., after 12 min exposure.
For the graphical determination of the Eg-t dependence, it is assumed
that um 5 3, which corresponds to a surface coverage, A/Ao 5 0.75.
Figure 11 shows u, derived from Eq. 18 and 20 using the assumptions described above and E g, determined graphically using Fig. 9, as
a function of time during the growth and subsequent dissolution of the
NiS deposit. Superimposed in the Eg-t plot is the measured potential
transient of nickel when rotated at 1000 rpm in 93.5 wt % H2SO4 at
608C. The theoretical and experimental curves are clearly of the same
form, and the agreement is good except at long times. It appears that
the galvanic interaction theory fit the facts reasonably well despite the
fact that the exact NiS nucleation and growth process is unknown.
Conclusions
Potentiodynamic polarization and corrosion potential studies on
a nickel RCE complemented with ex situ X-ray spectroscopic measurements led to the following conclusions regarding the corrosion
mechanism of nickel in hot concentrated H2SO4.
1. Concentrated 93.5 wt % H2SO4 acts as an oxidizing acid when
in contact with nickel. The cathodic process involves the reduction
of H2SO4 molecules to various sulfur-containing species, SO2, elemental sulfur, and H2S. Corrosion leads to the formation of several
surface products including NiS, NiSO4, and elemental sulfur, none
of which is protective.
2. Corrosion of nickel is a mass-transport-controlled process with
the convective diffusion of nickel cations from a saturated layer of
NiSO4 as its rate-determining step.
3. Prior to establishing a steady-state, the corrosion of nickel proceeds with the formation and subsequent dissolution of a NiS
deposit. When present on the surface, NiS establishes a galvanic
interaction with the uncovered metal, significantly polarizing the
nickel anodic dissolution reaction. The anodic dissolution rate is
essentially unaffected during the galvanic-induced polarization since
the rate is mass transport controlled.
Acknowledgments
The financial support of the Natural Science and Engineering
Council of Canada is gratefully appreciated, both in the provision of
experimental supplies and in the provision of a student stipend. The
experimental assistance of M. J. Graham, G. I. Sproule, and P.
Figure 11. Theoretical changes during growth and dissolution of NiS deposit
on nickel in 93.5 wt % H2SO4 at 608C (1000 rpm). (a) u as a function of time,
(b) NiS-nickel galvanic potential as a function of time superimposed with the
measured corrosion potential of a nickel RCE under identical conditions.
Schmuki of the National Research Council Canada and J. McAndrew of McMaster University during the XPS and XRF analyses,
respectively, is gratefully appreciated.
The Pulp and Paper Research Institute of Canada assisted in meeting the
publication costs of this article.
List of Symbols
Mass transfer correlation
Cb
bulk solution concentration, mol/m3
Cs
saturation concentration, mol/m3
d
diameter of RCE, m
D
diffusion coefficient, m2/s
F
Faraday’s constant, 9.6485 3 104 C/equiv
il
limiting anodic current density, A/m2
k
mass transfer coefficient, m/s
n
number of electrons in reaction
Re
Reynolds number, rvd/m
Sc
Schmidt number, m/rd
Sh
Sherwood number, kd/D
v
fluid velocity, vd/2, m/s
m
fluid viscosity, poise
r
fluid density, kg/m3
v
rotational speed, rad/s
Kinetic modeling
A
area covered by deposit, m2
Ao
total area, m2
Cb
bulk solution concentration, mol/m3
Cs
saturation concentration, mol/m3
D
diffusion coefficient, m2/s
Eg
galvanic potential, V
F
Faraday’s constant, 9.6485 3 104 C/equiv
Downloaded on 2016-09-18 to IP 130.203.136.75 address. Redistribution subject to ECS terms of use (see ecsdl.org/site/terms_use) unless CC License in place (see abstract).
3646
Journal of The Electrochemical Society, 147 (10) 3637-3646 (2000)
S0013-4651(99)12-003-2 CCC: $7.00 © The Electrochemical Society, Inc.
h
i
I
M
n
r
rm
ro
t
V
d
r
u
td
tg
height of disk cap, m
anodic current density of uncovered metal, A/cm2
anodic current of uncovered metal, A
molecular weight of deposit, kg/mol
number of electrons in reaction
radius of disk deposit, m
maximum radius of disk deposit, m
radius of metal surface, m
time, s
volume of disk deposit, m3
diffusion layer thickness, m
density of deposit, kg/m3
surface area ratio, A/(Ao-A)
dissolution time constant, rdh/DMCs
growth time constant, nFrh/2iM
References
1. M. Davies, Mater. Perform., 30, 57 (1991).
2. C. P. Dillon, in Materials Selector for Hazardous Chemicals, Vol. 1: Concentrated
Sulfuric Acid and Oleum, MTI Publication MS-1, MTI, St. Louis, MO (1997).
3. C. M. Schillmoller, Chem. Eng. Prog., 94, 38 (1998).
4. R. Matsuhashi, E. Sato, S. Abe, and H. Abo, Corros. Eng., 39, 89 (1991).
5. J. R. Kish, B. Ives, and J. R. Rodda, in Proceedings of the 13th International Corrosion Congress, Australasian Corrosion Association, Melbourne, Australia, p. 812,
(CD-ROM Version) (1996).
6. G. M. Cameron, Chem. Eng. Prog., 78, 71 (1982).
7. D. R. McAlister, A. G. Corey, L. J. Ewing, and S. A. Ziebold, Chem. Eng. Prog.,
77, 34 (1986).
8. T. F. Young, L. F. Maranville, and H. M. Smith, in The Structure of Electrolyte Solutions, W. J. Hammer, Editor, John Wiley, New York (1959).
