Electrodeposited hard-magnetic Fe50Pd50 nanowires from an ammonium-citrate-based bath

Journal of Alloys and Compounds 605 (2014) 71–79

Contents lists available at ScienceDirect

Journal of Alloys and Compounds

journal homepage: www.elsevier .com/locate / ja lcom

Electrodeposited hard-magnetic Fe50Pd50 nanowiresfrom an ammonium-citrate-based bath

http://dx.doi.org/10.1016/j.jallcom.2014.03.1560925-8388/� 2014 Elsevier B.V. All rights reserved.

⇑ Corresponding author. Address: Department for Nanostructured Materials,Jozef Stefan Institute, Jamova 39, 1000 Ljubljana, Slovenia. Tel.: +386 1 4773 945;fax: +386 1 4773 221.

E-mail address: [email protected] (D. Pecko).

Darja Pecko a,⇑, Kristina Zuzek Rozman a, Nina Kostevšek a, M. Shahid Arshad a, Boštjan Markoli b,Zoran Samardzija a, Spomenka Kobe a

a Department for Nanostructured Materials, Jozef Stefan Institute, Ljubljana, Sloveniab Department for Materials and Metallurgy, Faculty of Natural Science and Engineering, University of Ljubljana, Slovenia

a r t i c l e i n f o

Article history:Received 20 August 2013Received in revised form 24 February 2014Accepted 26 March 2014Available online 3 April 2014

Keywords:ElectrodepositionFe–Pd nanowiresL10 phaseMagnetic properties

a b s t r a c t

Fe–Pd nanowires were synthesised in anodic alumina templates by applying both potentiostatic andpulsed electrodeposition regimes. When using potentiostatic deposition, only fragmented nanowireswere obtained; however, the use of pulse deposition was shown to be effective for producing solid nano-wires. In order to achieve this, different on-times for the deposition and off-times between the pulses inan electrolyte with a constant concentration of Fe(III) and Pd(II) ions at pH 9 were employed. Homoge-neous nanowires with the composition Fe55±5Pd45±5, lengths of 2.5 lm and diameters of 200 nm weresynthesised under the following pulsed conditions Eon = �1.4 V, ton = 2 s and Eoff = �0.1 V, toff = 10 s for5000 cycles. The as-deposited nanowires had a fcc crystal structure and were magnetically soft(HC � 5 kA/m) with the easy axis of magnetization perpendicular to the long axes of the nanowires,mainly due to the dipolar coupling within the template. In order to promote the ordering into the L10

phase, annealing in the temperature range 400–700 �C for 1–9 h in Ar + 7% H2 was performed. The highestcoercivity of 122 kA/m was achieved by annealing at 600 �C for 5 h.

� 2014 Elsevier B.V. All rights reserved.

1. Introduction

One-dimensional nanostructures, such as nanowires, nanorodsand nanotubes, have been the topic of many investigations dueto their small size and unique and tuneable properties, all of whichmake them appropriate for a very broad range of applications, suchas electronic (in nano/micro-electro-mechanical systems), thermo-electric devices and for sensing in biology and chemistry [1–3].There have been many methods used for the synthesis of one-dimensional nanostructures, such as: chemical vapour deposition[4,5], vapour–liquid–solid (VLS) growth [6], lithographically pat-terned nanowire electrodeposition [7], reductive sulphidization[8], hydrothermal reduction [9], sol–gel [10] and template-assistedelectrodeposition. Template-assisted electrodeposition is, in com-parison to the other methods, highly selective, since it relies onelectron transfer, which is the fastest along the highest conductiv-ity path. Therefore, it offers the possibility to deposit differentnanostructures with a high technological potential. When using

nanoporous membranes it is possible to deposit and tailor thenanostructure’s properties with high aspect ratios that are denselyand continuously packed and possess a high crystallinity [11]. Themost commonly used membranes are polycarbonate (PC) andanodized alumina oxides (AAO). The AAO-based template has beenconsidered as an ideal template as it possesses many desirablecharacteristics, including tuneable pore dimensions, good mechan-ical strength, thermal stability and ordered nanopores with a highdensity (109–1011 cm2) [11,12]. Successful depositions of metalnanowires, such as Ni, Pd, Cu, [2,13–15] and different alloys,Co–Zn, Co–Cu, Cd–Se, Co–Pt, Fe–Pt [16–19] and Fe–Pd [3,20–23],have been performed in AAO templates. Nanostructures with anordered L10 phase have attracted a lot of attention in the pastfew years due to their large magnetocrystalline anisotropy alongthe c-axis of their tetragonal crystal structure, which makes themappropriate for high-density magnetic recording media [24–26].So far, many studies and investigations have been made on theCo–Pt and Fe–Pt systems because they have higher magnetocrys-talline anisotropies (6.6 and 4.9 MJ/m3) [25]. In comparison toCo–Pt and Fe–Pt, Fe–Pd (1.8 MJ/m3) [23] possesses a lower a mag-netocrystalline anisotropy; however, the Fe–Pd alloy exhibits alower disorder–order transition temperature, and the low cost ofPd in comparison with Pt makes this material interesting for

72 D. Pecko et al. / Journal of Alloys and Compounds 605 (2014) 71–79

potential applications in perpendicular magnetic recording media[25,27]. In this paper we have synthesised near-equiatomicFe55±5Pd45±5 nanowires using the template-assisted electrodeposit-ion method. The main challenge in the electrodeposition of alloysis to co-deposit two metals with a large difference in their standardreduction potential, as in the case of Fe2+ and Pd2+ (DE � 1.4 V).This can be achieved by using a complexing agent, which forms acomplex with the metal and shifts the reduction potentials of themetals closer together, thus making the co-deposition possible.In our case, the main problem lies in finding an appropriate com-plexing agent that would shift the reduction potential of palladium(E0

Pd2þ=Pd ¼ 0:95 V) towards the less-noble iron (E0Fe2þ=Fe ¼ �0:447 V),

while keeping the electrolyte chemically stable and preventing theprecipitation of heavy, soluble hydroxides of both metals. Untilnow there have been successful depositions performed from bathsbased on sulphosalicylic acid [20,25,28], ammonium tartrate [29]and ammonium citrate [24,30,31]. Here, ammonium citrate wasused to complex the iron (logK(Fe(II)Cit) = 15.5 and logK(Fe(III)-Cit) = 25.0 [32]) and the ammonia to complex the palladium(logb4[Pd(NH3)4]2+) = 29.6) [33]. Furthermore, for the depositionof nanowires an Fe(III)-based electrolyte was chosen, because ofthe higher stability of Fe(III)–citrate complexes in comparison withFe(II)–citrates [31] and to avoid the partial oxidation of Fe(II)–Fe(III) by dissolved oxygen at the anode during deposition thatcould lead to an unknown concentration of various species in thesolution [34]. Equiatomic Fe–Pd nanotubes have been successfullydeposited and characterized in our previous study [35]; however,the nanotube ordering was lost upon annealing due to the ther-mally unstable PC template, which made any discussion aboutthe magnetic properties measured parallel and perpendicular tothe long tube/wire axis impossible. In this study we synthesisednear-equiatomic 200-nm-thick Fe55±5Pd45±5 nanowires using thetemplate-assisted electrodeposition method in thermally stableAAO templates. The present study analyses and proposes therequirements (such as the electrolyte concentration, the depositionpotential and, subsequently, the pulse-plating conditions) for thesuccessful deposition of morphologically consistent and composi-tionally homogenous nanowires. Furthermore, equiatomicFe55±5Pd45±5 nanowires with a cubic fcc crystal structure weretransformed into the tetragonal fct or L10 ordered structure afterannealing at 600 �C, with a coercivity of HC = 122 kA/m. These goodmagnetic properties, low ordering temperature and the shortannealing times make Fe55±5Pd45±5 nanowires interesting for manyapplications in nano(micro)-electro-mechanical systems or as apotential material for perpendicular recording.

