Wear 376-377 (2017) 12981306

Contents lists available at ScienceDirect

Wear
journal homepage: www.elsevier.com/locate/wear

Micro abrasion-corrosion of ferritic stainless steels

W.S. Labiapari a,b, M.A.N. Ardila a, H.L. Costa a, J.D.B. de Mello a,n
a
Universidade Federal de Uberlndia, Uberlndia, Brazil
b
Aperam South America, Brazil

art ic l e i nf o a b s t r a c t

Article history: The synergistic effect between abrasion and corrosion has been widely investigated and debated, where
Received 21 December 2016 according to some works this synergy accentuates wear, and for others attenuates it. Despite the con-
Received in revised form siderable effort to understand the synergy between abrasion and corrosion, little progress has been made
19 January 2017
to understand this phenomenon for ferritic stainless steels. This paper analyses the micro abrasion-
Accepted 20 January 2017
corrosion performance of ferritic stainless steel with different chemical compositions (11%wt Cr with and
without Ti stabilization; 16%wt Cr with and without Nb stabilization) and, for comparative purposes,
Keywords: austenitic stainless steel (18%wt Cr
8%wt Ni) and carbon steel (0.2%wt C). The specimens were tested
Micro abrasion-corrosion for corrosion (turbulent and aerated environment), micro abrasion and micro abrasion-corrosion. For the
Synergy
corrosion tests, a 1 N H2SO4 solution was used, a 10%wt mixture of SiO2 in distilled water for the micro
ferritic stainless steels
abrasion tests, and an abrasive-corrosive environment of 10%wt SiO2 in 1 N H2SO4 for the micro abrasion-
Friction coefcient
Ti and Nb stabilization corrosion tests. In the micro abrasive wear tests there was not a clear trend between the different ma-
terials tested, despite differences in their chemical composition, mechanical properties and micro-
structure. On the other hand, their performance was ruled by their chemical composition, in particular
the Chromium content, under abrasion-corrosion conditions. This indicated the predominant role played
by corrosion in abrasive-corrosive environments in this particular tribo-system. For all materials tested,
micro abrasion wear coefcients were higher (4  ) than those measured under abrasion-corrosion
conditions. Friction coefcients could also be measured by a 3D load cell strategically positioned in the
specially developed micro abrasion-corrosion device, showing a strong reduction (2  ) in friction
coefcient under abrasion-corrosion conditions when compared with solely abrasion conditions. This
was attributed to the formation of a corrosion product, mainly constituted of iron sulphate (identied via
FTIR), which has lubricant properties.
& 2017 Elsevier B.V. All rights reserved.

1. Introduction passivating layers present in the metal surface. Exposed metal

In various engineering applications, mechanical components
are simultaneously subjected to a combination of mechanical wear
and corrosion [1,2]. Due to their high corrosion resistance, stain-
less steels could be interesting candidates, despite their relatively
low mechanical resistance when compared to other hard wear-
resistant materials. For example, in conditions of moderate cor-
rosion in combination with mechanical wear, such as in the initial
stages of sugar cane plants for ethanol production, a previous work
proved an exceptional performance of inexpensive ferritic stain-
less steels at low cost-benet ratios [3].
In systems where corrosion and mechanical wear occur si-
multaneously, phenomena such as plastic deformation inuence
mechanical failure by wear, but also lead to the removal of
n
Corresponding author. Tel.: 55 34 3239 4409; fax: 55 34 3239 4273.
E-mail address: demello@ufu.br (J.D.B. de Mello).
surfaces may be highly reactive to the environment, which can
accelerate corrosion [4]. Moreover, the presence of stresses chan-
ges the electrochemical response of mechanical components and
structures [5].
Typically, corrosion and abrasion are investigated separately.
However, attempts to include the inuence of a corrosive en-
vironment on wear via a mathematical approach have incurred in
great dispersion and considerable errors [6]. According to Wood
et al. [7], these errors may be reduced for tests that impose the
combined effect of corrosion and wear in situ. In this case, the
mechano-electrochemical results dispense the mathematical ap-
proach for the synergistic effects. The combined effect between
abrasion and corrosion, the so-called synergistic effect, has been
widely debated in the literature. According to some authors,
abrasion accentuates corrosion [2,611], whereas for others abra-
sion attenuates it [9,10]. However, even if considerable effort has
been done to understand the synergy between abrasion and

