- Email: [email protected]
Dendritic Ag–Fe nanocrystalline alloy synthesized by pulsed electrodeposition and its characterization
Applied Surface Science 316 (2014) 491–496
Contents lists available at ScienceDirect
Applied Surface Science journal homepage: www.elsevier.com/locate/apsusc
Dendritic Ag–Fe nanocrystalline alloy synthesized by pulsed electrodeposition and its characterization Kalavathy Santhi a,c , T.A. Revathy a , V. Narayanan b , A. Stephen a,∗ a b c
Material Science Centre, Department of Nuclear Physics, University of Madras, Guindy Campus, Chennai 600025, India Department of Inorganic Chemistry, University of Madras, Guindy Campus, Chennai 600025, India Department of Physics, Women’s Christian College, Chennai 600006, India
a r t i c l e
i n f o
Article history: Received 3 June 2014 Received in revised form 4 August 2014 Accepted 7 August 2014 Available online 15 August 2014 Keywords: Ag–Fe nanoparticles Nanocrystalline alloy Dendrite morphology Pulsed electrodeposition Microstructure Magnetic materials
a b s t r a c t Synthesis of dendrite shaped Ag–Fe alloy nanomaterial by pulsed electrodeposition route was investigated. The alloy samples were deposited at different current densities from electrolytes of different compositions to study the influence of current density and bath composition on metal contents in the alloy, which was determined by ICP-OES analysis. The XRD studies were carried out to determine the structure of these samples. Magnetic characterization at room temperature and during heating was carried out to understand their magnetic behaviour and to confirm the inferences drawn from the XRD results. The XPS spectra proved the presence of Fe and Ag in the metallic form in the alloy samples. The FESEM and TEM micrographs were taken to view the surface morphology of the nanosized particles. © 2014 Elsevier B.V. All rights reserved.
1. Introduction Ever since the discovery of giant magneto resistance (GMR) in nanocrystalline magnetic materials with large anisotropy, synthesizing such materials in different combinations by different methods and investigating their properties have become an important area of research. Magnetic properties of transition metals can be altered and their stability can be improved by alloying them with noble metals. Nanocrystalline Fe–Cu, Fe–Ag, Co–Ag, etc., exhibit higher anisotropy and hence show greater GMR [1–4]. At the same time, systems like Fe–Ag and Co–Ag are thermally immiscible as per their phase diagrams under equilibrium conditions [5,6]. In the case of Ag–Fe the solid solubility is insignificant and that in the liquid state is very low. This is due to the large positive heat of mixing, which is 42 kJ/mol in the solid state at equiatomic composition and 28 kJ/mol in liquid state [7]. Though this system has been prepared by a few methods such as magnetron sputtering, pulsed laser deposition, vapour quenching, sol–gel method, molecular beam epitoxy, etc., and its properties investigated [8–12], reports on its electrodeposition are scarce. The stability of electrodeposited Ag–Fe alloy has
∗ Corresponding author. Tel.: +91 44 2220 2802; fax: +91 44 2235 3309. E-mail addresses: stephen [email protected], stephen [email protected] (A. Stephen). http://dx.doi.org/10.1016/j.apsusc.2014.08.039 0169-4332/© 2014 Elsevier B.V. All rights reserved.
been reported by Roy et al. [13]. In this work dendrite shaped Ag–Fe alloy nanomaterial is prepared in different compositions by pulsed electrodeposition at different current densities from electrolytes not reported so far. 2. Materials and methods Electrodeposition is one of the best suited ‘bottom-up’ approaches for synthesizing alloy nanomaterials with high purity. The metals forming the alloy can be deposited simultaneously after optimization of the parameters such as the current density, bath composition, pH, addition of complexing agents, temperature, etc. Since Ag and Fe are two metals with a large difference in their standard electrode potentials, their simultaneous deposition to form an alloy, can be carried out in the presence of a complexing agent which reduces the standard electrode potential of the more noble metal. The standard electrode potential of Fe (−0.44 V) is far below that of Ag (+0.799 V). Hence the complexing agent, thiourea (0.05 M) was added to the electrolyte along with 0.2 M of sodium gluconate which served as the additive agent. The electrolyte contained 0.01 M of AgNO3 with 0.05 M of Fe(NO3 )3 for bath 1 in which the metal salts ratio was 1:5 and 0.01 M of AgNO3 with 0.1 M of Fe(NO3 )3 for bath 2, having the ratio of 1:10. Sodium perchlorate (0.2 M) was added to the bath in order to improve the ionic conductivity and enhance deposition. As the pH of the bath was very
492
K. Santhi et al. / Applied Surface Science 316 (2014) 491–496
low, it was increased to 4 before every deposition by adding a few drops of diluted ammonia. The pulse currents of densities of 30 mA/cm2 , 40 mA/cm2 and 50 mA/cm2 were passed through the electrolyte with 1 ms Ton and 9 ms Toff at a duty cycle of 10% to deposit samples 1A, 1B, 1C from bath 1 and 2A, 2B, 2C from bath 2. Passage of pulse current not only ensures uniform distribution of ions before every pulse during Ton , but also enables fresh nucleation during the ON time and formation of nanosized particles. The bath was stirred continuously using a magnetic stirrer. A stainless steel substrate was used as the cathode while a graphite plate formed the anode. This is a slightly modified method of what has been reported by the authors for the synthesis of Ag–Ni alloy nanomaterial [14]. The structure of the deposits were investigated using a Rich Siefert 3000 X-ray diffractometer (XRD) with Cu K␣1 radiation ( = 0.15406 nm). The composition of the metals in the alloy samples was also analysed using Perkin Elmer Optima 5300 DV, ICP-OES. The field emission scanning electron microscope (FESEM) Hitachi, SU6600, equipped with energy dispersive X-ray spectrometer (EDX) was used for observing the morphology of Ag–Fe samples. The microstructure of the deposit Ag–Fe 2B was determined using the Transmission Electron microscopy (HRTEM) on a JEOL model JEM 2011 at an accelerating voltage of 200 kV. The Xray photoelectron spectroscopy (XPS) measurements were made with Omicron ESCA Probe spectrometer with monochromatized Al K␣ X-rays (h = 1486.6 eV). The vibrating sample magnetometer (VSM) EG&G, PARC Model 4500 was used to analyse the magnetic behaviour of the samples both at room temperature and during heating. 