Microstructure, biocorrosion and cytotoxicity evaluations of rapid solidified Mg–3Ca alloy

ribbons as a biodegradable material

This article has been downloaded from IOPscience. Please scroll down to see the full text article.

2010 Biomed. Mater. 5 035013

(http://iopscience.iop.org/1748-605X/5/3/035013)

Download details:

IP Address: 162.105.195.240

The article was downloaded on 31/05/2010 at 13:56

Please note that terms and conditions apply.

View the table of contents for this issue, or go to the journal homepage for more

Home Search Collections Journals About Contact us My IOPscience

IOP PUBLISHING BIOMEDICAL MATERIALS

Biomed. Mater. 5 (2010) 035013 (7pp) doi:10.1088/1748-6041/5/3/035013

Microstructure, biocorrosion andcytotoxicity evaluations of rapid solidifiedMg–3Ca alloy ribbons as a biodegradablematerialX N Gu1, X L Li2, W R Zhou1, Y Cheng3 and Y F Zheng1,4

1 State Key Laboratory for Turbulence and Complex System and Department of Advanced Materials andNanotechnology, College of Engineering, Peking University, Beijing 100871, People’s Republic of China2 Center for Biomedical Materials and Engineering, Harbin Engineering University, Harbin 150001,People’s Republic of China3 Center for Biomedical Materials and Tissue Engineering, Academy for Advanced InterdisciplinaryStudies, Peking University, Beijing 100871, People’s Republic of China

E-mail: yfzheng@pku.edu.cn

Received 2 March 2010Accepted for publication 29 April 2010Published 27 May 2010Online at stacks.iop.org/BMM/5/035013

AbstractRapidly solidified (RS) Mg–3Ca alloy ribbons were prepared by the melt-spinning techniqueat different wheel rotating speeds (15 m s−1, 30 m s−1 and 45 m s−1) with the as-cast Mg–3Caalloy ingot as a raw material. The RS45 Mg–3Ca alloy ribbon showed a much more fine grainsize feature (approximately 200–500 nm) in comparison to the coarse grain size (50–100 μm)of the original as-cast Mg–3Ca alloy ingot. The corrosion electrochemical tests in simulatedbody fluid indicated that the corrosion rate of the as-cast Mg–3Ca alloy was strongly reducedby the RS procedure and tended to be further decreased with increasing wheel rotating speeds(1.43 mm yr−1 for RS15, 0.94 mm yr−1 for RS30 and 0.36 mm yr−1 for RS45). The RSMg–3Ca alloy ribbons showed more uniform corrosion morphology compared with the as-castMg–3Ca alloy after polarization. The cytotoxicity evaluation revealed that the threeexperimental as-spun Mg–3Ca alloy ribbon extracts did not induce toxicity to the L-929 cells,whereas the as-cast Mg–3Ca alloy ingot extract did. The L-929 cells showed more improvedadhesion on the surfaces of the three as-spun Mg–3Ca alloy ribbons than that of the as-castMg–3Ca alloy ingot.

(Some figures in this article are in colour only in the electronic version)

1. Introduction

Recent studies have shown that magnesium alloys exhibit goodbiocompatibilities, osteoinductivity and biodegradabilitieswithin bone [1–4]. However, the application of magnesiumalloys produced by traditional ingot metallurgy processing islimited by their fast corrosion rates [3]. In general, the mainreason for the low corrosion resistance of Mg alloys is internalgalvanic attack due to impurities and/or alloying elements

4 Author to whom any correspondence should be addressed.

[5]. Rapid solidification (RS) processing can result in therefinement of both matrix grains and intermetallic particles,extension of solid solubility, improved chemical homogeneity[6, 7] with a consequent suppression of internal galvaniccorrosion and thus enhancement of corrosion resistance [6, 7].For example, RS Mg–Y ribbons exhibited enhanced passivityproperties in 0.01 M NaCl electrolyte (pH = 12) compared topure Mg [6].

