ISIJ International, Vol. 57 (2017), No. 2

© 2017 ISIJ349

ISIJ International, Vol. 57 (2017), No. 2, pp. 349–357

* Corresponding author: E-mail: debabrata.pradhan@tatasteel.comDOI: http://dx.doi.org/10.2355/isijinternational.ISIJINT-2016-343

Optimizing Surface Roughness to Improve Steel-rubber Adhesion Strength through Electro-cleaning Simulation

Debabrata PRADHAN,* Rita GHOSH, Atanu BANERJEE and Monojit DUTTA

Research and Development Department, Tata Steel Ltd, Jamshedpur-831001, India.

(Received on June 10, 2016; accepted on September 28, 2016)

In this study, we have simulated the electro-cleaning process considering the actual surface topography of steel sheet to determine the final topography and roughness of the steel surface (Rz, distance between the surface peak and the valley) after different duration of electro-cleaning. COMSOL Multi-Physics® simu-lation module with moving mesh technique was used for the electro-cleaning simulation. The effect of applied current density (A/m2), electrolyte concentration (H2SO4, g/l) and temperature (°C) on the change in roughness (ΔRz, μm) of steel surface was evaluated by the simulation. Simulation results show that the current density has the highest effect towards the change in roughness of steel surface. The change in roughness increases with increase in current density. The current density of 500–700 A/m2 is required for appreciable electro-cleaning roughness in this study. An electrolyte concentration of 40–60 g/l would pro-vide the maximum change in roughness for any fixed applied current density. The change in roughness also increases with increasing the electrolyte temperature. Lab scale experimentation showed good agree-ment with the simulated results. Finally, it was determined that 600 A/m2 of applied current density and 4.75 μm of steel surface final roughness (Rz,f) would provide the maximum adhesion strength of 110 N between the steel and rubber.

KEY WORDS: electro-cleaning; current density; roughness; peel force; adhesion strength.

1. Introduction

The motor tire bead (MTB) wire is a very important ele-ment for pneumatic tires which provides tire’s mechanical strength and structural integrity due to its adhesion during vulcanization with the rubber. Thus, steel-rubber interface and the proper adhesion between them play the most crucial role creating the strength and the durability of the tire.

During vulcanization/curing process, sulfur which is added as a cross linking agent in rubber reacts with the steel surface and forms the interface.1) But the adhesion force comes due to the combined effect of chemical bonding and mechanical interlocking at the interface.2,3)

Van Ooij et al. reported that uncoated steel surface does not adhere with the rubber due to the lack of chemical bonding between them.4) Therefore, different surface treat-ments such as metallic alloy coating, thin polymer coating, bonding agents, selective surface oxidation and micro-roughening on the bare steel surface have been investigated to improve the adhesion strength between the steel and rubber.5–13) The recent study by Banerjee et al. have con-cluded the improvement of adhesion strength by applying Cu–Sn alloy coating on bare steel surface.14–19) Most of the literatures suggest copper as a preferred coating to improve the interfacial bonding with rubber during vulcanization because it reacts with sulfur present in the rubber to form

copper sulfide at the interface.4,20–22)

For improving the adhesion strength between the steel and rubber, steel surface initial roughness plays an impor-tant role in obtaining the uniform metallic coating coverage and the quality of the interface. The uniform coating cover-age on rough steel surface with a thin layer of Cu or Cu–Sn alloy coating through cementation or displacement method could be challenging.23) Our previous study suggest that displacement coating technique in a wire coating line pro-duces inhomogeneous coating coverage due to the smaller residence time (~2 s) in coating bath.14) The coating covers mostly the roughness peaks of the surface profile those are anodic by nature leaving the deep uncoated valleys due to the lack of solution penetration in the recesses.14) Similarly, Cho et al. have also reported the uncoated regions during 90 nm thick Cu coating.20,21) Unfortunately, their solution with the application of zinc coating prior to the Cu coating did not succeed much due to the dissolution of highly anodic zinc coating in the subsequent acidic Cu coating baths.20,21)