http://dx.doi.org/10.1016/j.wear.2017.01.083
0043-1648/& 2017 Elsevier B.V. All rights reserved.
 W.S. Labiapari et al. / Wear 376-377 (2017) 12981306
corrosion, little progress has been made to quantify this phe-
nomenon for stainless steels [6].
The corrosion resistance of stainless steels is attributed to the
formation of a protective passivated layer and they are generally
regarded as materials that are easily repassivated. The dynamics
involving removal of the passive layer by mechanical action and
repassivation plays an important role on the abrasion-corrosion
resistance of stainless steels. The complex tribochemical me-
chanisms of stainless steels depend on the microstructure and
chemical composition of the material surface, the solution pH, the
abrasive (size, type and concentration), and the imposed electro-
chemical conditions [7].
Bello et al. [6] performed abrasion-corrosion tests for AISI 304,
AISI 316 and duplex stainless steel using SiC (diameter 4.5 mm)
abrasive in a 3.5%wt sodium chloride solution in distilled water.
The abrasion corrosion tests showed higher passivation current
densities than pure corrosion tests and this was attributed to the
removal of the passive layer.
Another point regarding abrasion corrosion tests of passivating
materials such as stainless steels is that they usually show more
dispersion of the electrochemical current than pure corrosion
tests, which is partly attributed to competition between the re-
moval of the passive layer and repassivation [7].
Several studies have measured abrasion-corrosion for bioma-
terials and various coatings [7,8,10], including CoCrMo alloys and
duplex stainless steels, materials that are easily repassivated.
Those works showed higher current densities under abrasion-
corrosion conditions when compared with pure corrosion condi-
tions, due to removal or damage of the passive layer.
For carbon steels, current density increases with the applied
normal load, showing the abrasive effect on the passive lm [2].
For WC-metal composites tested using basic slurries, the me-
tallic matrices that formed passive layers showed higher resistance
under abrasion-corrosion conditions when compared with pure
abrasion conditions, i.e., a negative synergy between abrasion and
corrosion. On the other hand, for the matrices where corrosion
resulted in the loss of matrix support for the carbides, resistance
decreased under abrasion-corrosion conditions, showing a positive
synergy between abrasion and corrosion [9].
Most studies about abrasion-corrosion resistance of stainless
steels are mainly focused on more traditional materials, such as
AISI 304 and AISI 316 stainless steels, and, more recently, on AISI
2205 duplex stainless steels. The investigation of cost effective
ferritic stainless steels for those applications is often neglected.
Despite this, they have found an important application niche in
the biofuel industry [3], which certainly involves abrasion and
corrosion. Within ferritic stainless steels, it is relevant to in-
vestigate the effect of Cr content on tribocorrosion, as well as the
effect of stabilization. The cheapest stainless steel is 11Cr (DIN
14003) and it has vast application in the sugar-cane biofuel in-
dustry. 11CrTi (ASTM S40910) is also a low-cost ferritic stainless
steel stabilized with Ti, largely used in automobile exhausts sys-
tems. 16Cr (ASTM S43000) steels are slightly more expensive fer-
ritic stainless steel, mainly used in the cutlery industry, but they
are still cheaper than austentic steels. 16CrNb (ASTM S43000) is
the same stainless steel stabilized with Nb, used in cutlery and
stamping. On the other hand, austenitic stainless steels, such as
Table 1
Industrial hot rolling and annealing conditions.
Specimen 11Cr DIN 14003
Final hot rolling temperature (C) 10007 50
Annealing soaking temperature (C) 7607 50
Annealing time 8h
1299
18Cr8Ni (AISI 304), present higher cost, but are extremely versatile
in their use, with high corrosion resistance, good formability and
weldability.
This paper analyses the micro abrasion-corrosion performance
of ferritic stainless steel with different chemical compositions:
 11%wt Cr with (11CrTi) and without Ti (11Cr) stabilization;
 16%wt Cr with (16CrNb) and without Nb (16Cr) stabilization;
 For comparative purposes, one austenitic stainless steels with
18%wt Cr and 8%wt Ni (18Cr8Ni) and one carbon steel with 0.2%
wt C (ASTM A36).
The specimens were characterized in terms of corrosion, micro-
abrasion and micro-abrasion-corrosion. For all materials tested,
friction coefcients were measured by a 3D load cell.
2. Methodology
2.1. Materials
Four ferritic stainless steels were selected for this work and
were compared with one austenitic stainless steel for its great
versatility in terms of use. The ferritic stainless steels had different
Cr contents and were tested both in stabilized (with Ti or Nb) and
non-stabilized conditions. All the stainless steel specimens were
used after industrial hot rolling and annealing (see hot rolling
temperatures and annealing conditions in Table 1). The stainless
steels were also compared with a commercial low carbon steel,
ASTM A36, in the hot rolled condition, due to its low cost and large
applicability in engineering applications.
Their chemical composition, which was evaluated by different
techniques [infrared absorption, (Leco, CS444s), thermo con-
ductivity (Leco, TC436s), X-ray uorescence spectrometry (Ther-
mo ARL, 9900) and optical emission spectrometry (Thermo ARL,
4460)] is presented in Table 2.
The specimens were thoroughly characterized in terms of
mechanical properties [tensile tests (Instron, 5583) and Vickers
hardness (Instron Wolpert, Testor 930s)] and microstructure, de-
tails can be found in [12]. Table 3 summarizes the main mechan-
ical properties of the materials tested.
For each material, samples were cut into coupons with di-
mensions of 35x25  5 mm3 and then sanded (grit sizes of 220 and
600). After this step, they were subjected to ultrasonic cleaning in
acetone for 15 min and dried. All materials were tested for pure
corrosion, pure microabrasion and microabrasion-corrosion.
2.2. Corrosion tests
A standard 2 cm2 of active area for the corrosion and abrasion-
corrosion tests was obtained using nail varnish and wax. A copper
wire bonded by a silver glue established the electrical contact. To
ensure the solidication and xation of the wax on the specimen
and consequently the reproducibility of the results, the entire
preparation process occurred at least 8 h before each test.
The electrolyte was a 1 N H2SO4 (F. Maia Ind. e Com. Ltda -
Brazil) solution in distilled water. A platinum gauze was the
11CrTi ASTM S40910 16Cr 16CrNb ASTM S43000 18Cr8Ni AISI 304
950 7 50 1000 750 950 7 50 11007 50
950 7 30 780 720 930 7 20 10407 20
30 s 8h 30 s 30 s
1300 W.S. Labiapari et al. / Wear 376-377 (2017) 12981306