2. Experimental

The depositions of the Fe–Pd nanowires and the cyclic voltammetry measure-ments were carried out using a Gamry, Reference 600 potentiostat/galvanostatequipped with PHE200 software. A stable electrolyte was prepared from the300-mM FeCl3 (Sigma Aldrich, 99.9%), 3-mM PdCl2 (Sigma Aldrich, 99%), 0.4-M(NH4)2C6H6O7 (Sigma Aldrich, 99%) and 0.5-M NH3(aq) (Merck, 25%) for the pHadjustment to 9. The excess of ammonia and ammonium citrate ensured that allthe Pd(II) and Fe(III) in the electrolyte were present in the complex form: Pd wasconverted in the [Pd(NH3)4]2+ complex [36] and Fe in the Fe(III)–citrate complex.The abbreviation Fe(III)Cit will be used throughout the paper for the Fe(III)–citratecomplexes. The Fe–Pd nanowires were electrochemically deposited in the commer-cially available AAO membranes (Whatman, Anapore, pore diameter 200 nm). Theconductivity of the membranes was achieved by first sputtering the back side ofthe membrane with 20 nm of Pt, which was followed by electrochemically depos-ited Au at �0.7 V for 20 min (the back side of the pores was completely covered,to promote deposition from the top only). The AAO template was then glued withCu-tape on the Al-foil sheet and used as a cathode. A circular platinum mesh wasused as a counter electrode. The nanowires were deposited in the potential regimefrom �1.4 V to �1.6 V, measured versus a Ag/AgCl/3 M KCl//(SSCE) at room temper-ature in the potentiostatic and pulse-plating modes, where the pulse-plating condi-tions ton (2–120 s) and toff (10–240 s) were varied. The cyclic voltammetry (CV)measurements were recorded on the flat substrate (SiO2/Cr/Au) and on the AAOtemplate (with the same conductive layer (Pt/Au) as used for the deposition of

the nanowires) with a scanning rate of 50 mV/s. The as-deposited nanowires weresubsequently annealed in a reducing atmosphere (Ar + 7% H2) from 400 to 700 �Cfor 1–9 h. The Fe–Pd nanowires were characterized using a field-emission-gunscanning electron microscope (FEG-SEM) JEOL JSM-7600F equipped with an OxfordInstruments INCA system for energy-dispersive X-ray spectroscopy (EDXS) with a20-mm2 silicon drift detector (SDD). The chemical composition of the Fe–Pd nano-wires was determined by quantitative EDXS analyses, which were carried out withan SEM accelerating voltage of 6.5 kV and with a 100-s spectrum-acquisition time.The SDD efficiency/response was calibrated for the 6.5-kV experimental set-upusing pure silicon as a standard. The EDS spectra were quantified using the XPPmatrix-correction procedure and a stored-standards database for the Fe La andPd La spectral lines. A SEM accelerating voltage of 6.5 kV was sufficient to detectFe and Pd and also to perform a quantitative analysis of the Fe–Pd alloy using theFe La spectral line with an energy of 0.704 keV and the Pd La spectral line withan energy of 2.838 keV (with LIII edge at 3.187 keV). Consequently the overvoltageratio was kept at P2 for both elements (Fe, Pd), which is appropriate for the EDXSanalysis. The crystal structures of the as-deposited and annealed Fe–Pd nanowiresinside the alumina templates were investigated by XRD using a diffractometer (D4Endeavor, Bruker AXS, Karlsruhe, Germany) with Bragg–Brentano geometry, a CuKa radiation source and a Sol-X energy-dispersive detector. The magnetic measure-ments were performed using a vibrating-sample magnetometer (VSM LakeShore7003) with magnetic fields in the range ±1200 kA/m.

3. Results and discussion

3.1. Characterization of the electrochemical process

Prior to the electrodeposition, cyclic voltammetry measure-ments of single-metal-ion electrolytes containing 300-mM Fe(III)-Cit, 3 mM [Pd(NH4)2]2+ and an electrolyte containing both metalswere performed in order to find the optimum potentials for a suc-cessful deposition of the alloy. Before the CV measurements thesolution was de-aerated by bubbling with N2 gas for 15 min inorder to avoid any oxygen-reduction peaks.

The blue curve in Fig. 1a represents the CV of the 3-mM[Pd(NH3)4]2+ solution at pH 9 recorded on the AAO template, wherean increase of the cathodic current C1Pd at potential �0.3 V isobserved, and is attributed to the reduction of the [Pd(NH3)4]2+ +2e�? Pd0. The reduction of Pd is, up to a potential of �0.4 V, acti-vation controlled and is followed by a diffusion-controlled deposi-tion for potentials more negative than �0.4 V. An exponentialincrease in the cathodic current C1H at potentials more negativethan �1.0 V is observed and is attributed to the decomposition ofH2O, according to the reaction 2H2O + 2e�? H2 + 2OH� and conse-quently to hydrogen evolution reaction (HER). The red curve inFig. 1a represents the solution containing 300-mM Fe(III)Cit atpH 9 recorded on the AAO template, where one cathodic peakC1Fe (E = �0.7 V, jC = �0.9 mA) is observed and is attributed to thereduction of Fe(III)Cit to Fe(II)Cit�, which is continued with thereduction of Fe(II)Cit to elemental Fe together with HER, at poten-tials more negative then�1.0 V, C2Fe+H. The red curve in Fig. 1b rep-resents the CV of the complete electrolyte, containing both metals,300 mM Fe(III)Cit and 3 mM [Pd(NH3)4]2+

, recorded on the AAOtemplate. The cathodic peak C1Fe+Pd indicates a reduction consistingof two process, i.e., [Pd(NH3)4]2+ to Pd0 and at the same time theFe(III)Cit ions are reduced to Fe(II)Cit�, which is then further reducedto elemental Fe, together with HER, (C2Fe+H), at a potential morenegative than �0.9 V.