Therefore, electro-cleaning, more precisely electro-pol-ishing process could be an important step to produce smooth surface by reducing the surface roughness. It is generally applied instead of abrasive polishing entirely or as a final finishing process after polishing. In this electro-chemical surface finishing process, the metal is removed or selec-tively dissolved from the workpiece at the high points of the microscopic surface roughness (peaks) compared to the valleys by means of an external electric field. This anodic dissolution process preferentially reduces the peaks of sur-

ISIJ International, Vol. 57 (2017), No. 2

© 2017 ISIJ 350

face protuberances due to higher current densities associated with the peaks, resulting faster dissolution and smoothening surface.24) Thus, by controlling the roughness of steel sur-face a uniform coating coverage could be obtained for sub-sequent metallic coating for rubber adhesion. In our recent study, we have investigated the effect of electro-cleaning current density on the surface roughness of steel wire and correlated with the steel wire-rubber adhesion strength.25) But the effect of all electro-cleaning parameters such as current density, electrolyte concentration, temperature and their combination on the surface roughness have never been explored. Unfortunately, very limited information exists on the optimization of electro-cleaning parameters for improve-ment of adhesion strength between the steel surface and rubber.

In this context, we have modeled and simulated the effect of electro-cleaning parameters on the steel surface roughness using COMSOL Multi-Physics® electrochemi-cal tool. The simulated parameters were validated through electro-cleaning experiments. Finally, we have correlated and optimized the electro-cleaning surface roughness and current density with the adhesion strength between the steel and rubber.

2. Eletro-cleaning Simulation of Steel Surface

2.1. Working EquationsFigure 1 shows the model geometry for electro-cleaning

or the anodic dissolution process of steel surface. For bet-ter illustration, a red rectangular selected area in Fig. 1 is presented in the simulation results in the following sections.

The model geometry contains a single electrolyte (H2SO4) domain where the steel surface and an inert flat surface were used as anode (dissolution) and cathode, respectively. The other two boundaries represent as insulating surfaces where charge-neutrality was applied. The electro-cleaning reac-tions at the electrode surface are shown according to the following Eqs. (1), (2).

The anodic reaction or dissolution of steel surface could be expressed according to Eq. (1).

Fe Fe e VEFe Fe

→ + = +++++2 0 440

/. vs. SHE ..... (1)

The reduction or the hydrogen evolution reaction took place at the cathode surface.

H e H VEH H

+ + → =+½/

vs. SHE2 20 0 .......... (2)

The governing equations and assumptions were made for this simulation work:

1. Secondary current distribution was assumed where transport of ions took place through the migration (electrode kinetic) and solution resistance. The electrolyte (H2SO4) was mixed well to eliminate concentration gradients or diffusion mode of ion transfer.

2. The local electrolyte potential El (V) over the electro-lyte domain was solved according to Ohm’s law:

j El l= ∇σ . ................................. (3)

∇ =. jl 0 ................................... (4)

Where, jl is the local electrolyte current density vector (A/m2) and σ is the electrolyte conductivity for a particular concen-tration of H2SO4 which is assumed to be constant.

3. The following condition is valid for all insulated boundaries except electrode surface:

N jl. = 0 ................................... (5)

Where, N is the normal vector, pointing out of the domain.4. Butler-Volmer equation was used to model the dis-

solution reaction according to in Eq. (1) and to determine the local current density distribution at electrode surface and electrolyte.

j jnF

RTj

nF

RTloc Fe

a c, exp exp=

− −

0 0α η α η ....... (6)

Where, jloc,Fe is the local current density (A/m2), j0 is the anodic exchange current density (A/m2), η is the activation overpotential (η = − − ++E E EFe l Fe Fe/

0 ), EFe is the local anode/electrode potential (V), El is the local electrolyte potential (V), E

Fe Fe/ ++0

is the equilibrium potential, T is the electro-cleaning temperature (K), αa is the anodic charge transfer coefficient, αc is the anodic charge transfer coefficient, n is the number of electrons for anodic reaction, F is the Faraday constant (96 500 coulomb/equiv.) and R is the universal gas constant (8.314 J/mole K).

For higher values of η, Eq. (6) was simplified to Eq. (7) and rearranged further to Tafel equation (Eq. (8))

j jnF

RTloc Fe

a, exp=

0α η ....................... (7)

η βa aloc Fej

j= log ,

0

............................ (8)

Where, βa is the Tafel’s slope and expressed as:

βα

aa

RT

nF=

2 303.

