New approach to formation of coatings on Mg-Mn-Ce alloy using acombination of plasma treatment and spraying of fluoropolymer

D.V.Mshtlyr, K.V.Ndri,*, I.M.Imshinetskiy, S.L.Sinebryukhov, S.V.Gnedenkov

aInstitute of Chemistry FEB RAS, 159 Pr.100-letiya, Vladivostok 690022, Russia

b Far Eastern Federal University, 10 Ajax Bay, Russky Island, Vladivostok 690922, Russia

Abstract This paper presents a method of the formation of composite polymer-containing coatings on a Mg-Mn-Ce magnesium alloy by forming a ceramic-like layer using plasma electrolytic oxidation (PEO) and subsequent spraying superdispersed polytetrafluoroet ylene suspension.The coating composition and their morphological features were studied by SEM, EDS, GDOES, and XRD.The presented data confir the embedding of fluoropolyme in the PEO coating.The evaluation of the corrosion properties of the formed composite polymer-containing coatings indicates a decrease in the corrosion current density by more than 3 orders of magnitude in comparison with the base PEO coating.The incorporation of a fluoropolyme in a PEO layer by more than 32% increases the load value at which abrasion of the coating to the substrate occurs and reduces the wear of the coating by more than 27 fold in comparison with the PEO layer.It has been established that composite coatings possess hydrophobic properties: the value of the contact angle attains 152 °.

? 2021 Chongqing University.Publishing services provided by Elsevier B.V.on behalf of KeAi Communications Co.Ltd.

This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)

Peer review under responsibility of Chongqing University

Keywords: Plasma electrolytic oxidation; Polytetrafluoroethylene Composite coatings; Corrosion; Wear; Superhydrophobicity.

1.Introduction

Currently,much attention is paid to the production of magnesium and magnesium alloys.The low density of these materials in combination with high specifi strength,high vibration absorbing ability and a number of important physicochemical properties make it possible to consider magnesium alloys as a valuable structural material for various industries [1-3]such as mechanical engineering, instrument engineering,aircraft manufacturing and the production of space technology, as well as medicine, etc [3-7].However, despite the advantages of magnesium alloys, these materials are electrochemically active[1]that contributes to them high susceptibility to corrosion damage as a result of exposure to aggressive media.The oxidation of magnesium under ambient air does not cause the formation on its surface of a sufficientl protective oxide layer, which could reduce the dissolution of the alloy [1,8].Additionally, magnesium and its alloys are materials with low wear resistance [9,10].In this connection, there is a high probability of destruction of the magnesium parts in the friction units if the contacting medium contains any abrasive components.Thus, the use of magnesium alloys in some mechanisms and machines is limited, on the one hand,by high corrosion activity, and on the other, by insufficien wearproof.

Today, the coatings formation is widely used as a way to protect the material against corrosion and wear [11-16].Of particular interest among existing methods for the surface treatment of magnesium alloys is plasma electrolytic oxidation (PEO) [5,17-23].The principle of this method is formation on the interface “processed metal / electrolyte” under the high electric fiel strength (up to 1-10 MV/cm) of shortlived plasma discharges, in the channel of which substances are transferred from the electrolyte to the substrate material,and their further interaction occurs [24,25].Note that after the discharge attenuation a sharp cooling of the breakdown zone from several thousand degrees down to the electrolyte temperature is realized.This effect has a significan influenc on the physicochemical properties of the formed surface layers[25,26].Thus,the PEO allows forming the multifunctional wear and corrosion-resistant,dielectric coatings on the surface of various metals and alloys [5,27-32].

Due to the sharp cooling of the breakdown zone and the gas evolution during the PEO process, the PEO coating's surface is developed, and some defects are present,such as pores and microcracks[33-36].Through these defects,the corrosive medium can penetrate to the substrate material, which further leads to its destruction.On the other hand, the presence of a developed surface makes it possible to use it as a base to be fille with various materials with purpose to form the composite coatings[3,37-44].It is rational to use polytetraflu oroethylene (PTFE) as an inert material, which possesses a number of important properties.PTFE is chemically stable in a wide range of temperatures, has high anti-friction and electrical insulating properties, and low surface energy as well[45-47].Composite coatings (CC), obtained on the basis of PEO layers using PTFE, have enhanced protective characteristics in comparison with PEO layers [48-53].

In previous works, we used the dipping in a suspension of superdispersed polytetrafluoroet ylene (SPTFE) as a method of forming the polymer-containing layers [48,53-55].However, in some cases, formation of surface layers by dipcoating is difficult Limiting factors include large square of processed detail, absence of the required areas for the installation of baths with suspensions, necessity of low consumption of substances, etc.In this regard, we have developed a method of forming the polymer-containing layers, in which the fluoropolyme component is applied on the base PEO coating by spraying (spray-coating method).The spray-coating method is universal, it allows to signifi cantly reduce material consumption, as well as easy to automate.Thus, Mateus et al.used thermal spraying of a Al2O3-TiO2ceramics and PTFE/poly(tetrafluoroet ylene-coperfluoropro ylvinylether (PFA) powders [56].Among the coatings formed by this method, the lowest coefficien of friction was obtained for the Al2O3-TiO2+ PFA samples,which, according to the authors, is a consequence of the larger amount of the incorporated polymer in these surface layers in comparison with others.In another work [57], flu oroethylene propylene hydrophobic coatings on an aluminum substrate were obtained by cold spraying.By thermal spraying, Leivo et al.have obtained polyvinylidene fluoride ethylene chlorotrifluoroet ylene, perfluoroal oxy alkane, and fluo rinated perfluoroet ylenepropylene coatings, which have significan corrosion resistance [58].However, for the best of our knowledge,the formation of coatings by a combination of plasma electrolytic oxidation and spraying of fluoropolymer has not previously been used.In accordance with this, presented article provides detailed data on the described method,as well as on the properties of the obtained composite coatings.