The blue curve in Fig. 1b represents the CV of the completeFe–Pd electrolyte recorded on the flat substrate. The main idea ofrecording on a flat substrate was to help us with the determinationof the reduction peaks, which are usually not well defined on a por-ous substrate [1]. One cathodic peak C1Fe+Pd is observed and a cur-rent increase C2Fe+H. In a comparison of both CV curves, recordedon the AAO template and the flat substrate, there are two majordifferences. The first noticeable difference is the current decreasein the CV of the AAO template, which may be due to the lower elec-trochemically available surface area of the AAO in comparison withthe flat substrate [13,37]. The other noticeable difference is the

Fig. 1. Cyclic voltammetry response electrolytes containing 3-mM [Pd(NH3)4]2+ (blue curve), 300-mM Fe(III)Cit (red curve) recorded on AAO template (a) and a completeelectrolyte 300-mM Fe(III)Cit and 3-mM [Pd(NH3)4]2+, recorded on AAO template (red curve) and flat substrate (blue curve) (b). All the electrolytes contain 0.4-M(NH4)2C6H6O7 with pH 9 and the measurements were recorded with a scanning rate of 50 mV/s with the potential measured versus SSCE. (For interpretation of the referencesto colour in this figure legend, the reader is referred to the web version of this article.)

D. Pecko et al. / Journal of Alloys and Compounds 605 (2014) 71–79 73

peak shift C1Fe+Pd to more positive potentials for the AAO template.This behaviour might be due to the higher electro-activity of Au inthe template in comparison with the flat substrate. Similar behav-iour was observed in the work of El-Deab and Ohsaka [38] wherethe overpotential for the electrochemical reduction of oxygenwas decreased when an electrode containing Au nanoparticleswas used instead of a Au bulk electrode. Because of the highHER, which overlaps with the reduction peaks, any determinationof the latter is challenging; however, we can assume that the co-deposition of Pd and Fe would be possible from a potential of�0.9 V onwards.

3.2. Electrodeposition of nanowires

Based on the CV results, the depositions were performed atpotentials more negative than �1.0 V. At potentials more positivethan �1.3 V the Fe–Pd nanowires were Pd rich; therefore, poten-tials more negative than �1.3 V were applied to boost the Fe con-tent in the nanowires. With the potentiostatic conditions (Fig. 2arepresents the nanowires deposited at �1.4 V for 10,000 s) onlydendrite-like structures were deposited and although manyparameters such as the deposition potential, the electrolyte con-centration and the buffer additives were varied, the formation ofhomogenous nanowires was not achieved [15,20,22]. Fig. 2b showsthe current versus time characteristics (i–t characteristics) of thenanowires deposited at �1.4 V. At the beginning of the depositiona high current is obtained (iC = 40 mA) due to the charging of theelectrical double layer, but then gradually it decreases to 15 mA,because of the slow mass transport through the deep and narrowpores. However, after 500 s of the deposition, a constant currentis observed, which is assigned to the nanowire growth undernearly steady-state conditions [2,22]. The main problem with thepotentiostatic deposition in high-aspect-ratio membranes is thatonly at the beginning, when the concentration gradient is thehighest, is the diffusion fast. With time a total overlapping ofthe diffusion fields outside of each pore is expected [39,40] andthe ions have to travel through a long pore to the bottom of thepore. These growth instabilities that occur when the growth rateis limited by the rate of diffusion of the ions to the reactioninterface result in a dendritic deposit structure and in non-homogeneous pore filling. Another parameter that can stronglycontribute to the non-homogeneous deposition is a simultaneoushydrogen evolution at the cathode, with hydrogen bubbles par-tially or totally blocking the pores and consequently disturbingthe diffusion and deposition into them. This can be avoided byusing pulse electrodeposition, which enables similar concentrationgradients at the beginning of each pulse and reduces the blocking

of the pores by hydrogen-gas evolution. It is a well-known tech-nique to compensate for the slow diffusion-driven transport inhigh-aspect-ratio membranes [2]. The main principle of pulsedeposition is that the metal is deposited during the ton (at Eon),whereas during the toff (at Eoff), which is also called the relaxationtime, the ion concentration recovers and all the gas bubbles formedduring the ton are removed from the reaction spot and from thepores. Based on that, the Eoff potential was chosen to be �0.1 V,because from the cyclic voltammetry no reaction in the cathodicpart appears at this potential. In Fig. 2c the nanowires depositedunder pulse-plating conditions with the potential Eon = �1.4 V for120 s and Eoff = �0.1 V for 240 s are shown and the i–t characteris-tic (only the first few pulses) is shown in Fig. 2d. It is clear thatthese nanowires are not completely solid, despite the use of thepulse-plating regime. The reason most probably lies in the too longdeposition time ton, which consequently leads to a depletion of themetal ions at the electrode, with hydrogen bubbles still being pres-ent at the reaction spot and in the pore channels. The inset ofFig. 2d shows the cathodic part of the second pulse, where it isclearly observed that the current drops and reaches a plateau after20 s of deposition. Therefore, for further experiments ton was dras-tically shortened in such a way that the deposition is finishedbefore the cathodic current decreases to the diffusion-limitedregion. Our experiments have shown that with a shorter ton, homo-geneous and solid nanowires were obtained (Fig. 2e) when thedeposition was performed at Eon = �1.4 V for ton = 2 s and Eoff =�0.1 V for toff = 10 s for 5000 cycles. The corresponding i–t charac-teristics (only the first few pulses) are shown in Fig. 2f. At thepotential Eon = �1.4 V, the cathodic current is lower (�9 – 7 mA)in comparison with the potentiostatic deposition (�40–15 mA) orthe pulse deposition with ton = 120 s (�38–8 mA), which is inagreement with the much shorter, but homogenous, nanowires.When the potential Eoff = �0.1 V was applied, at the beginningthe anodic current (iA � 2 mA) is observed, which decreases to zeroafter 5 s and remains zero until the next pulse. The anodic currentis a result of several processes; firstly, it can be attributed to thecharging of the double layer, which takes milliseconds and, sec-ondly, it can result in some dissolution/oxidation reactions. Disso-lution of the previously deposited metals is not desired, since it cancause an inhomogeneous wire morphology and some variations inthe composition. For the explanation of the dissolution behaviour adetailed electrochemical study would be necessary; however, webelieve this is beyond the scope of this study and it will be pub-lished elsewhere. Nevertheless, based on these experiments weassume that the anodic current responds to the some dissolutionof the Fe and possibly to the oxidation of Fe(II)Cit to Fe(III)Cit orto the oxidation of the adsorbed hydrogen in the Fe-Pd deposits.

Fig. 2. FEG-SEM images of the deposited Fe-Pd nanowires from the electrolyte: 300-mM Fe(III)Cit and 3-mM [Pd(NH3)4]2+ with pH 9 in the AAO with different depositionconditions with corresponding current transient i–t: (a and b) potentiostatic deposition at �1.4 V with i–t, (c and d) pulsed deposition at Eon = �1.4 V, ton = 120 s, Eoff = �0.1 Vtoff = 240 s with i–t, and (e and f) pulsed deposition at Eon = �1.4 V, ton = 2 s, Eoff = �0.1 V toff = 10 s with i–t. The inset in (d) shows the cathodic part of the second pulse.