The Mg–Ca alloy was recently developed by the presentauthor as a potential bioabsorbable implant material showingno in vitro cytotoxicity and good in vivo biocompatibility

1748-6041/10/035013+07$30.00 1 © 2010 IOP Publishing Ltd Printed in the UK

Biomed. Mater. 5 (2010) 035013 X N Gu et al

within the rabbit’s bone [3]. It was found that the corrosionresistance of the Mg–Ca alloy reduced with increasing Cacontent since the maximum solubility of Ca in Mg at 789.5 Kwas 1.34 wt% [8]. The Mg–Ca alloy with high Ca contentwould form a secondary phase Mg2Ca, which acted as anefficient cathode and led to galvanic corrosion when couplingwith the Mg matrix [9]. For instance, the bulk as-cast Mg–3Caalloy ingot had a corrosion rate of 25 mm yr−1 in simulatedbody fluid (SBF) [3]. In the present study, as-spun Mg–3Ca alloy ribbons were fabricated at different wheel rotatingspeeds, and the effectiveness of the RS technique on theimprovement of corrosion resistance was investigated priorto the fabrication of RS bulk alloys, with the bulk as-castMg–3Ca alloy ingot as contrast.

2. Experimental details

2.1. Preparation of the as-spun Mg–3Ca alloy ribbon

A master alloy ingot of nominal composition, 97 wt% Mg and3 wt% Ca, was melted and cast in a mixed gas atmosphereof SF6 and CO2 using a mild steel crucible. The analyzedchemical composition of the prepared alloy was about96.79 wt% for Mg and 3.21 wt% for Ca. Three kinds of RSribbons were prepared by a single roller melt-spinning methodin an argon atmosphere. The RS ribbons were obtained atdifferent wheel rotating speeds 15.0 m s−1 (RS15), 30.0 m s−1

(RS30) and 45.0 m s−1 (RS45), respectively.

2.2. Microstructural characterization

Selected sections of the melt-spun ribbons were examinedby an x-ray diffractometer (Rigaku DMAX 2400) usingCu Kα radiation for microstructure characterization. Thinsamples for a transmission electron microscope (TEM, Tecnai20) observation were prepared using an ion beam thinningtechnique. The microstructure of the as-cast Mg–3Ca alloyingot was characterized by an optical microscope (OlympusBX51 M).

2.3. Electrochemical tests

The experimental samples were ultrasonically cleaned inacetone, absolute ethanol and distilled water prior to theelectrochemical and cytotoxicity tests. The electrochemicalbehavior was carried out at (37 ± 0.5) ◦C in SBF (NaCl 8.035 g,NaHCO3 0.355 g, KCl 0.225 g, K2HPO4·3H2O 0.231 g,MgCl2·6H2O 0.311 g, CaCl2 0.292 g, Na2SO4 0.072 g and(HOCH2)3CNH2 6.118 g, pH = 7.4 [10]) using a saturatedcalomel electrode (SCE) as a reference electrode and platinumas a counter electrode with a corrosion measurement system(CHI660C). The exposed area of the samples to the solutionwas 0.385 cm2. The open circuit potential was measuredand potentiodynamic polarization tests were carried out at ascan rate of 1 mV s−1 with the initial potential approximately300 mV below the corrosion potential. The electrochemicalimpedance spectroscopy (EIS) measurement was carried outfrom 100 mHz to 100 kHz at open circuit potential (OCP)values. The corrosion rate was calculated according to ASTM

G102-89. An average of three measurements was taken foreach group. Changes in the surface morphologies of thesamples were characterized using an environmental scanningelectron microscope (ESEM; Quanta 200FEG), equipped withan energy-dispersive spectrometry (EDS) attachment.

2.4. Cell experiments

L-929 cell lines were cultured under standard cell cultureconditions (37 ◦C, 5% CO2 and 95% rH) in Dulbecco’sModified Eagle’s Medium (DMEM) with 10% fetal bovineserum (FBS), 100 U ml−1 penicillin and 100 μg ml−1

streptomycin. Both the indirect and direct cell assays werecarried out in the cell experiments after at least 2 h ultraviolet-radiation sterilization of the samples.

