Microstructures and electrochemical properties of Mg 49 Ti 6 Ni (45-x) M x (M = Pd and Pt) alloy electrodes

Microstructures and electrochemical properties ofMg49Ti6Ni(45-x)Mx (M=Pd and Pt) alloy electrodesF.R. Nikkuni1, S.F. Santos2,*,† and E.A. Ticianelli1

1Instituto del Química de São Carlos, USP. Av. Trab. São-carlense, 400, CP 780, São Carlos, SP, CEP: 13560-970, Brazil2Centro de Engenharia, Modelagem e Ciências Sociais Aplicadas, UFABC, R. Santa Adélia, 166, Santo André, SP, CEP 09.210-170, Brazil

SUMMARY

Magnesium – nickel alloys have been considered an alternative for AB5 and AB2-type alloys in nickel–metal hydride bat-teries due to their larger maximum discharge capacities, but their low stability in alkaline solution has hindered their use incommercial cells. Aiming to improve the electrode performance of the Mg55Ni45 alloy, we investigated the simultaneousaddition of Ti and a noble metal (Pd and Pt) as alloying elements. The investigated system has general compositionMg49Ti6Ni(45-x)Mx, where M is Pd or Pt, and x assumed values of 0, 2.0 and 4.0 at.%. The electrochemical measurementsshowed that the Mg49Ti6Ni41Pd4 alloy has the best electrode performance among the studied alloys, reaching 431 mA h g�1

of maximum discharge capacity at the first cycle of charge/discharge. After 10 and 20 cycles, this alloy presented relativedischarge capacities of 84 and 77% of the initial one, respectively. The electrode performance of the investigated alloys isdiscussed in light of results of structural characterization by transmission electron microscopy and X-ray diffraction.Copyright © 2013 John Wiley & Sons, Ltd.

KEY WORDS

metal hydrides; nanostructured materials; mechanical alloying; Ni-MH batteries

Correspondence

*S.F. Santos, Centro de Engenharia, Modelagem e Ciências Sociais Aplicadas,UFABC, R. Santa Adélia, 166, Santo André – SP, CEP09.210-170, Brazil.†E-mail: [email protected]

Received 28 August 2012; Revised 22 October 2012; Accepted 25 November 2012

1. INTRODUCTION

Nickel – metal hydride (Ni-MH) batteries have been avail-able commercially since 1989, and their annual productionis over 1 billon of cells worldwide [1]. Despite this largeproduction, nowadays, Ni-MH batteries experience a hardcompetition with Li ion cells which have larger gravimetricand volumetric energy storage capacities. Moreover, thedevelopment of novel portable electronic devices, minia-turization and new technological products such as electricvehicles (EV) and hybrid EV impose a demand forbatteries with increasing energy densities [2,3]. Fulfillingthis demand is a major challenge for modern batteries.

Increase of the energy density of Ni-MH cells results inthe development of novel electrode materials capable todeliver high discharge capacities (large hydrogen storagecapacities) and presents fast charge/discharge reactionkinetics and large stability in alkaline electrolytes. Sincelate 90s, large discharge capacities have been reported forMg-Ni alloy electrodes [2]. Unfortunately, the practicaluse of these alloys in batteries is hindered by the fast decayof the discharge capacity over cycling. Thus, a number of

investigations started aiming to overcome the shortdurability of these alloy electrodes. From the materialsscience viewpoint, the several investigated approachescan be classified in two main groups: (i) modification ofthe surface chemical composition, and (ii) modificationof the bulk chemical composition of the alloy particles.Until now, the best results were obtained adopting thesecond approach.

Concerning bulk chemical modifications, several transi-tion metals have been probed as third elements in order tomodify the electrode performances of Mg-Ni alloys, suchas Ti, Nb, Cr, Zr, V, among others [4–6]. Promising resultswere reported for the ternary Mg-Ni-Ti alloys [7]. Theobserved improvement on cycle life performance in thesealloys has been ascribed to the formation of TiO2 on theparticles’ surface retarding the corrosion of magnesium.

