In vitro corrosion properties of HTHEed Mg-Zn-Y-Nd alloy microtubes for stent applications: Influence of second phase particles and crystal orientation

Peihu Du, Di Mei, Tsuyoshi Furushim, Shijie Zhu, Liguo Wng, Yifn Zhou,Shokng Gun

a School of Materials Science and Engineering, Zhengzhou University, Zhengzhou 450001, China

b Henan Key Laboratory of Advanced magnesium Alloys, Zhengzhou 450002, China

cKey Laboratory of Advanced Materials Processing and Mold Ministry of Education, Zhengzhou 450002, China

d Department of Mechanical and Biofunctional Systems, Institute of Industrial Science, The University of Tokyo, 4-6-1, Komaba, Meguro, Tokyo 153-8505,Japan

e Magnesium Innovation Centre - MagIC, Institute of Materials Research, Helmholtz-Zentrum Geesthacht, Geesthacht 21502, Germany

Abstract

Keywords: Mg-Zn-Y-Nd alloy; Microtubes; Heat treatment; Hot extrusion; Corrosion properties; Cardiovascular stent.

1.Introduction

Mg-base biomedical implants, including vascular stents and bone-fraction implants, have attracted worldwide attention since there is no requirement for a secondary removal surgery [1,2].Mg alloy vascular stents can be completely absorbed into the human body and pose no long-term risks of in–stent restenosis and thrombosis [3,4].However, owning to the complex service environment and the insufficient intrinsic corrosion resistance of Mg alloys, the structure and mechanical integrity of Mg alloy stents cannot be easily maintained during the service period (3 to 6 months) [2,5–7].The complexin vivoenvironment is characterized by dynamic chemical and physiological processes [8,9]; and mechanical and thermal loadings [2,5,10]; all of which have a significant impact on the corrosion behavior of stents [11–13].

Considering the complicatedin vivoenvironment,some encouraging results have been obtained to acquire a better corrosion performance from Mg alloys, including micro-alloying[14–16], microstructural control [17–19], and surface modification [20,21].Different from coating, which gives an extra protective layer on the surface of Mg alloys [21,22], microalloying and microstructural control are aimed to improve the anodic stability of Mg alloys so to reduce the dissolution of Mg alloys caused by galvanic corrosion (from the viewpoint of electrochemistry) [23–27].However, owing to the relatively low solubility of alloying elements in Mg, microalloying can only slightly improve the corrosion resistance[28,29].Meanwhile, micro-alloying leads to the subsequent formation of deleterious second phases, causing an accelerated localized corrosion [14,30,31].It should be pointed out that the localized corrosion will not be initiated when the size of second phase particles are smaller than a critical value [27,28,32].That is, by controlling the size of second phase particles (SPPs), the localized corrosion of Mg alloys can be alleviated.Grain size is the second factor that can change the corrosion rate of Mg alloys [17,25].Normally, the corrosion rate of Mg alloys decreases with the decrease of grain size.Grain orientation also affects the corrosion resistance of Mg alloys.In Song’s work, grains oriented to basal plane were stable enough to resist corrosion,compared with grains with non-basal crystal planes [24,33].Thus, to achieve a high corrosion resistance of Mg alloy microtubes,specific attention should be paid to the size of SPPs,grain size and orientation.In the work of Zhu et al.[17], severe plastic deformation,such as cyclic extrusion compression and equal channel angular pressing, was applied to acquire an improved corrosion resistance of Mg alloys by refining grains and SPPs; Ge et al.[34]proposed to acquire finegrained microtubes by low-temperature hot extrusion.However, Mg alloy microtubes at present are usually manufactured by cold drawing due to the requirements on dimension accuracy [3,35,36], the control of microstructure to acquire an improved corrosion resistance seems not to be taken into consideration.