74 D. Pecko et al. / Journal of Alloys and Compounds 605 (2014) 71–79

The dissolution of Pd at �0.1 V is not considered, since Pd did notdissolve even if potentials up to +1.0 V were applied. Therefore,despite some anodic current in the first 5 s in the second pulse,the morphology of the wires is homogenous; however, it mayinfluence the composition of the deposited nanowires, which willbe further explained in the manuscript.

Fig. 3 shows 2.5 lm-long nanowires from Fig. 2e. Fig. 3a showsthe nanowires’ tops, after polishing away the excess of the AAOtemplate, which allows us to determine the filling factor of theFe-Pd nanowires in the AAO template. The filling factor was calcu-lated to be nearly 97%. Fig. 3b and c shows free-standing nanowiresafter the dissolution of the alumina template in 10-M sodiumhydroxide for 3 h. The bottoms of the wires (�500 nm) are den-dritic due to the branched pores of the alumina template that hap-pen during the barrier-layer thinning procedure [41]. However,

above that branched part the wires are homogeneous with asmooth surface, uniformly long and nanocrystalline, as seen inFig. 3c, which shows the tops of the nanowires.

3.3. Characterization of deposited Fe–Pd nanowires

The curves presented in Fig. 4 shows the compositional varia-tions of the Fe content along the length of the Fe–Pd nanowireswith respect to the applied deposition potential Eon from �1.4 Vto �1.6 V, with all the other parameters being kept the same. Thegiven average/mean values were obtained from at least three ana-lysed nanowires that were selected at random. The nanowiresdeposited at a potential of �1.4 V are approximately 2.5 lm longwith an average composition of Fe55±5Pd45±5. Despite the pulse-plating regime, the composition of these wires varies along and

Fig. 3. FEG-SEM images of the Fe55±5Pd45±5 nanowires deposited at Eon = �1.4 V for 2 s and Eoff = �0.1 V for 10 s for 5000 cycles (a) top view of the Fe55±5Pd45±5 nanowires afterpolishing away the excess of AAO template, (b) side view and (c) top view of the free standing Fe55±5Pd45±5 nanowires after dissolution of the AAO template in 10-M NaOH for3 h.

Fig. 4. Variation of the Fe content in Fe-Pd nanowires with respect to the nanowirelength and to the applied deposition potential (Eon).

Fig. 5. The XRD pattern of the (a) as-deposited, (b) annealed at 600 �C for 1 h and (c)annealed at 700 �C for 1 h Fe55±5Pd45±5 nanowires.

D. Pecko et al. / Journal of Alloys and Compounds 605 (2014) 71–79 75

between the wires. Consequently, the largest scattering from theaverage values is present at the �1.4 V potential, as displayed bythe error bars, which correspond to a ±1 standard deviation ofthe measurements. In contrast to that, the nanowires depositedat more negative potentials have a more uniform composition.The nanowires deposited at Eon = �1.45 V are around 4 lm longwith a composition Fe74±2Pd26±2; the nanowires deposited at Eon =�1.5 V are around 5 lm long with a composition Fe75±3Pd25±3; andthe nanowires deposited at Eon = �1.6 V are 5.3 lm long with acomposition Fe84±4Pd16±4. The length of the wires increases withan increase of the applied potential.

The reason for the large variation in composition along thelength of the nanowires deposited at a potential of �1.4 V mightbe due to the small anodic current that appears in the second pulse(Eoff = �0.1 V), which could result from a dissolution/oxidationreaction, but in this case large variations in the composition shouldbe present in the deposited nanowires, regardless of the appliedEon potential, Therefore, the appearance of the anodic current atthe Eoff cannot be the major reason for the composition variation,but it is most probably due to the combination of two depositionreactions regimes at �1.4 V. As observed from the cyclic-voltam-metry studies (Fig. 1a, blue curve), the reduction of palladiumstarts at a potential of �0.2 V and it reaches a maximum currentat a potential of �0.3 V. For the reduction of Fe(II) due to the stronghydrogen evolution and the consequent overlapping with thereduction peaks indicating Fe(II), we can only assume that thereduction of Fe(II) into elemental Fe starts at a potential more neg-ative than �0.9 V (Fig. 1a, red curve). Furthermore, the reduction ofpalladium starts as activation controlled (potential more positivethan �0.4 V), but becomes controlled by diffusion at potentials

lower than �0.4 V. Therefore, at a potential of �1.4 V the deposi-tion of palladium is already in the diffusion-controlled regime;however, the deposition of Fe is most probably activation con-trolled, which results in an Fe content of 50 at.%. Moreover, whenthe applied potential is increased by only 0.05 V (Eon = �1.45 V),the composition changes by 20 at.% to Fe74±2Pd26±2. At this poten-tial the reduction of Fe reaches its maximum current and it entersinto the diffusion-limited zone, which is in agreement with thecomposition of the nanowires deposited at potentials of �1.45 Vonwards. Here, the composition becomes independent of theapplied potential, since the mass transport becomes the limitingfactor. A large variation in the composition of the wires depositedat �1.4 V is therefore most probably due to the combination of thetwo different reactions regimes: the diffusion-controlled deposi-tion of palladium and the activation-controlled deposition of iron.A similar gradient composition when two reaction regimes wereinvolved was observed by Llavona et al. [42] in Ni–Fe nanowires.

Nanowires deposited at �1.4 V were annealed up to 700 �C inorder to transform the low-anisotropy fcc Fe–Pd phase to highlyanisotropic fct Fe–Pd phase. The presence of different crystalphases upon annealing was monitored using the XRD.

Fig. 5 shows the XRD diffractograms of the as-deposited (a) andannealed Fe55±5Pd45±5 nanowires at 600 �C (Fig. 5b) and 700 �C(Fig. 5c) for 1 h. The XRD diffractograms were recorded for thenanowires embedded in the alumina template from the top ofthe membrane in order to avoid the high Au signal. The XRDdiffractogram of the as-deposited nanowires consists of the fccFe–Pd phase, with the most intensive (111) reflection and a(200) reflection. The ratio between (111) and (200) in comparisonwith the theoretical peak intensity ratio [43] is much smaller,indicating that the crystals have a preferred [111] orientation. Inaddition, the texture coefficient confirms the preferred crystalorientation as follows [44]:

Fig. 6. Hysteresis loops measured on an array of as-deposited Fe55±5Pd45±5

nanowires with the magnetic field applied parallel (red curve) and perpendicular(blue curve) to the long wire axis. (For interpretation of the references to colour inthis figure legend, the reader is referred to the web version of this article.)