Noble metals are also under investigation as alloyingelements in Mg-Ni alloy electrodes. Mg-Ni-Pd is an inter-esting alloy system which has shown improvements oncycling stability of the electrodes, but usually accompaniedby some decay of the maximum discharge capacity. Thebest electrochemical results were reported for the alloys

INTERNATIONAL JOURNAL OF ENERGY RESEARCHInt. J. Energy Res. 2013; 37:706–712

Published online 14 January 2013 in Wiley Online Library (wileyonlinelibrary.com). DOI: 10.1002/er.3008

Copyright © 2013 John Wiley & Sons, Ltd.706

with nearly 10 at.% of Pd [8,9]. Other noble metal investi-gated as alloying element for Mg-Ni alloy electrodes isplatinum. Souza et al. [10] reported that the Mg49Ni49Pt2alloy electrode exhibited the same maximum dischargecapacity obtained by the Mg50Ni50 alloy (364 mAh g�1)and superior cycling performance (nearly twice thedischarge capacity of the bare MgNi alloy after 10 cycles).

More recently, Mg-Ni-based quaternary alloys startedto be investigated for anode application in Ni-MHbatteries. The Mg0.9Ti0.1Ni1-xPdx alloy system was investi-gated by Tian et al. [11]. These authors observed animprovement on cycling durability accompanied by someprogressive decay of the maximum discharge capacity,when the Pd content was increased. The best cycling stabil-ity was obtained for Mg0.9Ti0.1Ni0.85Pd0.15, which retained76.8% of its initial discharge capacity after 20 charge/dis-charge cycles. This value was appreciably superior to thatobserved for the Mg0.9Ti0.1Ni alloy after the same numberof cycles (31.4%). However, the initial discharge capacityfor the Mg0.9Ti0.1Ni alloy was 455 mAh g�1, while forthe alloy with 0.15 of Pd (7.5 at.%), the capacity was only184.2 mAh g�1.

In a previous paper, we reported the results obtained forthe Mg(55-x)TixNi(45-x)Pty alloy system. The addition ofonly 2 at.% of Pt to the Mg55Ni45 bare alloy was enoughto increase both the maximum discharge capacity andcycle life performance of the alloy electrode. A synergeticeffect on the electrode performance was also observedwhen both Ti and Pt are added to the Mg-Ni alloy [12].In the present paper, we report the investigation of a newMg-Ni-based alloy system with nominal compositionMg49Ti6Ni(45-x)Mx, where M is a noble metal (Pt or Pd).The investigated alloys have their electrode performancesevaluated by repeated cycles of galvanostatic charge anddischarge with hydrogen and their microstructures inves-tigated by transmission electron microscopy (TEM) andX-ray diffraction (XRD). These experimental results werediscussed aiming to correlate the electrode performanceswith the respective microstructure characteristics.

2. EXPERIMENTAL PROCEDURES

The alloy nominal composition of Mg49Ti6Ni(45-x)Mx,where M is Pd or Pt, and x assumed values of 0, 2 and 4at. %. The amount of noble metal in these alloys was keptlow in order to limit the costs of raw material. A binaryMg55Ni45 was also processed to compare its propertieswith those of the investigated alloys. The alloys weresynthesized by mechanical alloying using a Spex 8000shaker mill using high purity elemental powders (purityover 99%) as raw material. The main processingparameters employed were ball to powder weigh ratio of10:1 and 6 h of milling.

The samples were characterized by XRD diffraction,using a Siemens D5005 diffractometer (Cu-Ka radiation),and TEM, using a Phillips CM 120 microscope (LaB6

filament) operating at 120 KV.