2.Materials and methods

2.1.Samples

The MA8 magnesium alloy (Mg-Mn-Ce system, in wt.%:1.12 Mn;0.14 Ce;0.11 Si;0.02 Cu;Mg:balance)was used as material for the manufacture of the samples under study.The sample dimensions were 30 × 15 × 1 mm3.Before applying coatings in order to standardize the surface, the samples were machined with sanding papers from #400 to #2000 sequentially.Then the samples were washed with distilled water and alcohol using an ultrasonic bath.

2.2.Coatings formation

In accordance with the previously developed principles of purposeful plasma-electrochemical synthesis of coatings on the surfaces of metals and alloys [23]and conclusions made from the analysis of literature data [16,59,60], a silicatefluorid electrolyte containing 15 g/l sodium orthosilicate(Na4SiO4) and 5 g/l sodium fluorid (NaF) was used to form the base PEO layer.

The samples were processed in two stages in a bipolar mode, in which the anode pulses periodically alternate with the cathodic ones [23,54].At the firs stage, the voltage in the anode component was increased potentiodynamically from 20 to 240 V at a sweep rate of 1.05 V/s, the cathode component was fi ed potentiostatically at -40 V (Fig.S1).At the second stage, the voltage was changed potentiodynamic in the anodic part from 240 to 200 V at a sweep rate of 0.07 V/s, in the cathodic one from -40 to -10 V at a sweep rate of 0.05 V/s(Fig.S1).The duration of the firs and second stages was 200 and 600 s, respectively.The ratio of the duration of the anode and cathode pulses was equal to 1, and the duty cycle was 50%, the frequency was 300 Hz.The breakdown voltage was 170-180 V, and the breakdown time was 140-150 s.The electrolyte temperature (10 ± 1 °C) was maintained using a ChillerSmart H150-3000 (LabTech Group, UK).

To form composite coatings in this work, we used the Forum? superdispersed polytetrafluoroet ylene (SPTFE) (Institute of Chemistry FEB RAS, Russia) obtained by the method of thermogradient synthesis of fluoroplast- (MITO, Russia)[55].Particles' size of the SPTFE powder ranged from 0.2 to 1.5μm (average size is 0.35μm).In order to increase the manufacturability of applying the composite layer, we used a 15 wt.% suspension of SPTFE powder in isopropanol [54].

Composite coatings were formed by spraying the fluo ropolymer suspension using a home-made spray gun (Fig.1)at a pressure of 0.3 MPa.The initial fl w velocity was 123 m/s.The distance from the PEO coated sample to the nozzle was equal to 20 cm.The spraying time was 5 min.Then, the samples were dried in air at room temperature and subjected to heat treatment at 310 °C for 15 min in a furnace L9/13/B180 (Nabertherm, Germany) and further cooling with the furnace.

In order to study the influenc of the multiplicity of processing the PEO layer with fluoropolyme material, SPTFEwere applied one (CC-1x), two (CC-2x) and three (CC-3x)times.Heat treatment was carried out after each polymer deposition in order to soften the polymer and embed it into the porous part of the coating.The use of such two-stage technique of polymer application made it possible to reduce the possibility of the formation of stress concentration points in the formed composite coating.According to the data of previous experiments, in the case of an increase in the amount of applied SPTFE by increasing the spraying time or the volume of the sprayed substance, a thick polymer fil is formed on the surface of the composite layer, the heat treatment of which leads to cracking of the polymer-containing layer and its partial peeling.This, in turn, leads to a decrease in protective properties.On the other hand, layer-by-layer spraying,alternating with heat treatment, ensures the gradual incorporation of the polymer into the pores of the PEO layer, thereby reducing the possibility of cracking.

Fig.1.Schematic representation of a process of applying a fluoropolyme material to a PEO-coating.

2.3.Study of the composition and morphology of coatings

To study the morphological features of the samples, an Evex Mini-SEM (Evex Analytical Instruments, USA) scanning electron microscope (SEM) was used.To evaluate the visible porosity of the coatings, the SEM images were processed using the ImageJ software (National Institutes of Health, USA).The porosityPof the coatings was calculated as the percentage of the area occupied by the pores to the total area.At processing the SEM image by the ImageJ, the sensitivity threshold was chosen, so that visually all the pores in the coating were marked.Coatings' cross-sections were evaluated using an EVO 40 (Carl Zeiss, Germany) scanning electron microscope.Additionally, this microscope was equipped with an INCA X-act device (Oxford Instruments, UK) for elemental analysis by energy dispersive spectroscopy (EDS).

The phase composition of the surface layers was determined in the “Far East Center for Structural Research” on a Rigaku X-ray diffractometer (XRD) (SmartLab, Japan), using Cu-Kαradiation.During the analysis, a Bragg-Brentano geometry focusing was used in the 2θangle range from 10 °to 80 ° with a step of 0.02 ° and an exposure time of 1 s at each point.During XRD analysis the “EVA” search program with the “PDF-2′′data bank was used.

For a detailed study of the elements distribution over the thickness of the coatings the glow discharge optical emission spectrometry (GDOES) was used using a GDS 850A device(LECO, USA).The cathode voltage was 1000 V and ion current was 30 mA, the diameter of the anode was 4 mm.

The investigation of the surface microrelief was carried out by optical laser profilomer using the OSP370 instrument installed on the M370 workstation(Princeton Applied Research,USA).Image analysis was performed using the Gwyddion 2.45 software.