76 D. Pecko et al. / Journal of Alloys and Compounds 605 (2014) 71–79

TCðhk lÞ ¼ Ihkl=Ihkl 1=NX

n Ihkl=Ihklð Þ� �

ð1Þ

where TC(hkl) is the texture coefficient of the hkl plane, I(hkl) isthe measured relative intensity of the (hkl) plane, I� is the relativeintensity of the corresponding plane and N is the reflection number.According to Eq. (1) the deviation of the texture coefficient (TC)from unity implies the preferred crystal orientation. The value ofTC = 1 represents the wires with randomly orientated crystals,while higher values indicate a preferred crystallographic orienta-tion perpendicular to the hkl plane [45]. For the as-depositedFe–Pd nanowires the TC (111), (200) are 1.42 and 0.58, respec-tively. This means that the wires have a weak preferred crystal ori-entation in the [111] direction and due to the recording of the XRDfrom the top of the membrane the [111] direction is parallel to thelong wire axis. Using the Scherrer equation we have calculated thatthe size of the crystallites of the as-deposited nanowires is less than10 nm. The diffractogram of the wires annealed at 600 �C for 1 hshows the development of the L10 phase with a distinctive (200)peak splitting into (200) and (002), together with the reflectionsof a-Fe (bcc-solid solution (SS) of Fe, Pd) and FePd3 (L12). The super-structural peak (110) of the L10 that should appear at 2H =33.026� is missing; however, the peak (220) at 2H = 68.997� (takenfrom the card PDF-021440) is also not visible, which may be due tothe [111] texture of the crystals in the nanowires [46,47]. Accord-ing to Eq. (1), the TC (111), (200), (002), (202) are 1.65, 0.85,0.83 and 0.64, which shows that the preferred orientation of thecrystals remains in the [111] direction even after annealing, withthe size of the crystallites increased to around 100 nm (calculatedusing Scherrer’s equation). After annealing at 700 �C for 1 h, the pre-dominant peaks correspond to the soft-magnetic L12, a-Fe (bcc-SSFe, Pd) and c(Fe, Pd) (fcc-SS Fe, Pd) phases, with some remains ofthe L10 phase observed and the size of crystallites increased toaround 200 nm. The calculated lattice parameter a of the a-Fe

(a = 2.87 Å) remains the same for both annealing temperatures,with similar reflection intensities, which indicates that the a-Fe

does not change upon annealing and it is presented in the sameamount for both annealed samples. Therefore, since the intensityof a-Fe also remains the same at 700 �C, it can be assumed that mostof the L10 phase separates into two phases, i.e., an Fe-rich phasec(Fe, Pd) (a = 3.592 Å) and a Pd-rich phase L12, after annealing at700 �C.

The major reason for the appearance of many phases in the dif-fractogram after annealing is the deviation in the compositionalong the wires (Fig. 4) and also between the wires Fe55±5Pd45±5.Due to the deviation in the composition we are moving on the xaxis of the Fe–Pd phase diagram [48,49] and according to Fig. 4,where the composition along the wire and between the wireschanges from the 37–65 at.% of Fe, many phases can form uponannealing. Therefore, the appearance of both aFe and c(Fe,Pd)(enriched with Pd) is not surprising at all. By annealing at 600 �Cand varying the chemical composition between 37 and 65 at.%Fe, we would expect the coexistence of aFe (enriched with Pd)and L10 (FePd) whenever there is less than 50 at.% of Pd dissolvedin the as-deposited nanowires. The appearance of the a-Fe withannealing in Fe-Pd in a similar composition was also observed by[24,26,50]. At 700 �C the amount of Pd within aFe should grow asthe cðFe;PdÞð46 at:% PdÞ aFeð1:0 at:% PdÞ þ FePdð48:5 at:% PdÞ

� �eutectoid

reaction is reversed leading to the simultaneous appearance of aFe

and c(Fe, Pd). Whenever there is more than 50.0 and less thanapprox. 60.0 at.% Pd dissolved in the as-deposited nanowires, weare well within the one-phase region of L10 (FePd) of the Fe–Pdphase diagram as long as the annealing temperatures are below700 �C. When the annealing temperatures are close to 700 �C andthe Pd content is between 48.5 and approx. 54.0 at.% we findourselves on the very border of the two-phase region of(c(Fe, Pd) + L10), which leads to the simultaneous coexistence of

these two phases, as in our case, Fig. 5c. On the other hand, whenthere is more than 60 at.% of Pd dissolved in the as-deposited nano-wires the formation of L12 upon annealing is expected, and wasalready observed in our previous study [35]. Furthermore, a com-parison of the lattice parameters of pure metals, i.e., a-Fe(a = 2.861 Å, taken from card PDF-870722), c-Fe (a = 3.57 Å) andfcc-Pd (a = 3.94 Å) [51] with the a-Fe and c(Fe, Pd) from the XRDdiffractogram, confirms the formation of Fe-Pd solid solutions afterannealing. The lattice parameters for a-Fe and c(Fe, Pd) areenlarged due to the incorporation of Pd, with a larger atomic radiusin the crystal structure of the a-Fe and c-Fe. However, the changesin the lattice parameters are small, which indicates that theamount of dissolved Pd in the crystal structure of both iron allo-tropes is small and is not completely in agreement with the equi-librium phase diagram. This is not surprising as the diffusioncoefficient for the Pd is much smaller than that of the Fe at variousannealing temperatures and times leading to non-equilibrium orsmaller amounts of Pd being dissolved in the iron allotropes andthus to deviations from the equilibrium constitution. Finally, itshould be pointed out that the phase diagram describes the equi-librium conditions, but due to the non-equilibrium conditions(annealing and high cooling rates) and non-uniform chemical com-position of nanowires deviations are to be expected.

Fig. 6 shows the hysteresis loop of the as-deposited Fe–Pdnanowires inside the alumina template with the composition ofFe55±5Pd45±5. The hysteresis loops were measured with a magneticfield parallel and perpendicular to the nanowire’s long axes. Theas-deposited nanowires show soft-magnetic behaviour measuredin both field directions with low coercivities Hck = 5 kA/m andHc? = 4 kA/m. The overall easy magnetization axis was found tolie perpendicular to the nanowire’s long axis, which is apparentfrom the shape of the hysteresis loops. According to the mean-fieldmodel [52], the magnetic behaviour and the distribution of themagnetization easy axis of the nanowires in an array depends onthe effective anisotropy. This model yields a demagnetizing factorparallel to the wire axis equal to the porosity P of the membranesand a factor equal to (1 � P)/2 [53] in the perpendicular direction.The porosity of the template in the case of infinitely long wires(aspect ratio >10), can be calculated as P = a2/(a + s)2, where a isthe side length of a single square wire and s is the distancebetween the wires [52]. For the case of the as-deposited fccFe–Pd nanowires with a diameter of 200 nm, the wires’ separation

Fig. 8. Hysteresis loops measured on an array of annealed Fe55±5Pd45±5 nanowiresat 600 �C for 1 h in Ar + 7% H2 with the magnetic field applied parallel (red curve)and perpendicular (blue curve) to the wire axis. The inset shows the dependence ofcoercivity on the annealing time at 600 �C. (For interpretation of the references tocolour in this figure legend, the reader is referred to the web version of this article.)