3. Results and discussion 3.1. Composition analysis The ICP-OES analysis reveals the metal contents in the alloy samples deposited using different current densities from the two electrolytic compositions. These values are presented in Table 1. Samples 1A, 1B and 1C deposited from bath I, have lower Fe content compared to that in 2A, 2B and 2C obtained from bath II, as expected. It is also observed that the Fe content increases with increase in current density. The samples from bath I show wide variations in the Fe content with changes in current density, whereas the samples from bath II have the Fe at.% lying within the range, from 86.6 and 94.2 for the three current densities. Therefore at the lower bath concentration, the current density influences the metal composition drastically. But at the higher concentration of the bath, the current density seems to be less influential. Though the other parameters such as duty cycle of the current pulse, pH, temperature of the bath, the presence of additives, etc., influence the size and the morphology of the deposits in addition to having little effect on the metal contents, the bath composition and the current density are the two parameters which greatly alter the alloy composition. 3.2. Structural analysis The XRD patterns of the samples Ag–Fe 1A, 1B, 1C and 2A, 2B, 2C are shown in Fig. 1a–c and d–f respectively. The samples 1A, 1B, 1C prepared from 1:5 composition of the electrolyte, exhibit fcc structure peaks at the same positions as fcc Ag peaks according to JCPDS file no. 893722 whereas the samples 2A, 2B, 2C obtained from 1:10 composition have a structure similar to bcc Fe (JCPDS no. 851410). In the samples deposited from bath 2, the most prominent fcc Ag(1 1 1) peak is not visible. The presence of fcc peaks can be attributed to the mixing of Fe atoms whose atomic radius
Fig. 1. XRD patterns of samples 1A, 1B, 1C deposited from bath 1 showing mixed fcc and bcc structure peaks and 2A, 2B, 2C from bath 2 showing bcc Fe peaks.
is smaller, into the fcc Ag lattice. Similar formation of fcc structure of Ag–Fe alloy nanomaterials have been reported in literatures [8,13,15]. But, since the peak positions of bcc Fe (1 1 0), (2 0 0) and (2 1 1) are very close to the fcc peaks (2 0 0), (2 2 0) and (3 1 1) of Ag–Fe alloy, they are likely to have submerged. Hence the coexistence of Fe rich bcc phase of the alloy cannot be ruled out. In the case of samples 1A, 1B, 1C deposited from lower Fe concentration in the electrolyte the fcc phase is more prominent while the samples 2A, 2B, 2C deposited from higher concentration of Fe, exhibit bcc structure (JCPDS file no. 851410). Kataoka et al. have reported fcc structure for the vapour quenched Fe1−x –Agx alloy for x > 0.6, coexistence of fcc and bcc phases for 0.14 < x < 0.6 and bcc peaks for x < 0.14 [16]. This is in close agreement with what is observed in the present study. The mixed fcc and bcc phases of the samples prepared from bath 1 and the bcc phase of the alloy samples obtained from bath 2 are confirmed by their Curie transitions. 3.3. Surface morphology The FESEM micrographs of four samples 1A, 1B, 2A and 2B are shown in Fig. 2a–d respectively. All the samples show a beautiful leaf-like dendritic growth. Formation of dendrites is attributed to the state where there is growth of protrusion under charge transfer control while the deposition on the rest of the cathode is under diffusion control. Dendrites with ordered nanometric structures grow as branches when a critical overvoltage is exceeded. When metals are deposited at high current densities as in pulsed electrodeposition, there is a tendency to form powdered deposits which are dendritic. Factors such as decrease in concentration of the depositing ions, temperature, stirring, increase in concentration of the supporting electrolytes and viscosity of the bath have been reported to favour dendritic growth [17]. The particle sizes of these samples range from 25 to 30 nm. The EDX spectrum taken on a single particle presented in Fig. 2e and f reveals the presence of both Ag and Fe. This is an evidence for the mixing of Ag and Fe atoms to form the alloy particles. The EDX spectrum indicates the presence of oxygen which is due to surface oxidation as commonly observed in nanosized particles. There is also a meagre amount of sulphur present on the surface which is likely to be from the complexing agent thiourea that gets adsorbed on the deposit just when the current is switched off before the removal of the substrate [18]. The TEM image of the sample 2B is depicted in Fig. 3. It
K. Santhi et al. / Applied Surface Science 316 (2014) 491–496
493
Table 1 Composition of metals in pulsed electrodeposited Ag–Fe alloy samples as obtained from ICP-OES analysis. Sample
Molar ratio of AgNO3 :Fe(NO3 )3 in electrolyte
Current density (mA/cm2 )
Ag content (at.%)
Fe content (at.%)
1A 1B 1C 2A 2B 2C
1:05 1:05 1:05 1:10 1:10 1:10
30 40 50 30 40 50
87.5 40.3 24.3 13.4 11.2 05.8
12.6 59.7 75.7 86.6 88.8 94.2
Fig. 2. (a)–(d) FESEM images depicting dendritic growth of the alloy samples 1A, 1B and 2A, 2B. (e)–(f) show the EDX spectrum which indicate the presence of Ag and Fe in a single particle.
494
K. Santhi et al. / Applied Surface Science 316 (2014) 491–496
Fig. 3. TEM micrographs with the corresponding EDX spectra of the sample 2B confirming the dendrite morphology of the sample.
Fig. 4. (a) Wide scan XPS spectrum of 2B sample. (b–d) High resolution spectra of Ag 3d, Fe 2p and oxygen 1 s regions.