For the purpose of achieving a high corrosion resistance of Mg alloy microtubes, a HTHE process is designed in this study.The HTHE process is featured by a long-time and high-temperature heat treatment, and a large-reduction-ratio hot extrusion.The long-time and high-temperature heat treatment was to dissolve alloying elements and reduce the size of SPPs, and the large-reduction-ratio hot extrusion was to acquire fine-grained and strongly textured microtubes.In this study, the size and distribution of SPPs, grain size and orientation of HTHEed microtubes were observed by optical microscopy (OM), scanning electron microscopy (SEM) and electron back-scattered diffraction (EBSD).Corrosion performance of HTHEed microtubes was evaluated ed byin vitrocorrosion testing, since corrosion testing is an easy approach to assess the corrosion resistance of Mg alloys even some differences exist betweenin vivoandin vitroperformance [37].Finally, the relationship between microstructure and corrosion properties was deduced.

2.Materials and methods

2.1.Fabrication of HTHEed microtubes

In this study, the as-cast Mg-Zn-Y-Nd alloy, produced by a semi-continuous casting process, was used as raw material(chemical composition is shown in Table 1).The microstructure of as-cast Mg-Zn-Y-Nd alloy is shown in Fig.1a.It can be observed that the second phases are distributed inside grains and along boundaries.The chemical composition of second phases in the as-cast condition is dominated by Mg,Nd and Zn (as shown by the EDS result in Fig.1(b)).The HTHE process was used to fabricate microtubes.The HTHE process is characterized by a long-time and high-temperature heat treatment (450°C for 10-days) followed by hot extrusion (400°C; extrusion ratio of 360:1), as shown in Fig.2.Fig.3 shows photographs of HTHEed microtubes (length,800mm;outer diameter,2.8mm;wall thickness,0.12mm)and the cross section.

Table 1Chemical composition of as-cast Mg-Zn-Y-Nd alloy.

2.2.Microstructural characterization

Specimens for microstructural characterization were coldmounted along the longitudinal direction.Specimens were ground (#100, #200, #400, #600, #800, and #1000 SiC papers) and polished with Al2O3 suspension (0.1μm).For OM observation, specimens were etched with picric acid solution(picric acid, 4.2g; glacial acetic acid, 10mL; ethanol, 70ml;water, 20mL).SEM (FEI Quanta-200) with an acceleration voltage of 20kV was carried out to observe the distribution of second phases.The chemical composition of second phases is determined by energy dispersive spectroscopy (EDS).The shape and size of SPPs of HTHEed microtubes were determined by TEM (JEOL JEM-2100).The crystal orientation of HTHEed microtubes was identified by EBSD (FIB, Zeiss Auriga).The specimens prepared for EBSD were finished by chemical polishing (room temperature, 90 s, solution composed of 180mL of ethanol, 50mL of glycerin, and 80mL of phosphoric acid).The directions of HTHEed microtube are defined as thickness direction (TD), circumferential direction(CD), and extrusion direction (ED).

2.3. In vitro corrosion testing

In vitrocorrosion testing was conducted in simulated body fluid (SBF, the composition is shown in Table 2) maintained in a water bath (37°C).According to the American Society for Testing and Materials standard (G31–72), the ratio of solution volume to specimen surface area is set at 30mL/cm2.In this study, in order to evaluate the influence of texture on thecorrosion properties, the LS and CS of HTHEed microtubes were tested separately.

Table 2Ion concentration of SBF.

Table 3Fitted electrochemical parameter for the polarization curves.

Fig.1.(a) Microstructure of as-cast Mg-Zn-Y-Nd alloy (SEM, with an acceleration voltage of 20kV and a working distance of 14.1mm); (b) and (c) EDS result showing the chemical composition of second phases.

Fig.2.Schematic illustration of HTHE process.

Fig.3.(a) As-cast Mg-Zn-Y-Nd alloy ingot to billet to HTHEed microtubes;(b) photo of the cross section of the HTHEed microtube; (c) photo of the longitudinal section of HTHEed microtube.