For the indirect cell assay, the extraction ratio was1.25 cm2 ml−1 in DMEM and the extraction of the sampleswas carried out in a humidified atmosphere with 5% CO2 at37 ◦C for 72 h. The supernatant fluid was withdrawn andcentrifuged to prepare the extraction medium, and then storedat 4 ◦C before the cell experiment. The cell cultures wereinoculated at the density of 5 × 103 cells/well and grown for24 h. Then the culture medium was replaced by the extracts.After incubation with the extraction medium for 1, 2 and4 days, the cell activity was determined by a cell proliferationkit (MTT; Sigma) assay. The detailed procedure of the indirectcell assay was described in our previous work [11]. Theconcentrations of Mg in the extraction mediums were alsomeasured using inductively coupled plasma atomic emissionspectrometry (Leeman, Profile ICP-AES).

For the direct observation of the cell proliferationbehavior, 5 × 104 cell ml−1 L-929 cells were seeded ontothe as-cast Mg–3Ca alloy samples and as-spun Mg–3Ca alloyribbons (RS15, RS30 and RS45) and incubated for 2 days at37 ◦C with 5% CO2 in humidified air in 24 well plates. At theend of the cultures, the samples were rinsed three times withPBS and the adhered cells were fixed in 2.5% glutaraldehydesolutions for 1 h at room temperature followed by dehydrationin a gradient ethanol/distilled water mixture (50%, 60%,70%, 80%, 90% and 100%) for 10 min each and dried ina hexamethyldisilazane (HMDS) solution. The surface of celladhered experimental samples were observed by ESEM. Threeparallel samples were utilized for each experimental condition.

3. Results and discussion

The as-spun Mg–3Ca alloy ribbons obtained at differentwheel speeds are approximately 330 μm thick for RS15,140 μm thick for RS30 and 80 μm thick for RS45. As thewheel speed increases, the thickness of the ribbons inherentlydecreases. Figure 1 shows the XRD patterns for the as-castMg–3Ca alloy ingot and as-spun Mg–3Ca alloy ribbons atdifferent wheel speeds. The RS15, RS30 and RS45 Mg–3Ca alloy ribbons show only an α-Mg matrix phase, whereasboth the α-Mg and Mg2Ca phases can be seen for the as-cast Mg–3Ca alloy sample. Figure 2(a) shows the opticalmicrograph of the as-cast Mg–3Ca alloy sample and figure 2(b)illustrates the TEM micrograph of the as-spun RS45 Mg–3Ca alloy ribbon. Clearly, the latter has a very fine

2

Biomed. Mater. 5 (2010) 035013 X N Gu et al

10 20 30 40 50 60 70 80

Mg

2Ca(

103)

Inte

nsi

ty/a

.u.

2theta/deg

αM

g(1

12)

αM

g(1

03)

αM

g(1

10)

αM

g(1

02)αM

g(1

01)

αMg

(002

)αM

g(1

00)

RS45

RS30

RS15

as-cast

Mg

2Ca(

112)

Mg

2Ca(

110)

Figure 1. X-ray diffraction patterns of the as-cast Mg–3Ca alloyingot and as-spun Mg–3Ca alloy ribbons produced at differentwheel rotating speeds.

grain size, 200–500 nm, compared to the coarse grain size,50–100 μm, of the as-cast Mg–3Ca alloy sample. Somenanoscale precipitates were observed along the grain boundaryof the RS45 Mg–3Ca alloy ribbon. Due to the high stabilityof secondary phases in SBF, this microstructure may be morefavorable for degradable biomaterials compared to the highvolume fraction of secondary phases for the as-cast Mg–3Caalloy [12].

Figure 3(a) shows the OCP as a function of immersiontime for the as-cast Mg–3Ca alloy ingot and as-spun Mg–3Ca alloy ribbons obtained at different wheel speeds tested inSBF at 37 ◦C. The as-spun Mg–3Ca alloy ribbons show morepositive and more stabilized OCP values than that of the as-cast Mg–3Ca alloy ingot. In addition, the OCP values of theas-spun Mg–3Ca alloy ribbons move to a positive value withincreasing wheel speeds (approximately −1.72 V for RS15,−1.57 V for RS30 and −1.55 V for RS45).