K. Santhi et al. / Applied Surface Science 316 (2014) 491–496
495
Table 2 Magnetization values and Curie temperatures of Ag–Fe deposits obtained from the two electrolytic baths at various current densities. Sample
Fe content (at.%)
Ms (A m2 /kg)
Hc (kA/m)
1A 1B 1C 2A 2B 2C
12.6 59.7 75.7 86.6 88.8 94.2
1.9 5.6 12.3 46.6 49.8 53.9
1.6 3.4 4.9 5.2 4.6 3.9
Tc1 (◦ C) 257 266 312 – – –
Tc2 (◦ C) 502 515 531 604 615 622
Fig. 5. Magnetic hysteresis loops indicating the ferromagnetic behaviour of all the deposits.
also confirms the leaf-like growth of the particles of size of about 25 nm. 3.4. Surface elemental analysis using XPS XPS analysis of the as prepared alloy sample 2B was done to determine the chemical composition at the surface and to analyse the electronic charge states of the metals present. XPS can be used to distinguish between metallic and different oxidation states of the elements so that the presence of the zero valent metal species in alloy and bimetallic systems is confirmed [19]. Fig. 4a presents the wide scan spectrum of the sample indicating the presence of the 3d and 3p states of Ag and the 2p state of Fe along with the oxygen peak. Fig. 4b shows Ag 3d5/2 and Ag 3d3/2 peaks of metallic silver at binding energies 368.3 eV and 374.3 eV with a difference of 6.0 eV [20]. The slow scan spectrum of 2p Fe region shown in Fig. 4c has two peaks accompanied by various shoulders. The two prominent peaks at 707.75 eV and 720.95 eV with an energy gap of 13.2 eV are from the 2p3/2 and 2p 1/2 states of metallic Fe [21,22]. The two resolved peaks at 711.5 eV and 725.1 eV match with the binding energies of Fe 2p3/2 and Fe 2p1/2 states of Fe3+ ions. The fine structure splitting of these two components spaced at 13.6 eV indicates that these Fe3+ ions originate from the ␥-Fe2 O3 molecules on the surface of the nanoparticles [23]. The two shoulders on the higher energy side at 716.0 eV and 730.2 eV represent the satellite peaks associated with Fe 2p. These satellite peaks also indicate bonding of oxygen with Fe resulting in formation of iron oxide on the surface. Fig. 4d shows the oxygen peak at 530.2 eV, which can be attributed to its bonding with Fe [24].
Fig. 6. Magnetic behaviour of samples 1A, 1B and 1C under heat treatment. These samples of mixed phases exhibit two Curie transitions.
Ag with Fe which reduces the magnetic interactions. In addition to this, the presence of a small amount of sulphur coming from the complexing agent, surface oxidation and the size effect of the nanoparticles also cause a decrease in the magnetization values. The temperature dependence of magnetization of the samples was analysed while heating them at a constant rate in the vibrating sample magnetometer. The two sets of samples exhibit different behaviour with increase in temperature. Samples 1A, 1B, and 1C prepared from bath 1, show two Curie transitions, one above 250 ◦ C and the other above 500 ◦ C as seen from Fig. 6. The two
3.5. Magnetic characterization The results obtained from vibrating magnetometer for all the Ag–Fe samples of both fcc and bcc structure indicate their ferromagnetic nature as shown in Fig. 5. All the samples exhibit ferromagnetic nature. The saturation magnetization values ( s ), the coercivity (Hc ) and the Curie temperatures (Tc ) of the samples are presented in Table 2. Comparing the s values of the two sets of samples, Ag–Fe alloys show a huge increase in the s values when the bath concentration is higher. At this concentration the variations in s due to current densities is less pronounced. The saturation magnetization values of the samples are found to be less compared to that of pure Fe. This could be due to the new structures formed on account of mixing of the non-magnetic
Fig. 7. Magnetic behaviour of samples 2A, 2B and 2C with increase in temperature and the Curie transitions.
496
K. Santhi et al. / Applied Surface Science 316 (2014) 491–496
Curie transitions indicate that the alloy samples are composed of two phases. Hence their structure is a combination of Ag rich fcc Ag–Fe phase and the Fe rich bcc Ag–Fe phase. The samples deposited from bath 2 with higher Fe content have greater saturation magnetization values as can be seen in Table 2 and they show single Curie transition above 600 ◦ C (Fig. 7), indicating that these samples possess single bcc phase as also observed from XRD. These values are similar to the Curie temperature reported by Kataoka et al. for fcc and bcc Ag–Fe alloy [25]. In the case of samples 2A, 2B and 2C, a well defined Hopkinson peak indicating an increase in magnetization just before Curie transition can be observed. The origin of Hopkinson effect in nanocrystalline materials with a single phase has been explained in literatures [26,27]. The initial susceptibility in these materials can be expressed as √ i = Ms / K1 where Ms is the saturation magnetization and K1 is the crystalline anisotropy constant. At a constant low applied field, both Ms and K1 decrease with increase in temperature. But as the temperature increases the anisotropy constant, K1 decreases more rapidly than Ms , and tends to zero at the Curie point. Due to this reason the permeability increases and attains a maximum causing the Hopkinson peak to appear near the Curie temperature. This distinct phenomenon of soft ferromagnetic materials is clearly speculated in the Fe rich bcc alloy samples whereas the samples of mixed phases from bath 1, do not exhibit this behaviour. 4. Conclusion Synthesis of the immiscible Ag–Fe alloy nanomaterial in various compositions has been effectively accomplished by pulsed electrodeposition. It is found that by changing the bath composition and the current density, the metal contents in the alloy can be controlled. The samples prepared from lower Fe salt concentration in the bath contain more of Ag rich fcc phase along with the Fe rich bcc phase while the samples obtained from higher Fe salt concentration exhibit only the bcc Fe structure. The Fe content in the samples is primarily bath concentration dependent and for the lower bath concentration, the current density is found to have considerable effect. But at the higher bath concentration, variation due to current density is less pronounced. Therefore even the structures and the magnetic behaviour of the two sets of samples are very much different from each other. This conclusion drawn from the XRD results and the VSM measurements is well supported by the observed Curie transitions of the two sets of samples. The FESEM and TEM micrographs clearly depict that the particles are 25–30 nm in size and the crystals have grown with dendrite morphology. Acknowledgement The authors wish to thank Mr. B. Sounderarajan for his support in carrying out the magnetic measurements. We acknowledge the FESEM facilities provided by the National Centre for Nanoscience and Nanotechnology, University of Madras and the ICP-OES facility provided by DST and SAIF, IIT Madras.