D. Pecko et al. / Journal of Alloys and Compounds 605 (2014) 71–79 77

of 100 nm and a filling factor of nearly 97% the calculated porosityP is �40%. Taking into account this porosity the demagnetizationfactor in the direction parallel to the nanowires is 0.4, whereas inthe perpendicular direction the demagnetization factor is 0.3, indi-cating the parallel and perpendicular hysteresis curves in Fig. 6should be very close. The observed disagreement might be attrib-uted to the additional anisotropy, i.e., the magnetocrystallineanisotropy (MCA) that modifies the hysteresis loop. A similar phe-nomenon was reported by Forster et al. [54], they concluded that itis important to take into consideration the specific arrangement ofthe magnetocrystalline anisotropy in the nanowires. Thus, the ori-entation of the MCA easy axis with respect to the nanowire axisgreatly affects the hysteresis properties of the nanowires. In orderto clarify the contribution of the MCA to the magnetic behaviour ofthe Fe–Pd nanowires, we have to estimate its contribution to theaverage effective anisotropy, which depends on the angle of themagnetocrystalline easy axis to the long nanowire axis. Theobserved texturing of crystallites in the [111] (TC = 1.45, Fig. 5a)direction was found to be parallel to the long nanowire axis.According to the literature [55] the magnetocrystalline easy axisof the fcc Fe-Pd is in the direction [111] and thus making an anglewith the [111] of 35.3�. Recently Ivanov et al. [56] simulated thatwhen the angle of the magnetocrystalline easy axis with the nano-wire axis is smaller than 70� the MCA reinforces the shape anisot-ropy and when the angle is higher the MCA competes with theshape anisotropy. Therefore, in accordance with Ivanov’s et al.[56] simulations the shape of the hysteresis loops of the Fe–Pdnanowires is not affected by the influence of the MCA, but this dis-agreement might be due to some irregularities in the nanowires,the roughness of the nanowires, which can modify the dipolarenergy and therefore influence the demagnetization factor in twodirections, as was observed by Dahmane et al. [53].

In order to improve the coercivity of the as-depositedFe55±5Pd45±5 nanowires, they were annealed at temperatures from400 to 700 �C for 1–9 h in forming gas. The coercivity dependenceon the annealing temperature is shown in Fig. 7. The coercivity is�10 kA/m in both measured directions, when temperatures lowerthan 500 �C were used. However, a higher coercivity Hc? = 22 kA/mand Hck = 28 kA/m was achieved when annealing up to 550 �C. Thehighest coercivity of Hc? = 118 kA/m and Hck = 98 kA/m wasachieved after annealing at 600 �C for 1 h, with isotropic magneticbehaviour (Fig. 8). Increasing the annealing temperature further to700 �C results in a sharp decrease in the coercivity in bothmeasured directions. With the intention to improve the coercivity

Fig. 7. Temperature dependence of coercivity for an array of the Fe55±5Pd45±5

nanowires, which were annealed for 1 h in a reducing atmosphere (Ar + 7% H2).

even further, the annealing time at 600 �C was optimized. The insetof Fig. 8 shows the dependence of HC on the annealing time (1–9 h)at 600 �C in forming gas. After annealing for 5 h, the HC slightlyimproved to 122 kA/m, but after annealing for 9 h the coercivitystarted to decrease. Thus, for the Fe–Pd nanowires the maximumcoercivity was obtained at 600 �C with 5 h of annealing.

To summarize, the as-deposited and annealed up to 500 �CFe55±5Pd45±5 nanowires exhibit low coercivities due to the strongdipolar interactions that completely suppress the shape anisotropy(2pMS � 6500 kA/m, Ms = 1040 emu/cm3) [57]. Furthermore,annealing at 550 �C results in a coercivity increase, because ofthe L10 ordering (280 kA/m [35]). With a further increase of theannealing temperature to 600 �C, the ordering in the L10 phase getsmore pronounced (observed with the XRD, Fig. 5b), giving rise tocoercivity in both field directions. The isotropic magnetic behav-iour is expected from the texture in the [111] direction, wherethe magnetocrystalline easy axis of the fct structure [111] is tiltedfor the c/a = 0.96 by 52.6� [58]. Furthermore, the formation of theL10 phase is triggered and governed by the annealing temperatureand time, increasing the temperature above 600 �C and the time ofthe annealing (t > 5 h), results in a sharp decrease in the Hc, due tothe formation of soft-magnetic phases L12, c(Fe, Pd) and the signif-icant grain growth as calculated from the XRD results. Since themagnetization reversal in Fe–Pd nanowires most probablyproceeds through a mechanism of nucleation and wall propaga-tion, thus in larger grains (100–200 nm in our case) there are fewergrain boundaries resulting in less reversal-magnetization domain-pinning centres, which lead to lower coercivity. Additionally, theprobability of reverse-domain nucleation is higher due to a largersurface and more defects where the reverse domains normallystart to nucleate, as was also observed in our previous studies onCo–Pt nanotubes [59] and Fe–Pd thin films [60] and nanotubes[35].

4. Conclusions

The synthesis of Fe–Pd nanowires was investigated from theelectrochemical point of view. Potentiostatic deposition (due tothe discrepancy in the kinetics, i.e., the activation and the masstransport) was found not to be appropriate for achieving solidnanowires; however, a pulse-plating regime with a deposition time

78 D. Pecko et al. / Journal of Alloys and Compounds 605 (2014) 71–79

of ton = 2 s and an equilibrium time toff = 10 s was shown to be themost effective for the deposition of morphologically continuousand compositionally uniform Fe–Pd nanowires. The compositionof the nanowires varied with the applied deposition potential. Atlow over-potentials (�1.4 V) a near-equiatomic compositionFe55±5Pd45±5 was observed, with a large deviation in the composi-tion attributed to the mixed activation and diffusion regimes.Because of the change in the reaction regime to the mass transportlimited reaction, the applied potentials above �1.45 V resulted inFe-rich nanowires with a consistent composition, which ispotential independent. Morphologically consistent equiatomicFe55±5Pd45±5 nanowires exhibit a soft-magnetic behaviour with aneasy axis aligned perpendicular to the nanowire axis, due the highporosity of the AAO template leading to a large contribution of thedipolar interactions. The coercivity significantly increases afterannealing at 600 �C for 1 h in forming gas due to the ordering intothe L10 phase, which was confirmed by the XRD. The highest coer-civity HC = 122 kA/m was achieved after annealing at 600 �C for 5 h.Annealing above 600 �C resulted in a coercivity decrease, becauseordering into the L10 phase was accompanied by other side reac-tions, like the formation of soft-magnetic phases and grain growth.

Acknowledgments

This work was supported by Slovenian Research Agency (ARRS).Special thanks also to Medeja Gec for the preparation of the FEGSEM sample.

References

[1] S. Valizadeh, J.M. George, P. Leisner, L. Hultman, Electrochemical deposition ofCo nanowire arrays; quantitative consideration of concentration profiles,Electrochim. Acta 47 (2001) 865–874.

[2] K. Nielsch, F. Muller, A.P. Li, U. Gosele, Uniform nickel deposition into orderedalumina pores by pulsed electrodeposition, Adv. Mater. 12 (2000) 582–586.

[3] N. Tas�altın, S. Öztürk, N. Kılınç, H. Yüzer, Z.Z. Öztürk, Fabrication of Pd–Fenanowires with a high aspect ratio by AAO template-assistedelectrodeposition, J. Alloys Comp. 509 (2011) 3894–3898.