2.4.Study of electrochemical properties

The electrochemical properties of samples were studied by potentiodynamic polarization and electrochemical impedance spectroscopy (EIS) using the VersaSTAT MC electrochemical system (Princeton Applied Research, USA).The measurements were carried out in a three-electrode cell at room temperature in a 3.5% NaCl solution.As a counter electrode,a platinized niobium mesh was used.As a reference electrode,a saturated calomel electrode (SCE) was used.Exposure to electrolyte surface area of the samples was 1 cm2.Before EIS tests, open circuit potential (OCP) measurements were carried out for 15 min.During EIS test the sinusoidal signal had an amplitude of 10 mV (rms).The measurement was carried out at a stabilized value of the electrode potential in the frequency range from 0.01 Hz to 0.1-1 MHz with a logarithmic sweep of 7 points per decade.

To evaluate the dynamics of changes in the electrochemical properties of the obtained coatings, the samples were kept in a 3.5% NaCl solution at room temperature for 24 h.Additionally, changes in the corrosion resistance of coatings were investigated by long-term (7 day) exposure of the samples in a 3.5% NaCl solution at room temperature.

Potentiodynamic measurements were carried out with a potential sweep rate of 1 mV/s in the range fromEC-0.15 V toEC+ 0.50 V.The Levenberg-Marquardt method was used to fi the experimental dependence of current densityIon potentialE(LEV) with following Eq.(1) [61-63]:

This method makes it possible to obtain the best fi values of corrosion potentialEC, corrosion current densityIC, the slope of the cathodic polarization curveβc, and the slope of the anodic polarization curveβa.

The polarization resistanceRPwas determined in a separate experiment with potentiodynamic polarization of the sample in the potential regionΔE=EC± 20 mV with a scan rate of 0.167 mV/s, in which the linear dependenceI=f(E) is observed:

The study of the electrochemical properties of the samples was carried out on 3 samples with the same type of surface treatment.

2.5.Evaluation of coatings wear resistance

The wear resistance of the coatings was studied using a TRB-S-DE device (CSM Instruments, Switzerland).Using this device, wear and friction coefficien were calculated.

The tests were carried out at room temperature at a sliding speed of 50 mm/s.The tribological behavior of the coatings was studied under dry friction conditions at a load of 10 N.A corundum ball (α-Al2O3) was used as a counterbody.The cross-section of the wear track was evaluated using a Surtronic 25 profilomete (Taylor Hobson Ltd, UK).Thus,the wear rate was calculated using the Eq.(3):

whereΔVis the worn volume(mm3),Nis the distance moved(m), andFis the normal load (N).

The volume loss of the sample was calculated according to the Eq.(4):

whereLis the length of the of the wear track (mm),Sis cross-section area of the wear track (mm2).

In all the experiments carried out, the wear of the counterbody was not detected and taken into account in the calculation.Wear measurements was carried out on three samples with the same type of coating.

2.6.Evaluation of coatings adhesive properties

The adhesion characteristics of the coatings were evaluated using a Revetest Scratch Tester (CSM Instruments, Switzerland).

The study adhesion by scratch-testing was carried out by measuring the critical load at which coating failure was observed.The indenter was a diamond Rockwell indenter with an apex angle of 120 ° and a radius of curvature of 200μm.The length of indenter movement along the sample surface was 5 mm with increasing load from 1 N to 20 N and loading rate 5 N/min.Adhesive properties were studied on three samples with the same type of coating.

2.7.Assessment of wettability

The wettability of the obtained coatings was studied by the method of sessile drop using a DSA100 device (Krüss, Germany).This technique allows to measure the optical contact angle (CA).The angle between the base line of the drop and the tangent to the boundary of the drop at the point of the three phases contact is measured according to [64].

The distilled water was used as a tested liquid, the drop volume was equal to 5 μl.In our work, for calculation of CA the Young-Laplace fi was used [64,65].With this fit parameters of an equation system, which models the shape of the sessile drop, are determined by means of numerical analysis.Additionally, for samples with CA exceeding 150 °, the contact angle hysteresis (CAH) was measured in order to confir their superhydrophobic properties.The measurement of CAH was carried out in accordance with the method described in[66]and consisted in measuring the difference between an advancing contact angle (ACA) and a receding contact angle(RCA).The volume of the initial drop was 10μl; this is due to the need to achieve a change in the contact line between drop and surface.Water was gradually added into the drop(with a rate of 0.05μl/s), the ACA measurement was carried out, when the drop shape did not change and the contact line began to increase.After measuring ACA, water was being removed from the droplet with a decrease in its volume.RCA measurement was carried out, when the droplet shape did not change during aspiration and the contact line began to decrease.Thus, CAH was calculated, using Eq.(5) [66]:

The contact angle and contact angle hysteresis were measured at fi e different points on the surface of the samples with different types of surface treatment.

3.Results and discussion

3.1.Composition and morphology of coatings

Data analysis of SEM images presented in Fig.2 indicates the significan change in the morphology of the coatings after spraying SPTFE particles onto the base PEO layer.For the PEO coating, a typical surface structure with pores and microdefects is observed, the sizes of which differ significantl(from 0.5-1μm to 7μm) (Fig.2a).The presence of such pores is a consequence of the sharp cooling of the breakdown zone after discharge decay up to the electrolyte temperature(about 10 °C), as well as gas evolution during PEO.After a single application of the organofluorin material, the number of microdefects significantl decreases, and visible porosityPdecreases accordingly from 18 to 6%, while the pores present in the base PEO coating are sealed with a fluoropolyme(Fig.2b).A further increase in the multiplicity of processing with SPTFE of the magnesium alloy specimens leads to a gradual decrease in the heterogeneity of the samples' surface and increase in the continuity of the polymer-containing coating (Fig.2c, d).The porosity decreases to 3% and 2%for CC-2x and CC-3x, respectively (Fig.2c, d).Note that,regardless of the multiplicity of deposition, a few cracks are present in the structure of composite coatings (Fig.2b-d),the presence of which is caused by an uneven distribution of stresses after heat treatment.However, the number of these cracks is small (Fig.2b-d).