References [1] J.Q. Xiao, J.S. Jiang, C.L. Chien, Giant magnetoresistance in multilayer magnetic systems, Phys. Rev. Lett. 68 (1992) 3749–3752. [2] A. Tsoukatos, H. Wan, G.C. Hadjipanayis, K.M. Unruh, Z.G. Li, Giant magnetoresistance studies in (Fe,Co)-Ag films, J. Appl. Phys. 73 (1993) 5509–5511. [3] C. Revenant-Brizard, J.R. Regnard, J. Mimault, O. Proux, B. Dieny, B. Mevel, Nanostructure of giant magnetoresistance heterogeneous alloys Ni0.20 Ag after annealing, J. Phys. IV France 7 (1997), C2-1111–C2-1113. [4] S. Arana, N. Arana, N.J. Gracia, E. Castano, High sensitivity linear position sensor developed using granular Ag–Co giant magnetoresistances, Sens. Actuators A 123–124 (2005) 116–121. [5] L.J. Swartzendruber, Bull. Alloy Phase Diagr. 5 (1984) 560. [6] B. Hugh (Ed.), Alloy Phase Diagram ASM Handbook, ASM International, Cleveland, OH, 1992. [7] F.R. De Boer, R. Boom, W.C. Mattens, A.R. Midema, A.K. Niessen, Cohesion in Metals: Transition Metal Alloys, North-Holland, Amsterdam, 1988. [8] R.L. Zong, S.P. Wen, F. Zeng, Y. Gao, C. Song, B. He, F. Pan, Fcc solid solution alloy films formed in Fe–Ag system and their mechanical behaviors, Appl. Surf. Sci. 253 (2007) 2993–2998. [9] M. Stomer, H.-U. Krebs, Structure of laser deposited alloys, J. Appl. Phys. 78 (1995) 7080–7087. [10] G. Xiao, J.Q. Wang, Magnetic properties of metallic Co- and Fe-based granular alloys, J. Appl. Phys. 75 (10) (1994) 6604–6606. [11] J.M. Soares, J.H. de Aroujo, F.O.C. de Assis, T. Dunelow, M.M.X. Junior, J.M. Sasaki, Magnetic properties of the granular alloy Fe10 Ag90 as a function of annealing temperature, Mater. Res. 8 (2005) 347–350. [12] N. Thangaraj, C. Echer, M. Kannan, Krishnan, R.F.C. Farrow, R.F. Marks, S.S.P. Parkin, Giant magnetoresistance and microstructural characteristics of epitaxial Fe–Ag and Co–Ag granular thin films, J. Appl. Phys. 79 (1994) 6900–6902. [13] M.K. Roy, P.M.G. Nambissan, H.C. Verma, Structural, thermal stability and defect studies of Fe–Ag alloy prepared by electrodeposition technique, J. Alloys Compd. 345 (2002) 183–188. [14] K. Santhi, S.N. Karthick, H.-J. Kim, M. Nidhin, V. Narayanan, A. Stephen, Microstructure analysis of the ferromagnetic Ag–Ni system synthesized by pulsed electrodeposition, Appl. Surf. Sci. 258 (2012) 3126–3132. [15] L. Dimesso, G. Miehe, H. Fuess, H. Hahn, Preparation of nanostructured granular Ag–Co and Ag–Fe alloys by gas flow condensation technique, J. Magn. Magn. Mater. 191 (1999) 162–168. [16] N. Kataoka, K. Sumiyama, Y. Nakamura, Magnetic properties of highconcentration Ag–Fe alloys produced by vapour quenching, J. Phys. F: Met. Phys. 15 (1985) 1405–1411. [17] L.P. Bicellii, B. Bozzini, C. Mele, L. D’Urzo, A review of nanostructural aspects of metal electrodeposition, Int. J. Electrochem. Sci. 3 (2008) 356–408. [18] J. Garcia-Torres, E. Gómez, E. Vallés, Modification of magnetic and structural properties of Co and Co–Ag electrodeposits by sulphur incorporation, Mater. Chem. Phys. 122 (2010) 463–469. [19] X. Qingmei, Z. Yao, J. Liu, R. Hai, H.Y. Oderji, H. Ding, Synthesis and characterization of Ag–Ni bimetallic nanoparticles by laser-induced plasma, Thin Solid Films 519 (2011) 7116–7119. [20] J.F. Moulder, W.F. Stickle, P.E. Sobel, K.D. Bomben, in: J. Chastain, R.C. King Jr. (Eds.), Handbook of X-ray Photoelectron Spectroscopy, ULVAC-PHI Inc./Physical Electronics USA Inc., Japan/USA, 1993. [21] A. Lebugle, U. Axelson, R. Nyholm, N. Martensson, Experimental L and M core level binding energies for the metals 22 Ti to 30 Zn, Phys. Scr. 23 (1981) 825. [22] K. Asami, A precisely consistent energy calibration method for X-ray photoelectron spectroscopy, J. Electron Spectrosc. Relat. Phenom. 9 (1976) 469–478. [23] N.S. McIntyre, D.G. Zetaruk, X-ray photoelectron spectroscopic studies of iron oxides, Anal. Chem. 49 (1977) 1521–1529. [24] C.D. Wagner, D.A. Zatco, R.H. Raymond, Use of oxygen KLL Auger lines in identification of surface chemical states by electron spectroscopy for chemical analysis, Anal. Chem. 52 (1980) 1445. [25] N. Kataoka, K. Sumiyama, Y. Nakamura, Thermal stability of nonequilibrium crystalline Fe–Ag alloys produced by vapor quenching, Acta Metall. 37 (1989) 1135–1142. [26] B.D. Cullity, C.D. Graham, Introduction to Magnetic Materials, 2nd ed., John Wiley & Sons Inc., New Jersey, 2009. [27] S. Chikazumi, Physics of Magnetism, Wiley, New York, 1964, pp. 247.