Table 2 0.7
0.56
Chemical composition of the specimens.
0.59 0.60
0.6
0.53 0.51
Specimen 11Cr 11CrTi 16Cr 16CrNb 18Cr8Ni A36 0.51

Friction Coefficient
0.5
C (%wt) 0.011 0.009 0.049 0.025 0.055 0.138
Cr (%wt) 11.23 11.29 16.10 16.19 18.28 0.01 0.4
Ni (%wt) 0.31 0.12 0.27 0.19 8.01 0.01
Nb (%wt) 0.006 0.002 0.014 0.416 0.005 0.001 0.3
Ti (%wt) 0.003 0.144 0.003 0.004 0.001 0.001
N (%wt) 0.0145 0.0087 0.0528 0.0202 0.0421 0.0026 0.2

0.1
Table 3
Mechanical properties of the materials tested. 0.0
18Cr8Ni A36 16CrNb 11CrTi 16Cr 11Cr
Specimen Vickers hard- Yield stress Ultimate Elongation to Materials
ness (MPa) (MPa) stress (MPa) fracture (%)
Fig. 3. Average friction coefcients during the pure microabrasion tests.
11Cr 14917 39 323 7 6 4127 2 38 71
11CrTi 14717 29 3167 4 4007 1 40 72 area, creating a saline bridge. The anodic polarization curve was
16Cr 1638 7 10 343 7 4 488 7 3 317 2
obtained using an increasing potential rate of 1 mV/s, which gives
16CrNb 1510 710 3367 3 4477 2 36 72
18Cr8Ni 19127 29 3477 9 7067 7 62 72 a test duration of 42 min. ASTM G5-94 (2004) [13] recommends a
ASTM A36 1402 7 10 305 7 4 4457 3 34 71 rate of 0.167 mV/s, which gives a test duration of 4 h. The reduc-
tion of the test time and consequent increase in the potential
variation rate was necessary because the subsequent micro-abra-
7 100
Fraction sion corrosion tests required a test time of around 40 min.
Accumulated 90 Therefore, the use of this higher rate ensures the imposition of
6
80 equivalent electrochemical conditions in both the corrosion and
5 abrasion-corrosion tests. Moreover, the rate of 1 mV/s has already
Accumulated [%]

70
Fractoin [%]

been used in other works [9,1416] or even higher values (2 mV/s)

60
4 as reported by Perret et al. [17].
50 For each condition tested, at least 3 repetitions were performed
3
40 to ensure the reproducibility of the results and the same applies to
30 the subsequent pure microabrasion and microabrasion-corrosion
2
tests.
20
1
10
2.3. Micro-abrasion tests
0 0
0.00
0.49

0.67
0.78
0.91
1.06
1.24
1.44

1.95
2.28
2.65
3.09

4.19

5.69
6.63
7.72
9.00
0.58

1.68

3.60

4.88

12.21
10.48

The so-called pure abrasion tests were carried out in a specially

Granulometry [mm] designed hybrid equipment, which allows both pure microabra-
sion and micro-abrasion-corrosion tests to be carried out. The test
Fig. 1. SiO2 particle size distribution analyzed by laser granulometry and abrasive
morphology by SEM.
rig consists of a xed-ball micro-abrasion tester, where the spe-
cimen is supported by a three-axis load cell, enabling the mea-
counter electrode and a saturated calomel electrode (SCE) was the surement of forces and moments acting on the contact. This test
reference electrode. The corrosion tests were carried out using a rig is well described in a previous paper [18].
Abrasive slurries composed of silica (SiO2) particles (Sigma-
potentiostat/galvanostat model Biologic SP150. During the tests,
Aldrich) suspended in distilled water, at a concentration of
the specimens were partially submerged in the solution. The same
10 g cm
3, were used. Distilled water can be considered as a
solution was then pumped at a rate of 1.7 ml/min over the exposed
neutral medium, leading to negligible corrosion of stainless steels,

2.0 10000
11Cr
3,030.0 3,158.7
11CrTi
1.5 16Cr
Potential [V] vs. SCE

16CrNb 1000
Ip [mA/cm2].10-3

1.0 18Cr8Ni
A36
0.5 100
35.7
0.0 19.8
11.5
10
-0.5

-1.0 1
1.E-05 1.E-03 1.E-01 1.E+01 1.E+03 18Cr8Ni 16Cr 16CrNb 11Cr 11CrTi
Current Density [mA/cm2] Materials
-a- -b-
Fig. 2. Abrasion behaviour. a- Typical potentiodynamic polarization curves. b- Average passivation current density.
 W.S. Labiapari et al. / Wear 376-377 (2017) 12981306 1301

60 60

50 50

k [m3/N.m].10-15
k [m3/N.m].10-15
40 40

30 30

20 20 Test 1
Test 2
Test 1 Test 3
10 Test 2 10
Test 3
0 0
0 3 6 9 12 15 18 21 24 27 30 33 0 3 6 9 12 15 18 21 24 27 30 33
Time [min] Time [min]
-a- -b-
Fig. 4. Examples of the evolution of the micro-abrasive wear coefcient k with time: (a) 11Cr, (b) 11CrTi.

