Influence of Plasma Electrolytic Polishing Conditions on Surface Roughness of Steel

Anodic polishing of St45 medium carbon steel (0.45 wt % C) and 12Cr18Ni10Ti austenitic stainless steel in aqueous electrolytes based on ammonium chloride or ammonium sulfate is studied under conditions of natural or force convection. Surface roughness, weight loss of the samples and the current resulted from polishing are measured. A possibility of plasma electrolytic polishing using a setup for the anodic thermal chemical treatment of compounds with a longitudinal electrolyte flow is shown. Polishing regimes are found to decrease the initial surface roughness of the steels from Ra = 1.0 to 0.19–0.22 μm in 5 minutes at a sample weight loss 0.5–0.7 mg/s. The minimal surface roughness is obtained using a 3% solution of ammonium chloride at a flow rate of 0.8 L/min, voltage of 300 V, and electrolyte temperature of 80°C.


INTRODUCTION
Plasma electrolytic polishing (PEP) in aqueous electrolytes is one of the leading processes of plasma electrolytic treatment used in modern industry. Among its advantages are high processing speeds, low roughness of the polished surface, and the absence of aggressive and toxic reagents. This technology is basically implemented in inexpensive solutions of neutral salts [1]. Obviously, this method does not create residual stresses on the surface of the component. So far, processing regimes and electrolyte compositions for PEP of various metals and alloys have been developed. Some physical and chemical features of the PEP have been elucidated [2,3]. At the same time, many of the basic aspects of the PEP remain unexplored. These include the role of electrical discharges and electrochemical dissolution in the removal of material from the surface of the workpiece, the nature of the current passing through the vapour gaseous envelope (VGE), the properties and conditions of formation of the surface oxide film, as well as other aspects associated with these processes. There are many published hypotheses about the electrical discharges which are called glow [4], spark [5], arc [6], and even streamer ones [7], however, without experimental evidence. Some assumptions, for example, the accumulation of a positive charge in a VGE and its explosive interaction with electrons [8], are not fully correlated with the accepted motion of charges in an electric field. The accumulation of electrolyte anions on the surface of a gas bubble supposedly capable of existing inside the VGE is also doubtful [9].
It can be considered generally accepted that a contact between the electrolyte solution and the workpiece surface occurs in the voltage range used for PEP. On the contrary, it is impossible to explain a sharp decrease in the anode temperature during the transition from the stationary heating mode, observed at lower voltages, to the polishing one. As a rule, this contact is described by the "bridges" of the electrolyte which penetrate through the VGE to anode [10]. The relationship between the frequency characteristics of the current and the properties of the polished surface is found by means of the spectral analysis of current pulsations during PEP together with the study of acoustic oscillations. This regularity can be used to monitor the on-going process [11]. A pilot setups of PEP has been developed therein the EPP treatment is carried out under the conditions of natural convection of aheated electrolyte or with its mixing [2]. Most studies were performed on flat samples or industrial components.
This article focuses on PEP in an electrolyzer with controlled hydrodynamics, which was developed for plasma electrolytic processes of diffusion saturation of metals and alloys with nitrogen, carbon, or boron [12]. These treatments are carried out in a stationary heating mode, where it is necessary to stabilize the temperature of the cooled electrolyte, which is achieved by a longitudinal flow of a cylindrical sample with the electrolyte circulating through a heat exchanger. Plasma electrolytic nitriding, carburizing, boronization, and others increase the wear resistance of steel and titanium alloys, which can be further enhanced by PEP. Performing two operations on the same setup can significantly reduce the processing time and the treatment cost. The aim of this work is to assess the role of hydrodynamic conditions in the electrolyzer on the PEP results of St45 medium carbon steel in aqueous solutions of ammonium chloride and ammonium sulphate. The study of the influence of the conditions of longitudinal flow around the samples, voltage, processing time, the concentration of electrolyte components, and the solution temperature on the surface roughness, material removal rate of samples, and current will provide information on the feasibility of PEP using a setup for anodic thermal chemical treatment.