The electrochemical tests were carried out using a three-electrode cell configuration, with a Pt counter electrode, anHg/HgO reference electrode and a 6 mol/L KOH electro-lyte. The working electrodes were prepared by cold press-ing a mixture of 0.1 g of active material (alloy powder) and0.1 g of carbon black (Vulkan XC-72R) containing 33 wt.% of polytetrafluoroethylene (Teflon T-30, E.I. Dupont)binder in both sides of a 2 cm2 nickel screen. The electro-des were charged during 3 h with a current density of 200mA/g of active material and discharged with a current den-sity of 20 mA/g. The cut-off potential was fixed as �0.50V (vs. Hg/HgO, KOH 6 mol/L).

3. EXPERIMENTAL RESULTS ANDDISCUSSION

3.1. Discharge capacity and cycle lifeperformance

The discharge capacities versus cycling number of theinvestigated alloys are shown in Figure 1. All investigatedalloys exhibited the maximum discharge capacity at thefirst cycle indicating their activated states. Conversely,the values of maximum discharge capacity attained bythe alloys vary depending on their chemical composition.The Mg55Ni45 binary alloy attained the lowest electrodeperformance, as indicated by its maximum dischargecapacity of about 250 mA.h/g.

Figure 1 shows that the addition of Ti increases thedischarge capacity of the electrodes, as can be observed forthe Mg55Ti6Ni45 alloy. It is also observed that the additionof 2 at.% of noble metal further increases the dischargecapacity. In the case of the Mg49Ti6Ni43Pt2 alloy, themaximum discharge capacity increased from 250 (binaryalloy) to 445 mA.h/g. In the case of the Mg49Ti6Ni43Pd2alloy, this value was 457 mA.h/g. Increasing the content ofnoble metal in the alloys up to 4 at.% resulted in a slightdecrease of the maximum discharge capacity for theMg49Ti6Ni41Pd4 alloy. This decay was more marked for

Figure 1. Discharge capacity versus cycling number of theMg55Ni45 and Mg49Ti6Ni(45-x)Mx alloys.

Novel Mg49Ti6Ni(45-x)Mx electrode alloys F. R. Nikkuni, S. F. Santos and E. A. Ticianelli

707Int. J. Energy Res. 2013; 37:706–712 © 2013 John Wiley & Sons, Ltd.DOI: 10.1002/er

the Mg49Ti6Ni41Pt4 alloy. This trend is most likely to berelated to the decrease in relative weight of Mg (hydride-forming element) in the quaternary alloys when the amountof noble metal is increased from 2 to 4 at.% of noble metal.The decay of the maximum discharge capacity is lesspronounced for the alloy with Pd than that containing Pt,which can be related to both the lower atomic weight of Pdand the high hydrogen solubility in Pd. From technologicalviewpoint, keeping the amount of noble metal in the alloysas low as possible is important to limit the cost of rawmaterial.

The degradation rate of different alloy electrodesinvestigated can be better observed in Table I whichshows the maximum discharge capacity and the relative(retained) discharge capacity after 10 and 20 cycles.Herein, relative discharge capacity of the electrode isdefined as the discharge capacity after a certain numberof galvanostatic cycles of charge and discharge dividedby the maximum discharge capacity (in percentage).Mg55Ni45 and Mg49Ti6Ni45 alloys exhibited very lowcycling stability, as expected. After 10 cycles, the relativedischarge capacity was below 40% in both cases. The addi-tion of a noble metal (Pd or Pt) strongly affected thecycling stability of the alloy electrodes. Table I shows thatthe addition of only 2 at.% of noble metal was enough tosignificantly decrease the degradation rate of the electro-des. The Pt and Pd-containing alloys with 2 at.% of noblemetal have similar degradation rates. When the amount ofnoble metal is increased up to 4 at.%, the degradation ratefor the Pd-containing alloy electrode was reduced evenmore, but this trend was not followed by the alloy with Pt.

Considering both criteria maximum discharge capacityand cycling stability, the best electrodes’ performancewas accomplished by the Mg49Ti6Ni41Pd4 alloy (Table I).