Evaluation of the XRD data of the samples with PEO coatings revealed the presence of periclase (MgO, JCPDS Card No: 00-045-0946) and forsterite (Mg2SiO4, JCPDS Card No:01-084-1402) (Fig.3a).The presence of these compounds is a consequence of the interaction of electrolyte components with the substrate material during the oxidation process and due to the occurrence of the reactions presented below.

During the electrolysis of water,oxygen evolution,and dissolution of the magnesium alloy, an oxide-containing fil is formed on the sample surface:

Fig.2.SEM images of a PEO coating (a) and composite coatings obtained on magnesium alloy MA8 by single (b), double (c), and triple (d) superdispersed polytetrafluoroethylen isopropanol suspension spraying. P indicates the porosity of the coatings.

Sodium orthosilicate contained in the electrolyte undergoes hydrolysis with the formation of orthosilicic acid, which decomposes into silicon dioxide and water under the influenc of plasma discharges:

Then SiO2interacts with MgO, resulting in the formation of magnesium orthosilicate:

The processes of formation of a PEO coating in a silicatefluorid electrolyte are described in more detail in [23].

The diffractogram of the CC-1x contains a PTFE pattern(Fig.3b, JCPDS Card No: 01-047-2217), the presence of which is due to the incorporation of a fluoropolyme into the base PEO layer.

The presence of magnesium (JCPDS Card No: 00-035-0821) in the diffraction patterns (Fig.3) is explained by the low absorption capacity of the PEO layer and the penetration of X-rays to the Mg substrate during the analysis.

Fig.4.SEM image of a cross section of a CC-1x and maps of the distribution of elements within the coating thickness: magnesium, oxygen, silicon, and fluorine according to EDS data.

In previous works, we have investigated cross-sections of a PEO coating, as well as a number of polymer-containing coatings [42,53,54].In order to detailed study, the elemental composition of the composite polymer-containing coating obtained by a single application of SPTFE (CC-1x), an energy-dispersive spectroscopy of a cross-section was carried out(Fig.4).The coating contains magnesium(substrate material), evenly distributed within the coating thickness (Fig.4).The presence of oxygen and silicon in the composition of the surface layer (Fig.4) is a consequence of the formation of magnesium oxide and silicate (Fig.3) during PEO.The presented results are fully confirme by the previously obtained data [23,54].The fluorin distribution as a SPTFE component indicates sealing the pores and the formation of a continuous polymer layer (Fig.4).Fluorine is unevenly distributed within the composite layer.The largest amount of fluorin is observed on the coating surface (Fig.4), which is due to the formation of a thin polymeric fil during the spraying and subsequent heat treatment of the CC.Additionally, fluorin embedded into the base PEO layer (Fig.4), which confirm the assumption that the microdefects of the PEO coating are sealed with fluoropolyme material.The presence of fluorin in the non-porous sublayer at the coating / substrate interface is due to the incorporation of this element from the electrolyte and the formation of MgF2during the plasma electrolytic oxidation process [23,54].The absence of the MgF2on XRDpatterns is due to the fact that the amount of this substance is less than the detection limit of compounds for XRD analysis(about 10%).

Fig.5.Distribution of elements within the thickness of a PEO coating (a) and a CC-1x (b), according to GDOES data.

Evaluation of the content of such elements as magnesium,oxygen and silicon in the base PEO layer by the GDOES revealed the certain regularities in their distribution within the coating thickness.Thus, a large amount of silicon is in the outer part of the coating (Fig.5a), with thickness about 5 μm.The concentration of Si decreases sharply toward the coating/substrate interface.Thereby, most of the MgSiO4is in the near-surface layer.Apparently, the transport of siliconcontaining compounds in the channels of plasma discharges to the substrate material is somewhat difficul because of their large size.Oxygen and magnesium are evenly distributed within the coating thickness (Fig.5a).At a depth of about 28 μm, a sharp decrease in the oxygen content and an increase in the magnesium content are observed(Fig.5a).Based on this, it can be concluded that the thickness of the PEO coating is near to 28μm.A similar elements distribution within the thickness was observed also for composite coatings (Fig.5b).In addition, for CC-1x, the presence of carbon in the outer part of the coating is observed (Fig.5b), which is a consequence of the incorporation of SPTFE.Note that the obtained data are in good agreement with the EDS results presented in Fig.4.

Incorporation of the fluoropolyme into the base PEO layer did not cause a significan change in thickness, since during heat treatment, the polymer penetrated into the pores of the coating, and only a thin polymer fil of about 2-3μm was formed on the surface (Figs.4, 5b).

The study of the surface topography of the samples revealed the certain differences in their microrelief.Thus, it was found that PEO coatings have the smaller difference of the roughness height in comparison with composite layers(Fig.6, Table 1).

The polymer-containing coatings have a more continuous structure (Fig.2).However, due to the presence of a thin fil of polytetrafluoroet ylene on the surface of the CC, on whichagglomerates of fused particles are clearly visible(Fig.2b-d),such layers have a large roughness.It should be noted that an increase in the multiplicity of SPTFE deposition from one to two leads to an increase in the surface roughness (Fig.6b,c, Table 1), which is also confirme by the analysis of SEM images (Fig.2b, c).

Table 1Roughness parameters (arithmetical mean height Sa and root mean square height Sq) of samples with different types of surface treatment.

Furthermore, for CC-3x, the roughness of the coatings decreases, an alignment of the microrelief is observed (Figs.2c,d, 6c, d, Table 1), due to the fillin of the irregularities of the coating with the polymer.Thus, using the multiplicity of spraying, it is possible to vary the microrelief of the coatings within a wide range, thereby varying the properties of the surface layer.

