To Li,Zhongjun Leng,Shifng Wng,Xito Wng,Rez Ghomshchi,c,Yunsheng Yng,b,Jixue Zhou,*

a Shandong Provincial Key Laboratory of High Strength Lightweight Metallic Materials,Advanced Materials Institute,Qilu University of Technology (Shandong Academy of Sciences),Jinan 250014,China

b Institute of Metal Research,Chinese Academy of Sciences,Shenyang 110016,China

cSchool of Mechanical Engineering,The University of Adelaide,Adelaide SA 5005,Australia

Abstract In this study,Mg-6.0Zn-3.0Sn-0.5Mn(ZTM630)magnesium alloy was pre-activated by colloidal Ti,oxalic acid,and phosphoric acid,and a phosphate conversion coating(PCC)was prepared on the alloy surface.The morphology and corrosion resistance of the prepared PCCs were investigated.Surface morphology studies showed that the phosphate crystals that formed the coating were the smallest for the sample pre-activated by phosphoric acid.The coating on the colloidal Ti and the phosphoric acid samples had the largest and the smallest thickness and surface roughness,respectively.The reason for the discrepancy was analyzed by comparing the surface morphologies of alloy samples immediately after the pre-activation treatment and various phosphating treatments.X-ray diffraction analysis revealed that all three PCCs contained the same compounds.The corrosion resistance time from the copper sulfate drop test and the electrochemical data from the potentiodynamic polarization curves showed that the coating pre-activated by phosphoric acid had the best corrosion resistance.Finally,the 1500h neutral salt spray corrosion test confirme that the phosphating treated magnesium alloy,which was pre-activated by phosphoric acid,exhibited excellent corrosion resistance and behavior.? 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:Phosphate conversion coating;Surface pre-activator;Magnesium alloy;Morphology;Corrosion.

1.Introduction

Magnesium alloys are considered the most promising lightweight metal materials in the 21st century due to their low density,high specifi strength,and unique damping and electromagnetic properties[1-3].However,a major problem facing the application of magnesium alloys is their poor corrosion resistance[4,5],which is most effectively improved by surface coating[6-8].Phosphate conversion coating(PCC)has been widely adopted because of its low cost and easy operation,especially in engineering applications[9-12].

Various types of PCCs with different structures prepared on the surface of magnesium alloys in the past decade to enhance anti-corrosion properties include the Zn series[13,14],Mn series[15,16],Ca series[17,18],Mg series[19,20],Sr series[21,22],and their combined series[23,24].Zhou et al.[25]obtained a PCC on a die-cast AZ91D Mg alloy by immersing in a bath based on manganese dihydro phosphate.This PCC was identifie as having an amorphous structural characteristic.Su et al.[26]prepared a Ca-series PCC on an AZ60 Mg alloy with a coating made of fla e crystals.By controlling the pH and temperature of the phosphating solution,the size of the fla es could be adjusted,and the performance of the coating could be affected.Further,PCCs with fl wer-like and equiaxed crystal structures have been widely reported[13,19].In addition to traditional engineering applications,PCC has been used in functional materials in recent years.For example,PCC on the surface of a biomedical Mg alloy improves the corrosion resistance and biocompatibility of the alloy[27-29].Furthermore,Cui et al.[30]obtained a calcium phosphate coating with corrosion resistance and antibacterial dual characteristics by tin dioxide doping.In short,the phosphating treatment of magnesium alloys has become an important treatment method for performance modification However,regardless of the application field optimization of PCC morphology remains the decisive factor for improving the coating performance.

In response to environmental pressures,we developed a green phosphating technology for magnesium alloys free of fluorine chromium,and nitrite[31].After undergoing surface pre-activation,the phosphate grain in the PCC was greatly refined and the growth orientation of the phosphate grains changed,improving the corrosion resistance of the coating.Thus,the alloy surface should be subjected to preactivation before phosphating to accelerate PCC formation and to improve its quality.However,the current selection of pre-activators has not been systematically studied for the aforementioned green phosphating technology.