20 in the abrasioncorrosion tests. For that, an electrolyte of 1 N

18 17.3 H2SO4 instead of distilled water was used in the abrasive slurry at
16 13.6 the same abrasive concentration (10%wt. SiO2 4.5 mm).
To minimize the effect of transients, the abrasioncorrosion
14
k [m3/N.m].10-15

12
10
8 corrosive slurry assured a pickling process that was able to remove
6
4
2 section.
0 After that, a potentiodynamic corrosion test was triggered, so
A36 11CrTi
Fig. 5. Average wear coefcients k.
so that the microabrasion tests carried out using this abrasive
slurry were considered pure microabrasion tests [18]. The granu-
lometric distribution of the abrasive particles, determined by laser
granulometry (Malvern Instruments - Mastersizers), is shown in
Fig. 1, where the abrasives have semi angular shape and an
asymmetric normal size distribution, mean particle size of
3.4 70.2 mm, 85% of the particles with diameter between 1 and
10 mm.
The counter body used was a zirconia ball (S Esferas- Brazil)
with diameter 25.4 mm. As reported by some authors [20] and
quantied by Costa et al. [21], the ball roughness strongly inu-
ence the wear mechanisms and wear rates. In order to maintain
this inuence negligible, the arithmetic surface roughness (Ra) was
controlled and kept in the range 0.32 mm o Ra o0.38 mm. The ro-
tary speed was set to 150 rpm, and the normal load was set to
1.42 N.
For each test, wear coefcients (k) were calculated as k (b4
/64R)/(s.N), where b is the diameter of the wear scar, R is the
sphere radius, s is the sliding distance and N is the normal load, as
proposed in [19]. The tests were interrupted at every 3 min to
evaluate the evolution of k with time. After the tests, the spherical
shape of the wear craters was veried by 3D laser interferometry,
since this equation for k only applies to spherical craters.
2.4. Micro-abrasion-corrosion tests
The microabrasion tests were carried out using the electro-
chemical cell present in the same hybrid equipment used for the
pure abrasion tests. The same imposed parameters used for the
separate abrasion and corrosion tests were applied simultaneously
12.7
tests were performed only after the abrasion process reached a
11.3
10.2 steady state condition. For that, the tests started without applying
9.0 any external potential difference during 15 min. The use of the
the passive layer. This initial time of 15 min was established as the
time necessary to reach a steady-state abrasion condition under
pure microabrasion conditions, as it will be shown in the Results
11Cr 16CrNb 16Cr 18Cr8Ni
that a micro-abrasion-corrosion test could be carried out. The
Materials conditions for the potentiodynamic scan were identical to those
used in the pure corrosion tests.
In order to verify whether corrosion occurs mainly in the wear
crater, after the abrasion-corrosion test, one sample (11Cr) was
immersed in a solution containing potassium ferricyanide, which
helps to evidence the corrosion products. In fact, the corrosion
products ow from the wear crater, evidencing that corrosion and
abrasion indeed occur in the same location [12].
After the tests, the surfaces were investigated by optical mi-
croscopy and scanning electron microscopy (SEM- Philips, XL-30s)
to identify wear mechanisms.
3. Results and discussion
3.1. Corrosion tests
Fig. 2 synthesizes the corrosion behaviour of the materials in-
vestigated in this work. In Fig. 2-a, typical potentiodynamic po-
larization curves for all studied steels are illustrated. Different
repetitions for each material tested showed very stable and re-
producible results, so that for clarity only one curve is presented
for each material. Fig. 2-b presents the average value of the pas-
sivation current density for the stainless steels. The arrow in this
gure indicates that materials with lower passivation current
densities exhibit higher corrosion resistance. There is a strong
inuence of the Chromium content on the passivation current
density and the shape of the passivation curves. The higher Cr
content (18Cr8Ni steel) induced the lower passivation current
whereas the higher passivation currents corresponded to the
lower Cr content regardless of the stabilization by Ti (11CrTi and
11Cr). The effect of stabilization by Nb on current density was also
negligible for the intermediate Cr content, 16%wt [22]. The low
1302 W.S. Labiapari et al. / Wear 376-377 (2017) 12981306

Fig. 6. SEM of wear craters after the pure abrasion tests: (a) A36 (b) 18Cr8Ni (c) 16Cr.