3.2. Microstructures

The results of XRD are shown in Figure 2. As a generalfeature, all the investigated alloys presented highly broad-ened diffraction peaks. This behavior is related to the pres-ence of residual strain and reduction in crystallite sizes.Such feature is expected for materials synthesized bymechanical alloying. The Mg2Ni and MgNi2 intermediatephases were identified in all samples, and there is anevidence of Pt phase in the alloys containing this element.The presence of unalloyed Pt is indicated by the increase inintensity of the diffraction peak close to 40o which corre-sponds to (200) plane of Mg2Ni (Figure 2). In the alloys

containing Pt, this peak is overlapped to the (111) planeof Pt so increasing its intensity. This trend becomes morepronounced as the Pt content in the alloy raises. TheXRD patterns also indicate the presence of amorphousphase as indicated by an overlapping of broadened crystal-lite peaks and broad band around 35 to 55o.

Figures 3 to 6 show the TEM images of theMg49Ti6Ni45-xMx alloys. All bright field and dark fieldimages indicate that the alloys are composed by sub-micrometric particles, each one composed by small nano-crystals and an amorphous phase (in agreement to XRDresults in Figure 2). The phases Mg2Ni and MgNi2were identified in all the alloys through the selected areaelectron diffraction patterns (SAEDP) in agreement to theXRD results. In the SAEDP, there is an indication ofdiffraction rings which may be related to Pd and Pt phases,but in these cases, these diffraction rings are overlappedwith those of Mg2Ni and MgNi2 phases.

The values of mean crystallite sizes with respectivestandard deviations are shown in Table II, while thecrystallite size distribution of each alloy is shown inFigure 7.These values were obtained from the TEMimages in dark field mode. The values for mean crystal-lite sizes are ranging from 7.3 to 10.5 nm for the alloyscontaining noble metals, indicating that only 6 h of ballmilling was enough to obtain alloys with refined micro-structures. An interesting trend is that the alloys withsmaller mean crystallite sizes displayed the larger valuesof maximum discharge capacity. The alloys with smaller

Table I. Electrode performances of the investigated alloys.

AlloyMax. discharge capacity

(mA.h.g�1)Relative discharge capacity

after 10 cycles (%)Relative discharge capacity

after 20 cycles (%)

Mg55Ni45 250 37 24Mg49Ti6Ni45 441 38 26Mg49Ti6Ni43Pd2 457 69 59Mg49Ti6Ni41Pd4 431 84 77Mg49Ti6Ni43Pt2 445 66 55Mg49Ti6Ni41Pt4 376 69 59

Figure 2. XRD patterns of the investigated alloys.

Novel Mg49Ti6Ni(45-x)Mx electrode alloysF. R. Nikkuni, S. F. Santos and E. A. Ticianelli

708 Int. J. Energy Res. 2013; 37:706–712 © 2013 John Wiley & Sons, Ltd.DOI: 10.1002/er

crystallite sizes have larger fractions of grain boundariesand therefore more diffusion short-circuits favoringhydrogen diffusion. This feature of the microstructuremight be associated to the larger discharge capacitiessince the inner hydrogen atoms can reach more quicklythe surface of the particle. [13]. However, it is necessaryto beware regarding the implication of the average crys-tallite size on the discharge capacity of the alloy

electrodes since each alloy has its own standard devia-tion for this value as well as different distribution ofcrystallite sizes, as shown in Table II and Figure 7.

The microstructural characterization indicated that allinvestigated alloys are partially amorphous. This lowcrystallinity of the alloys increases its electrochemicalproperties of Mg-Ni based alloy electrodes, as depicted inSantos et al. [2] and references therein.

Figure 3. (a) Bright field and (b) dark field images and respective (c) SAEDP of the Mg49Ti6Ni43Pt2 alloy.

Figure 4. (a) Bright field and (b) dark field images and respective (c) SAEDP of the Mg49Ti6Ni41Pt4 alloy.

Figure 5. (a) Bright field and (b) dark field images and respective (c) SAEDP of the Mg49Ti6Ni43Pd2 alloy.

Figure 6. (a) Bright field and (b) dark field images and respective (c) SAEDP of the Mg49Ti6Ni41Pd4 alloy.