The pre-activator is crucial for obtaining a better preactivation effect.Colloidal titanium phosphate(colloidal Ti)is a common pre-activator in the fiel of iron phosphating that works primarily by adsorption on the alloy surface as a nucleation point[32].Wolpers et al.[32]stated that the coagulation and sedimentation of colloidal Ti on the alloy surface could affect the reaction kinetics of phosphating and,ultimately,the density and size of crystals.However,very few studies report the application of colloidal Ti to magnesium alloy phosphating.Moon et al.[33]used colloidal Ti to pretreat a magnesium alloy when preparing a zinc phosphate coating,but the treatment effect was not discussed.Additionally,acid is a commonly used pre-activator in the preparation of chemical conversion coatings[11].Jiang et al.[34]compared the effect of alkali/acid pretreatment on the morphology and properties of as-deposited CaP coatings on magnesium alloys and revealed that alkali and acid have different influence on the PCC phase component,morphology,thickness,and corrosion resistance.The coating pretreated with HF acid exhibited the best corrosion resistance.HF acid is widely used to pick the alloy surface owing to its strong erosion ability.However,flu oride ions have a strong destructive effect on the environment and are heavily restricted by current emission standards.Thus,a goal of our green phosphating technology was to eliminate fluorin from the process.Actually,both organic and inorganic acids react with the magnesium alloy surface,indicating that these acids can both acts as pre-activators to form nucleation sites or change the surface characteristics,thereby accelerating the phosphating process.Oxalic acid,a typical small molecular organic acid,is usually a degradation product of macromolecular organic matter.Therefore,from the perspective of wastewater treatment,the use of small molecular oxalic acid is more environmentally friendly than that of large molecular organic acids.Phosphate reagents are essential when preparing the PCC,and the use of phosphoric acid to pre-activate the sample would minimize additional environmental pollution from the pre-activation process.However,to date,no studies have compared colloidal titanium,oxalic acid,and phosphoric acid for the pre-activation treatment of magnesium alloys.Thus,the effectiveness of these pre-activators needs to be determined.

To improve the performance of our newly developed green PCC,three pre-activators,namely colloidal Ti,oxalic acid,and phosphoric acid,were selected to pre-activate and subsequently phosphatize ZTM630 magnesium alloy.By analyzing the morphology and corrosion resistance of the PCCs,a pre-activator suitable for the novel green phosphating of magnesium alloy was determined.A long-term 1500h salt spray corrosion test was conducted to detect the real corrosion behavior of the coated alloy.

2.Experimental

2.1.Materials and preparation of PCCs

Mg-6.0Zn-3.0Sn-0.5Mn(ZTM630)extruded magnesium alloy plates with a size of 20mm×20mm×3mm were f rst washed in a 60%NaOH solution and ground using#2000 SiC paper.Then,the samples were rinsed with clean water,ultrasonically cleaned with absolute ethanol for 5min,and dried with hot air for subsequent use.The pre-activation solutions prepared were a 0.2wt.%colloidal Ti solution,a 5wt.%oxalic acid solution,and a 10vol.% phosphoric acid solution.The cleaned alloy plates were dipped in the pre-activation solutions for pre-activation treatment at room temperature for 20s.The specimens were then subjected to phosphating for 20min;the temperature of the phosphating solution was maintained at 90°C using an electric-heated thermostatic water bath(Taisite DK-98-II,China).The phosphating solution contained 40g/L of Mn(H2PO4)2·2H2O,0.5g/L of sodium citrate,and 0.2g/L of hexamethylenetetramine.The pH of the phosphating solution was adjusted to 2.3 using a 10vol.% phosphoric acid solution and a 0.1mol/L sodium hydroxide solution.Finally,the specimens were cleaned and dried for characterization.Colloidal Ti(TZD-1019)was purchased from Shandong Tonghu Trading Co.Ltd.,China.All other chemicals used in this study were of reagent grade from the China National Pharmaceutical Group Corporation.All solutions mentioned in this study refer to aqueous solutions.