Figure 3(b) shows the representative potentiodynamicpolarization curves of the as-cast Mg–3Ca alloy ingot andas-spun Mg–3Ca alloy ribbons obtained at different wheelspeeds tested in SBF at 37 ◦C. It is noted that corrosioncurrent densities (Icorr) decrease after the RS process, as listedin table 1. In addition, the cathodic polarization current

(a) (b)

Figure 2. (a) Optical micrograph of the as-cast Mg–3Ca alloy ingot; (b) TEM micrograph of the as-spun RS45 Mg–3Ca alloy ribbon.

of the as-spun Mg–3Ca alloy ribbons is much lower thanthat of the as-cast Mg–3Ca alloy ingot. That is to say, thecathodic reaction resistance is improved by the increment ofcooling rate in the corrosion process, and thus, also leads tothe increased corrosion resistance of the Mg–3Ca alloy. Inaddition, Icorr of the as-spun Mg–3Ca alloy ribbons decreaseswith increasing wheel speeds (table 1). Figure 4 shows thecorrosion morphologies of the as-cast Mg–3Ca alloy ingotand the as-spun Mg–3Ca alloy ribbons obtained at differentwheel speeds after the polarization tests. It can be seenthat the surface of the as-cast Mg–3Ca alloy is non-uniform,whereas the as-spun Mg–3Ca alloy ribbons exhibit a relativeeven corrosion surface with some visible cracks. In addition, acompact and homogeneous corrosion product layer is observedon the surface of the RS45 sample, as shown in figure 4(d).The EDS results indicate that the compositions of the surfacelayer on the as-cast Mg–3Ca alloy ingot and as-spun Mg–3Caalloy ribbons are C, O, Mg, Cl and Ca. The Cl content is muchhigher at the surface of the as-cast Mg–3Ca alloy ingot thanthat of the as-spun Mg–3Ca alloy ribbons.

Figure 3(c) presents the Nyquist plots of the as-cast Mg–3Ca alloy ingot and as-spun Mg–3Ca alloy ribbons obtainedat different wheel speeds tested in SBF at 37 ◦C. Two timeconstants can be seen in the Nyquist plots: one capacitanceloop at high frequency and the other at low frequency, whichare related to the electric double layer and the surface film,respectively. The corresponding equivalent circuit is shown infigure 3(d). The electron transfer resistance Rt is parallel to theelectric double layer capacity Cdl, and the film capacity Cf isparallel to the film resistance Rf . The fitting results are listedin table 2. Both Rf and Rt values for the as-cast Mg–3Ca alloyincrease after the rapid solidification process, indicating theirslower corrosion rates. Moreover, the corrosion rate of theribbon decreases with increasing wheel speeds, showing thesame variation sequence as the potentiodynamic polarization.Izumi [12] also indicated that the corrosion rates of RS Mg–Zn–Y alloys in a 0.17 M NaCl neutral aqueous solution(pH 6.8) at 298 K decreased with increasing wheel speed.

The different corrosion behavior of the as-cast Mg–3Ca alloy ingot and as-spun Mg–3Ca alloy ribbons may beattributed to the improved effectiveness of corrosion resistancefor the surface films formed on as-spun ribbons than that onthe as-cast Mg–3Ca alloy ingot, due to the extended Eb –Ecorr values and increased Rf values for RS30 and RS45

3

Biomed. Mater. 5 (2010) 035013 X N Gu et al

0 1000 2000 3000 4000-2.1

-2.0

-1.9

-1.8

-1.7

-1.6

-1.5

-1.4

RS45RS30

RS15

as-cast

Po

ten

tial

/V

Time/sec

(a)

-8 -7 -6 -5 -4 -3 -2 -1

-2.2

-2.0

-1.8

-1.6

-1.4

-1.2

-1.0

-0.8

-0.6

RS45

RS30

RS15

as-castPo

ten

ial/V

Log Current Density/A.cm-2

(b)

0 100 200 300 400 500-100

-80

-60

-40

-20

0

20

40

60

80

100

120

140

160

180

200

as-cast RS15 RS30 RS45

-Z''/

oh

m.c

m2

Z'/ohm.cm2

(c)

(d)

Figure 3. (a) OCP, (b) representative potentiodynamic polarization curves and (c) Nyquist plots of the as-cast Mg–3Ca alloy ingot andas-spun Mg–3Ca alloy ribbons (RS15, RS30 and RS45) tested in SBF. (d) Equivalent circuit of the sample/solution system.

Table 1. The electrochemical data obtained from potentiodynamic polarization measurement.