Fig.4.(a) Cold-mounted microtubes to exposure CS for conducting hydrogen evolution tests;(b)schematic illustration showing the hydrogen evolution tests on CS.

To observe the corrosion morphology, the corrosion products were removed in a chromic acid solution (Cr2O3,200g/L)in an ultrasonic cleaner for 3 min after the immersion test.The corrosion rate was evaluated by the hydrogen evolution test [38].The hydrogen evolution tests lasted for 120 h and the testing solution was renewed every 24 h.Because of the small area of CS, three microtubes are cold mounted to exposure a total of six CSs to SBF (one sample), as shown Fig.4.Three parallel samples were tested to obtain an average value, and the hydrogen evolution rateΔV(mL/cm2/h)was calculated according to Eq.(1):

where,Vtiis the total volume of hydrogen evolution collected at timeti,Sis the sample area exposed to immersion solution.The potentiodynamic polarisation curves were measured by an electrochemical workstation (RST5200) at a scan rate of 0.5mV/s.

3.Results

3.1.Microstructure of HTHEed microtubes

Fig.5(a) shows the OM of HTHEed microtubes, the grain size is not uniform, with coarse grains of around 20μm and fine grains of around 10μm.Twins cannot be observed inside grains.Because hot extrusion is conducted at 400°C with a high extrusion ratio (360:1), a full dynamic recrystallization can be expected.Inside grains, SPPs are distributed uniformly(as shown in Fig.5(a) and 5(b)), but the size and shape of SPFs is not uniform, with a range from 200 to 2000nm (as shown in Fig.5(c),5(d)and 5(e)).SPPs are mainly composed by Mg and Y, as shown in Fig.5(f) and 5(g).Similar to other works, HTHEed microtubes are strongly textured after hot extrusion, with the basal plain oriented parallel to the LS(as shown in Fig.6).

3.2.Corrosion performance

Fig.7 shows the macro-morphologies of the immersed HTHEed microtubes, and the corresponding corrosion morphology of the LS and CS.From Fig.7(a), 7(d), 7(g), and 7(j), it can be observed that HTHEed microtubes maintain the integrity of structure after 40 hours’ immersion in SBF.However, the corrosion morphologies after removing the corrosion products show that a huge difference between the LS and CS exists.Some part of the LS is initially attacked by localized corrosion while the other part retains the original surface, as shown in Fig.7(b).With the increase of immersion time, the corrosion morphology keeps unchanged because both localized and uniform corrosion areas can be observed, as shown in Fig.7(e), 7(h), and 7(k).For the CS, localized corrosion can be observed on the whole surface, as shown in Fig.7(c),7(f), 7(i), and 7(l).Compared the corrosion morphologies of the LS and CS, it can be concluded that localized corrosion on the LS has been alleviated to some degree.

For Mg alloys, the existing of localized corrosion leads to an accelerated corrosion rate [28,6].Based on the different corrosion behaviors of the LS and CS, the corrosion rates of the LS and CS were further tested by hydrogen evolution tests, and the results are shown in Fig.8.For the LS,the hydrogen evolution rate remains constant throughout the whole process, around 0.1mL/cm2/h.In comparison, the hydrogen evolution rate of the CS is around 1.1mL/cm2/h during the first 4 h.With increase of time, the hydrogen evolution rate of the CS decreases rapidly, to 0.1mL/cm2/h after 80 h.Fig.8 also shows that the hydrogen evolution rate of as-cast sample is similar to the CS, which means the corrosion resistance of the CS is not increased.The decrease of the hydrogen evolution rate is suspected to be related to the generation of corrosion product layer.A number of published works have shown the reaction between Mg alloys and SBF leads to the generation of a compact corrosion layer (phosphate and carbonate product layer), which protects Mg alloys from further corrosion [39–41].

