Investigation of mutual effects among additives in electrolyte for plasmaelectrolytic oxidation on magnesium alloys

Lingyun An, Ying Ma, Le Sun, Zhanying Wang, Sheng Wang

State Key Laboratory of Advanced Processing and Recycling of Non-ferrous Metals, Lanzhou University of Technology, Lanzhou 730050, China

Abstract Plasma electrolytic oxidation (PEO) coatings were prepared on AZ91D magnesium alloys in alkaline silicate-based electrolyte with and without additives.The mutual effects among additives including TiC particles, dispersant polyethylene glycol 6000 (PEG6000) and anionic surfactant sodium dodecyl sulfate (SDS) were studied based on orthogonal experiment.The content and distribution of TiC deposited in the coatings were measured by EPMA and EDS.The thicknesses, phase compositions, microstructures and corrosion resistances of the coatings were examined by using TT260 eddy current thickness gage, XRD, SEM and electrochemical test, respectively.The results show that the experiment design of this study is the key to study the mutual effects among these additives.Each additive and their interactions all remarkably influenc TiC content and corrosion resistance of the coatings.Smaller size TiC is much easier to migrate towards the anode,and the interaction between PEG6000 and SDS both effectively prevents its agglomeration and increases the number of its negative surface charges, which further increase the migration rate and the deposited uniformity of TiC and make TiC have more opportunity to deposit in the discharge channel.Thus, when smaller size TiC, PEG6000 and SDS are all added into the electrolyte, they could improve the anti-corrosion property of the coating to the largest extent attributed to higher TiC content and the densest microstructure of the coating.

Keywords: Magnesium alloys; Plasma electrolytic oxidation; Additives; Mutual effects; Orthogonal experiment; Corrosion resistance.

1.Introduction

Plasma electrolytic oxidation (PEO), as a relatively novel surface treatment technique, possesses a large number of advantages such as less pre-treatment process, easier operation and environmentally friendly electrolyte.More importantly, it could directly fabricate ceramic-like coatings on the surface of magnesium, aluminum, titanium and their alloys, and the obtained coating can greatly improve the corrosion resistance,wear resistance and insulating property of these metals [1-5].

The corrosion resistance of PEO coatings could be affected by a series of process parameters like electrolyte, electrical parameters and treatment time, in which the influenc of the electrolyte is the most important [6-11].At present, some primary electrolyte systems used for preparing PEO coatings on magnesium alloys include silicate, phosphate and aluminate as well as their complex systems.Nevertheless, the PEO coatings prepared on magnesium alloys in these electrolyte systems are mainly composed of MgO phase whose stability and anti-corrosion property are poor in harsh service environments, so the substrate can’t be protected effectively.Meanwhile, these coatings also inevitably possess porous features,which provide the channel for the corrosion medium to penetrate into the coatings and corrode the substrate.For these reasons, additives have obtained the widespread attention of many researchers.

Some researchers carried out the investigation by adding CeO2, ZrO2, TiO2, PTFE and SiC particles into the electrolytes and the results showed that these anti-corrosion phases could be incorporated into the coating.Meanwhile,these particles were fille into the micro-pores of the coating through the electrophoresis and diffusion process, increasing the compactness of the coating.As a result, the property of the coatings was noticeably improved [12-17].Q.P.Tran et al.[18] prepared diamond powder incorporated oxide layers by adding different concentration of diamond powder into the electrolyte,and the results suggested that the thickness,roughness, corrosion and wear resistance of the coatings improved with increasing content of diamond powder in the electrolyte from 0 to 6g/L.S.V.Gnedenkov et al.[19] pointed out that the addition of SiO2or ZrO2to silicate-fluorid electrolyte leaded to an increment in the coating thickness formed on MA8 Mg alloy via PEO process, which is beneficia to the improvement of the anti-corrosion performance of the coatings.Further, S.Fatimah et al.[20] investigated that the influenc of dual incorporation of SiO2and ZrO2nanoparticles on the microstructure and corrosion resistance of the coating.The results indicated that the coatings containing both ZrO2and SiO2nanoparticles exhibited excellent corrosion protection properties due to the combination of their roles as the micropores blocker and cracks fille so that the microstructural defects were minimized.The literature [21] found that the corrosion resistance of the coatings has not been improved after depositing Si3N4particles into the coatings, and the dense PEO coating could not be produced in the electrolyte containing Si3N4particles with average size of 0.02μm.Whereas,M.M.Krishtal et al.[22] studied that the effect of dispersed different sized SiO2nanoparticles on the composition and properties of PEO coatings formed on magnesium 9995A.And the results showed that the fine dispersed 25nm SiO2are the most chemically reactive, and the obtained coating demonstrated the best wear properties and corrosion resistance.