Removed B m
Material 1.2 m Profile BB
0.2
1.0
Height

0
0.8 -0.2 Removed
0.6 -0.4 Material
0.4 -0.6
83.5 m
1.3 m -0.8
0.2
0 10 20 30 40 50 60 70 80 m
0.0 Length [m]
B 35.2 m
-a- -b-
Fig. 7. Topographic analysis of the wear craters evidencing large portions of material removal, 18Cr8Ni: (a) 3D map and (b) 2D prole.

carbon steel (A36) did not exhibit passivation in this medium, as
normally reported in the literature [23,24].
3.2. Micro-abrasion tests
Due to the use of the 3D load cell, friction coefcients could be
obtained during the tests. Average friction coefcients were
computed for each material (Fig. 3).
No signicant difference was detected between the friction
coefcients presented by the different materials.
Fig. 4 exemplies the evolution of the microabrasion wear
coefcient k with time for two of the materials tested. It shows
that after a transient period of around 15 min, a steady-state re-
gime is achieved. This initial transient period during microabra-
sion tests has been widely reported in many other works
[21,25,26]. Similar behaviour was observed for all the materials
tested.
Average wear coefcients were calculated for each material. To
compute the average values, all points in the steady-state regime
(after 15 min of test) for each repetition (3) were used, which gives
18 values for each material. Fig. 5 summarizes the average wear
coefcients k obtained.
The ferritic steels showed higher abrasion resistance than the
low carbon steel A36. The best performance was presented by the
austenitic stainless steel. Attempts were carried out to correlate
wear coefcients with hardness and some microstructural char-
acteristics, such as grain size, but no clear pattern was identied.
Fig. 6 shows SEM images of some selected wear craters pro-
duced after the microabrasion tests. In all tests the prevailing
mechanism was the sliding of the abrasive particles (grooving)
[27]. It is noteworthy the detachment of large portions of material
for the carbon steel (A36, Fig. 6-a) and for the austenitic stainless
steel (18Cr8Ni, Fig. 6-b), indicated by arrows. Similar localized
removal did not occur for the ferritic stainless steels (Fig. 6-c),
 W.S. Labiapari et al. / Wear 376-377 (2017) 12981306 1303

2.0 2.0
18Cr8Ni: Aerated - Turbulent 16CrNb: Aerated - Turbulent
18Cr8Ni: Abrasion - Corrosion 16CrNb: Abrasion - Corrosion
1.5 1.5

Potential [V] vs. SCE

Potential [V] vs. SCE
1.0 1.0

0.5 0.5

0.0 0.0

-0.5 -0.5

-1.0 -1.0
1.E-06 1.E-04 1.E-02 1.E+00 1.E+02 1.E-06 1.E-04 1.E-02 1.E+00 1.E+02
Current Density [mA/cm2] Current density [mA/cm2]

-a- -b-
1.0
0.9 Abrasion
Abrasion-Corrosion
0.8
Friction Coefficient

0.7
0.6
0.5
0.4
0.3
0.2
0.1
0.0
0 30 60 90 120 150 180
Time [s]
-c-
Fig. 8. Typical results for the microabrasion tests. (a) potentiodynamic curve, 18Cr8Ni, (b) potentiodynamic curve, 16CrNb. (c) Evolution of friction coefcient with test time,
18Cr8Ni.

90 0.8
Abrasion 81.3 Abrasion
80 0.7 Abrasion-Corrosion
Abrasion-Corrosion
70 63.5 0.59 0.60
Friction Coefficient

0.6
k [m3/N.m].10-15

0.53 0.51 0.55

60 0.52
50.6 0.5
50 0.33
42.0 0.4
40 36.7 0.28 0.31 0.34
33.9 0.26 0.27
0.3
30

17.3 0.2
20
13.6 12.7 11.3 10.2 9.0
10 0.1

0 0.0
A36 11CrTi 11Cr 16CrNb 16Cr 18Cr8Ni A36 11CrTi 11Cr 16CrNb 16Cr 18Cr8Ni
Materials Materials
-a- -b-
Fig. 9. Evaluation of (a) wear rate coefcient k, and (b) friction coefcient in pure micro-abrasive and micro-abrasive-corrosive environment.