MATERIALS AND METHODS
Samples-anodes of St45medium carbon steel (0.45 wt % C) and 12Cr18Ni10Ti stainless steel with a diameter of 10 mm and a length of 15 mm were processed in cylindrical electrolyzers under various hydrodynamic conditions. Figure 1 shows a setup with a longitudinal vertical flow around the side surface of the sample with the electrolyte that overflowed the edge of the electrolyzer into the sump, whence it was pumped into the heat exchanger. Figure 2 presents a more common scheme for PEP where polishing is carried out under conditions of natural convection. The electrolytes used were aqueous solutions of ammonium sulphate and ammonium chloride. Their temperature was controlled with a thermocouple at a distance of 30 mm from the surface to be treated. The electrolyte volume was 2 litres, which made it possible to consider their depletion insignificant in the measurement process that did not exceed 30 minutes. All of the samples were sanded with emery paper to a surface roughness of R a = 1.00 ± 0.10 μm and cleaned with acetone in an ultrasonic bath before processing. The heater was turned off when the required electrolyte temperature was reached. Further stabilization of the solution temperature was achieved by the generation of heat in the system due to the passage of current and the re-activation of the heater if necessary. The electrolyte flow rate in the case of its circulation was monitored using a LZT M15 flowmeter and maintained at 0.8 L/min.
The samples were submerged in the electrolyte to the depth of the upper end equal to 2 mm after applying the voltage, unless otherwise indicated. A part of the current lead immersed in the electrolyte was insulated with a Teflon sleeve to prevent the dissolution of the current lead and electrolyte spraying, which is characteristic of PEP. The current and voltage values were measured using a two-channel analog-to-digital converter with simultaneous recording of voltage and current with a frequency of 2 Hz. To convert the current into voltage, a 1-Ω shunt with the 1% accuracy and a 10 W dissipation power was used.
Surface roughness was measured with a TR-200 profilometer. The change in the weight of samples was determined using an electronic analytical balance Cit-izonCY224C, with an accuracy of ±0.0001 g, after washing the samples with distilled water to remove traces of salts.

RESULTS AND DISCUSSION
The largest decrease in the parameter Ra is reached with PEP at 275-325 V in sulphate electrolyte and at 300-325 V in chloride electrolyte, regardless of hydrodynamic conditions (Fig. 3). Solutions concentrations   are indicated in the weight percentage. As soon as voltage decreases from 300 to 275 V in chloride electrolyte, there emerge signs of transition to the stationary heating mode. This condition is also characterized by a slight decrease in the surface roughness and minimal dissolution of the sample material (Fig. 4). Such transition was not detected in a solution of ammonium sulphate; hence the studied voltage range corresponds to the polishing mode.
The longitudinal electrolyte flow around a sample significantly affects the resulting surface roughness only when ammonium chloride solution is used. PEP with sulphate electrolyte is fairly effective and approximately equal at all voltages regardless of hydrodynamic conditions. The forced flow of ammonium chloride leads to a contraction of the voltage range in which effective polishing is achieved. The best result is obtained at 300 V, with the initial roughness decrease of more than 5 times.
The expected consequence of the longitudinal flow is thinning of the VGE, which usually results in an increase in the current density and the likely intensification of anodic dissolution. Indeed, the forced circulation of electrolyte noticeably increases the weight loss of the samples after PEP of steel in an ammonium sulphate solution (curves 1 and 2 in Fig. 4). This regularity is even more pronounced when ammonium chloride is used which is characterized by a higher current efficiency than ammonium sulphate when steel is dissolved in the stationary heating mode [13]. Voltage has little effect on the weight loss during PEP under all conditions, which is an indirect argument in favour of the electrochemical mechanism for removing the sample material where current density is a determining parameter.
Measured dependences of current on time decrease (Figs. [5][6][7][8], which indicates an increase in the electrical resistance of a multiphase system that contains a series-connected electrolyte, a VGE, and a metal sample. The electrical conductivity of the electrolyte, even more of the steel sample, is orders of magnitude higher than that of the VGE. The surface of the samples remains light without visually observed iron oxides. Therefore, the VGE thickness is most likely to change due to insufficient stabilization of temperature in the near-anode electrolyte zone. This assumption is confirmed by the fact that the circulation of ammonium chloride solution raises the average current values by 25-30% after 5 min of PEP. This regularity is observed to a lesser extent in an ammonium sulphate solution where the average values of the current are always lower than those in the solution of ammonium chloride, other conditions being equal. In general, changes in current density correlate with the sample weight loss during PEP, which does not contradict with the electrochemical mechanism for the material removal. Polishing of St45 steel in a sulphate electrolyte with circulation of the solution occurs at lower  [14]. The surface roughness of St45 steel after its PEP rapidly decreases with the extension of the processing time ( Fig. 9), which corresponds to published data [4,15] and is modelled by an exponential function [16]. The minimum value of the parameter Ra is achieved in 5 minutes of PEP in a chloride electrolyte with its forced circulation. Similar results are obtained in an ammonium sulphate solution. The positive role of controlled hydrodynamics suggests that the used rate of the flow of a sample does not reduce the thickness of the diffusion electrolyte layer as an analogue of the counter electrode to the magnitudes of micro pro- 1 trusions on the sample surface. Otherwise, the difference in dissolution of protrusions and valleys can lead to a transition from the polishing mode to the etching one in order to develop the treated surface [17]. The linear dependence of the weight loss of the samples on the processing time does not either contradict the electrochemical nature of the polishing process (Fig. 10).
The variation in the concentration of ammonium salts permits to determine the electrolyte composi-tions providing the greatest reduction in roughness. In the case of sulphate electrolyte, a 5% solution concentration allows to reduce the surface roughness more than twice after PEP in 3 min (Fig. 11). On the other hand, a concentration of 3% is more preferred for minimizing the material removal. Reducing the concentration of ammonium chloride to the minimum value of 2%, providing a polishing mode, promotes surface roughness reduction up to the minimum value  Fig. 12). At the same time, the weight loss of the samples to be not too dependent on the electrolyte concentration.
Studies on the effect of the electrolyte temperature on the surface roughness and loss of the sample weight show that the greatest decrease in the Ra parameter is achieved at electrolyte temperatures of 70-80°C. The lowest removal of material linearly depends on temperature: the higher the solution temperature, the lower is the anodic dissolution rate (Fig. 13). An increase in the surface roughness with a rise of the electrolyte temperature from 70 to 90°C in paralleled with a monotonous decrease in the dissolution rate can be explained by additional processes on the sample surface, the study of which is beyond the scope of this work. Table 1 shows the PEP results obtained in a solution of ammonium chloride with addition of components used in polishing or those serving as nitrogen sources, carbon, or boron. Addition of oxalic acid or glycerol is found to decrease the weight loss of the samples. However, in these cases, the surface roughness does not change as compared to that at PEP in the chloride electrolyte without additives. Other additives deteriorate the surface roughness and the weight loss or intensify of anodic dissolution.
Finishing the discussion of the results obtained, we present the data on of PEP of 12Cr18Ni10Ti austenitic stainless steel in an ammonium sulfate solution for the conditions of a longitudinal flow of the electrolyte around the sample (Fig. 14). This composition is recommended by many authors for PEP of stainless steels [1][2][3]5], in contrast to ammonium chloride solutions with lower passivation of the treated surface [18]. An increase in the flow rate is established to promote a reduction in the surface roughness by more than a half. The movement of the electrolyte leads to compression of the VGE and an increase in the current density,  which is reflected in a growth of the dissolution rate and a decrease in the sample weight loss. An increase in the depth of immersion of the sample in the electrolyte is not justified since it adds to the roughness and intensity of steel dissolution (Fig. 15).