Contents lists available at ScienceDirect
Applied Surface Science journal homepage: www.elsevier.com/locate/apsusc
Dendritic Ag–Fe nanocrystalline alloy synthesized by pulsed electrodeposition and its characterization Kalavathy Santhi a,c , T.A. Revathy a , V. Narayanan b , A. Stephen a,∗ a b c
Material Science Centre, Department of Nuclear Physics, University of Madras, Guindy Campus, Chennai 600025, India Department of Inorganic Chemistry, University of Madras, Guindy Campus, Chennai 600025, India Department of Physics, Women’s Christian College, Chennai 600006, India
a r t i c l e
i n f o
Article history: Received 3 June 2014 Received in revised form 4 August 2014 Accepted 7 August 2014 Available online 15 August 2014 Keywords: Ag–Fe nanoparticles Nanocrystalline alloy Dendrite morphology Pulsed electrodeposition Microstructure Magnetic materials
a b s t r a c t Synthesis of dendrite shaped Ag–Fe alloy nanomaterial by pulsed electrodeposition route was investigated. The alloy samples were deposited at different current densities from electrolytes of different compositions to study the influence of current density and bath composition on metal contents in the alloy, which was determined by ICP-OES analysis. The XRD studies were carried out to determine the structure of these samples. Magnetic characterization at room temperature and during heating was carried out to understand their magnetic behaviour and to confirm the inferences drawn from the XRD results. The XPS spectra proved the presence of Fe and Ag in the metallic form in the alloy samples. The FESEM and TEM micrographs were taken to view the surface morphology of the nanosized particles. © 2014 Elsevier B.V. All rights reserved.
1. Introduction Ever since the discovery of giant magneto resistance (GMR) in nanocrystalline magnetic materials with large anisotropy, synthesizing such materials in different combinations by different methods and investigating their properties have become an important area of research. Magnetic properties of transition metals can be altered and their stability can be improved by alloying them with noble metals. Nanocrystalline Fe–Cu, Fe–Ag, Co–Ag, etc., exhibit higher anisotropy and hence show greater GMR [1–4]. At the same time, systems like Fe–Ag and Co–Ag are thermally immiscible as per their phase diagrams under equilibrium conditions [5,6]. In the case of Ag–Fe the solid solubility is insignificant and that in the liquid state is very low. This is due to the large positive heat of mixing, which is 42 kJ/mol in the solid state at equiatomic composition and 28 kJ/mol in liquid state [7]. Though this system has been prepared by a few methods such as magnetron sputtering, pulsed laser deposition, vapour quenching, sol–gel method, molecular beam epitoxy, etc., and its properties investigated [8–12], reports on its electrodeposition are scarce. The stability of electrodeposited Ag–Fe alloy has
∗ Corresponding author. Tel.: +91 44 2220 2802; fax: +91 44 2235 3309. E-mail addresses: stephen [email protected], stephen [email protected] (A. Stephen). http://dx.doi.org/10.1016/j.apsusc.2014.08.039 0169-4332/© 2014 Elsevier B.V. All rights reserved.
been reported by Roy et al. [13]. In this work dendrite shaped Ag–Fe alloy nanomaterial is prepared in different compositions by pulsed electrodeposition at different current densities from electrolytes not reported so far. 2. Materials and methods Electrodeposition is one of the best suited ‘bottom-up’ approaches for synthesizing alloy nanomaterials with high purity. The metals forming the alloy can be deposited simultaneously after optimization of the parameters such as the current density, bath composition, pH, addition of complexing agents, temperature, etc. Since Ag and Fe are two metals with a large difference in their standard electrode potentials, their simultaneous deposition to form an alloy, can be carried out in the presence of a complexing agent which reduces the standard electrode potential of the more noble metal. The standard electrode potential of Fe (−0.44 V) is far below that of Ag (+0.799 V). Hence the complexing agent, thiourea (0.05 M) was added to the electrolyte along with 0.2 M of sodium gluconate which served as the additive agent. The electrolyte contained 0.01 M of AgNO3 with 0.05 M of Fe(NO3 )3 for bath 1 in which the metal salts ratio was 1:5 and 0.01 M of AgNO3 with 0.1 M of Fe(NO3 )3 for bath 2, having the ratio of 1:10. Sodium perchlorate (0.2 M) was added to the bath in order to improve the ionic conductivity and enhance deposition. As the pH of the bath was very
492
K. Santhi et al. / Applied Surface Science 316 (2014) 491–496
low, it was increased to 4 before every deposition by adding a few drops of diluted ammonia. The pulse currents of densities of 30 mA/cm2 , 40 mA/cm2 and 50 mA/cm2 were passed through the electrolyte with 1 ms Ton and 9 ms Toff at a duty cycle of 10% to deposit samples 1A, 1B, 1C from bath 1 and 2A, 2B, 2C from bath 2. Passage of pulse current not only ensures uniform distribution of ions before every pulse during Ton , but also enables fresh nucleation during the ON time and formation of nanosized particles. The bath was stirred continuously using a magnetic stirrer. A stainless steel substrate was used as the cathode while a graphite plate formed the anode. This is a slightly modified method of what has been reported by the authors for the synthesis of Ag–Ni alloy nanomaterial [14]. The structure of the deposits were investigated using a Rich Siefert 3000 X-ray diffractometer (XRD) with Cu K␣1 radiation ( = 0.15406 nm). The composition of the metals in the alloy samples was also analysed using Perkin Elmer Optima 5300 DV, ICP-OES. The field emission scanning electron microscope (FESEM) Hitachi, SU6600, equipped with energy dispersive X-ray spectrometer (EDX) was used for observing the morphology of Ag–Fe samples. The microstructure of the deposit Ag–Fe 2B was determined using the Transmission Electron microscopy (HRTEM) on a JEOL model JEM 2011 at an accelerating voltage of 200 kV. The Xray photoelectron spectroscopy (XPS) measurements were made with Omicron ESCA Probe spectrometer with monochromatized Al K␣ X-rays (h = 1486.6 eV). The vibrating sample magnetometer (VSM) EG&G, PARC Model 4500 was used to analyse the magnetic behaviour of the samples both at room temperature and during heating. 