[4] S. Hofmann, C. Ducati, R.J. Neill, S. Piscanec, A.C. Ferrari, J. Geng, R.E. Dunin-Borkowski, J. Robertson, Gold catalyzed growth of silicon nanowires by plasmaenhanced chemical vapor deposition, J. Appl. Phys. 94 (2003) 6005–6012.

[5] C.-W. Hsu, A. Ganguly, C.-P. Chen, C.-C. Kuo, P.P. Paskov, P.O. Holtz, L.-C. Chen,K.-H. Chen, Optical properties of functionalized GaN nanowires, J. Appl. Phys.109 (2011) 053523–053526.

[6] K.W. Kolasinski, Catalytic growth of nanowires: vapor–liquid–solid, vapor–solid–solid, solution–liquid–solid and solid–liquid–solid growth, Curr. Opin.Solid State Mater. Sci. 10 (2006) 182–191.

[7] E.J. Menke, M.A. Thompson, C. Xiang, L.C. Yang, R.M. Penner, Lithographicallypatterned nanowire electrodeposition, Nat. Mater. 5 (2006) 914–919.

[8] X. Fu, H. Zou, L. Zhou, A novel synthesis route of Ag2S nanotubes by sulfidizingsilver nanowires in ambient atmosphere, J. Nanosci. Nanotechnol. 10 (2010)5851–5856.

[9] Z. Liu, Y. Yang, J. Liang, Z. Hu, S. Li, S. Peng, Y. Qian, Synthesis of coppernanowires via a complex-surfactant-assisted hydrothermal reduction process,J. Phys. Chem. B 107 (2003) 12658–12661.

[10] N.I. Kovtyukhova, T.E. Mallouk, T.S. Mayer, Templated surface Sol–Gelsynthesis of SiO2 nanotubes and SiO2-insulated metal nanowires, Adv.Mater. 15 (2003) 780–785.

[11] H. Schlörb, V. Haehnel, M.S. Khatri, A. Srivastav, A. Kumar, L. Schultz, S. Fähler,Magnetic nanowires by electrodeposition within templates, Phys. Status SolidiB 247 (2010) 2364–2379.

[12] H. Liu, F. Wang, Y. Zhao, J. Liu, K.C. Park, M. Endo, Synthesis of iron–palladiumbinary alloy nanotubes by template-assisted electrodeposition from metal-complex solution, J. Electroanal. Chem. 633 (2009) 15–18.

[13] K.S. Napolskii, I.V. Roslyakov, A.A. Eliseev, D.I. Petukhov, A.V. Lukashin, S.-F.Chen, C.-P. Liu, G.A. Tsirlina, Tuning the microstructure and functionalproperties of metal nanowire arrays via deposition potential, Electrochim.Acta 56 (2011) 2378–2384.

[14] N.J. Gerein, J.A. Haber, Effect of ac electrodeposition conditions on the growthof high aspect ratio copper nanowires in porous aluminum oxide templates, J.Phys. Chem. B 109 (2005) 17372–17385.

[15] K. Kim, M. Kim, S.M. Cho, Pulsed electrodeposition of palladium nanowirearrays using AAO template, Mater. Chem. Phys. 96 (2006) 278–282.

[16] M.A. Kashi, A. Ramazani, Z. Fallah, Magnetic properties improvement throughoff time between pulses and annealing in pulse electrodeposited CoZnnanowires, J. Alloys Comp. 509 (2011) 8845–8849.

[17] M. Almasi Kashi, A. Ramazani, F.A. Najafabadi, Z. Heydari, Controlled Cucontent of electrodeposited CoCu nanowires through pulse features andinvestigations of microstructures and magnetic properties, Appl. Surf. Sci. 257(2011) 9347–9350.

[18] C.-M. Shen, X.-G. Zhang, H.-L. Li, DC electrochemical deposition of CdSenanorods array using porous anodic aluminum oxide template, Mater. Sci.Eng., A 303 (2001) 19–23.

[19] Y.H. Huang, H. Okumura, G.C. Hadjipanayis, D. Weller, CoPt and FePtnanowires by electrodeposition, J. Appl. Phys. 91 (2002) 6869–6871.

[20] V. Haehnel, S. Fähler, L. Schultz, H. Schlörb, Electrodeposition of Fe70Pd30

nanowires from a complexed ammonium–sulfosalicylic electrolyte with highstability, Electrochem. Commun. 12 (2010) 1116–1119.

[21] H. Hu, C. Yang, J. Chen, G. Wu, Magnetic properties of Fe0.95Pd0.05 nanowirearrays, J. Magn. Magn. Mater. 320 (2008) 2305–2309.

[22] B. Jeon, S. Yoon, B. Yoo, Electrochemical synthesis of compositionallymodulated FexPd1�x nanowires, Electrochim. Acta 56 (2010) 401–405.

[23] X.L. Fei, S.L. Tang, R.L. Wang, H.L. Su, Y.W. Du, Fabrication and magneticproperties of Fe–Pd nanowire arrays, Solid State Commun. 141 (2007) 25–28.

[24] S.C. Hernández, B.Y. Yoo, E. Stefanescu, S. Khizroev, N.V. Myung,Electrodeposition of iron–palladium thin films, Electrochim. Acta 53 (2008)5621–5627.

[25] F.M. Takata, G. Pattanaik, W.A. Soffa, P.T.A. Sumodjo, G. Zangari, Synthesis ofL10 Fe–Pd films by electrodeposition and thermal annealing, Electrochem.Commun. 10 (2008) 568–571.

[26] H.-P. Lin, J.-C. Kuo, Texture and magnetic properties of electrodeposited FePdfilms, Mater. Lett. 65 (2011) 3537–3539.

[27] C. Issro, W. Puschl, W. Pfeiler, R. Rogl, W.A. Soffa, M. Acosta, G. Schmerber, R.Kozubski, V. Pierron-Bohnes, Temperature-driven changes of order andmagnetism in FePd thin films and thin foil, Scripta Mater. 53 (2005) 447–452.

[28] K.J. Bryden, J.Y. Ying, Electrodeposition synthesis and hydrogen absorptionproperties of nanostructured palladium–iron alloys, Nanostruct. Mater. 9(1997) 485–488.

[29] S. Doi, F. Wang, K. Hosoiri, T. Watanabe, Preparation and characterization ofelectrodeposited Fe–Pd binary alloy film, Mater. Trans. 44 (2003) 649–652.

[30] M. Rezaei, M. Ghorbani, A. Dolati, Electrochemical investigation ofelectrodeposited Fe–Pd alloy thin films, Electrochim. Acta 56 (2010) 483–490.

[31] D. Liang, J.J. Mallett, G. Zangari, Underpotential codeposition of Fe–Pt alloysfrom an alkaline complexing electrolyte: electrochemical studies, J.Electrochem. Soc. 158 (2011) D149–D157.

[32] A. Ringbom, Complexation in Analytical Chemistry, in: E. Robert (Ed.), KriegerPublishing Company, 1979.