It is well known the fast degradation rate of thedischarge capacities of Mg-Ni alloy electrodes [2–5,14–17].Previous investigations [18] associate this behavior to thelow stability of Mg in alkaline solution, resulting in theformation of Mg(OH)2 on the particles’ surfaces. The forma-tion of Mg(OH)2 causes loss of active hydride forming mate-rial. Moreover, the formation of Mg hydroxide passive filmon the particles’ surface decreases the electron transferreaction kinetics at the particle/electrolyte interface.

In the case of ternary Mg-Ni-Ti, the slower degradationkinetics of these electrodes when compared to Mg-Ni oneswas ascribed to a preferential oxidation of Ti resulting onthe formation of a TiO2 passive film on the particles’surface. This film could retard the formation of magnesiumhydroxide [19,20]. Similar corrosion behavior wasobserved for binary Mg-Ti thin films [21,22]. Moreover,it is necessary to obtain a homogeneous distribution of Tielement throughout the microstructure to attain good corro-sion properties; otherwise, heterogeneous distribution of Ticould give rise to micro-galvanic corrosion pairs betweenTi-rich and Ti-poor regions [22].

In the present study, the partial substitution of Ti for Mg(Mg49Ti6Ni45 alloy) did not result in any appreciableimprovement on the cycle life performance of this alloyelectrode. This is probably due to the small amount of Tiadded to the alloy which may be too low to cause enrich-ment on the surface region of the particles with Ti.Conversely, substantial increase of the maximum dischargecapacity was obtained by the addition of Ti.

The Mg-Ni-Ti-Pd system was investigated by Tian et al.[11,23]. These authors synthesized fully amorphous Mg

(0.9�x)Ti0.1PdxNi alloys (x = 0.04 to 0.1) bymechanical alloy-ing and observed improvements on cycling stability with theaddition of Pd, comparing to the ternary Mg0.9Ti0.1Ni alloy.These authors ascribed these improvements to the formationof dense stable passive films on the particle surfaces, prevent-ing the Mg oxidation. Surface analyses indicated a decreaseof Mg oxidation when the Pd content increased and theformation of Mg(OH)2, NiO, PdO and TiO2 oxides on thesurface of the particles [23]. The authors ascribed thedecrease in Mg oxidation/hydroxidation to the formation ofthis complex oxide layer. The alloys investigated in the pres-ent study are also Mg-Ni alloys modified with Ti and a noblemetal (in our case Pd or Pt). Thus, the formation of a similarprotective layer on the surface of the particles is expected,too. Moreover, the addition of noble metals as Pt and Ptshould dislocate the corrosion potential of the investigatedalloys reinforcing their cathodic nature. Nakagawa et al.[24] reported that Ti-Pd and Ti-Pt alloys with small amountof noble metals (0.2% of Pd and 0.5% of Pt) had significantimprove on their corrosion resistances when compared tounalloyed Ti.

For the discharge capacity, Zhao and Ma [5] in theirrecent review paper on Ni-MH anode materials introducedsome data of hydrogen diffusion coefficients for Mg2Niand Mg-Ti-Ni-Pd alloys, where it is possible to observethat the Mg-Ni-Ti-Pd alloys have diffusion coefficients ofone order of magnitude higher than those of the binaryMg-Ni alloys. In addition the fast kinetics of the hydrogenoxidation process on Pt and Pd is a well known phenome-non, as seen in many investigations as for example [25,26].Hence, in this work, the increase on the discharge capacityobtained with the incorporation of Ti and a noble metal inthe alloy is the result of the improvement of the electroca-talytic activity of the alloys and faster hydrogen diffusivityinto the quaternary alloys.

Comparing the results of cycling stability of the electro-des containing Pd and Pt, it is remarkable the best resultsobtained by the Pd addition. As aforesaid, there areevidences of the presence of unalloyed Pt phase fromXRD results. This may indicate that Pd was more homoge-neously dispersed on the microstructure, promoting a moreeffective cathodic protection of the particles’ surfaces.