3. Results and discussion 3.1. Composition analysis The ICP-OES analysis reveals the metal contents in the alloy samples deposited using different current densities from the two electrolytic compositions. These values are presented in Table 1. Samples 1A, 1B and 1C deposited from bath I, have lower Fe content compared to that in 2A, 2B and 2C obtained from bath II, as expected. It is also observed that the Fe content increases with increase in current density. The samples from bath I show wide variations in the Fe content with changes in current density, whereas the samples from bath II have the Fe at.% lying within the range, from 86.6 and 94.2 for the three current densities. Therefore at the lower bath concentration, the current density influences the metal composition drastically. But at the higher concentration of the bath, the current density seems to be less influential. Though the other parameters such as duty cycle of the current pulse, pH, temperature of the bath, the presence of additives, etc., influence the size and the morphology of the deposits in addition to having little effect on the metal contents, the bath composition and the current density are the two parameters which greatly alter the alloy composition. 3.2. Structural analysis The XRD patterns of the samples Ag–Fe 1A, 1B, 1C and 2A, 2B, 2C are shown in Fig. 1a–c and d–f respectively. The samples 1A, 1B, 1C prepared from 1:5 composition of the electrolyte, exhibit fcc structure peaks at the same positions as fcc Ag peaks according to JCPDS file no. 893722 whereas the samples 2A, 2B, 2C obtained from 1:10 composition have a structure similar to bcc Fe (JCPDS no. 851410). In the samples deposited from bath 2, the most prominent fcc Ag(1 1 1) peak is not visible. The presence of fcc peaks can be attributed to the mixing of Fe atoms whose atomic radius
Fig. 1. XRD patterns of samples 1A, 1B, 1C deposited from bath 1 showing mixed fcc and bcc structure peaks and 2A, 2B, 2C from bath 2 showing bcc Fe peaks.
is smaller, into the fcc Ag lattice. Similar formation of fcc structure of Ag–Fe alloy nanomaterials have been reported in literatures [8,13,15]. But, since the peak positions of bcc Fe (1 1 0), (2 0 0) and (2 1 1) are very close to the fcc peaks (2 0 0), (2 2 0) and (3 1 1) of Ag–Fe alloy, they are likely to have submerged. Hence the coexistence of Fe rich bcc phase of the alloy cannot be ruled out. In the case of samples 1A, 1B, 1C deposited from lower Fe concentration in the electrolyte the fcc phase is more prominent while the samples 2A, 2B, 2C deposited from higher concentration of Fe, exhibit bcc structure (JCPDS file no. 851410). Kataoka et al. have reported fcc structure for the vapour quenched Fe1−x –Agx alloy for x > 0.6, coexistence of fcc and bcc phases for 0.14 < x < 0.6 and bcc peaks for x < 0.14 [16]. This is in close agreement with what is observed in the present study. The mixed fcc and bcc phases of the samples prepared from bath 1 and the bcc phase of the alloy samples obtained from bath 2 are confirmed by their Curie transitions. 3.3. Surface morphology The FESEM micrographs of four samples 1A, 1B, 2A and 2B are shown in Fig. 2a–d respectively. All the samples show a beautiful leaf-like dendritic growth. Formation of dendrites is attributed to the state where there is growth of protrusion under charge transfer control while the deposition on the rest of the cathode is under diffusion control. Dendrites with ordered nanometric structures grow as branches when a critical overvoltage is exceeded. When metals are deposited at high current densities as in pulsed electrodeposition, there is a tendency to form powdered deposits which are dendritic. Factors such as decrease in concentration of the depositing ions, temperature, stirring, increase in concentration of the supporting electrolytes and viscosity of the bath have been reported to favour dendritic growth [17]. The particle sizes of these samples range from 25 to 30 nm. The EDX spectrum taken on a single particle presented in Fig. 2e and f reveals the presence of both Ag and Fe. This is an evidence for the mixing of Ag and Fe atoms to form the alloy particles. The EDX spectrum indicates the presence of oxygen which is due to surface oxidation as commonly observed in nanosized particles. There is also a meagre amount of sulphur present on the surface which is likely to be from the complexing agent thiourea that gets adsorbed on the deposit just when the current is switched off before the removal of the substrate [18]. The TEM image of the sample 2B is depicted in Fig. 3. It
K. Santhi et al. / Applied Surface Science 316 (2014) 491–496
493
Table 1 Composition of metals in pulsed electrodeposited Ag–Fe alloy samples as obtained from ICP-OES analysis. Sample
Molar ratio of AgNO3 :Fe(NO3 )3 in electrolyte
Current density (mA/cm2 )
Ag content (at.%)
Fe content (at.%)
1A 1B 1C 2A 2B 2C
1:05 1:05 1:05 1:10 1:10 1:10
30 40 50 30 40 50
87.5 40.3 24.3 13.4 11.2 05.8
12.6 59.7 75.7 86.6 88.8 94.2
Fig. 2. (a)–(d) FESEM images depicting dendritic growth of the alloy samples 1A, 1B and 2A, 2B. (e)–(f) show the EDX spectrum which indicate the presence of Ag and Fe in a single particle.
494
K. Santhi et al. / Applied Surface Science 316 (2014) 491–496
Fig. 3. TEM micrographs with the corresponding EDX spectra of the sample 2B confirming the dendrite morphology of the sample.
Fig. 4. (a) Wide scan XPS spectrum of 2B sample. (b–d) High resolution spectra of Ag 3d, Fe 2p and oxygen 1 s regions.