5. On the anode, the electrolyte current density was set to be the local current density of the iron dissolution reaction:

N j j jl loc Fe app. ,= = ........................... (9)

Where, japp is applied current density vector (A/m2).6. The dissolution at the anode surface with a velocity in

the normal direction (x) was evaluated according to:Fig. 1. The schematic model geometry for eletro-cleaning simula-

tion of steel surface. (Online version in color.)

ISIJ International, Vol. 57 (2017), No. 2

© 2017 ISIJ351

Nx

t

r Mdl w..∂

∂=

ρ............................ (10)

Where, rdl is the dissolution rate of the steel electrode, Mw is molar mass of iron (0.0558 kg/mol) and ρ is the density of iron (7 874 kg/m3).

Rdl was evaluated using the following equation:

ru j

nFdl

app=. ............................... (11)

Where, u is the stoichiometric coefficient.7. The hydrogen evolution reaction on the cathode will

contribute to the local current density at the cathode surface but for anode surface it was neglected.

2.2. Electro-cleaning Parameters and Roughness Mea-surement

The steel electrode kinetic parameters used for the simu-lation are listed in Table 1. Most of the parameters were collected from literature.26) Table 2 lists the variable param-eters to evaluate their effect on steel surface roughness.

Figure 2 shows the pictorial view for roughness (Rz, μm) which was calculated from the peak and valley distance measurement:

Roughness R peak distance valley distancez( ) ( )= − / 2

........................................ (12)

The change in roughness (ΔRz, μm) was measured from the difference in initial and final roughness after electro-cleaning (see Fig. 2).

Change in roughness R

Initial roughness R Final

z

z i

∆( )( )

=

−, rroughness Rz f ,( ) ... (13)

2.3. SteelSurfaceRoughnessProfileforSimulationThe initial surface roughness profile of steel plays an

important role in creating the 2-dimentional model geometry and determining change in roughnesses after simulation. As presented in Fig. 1, the roughness profile was taken from the actual steel surface using Olympus Laser Confocal Microscope, OLS 4100® before electro-cleaning experiment described elsewhere.24) The steel surface was polished with 120 grit SiC sandpaper to create average initial roughness (Rz,i) of 7.7 μm. The initial roughness of the selected portion of the model geometry in Fig. 1 varies closely to 7.7 μm as the roughness profile was exported from laser confocal microscope to the simulation software.

3. Experimental Procedure

After obtaining the optimized electro-cleaning parameters for each applied current density from simulation results, the lab-scale experiments were conducted to validate the simulation results as well as to carry out further steel-rubber adhesion study.

3.1. Sample PreparationCold rolled closed annealed steel sheet samples were cut

into the required size (12 cm×2.4 cm) and cleaned with acetone to remove the oil. The samples were then polished with 120 grit SiC sandpaper to achieve the required initial roughness (Rz,i) of 7.7 μm. The samples were finally cleaned with acetone and dried before electro-cleaning experimenta-tion.

3.2. Electro-cleaning and Cu–Sn Alloy CoatingThe electro-cleaning and subsequent Cu–Sn alloy coat-

ing on steel surface were conducted before the vulcaniza-tion test with the rubber. The electro-cleaning solution was prepared by adding H2SO4 (98%) into the industrial de-mineralized water. The solution was properly mixed with a magnetic stirrer before the experiment. The electro-cleaning experiment was conducted in a 2-electrode Pyrex® glass cell (500 ml) where steel sample and stainless steel sheet were used as anode and cathode, respectively. The DC power source (AUTONIX®, India) supplied the constant current through the cell. A predetermined area (5 cm × 2.5 cm) of

Table 1. Steel electrode kinetic parameters used for electro-cleaning simulation.26)

Name Value Description

EFe Fe/ ++0 −0.44 V vs. SHE Iron oxidation equilibrium potential

j0 0.00423 A/m2 Iron oxidation exchange current density

βa 0.41 V vs. SHE Tafel slope for Iron oxidation

n 2 Number of participating electron for iron oxidation

αa 0.5 Anodic charge transfer coefficient

Table 2. Variable parameters used for electro-cleaning simula-tion.