4. CONCLUDING REMARKS

In this paper, a new alloy series of the Mg-Ni-Ti-M system(M=Pd or Pt) was introduced. Simultaneous addition of Tiand noble metal improved the electrochemical propertiesof the alloy electrodes.

Comparing both Pd and Pt, the former element wasmore effective to improve the electrode performance ofthe alloys. Considering both maximum discharge capacityand cycling stability, the best results were attained by theMg49Ti6Ni41Pd4 alloy which accomplished a maximumdischarge capacity of 431 mA.h.g�1 and retained 84 and

Figure 7. Crystallite size distributions of the Mg49Ti6Ni(45-x)Mx alloys.

Table II. Average crystallite sizes of the Mg49Ti6Ni(45-x)Mx

alloys.

Alloy Average crystallite size (nm) Std. dev. (nm)

Mg49Ti6Ni43Pd2 7.3 2.5Mg49Ti6Ni41Pd4 9.4 5.2Mg49Ti6Ni43Pt2 8.9 4.7Mg49Ti6Ni41Pt4 10.5 13.6



77% of this initial discharge capacity after 10 and 20 cyclesof charge/discharge, respectively.

The microstructural characterization indicates a correla-tion between larger discharge capacities with smaller crys-tallite sizes. Moreover, the low crystallinity of the alloyelectrodes also collaborates to the electrochemical perfor-mance of the alloys.

ACKNOWLEDGEMENTS

The authors would like to thank the Brazilian institutionsCAPES, CNPq and FAPESP for supporting this work.

REFERENCES

1. Fetcenko MA, Ovshinsky SR, Reichman B, Young K,Fierro C, Zallen A, Mays W, Ouchi T. Recentadvances in NiMH battery technology. Journal ofPower Sources 2007; 165:544–551.

2. Santos SF, Ishikawa TT, Ticianelli EA. Fundamentalsand technological applications of hydrogen – absorb-ing Mg amorphous alloys. In Amorphous Materials:Research, Technology and Applications, JR Telle,NA Pearstine (eds). Nova Science: New York, 2009;219–237.

3. Notten PHL, Ouwerkerk M, van Hal H, Beelen D,Keur W, Zhuo J, Feil H. High energy density strate-gies: from hydride-forming materials research tobattery integration. Journal of Power Sources 2004;129:45–54.

4. Santos SF, de Castro JFR, Ishikawa TT, Ticianelli EA.Effect of transition metal additions on the electrochem-ical properties of a MgNi-based alloy. Journal ofAlloys and Compounds 2007; 434:756–759.

5. Zhao X, Ma L. Recent progress in hydrogen storagealloys for nickel/metal hydride secondary batteries.International Journal of Hydrogen Energy 2009;34:4788–4796.

6. Zhang YH, Zhao DL, Han XY, Li BW, Dong XP,Wang XL. Electrochemical characteristics of mechan-ical alloyed (Mg1�xZrx)2Ni (x = 0–0.1) electrodealloys. International Journal of Hydrogen Energy

2007; 32:2830–2835.7. Ruggeri S, Roué L, Huot J, Schulz R, Aymard L,

Tarascon JM. Properties of mechanically alloyedMg-Ni-Ti ternary hydrogen storage alloys for Ni-MHbatteries. Journal of Power Sources 2002; 112:547–556.

8. Yamaura SI, Kim HY, Kimura H, Inoue A, Arata Y.Electrode properties of rapidly solidified Mg67Ni23Pd10amorphous alloy. Journal of Alloys and Compounds

2002; 347:239–243.

9. Ma T, Hatano Y, Abe T, Watanabe K. Effects of Pdaddition on electrochemical properties of MgNi. Journalof Alloys and Compounds 2004; 372:251–258.

10. Souza EC, de Castro JFR, Ticianelli EA. A new elec-trode material for nickel-metal hydride batteries:MgNiPt alloy prepared by ball-milling. Journal ofPower Sources 2006; 160:1425–1430.