It is apparent to all that the addition of particles, especially ultra-fin particles, is apt to cause the unstability of the electrolyte, which may lead to the unstability of PEO process and decrease the compactness of the coating.As what was referred in literatures [23-26], the stability of the PEO process can usually be improved greatly by doping organic additives such as tannic acid, glycerol, potassium acid phthalate and EDTA-Na into the electrolyte.Accordingly, the uniform, compact and excellent coating can be obtained.What’s more, the authors in [27] pointed out that organic additives,mainly through the combined effect of inhibition, arc suppression, and surfactant, affect the anodizing process and the performance of the oxide film

The current studies, however, mainly focus on the influ ences of ultra-fin particles and organic additives, as individual variables, on the PEO coatings.It is lack of the enough information about the influence of the organic additives on the coating containing micro/nano-particles, especially the reports about the effects of the mutual effects between ultra-fin particles and organic additives are still fewer [28].

The aim of this study is to study the mutual effects between ultra-fin particles and organic additives in detail.In this paper, TiC particles were added to deposit them into the coating due to their high chemical stability and excellent anticorrosion property.Meanwhile, in order to effectively impede the agglomeration of TiC particles in the electrolyte and increase their negative surface charges,organic additives including polyethylene glycol 6000 (PEG6000) and sodium dodecyl sulfate (SDS) were doped, which are the most common and widely used dispersant and anionic surfactant respectively.Then based on the orthogonal experiment design, the effects of TiC size, PEG6000 and SDS on the TiC content and the corrosion resistance of the coatings were investigated, especially the influenc of their interactions.In addition, influenc mechanisms of the mutual effects among these additives on the coatings were discussed.

Table 1Factors and levels of orthogonal experiment.

2.Experimental

2.1.Materials and preparation of coatings

An ingot of AZ91D magnesium alloy cut into the specimens of 30mm×20mm×10mm was used as the substrate material and its normal chemical composition are as follows(mass fraction): Al 8.3% ～9.7%, Zn 0.35% ～1.0%, Mn 0.17% ～0.27%, Si 0.1%, Cu 0.03%, Ni 0.002%, Fe 0.005%,and Mg balance.Before PEO treatment,each sample was successively ground by a series of SiC abrasive paper, and then was washed with distilled water and dried under normal atmospheric conditions.PEO process was performed for 21min under the constant voltage mode using a homemade bipolar pulsed power supply with a square waveform output shown in Fig.1.In Fig.1, the frequency and duty ratio of the positive pulse are the same as those of the negative pulse, and they are 800Hz and 20% respectively.At the same time, the value of the positive pulse voltage is fi ed at 450V and the negative one is set as zero invariably.The sample was used as the anode and a stainless steel served as the cathode.During this process, the electrolyte temperature was maintained below 20°C by a water cooling system.

The basic electrolyte, define as No.0 scheme in the present study, was an alkaline silicate-containing system, and was composed of 15g/L Na2SiO3,13g/L KF and 2g/L NaOH.Three kinds of additives, namely, 5g/L of different sized black-gray TiC particles, dispersant polyethylene glycol 6000(PEG6000) and anionic surfactant sodium dodecyl sulfate(SDS), were added into the base electrolyte.And TiC size(A), PEG6000 (B) and SDS (C) were chosen as three factors of this experiment.The orthogonal experiment with three factors and two levels was designed based on orthogonal table L8(27), and the mutual effects among factors (A×B,A×C, andB×C) were taken into account.Each factor along with their corresponding levels is listed in Table 1.Meanwhile, column name of orthogonal table L8(27) is given in Table 2, and the detailed experiment schemes are shown in Table 3.Furthermore,the content of Ti element and the corrosion current density of the coatings, analyzed with the methods of range and variance analysis by using Excel software,were used as evaluation indexes.

Fig.1.Sketch diagram of voltage output waveform of bipolar pulsed power supply.

Table 2Column name of orthogonal design table L8(27).