Name Value Description

japp 200–700 A/m2 Applied current density

t 1–500 s Electro-cleaning time

Celectrolyte* 5–200 g/l H2SO4 concentration

T 40–100°C Electrolyte temperature

*Conductivity (σ, s/m) of the solution collected from literature and used for simulation.27)

Fig. 2. Schematic view for steel surface roughness calculation before and after simulation from the current density distri-bution map. (Online version in color.)

ISIJ International, Vol. 57 (2017), No. 2

© 2017 ISIJ 352

steel anode was dipped into the electrolyte to achieve the required amount of current density. The test was carried out using optimized concentration of H2SO4 (50 g/l), electrolyte temperature (50°C) and electro-cleaning duration of 500 s. The only variable for the experiments is the applied current density which varied from 300 A/m2 to 700 A/m2 to evaluate its effect on steel surface roughness. The applied electro-cleaning current density and the average voltage reading are listed in Table 3. At least five samples for each applied current density were prepared to ensure the repeatability of the electro-cleaning process.

For adhesion test with rubber, Cu–Sn alloy coating was applied on the electro-cleaned surface through substitution or cementation technique. Cu–Sn alloy coating was car-ried out by dipping the electro-cleaned steel samples in electrolytic bath containing 2.15% of conc. H2SO4, 0.25 g/l of SnSO4 and 35 g/l of CuSO4.5H2O. The coating experi-ment was conducted with a dipping time of 10 s at room temperature. Al least five coated samples were prepared for adhesion test with rubber to ensure the repeatability.

3.4. Surface Roughness (Rz) MeasurementThe surface roughness of steel before and after the electro-

cleaning experiments were measured using laser confocal microscope (LEXT OLS4000®; Olympus, Japan) attached with 3-D imaging and quantification software (OLS4000, version 2.2.4, Olympus Corporation, Japan). The roughness (Rz) was calculated from the height difference between the highest peak and lowest valley of the roughness profile. To measure the most accurate Rz, several line profile rough-nesses on the steel surface were measured and the average value was reported as mentioned elsewhere.18)

3.5. Steel-rubber Adhesion TestRubber peel-off test of the vulcanized sample was per-

formed to measure the adhesion strength between the steel and rubber. For comparison, adhesion strength for few base samples (not electro-cleaned) coated with Cu–Sn alloy were also measured. Laboratory made Styrene Butadiene rubber (SBR) formulation was used for the vulcanization experi-ments.16) The steel sheet sample with an area of 5 × 2.5 cm2 was sandwiched with a 2 mm thick rubber pad prepared from bead rubber formulation in the curing press. On the uncoated parts of the sample, silicone oil was sprayed to avoid the adhesion between rubber and the steel substrate. The samples were vulcanized in an electrically heated press at 150°C for 28 minutes at a pressure of 13 MPa. The peel-off test was carried out in a tensile tester (Universal Test-ing Machine-1445; Zwick GmbH & Co. Ulm, Germany) at

room temperature. The force vs. displacement information was continuously recorded during the peeling at 180° with constant crosshead speed of 100 mm/min. For repeatability, at least five samples were tested for each set of experiments and the average peel-off force of three close readings was reported.

4. Results and Discussion

4.1. Electro-cleaning Simulation4.1.1. Effect of Applied Current Density

Figure 3 shows the variation of change in roughness (ΔRz) with the applied current density for different electro-cleaning times (3–30 s) using electrolyte concentration of 50 g/l and temperature of 50°C. It shows that the change in roughness increases with increasing the applied current density which is consistent with our earlier investigation.25) The curves also show almost linear relationship between the applied current density and the change in roughness. But, the change in roughness does not vary linearly with electro-cleaning time as shown in Fig. 4. Initial steep change in roughness was observed for all applied current density up to the duration of 15 s. Afterwards, the rate in roughness change reduces. The probable explanation of this observation could be the initial rapid dissolution of the surface peaks which have higher current density compared to the valleys.