Ecorr (V) Icorr (μA cm−2) Eb – Ecorr (mV) vcorr (mm yr−1)

As-cast Mg–3Ca alloy ingot −1.838 ± 0.02 929.3 ± 60.1 419.7 ± 71.8 21.0 ± 1.4RS15 Mg–3Ca alloy ribbon −1.591 ± 0.06 74.2 ± 9.9 264.1 ± 20.1 1.68 ± 0.2RS30 Mg–3Ca alloy ribbon −1.530 ± 0.05 55.6 ± 14.3 531.7 ± 46.3 1.25 ± 0.3RS45 Mg–3Ca alloy ribbon −1.516 ± 0.04 17.1 ± 4.9 621.7 ± 51.0 0.39 ± 0.1

Table 2. EIS fitted results of as-cast Mg–3Ca alloy ingot and as-spun Mg–3Ca alloy ribbons obtained at different wheel rotating speeds.

Rs Cf Rf Cdl Rt Rp vcorr

(� cm2) (μF cm−2) m (� cm2) (μF cm−2) n (� cm2) (� cm2) (mm yr−1)

As-cast Mg–3Ca alloy ingot 17.28 2.32 0.5590 23.01 206.5 0.8795 162.5 185.51 18.04RS15 Mg–3Ca alloy ribbon 19.49 77.83 0.5984 69.36 33.60 0.8696 212.7 282.06 2.19RS30 Mg–3Ca alloy ribbon 17.37 36.05 0.5941 89.88 32.89 0.8655 292.4 382.28 1.49RS45 Mg–3Ca alloy ribbon 20.13 8.82 0.6125 146.5 14.27 0.8015 312.5 459 0.43

than the as-cast Mg–3Ca alloy ingot. It is reported thatthe enhancement of microstructural and electrochemicalhomogeneities by the RS process would result in modificationof the surface films and an increasing resistance to localbreakdown of the films, regardless of the same alloycomposition [6, 13]. On the other hand, the decreasing surfacearea ratio of second phase Mg2Ca to Mg in the as-spun Mg–3Ca alloy ribbons, suggesting a decreasing amount of galvaniccoupling, is believed to result in the reduced corrosion rate for

the as-spun Mg–3Ca alloy ribbons. The fine grain size forthe as-spun Mg–3Ca alloy ribbons may also contribute to theirreduced corrosion rate. Ambat [14] reported that a die-castmaterial with smaller grain size and fine β phase showed lowercorrosion rate and better passivation ability compared to theingot sample. The corrosion rates of the RS15, RS30 and RS45Mg–3Ca alloy ribbons are much slower than that of the as-castMg–3Ca alloy ingot, and at the same level as an as-rolled Mg–1Ca alloy (1.63 mm yr−1) and an as-extruded Mg–1Ca alloy

4

Biomed. Mater. 5 (2010) 035013 X N Gu et al

(a) (b)

(c) (d )

Figure 4. SEM images showing the surface morphologies of the (a) as-cast Mg–3Ca alloy, (b) RS15, (c) RS30 and (d) RS45 Mg–3Ca alloyribbons after the potentiodynamic polarization test.

0

20

40

60

80

100

120

140

160

cell

viab

ility

/%

culture time/d

as-cast RS15 RS30 RS45

1 2 4

Figure 5. L-929 cell viability expressed as a percentage of theviability of cells in the negative control after 1, 2 and 4 days ofculture in the extracts of as-cast Mg–3Ca alloy ingot and threeas-spun Mg–3Ca alloy ribbons (RS15, RS30 and RS45).

(1.74 mm yr−1) [3]. The RS45 Mg–3Ca alloy ribbonshowed a Icorr value comparable with that of the AZ91Caalloy [15].