K. Santhi et al. / Applied Surface Science 316 (2014) 491–496
495
Table 2 Magnetization values and Curie temperatures of Ag–Fe deposits obtained from the two electrolytic baths at various current densities. Sample
Fe content (at.%)
Ms (A m2 /kg)
Hc (kA/m)
1A 1B 1C 2A 2B 2C
12.6 59.7 75.7 86.6 88.8 94.2
1.9 5.6 12.3 46.6 49.8 53.9
1.6 3.4 4.9 5.2 4.6 3.9
Tc1 (◦ C) 257 266 312 – – –
Tc2 (◦ C) 502 515 531 604 615 622
Fig. 5. Magnetic hysteresis loops indicating the ferromagnetic behaviour of all the deposits.
also confirms the leaf-like growth of the particles of size of about 25 nm. 3.4. Surface elemental analysis using XPS XPS analysis of the as prepared alloy sample 2B was done to determine the chemical composition at the surface and to analyse the electronic charge states of the metals present. XPS can be used to distinguish between metallic and different oxidation states of the elements so that the presence of the zero valent metal species in alloy and bimetallic systems is confirmed [19]. Fig. 4a presents the wide scan spectrum of the sample indicating the presence of the 3d and 3p states of Ag and the 2p state of Fe along with the oxygen peak. Fig. 4b shows Ag 3d5/2 and Ag 3d3/2 peaks of metallic silver at binding energies 368.3 eV and 374.3 eV with a difference of 6.0 eV [20]. The slow scan spectrum of 2p Fe region shown in Fig. 4c has two peaks accompanied by various shoulders. The two prominent peaks at 707.75 eV and 720.95 eV with an energy gap of 13.2 eV are from the 2p3/2 and 2p 1/2 states of metallic Fe [21,22]. The two resolved peaks at 711.5 eV and 725.1 eV match with the binding energies of Fe 2p3/2 and Fe 2p1/2 states of Fe3+ ions. The fine structure splitting of these two components spaced at 13.6 eV indicates that these Fe3+ ions originate from the ␥-Fe2 O3 molecules on the surface of the nanoparticles [23]. The two shoulders on the higher energy side at 716.0 eV and 730.2 eV represent the satellite peaks associated with Fe 2p. These satellite peaks also indicate bonding of oxygen with Fe resulting in formation of iron oxide on the surface. Fig. 4d shows the oxygen peak at 530.2 eV, which can be attributed to its bonding with Fe [24].
Fig. 6. Magnetic behaviour of samples 1A, 1B and 1C under heat treatment. These samples of mixed phases exhibit two Curie transitions.
Ag with Fe which reduces the magnetic interactions. In addition to this, the presence of a small amount of sulphur coming from the complexing agent, surface oxidation and the size effect of the nanoparticles also cause a decrease in the magnetization values. The temperature dependence of magnetization of the samples was analysed while heating them at a constant rate in the vibrating sample magnetometer. The two sets of samples exhibit different behaviour with increase in temperature. Samples 1A, 1B, and 1C prepared from bath 1, show two Curie transitions, one above 250 ◦ C and the other above 500 ◦ C as seen from Fig. 6. The two
3.5. Magnetic characterization The results obtained from vibrating magnetometer for all the Ag–Fe samples of both fcc and bcc structure indicate their ferromagnetic nature as shown in Fig. 5. All the samples exhibit ferromagnetic nature. The saturation magnetization values ( s ), the coercivity (Hc ) and the Curie temperatures (Tc ) of the samples are presented in Table 2. Comparing the s values of the two sets of samples, Ag–Fe alloys show a huge increase in the s values when the bath concentration is higher. At this concentration the variations in s due to current densities is less pronounced. The saturation magnetization values of the samples are found to be less compared to that of pure Fe. This could be due to the new structures formed on account of mixing of the non-magnetic
Fig. 7. Magnetic behaviour of samples 2A, 2B and 2C with increase in temperature and the Curie transitions.
496
K. Santhi et al. / Applied Surface Science 316 (2014) 491–496
Curie transitions indicate that the alloy samples are composed of two phases. Hence their structure is a combination of Ag rich fcc Ag–Fe phase and the Fe rich bcc Ag–Fe phase. The samples deposited from bath 2 with higher Fe content have greater saturation magnetization values as can be seen in Table 2 and they show single Curie transition above 600 ◦ C (Fig. 7), indicating that these samples possess single bcc phase as also observed from XRD. These values are similar to the Curie temperature reported by Kataoka et al. for fcc and bcc Ag–Fe alloy [25]. In the case of samples 2A, 2B and 2C, a well defined Hopkinson peak indicating an increase in magnetization just before Curie transition can be observed. The origin of Hopkinson effect in nanocrystalline materials with a single phase has been explained in literatures [26,27]. The initial susceptibility in these materials can be expressed as √ i = Ms / K1 where Ms is the saturation magnetization and K1 is the crystalline anisotropy constant. At a constant low applied field, both Ms and K1 decrease with increase in temperature. But as the temperature increases the anisotropy constant, K1 decreases more rapidly than Ms , and tends to zero at the Curie point. Due to this reason the permeability increases and attains a maximum causing the Hopkinson peak to appear near the Curie temperature. This distinct phenomenon of soft ferromagnetic materials is clearly speculated in the Fe rich bcc alloy samples whereas the samples of mixed phases from bath 1, do not exhibit this behaviour. 4. Conclusion Synthesis of the immiscible Ag–Fe alloy nanomaterial in various compositions has been effectively accomplished by pulsed electrodeposition. It is found that by changing the bath composition and the current density, the metal contents in the alloy can be controlled. The samples prepared from lower Fe salt concentration in the bath contain more of Ag rich fcc phase along with the Fe rich bcc phase while the samples obtained from higher Fe salt concentration exhibit only the bcc Fe structure. The Fe content in the samples is primarily bath concentration dependent and for the lower bath concentration, the current density is found to have considerable effect. But at the higher bath concentration, variation due to current density is less pronounced. Therefore even the structures and the magnetic behaviour of the two sets of samples are very much different from each other. This conclusion drawn from the XRD results and the VSM measurements is well supported by the observed Curie transitions of the two sets of samples. The FESEM and TEM micrographs clearly depict that the particles are 25–30 nm in size and the crystals have grown with dendrite morphology. Acknowledgement The authors wish to thank Mr. B. Sounderarajan for his support in carrying out the magnetic measurements. We acknowledge the FESEM facilities provided by the National Centre for Nanoscience and Nanotechnology, University of Madras and the ICP-OES facility provided by DST and SAIF, IIT Madras.