After electro-cleaning simulation for the duration of 500 s, it was calculated that the change in roughness using 700 A/m2 applied current density is four times higher compared 200 A/m2 applied current density (5.10 μm vs. 1.25 μm). This behavior could be explained from the elec-trolyte current distribution for two different applied current densities of 200 A/m2 and 700 A/m2 shown in Fig. 5. The peak area in the roughness profile shows higher current den-sity compared to the valley area which signifies the higher dissolution rate of peak compared to valley of the steel surface profile. For instance, the peak area reached the maxi-

Table 3. Variation of applied current density and average voltage reading for electro-cleaning experimentation.

Current density (A/m2) Av. applied current (A) Av. voltage reading (V)

300 0.675 0.75

400 0.9 1.03

500 1.125 1.22

600 1.35 1.52

700 1.575 1.60

Fig. 3. Variation of change in roughness with applied current den-sity for different electro-cleaning times for electrolyte concentration of 50 g/l and temperature at 50°C. (Online version in color.)

ISIJ International, Vol. 57 (2017), No. 2

© 2017 ISIJ353

of electro-cleaning simulation. In comparison between the two different applied current densities of 200 A/m2 and 700 A/m2, the higher change in roughness for 700 A/m2 could be justified by the higher current density at the peak area (>750 A/m2) compared to 250 A/m2 peak current density for 200 A/m2 applied current density. Therefore, it could be concluded that the electro-cleaning current density plays a significant role in roughness change.

4.1.2. Effect of Electrolyte (H2SO4) ConcentrationThe effect of electrolyte concentration (H2SO4, g/l) on

the change in roughness for different duration of electro-cleaning times (3–30 s) using constant applied current density of 600 A/m2 and temperature of 50°C is shown in Fig. 6. It clearly shows that the roughness change increases with increase in H2SO4 concentration and reaches a peak at 50 g/l. Afterwards, the change in roughness reduces with increase in H2SO4 concentration. For example, after 500 s of electro-cleaning simulation, it was determined that the change in roughness is 4.0 μm for 50 g/l electrolyte compared 3.25 μm for 100 g/l electrolyte. This behavior could be explained by the higher electrical conductivity of 50 g/l H2SO4 electrolyte (86 s/m at 50°C) compared to the lower electrical conductivity when concentration increased to 100 g/l (1.38 s/m at 50°C).27) Highly conductive elec-trolyte provided higher current density at the peak region of surface topography as well as the dissolution rate of steel and hence, increased the change in roughness. On the other hand, the change in roughness varied almost linearly with electro-cleaning time for all electrolyte concentrations shown in Fig. 7.

4.1.3. Effect of Electro-cleaning TemperatureFigure 8 depicts the variation of change in roughness

with the electro-cleaning temperature for different duration of electro-cleaning (3–30 s) using constant applied cur-rent density of 600 A/m2 and electrolyte concentration of 50 g/l. It shows that the change in roughness increases with increase in electro-cleaning temperature. As the electro-

Fig. 5. Electrolyte current density distribution for different applied current densities (200 A/m2 and 700 A/m2) after 500 s of simulation for electrolyte concentration of 50 g/l and temperature at 50°C. (Online version in color.)

Fig. 4. Variation of change in roughness with electro-cleaning time for different applied current densities for electrolyte concentration of 50 g/l and temperature at 50°C. (Online version in color.)

mum current density ~250 A/m2 compared to ~150 A/m2 after 500 s of simulation for an applied current density of 200 A/m2. Thus, the final roughness decreases after 500 s

Fig. 6. Variation of change in roughness with electrolyte concen-tration for different electro-cleaning times for applied cur-rent density of 600 A/m2. (Online version in color.)

ISIJ International, Vol. 57 (2017), No. 2

© 2017 ISIJ 354

Fig. 9. Combined effect of electrolyte concentration and current density on change in roughness for different electro-cleaning times at electro-cleaning temperature of 50°C. (Online version in color.)

Fig. 8. Variation of change in roughness with electro-cleaning temperature for different electro-cleaning times with fixed applied current density of 600 A/m2. (Online version in color.)

Fig. 7. Variation of change in roughness with electro-cleaning time for different electrolyte concentrations for applied current density of 600 A/m2. (Online version in color.)

cleaning temperature increased the conductivity of the electrolyte increased which improved the overall current density at the steel-electrolyte interfaces. Hence, the rate

of steel dissolution at the roughness peak increased which in turn increased the change in roughness. But, the change in roughness does not vary much for the higher duration of

ISIJ International, Vol. 57 (2017), No. 2

© 2017 ISIJ355

electro-cleaning. For example, the change in roughness is 4.5 μm at 100°C compared to 3.9 μm at 40°C after 500 s of electro-cleaning simulation.