4.Discussion

4.1.The initiation of localized corrosion

From Figs.7 to 8, it can observe that the localized corrosion is alleviated, and the corrosion rate maintains constant for the LS.The initiation of localized corrosion seems to be the key factor in determining the corrosion performance[14].In the work of Xu et al.[28], the initiation of corrosion is treated as a process of electrochemical reaction of Mg alloys in corrosive solution, and thus three factors (the alloying elements, the damage of compact corrosion product layer, and the existing of SPPs) are considered responsible for initiating corrosion.The existence of SPPs leading to localization of electrochemical reactions on Mg alloy surface,is the main factor initiating localized corrosion [6].Element addition, including Al and Zn, can slightly improve the corrosion resistance by retarding anodic reaction kinetic [29,38].Meanwhile, the existence of chloride ions, which destroy the compact corrosion product layer, could also be one factor initiating corrosion [37,42,43,38].Song et al.found that the presence of chloride ions increased the film free area and accelerated the electrochemical reaction rate [42].Esmaily et al.further confirmed that chloride ions tended to cause local failure of passive and quasi-passive films on metals, giving rise to localized corrosion [38].In this study, the existing of SPPs(mainly composed by Mg and Y element) is suspected to be the factor initiating localized corrosion.As shown in Fig.9(a),the amplified image shows that localized corrosion can be observed near the position of SPPs.

Fig.5.(a) OM of HTHEed microtube; (2) SEM images showing the distribution of SPPs (SEM, with an acceleration voltage of 20kV and a working distance of 22.2mm); (c), (d) and (e) TEM bright field images showing the shape and size of SPPs (TEM, with an acceleration voltage of 200kV); (f) and (g) EDS results and element table showing the chemical composition of SPP (point 1).

Fig.6.(a) Inverse pole figure of HTHEed microtube; (b) schematic illustration showing crystal orientation of HTHEed microtube.

Fig.7.Evolution of corrosion morphology of HTHEed microtubes as a function of time: (a), (d), (g), and (j) photo showing the HTHEed microtubes after 10, 20, 30, and 40 hours’ immersion; (b), (e), (h), and (k) SEM images showing the evolution of corrosion morphology of LS after 10, 20, 30, and 40 hours’immersion; (c), (f), (i), and (l) SEM images showing the evolution of corrosion morphology of CS after 10, 20, 30, and 40 hours’ immersion.

It should be pointed out that, there exists a critical size of features to create localized corrosion for SPPs according to the published works [27].For Mg-Zn-Y-Nd alloy, the transformation from localized corrosion to uniform corrosion can also be observed when SPPs are refined by severe plastic deformation [4,16,17].These results show that refinement of SPPs is an effective way to alleviated localized corrosion.In the present work, instead of severe plastic deformation,a long-time and high-temperature heat treatment process is used to refine SPPs.The effect of heat treatment is shown in Fig.10.With the increase of time, SPPs in HTHEed microtubes are refined and distributed uniformly.In comparison,coarse SPPs can be observed along the extrusion direction for samples with 3-days’ heat treatment (Fig.10(a)), and service and continuous localized corrosion occurs along extrusion direction (Fig.10(b)).The existence of the coarse SPPs can cause severe localized corrosion.Meanwhile, the refinement of SPPs in HTHEed microtubes alleviates the localizedcorrosion (Fig.7(k)).It should be pointed out that with the refinement of SPPs, the number of SPPs is observed to be increased dramatically.Even the serious localized corrosion does not occur on longitudinal section, the existence of SPPs can still accelerate the corrosion rate.

Fig.8.Hydrogen evolution rate as a function of time for the as-cast ingot and the LS and CS of HTHEed microtubes.