3.2.Electrochemical properties of coatings

Fig.7 shows data of the change of the OCP during 900 s of immersing the samples in a 3.5%NaCl solution.Based on the analysis of Fig.7,it can be concluded that the potential values for an uncoated magnesium alloy change insignificantl over the exposure time and remain in the range from -1.60 to-1.70 V (Fig.7), which corresponds to the ordinary value of the potential for magnesium and a number of magnesium alloys in 3.5% NaCl [1].

For PEO coated sample the electrode potential variates insignificantl , and the changes in its values is not observed almost (Fig.7).The increase in OCP in comparison with untreated sample is a consequence of the formation of a PEO layer with protective properties on the surface of the magnesium alloy.

Fig.6.Surface topography of a PEO coating (a) and composite coatings obtained on magnesium alloy MA8 by single (b), double (c), and triple (d) application of superdispersed polytetrafluoroethylene

Fig.7.Dynamics of changes in OCP for 15 min of exposure in 3.5% NaCl solution for samples with different types of surface treatment.

After a spraying of the fluoropolyme material onto a base PEO coating, the OCP of alloy with CC on the surface increases compared to PEO coating, and its initial value depends on the multiplicity of polymer spraying (Fig.7).Note that for CC-1x, after 300 s exposure in the electrolyte, the OCP sharply decreases to -1.42 V, and then fluctuate within the range from -1.53 to -0.89 V (Fig.7).Such significan fluctuation in the potential are a consequence of insufficien continuity of the applied polymer-containing layer and presence of not fully sealed pores, this leads to the appearance of local areas, in which the potential values are near to the values obtained for the PEO coating.With double deposition of SPTFE a smooth increase is observed in the range from-0.95 to -0.62 V (Fig.7).The highest potential (-0.25 V)was recorded for the polymer-containing layer obtained by a triple application of the fluoropolyme (Fig.7).Note that for CC-3x the highest values of the OCP for 900 s among all composite coatings were also obtained (Fig.7).An increase in the potential values and its stability during the immersing time is a consequence of sealing the porous part of the base PEO layer and the formation on the surface of CC-2x and CC-3x of an even polymer-containing layer with the least number of defects (Fig.2c, d).

The smaller number of fluctuation of the potential of the PEO coating over time in comparison with the samples with composite layer is a consequence of the fact that the corrosion rate of the samples with the PEO coatings is much higher.Thus, during the corrosion process, reaction products are quickly formed in the pores of the PEO coating; these compounds block the access of an aggressive medium to the substrate.This, in turn, leads to a decrease in the number of fluctuation in the potential of the PEO coating over time.

The corrosion properties of the samples were investigated by potentiodynamic polarization test (Fig.8).The formation of a PEO coating on the surface of the magnesium alloy significantl reduces the corrosion current densityICand increases the polarization resistanceRP(Fig.8, Table 2), which significantl reduces the corrosion of the processed material.Evaluation of the data of electrochemical tests indicates the positive effect of incorporating polytetrafluoroet ylene into the PEO coating (Fig.8, Table 2).According to the presented data,a single application of SPTFE reduces the corrosion cur-rent density by 2 orders of magnitude in comparison with the PEO coating (Fig.8, Table 2).A further increase in the multiplicity of treatment with a fluoropolyme (CC-2x) leads to an enlarge in the polarization resistance by more than 3-fold in comparison with CC-1x (Fig.8, Table 2).

Table 2Corrosion properties (corrosion potential EC, corrosion current density IC, cathodic βc and anodic βa Tafel slops, polarization resistance RP) of samples with different types of surface treatment.

Fig.8.Polarization curves for samples with different types of surface treatment.

The highest protective properties were demonstrated by samples with triple application of SPTFE.For this polymercontaining layer, the corrosion current density decreases, and the polarization resistance increases by more than 6 and 3 orders of magnitude in comparison with the uncoated magnesium alloy and the base PEO coating, respectively (Fig.8,Table 2).

Electrochemical impedance spectroscopy data are presented in Bode plot (dependences of the impedance modulus |Z| and phase angleθon frequencyf) and Nyquist plot(dependence of the imaginary part of the impedanceZ′′on the real part of the impedanceZ′).Based on the analysis of the EIS results, it can be concluded that the formation of a PEO coating on the surface of the magnesium alloy leads to an increase in the impedance modulus at low frequencies|Z|f=0.01Hzby almost 2 orders of magnitude (Fig.9a,Table 3).The presence of a fluoropolyme in the CC-1x leads to an increase in|Z|f=0.01Hzby more than 2 orders of magnitude in comparison with the base PEO coating (Fig.9a,Table 3).An increase in the number of treatments of the composite layer up to twofold with superdispersed polytetrafluoroet ylene (CC-2x) makes it possible to increase the values of|Z|f=0.01Hzby 28 times in comparison with CC-1x (Fig.9a, Table 3).The highest values of the impedance modulus were obtained for polymer-containing layers treated with a fluoropolyme by three times (CC-3x).For this type of coatings|Z|f=0.01Hzis 7.6 × 108Ω× cm2, which is more than 6 orders of magnitude higher than the values obtained for a bare magnesium alloy (Fig.9a, Table 3).A significan increase in the|Z|f=0.01Hzfor composite coatings in comparison with the PEO layer is a consequence of the sealing of the pores and defects of the base PEO coating with the organofluorin material (Figs.2-5).

Table 3Calculated parameters of equivalent electrical circuits' elements for samples with different types of surface treatment.