4.1.4. Combined Effect of Applied Current Density and Electrolyte Concentration

The combined effect of current density and electrolyte concentration on the change in roughness were evaluated by varying the electrolyte concentration (5 g/l−200 g/l) and applied current density (200–700 A/m2). The electro-cleaning temperature was unchanged at 50°C. Figure 9 depicts the effect of electrolyte concentration on the change in roughness for different applied current density and electro-cleaning times (3–500 s). It shows that the change in roughness increases with increasing H2SO4 concentration and reaches a peak at 50 g/l similar to our earlier observa-tion. Afterwards, the change in roughness reduces with increasing the H2SO4 concentration. From the combined effect of applied current density and electrolyte concentra-tion, it could be concluded that optimized parameters for each applied current density varied from 300 A/m2 to 700 A/m2 where the maximum change in roughness could be obtained are: electrolyte concentration of 40–50 g/l and electro-cleaning temperature of 40–60°C.

4.2. Experimental ValidationFor validation, lab-scale electro-cleaning experiments

were carried out according to the maximum change in roughness for each applied current density obtained from COMSOL simulation shown in Fig. 9. The experiments were conducted by varying only the applied current density from 300 A/m2 to 700 A/m2 by fixing other parameters such as concentration of H2SO4 (50 g/l), electrolyte temperature (50°C) and electro-cleaning time (500 s). The change in roughness was listed in Table 4 for different applied current density. It clearly shows that with the increase in applied current density, the change in roughness increases i.e. more reduction of final roughness happened. The surface analy-sis of electro-cleaned sample at 700 A/m2 applied current density using laser confocal microscope shows smoothening or more reduction of surface roughness compared to the electro-cleaned sample produced at lower applied current density of 300 A/m2 (see Fig. 10). Furthermore, the line roughness profile of the steel surface electro-cleaned at 700 A/m2 shows less serration and lower roughness compared

to 300 A/m2 which also confirms the higher reduction in surface roughness due to the faster dissolution of the peaks.

For validation, the simulated and experimentally mea-sured changes in roughnesses were presented in Fig. 11. The change in roughness depicts very close variation within 1 μm and shows similar trend in both simulated and experimental curves. The little variation could be explained from the smaller selected length of line roughness pro-file (200 μm) and smaller geometrical area of electrolyte (200 μm × 300 μm) used for simulation which did not interpret the entire electrode area of electro-cleaning experi-mentation.

Fig. 10. Surface topograph and line roughness profile of electro-cleaned steel surface analyzed by laser confocal micro-scope for different applied current densities (300 A/m2 and 700 A/m2). (Online version in color.)

Table 4. Steel surface roughness after different electro-cleaning experiments conducted at different applied current density using fixed concentration of H2SO4 (50 g/l), electrolyte temperature (50°C) and electro-cleaning time of 500 s.

Applied current density (A/m2)

Av. initial roughness (Rz,i), μm

Av. final roughness (Rz,f), μm

Change in roughness

(ΔRz)

300 7.7 6.13 1.57

400 7.7 6.07 1.63

500 7.7 5.91 1.79

600 7.7 4.78 2.92

700 7.7 3.96 3.74

Fig. 11. Comparison between simulated and lab-scale experimen-tation result for the electro-cleaning duration of 500 s. (Online version in color.)

ISIJ International, Vol. 57 (2017), No. 2

© 2017 ISIJ 356

4.3. Effect of Parameters on Steel-rubber Adhesion Strength

The adhesion strength between steel-rubber was mea-sured through peel-off test by varying the electro-cleaning current density and the average final roughness. Figure 12 shows the variation of the average peel force (N) at the failure for different electro-cleaned samples as a function of applied current density. For comparison, the peel force for base sample (not electro-cleaned) was included. It shows that all the electro-cleaned samples have higher peel-off forces compared to the base sample (94 N) which has higher average surface roughness (Rz,i = 7.7 μm). The peel force curve for electro-cleaned samples shows interesting behav-ior; it increases with increase in electro-cleaning current density up to a certain limit of 600 A/m2 and after which it reduces. The maximum peel force of 110 N was attained at the electro-cleaning current density of 600 A/m2. The nature of such a behavior could be explained from the correlation between the peel force and the average final roughness (Rz,f) of electro-cleaned steel surface (see Table 4) shown