which showed lower wear coefcients than the carbon steel and whole potential range investigated (42 min) was longer than the
the austenitic stainless steel. A 3D topographic assessment of the transient period in the pure microabrasion tests (15 min), so that
worn surfaces by laser interferometry (Fig. 7) conrms that the the nal portion of the test should occur after a steady-state wear
regions indicated by arrows correspond indeed to areas with in- regime was achieved. First, it was observed for all material an
tense material removal. increase in the passivation current density when compared with
the pure corrosion tests. The curves obtained for pure corrosion
3.3. Micro-abrasion-corrosion test are again presented in this gure for comparison. In the literature
[2,7,28], this increased passivation current density Ip observed in
Fig. 8-a and Fig. 8-b shows some selected potentiodynamic abrasion corrosion tests is attributed to a competition between of
curves obtained during the microabrasion-corrosion tests. It the removal of the passive layer and repassivation. In our tests, the
should be again pointed out that the time necessary to scan the current density showed a slight increase in the passivation
1304 W.S. Labiapari et al. / Wear 376-377 (2017) 12981306
Fig. 10. FTIR analysis of 11Cr steel before and after immersion in 1 N H2SO4
solution.
region, i.e. a less effective repassivation occurred.
Fig. 8-c exemplies, for the stainless steel, a comparison be-
tween friction coefcients obtained during pure abrasion and
abrasion-corrosion tests. Under abrasion-corrosion conditions the
friction coefcients obtained were substantially lower. Similar
behaviour was observed for the other materials tested.
Average wear coefcients were calculated for the microabra-
sion-corrosion tests using the same methodology described for the
pure microabrasion tests. Fig. 9-a presents the values obtained,
which for comparison also presents the values obtained under
pure microabrasion conditions.
The rst strong fact evidenced by this gure is the substantially
lower wear coefcient (38  lower) observed under microabra-
sion-corrosion conditions when compared with pure microabra-
sion conditions. A similar tendency was also observed using NaCl
solutions by Bateni et al. [29] for AISI 304 stainless steels and AISI
1045 carbon steel and by Sinnett-Jones et al. [14] for AISI 304
stainless steels. This behaviour was associated to the change in the
repassivation kinetics due to the presence of the corrosive media
and to the external application of a potential difference.
The comparative behaviour between the different materials
was very different from that observed under pure abrasion con-
ditions, where the austenitic stainless steels had presented the
worst behaviour. Under abrasion-corrosion conditions, the in-
crease in Chromium content resulted in a steady reduction of the
wear coefcient. Higher Cr contents increase the stability of the
passive lm in most corrosive environments [22]. Under abrasion-
corrosion conditions, the passive lm stability seems to govern the
behaviour of stainless steels [6,14]. The increase in Cr content
could improve the lm stability by either hindering depassivation
Fig. 11. Friction coefcients measured during microabrasion-corrosion tests of 18Cr8Ni:(a) started with water slurry and replaced by 1 N H2SO4 slurry; (b) started with 1 N
H2SO4 slurry and replaced by water slurry.
or accelerating repassivation. The exact mechanism still needs to
be elucidated and should be a niche of further research.
Fig. 9-b compares the average friction coefcients of each
material measured during the pure abrasion and the micro-abra-
sion tests. No signicant difference was observed between the
different materials, which was conrmed by Analysis of Variance
(ANOVA). This gure conrms the lower friction coefcients under
abrasion-corrosion conditions when compared with pure
microabrasion.
When Wu et al. [30] measured the friction coefcient of AISI
304 stainless steel during sliding corrosion tests (fretting condi-
tions) using H2SO4 solution, they found friction coefcients be-
tween 0.25 and 0.3, which is the same range found here under
microabrasion-corrosion conditions.
According to Kolesnichenko et al. [31], the formation of FeSO4
in tribocorrosive systems containing sulfuric acid is the main
factor responsible for the reduction of the friction coefcient. Re-
ciprocating sliding-corrosion tests of iron also showed a reduction
of the friction coefcient when sliding in H2SO4 solutions when
compared with pure sliding tests using water [32]. Rowson and
Azouz [33] associated friction reduction to the formation of FeSO4
as an extreme pressure agent to the contact. We had tried, using
FTIR, to detect the presence of FeSO4 in the wear scars. It was
impossible to detect any variation by comparing the spectrum
obtained outside and within the wear mark. In fact, the instrument
indicated that the bands were very weak and therefore unreliable.
The thickness of FeSO4 lms formed on iron surfaces in the
presence of H2SO2 was measured by Auger microscopy in the lit-
erature as around 3 nm [34]. Therefore, it is difcult to identify this
lm, but it induces signicant smoothening of the surface. In
presence of a H2SO4 solution as electrolyte, iron dissolution leads
to the formation of FeSO4 on the metallic surface, as indicated by
the reaction shown in Eq. (1).
Fe(s) + H2 SO4 (aq) FeSO4 (aq) + H2 (g) (1)
To demonstrate the formation of this lm, a 11Cr steel sample
was immersed in a 1 N H2SO4 solution at room temperature for
1 h. Then the sample was subjected to FTIR analysis, Fig. 10.
The FTIR spectrum identied humidity bands in the region
between 3570 and 2940 cm
1 and 1650 cm
1, and more im-
portantly, it conrmed the formation of FeSO4 in the region be-
tween 1168 and 1068 cm
1.
In order to verify the kinetics response of the effects of corro-
sion on friction coefcient, two further test sequences were car-
ried out. In the rst, the abrasion-corrosion test started using a
slurry of abrasive and water, which was then changed to a slurry of