Figure 5 shows the cell viability of cultures in theindividual extraction media of the as-cast Mg–3Ca alloy ingotand the three as-spun Mg–3Ca alloy ribbons for 1, 2 and 4days. The pH values of the extracts are 9.28 ± 0.14 for the as-cast Mg–3Ca alloy, 8.36 ± 0.10 for the RS15 ribbon, 8.30 ±

0.11 for the RS30 ribbon and 8.32 ± 0.06 for the RS45 ribbon.The as-cast Mg–3Ca alloy extract shows toxicity to the L-929 cells, which may be due to its poor corrosion resistance,causing a large change in the pH value of the culture medium[11, 16]. In addition, the Mg concentrations of the extractsare (453 ± 47) μg ml−1 for the as-cast Mg–3Ca alloy, (221 ±32) μg ml−1 for the RS15 ribbon, (178 ± 36) μg ml−1 for theRS30 ribbon and (142 ± 29) μg ml−1 for the RS45 ribbon,which are much lower than the IC50 of magnesium ions forMG63 osteoblasts [17]. It is reasonable to believe that thesehigh Mg ion concentrations will not cause toxicity to the cells.In the case of the as-spun Mg–3Ca alloy ribbons, there is nosignificant difference of cell viabilities between the negativecontrol and the RS15, RS30 and RS45 Mg–3Ca alloy ribbongroups, showing that the three as-spun Mg–3Ca alloy ribbonshave no toxicity to the L-929 cells during the 4 day cultureperiod.

Figure 6 presents the morphologies of the L-929 cellsafter 4 days of incubation in the as-cast Mg–3Ca alloy ingotand the three as-spun Mg–3Ca alloy ribbon extraction media,respectively. It can be seen that the L-929 cells cultured in thethree as-spun Mg–3Ca alloy ribbon extracts exhibit a flattenedspindle shape, whereas cells cultured in the as-cast Mg–3Caalloy ingot extract show a round shape.

Figure 7 presents the morphologies of the L-929 cellsafter 2 days of incubation on the surfaces of the as-cast Mg–3Ca alloy ingot and the three as-spun Mg–3Ca alloy ribbonsamples. It can be seen that the L-929 cells exhibited differentresponses to the different materials. No attached cells can beobserved on the surface of the as-cast Mg–3Ca alloy sample,only the corrosion products. For the three as-spun Mg–3Ca

5

Biomed. Mater. 5 (2010) 035013 X N Gu et al

(a) (b)

(c) (d)

Figure 6. Optical micrographs of L-929 cells cultured with the (a) as-cast Mg–3Ca alloy ingot extract, (b) RS15 Mg–3Ca alloy ribbonextract, (c) RS30 Mg–3Ca alloy ribbon extract and (d) RS45 Mg–3Ca alloy ribbon extract for 4 days.

(a) (b)

(c) (d)

Figure 7. SEM micrographs of L-929 cells cultured on the (a) as-cast Mg–3Ca alloy ingot sample, (b) RS15 Mg–3Ca alloy ribbon sample,(c) RS30 Mg–3Ca alloy ribbon sample and (d) RS45 Mg–3Ca alloy ribbon sample for 2 days.

alloy ribbons, many L-929 cells can be seen on their surfacesmaintaining the round or spindle-like morphologies after2 days of incubation. This may be because Mg ions are

active in cell adhesion mechanisms [18, 19]. However, the fastcorrosion of the as-cast Mg–3Ca alloy ingot, resulting in a fastincrease in the pH value of the DMEM, is undesirable for cell

6

Biomed. Mater. 5 (2010) 035013 X N Gu et al

adhesion, growth and proliferation [20]. Thus, the RS process,with the Mg ion releasing and a tolerant pH change, mightsomehow benefit the cell adhesion and proliferation.

4. Conclusions

The present study investigated the microstructure andcorrosion resistance of rapid solidified Mg–3Ca alloy ribbons.The rapid solidified Mg–3Ca alloy ribbons showed much finergrain features than the as-cast Mg–3Ca alloy ingot. Thecorrosion results showed that the corrosion resistance for theas-spun Mg–3Ca alloy ribbons in SBF was improved comparedwith the as-cast Mg–3Ca alloy ingot and the corrosionrates decreased with increasing wheel rotating speeds. Thecytotoxicity evaluation tests showed that the RS15, RS30 andRS45 Mg–3Ca alloy ribbons indicated no toxicity to the L-929 cells, whereas the as-cast Mg–3Ca alloy ingot extractdid. In addition, improved adhesion of the L-929 cells canbe seen on the surfaces of the RS15, RS30 and RS45 Mg–3Ca alloy ribbons compared with the as-cast Mg–3Ca alloyingot. Therefore, the rapid solidification procedure may be apromising technique to improve the corrosion resistance andcytocompatibility of the Mg–Ca alloy.