CONCLUSIONS
(1) PEP of St45 medium-carbon steel can be performed using a setup for anodic diffusion saturation of metals and alloys with nitrogen, carbon, or boron under conditions of controlled hydrodynamics with a longitudinal flow of samples with an electrolyte. The circulation of the ammonium chloride solution is established to narrow the range of operating voltages for PEP to 300-325 V, but provides the best surface roughness. Hydrodynamic conditions do not affect the surface roughness obtained after PEP in a solution of ammonium sulphate.
(2) The greatest decrease in the surface roughness of St45 medium carbon steel from the original R a 1.0 μm to 0.19 μm is achieved in 5 minutes under conditions of a longitudinal flow of a sample using 3 wt % solution of ammonium chloride at a flow rate of 0.8 L/min, voltage 300 V, and the electrolyte temperature of 80°С. The rate of the weight loss is 0.73 mg/s under these conditions. PEP of St45 steel provides the surface roughness of 0.26 μm (325 V) and a lower weight loss rate of 0.54 mg/s in the same electrolyte, but in the case of natural convection.
(3) Close values of the surface roughness of St45 steel (0.22-0.25 μm) were obtained after PEP in a 3% solution of ammonium sulphate at 275-350 V, the electrolyte temperature 80°С and its flow rate 0.8 L/min. The removal of the material during PEP in a sulphate electrolyte is slightly lower than in a chloride electrolyte and amounts to 0.6 ± 0.03 mg/s with an electrolyte flowing around the sample and to 0.5 ± 0.02 mg/s without it. The minimal surface roughness is reached when the concentration of an ammonium sulphate is 5 wt %, and the lowest weight loss of St45 steel at 3 wt % of the electrolyte concentration. The recommended temperature of a sulphate solution is 70°C, its increase leads to a growth of the surface roughness.
(4) The rate of the surface roughness reduction decreases rapidly in all studied regimes. The PEP time can be limited to 5 minutes. Additives of glycerol or oxalic acid in an amount of 3 wt % help reduce the weight loss of St45 steel without deteriorating the surface roughness.