Fig. 12. Analysis by SEM of the surfaces of steel samples after testing 18Cr8Ni: (a) micro-abrasion-corrosion and (b) micro-abrasion test.
4000
Microhardeness HV0.05 [MPa]
Before test
3500
Abrasion
Abrasion - Corrosion
3000
2527
2500 2400
2145 2301
2143 2186 2186
2143
1971 1971
2000 1842 1842
1500
1000
500
0 sured under abrasion-corrosion than under pure abrasion, since
A36 11CrTi 11Cr 16CrNb
Materials
Fig. 13. Micro hardness Vickers HV0,05 at the samples of the material measured
within the wear scar before and after micro-abrasion and micro-abrasion-corrosion
test.
abrasive in 1 N H2SO4 solution. In the second, the abrasion-cor-
rosion test started using a slurry of abrasive in 1N H2SO4 solution,
which was then changed to a slurry of abrasive and water. The
results are shown in Fig. 11. Both curves show a short interval
between the two situations that corresponds to the time necessary
to remove the test load and change the slurry. When H2SO4 is
added to the slurry (Fig. 11-a), the lm seems to form very
promptly, reducing friction coefcient. In Fig. 11-b, abrasion re-
moves the lm, which is not restored in the absence of H2SO4,
causing friction to increase.
We believe that the FTIR technique does not present sufcient
sensitivity and is therefore not suitable for the detection of FeSO4
in the wear marks produced by abrasion-corrosion. More sensitive
techniques such as Auger Microscopy, XPS and TOF-SIMS are ne-
cessary and will be used in further investigations.
Fig. 12 compares for the austenitic stainless steel the wear
crater produced after abrasion-corrosion (Fig. 12-a) with that
produced after pure abrasion (Fig. 12-b). This gure shows that
abrasion-corrosion leads to the formation of a signicantly
smoother surface, which gives lower friction. On this way, it was
observed the smoother surface, Fig. 12-a compared to Fig. 12-b, as
a consequence of low friction coefcient.
The reduction of friction coefcient under microabrasion cor-
rosion conditions decreases the amount of mechanical energy that
is dissipated in the active interface in the form of friction. Vickers
microhardness measurements were carried out inside the wear
W.S. Labiapari et al. / Wear 376-377 (2017) 12981306 1305
scars for all the materials tested under pure microabrasion and
microabrasion-corrosion conditions. The results are shown in
3255
2972
2972 Fig. 13.
2810
First, all the samples showed an increase in hardness when
2363 2363
compared with microhardness of the samples before the tests.
Strain hardening is a common phenomenon during abrasion of
metals and has been vastly reported in the literature [35]. How-
ever, strain hardening was less intense under abrasion-corrosion
conditions than under pure abrasion conditions. Comparing pure
microabrasion, which is mechanically dominated, with abrasion-
corrosion, less energy is used in mechanical phenomena such as
strain hardening. This agrees with the lower friction values mea-
16Cr 18Cr8Ni
abrasion-corrosion only needs frictional energy for the mechanical
removal of passive lm, but not for the tribocorrosion of the active
areas.
4. Conclusions
Abrasion-corrosion of stainless steels was assessed using xed-
ball microabrasion tests with 1 N H2SO4 solution slurry and si-
multaneous imposition of potentiodynamic curves. The results
showed that:
1. The mechanical effects of turbulence and abrasion accelerate
the corrosion process, mainly evidenced by increase in the
passivation current density.
2. Although the austentic stainless steel stainless steel (18Cr8Ni),
with the highest Cr content within the materials tested, showed
the worst performance under pure abrasion conditions, it pre-
sented the best performance in the microabrasion-corrosion
tests. Under abrasion-corrosion conditions, the tribological
performance improved with the increase in Cr content.
3. The test rig allowed the measurement of friction coefcients
during microabrasion tests. This allowed to detect a reduction in
friction coefcient during abrasion-corrosion conditions when
compared with pure abrasion tests. Friction reduction was at-
tributed to the presence of FeSO4 as a corrosion product, which
was conrmed by FTIR analysis.
4. Microhardness Vickers measurements inside the wear scars
showed a less intense strain hardening under abrasion corr-
osion conditions than under pure abrasion conditions. This
is probably due to the reduction in the energy dissipated as fric-
tion in the contact and therefore available for plastic deformation.
1306 W.S. Labiapari et al. / Wear 376-377 (2017) 12981306
Acknowledgements
The authors would like to thank Capes/Proex, CNPq (Gant
477286/2011-9), CBMM S.A. (Grant FEMEC01-2011) and Fapemig
for nancial support.
References
[1] M.M. Stack, K. Chi, Mapping sliding wear of steels in aqueous conditions, Wear
255 (2003) 456465.
[2] M.M. Stack, H. Jawan, M.T. Mathew, On the construction of micro-abrasion
maps for a steel/polymer couple in corrosive environments, Tribol. Int. 38
(2005) 848856.
[3] W.S. Labiapari, C.M. Alcntara, H.L. Costa, J.D.B. de Mello, Stainless steel as an
antiwear material for the bio-fuel industry, Wear 302 (2013) 15361545.
[4] D. Landolt, S. Mischler, M. Stemp, S. Barril, Third body effects and material