References [1] J.Q. Xiao, J.S. Jiang, C.L. Chien, Giant magnetoresistance in multilayer magnetic systems, Phys. Rev. Lett. 68 (1992) 3749–3752. [2] A. Tsoukatos, H. Wan, G.C. Hadjipanayis, K.M. Unruh, Z.G. Li, Giant magnetoresistance studies in (Fe,Co)-Ag films, J. Appl. Phys. 73 (1993) 5509–5511. [3] C. Revenant-Brizard, J.R. Regnard, J. Mimault, O. Proux, B. Dieny, B. Mevel, Nanostructure of giant magnetoresistance heterogeneous alloys Ni0.20 Ag after annealing, J. Phys. IV France 7 (1997), C2-1111–C2-1113. [4] S. Arana, N. Arana, N.J. Gracia, E. Castano, High sensitivity linear position sensor developed using granular Ag–Co giant magnetoresistances, Sens. Actuators A 123–124 (2005) 116–121. [5] L.J. Swartzendruber, Bull. Alloy Phase Diagr. 5 (1984) 560. [6] B. Hugh (Ed.), Alloy Phase Diagram ASM Handbook, ASM International, Cleveland, OH, 1992. [7] F.R. De Boer, R. Boom, W.C. Mattens, A.R. Midema, A.K. Niessen, Cohesion in Metals: Transition Metal Alloys, North-Holland, Amsterdam, 1988. [8] R.L. Zong, S.P. Wen, F. Zeng, Y. Gao, C. Song, B. He, F. Pan, Fcc solid solution alloy films formed in Fe–Ag system and their mechanical behaviors, Appl. Surf. Sci. 253 (2007) 2993–2998. [9] M. Stomer, H.-U. Krebs, Structure of laser deposited alloys, J. Appl. Phys. 78 (1995) 7080–7087. [10] G. Xiao, J.Q. Wang, Magnetic properties of metallic Co- and Fe-based granular alloys, J. Appl. Phys. 75 (10) (1994) 6604–6606. [11] J.M. Soares, J.H. de Aroujo, F.O.C. de Assis, T. Dunelow, M.M.X. Junior, J.M. Sasaki, Magnetic properties of the granular alloy Fe10 Ag90 as a function of annealing temperature, Mater. Res. 8 (2005) 347–350. [12] N. Thangaraj, C. Echer, M. Kannan, Krishnan, R.F.C. Farrow, R.F. Marks, S.S.P. Parkin, Giant magnetoresistance and microstructural characteristics of epitaxial Fe–Ag and Co–Ag granular thin films, J. Appl. Phys. 79 (1994) 6900–6902. [13] M.K. Roy, P.M.G. Nambissan, H.C. Verma, Structural, thermal stability and defect studies of Fe–Ag alloy prepared by electrodeposition technique, J. Alloys Compd. 345 (2002) 183–188. [14] K. Santhi, S.N. Karthick, H.-J. Kim, M. Nidhin, V. Narayanan, A. Stephen, Microstructure analysis of the ferromagnetic Ag–Ni system synthesized by pulsed electrodeposition, Appl. Surf. Sci. 258 (2012) 3126–3132. [15] L. Dimesso, G. Miehe, H. Fuess, H. Hahn, Preparation of nanostructured granular Ag–Co and Ag–Fe alloys by gas flow condensation technique, J. Magn. Magn. Mater. 191 (1999) 162–168. [16] N. Kataoka, K. Sumiyama, Y. Nakamura, Magnetic properties of highconcentration Ag–Fe alloys produced by vapour quenching, J. Phys. F: Met. Phys. 15 (1985) 1405–1411. [17] L.P. Bicellii, B. Bozzini, C. Mele, L. D’Urzo, A review of nanostructural aspects of metal electrodeposition, Int. J. Electrochem. Sci. 3 (2008) 356–408. [18] J. Garcia-Torres, E. Gómez, E. Vallés, Modification of magnetic and structural properties of Co and Co–Ag electrodeposits by sulphur incorporation, Mater. Chem. Phys. 122 (2010) 463–469. [19] X. Qingmei, Z. Yao, J. Liu, R. Hai, H.Y. Oderji, H. Ding, Synthesis and characterization of Ag–Ni bimetallic nanoparticles by laser-induced plasma, Thin Solid Films 519 (2011) 7116–7119. [20] J.F. Moulder, W.F. Stickle, P.E. Sobel, K.D. Bomben, in: J. Chastain, R.C. King Jr. (Eds.), Handbook of X-ray Photoelectron Spectroscopy, ULVAC-PHI Inc./Physical Electronics USA Inc., Japan/USA, 1993. [21] A. Lebugle, U. Axelson, R. Nyholm, N. Martensson, Experimental L and M core level binding energies for the metals 22 Ti to 30 Zn, Phys. Scr. 23 (1981) 825. [22] K. Asami, A precisely consistent energy calibration method for X-ray photoelectron spectroscopy, J. Electron Spectrosc. Relat. Phenom. 9 (1976) 469–478. [23] N.S. McIntyre, D.G. Zetaruk, X-ray photoelectron spectroscopic studies of iron oxides, Anal. Chem. 49 (1977) 1521–1529. [24] C.D. Wagner, D.A. Zatco, R.H. Raymond, Use of oxygen KLL Auger lines in identification of surface chemical states by electron spectroscopy for chemical analysis, Anal. Chem. 52 (1980) 1445. [25] N. Kataoka, K. Sumiyama, