Research article

Received: 27 June 2012 Revised: 31 December 2012 Accepted: 3 February 2013 Published online in Wiley Online Library: 2 April 2013

(wileyonlinelibrary.com) DOI 10.1002/sia.5257

Corrosion resistance and cytotoxicity of aMgF2 coating on biomedical Mg–1Ca alloyvia vacuum evaporation deposition methodN. Li,a Y. D. Li,b Y. B. Wang,a M. Li,c Y. Cheng,c Y. H. Wuc and Y. F. Zhenga,c*

To reduce the biocorrosion rate and enhance the biocompatibility by surface modification, MgF2 coatings were prepared onMg–1Ca alloy using vacuum evaporation deposition method. The average thickness of the coating was about 0.95 mm. Theresults of immersion test and electrochemical test indicated that the corrosion rate of Mg–1Ca alloy was effectively decreasedafter coating with MgF2. The MgF2 coating induced calcium phosphate deposition on Mg–1Ca alloy. After 72 h culture, MG63cells and MC3T3-E1 cells were well spread on the surface of the MgF2-coated Mg–1Ca alloy, while few cells were observed onuncoated Mg–1Ca alloy samples. In summary, MgF2 coating showed beneficial effects on the corrosion resistance and thusimproved cell response of theMg–1Ca alloy effectively and should be a good surface modificationmethod for other biomedicalmagnesium alloys. Copyright © 2013 John Wiley & Sons, Ltd.

Keywords: Mg–1Ca alloy; magnesium fluoride; surface modification; corrosion resistance; biocompatibility

* Correspondence to: Y. F. Zheng, Department ofMaterials Science and Engineering,College of Engineering, Peking University, Beijing 100871, China E-mail:yfzheng@pku.edu.cn

a State Key Laboratory for Turbulence and Complex System and Department ofMaterials Science and Engineering, College of Engineering, Peking University,Beijing 100871, China

b DongGuan EONTEC Co., Ltd, Yin Quan Industrial District, Qing Xi, DongGuan523662, China

c Center for Biomedical Materials and Tissue Engineering, Academy forAdvanced Interdisciplinary Studies, Peking University, Beijing 100871, China

121

Introduction

As a lightweight metal with mechanical properties similar tonatural bone, a natural ionic presence with significant functionalroles in biological systems, and a material that can degrade in theelectrolytic environment of the body, magnesium-based implantshave the potential to serve as biocompatible, osteoconductive,degradable implants for load-bearing applications.[1] However,the medical applications of Mg alloys had been hindered owingto the fast corrosion ratemismatched with the tissue healing rate.[2]

In order to slow down the corrosion rate of magnesium alloys,as well as to maintain their mechanical integrate and to improvetheir biocompatibility, various kinds of surface modification tech-niques have been developed, such as alkaline heat treatment,[3,4]

electrodeposition,[5–7] polymer coating,[8] phosphating treatment [9]

and microarc oxidation.[10,11]

Fluorine is a natural component of the human skeleton andteeth. This is a rationale for using F supplements for treatmentof osteoporosis, as F incorporation into bone increases the sizeof the apatite crystals and larger crystals are more resistant toosteoclastic attack.[12] In addition, fluoride is one of the fewknown agents that can stimulate osteoblast proliferation andincrease new mineral deposition in cancellous bones.[12,13]

Fluoride treatment of the implant surface was employed todegrade cytotoxicity of titanium implants [14] and promoteosseointegration in the early phase of healing following dentalimplant installation.[15]

Fluoride treatments also have been extensively reported toimprove the corrosion resistance of Mg and its alloys in industrialenvironments [16–18] and in biomedical field.[19–24] Witte et al. [20]

found MgF2 coating could reduce in vivo corrosion rate ofLAE442 magnesium alloy with no elevated fluoride concentrationin the adjacent bone. Drynda et al. [23] developed fluoride-coatedmagnesium-calcium alloys that exhibit improved mechanical fea-tures, decreased degradation kinetics and a good biocompatibility

Surf. Interface Anal. 2013, 45, 1217–1222

toward vascular cells in vitro. Thomann et al. [21] reported fluoride-coatedMg0.8Ca implants possess good clinical tolerance in a rabbitmodel and show a slight reduction of the degradation velocity incomparison to uncoated Mg0.8Ca implants. Yan et al. [24] success-fully prepared a flourinated coating on AZ31B magnesium alloy,which could maintain the mechanical property of the alloy for aslong as 45 days in SBF.

Fluoride treatment of magnesium alloys is usually conductedby immersing the samples in hydrofluoric acid. This method istime-consuming (several hours to several days), and the treat-ment should be very carefully performed since hydrofluoric acidis very toxic. Vacuum evaporation deposition is a time-efficientsurface modification technique. MgF2 coatings on magnesiumalloys prepared by vacuum evaporation deposition method havenot been reported.

In the present study, MgF2 coatings were prepared on Mg–1Caalloy using vacuum evaporation deposition method. Themicrostructure and in vitro corrosion properties of MgF2-coatedMg–Ca alloy were assessed. The surface biocompatibility of theMgF2 coatings was evaluated by the direct assays.

Copyright © 2013 John Wiley & Sons, Ltd.

7

N. Li et al.

1218

Material and methods

Preparation of samples

The Mg–1Ca (1wt.%) alloy ingot was cast by commercial pureMg (99.98%) and Ca (99.95%) in a crucible under a mixed gasatmosphere of SF6 and CO2, then was extruded into rod at210 �C with a reduction ratio of 17. The experimental Mg–1Caalloy samples were machined from the extruded rod with 12 mmin diameter and 2 mm in thickness, mechanically polished up to2000 grit, ultrasonically cleaned in acetone, absolute ethanol anddistilled water.For the deposition of the MgF2 film, a custom-made multi-

function coating machine (Technol Science Co., Ltd., China) wasemployed. MgF2 powder was placed in a molybdenum boatand evaporated at a base pressure of 0.1 Pa. The deposition timewas 3 min.The surface and cross-sectional morphologies of MgF2 coatings

were characterized by environmental scanning electronmicroscopy (ESEM, Quanta 200FEG), equipped with an energy-dispersive spectroscopy (EDS) attachment. The grazing incidentX-ray diffraction (GIXRD) was conducted on a Rigaku Dmax2500-VX-ray diffractometer with Cu Ka radiation to identify constituentphases of the coating. The incident angle was 1� and the scanningrate was 4�/min.

Electrochemical test

A three-electrode cell was used for electrochemical measure-ments with a platinum counter-electrode and a saturatedcalomel electrode as the reference electrode. The tests wereperformed in Hank’s solution (NaCl 8.00 g/L, KCl 0.40 g/L, CaCl20.14 g/L, NaHCO3 0.35 g/L, MgSO4�7H2O 0.20 g/L, Na2HPO4�12H2O0.12 g/L, KH2PO4 0.06 g/L, pH= 7.4) at 37� 0.5 �C using a CHI660Celectrochemistry workstation. The exposed area of the workingelectrode to the electrolyte was 1.13 cm2. Potentiodynamicpolarization scans were performed at a scanning rate of 1 mV/s,and the initial potential was 300 mV below the open circuitpotential (OCP). The electrochemical impedance spectroscopymeasurement was carried out from 100 mHz to 100 kHz atOCP values. An average of three measurements was taken foreach group.

Immersion test

The immersion test was carried out in Hank’s solution at 37� 0.5 �Caccording to ASTM-G31-72. The pH value of the solution and the

Figure 1. SEM micrographs of (a) surface morphology and (b) cross-section

wileyonlinelibrary.com/journal/sia Copyright © 2013 John

hydrogen evolution rate were monitored during the immersiontest. An average of three measurements was taken. After 20 dimmersion, the samples were removed from of the solution,rinsed with distilled water and dried in air. Changes on thesurface microstructure of coated and bare samples before andafter immersionwere checked by ESEM and an X-ray diffractometer(Rigaku DMAX 2400) using Cu Ka radiation.

Direct cell adhesion assay

Human osteosarcoma cells (MG63) and mouse osteoblast-likecells (MC3T3-E1) were used in the direct cell adhesion assay.MG63 cells were cultured in MEM, while MC3T3-E1 cells were cul-tured in a-MEM. Both culturemedia contained 10%FBS, 100 U �ml�1

penicillin and 100 mg �ml�1 streptomycin. Coated and uncoatedMg–1Ca alloy samples were settled in the bottom of each well of24-well cell culture plates, and 1 ml cell suspensions were seededin each well at a cell density of 6� 104. After 72 h culture in ahumidified atmosphere with 5% CO2 at 37 �C, the samples wererinsed with PBS for three times and then fixed in 2.5% glutaralde-hyde solution for 2 h at room temperature. The samples werethen dehydrated in a gradient ethanol/distilled water mixture(50, 60, 70, 80, 90 and 100%) for 10 min each and dried in ahexamethyldisilazane solution. The morphologies of the cellsadhered to the surfaces of samples were observed using ESEM.

Results and discussion

Characterization of the MgF2 coating

The top-view surface and cross-sectional morphologies ofMgF2 coatings on the Mg–1Ca alloy samples are shown in Fig. 1,revealing an average coating thickness of about 0.95 mm. The linescanning of elements distribution across the coating and thesubstrate depicted in Fig. 1(b) demonstrated elements existedin the coating are mainly Mg and F. GIXRD patterns in Fig. 2 provethe coating was MgF2. The MgF2 coatings were transparent andcould not be identified with the naked eye. The crack formedwhile cooling due to the different thermal expansion coefficientof the film and substrate.

Electrochemical corrosion

Figure 3(a) shows the OCP–time curves of the bare and MgF2-coated Mg–1Ca alloy samples exposed to Hank’s solution for 1 h.The OCP curve recorded for the untreated sample was more

al morphology with EDS line scan of the MgF2-coated Mg–1Ca alloy.

Wiley & Sons, Ltd. Surf. Interface Anal. 2013, 45, 1217–1222

Figure 2. GIXRD patterns of the MgF2-coated Mg–1Ca alloy.

Corrosion resistance and cytotoxicity of MgF2 coating on Mg–Ca alloy

positive than the MgF2-coated one, and both samples exhibited asimilar variation trend with the extension of immersion time.Figure 3(b) shows typical potentiodynamic polarization curves ofthe bare and MgF2-coated samples after OCP measurements. TheMgF2-coated Mg–1Ca alloy sample showed reduced kinetics ofanodic and cathodic reactions than that of untreated Mg–1Casample. The corrosion current density of the MgF2-coatedMg–1Ca alloy was half ((6.06� 1.00) mA�cm�2) of that of the bareMg–Ca alloy ((11.58� 2.31) mA�cm�2), and the corrosion ratescalculated according to ASTM-G102-89 were (0.14� 0.02)mm�yr�1 and (0.26� 0.05) mm�yr�1, respectively. Two timeconstants can be seen in the Nyquist plots in Fig. 3(c): one capaci-tance loop at high frequency and the other at low frequency,which are related to the electric double layer and the surface film,respectively. The polarization resistance of the samples can beobtained from the diameter of the near semi-circular spectra

Figure 3. (a) Open circuit potential, (b) potentiodynamic polarization curvesHank’s solution.

Surf. Interface Anal. 2013, 45, 1217–1222 Copyright © 2013 Joh

which was related to the corrosion rate. The MgF2-coatedMg–1Ca alloy exhibited a much higher polarization resistance thanthe bare Mg–1Ca alloy.

Immersion corrosion

Figure 4 presents the variation of the pH value of the immersionsolution and the hydrogen evolution volume during the immersionperiod. As shown in Fig. 4(a), hydrogen evolution volume of bothkinds of samples displayed approximately linear relationship withthe immersion time, and the bare Mg–1Ca alloy released muchmore hydrogen than the MgF2-coated one. Smaller increase in pHvalue of the immersion solution also indicated the MgF2 coatingprotected the Mg–1Ca alloy from fast degradation (Fig. 4(b)).

Figure 5 displays the surface morphology of the untreatedMg–1Ca alloy sample and MgF2-coated Mg–1Ca alloy samplesafter 20 d immersion in Hank’s solution. It can be seen thatuntreated Mg–1Ca alloy sample exhibited a severe corrosionattack with some deep pits visible (Fig. 5(a)). The EDS analysis inFig. 5(c) and Fig. 5(d) reveals the existed elements on the corrodedsample surface were mainly C, O and Mg, with Cl detected at thecorrosion pit position 1 and trace Ca and P detected at area 2. Thesurface morphology of the MgF2-coated sample didn’t changemuch after 20 d immersion (Fig. 5(b)), and the EDS analysis in Fig. 5(e) and Fig. 5(f) reveals compounds that rich in Ca and P haddeposited on the MgF2 film. This enhanced deposition of calciumphosphate might suggest an accelerated mineral deposition rateon the Mg–1Ca alloy sample when implanted into bone.

The XRD patterns of the untreated Mg–1Ca alloy sample andMgF2-coated Mg–1Ca alloy samples after 20 d immersion inHank’s are shown in Fig. 6. The resulting corrosion products foruntreated Mg–1Ca alloy sample after 20 d immersion are mainlyMg(OH)2, with the peak intensity of a-Mg was remarkablydecreased. In contrast, only small amount of Mg(OH)2 can be

and (c) Nyquist plots of bare and MgF2-coated Mg–1Ca alloy samples in

n Wiley & Sons, Ltd. wileyonlinelibrary.com/journal/sia

1219

Figure 4. (a) The hydrogen evolution volume and (b) the change of pH value of Hank’s solution incubating bare and MgF2-coated Mg–1Ca alloysamples as a function of the immersion time.

Figure 5. The surface morphologies of (a) bare Mg–1Ca alloy, (b) MgF2-coated Mg–1Ca alloy samples after 20 d immersion in Hank’s solution and EDSspectrum of (c) area 1 in (a), (d) area 2 in (a), (e) area 3 in (b) and (f) area 4 in (b).

N. Li et al.

wileyonlinelibrary.com/journal/sia Copyright © 2013 John Wiley & Sons, Ltd. Surf. Interface Anal. 2013, 45, 1217–1222

1220

Figure 6. The XRD patterns of bare and MgF2-coated Mg–1Ca alloy sam-ples after 20 d immersion in Hank’s solution.

Corrosion resistance and cytotoxicity of MgF2 coating on Mg–Ca alloy

detected on the surface of the MgF2-coated Mg–1Ca alloysamples.

The MgF2 coating insulates the Mg–1Ca alloy bulk from theelectrolyte and thereby decreases the corrosive attack of chlorideions. Therefore, the corrosion of the Mg–1Ca alloy bulk cannotstart before the protective coating has dissolved or peeled. Thevalid life of the surface coating guarantees not only the reducedcorrosion rate but also the mechanical integrity of magnesiumalloy implants.[11] The results of immersion test indicates thein vitro valid life of the MgF2 coating is over 20 days, which isbetter than that of the alkaline-heated and chitosan coatingson Mg–1Ca alloy sample (about 10 days’ validity).[4,25] Normally,

Figure 7. Morphologies of MG63 cells on (a) bare Mg–1Ca alloy, (b) MgF2-ccoated Mg–1Ca alloy.

Surf. Interface Anal. 2013, 45, 1217–1222 Copyright © 2013 Joh

implants should maintain their mechanical integrity before thereparative phase for strong healing union of fracture boneapproximately 3–4 months [26] avoiding the second fracture. Ifthis MgF2 layer alone cannot guarantee the maintenance of themechanical strength for that long, double layer could beprepared on magnesium alloys, and the MgF2 layer acts as aninterlayer between the substrate and the outer layer, sucn ashydroxyapatite.[27]

Cell experiments

The morphologies of MG63 and MC3T3-E1 cells cultured on thebare Mg–1Ca alloy sample and MgF2-coated Mg–1Ca alloysamples for 72 h are shown in Fig. 7. Difference in the responseof the cells to bare Mg–1Ca alloy sample and MgF2-coatedMg–1Ca alloy samples is obvious. The untreated Mg–1Ca alloysample suffered pitting corrosion attack, and cell adhesionprocesses were hindered. Only a few MG63 cells with unhealthyround shape and hardly any MC3T3-E1 cells can be seen on thebare Mg–1Ca surface. The MgF2 film effectively retard corrosionof the Mg–1Ca substrate and prevent alkalization of the culturemedium, so both MG63 and MC3T3-E1 cells exhibited healthyconfiguration with a cytoplastic and extended morphology andseveral pseudopodium contacting the sample surface.

It is reported that F� only showed cytotoxicity to UMR-106 cellsat high concentration (10�3M).[28] The F� released from MgF2coating could not reach that level since the low solubility ofMgF2 (0.13 g/l).

oated Mg–1Ca alloy and MC3T3 cells on (c) bare Mg–1Ca alloy, (d) MgF2-

n Wiley & Sons, Ltd. wileyonlinelibrary.com/journal/sia

1221

N. Li et al.

1222

Conclusions

MgF2 film was successfully deposited on the surface of Mg–1Caalloy by vacuum evaporation deposition method. The results ofimmersion test and electrochemical test indicated that theMgF2-coated samples exhibited much lower degradation ratethan the uncoated ones. The MgF2 coating induced calciumphosphate deposition on Mg–1Ca alloy. In the direct cell assay,MG63 cells and MC3T3-E1 cells adhered and spread well on thesurface of the MgF2-coated samples after 72 h culture.