There is a band in the mid-frequency region with a minimum of about -50 ° in the Bode plot for uncoated sample(Fig.9b).At low frequencies, this plot goes into the area of positive values (Fig.9b).The Nyquist plot for an uncoated magnesium alloy sample is characterized by a capacitive semicircle at medium and high frequencies and an inductive semicircle at low frequencies (Fig.9d).The presence of a capacitive component is a consequence of the resistance to charge transfer at the metal / electrolyte interface, as well as the capacity of the electrical double layer.The presence of the inductive part on Nyquist plot and positive values of phase angle at low frequencies on Bode plot are explained by dissolving of the alloy and the adsorption of intermediate products during the corrosion process [67].Note that the electrochemical behavior of the sample at low frequencies was irreproducible, which is probably caused by the destruction of the sample during the pitting corrosion.In this regard,the frequency range used for the analysis and fittin of experimental data was deliberately limited, which is a common practice in impedance studies [67].Thus, the spectrum of an uncoated magnesium alloy was fitte using a simple equivalent electrical circuit (EEC) with oneR2-CPE2, whereR2is the charge transfer resistance andCPE2is the capacitance of the double electrical layer (Fig.10a).In this and subsequent EEC, the symbolRsindicated the resistance of the electrolyte, and its value for all studies equal to 37 ± 2Ω.

To fi the experimental impedance data in this work, a constant phase element (CPE) is used instead of the ideal capacitance [42].The use ofCPEin EEC is due to the heterogeneity of the systems under study.TheCPEimpedance can be calculated using the Eq.(13):

whereωis angular frequency (ω= 2πf),jis the imaginary unit,nis the exponential coefficientQis the frequency independent parameter.

Fig.9.Bode plots (dependence of impedance modulus |Z| (a) and phase angle θ (b) on frequency f) and Nyquist plots (dependence of the imaginary part of the impedance Z′′ on the real part of the impedance Z′) (c, d) for samples with different types of surface treatment.The symbols indicate the experimental data, and the lines is the fittin curves calculated in accordance with the proposed models.

The dependence of phase angle on frequency of a PEO coated sample has two bands, one of which is located in the high-frequency region with an extremum about -60 °, and the second one is in the mid-frequency region with a minimum of about -40 ° (Fig.9b).The Nyquist plot of a sample with PEO layer also has two capacitive semicircles in the region of medium and high frequencies (Fig.9d).The presence of two time constants in this dependence is explained by the structure of the PEO coating.As is known [54], a PEO coating consists of two layers: an outer porous layer (time constant in the high-frequency region) and a thin inner poreless sublayer(time constant in the mid-frequency region).To describe the presented dependence, an EEC with twoR-CPEcircuits was used (Fig.10b).Here, theR1-CPE1element describes the geometric capacitance including the outer porous part of the PEO coating, and theR2-CPE2element describes the inner poreless one.

The spectra of composite coatings also contain two time constants (Figs.9b, c, S2).For CC-1x, these constants practically merge; however, just like for CC-2x and CC-3x, they are located in the high-frequency and mid-frequency regions(Figs.9b, c, S2).Note that the extremums on the graphsθ(f)for polymer-containing layers are located higher than those for the base PEO layer (Fig.9b), that is connected with more capacitance character of composite coatings.On the Nyquist plots for various types of coatings, the values ofZ' andZ′′increase in the following order: PEO coating →CC-1x →CC-2x → CC-3x (Figs.9c, d, S2).The data presented in Figs.9 and S2 indicate an improvement in the corrosion resistance of composite coatings in comparison with the PEO layer.To describe the charge transfer at the electrolyte /polymer-containing coating interface, a two-R-CPE-circuits was used, as for the base PEO coating.TheR1-CPE1circuit describes the outer porous part of the composite coating,namely the PEO layer fille with polymer, and theR2-CPE2element describes the inner poreless part (Fig.10c).

Fig 10.Models of the coatings structure and corresponding equivalent electrical circuits used to fi experimental impedance data: one-R-CPE-circuit for magnesium alloy without coating (a), two-R-CPE-circuit for a PEO coating (b) and composite coatings (c).

Analysis of the change in the calculated parameters of equivalent electrical circuits' elements indicates a significan effect of the coatings' formation method on charge transfer at the electrode / electrolyte interface.After applying the fluoropolyme material on the base PEO layer, a significan increase in the resistance of the outer porous layerR1is observed, while theQ1value is decreased (Table 3)that is the result of the increase in outer layer thickness due to SPTFE spraying.Additionally, it is observed an increase ofn1, which indicates an increase inthe homogeneity of the coating.Note that for CC-2x,higher values ofR1andn1as well as lower values ofQ1were obtained in comparison with CC-3x (Table 3).Apparently, this is a consequence of the decrease in the pore depth (as a result of fillin with a polymer) for a composite layer obtained by threefold treatment with SPTFE in comparison with CC-2x.ResistanceR2of the poreless inner layer increases with an increase in the multiplicity of treatment of the base PEO coating with a fluoropolyme (Table 3).Moreover, a gradual decrease inQ2is observed; this is a consequence of the polymer sedimentation on the bottom of pores of the PEO layer and an increase its inner sublayer thickness(Fig.10, Table 3).The increase in then2in the following order: CC-1x →CC-2x →CC-3x confirm the greater homogeneity of the formed poreless sublayer due to the penetration of the polymer to the bottom of the pores during heat treatment (Fig.10, Table 3).

In order to study the dynamics of changes in the electrochemical properties of composite coatings, the samples were kept in a 3.5% NaCl solution for 24 h.The evaluation of the change in the values of the impedance modulus, measured at a frequency off= 0.01 Hz, after 5, 10 and 24 h of exposure,indicates the saving the high protective properties of polymercontaining layers: the|Z|f=0.01Hzis more than 2 orders of magnitude higher than that obtained for the base PEO coating after 15 min of exposure time (Figs.9, 11, Tables 3, 4).Note that, regardless of the exposure time, the corrosion resistance of coatings increases in the following order: CC-1x →CC-2x →CC-3x (Fig.11), which is the result of an increase in the amount of applied fluoropolyme, the uniformity and thickness of the polymer layer and,as a consequence,the protective properties of composite coatings.However, changes in the values of|Z|f=0.01Hzfor each type of composite coating are different.Thus, for CC-1x in the firs 5 h of exposure|Z|f=0.01Hzincreased to 1.6 × 107Ω× cm2, and then gradually decreased (Fig.11a, c, e, Table 4).