Another role of heat treatment is to dissolve alloying element of Zn and Nd into the matrix, which can also contribute to the improvement of the corrosion resistance of Mg-Zn-YNd alloy [14,28,38].The dissolution of Zn and Nd is judged based on the change of the chemical composition of SPPs:in as-cast condition, second phases are mainly composed of Mg, Nd and Zn (Fig.1(b)), but the chemical composition changes into Mg and Y after the HTHE process.The change of the chemical composition of the second phases is suspect to related to the participation condition.In the work of Du et al., the chemical composition of the second phases of Mg-Zn-Y-Nd alloys was identified with the change of the content of alloying elements (Zn, Y, and Nd), and observed that major second phases were W-phase (Mg3Zn3Y2), and fine precipitates were Mg24Y5[44].Zhao et al.further confirmed that Mg24Y5was precipitated at low contents of Y element[45].Because a long-time, high temperature heat treatment in HTHE process, the dissolution of Y element during heat treatment(a high solubility of Y element at 450°C, around 7.5wt.% [15]) led to a low mole fraction of Y element and the generation of Mg24Y5.The published works have shown that the precipitated Mg24Y5particles are characteriszed by a face-centered cubic (fcc) structure with a lattic parameterαof 0.52nm [46].

Fig.10.(a) the distribution of SPPs after 3-days’ heat treatment; (b) the corresponding corrosion morphology of microtubes after 72 h’ immersion in SBF (Fig.10(b) is referred from the work of Wang et al.[18]).

4.2.Influence of texture

Grain size and orientation can affect the corrosion resistance of Mg alloys by retarding anodic reaction kinetics: Aung et al.found that the corrosion potential shifted from -1.48 to -1.38V with grain size decreasing from 250 to 65μm, and the corresponding hydrogen evolution volume decreases from 46 to 36.5ml/cm2[25]; Song et al.suggested that basal-oriented grains are about 60mV more positive than non-basal-oriented grains for pure polycrystalline Mg in the Mg(OH)2saturated solution [24].Grain refinement can be easily achieved in manufacturing microtubes for stent application.Through cold drawing and in-pass annealing,Wang et al.acquired microtubes with grain size of 4μm [35]; Ge et al.also acquired ultrafine-grained Mg tubes by low-temperature hot extrusion [34].Compared to grain refinement, some researchers have proposed to design microtubes with specific crystal orientation [3, 47], but changing grain orientation to achieve high corrosion performance of Mg-alloy microtubes has never been reported.

Fig.9.(a) amplified image showing initiation of localized corrosion caused by SPPs from area 1 of Fig.7(k); (b) schematic illustration showing the initiation of localized corrosion by SPP.

Fig.11.Polarization curves of the as-cast ingot and the LS and CS of HEHTed microtubes.

In the present work, microtubes were manufactured by the HTHE process, and microtubes with strong texture were acquired after hot extrusion.The basal plane of HTHEed microtubes is oriented parallel to the LS, as shown in Fig.6.With such a crystal orientation, the LS is more anodically stable than the CS.As shown in Fig.11, the corrosion potential or open-curve potential of the LS is about 150mV more positive than that of the CS, and the LS has much lower anodic polarization current density than the CS (Table 3).Meanwhile, the corrosion potential and anodic polarization current density of as-cast sample and the CS are similar to each other, which means similar corrosion resistance for as-cast sample and the CS.

The electrochemical results further prove the influence of texture on the corrosion properties of the LS and CS.However, it should be pointed out that the measured corrosion potentials of the LS and CS are influenced by both grain orientation and SPPs.Given the small size of SPPs, they do not critically influence corrosion potential.The measured corrosion potentials of the LS and CS can still reflect the influence of texture on anodic stability of the LS and CS to a certain degree, indicating that the LS is more anodically stable than the CS.Since the LS is more anodically stable than the CS,the localized corrosion can be observed to be alleviated on the LS.Fig.12 illustrates the corrosion mechanism of the LS and CS.SPPs are considered as the main factor in initiating localized corrosion.Both small and large size SPPs can initiate localized corrosion due to the low corrosion potential of CS,while only large size SPPs can initiate localized corrosion due to a relatively high corrosion potential of LS.

4.3.Comparison between different manufacturing processes

At present, a great deal of work has been dedicated to manufacturing Mg alloy microtubes for stent applications.The manufacturing processes include cold drawing [3,35,36], cold rolling [48], dieless drawing [49–51], and double extrusion[52].Cold drawing is usually recommended, but the resultsshow that cracks can be found in cold-drawn microtubes,which may account for the rapid corrosion rate and severe localized corrosion [53].Moreover, compared with the HTHE process, the texture of microtubes changes after cold-drawing and cold-rolling, causing both the LS and CS to be oriented to non-basal planes (less corrosion resistant) [3,48].However,it should be pointed out that the microtube manufactured by cold drawing have the characteristics of the high dimensionaccuracy and good surface-quality, which are not easy to achieve by other manufacturing methods.Dieless drawing is a thermal process, but at present the corrosion performance of dieless drawn tubes is unclear.Hot extrusion is a process recommended by many researchers: in the work of Du et al.[54], the influence of extrusion ratio and extrusion pass are investigated, which shows that the refinement of SPPs and grain size caused by the increase of extrusion ratio and pass could obviously improve the corrosion resistance; the a In the work of Lu et al.[52], a double extrusion process is introduced, by which refined grains and strong texture of microtubes can be acquired, but the refinement of SPPs seems not to be taken into consideration in the double extrusion process; in the work of Sheng et al.[55], annealing treatment was proved to improve the microstructure of hot-extruded Mg-Zn-Y-Nd alloys.Based on the studies about hot extrusion, a HTHE process (with long-time and high-temperature heat treatment,large extrusion ratio hot extrusion)is proposed to manufacture Mg alloy microtubes, by which refined SPPs and strong-texture HTHEed microtubes are acquired.Finally,the corrosion resistance of the LS is enhanced by alleviating localized corrosion.It should be noted that we cannot simply answer the question that which manufacturing method is bettwe, since different manufacturing mthods have unique advantages.However, even it is still unclear whether HTHEed microtubes can solve the problem of the rapid corrosion rate of severe localized corrosion in clinical tests, the corrosion behavior of stents (made from HTHEed microtubes) can be expected to be controlled to some degree, based on the difference of corrosion resistance between the LS and CS.

Fig.12.Schematic illustration showing the corrosion mechanism on the LS and CS.

5.Conclusion

In this study, a HTHE process was proposed to manufacture Mg-Zn-Y-Nd alloy microtubes for stent applications.The HTHE process is characterized by long-time and hightemperature heat treatment and large-extrusion-ratio hot extrusion.The microstructure and corrosion properties of HTHEed microtubes are characterized, and the relationship between microstructure and corrosion properties was discussed.The following conclusion are made:

(1) After the HTHE process, the SPPs are refined and distributed in matrix, while the size and shape of SPPs are not uniform.Strong texture of HTHEed microtubes can be acquired, with basal planes oriented parallel to the LS.

(2) The corrosion morphology of the LS and CS shows that localized corrosion on the LS is alleviated compared to the CS, and hydrogen evolution tests show that the LS is more corrosion resistant than the CS.

(3) SPPs could be initiating localized corrosion.The more positive corrosion potential of the LS over the CS could be the reason that of localized corrosion on the LS is alleviated.

(4) Based on the different corrosion properties of the LS and CS, the corrosion behavior of stents is expected to be controlled.

Acknowledgment

The authors are grateful for the financial support of Key Projects of the Joint Fund of the National Natural Science Foundation of China (Grant No: U1804251),the National Key Research and Development Program of China (2016YFC1102403,2018YFC1106703 and 2017YFB0702504).Peihua Du thanks China Scholarship Council for the award of fellowship and funding (No.201707040058).Di Mei thanks China Scholarship Council for the award of fellowship and funding (No.201607040051).