Acknowledgements

This work was supported by the National Basic Research Programof China (973 Program) (Grant No. 2012CB619102), the NationalHigh Technology Research and Development Program of China(863 Program) under grant number 2011AA030101 and2011AA030103, the National Science Fund for DistinguishedYoung Scholars (Grant No. 51225101), the National NaturalScience Foundation of China (No. 31170909), the Research Fundfor the Doctoral Program of Higher Education under Grant No.20100001110011, theNatural Science Foundation of HeilongjiangProvince (ZD201012) and Guangdong Province innovation R&Dteam project (no. 201001C0104669453).

References[1] M. P. Staiger, A. M. Pietak, J. Huadmai, G. Dias, Magnesium and its

alloys as orthopedic biomaterials: A review, Biomaterials 2006, 27,1728–1734.

[2] Z. J. Li, X. N. Gu, S. Q. Lou, Y. F. Zheng, The development ofbinary Mg-Ca alloys for use as biodegradable materials within bone,Biomaterials 2008, 29, 1329–1344.

[3] L. C. Li, J. C. Gao, Y. Wang, Evaluation of cyto-toxicity and corrosionbehavior of alkali-heat-treated magnesium in simulated body fluid,Surf. Coat.Technol. 2004, 185, 92–98.

[4] X. N. Gu, W. Zheng, Y. Cheng, Y. F. Zheng, A study on alkaline heattreated Mg-Ca alloy for the control of the biocorrosion rate, ActaBiomater. 2009, 5, 2790–2799.

[5] C. L. Wen, S. K. Guan, L. Peng, C. X. Ren, X. Wang, Z. H. Hu, Character-ization and degradation behavior of AZ31 alloy surface modified bybone-like hydroxyapatite for implant applications, Appl. Surf. Sci.2009, 255, 6433–6438.

[6] J. N. Li, Y. Song, S. X. Zhang, C. L. Zhao, F. Zhang, X. N. Zhang, L. Cao,Q. M. Fan, T. T. Tang, In vitro responses of human bone marrowstromal cells to a fluoridated hydroxyapatite coated biodegradableMg-Zn alloy, Biomaterials 2010, 31, 5782–5788.

[7] H. X. Wang, S. K. Guan, X. Wang, C. X. Ren, L. G. Wang, In vitrodegradation and mechanical integrity of Mg-Zn-Ca alloy coated withCa-deficient hydroxyapatite by the pulse electrodeposition process,Acta Biomater. 2010, 6, 1743–1748.

[8] H. M. Wong, K. W. K. Yeung, K. O. Lam, V. Tam, P. K. Chu, K. D. K. Luk,K. M. C. Cheung, A biodegradable polymer-based coating tocontrol the performance of magnesium alloy orthopaedic implants,Biomaterials 2010, 31, 2084–2096.

[9] L. P. Xu, E. L. Zhang, K. Yang, Phosphating treatment and corrosionproperties of Mg-Mn-Zn alloy for biomedical application, J. Mater.Sci. Mater. Med. 2009, 20, 859–867.

wileyonlinelibrary.com/journal/sia Copyright © 2013 John

[10] Y. M. Wang, F. H. Wang, M. J. Xu, B. Zhao, L. X. Guo, J. H. Ouyang,Microstructure and corrosion behavior of coated AZ91 alloy bymicroarc oxidation for biomedical application, Appl. Surf. Sci. 2009,255, 9124–9131.

[11] X. N. Gu, N. Li, W. R. Zhou, Y. F. Zheng, X. Zhao, Q. Z. Cai, L. Ruan,Corrosion resistance and surface biocompatibility of a microarc oxi-dation coating on a Mg-Ca alloy, Acta Biomater. 2011, 7, 1880–1889.

[12] J. Z. Ilich, J. E. Kerstetter, Nutrition in bone health revisited: A storybeyond calcium, J. Am. Coll. Nutr. 2000, 19, 715–737.

[13] Position of The American Dietetic Association: The Impact ofFluoride on Health, J. Am. Diet. Assoc. 2000, 100, 1208–1213.