[33] E.M. Baumgärtner, R.D. Gabe, Palladium–iron alloy electrodeposition. Part I.Single metal systems, Inst. Met. Finish. (2000).

[34] D.R.G. M.E. Baumgärtner, Palladium-iron alloy electrodeposition. Parti I. Singlemetal systems, 78, 2000, pp. 11–16.

[35] K.Z. Rozman, D. Pecko, L. Suhodolcan, P.J. McGuiness, S. Kobe, Electrochemicalsyntheses of soft and hard magnetic Fe50Pd50-based nanotubes and theirmagnetic characterization, J. Alloys Comp. 509 (2011) 551–555.

[36] A. Narayan, L. Landstrom, M. Boman, Laser-assisted synthesis of ultra smallmetal nanoparticles, Appl. Surf. Sci. 208–209 (2003). 5–5.

[37] C. Xiaoli, Z. Qiang, L. Zhizhou, S. Ziying, J. Zhiyu, Cyclic voltammetry as a tool toestimate the effective pore density of an anodic aluminium oxide template,Nanotechnology 18 (2007) 215701.

[38] M.S. El-Deab, T. Ohsaka, Hydrodynamic voltammetric studies of the oxygenreduction at gold nanoparticles-electrodeposited gold electrodes, Electrochim.Acta 47 (2002) 4255–4261.

[39] C.G. Zoski, Handbook of Electrochemistry, Elsevier, Oxford, 2007.[40] N. Kostevšek, K.Z. Rozman, D. Pecko, B. Pihlar, S. Kobe, A comparative study of

the electrochemical deposition kinetics of iron–palladium alloys on a flatelectrode and in a porous alumina template, Electrochim. Acta. (2014).

[41] C.T.d. Sousa, Development of nanoporous alumina template forbiotechnological applications, Ph.D. Dissertation, 2011.

[42] Á. Llavona, L. Pérez, M.C. Sánchez, V. de Manuel, Enhancement of anomalouscodeposition in the synthesis of Fe–Ni alloys in nanopores, Electrochim. Acta106 (2013) 392–397.

[43] L. Yiping, G.C. Hadjipanayis, C.M. Sorensen, K.J. Klabunde, Structural andmagnetic properties of ultrafine Fe–Pd particles, J. Appl. Phys. 75 (1994) 5885–5887.

[44] E.A. Dalchiele, R.E. Marotti, A. Cortes, G. Riveros, H. Gómez, L. Martínez, R.Romero, D. Leinen, F. Martin, J.R. Ramos-Barrado, Silver nanowireselectrodeposited into nanoporous templates: study of the influence of sizeson crystallinity and structural properties, Physica E 37 (2007) 184–188.

[45] J. Liu, J.L. Duan, M.E. Toimil-Molares, S. Karim, T.W. Cornelius, D. Dobrev, H.J.Yao, Y.M. Sun, M.D. Hou, D. Mo, Z.G. Wang, R. Neumann, Electrochemicalfabrication of single-crystalline and polycrystalline Au nanowires: theinfluence of deposition parameters, Nanotechnology 17 (2006) 1922.

[46] R. Fathi, S. Sanjabi, N. Bayat, Synthesis and characterization of NiMn alloynanowires via electrodeposition in AAO template, Mater. Lett. 66 (2012) 346–348.

[47] J. Mallet, K. Yu-Zhang, S. Matefi-Tempfli, M. Matefi-Tempfli, L. Piraux,Electrodeposited L1(0) CoxPt1�x nanowires, J. Phys. D: Appl. Phys. 38 (2005)909–914.

[48] H. Okamoto, Desk Handbook, Phase Diagram for Binary Alloys, ASMInternational, Materials Park, OH 44073-0002, United States of America, 2000.

[49] D.E. Laughlin, K. Srinivasan, M. Tanase, L. Wang, Crystallographic aspects of L10

magnetic materials, Scripta Mater. 53 (2005) 383–388.

D. Pecko et al. / Journal of Alloys and Compounds 605 (2014) 71–79 79

[50] D.S. Li, H. Garmestani, S.-S. Yan, M. Elkawni, M.B. Bacaltchuk, H.J. Schneider-Muntau, J.P. Liu, S. Saha, J.A. Barnard, Effects of high magnetic field annealingon texture and magnetic properties of FePd, J. Magn. Magn. Mater. 281 (2004)272–275.

[51] E. Burzo, P. Vlaic, Magnetic properties of iron–palladium solid solutions andcompounds, J. Optoelectron. Adv. Mater. 12 (2010) 1869–1878.

[52] A. Encinas-Oropesa, M. Demand, L. Piraux, I. Huynen, U. Ebels, Dipolarinteractions in arrays of nickel nanowires studied by ferromagneticresonance, Phys. Rev. B 63 (2001) 104415.

[53] Y. Dahmane, L. Cagnon, J. Voiron, S. Pairis, M. Bacia, L. Ortega, N. Benbrahim, A.Kadri, Magnetic and structural properties of electrodeposited CoPt and FePtnanowires in nanoporous alumina templates, J. Phys. D: Appl. Phys. 39 (2006)4523.

[54] H. Forster, T. Schrefl, R. Dittrich, D. Suess, W. Scholz, V. Tsiantos, J. Fidler, K.Nielsch, H. Hofmeister, H. Kronmuller, S. Fischer, Magnetization reversal ingranular nanowires, IEEE Trans. Magn. 38 (2002) 2580–2582.

[55] H. Shima, K. Oikawa, A. Fujita, K. Fukamichi, K. Ishida, A. Sakuma, Lattice axialratio and large uniaxial magnetocrystalline anisotropy in L1_{0}-type FePd

single crystals prepared under compressive stress, Phys. Rev. B 70 (2004)224408.

[56] Y.P. Ivanov, M. Vázquez, O. Chubykalo-Fesenko, Magnetic reversal modes incylindrical nanowires, J. Phys. D: Appl. Phys. 46 (2013) 485001.

[57] Y. Hirotsu, S. Kazuhisa, Growth and atomic ordering of hard magnetic L10-FePt,FePd and CoPt alloy nanoparticles studied by transmission electronmicroscopy: alloy system and particle size dependence, J. Ceram. Process.Res. 6 (2005) 236–244.

[58] K.Z. Rozman, J. Kovac, P.J. McGuiness, Z. Samardzija, B. Markoli, S. Kobe,Microstructural, compositional and magnetic characterization ofelectrodeposited and annealed Co–Pt-based thin films, Thin Solid Films 518(2010) 1751–1755.

[59] K.Z. Rozman, S. Sturm, P.J. McGuiness, S. Kobe, Hard magnetic Co–Pt-basednanotubes produced via direct electroplating, magnetics, IEEE Trans. Magn. 45(2009) 4413–4416.

[60] D. Pecko, K.Z. Rozman, P.J. McGuiness, B. Pihlar, S. Kobe, Temperature-drivenmicrostructural, compositional, and magnetic changes in electrodeposited Fe–Pd thin films, J. Appl. Phys. 107 (2010) 09A712–713.