11. Tian QF, Zhang Y, Chu HL, Sun LX, Xu F, Tan ZC,Yuan HT, Zhang T. The electrochemical performancesof Mg0.9Ti0.1Ni1�xPdx (x = 0–0.15) hydrogen storageelectrode alloys. Journal of Power Sources 2006;159:155–1558.

12. de Castro JFR, Santos SF, Nikkuni FR, Ishikawa TT,Ticianelli EA. Structural and electrochemical charac-teristics of Mg(55-x)TixNi(45-y)Pty metal hydride elec-trodes. Journal of Alloys and Compounds 2010;498:57–61.

13. Fultz B, Frase HN. Grain boundaries of nanocrys-talline materials – their widths, compositions, andinternal structures. Hyperfine Interactions 2000;130:81–108.

14. Santos SF, de Castro JFR, Ishikawa TT, Ticianelli EA.Effect of mechanical coating with Ni and Ni–5% Al onthe structure and electrochemical properties of theMg–50% Ni alloy. Journal of Materials Science2008; 43:2889–2894.

15. Jiang JJ, Gasik M. An electrochemical investigationof mechanical alloying of MgNi-based hydrogenstorage alloys. Journal of Power Sources 2000;89:117–124.

16. Han SC, Goo NH, Jeong WT, Lee KS. Effects of Ti onthe cycle life of amorphous MgNi-based alloyprepared by ball milling. Journal of Alloys and

Compounds 2000; 306:219–226.17. Lee HY, Goo NH, Jeong WT, Lee KS. The surface

state of nanocrystalline and amorphous Mg2Ni alloysprepared by mechanical alloying. Journal of Alloys

and Compounds 2000; 313:258–262.18. Liu W, Lei Y, Sun D, Wu J, Wang Q. A study of the

degradation of the electrochemical capacity of amor-phous Mg50Ni50 alloy. Journal of Power Sources

1996; 58:243–247.19. Ruggeri S, Roué L, Huot J, Schulz R, Aymard L,

Tarascon JM. Properties of mechanically alloyedMg–Ni–Ti ternary hydrogen storage alloys forNi-MH batteries. Journal of Power Sources 2002;112:547–556.

20. Han SC, Lee PS, Lee JY, Zuttel A, Schlapbach L.Effects of Ti on the cycle life of amorphousMgNi-based alloy prepared by ball milling. Journalof Alloys and Compounds 2000; 306:219–226.

21. Baldwin KR, Bray DJ, Howard DG, Gardiner RW.Corrosion behavior of some vapour deposited



magnesium alloys. Materials Science and Technology1996; 12:929–944.

22. Mitchell T, Diplas S, Tsakiropoulos P. Characteriza-tion of corrosion products formed on PVD in situmechanically worked Mg–Ti alloys. Journal of Alloysand Compounds 2005; 392:127–141.

23. Tian QF, Zhang Y, Sun LX, Xu F, Tan ZC, Yuan HT,Zhang T. Effects of Pd substitution on the electroche-mical properties of Mg0.9�xTi0.1PdxNi (x=0.04–0.1)hydrogen storage alloys. Journal of Power Sources2006; 158:1463–1471.

24. Nakagawa M, Matsuya S, Udoh K. Effects of Fluorideand Dissolved Oxygen Concentrations on the CorrosionBehavior of Pure Titanium and Titanium Alloys. DentalMaterials Journal 2002; 21:83–92.

25. Lopes PP, Freitas KS, Ticianelli EA. Electrocatalysisof the hydrogen oxidation in the presence of CO onRhO2/C-supported Pt nanoparticles. ElectrochimicaActa 2010; 56:418–426.

26. Garcia AC, Paganin VA, Ticianelli EA. CO tolerance ofPdPt/C and PdPtRu/C anodes for PEMFC. Electro-chimica Acta 2008; 53:4309–4315.



Documents

Microstructures and electrochemical properties of Mg 49 Ti 6 Ni (45-x) M x (M = Pd and Pt) alloy electrodes