2.2.Morphology characterization

The surface morphology of the prepared PCCs was examined using a scanning electron microscope(SEM,Zeiss Evoma10,Germany)equipped with an energy dispersive spectrometer(EDS,Oxford X-Max50,England).The phase structure was detected using an X-ray diffractometer(XRD,PANalytical Empyrean,Netherlands)with a scan speed of 8°/min.The PCC thickness was tested using a coating thickness gauge(Shidaichuanghe CTG260,China)with an N1 model probe,and the surface roughnessRa was tested using a surface roughness tester(Shidaichuanghe TR221,China).Three detections were performed in parallel for each sample for thickness and surface roughness tests.

2.3.Corrosion behavior

The corrosion resistance time of the PCC was tested using the drop method.The drop solution was composed of 41g/L CuSO4·5H2O,35g/L NaCl,and 13mL/L of 0.1N HCl solution.The sample was flattened and a circle of approximately 1cm diameter was drawn on the surface with a crayon to prevent the solution from fl wing out;0.1mL of the solution was dropped into the circle,and timing began immediately.Timing was stopped when a reddish sign appeared in the droplet and recorded the holding time,which was set as the corrosion resistance time of the PCC.For the corrosion resistance time tests,three detections were performed in parallel for each sample.

The potentiodynamic polarization(PDP)curves were measured for the bare and coated samples in 3.5wt.% NaCl solution using an electrochemical workstation(Shanghai CH Instrument CHI660E,China).A three-electrode electrochemical cell was used.A saturated calomel electrode(SCE),a platinum electrode,and a sample were set as the reference,counter,and working electrodes,respectively.First,the electrode was stabilized in the solution for 30min.Next,the PDP curve was immediately measured at a scan rate of 1mV/s.For the PDP measurement,polarization started from a cathodic potential of-1.7V and stopped at an anodic potential,at which the anodic current increased dramatically.The electrochemical data were derived from the PDP curves by Tafel extrapolation.

Fig.1.Surface morphologies of pre-activated samples using different pre-activators:(a,b)colloidal Ti;(c,d)oxalic acid;(e,f)phosphoric acid.

Fig.2.Surface morphologies of three differently pre-activated alloy samples at 10s,20s,and 100s after immersion in the phosphating solution.

The neutral salt spray corrosion test was conducted in a salt spray test chamber(Huanke YWX-250,China)according to the standard ISO 9227:2006.The salt solution was a 50g/L sodium chloride solution.The temperature in the test chamber was set to 35 °C.The average collection rate for a horizontal collecting area of 80 cm2was 1.5mL/h.Samples with a size of 110mm×40mm×3mm were used.For the phosphating coated sample,PCC was prepared as described above.During the test,the test surface of the sample faced upwards and formed an angle of 20° in the vertical direction.The test adopted a continuous spray mode for a duration of 1500h.Photographs were taken at 24h intervals during the firs two weeks and at 48h intervals until the end of the test.After the 1500h salt spray corrosion test,the samples were dipped in a chromic acid solution,which consisted of 200g/L CrO3and 10g/L AgNO3,to remove the corrosion products[35].The samples were weighed before and after the salt spray corrosion test,and the weight losses were calculated.

3.Results and discussion

3.1.Morphology

To explain the surface morphology differences of the fina PCCs,the initial morphologies of the sample surface immediately after the pre-activation treatment were observed.The results are shown in Fig.1.Scratches caused by grinding remained on the colloidal Ti sample after the pre-activation treatment.Further,Ti-containing particles were adsorbed on the sample surface.All the original grinding scratches on the oxalic acid sample disappeared,and the sample surface was slightly corroded.The overall surface presented two forms:fla and pitted areas.Inside the fla area,the surface was smooth.However,inside the pitted area,many corrosion pits appeared(shown in Fig.1d),the size of which reached approximately 10μm.Fig.1e shows that the corrosion of the sample surface was more serious after pre-activation by the phosphoric acid solution;the grain boundaries were completely corroded,and the inside of the grain was also uniformly corroded.Fig.1f shows the corrosion morphology inside the grain.Tiny corrosion pits were densely distributed across the surface and measured approximately 0.5μm in size—much smaller than those on the oxalic acid sample.