Apparently, the increase in the values of the impedance modulus in the firs 5 h of the experiment is a consequence of the formation of corrosion products in some pores of the coating not completely sealed with polymer; these products partially block the pores, thereby reducing the access of an aggressive medium to the substrate.A similar trend is also observed for CC-2x and CC-3x (Fig.11a, c, e).However, for these coatings, the highest|Z|f=0.01Hzwas recorded after 10 h of exposure, which indicates the formation of corrosion products at a later stage of the experiment and, consequently,the higher corrosion resistance of these samples in whole(Fig.11a, c, e).On the Bode plots for all composite layers two time constants are clearly distinguishable in the highfrequency and mid-frequency regions (Fig.11b, d, f).Table 4 presents the calculated parameters of the applied EEC.For a composite coating obtained by single spraying of SPTFE, a gradual decrease in the resistance values of the porous layerR1and an increase inQ1with an increase in the exposure time in a 3.5% NaCl solution (Table 4) are observed, which is a consequence of the gradual penetration of electrolyte into the pores of the coating and the occurrence of the corrosion process.Additionally, there is a decrease ofn1(Table 4), which indicates an increase in coatings' heterogeneity as a result of the formation of corrosion products.A similar trend persists forR2(resistance of the inner sublayer of the coating) andQ2(analogous of the capacity of the inner sublayer of the coating) (Table 4).

For a polymer-containing coating obtained by double treatment with a fluoropolyme,R1has the highest values(1.1 × 105Ω× cm2) after 10 h of exposure (Table 4).For the same exposure time, the lowestQ1and the highestn1values were obtained (Table 4).The highest values ofR2andn2,as well as the lowest values ofQ2, were also obtained after 10 h of keeping the samples in the electrolyte.This change in the EEC parameters describing condition of the electrolyte /composite layer interface is apparently a consequence of the formation of corrosion products, which partially reduce the electrolyte's access to the substrate material.

For CC-3x, as well as for CC-2x, the highest values of the calculated parameters of the equivalent circuit are also observed after 10 h exposure in the corrosive environment(Table 4), which is explained by a similar mechanism of the corrosion process for both composite coatings.For all types of coatings after 24 h of exposure, a decrease in |Z|f=0.01Hzwas observed, which indicates the development of the corrosion process and partial penetration of an aggressive medium to the substrate (Figs.9, 11, Tables 3, 4).However, despite a slight decrease in the corrosion properties of composite lay-ers, all of them, regardless of the multiplicity, demonstrated high values of |Z|f=0.01Hz, which are more than 2 and 4 orders of magnitude higher than the values obtained for PEO coating and uncoated magnesium alloy, respectively (Figs.9,11, Tables 3, 4).

The evaluation of the resistance of various coatings to long-term exposure to a corrosive environment is presented in the form of changes in the values of the impedance modulus|Z|f=0.01Hzdepending on the exposure time (Fig.12).Based on the analysis of the presented data, it can be concluded that for the PEO coating the most significan decrease in the protective properties was observed in the firs 72 h of exposure(Fig.12a).Subsequently, the values of the impedance modulus became stationary(Fig.12a).This is due to the penetration of a corrosive medium to the substrate material, interaction with it with the subsequent formation of corrosion products,as noted above.Nevertheless, the values of the impedance modulus for the PEO coating even on the 7th day exceed the values obtained for the bare magnesium alloy (Figs.9a, e,12a, Table 3); this indicates that the PEO coating preserves a certain protective resource.

Note that all composite coatings demonstrated signifi cantly higher corrosion resistance compared to the PEO layer(Fig.12).However,the largest decrease in the protective properties for samples with CC during a 7 day exposure was observed in the firs 48 h (Fig.12b).This decrease is also due to the interaction of the corrosive environment with the magnesium substrate.Further, a gradual decrease in the values of the impedance modulus was observed, with an exit to stationary values after 120 h (Fig.12b).Note that even after 168 h of exposure in a corrosive environment, the impedance|Z|f=0.01Hzwas higher than for a sample with a PEO coating after 15 min of exposure to a 3.5% NaCl solution (Figs.9a,12, Table 3).

3.3.Tribological properties of coatings

Analysis of the results of tribology test indicates a significant effect of the fluoropolyme material on the wearproof of coatings, as well as on the character of the wear process.Thus, the base PEO coating was abraded during about 2200 cycles (Fig.13).A linear increase in the coefficien of friction is observed, which is a consequence of the uniform abrasion of the PEO layer during the experiment (Fig.13).For polymer-containing coatings, the wear process changes significantl.Regardless of the multiplicity of SPTFE spraying,two stages of abrasion can be distinguished for all composite layers (Fig.13):

1.Uniform abrasion of the outer polymer containing fil(SPTFE on the surface and in the pores) of the composite coating, characterized by low values of the friction coeffi cient of about 0.10-0.25.

2.Abrasion of the composite layer to metal, at which there is a sharp increase in the coefficien of friction to 0.60-0.70.

Note that the duration of the firs stage directly depends on the multiplicity of polymer deposition.Thus, CC-3x withstands the highest number of the cycles (about 130,000) until the second stage (Fig.13), duration of which varies insignificantly depending on the studied composite coating and is about 5000-7000 cycles (Fig.13).Thus, the wearproof of the samples is primarily affected by the duration of the firs stage,which is a consequence of the multiplicity of the SPTFE deposition.Each subsequent application of the organofluorin substance increases the thickness of the composite coating and ensures uniform distribution of the fluoropolyme in the pores and on the surface of the PEO layer (Fig.2).Low values of the coefficien of friction (about 0.10) during long time (Fig.13) make it possible to characterize the formed composite coatings as anti-friction.