2.2.Characterization of coatings

Size distribution and zeta potential of TiC particles in the alkaline silicate electrolyte were measured by using a laser particle size analyzer and a Zeta potential analyzer (Zetasizer Nano ZS, Malvern).The thicknesses of the coatings were investigated by TT260 eddy current thickness gage.The phase composition of the coatings was analyzed by X-ray diffractometer (XRD, D/MAX-2400, Rigaku) using a Cu Kαradiation source with a step length of 0.02°at a scan range from 20 to 80° (in 2θ).The content and distribution of Ti element in the coatings were examined by electron microprobe analyzer(EPMA-1600).The surface and cross-sectional morphologies of the coatings were studied by fiel emission scanning electron microscope (FESEM) with energy dispersive spectrometer (EDS) after they were coated with gold, and then on the obtained surface SEM images, the distribution of the micropores as well as the porosity of the coating surface was analyzed statistically by Image J software.

The corrosion resistance of the coatings was evaluated by CHI660 electrochemical workstation,including a typical three electrode system with samples as working electrode,saturated calomel electrode as a reference electrode, and platinum electrodes as an auxiliary electrode.The samples with an exposed area of 1 cm2were tested at a scan rate of 1mV/s in the scanning range of ?1.9 to ?1.3V after they were immersed in 3.5wt.% NaCl corrosive medium for 30min.

It is worth noting that the zeta potential of TiC particles in different electrolytes, the thickness, the content of Ti element and the corrosion current densities of the coatings were all the average values after removing abnormal values from the original measured values.

Fig.2.Zeta potential of TiC particles in different electrolytes obtained based on orthogonal experiment.

3.Results and discussion

3.1.Zeta potential and size distribution of TiC particles in different electrolytes

Fig.2 and Fig.3 separately show the zeta potential,size distribution and average size of TiC particles in the electrolytes obtained according to the orthogonal experiment schemes listed in Table 3.As can be seen from Fig.2 and Fig.3, compared with the No.7 electrolyte, the zeta potential of TiC particles in No.3 electrolyte is lower, at the same time their average size is larger (Fig.3b(3#)) and there are two intensity peaks in the diagram of size distribution of the particles (Fig.3a(3#)), indicating that the smaller size TiC carries more negative surface charges but their agglomeration is severer.

On the basis of the addition of smaller size TiC, adding anionic surfactant SDS significantl increases the number of the negative charge on TiC surface, as can be demonstrated by Fig.2(4#), TiC particles possess the lowest zeta potential (?34.4mV).However, On this basis, adding dispersant PEG6000 make zeta potential of TiC particles shift to the most positive, but their average diameter decreases (Fig.3b(1#)) and size distribution becomes narrow(Fig.3a(1#)).This suggests that adding PEG6000 decreases the negative charge amount of TiC surface, but it contributes to the dispersion of TiC particles in the electrolyte.Moreover,When PEG6000 and SDS are both added into the electrolyte containing smaller size TiC,these TiC particles not only carry the more negative charges, but also possesses better dispersity and smaller particle size, as shown in Fig.2(2#) and Fig.3(2#).

Table 3Experiment schemes and experimental results.

Fig.3.Size distribution (a) and average diameter (b) of TiC particles in different electrolytes obtained based on orthogonal experiment.

3.2.Macroscopic morphologies of coatings

Fig.4 presents macroscopic pictures of PEO coatings formed in the electrolyte with and without additives respectively.As can be seen from Fig.4, the No.0 coating obtained in the base electrolyte is gray-white.After doping additives into the electrolyte, the color of No.1 to No.8 coatings prepared according to the orthogonal experiment all deepens to a different extent.This is probably because the TiC particles added in the electrolyte deposit into the coating and change the coating’s color.

3.3.Element and phase composition of coatings

The composition and distribution of the element in PEO coating prepared in the No.2 electrolyte with additives are shown in Fig.5.As can be seen from Fig.5, the coating contains Mg, Al, Si, O, F and Ti elements, in which Mg and Al are from the substrate, and Si, O, F and Ti originate from the electrolyte, indicating that elements resulting from the substrate and electrolyte all take part in the coatingforming reaction.In addition, Al element exhibits a localized enrichment phenomena on the surface of the coating, while amount of F element is slightly higher at the interface between the coatings and the substrate.It is worth mentioning that the Ti element distributes on the surface and cross-section of the whole coating, but there are also a few orange-red “particlelike bright spots” that are rich in Ti element.