The initial nucleation state of the phosphate partially determines the fina morphology of the coating.Therefore,the evolution of the alloy surface morphology during the initial nucleation stage of the coating was characterized.Fig.2 shows the surface morphologies of three pre-activated alloy samples at 10s,20s,and 100s after immersion in the phosphating solution.After phosphating for 10s,no phosphate nuclei were found on the colloidal Ti sample.While abundant phosphate nuclei were found on the oxalic acid and phosphoric acid samples.The differences were as follows:(i)On the oxalic acid sample,the phosphate nuclei gathered in local areas;however,the phosphate nuclei were evenly dispersed on the surface of the phosphoric acid sample.(ii)The number of phosphate nuclei on the phosphoric acid sample was higher.After phosphating for 20s,crystal nuclei were also found on the colloidal Ti sample.Interestingly,although the number of crystal nuclei on the oxalic acid sample was larger than that at 10s,it still presented a gathering distribution.At this time,the number of phosphate nuclei on the phosphoric acid sample was the largest.When the phosphating process reached 100s,the oxalic acid and phosphoric acid samples were covered by dense phosphate crystals.However,the surface of the colloidal Ti sample contained only scattered,large phosphate particles.

Fig.3.Surface morphologies of the fina PCCs prepared at 90 °C for 20min treated with different pre-activators:(a,d)colloidal Ti;(b,e)oxalic acid;(c,f)phosphoric acid.

Fig.3 presents the surface morphologies of the fina prepared PCCs.A dense PCC without any pores or cracks eventually formed on the alloy surfaces with all three preactivators,but the coatings featured morphological differences.The coating of the colloidal Ti sample was spread over the surface of the alloy and exhibited a number of phosphate clusters.Each cluster was stacked with a series of phosphate crystal platelets,and a bulge formed at the center of the cluster due to excessive phosphate growth.The prepared PCCs on the oxalic acid and phosphoric acid samples were closely packed together with many equiaxed polyhedral crystals.Comparatively,the phosphoric acid sample had a smaller crystal size than the other samples.

According to previous studies[9,31,36],the phosphating process includes the peeling-off of the original oxide fil on the alloy,followed by electrochemical corrosion,nucleation,growth,and densification When pre-activated with oxalic acid and phosphoric acid solutions,the original oxide fil is completely peeled off the alloy by the strongly acidic environment during the pre-activation process.Thus,the peeling-off stage can be omitted after immersion into the phosphating solution,and subsequent stages can be performed directly to promote the rapid occurrence of phosphating nucleation.Further,preactivation changed the surface morphology of the alloy.The formation of corrosion pits increased the specifi surface area,which was directly related to the chemical reactivity of the surface,increasing the rate of electrochemical corrosion in the early stage of phosphating.Consequently,the consumption of hydrogen ions was accelerated,thereby promoting the ionization of(H2PO4)-to(HPO4)2-and(PO4)3-.The latter two ions could then react with manganese ions to form phosphate nuclei.This is also the cause of the gathering distribution of phosphate nuclei on the oxalic acid sample after phosphating for 10 and 20s,which was related to the two different surface morphological structures,namely fla area and pitted area,on the oxalic acid sample immediately after the pre-activation treatment.Liu et al.[37]once performed laser shock peening(LSP)pretreatment before preparing the PCC which also confirme that the increase in surface activity of AZ31 Mg alloy caused by LSP pretreatment was responsible for facilitating the deposition of phosphide.In summary,the combination of the omitted peeling-off stage and the increased specifi surface area shortened the nucleation time,increased the nuclei quantity,and reduced the size of the nuclei,which led to the formation of a dense PCC with the smallest phosphate crystal size on the phosphoric acid sample.The two effects of phosphoric acid pre-activation on phosphating nucleation are schematically compared in Fig.4.