Acknowledgments

This work was supported by the foundation of KeyLaboratory for Advanced Materials Processing Technology,Ministry of Education (2008001), National Natural ScienceFoundation of China (No. 30670560), Science and TechnologyProject of Guangdong Province (2008A030102006),ShenZhen Industry-Education-Research Cooperation fundingand Program for New Century Excellent Talents in University(NCET-07-0033).

References

[1] Witte F, Kaese V, Haferkamp H, Switzer H,Meyer-Lindenberg A, Wirth C J and Windhagen H 2005In vivo corrosion of four magnesium alloys and theassociated bone response Biomaterials 26 3557–63

[2] Witte F, Fischer J, Nellesen J, Crostack H-A, Kaese V,Pischd A, Beckmann F and Windhagen H 2006 In vitro andin vivo corrosion measurements of magnesium alloysBiomaterials 27 1013–8

[3] Li Z, Gu X, Lou S and Zheng Y 2008 The development ofbinary Mg–Ca alloys for use as biodegradable materialswithin bone Biomaterials 29 1329–44

[4] Xu L, Pan F, Yu G, Yang L, Zhang E and Yang K 2009 In vitroand in vivo evaluation of the surface bioactivity of a calciumphosphate coated magnesium alloy Biomaterials30 1512–23

[5] Song G and Atrens A 2007 Recent insights into themechanism of magnesium corrosion and researchsuggestions Adv. Eng. Mater. 9 177–83

[6] Yao H B, Li Y and Wee A T S 2003 Passivity behavior ofmelt-spun Mg–Y alloys Electrochim. Acta. 48 4197–204

[7] Yamasaki M, Hayashi N, Izumi S and Kawamura Y 2007Corrosion behavior of rapidly solidified Mg–Zn rare earthelement alloys in NaCl solution Corros. Sci. 49 255–62

[8] ASM International 1992 Alloy phase diagram ASM Handbookvol 3 (Materials Park, OH: ASM International, TheMaterials Information Company)

[9] Kim J-G and Kim Y-W 2001 Advanced Mg–Mn–Ca sacrificialanode materials for cathodic protection Mater. Corros.52 137–9

[10] Kokubo T and Takadama H 2006 How useful in SBF inpredicting in vivo bone bioactivity? Biomaterials27 2907–15

[11] Gu X, Zheng Y, Cheng Y, Zhong S and Xi T 2009 In vitrocorrosion and biocompatibility of binary magnesium alloysBiomaterials 30 484–98

[12] Kannan M B 2010 Influence of microstructure on the in-vitrodegradation behaviour of magnesium alloys Mater. Lett.64 739–42

[13] Izumi S, Yamasaki M and Kawamura Y 2009 Relationbetween corrosion behavior and microstructure ofMg–Zn–Y alloys prepared by rapid solidification at variouscooling rates Corros. Sci. 51 395–402

[14] Ambat R, Aung N N and Zhou W 2000 Evaluation ofmicrostructural effects on corrosion behavior of AZ91Dmagnesium alloy Corros. Sci. 42 1433–55

[15] Kannan M B and Raman R K S 2008 In vitro degradation andmechanical integrity of calcium-containing magnesiumalloys in modified-simulated body fluid Biomaterials29 2306–14

[16] Groos E, Walker L and Masters J R W 2006 Intravesicalchemotherapy: studies on the relationship between pH andcytotoxicity Cancer 58 1199–203

[17] Hallab N J, Vermes C, Messina C, Roebuck K A, Glant T Tand Jacobs J J 2002 Concentration- andcomposition-dependent effects of metal ions on humanMG-63 osteoblasts J. Biomed. Mater. Res. 60 420–33

[18] Hynes R O 1992 Integrins: versatility, modulation, andsignaling in cell adhesion Cell 69 11–25

[19] Paul W and Sharma C P 2006 Nanoceramic matrices:biomedical applications Am. J. Biochem. Biotechnol. 2 41–8

[20] Song G and Song S 2007 A possible biodegradablemagnesium implant material Adv. Eng. Mater. 9 298–302

7