Fig.6 illustrates the XRD patterns of the No.0 and No.2 PEO coatings, formed in the electrolyte without and with additives, respectively.It can be known from Fig.6 that the coatings are mainly composed of MgO, Mg2SiO4and MgF2,while the diffraction peaks of Mg stem from the substrate.Compared to the No.0 coating, the No.2 coatings obtained in the electrolyte with additives also contain TiC phase, which suggests that the TiC particles in the electrolyte indeed participate in the coating formation reaction and deposit into the coatings, and Ti element in the coating primarily exist in the form of TiC.

Fig.4.Macroscopic photographs of PEO coatings prepared (0#) in basic electrolyte and (1#～8#) based on orthogonal experiment schemes.

3.3.Orthogonal experimental results and analysis

3.3.1.Range analysis

Table 3 shows Ti content obtained by EPMA and corrosion current densities of the coatings prepared based on No.0 and orthogonal experiment schemes.In order to systematically study the effect orders of each additive and their interactions on Ti amount and corrosion resistance of the coatings,orthogonal experimental data are analyzed by range analysis approach, as shown in Table 4.In Table 4, k1 and k2 separately corresponds to the mean values of Ti content and corrosion current densities of the coatings at the corresponding level of each factor, and R is the difference between k1 and k2.The greater the R is, the more important the factor is.

Based on R values, the influenc sequences of each factor and their cross-actions on the content of TiC in the coatings are as follows: the size of TiC>SDS>PEG6000>cross-action between SDS and PEG6000>cross-action between TiC size and SDS>cross-action between TiC size and PEG6000.While for the corrosion resistances of the coatings,the decreasing order of the influence of three factors and their cross-actions are PEG6000, SDS, cross-action between SDS and PEG6000, TiC size, cross-action between TiC size and SDS, and cross-action between TiC size and PEG6000.Apparently, the primary and secondary order of the influenc of three factors and their cross-actions on the TiC amount and the corrosion resistance of the coatings is different, among which the influenc of TiC size on the TiC quantity is the greatest, while the factor of PEG6000 is taking the leading role on the corrosion resistance of the coatings.

3.3.2.Relationship between response parameters and factors

Fig.7 displays the relationships between response parameters and factors, which are plotted in accordance with the values of k1 and k2 in Table 4.It can be drawn from Fig.7 that,by decreasing the size of TiC particles, the coating’s TiC content and its corrosion resistance are both improved.With the addition of PEG6000, the amount of TiC in the coatings decreases, while the coatings’ anti-corrosion property enhances.The addition of SDS not just increases TiC quantity, but also improves the corrosion resistance of the coatings.These results suggest that the smaller size TiC particles are easier to enter into the coating to increase TiC content of the coating and further improve the anti-corrosion property of the coatings.Moreover, the addition of PEG6000 or/and SDS into the electrolyte containing TiC particles can assist TiC to enhance the anti-corrosion performance of the coatings.

Fig.5.Composition and distribution of element (a) on surface and (b) in cross-section of No.2 PEO coating.

Fig.6.XRD patterns of No.0 and No.2 PEO coatings.

Table 4Results of range analysis of orthogonal experiment.

Fig.7.Relationships between response parameters and each factor.

3.3.3.Variance analysis

In order to further explore the influenc degrees of TiC size, PEG6000, SDS and their cross-actions on the TiC content and the corrosion resistance of the coatings, the orthogonal experimental data are analyzed by the method of variance analysis, as shown in Table 5.In Table 5, SS is deviance quadratic sum of the factor, and df represents the corresponding degree of freedom.At the same time, MS denotes the mean square of the factor.Generally, the blank column in the orthogonal experiment table is used as the error column, and its MS corresponds to the Ms of the error.If the MS of a factor is smaller than that of the error, then it will be classifie into the error to get a new/total error.While F-value is define as the ratio of the MS of the factor to that of the total error.When F-value is higher than F0.01(df,dfe), the influenc of the factor is considered very significant remarked as “??”.And it will be significan (?) when the F-value is between F0.01(df,dfe) and F0.05(df,dfe).Otherwise, it will be non-significant

It can be seen from Table 5 that among all variables,only TiC size and SDS exhibit significan influence on the content of TiC in the coatings, while others’ influence are non-significant With respect to the corrosion resistance of the coatings, the influence of PEG6000, SDS and the crossaction between them are remarkable, and the size of TiC particles also has some influence but the others’ influence are regarded as the error.