According to the analysis of the data presented in Table 5,the incorporation of a polymer material into the PEO coating reduces the wear of the coatings at least by 1.8 fold.The highest wearproof was demonstrated by CC-3x sample, for it the wear was reduced by more than 27 times in comparison with the base PEO coating (Table 5).

3.4.Adhesive properties of coatings

The destruction process of the base PEO layer and composite coatings under the influenc of the applied load occurs according to a single mechanism: the indenter evenly penetrates into the formed coating, as the load grows, it is pressed into the coating material, which leads to its adhesivedestruction (Fig.14).Analysis of the adhesion properties of samples enables one to conclude that the fluoropolyme material has a positive effect on the adhesion of the coatings to substrate.LoadLC2, at which the beginning of peeling of coating areas was observed, is more than 30% higher for polymer-containing layer, than for the PEO coating (Fig.14,Table 5).Note that the beginning of the peeling of the coating for CC-1x and CC-2x occurred at approximately the same load: (6.0 ± 0.3) N and (6.1 ± 0.2) N, respectively (Fig.14b,c, Table 5).The highestLC2values were observed for CC-3x((7.0 ± 0.4) N), which is more than 52% higher than for the PEO layer (Fig.14d, Table 5).A significan increase in the values ofLC2is a consequence of an increase in the continuity and uniformity of the composite coating obtained by threefold deposition of the polymer in comparison with other polymer-containing layers (Fig.2).TheLC3value at which abrasion of the coating to the substrate occurs for the composite layers is more than 28% higher than for the PEO coating(Table 5); this is the result of the presence of a fluoropolyme on the CC surface and inside of pores, which has a low coefficien of friction and acts as a dry lubricant.The difference inLC3for polymer-containing coatings changes insignificantl(within 4%).Since, after the penetration of indenter into the outer porous layer fille with polymer, the destruction of the inner sublayer occurs at the same loads (Fig.14).

Fig.14.Optical images of scratches on the surface of a PEO coating (a) and composite coatings obtained by a single (b), double (c), and triple (d) application of superdispersed polytetrafluoroethylene

Fig.15.SEM images of composite coatings and optical images of a sessile drop on the surface of these coatings with a single (a), double (b) and triple (c)application of superdispersed polytetrafluoroethylene

The data obtained(Fig.14,Table 5)are in good agreement with the results of tribological tests, in which the duration of the second stage of abrasion also varied insignificantl(Fig.13).

3.5.Wettability of coatings

The assessment of the wettability of the coatings indicates a positive effect of the fluoropolyme on the hydrophobic properties of the coatings.Thus, after a single treatment with superdispersed polytetrafluoroet ylene, the previouslyhydrophilic PEO coating possesses the hydrophobic properties, while the CA values attains 148 ° (i.e., up on 100 ° more than for the PEO layer) (Table 6).Double spraying allows increasing the CA values up to 152 ° (Table 6).However, a further increase in the multiplicity of polymer treatment leads to a decrease in the contact angle by 14 °.This decrease in CA values is a consequence of changes in the morphological structure of coatings.It is well known that the surface structure directly affects its wettability [68].For CC-1x and CC-2x, a developed surface is observed, on which roughness of various levels is presented (Fig.15a, b).After increasing the multiplicity of SPTFE spraying and subsequent heat treatment up to three times, the morphology of the formed composite coating becomes less developed, i.e., more even and more continuous (Fig.15c), which leads to a decrease in the contact angle.

Table 6Wetting of samples with different types of surface treatment.

It is known that a surface, for which the CA value exceeds 150 °, and the CAH is less than 10 °, are superhydrophobic[69].For polymer-containing layers obtained by double deposition of a fluoropolyme, the CA was equal to 152 ° However, the measured CAH for CC-2x was equal to at least 11 °,which does not allows characterizing this coating as superhydrophobic.Apparently,the morphological structure formed on the surface of the polymer-containing layer is close to hierarchical (the presence of nanolevel roughness on the microlevel roughness) (Fig.15b).

4.Conclusion

For the firs time, composite coatings were formed on the Mg-Mn-Ce magnesium alloy by plasma electrolytic oxidation and subsequent spraying of isopropanol suspension of superdispersed polytetrafluoroet ylene.

The coating obtained by threefold polymer treatment reduces the corrosion current density by more than 3 orders of magnitude in comparison with the base PEO layer and by more than 6 orders of magnitude in comparison with the uncoated magnesium alloy.Polymer-containing layers have sufficien corrosion resistance during the exposure in a corrosive environment for 7 days.

Composite coatings have antifriction properties, reducing wear by more than 27 fold compared to the base PEO coating.The presence of polymer in the coating's composition allows an increase in the load values at which the abrasion of the coating to the substrate occurs by more than 32% in comparison with the base PEO coating.

CC-3x composite coatings possesses high hydrophobic properties, the contact angle attains 152 °, and the contact angle hysteresis is 11 °.

Funding

This work was supported by the Russian Science Foundation [Grant No.20-73-00280].

Acknowledgments

The authors express their sincere gratitude to their colleagues: Senior Researcher A.K.Tsvetnikov for the synthesis of superdispersed polytetrafluoroet ylene and the Head of the laboratory A.V.Gerasimenko for the XRD analysis of the samples.

Additionally, authors thank T.Mathenia and A.Isakov from the LECO Corporation for the elemental analyzes.

Supplementary materials

Supplementary material associated with this article can be found, in the online version, at doi:10.1016/j.jma.2021.07.020.