9. N. G. Zarakhani and M. I. Vinnik, Zh. Fiz. Khim., 36, 916 (1962).
10. R. J. Gillespie and E. A. Robinson, in Advances in Inorganic Chemistry and Radiochemistry, Vol. 1, p. 385, Academic Press, New York (1959).
11. R. J. Gillespie and E. A. Robinson, in Non-Aqueous Solvent Systems, T. C.
Waddington, Editor, p. 117, Academic Press, New York (1965).
12. R. J. Gillespie, in Inorganic Sulphur Chemistry, G. Nickless, Editor, p. 563, Elsevier, New York (1968).
13. N. Sridher, Mater. Perform., 27, 40 (1988).
14. B. Stypula and J. Banas, Electrochim. Acta, 38, 2309 (1993).
15. H. Hoffman, Z. Elektrochem., 27, 442 (1921).
16. F. Beck, Electrochim. Acta, 17, 2317 (1972).
17. J. R. Kish, Ph.D. Thesis, McMaster University, Hamilton, ON, Canada (1999).
18. H. A. Garrera, J. W. S. Carrozza, and A. J. Arvía, Electrochim. Acta, 13, 771 (1968).
19. M. G. Fontana and N. D. Greene, Corros. Eng., McGraw-Hill, New York (1967).
20.
21.
22.
23.
24.
25.
26.
27.
28.
29.
30.
31.
32.
33.
34.
35.
36.
37.
38.
39.
40.
41.
42.
43.
44.
45.
46.
47.
48.
49.
50.
51.
52.
53.
54.
C. W. Rogers, J. Chem. Soc., 254 (1926).
E. G. Kremko, I. A. Kakovskii, and L. D. Sheveleva, Prot. Met., 20, 73 (1985).
G. Gilli and F. Zucchii, Corros. Sci., 8, 801 (1968).
L. A. Poluboyarteseva, P. I. Zarubin, and V. M. Vobakovskii, J. Appl. Chem. USSR,
36, 1210 (1963).
B. T. Ellison and W. R. Schmeal, J. Electrochem. Soc., 125, 524 (1978).
T. N. Andersen, N. Vanorden, and W. J. Schlitt, Met. Trans. A., 11A, 1421 (1980).
S. W. Dean and G. D. Grab, Mater. Perform., 24, 22 (1985).
S. L. Pohlman and T. N. Andersen, Mater. Perform., 26, 41 (1987).
U. Ebersbach, K. Schwabe, and K. Ritter, Electrochim. Acta, 12, 927 (1967).
G. Gilli, P. Borea, F. Zucchi, and G. Trabanelli, Corros. Sci., 9, 673 (1969).
M. Turner, G. E. Thomson, and P. A. Brook, Corros. Sci., 13, 985 (1973).
W. D. Halsted and B. F. Lovey, J. Appl. Chem. Biotechnol., 27, 585 (1977).
A. J. Arvía and J. W. S. Carrozza, Electrochim. Acta, 11, 1641 (1966).
J. W. S. Carrozza, H. A. Garrera, and A. J. Arvía, Electrochim. Acta, 14, 205 (1969).
A. J. Arvía, H. A. Garrera, and J. W. S. Carrozza, Electrochim. Acta, 16, 79 (1971).
O. L. Riggs, Jr. and C. E. Locke, Anodic Protection Theory and Practice in the Prevention of Corrosion, Plenum Press, New York (1981).
M. Eisenberg, C. W. Tobias, and C. R. Wilke, J. Electrochem. Soc., 101, 306
(1954).
M. Rahmani and J. E. Strutt, Hydrodynamic Modelling of Corrosion of Carbon
Steel and Cast Irons in Sulfuric Acid, MTI Publication No. T-2, NACE, Houston,
TX (1992).
J. R. Kish, M. B. Ives, and J. R. Rodda, To be published.
D. C. Silverman and M. E. Zerr, Corrosion, 42, 633 (1986).
W. T. Hall, in Analytical Chemistry, Vol. II, 9th ed., John Wiley & Sons, New York
(1955).
F. D. Miles and T. Carson, J. Chem., 786 (1946).
J. Chastain, Handbook of X-Ray Photoelectron Spectroscopy, Perkin-Elmer, Eden
Prairie, MN (1992).
CRC Handbook of Chemistry and Physics, 46th ed., CRC Press, Cleveland (196566).
W. D. Halstead and B. F. Lovey, J. Appl. Chem. Biotechnol., 27, 585 (1977).
J. C. Hines and R. C. Williamson, Corros. Sci., 4, 201 (1964).
J. C. Hines and R. C. Williamson, Corros. Sci., 4, 221 (1964).
C. R. Wilke, Chem. Eng. Prog., 45, 218 (1949).
D. F. A. Koch, in Modern Aspects of Electrochemistry, No. 10, J. O’M. Bockris,
Editor, p. 211, Plenum, New York (1984).
N. E. Wadsworth, in Hydrometallurgical Process Fundamentals, R. G. Bautista,
Editor, p. 41, Plenum, New York (1984).
K. Osseo-Asare, Hydrometallurgy, 29, 61 (1992).
E. Peters, I. H. Warren, and H. Veltman, First Tutorial Symposium on Hydrometallurgy, The University of Denver, Denver, CO (1972).
V. K. Berry, L. E. Murr, and J. B. Hiskey, Hydrometallurgy, 3, 209 (1978).
International Critical Tables, Vol. 5, McGraw-Hill, New York (1929).
O. T. Fasulo, Sulfuric Acid-Use and Handling, McGraw-Hill, New York (1965).
Downloaded on 2016-09-18 to IP 130.203.136.75 address. Redistribution subject to ECS terms of use (see ecsdl.org/site/terms_use) unless CC License in place (see abstract).