As can be known from the aforementioned variance analysis, the impact of the cross-action between PEG6000 and SDS on the corrosion resistance of the coatings cannot be ignored.Therefore, it is necessary to investigate the level match between PEG6000 and SDS on the coatings’ anti-corrosion property, as listed in Table 6.At the same time their corresponding interaction diagram is displayed in Fig.8.It can be seen from Table 6 and Fig.8 that when B1 combines with C2, namely, PEG6000 and SDS are both added into the electrolyte with TiC particles, the corrosion current density of the coating is the lowest and its anti-corrosion capability is the strongest.

Table 5Results of variance analysis of orthogonal experiment.

Table 6 Level match of factor B and factor C for corrosion current density of coating.

Fig.8.Cross-action between PEG6000 and SDS.

It can be concluded from Table 4 or Fig.7 that when the quantity of TiC deposited in the coatings reaches the highest value, the optimum combination scheme among additives is A1B2C2, viz., 50nm of TiC, 0g/L of PEG6000 and 0.8g/L of SDS.This means that in order to obtain the coating with the highest TiC content, the electrolyte containing TiC particles should not be mixed with dispersant PEG6000.However,the results attained from Table 5 and Table 6 (Fig.8) illustrate that when PEG6000 is 10g/L and SDS is 0.8g/L, the corrosion resistance of the resultant coating is the best.As a result, in order to enhance the corrosion resistance of the coatings to the largest extent, the optimal combination among additives needs to be adjusted to A1B1C2, namely, both dispersant PEG6000 and anionic surfactant SDS should also be added when adding TiC particles into the base electrolyte.

Based on the above analysis, it can be recognized that the design of the orthogonal experiment and the level design of PEG6000 and SDS are the key to research mutual effects among TiC size, PEG6000 and SDS along with the necessity of existence of PEG6000 and SDS in the TiC particlescontaining electrolyte.Employing the orthogonal experimental design as well as the methods of the range analysis and the variance analysis,could study the influenc orders and degrees of TiC size, PEG6000 and SDS on the TiC content and the corrosion resistance of the coatings.More importantly, it could study mutual effects among these additives.And the interaction between PEG6000 and SDS has revised an exprimental result that PEG6000 should not be added into the electrolyte with the addition of TiC particles.In other words,the addition of both PEG6000 and SDS into the electrolyte containing TiC particles could most beneficiall assist TiC to improve the coating’s corrosion resistance

3.4.Morphologies of surface and cross section

The surface morphologies of No.0,No.7,No.3,No.1,No.4 and No.2 coatings and the distribution of the corresponding micro-pores are displayed in Fig.9.It can be seen from Fig.9 that after additives are added into the electrolyte, the diameter of the micro-pores on the surface of the coatings all reduces somewhat.However, the number of micro-pores increases on No.7 and No.3 coating surfaces, resulting in the increase of the coating’s surface porosity,as shown in Fig.10,No.3 coating has the highest surface porosity.Compared with other coatings, No.2 coating’s pore number is the fewest, and it exhibits the lowest surface porosity in Fig.10.

Here,No.2 coating’s surface is further magnified as shown in Fig.11(a).There are a few of particles on the surface and in several micropores of the coating.And EDS point analysis of these particles shows that they all contain relatively higher Ti content (Fig.11(b-f)), indicating that these particles may be TiC particles and they distribute randomly on the surface and in some micro-pores of the coating.Thus, it can be inferred that even though only a small number of TiC particles deposit into the micropores of the coating, the micropores will be fille by these TiC particles to some extent.This is advantageous to increase the compactness of the coating.

Fig.9.Surface morphologies and corresponding micro-pore distribution of PEO coatings obtained (0#) in basic electrolyte and (7#, 3#, 1#, 4# and 2#) based on orthogonal experimental schemes.

Fig.10.Surface porosity of PEO coatings obtained (0#) in basic electrolyte and (7#, 3#, 1#, 4# and 2#) based on orthogonal experimental schemes.

The cross-sectional morphologies of No.0, No.7, No.3,No.1, No.4 and No.2 coatings are shown in Fig.12.It can be seen from Fig.12 that the thicknesses of all coatings are almost the same, which change in the range of 19.4-21.1μm.Compared to No.0 coating, the local areas of No.7 and No.3 coatings exhibit a relatively loose microstructure, and the microstructure of the No.3 coating is looser.However, No.2 coating has the densest microstructure, the features of which has no obvious micro-pores and micro-cracks.This will prevent effectively the corrosive medium from penetrating the coating to damage the substrate.