Fig.4.Diagrams showing the effects of phosphoric acid pre-activation on phosphating nucleation:(a,b,c)sample without pre-activation;(a’,b’,c’)sample pre-activated with phosphoric acid.

The thickness and surface roughness of the prepared PCCs are presented in Fig.5.The colloidal Ti sample had the thickest coating,followed by the oxalic acid sample;the phosphoric acid sample had the thinnest coating—only half that of the colloidal Ti sample.This means that during the same phosphating time,half of the film-formin agent was consumed to create a fine-graine coating layer on the phosphoric acid sample.The surface roughness of the prepared PCCs followed the same pattern as that of the thickness and was consistent with the surface morphology observations shown in Fig.3.On the one hand,low surface roughness is generally beneficia for improving the wear resistance of coatings.On the other hand,the phosphating coating is usually only used as an intermediate layer.In practical applications,a fina spray paint treatment is usually performed on the outside of the phosphating coating.For subsequent painting,the relatively low surface roughness of the phosphating coating is benefi cial to improve the evenness and the distinctness of reflecte image(DOI)of the painting film

Table 1Electrochemical data of PDP curves.

Fig.6 shows the XRD patterns of the prepared PCCs.After comparing these results with standard powder diffraction fil cards,it can be concluded that all the three prepared PCCs consisted of the same compounds:Mn5(PO4)2(PO3OH)2·4H2O and MnHPO4·2.25H2O.Thus,after treatment with three different pre-activators,the growth rate and morphology of the coatings changed,but the formation mechanism of the PCCs was unaffected.

3.2.Corrosion resistance

Fig.7 presents the corrosion resistance times of the prepared PCCs.The phosphoric acid sample had the longest corrosion resistance time.The corrosion resistance time showed a trend opposite that of the thickness.The PCC on the phosphoric acid sample had only half the thickness of that on the colloidal Ti sample,but twice the corrosion resistance time.

Fig.8 shows the PDP curves of the prepared PCCs.The electrochemical data derived from the PDP curves are presented in Table 1.The results showed that the prepared PCCs significantl improved the corrosion resistance of the base alloy.The corrosion potential(Ecorr)of the samples from least to greatest was:the bare sample,the colloidal Ti sample,the oxalic acid sample,and the phosphoric acid sample.Thus,the phosphoric acid sample was the least susceptible to corrosion[38].Further,all three phosphating samples exhibited a pitting potential(Epit)in the anode region.For a coated sample,this characteristic potential corresponds to the potential for the breakdown of the coating film which usually occurs in the form of pitting corrosion.However,the phosphoric acid sample had the largestEpit.Moreover,according to Table 1,the phosphoric acid sample exhibited the largest linear polarization resistance(Rp)and the smallest corrosion current(icorr).This could be attributed to two reasons.First,as shown in Fig.3,the phosphoric acid sample had the smallest phosphate crystal size and the densest PCC,which could effectively prevent the penetration of corrosive media.Second,the phosphoric acid sample had the smallest surface roughness,which could reduce the contact and reaction of the sample surface with the corrosive media.In summary,all these results confirme that the coating on the phosphoric acid sample had the best corrosion resistance.

Fig.5.Thickness(a)and surface roughness(b)of the PCCs prepared at 90 °C for 20min.

Fig.6.XRD pattern of the PCCs prepared at 90 °C for 20min.

Fig.8.PDP curves of PCCs prepared at 90 °C for 20min.

Fig.7.Corrosion resistance time of PCCs prepared at 90 °C for 20min.

Fig.9.Surface morphologies of the samples at various times during the 1500h salt spray corrosion test.The upper row represents the bare sample without coating.The lower row represents the phosphating treated sample pre-activated by phosphoric acid.

Fig.10.(a)Corroded surface morphologies of the samples with corrosion products removed after a 1500h salt spray corrosion test;(b)weight loss results.