Fig. 13. Effect of average final surface roughness (Rz,f) on peel-off force for of electro-cleaned samples.

in Fig. 13. It shows that the peel force first increases with increase in average final roughness and reaches a peak at ~4.75 μm. Afterwards it decreases. It certainly concludes that maximum peel force could be obtained at the roughness of 4.75 μm which corresponds to the electro-cleaning cur-rent density of 600 A/m2. Further increase in applied current density beyond 600 A/m2 will reduce the final roughness below 4.75 μm which does not improve the peel-force. The variation of this peel force with electro-cleaning final roughness could be explained from the basic mechanism of the steel and rubber adhesion strength. The adhesion strength between the steel and rubber has two main compo-nents; the mechanical force at the inter-locking interfaces of steel-rubber and the chemical or electrochemical bond-ing between the rubber and the Cu–Sn coating layer during vulcanization.2,3) Higher roughness of steel surface enhances the mechanical inter-locking while on the other hand it creates non-uniform coating coverage of Cu–Sn layer and hence reduces the overall chemical bond strength during vulcanization at the uncoated region which is consistent with the literatures.14,15,18) Likewise, lower surface rough-ness enhances more chemical adhesion strength by creating more uniform coating coverage of Cu–Sn alloy layer while it reduces the mechanical inter-locking strength. Therefore, a balance between the mechanical inter-locking force and chemical bonding strength was obtained at the final rough-ness of ~4.75 μm which corresponds to the electro-cleaning current density of 600 A/m2 that provided the maximum adhesion strength of 110 N. In contrast, the higher surface roughness (Rz,i = 7.7 μm) of base sample showed poor adhesion strength due to non-uniform coating coverage of Cu–Sn layer and lower chemical bonding strength.

5. Conclusions

A numerical simulation model was developed using COMSOL Multi-Physics® tool to simulate the effect of electro-cleaning parameters such as applied current density, electrolyte concentration and temperature on the change in roughness (ΔRz, μm) of steel surface. It was determined from the simulation that the applied current density has the maximum effect in changing the final roughness of steel surface. The change in roughness increases with increase in current density due to the higher dissolution rate of the roughness peaks compared to the valleys. It was evalu-ated that the current density of 500–700 A/m2 is required to create appropriate electro-cleaning roughness. An opti-mum electrolyte concentration of 40–60 g/l is required to provide the maximum change in roughness. The change in roughness also increases with increase in electro-cleaning temperature. Lab scale experimentation using electrolyte concentration of 50 g/l, temperature of 50°C with varying electro-cleaning current densities (300–700 A/m2) showed good agreement with the simulated change in roughness of steel surface. The topographical investigation of electro-cleaned surface through the laser confocal microscope also confirmed the change in roughness with increasing electro-cleaning current density. Peel-off tests revealed that the steel-rubber adhesion strength increases with increase in electro-cleaning current density to a certain limit, beyond which it reduces. The electro-cleaning current density of

Fig. 12. Variation of peel-off force with applied current density for electro-cleaned (electrolyte concentration of 50 g/l and temperature at 50°C) and base samples (no electro-cleaning). (Online version in color.)

ISIJ International, Vol. 57 (2017), No. 2

© 2017 ISIJ357

600 A/m2 was determined to achieve the maximum steel-rubber adhesion strength of 110 N. Finally, it was concluded that the final electro-cleaning roughness (Rz,f) of 4.75 μm is required to create the maximum adhesion strength between steel and rubber of 110 N.

AcknowledgmentsThe authors are grateful to Puspanjali Mohanta and Nitu

Rani for assisting in lab-scale electro-cleaning experimen-tation. Authors would like to thank Puja Kumari for OLS roughness measurement. Authors also like to acknowledge the support from Dr. Tapas Laha and Dr. Anil K. Bhowmick, Rubber Technology Centre, IIT Kharagpur for adhesion tests.