Fig.11.Magnifie surface morphology of No.2 coating (a) and EDS point analysis of representative (b) particle 1, (c) particle 2, (d) particle 3, (e) particle 4 and (f) particle 5.

Fig.12.Cross-sectional morphologies of PEO coatings fabricated (0#) in basic electrolyte and (7#, 3#, 1#, 4# and 2#) based on orthogonal experimental schemes.

Fig.13.Magnifie cross-section morphology of No.2 coating (a) and EDS analysis of representative (b) point 1, (c) point 2, (d) point 3, (e) point 4 and (f)point 5.

The cross-section of the densest No.2 coating is further magnified as shown in Fig.13(a).A few of dark points irregularly distribute in the coating.Regarding to these dark points, EDS point analysis shows that they contain relatively higher Ti content (Fig.13 (b-e)).For comparison, bright area is also examined by EDS point scan and it is found that this representative point also contains Ti element, but its quantity is lower than that in the dark point (Fig.13(f)).In combination with EPMA (Fig.5(b)) and XRD analyses (Fig.6)),it can be deduced that the entire coating contains TiC.And most of TiC may firstl melt and then mix with other coatingforming substances, at last solidify under the rapid quenching of the cool electrolyte, forming the coating.Because TiC has high chemical stability and superior anti-corrosion property[29,30], its presence across the coating is propitious to improve the overall corrosion resistance of coatings.

Fig.14.Potentiodynamic polarization curves of No.0, No.7, No.3, No.1,No.4 and No.2 PEO coatings.

Table 7Fitted values of potentiodynamic polarization curves of PEO coatings shown in Fig.14.

3.5.Corrosion resistance of coatings

Fig.14 depicts the potentiodynamic polarization curves of No.0, No.7, No.3, No.1, No.4 and No.2 coatings.The corrosion potential (Ecorr), corrosion current density (Jcorr) and the linear polarization resistance (Rp) derived from potentiodynamic polarization curves are listed in Table 7.It can be seen from Fig.14 and Table 7 that, compared to No.0 coating,the Jcorrvalues of No.7 and No.3 coatings increase and their Rpvalues decrease, indicating the worse corrosion resistance.In contrast, the No.1, No.4 and No.2 coatings possess lower Jcorrand higher Rp, and thus they exhibit more excellent corrosion resistance.Among them, the anti-corrosion property of the No.2 coating is the best, as its Jcorris minimum, which is nearly one order of magnitude lower than No.0 coating, is approximately two orders of magnitude lower than the No.3 coating, and decreases by more than two orders of magnitude compared with No.7 coating.Meanwhile, Rpvalue of the No.2 coating is the highest, which is nearly one order higher than No.0 coating, is about 25 times of the No.3 coating, and improves by two orders in comparison with No.7 coating.This shows that, by adding TiC nanoparticles to the electrolyte to improve the corrosion resistance of the coatings,dispersant or anionic surfactant should also be added, and the corrosion resistance of the coatings will be improved more obviously when they are both added.

4.Discussion

The corrosion resistance of PEO coatings is affected by a series of microcosmic characteristic parameters such as the thickness, compactness, chemical composition and phase composition, and defects of coatings.Based on the surface and cross-section morphologies as well as the phase composition of coatings in this study, it can be known that there are no obvious defects on the surface of all coatings, and their thicknesses are similar.At the meantime, the content of highquality TiC deposited in the coatings dominates the coatings’phase composition.Consequently, the anti-corrosion performance of the coatings will primarily depend on TiC content and compactness of the coatings.

Compared to No.0 coating, the No.7 and No.3 coatings possess a lower anti-corrosion performance,which may be ascribed to their loose microstructure.Ultra-fin particles, such as submicro or nano-particles, usually have a smaller size and larger specifi surface area, and so they possess the higher surface free energy.This makes them easier to agglomerate when they are doped into the aqueous electrolyte [31].In the present study, after only adding TiC particles into the electrolyte, these TiC particles may unevenly adsorb and deposit into the coatings under the function of the strong electric field which destroys the uniformity and continuity of the coating and results in the disorderly deposition of the coating-forming material, as a result, the microstructure of the coatings becomes comparatively loose (Fig.12(7#) and (3#)).While the corrosion resistance of the No.3 coating is better than that of the No.7 coating, which is mainly due to its more highquality TiC.Although smaller size TiC are more likely to agglomerate (Fig.3(3#)) and make the compactness of the No.3 coating lower than that of the No.7 coating (Fig.12),they have a higher migration rate in the electric fiel because of their more negative surface charges (Fig.2(3#)), which enable them to deposit into the coating more easily.This can be demonstrated by the literature [32].Finally, under the coaction of low compactness and more high-quality phase, the No.3 coating shows more excellent corrosion resistance.