uxes in tribocorrosion systems involving a sliding contact, Wear 256 (2004)
517524.
[5] V. Vignal, N. Mary, P. Ponthiaux, F. Wenger, Inuence of friction on the local
mechanical and electrochemical behaviour of duplex stainless steels, Wear
261 (2006) 947953.
[6] J.O. Bello, R.J.K. Wood, J.A. Wharton, Synergistic effects of micro-abrasion
corrosion of UNS S30403, S31603 and S32760 stainless steels, Wear 263
(2007) 149159.
[7] R.J.K. Wood, D. Sun, M.R. Thakare, A. de Frutos, Rozas, J.A. Wharton, Inter-
pretation of electrochemical measurements made during micro-scale abra-
sion-corrosion, Tribol. Int. 43 (2010) 12181227.
[8] D. Sun, J.A. Wharton, R.J.K. Wood, Micro-abrasion mechanisms of cast CoCrMo
in simulated body uids, Wear 267 (2009) 18451855.
[9] M.R. Thakare, J.A. Wharton, R.J.K. Wood, C. Menger, Exposure effects of alka-
line drilling uid on the microscale abrasioncorrosion of WC-based hard-
metals, Wear 263 (2007) 125136.
[10] M.R. Thakare, J.A. Wharton, R.J.K. Wood, C. Menger, Investigation of micro-
scale abrasioncorrosion of WC-based sintered hardmetal and sprayed coating
using in situ electrochemical current-noise measurements, Wear 267 (2009)
19671977.
[11] A.J. Gant, M.G. Gee, A.T. May, The evaluation of tribo-corrosion synergy for
WCCo hardmetals in low stress abrasion, Wear 256 (2004) 500516.
[12] W.S. Labiapari, Abrasion-Corrosion of Ferritic Stainless Steels (Doctoral Thesis),
Universidade Federal de Uberlandia 2015, p. 186, p, Uberlandia, MG, Brazil (in
Portuguese).
[13] ASTM, Standard Reference Test Method for Making Potentiostatic and Po-
tentiodynamic Anodic Polarization Measurements, G5-94, 2004.
[14] P.E. Sinnett-Jones, J.A. Wharton, R.J.K. Wood, Micro-abrasioncorrosion of a
CoCrMo alloy in simulated articial hip joint environments, Wear 259 (2005)
898909.
[15] I. Garc a, D. Drees, J.P. Celis, Corrosion-wear of passivating materials in sliding
contacts based on a concept of active wear track area, Wear 249 (2001)
452460.
[16] H.B. Lee, D.S. Wuu, C.Y. Lee, C.S. Lin, Wear and immersion corrosion of NiP
electrodeposit in NaCl solution, Tribol. Int. 43 (2010) 235244.
[17] J. Perret, E. Boehm-Courjault, M. Cantoni, S. Mischler, A. Beaudouin, W. Chitty,
J.P. Vernot, EBSD, SEM and FIB characterisation of subsurface deformation
during tribocorrosion of stainless steel in sulphuric acid, Wear 269 (2010)
383393.
[18] M.B. Santos, W.S. Labiapari, M.A.N. Ardila, W.M. da Silva Jr, J.D.B. de Mello,
Abrasioncorrosion: new insights from force measurements, Wear 332333
(2015) 12061214.
[19] D.N. Allsopp, R.I. Trezona, I.M. Hutchings, The effects of ball surface condition
in the micro-scale abrasive wear test, Tribol. Lett. 5 (1998) 259264.
[20] P.H. Shipway, The role of test conditions on the microabrasive wear behaviour
of soda-lime glass, Wear 233 (1999) 191199.
[21] H.L. Costa, M.A.N. Ardila, W.S. Labiapari, W.M. Silva, J.D.B. de Mello, Effect of
surface topography on the dynamics of the abrasive particles during micro-
abrasion, Wear 324325 (2015) 129139.
[22] A.J. Sedriks, Corrosion of Stainless Steels, Wiley-Interscience Publication, New
York, 1996.
[23] Y.F. Cheng, J. Bullerwell, F.R. Steward, Electrochemical investigation of the
corrosion behavior of chromium-modied carbon steels in water, Electrochim.
Acta 48 (2003) 15211530.
[24] J.A.L. Dobbelaar, E.C.M. Herman, J.H.W. de Wit, The corrosion behaviour of
iron-chromium alloys in 0.5 M sulphuric acid, Corros. Sci. 33 (1992) 765778.
[25] W.M. da Silva, R. Binder, J.D.B. de Mello, Abrasive wear of steam-treated sin-
tered iron, Wear 258 (2005) 166177.
[26] J. D. B de Mello, A. A. Polycarpou, Abrasive wear mechanisms of multi-com-
ponents ferrous alloys abraded by soft, ne abrasive particles, Wear, 269,
2010, 911-920.
[27] R.I. Trezona, D.N. Allsopp, I.M. Hutchings, Transitions between two-body and
three-body abrasive wear: inuence of test conditions in the microscale
abrasive wear test, Wear 225 (1999) 205214.
[28] F. Ferrer, H. Idrissi, H. Mazille, P. Fleischmann, P. Labeeuw, A study of abrasion
corrosion of AISI 304L austenitic stainless steel in saline solution using
acoustic emission technique, NDT E Int. 33 (2000) 363371.
[29] M.R. Bateni, J.A. Szpunar, X. Wang, D.Y. Li, The effect of wear and corrosion on
internal crystalline texture of carbon steel and stainless steel, Wear 259
(2005) 400404.
[30] P.-Q. Wu, J.P. Celis, Electrochemical noise measurements on stainless steel
during corrosionwear in sliding contacts, Wear 256 (2004) 480490.
[31] L.F. Kolesnichenko, P.V. Trushko, Formation of boundary layers during friction
in the presence of sulfur, Sov. Powder Metall. Met. Ceram. 9 (1970) 993997.
[32] K. Miyoshi, Studies of mechanochemical interactions in the tribological be-
havior of materials, Surf. Coat. Technol. 43 (1990) 799812.
[33] D.M.Rowson, A.Azouz, An investigation of extreme pressure lms formed
during wear test, Vacuum 31 (1981) 593596.
[34] S. Mischler, E. Rosset, G.W. Stachowiak, D. Landolt, Effect of sulphuric acid
concentration on the rate of tribocorrosion of iron, Wear 167 (1993) 101108.
[35] K.L. Zum Gahr, Microstructure and Wear of Materials, Tribology Series 10,
Elsevier, Amsterdam, 560p, 1987.