REFERENCES

1) W. J. van Ooij: Rubber Chem. Technol., 57 (1984), 421.2) J. W. Cook, S. Edge and D. E. Packham: J. Adhes., 72 (2000), 293.3) D. A. Stout and J. Rife: J. Vac. Sci. Technol. A, 8 (1990), 2407.4) W. J. van Ooij: The Handbook of Rubber Bonding, Chapter 6:

Rubber-Brass Bonding, Rapra Tech., UK, (2003), 163.5) O. E. Adler: Nat Standard Co, Patent No. GB840693, July, (1960).6) S. Nakagawa, K. Kawasaki and M. Yoshikawa: Bridgestone Corp.,

Production of Metallic Wire Rod For Reinforcing Rubber Product, Patent No. JP1177390A, July, (1989).

7) S. Hachisuka: Tokyo Rope Mfg Co., Method for Producing a Rubber-reinforced Steel Wire, Patent No. US5389163, Feb., (1995).

8) T. W. Starinshak: Goodyear Tyre & Rubber Co., Process for Treating a Brass-plated Steel Wire, Patent No. US5118367, Jan., (1992).

9) L. F. Rakestraw, R. W. Bullard, J. E. Niesse and M. F. Bender: Monsanto Co., Method for Improving Hydrolytic Stability of the Bond between Steel Surfaces and Organic Adhesives, Patent No.

US3996074, Dec., (1976).10) R. M. Shemenski, D. K. Kim and T. W. Starinshak: Goodyear Tyre

& Rubber Co., Copper-zinc-iron Ternary Alloy Coated Steel Wire Reinforcers in Tires, Patent No. US4446198, May, (1984).

11) G. Haemers: Bekaert Sa Nv, Steel Wire Reinforcing Elements, Patent No. US4347290, Aug., (1982).

12) R. A. Friend: Nat Standard Co., Rubber Adherent Ternary Cu-Zn-Ni Alloy Coated Steel Wires, Patent No. US4226918 (A), Oct., (1980).

13) W. Coppens, D. Chambaere and W. Vanhee: Bekaert Sa Nv, A Steel Wire for the Reinforcement of Elastomers, Patent No. EP351909, Jan., (1990).

14) A. Banerjee, M. Dutta, S. Bysakh, A. K. Bhowmick and T. Laha: J. Adhes. Sci. Technol., 28 (2014), 987.

15) A. Banerjee, M. Dutta, A. K. Bhowmick and T. Laha: ISIJ Int., 54 (2014), 671.

16) A. Banerjee, M. Dutta, A. K. Bhowmick and T. Laha: J. Adhes. Sci. Technol., 28 (2014), 1610.

17) A. Banerjee, M. Dutta, A. K. Bhowmick and T. Laha: Mater. Sci. Forum, 783 (2014), 1536.

18) A. Banerjee, M. Dutta, S. Bysakh, A. K. Bhowmick and T. Laha: Appl. Surf. Sci., 313 (2014), 804.

19) A. Banerjee, R. Kumar, M. Dutta, S. Bysakh, A. K. Bhowmick and T. Laha: Surf. Coat. Technol., 262 (2015), 200.

20) P. L. Cho, G. Seo, G. S. Jeon and S. K. Ryu: J. Adhes., 73 (2000), 43.

21) P. L. Cho, G. S. Jeon, S. K. Ryu and G. Seo: J. Adhes., 70 (1999), 241.

22) Y. Y. Su and R. M. Shemenski: Appl. Surf. Sci., 161 (2000), 355.23) D. A. Stout: Appl. Surf. Sci., 15 (1983), 166.24) W. Tegart: The Electrolytic and Chemical Polishing of Metals: in

Research and Industry, Pergamon Press, Oxford, (1959).25) D. Pradhan, R. Ghosh, A. Banerjee and M. Dutta: Surf. Coat. Technol.,

286 (2016), 293.26) D. A. Jones: Principles and Prevention of Corrosion, 2nd Ed., Pren-

tice Hall, NJ, (1995).27) IC Controls: Conductivity Theory and Measurement, Technical notes,

Issue 4-1, (2005), http://www.iccontrols.com/files/4-1.pdf, (accessed 02-06-2016).