The copper sulfate drop test and electrochemical measurement revealed that the coating had the best corrosion resistance after pre-activated by phosphoric acid,primarily because of the fine phosphate crystal size of the coating and the more densely packed crystals(Fig.3).Consequently,invasion by the corrosive medium was more difficult Moreover,the consumption of the phosphating film-formin agent to form a PCC was minimal after pre-activation with phosphoric acid,meanwhile still ensuring the strong corrosion resistance of the PCC.Thus,pre-activation with phosphoric acid made the coating more economical and smoother,which improved the wear resistance and the quality of the subsequent painting.The pre-activation treatment with phosphoric acid generally had the most positive effect on the morphology and corrosion resistance of the PCC on the magnesium alloy compared to the use of colloidal Ti and oxalic acid.

To study the long-term corrosion behavior of the phosphating treated sample pre-activated by phosphoric acid,a neutral salt spray corrosion test was conducted for a duration of up to 1500h.Fig.9 shows the surface morphologies of the samples at various times during the test.Signs of accelerated corrosion appeared on the bare sample after 48h.The phosphating treated sample showed signs of accelerated corrosion at 500h.After the 1500h salt spray corrosion test,the surface of the bare sample was completely corroded,and numerous corrosion products had accumulated on the surface.However,the corroded area on the surface of the phosphating treated sample was approximately 50%.The remaining area was unetched and still covered by phosphate film

Fig.10 presents the corroded surface morphologies of the bare and phosphating treated samples with corrosion products removed after 1500h of salt spray corrosion test and the results of their weight loss.Both samples exhibited obvious filifor corrosion along the extrusion direction,due primarily to the presence of incomplete recrystallization zones along the extrusion direction in extruded samples,which were usually formed by incomplete recrystallization during the extrusion process[39].The dislocation energy inside these zones was not completely released,resulting in the high corrosion sensitivity of these structures[40].Comparatively,the corrosion area of the phosphating treated sample was significantl lower than that of the bare sample.Moreover,the phosphating treated sample showed a relatively weaker corrosion degree in the corrosion area in comparison with the bare sample.Additionally,in the large area protected by the PCC,the substrate was not corroded.Overall,the weight loss rate of the phosphating treated sample was 24% that of the bare sample.In summary,the results of the long-term salt spray corrosion test proved that the phosphating treated magnesium alloy pre-activated by phosphoric acid exhibited excellent corrosion resistance and behavior.

4.Conclusions

After comparing the effects of three pre-activators(colloidal Ti,oxalic acid,and phosphoric acid)on the morphology and corrosion resistance of a phosphate conversion coating,the following conclusions were drawn:

(1)During pre-activation,the pre-activators peeled off the original oxide fil and increased the specifi surface area to different degrees,which shortened the nucleation time,increased the nucleation quantity,and reduced the nuclei size.This effect was the strongest for phosphoric acid,followed by oxalic acid,and absent for colloidal Ti.

(2)After pre-activation with phosphoric acid,the phosphate crystals of the coating were the fines and most densely packed.The thickness and surface roughness of the prepared PCCs were greatest for the colloidal Ti samples,followed by the oxalic acid samples,and lastly the phosphoric acid samples.

(3)All three prepared PCCs consisted of the same phosphate compounds.Using different pre-activators did not alter the formation mechanism of the PCC,but signifi cantly changed the growth rate and morphology of the PCC.

(4)The corrosion resistance exhibited the opposite trend of the thickness and surface roughness.Of the three pre-activators,phosphoric acid was the best choice for preparing a PCC for a magnesium alloy.Phosphoric acid pre-activation allowed the PCC to consume the least amount of film-formin agent but obtained the best corrosion resistance.

Acknowledgments

This project was financiall supported by National Key Research and Development Program of China(Nos.2017YFB0103904,2016YFB0301105),National Natural Science Foundation of China(No.51804190),Youth Science Funds of Shandong Academy of Sciences(No.2020QN0022),Youth Innovation and Technology Support Program of Shandong Provincial Colleges and Universities(No.2020KJA002),and Jinan Science & Technology Bureau(No.2019GXRC030).