[14] S. F. Lamolle, M. Monjo, M. Rubert, H. J. Haugen, S. P. Lyngstadaas,J. E. Ellingsen, The effect of hydrofluoric acid treatment of titaniumsurface on nanostructural and chemical changes and the growthof MC3T3-E1 cells, Biomaterials 2009, 30, 736–742.

[15] T. Berglundh, I. Abrahamsson, J. P. Albouy, J. Lindhe, Bone healing atimplants with a fluoride-modified surface: an experimental study indogs, Clin. Oral Implant. Res. 2007, 18, 147–152.

[16] S. Verdier, N. van der Laak, S. Delalande, J. Metson, F. Dalard, Thesurface reactivity of a magnesium-aluminium alloy in acidic fluoridesolutions studied by electrochemical techniques and XPS, Appl. Surf.Sci. 2004, 235, 513–524.

[17] E. Gulbrandsen, J. Taftø, A. Olsen, The passive behaviour of Mg inalkaline fluoride solutions. Electrochemical and electron microscopi-cal investigations, Corros. Sci. 1993, 34, 1423–1440.

[18] A. Yamamoto, T. Terawaki, H. Tsubakino, Microstructures and corro-sion properties on fluoride treated magnesium alloy, Mater. Trans.2008, 49, 1042–1047.

[19] Y. Chen, Y. Song, S. Zhang, J. Li, H. Wang, C. Zhao, X. Zhang, Effect offluoride coating on in vitro dynamic degradation of Mg–Zn alloy,Mater. Lett. 2011, 65, 2568–2571.

[20] F. Witte, J. Fischer, J. Nellesen, C. Vogt, J. Vogt, T. Donath, F. Beckmann,In vivo corrosion and corrosion protection of magnesium alloy LAE442,Acta Biomater. 2010, 6, 1792–1799.

[21] M. Thomann, C. Krause, N. Angrisani, D. Bormann, T. Hassel,H. Windhagen, A. Meyer-Lindenberg, Influence of a magnesium-fluoride coating of magnesium-based implants (MgCa0.8) ondegradation in a rabbit model, J. Biomed. Mater. Res. A 2010, 93A,1609–1619.

[22] M. D. Pereda, C. Alonso, L. Burgos-Asperilla, J. A. del Valle, O. A.Ruano, P. Perez, M. A. F. L. de Mele, Corrosion inhibition of powdermetallurgy Mg by fluoride treatments, Acta Biomater. 2010, 6,1772–1782.

[23] A. Drynda, T. Hassel, R. Hoehn, A. Perz, F. W. Bach, M. Peuster, Devel-opment and biocompatibility of a novel corrodible fluoride-coatedmagnesium-calcium alloy with improved degradation kinetics andadequate mechanical properties for cardiovascular applications,J. Biomed. Mater. Res. A 2010, 93A, 763–775.

[24] T. T. Yan, L. L. Tan, D. S. Xiong, X. J. Liu, B. C. Zhang, K. Yang, Fluoridetreatment and in vitro corrosion behavior of an AZ31B magnesiumalloy, Mater. Sci. Eng. C-Mater. Biol. Appl. 2010, 30, 740–748.

[25] X. N. Gu, Y. F. Zheng, Q. X. Lan, Y. Cheng, Z. X. Zhang, T. F. Xi,D. Y. Zhang, Surface modification of an Mg-1Ca alloy to slow downits biocorrosion by chitosan, Biomed. Mater. 2009, 4, 044109.

[26] T. P. Rüedi, T. P. Ruedi, R. Buckley, C. G. Moran, AO Principles ofFracture Management, Thieme, 2007.

[27] J. H. Jo, B. G. Kang, K. S. Shin, H. E. Kim, B. D. Hahn, D. S. Park,Y. H. Koh, Hydroxyapatite coating on magnesium with MgF2interlayer for enhanced corrosion resistance and biocompatibility,J. Mater. Sci. Mater. Med. 2011, 22, 2437–2447.

[28] C. A. Grillo, F. Alvarez, M. A. Fernández Lorenzo de Mele, Biologicaleffects of magnesium particles degradation on UMR-106 cell line:Influence of fluoride treatments, Colloids Surf., B 2011, 88, 471–476.

Wiley & Sons, Ltd. Surf. Interface Anal. 2013, 45, 1217–1222