No.1 coating is more resistant to corrosion in comparison with No.0 coating.The reason may be its higher TiC content(Table 3) and relatively dense microstructure (Fig.12(1#)).Polyethylene glycol 6000 (PEG6000), as a nonionic polymeric dispersant, consists of a hydrophobic group and a hydrophilic group [33].When it is added to the electrolyte containing smaller size TiC, its hydrophobic group can adsorb on the surface of these smaller size TiC particles, blocking effectively their direct contact and agglomeration through its steric hindrance effect.Therefore, the cross-action between the smaller size TiC and PEG6000 improves TiC quantity in the coating and obviously avoids the deterioration of the coating structure.Eventually, No.1 coating has a relatively high corrosion resistance.

Fig.15.Schematic illustration of working mechanisms of TiC particles and its interactions with PEG6000 and SDS during PEO process.

The No.4 coating exhibits a superior anti-corrosion property, which is attributed to its highest TiC amount (Table 3).Sodium dodecyl sulfate (SDS), as an anionic surfactant, is liable to ionize in the aqueous electrolyte to generate dodecyl sulfate anion that adsorbs on the surface of TiC particles to increase the negative surface charge of them [33].On the basis of the addition of smaller size TiC, the addition of SDS can further increase negative surface charges of these smaller size TiC (Fig.2(4#)).Thus, their cross-action significantl quickens the migration of TiC particles in the electric fiel and increases the high-quality TiC content in the coating, which makes the No.4 coating have a higher corrosion resistance.

No.2 coating possesses the best performance in anticorrosion.As illustrated above,smaller size TiC have a higher migration rate and are much easier to deposit into the coating.And when PEG6000 and SDS are added into the electrolyte containing smaller size TiC, the cross-action between them not only effectively hinders the agglomeration of these smaller size TiC particles, but also increases their negative surface charge.This greatly augments the migration rate of these smaller size TiC and the content of high-quality TiC phase in the coating.At the same time,their interaction makes TiC particles uniformly deposit into the coating and increases their probability of entering and then depositing in the discharge channel, improving the compactness of the coatings,as shown in Fig.9(2#) and Fig.12(2#).As a result, under the coaction of higher TiC content and the densest microstructure, the optimum corrosion resistance coatings could be obtained in the electrolyte with the addition of smaller size TiC,PEG6000 and SDS.

In conclusion, the possible working mechanisms of different sized TiC and its interactions with PEG6000 and SDS are shown in Fig.15.

4.Conclusions

(1) The experiment design of this study is the key to study the mutual effects among TiC size, dispersant PEG6000 and anionic surfactant SDS.And the results show that each factor and their interactions all remarkably affect the coating.

(2) The influence of individual factor of TiC size,PEG6000 and SDS on the corrosion resistance of the coating are different, in which the influenc of TiC size is the smallest, while PEG6000 exhibits the greatest impact, followed by the SDS.By decreasing the size of TiC particles, the coating’s anti-corrosion property is improved.Moreover, addition of PEG6000 and SDS into the electrolyte containing TiC particles enhances the corrosion resistance of the coatings.In short, the design of two obviously different sizes of TiC particles as well as the level design of PEG6000 and SDS provides the guiding direction for their further application.

(3) The interaction between PEG6000 and SDS both effectively prevents the agglomeration of TiC particles and increases their negative surface charges, which not only increase the migration rate and deposited uniformity of TiC, but also enable TiC to deposit into the discharge channel more easily.Thus, under the optimum combination of A1B1C2, namely when smaller size TiC combines with PEG6000 and SDS, the prepared coating possesses high TiC content, the densest microstructure and the most excellent corrosion resistance.

Acknowledgments

The authors are grateful to the Department of Science &Technology of Gansu Province (China) for the support of the Creative Research Group Fund Grant (1111RJDA011), and the Open Fund (SKLAB02015006) from State Key Laboratory of Advanced Processing and Recycling of Non-ferrous Metals (China).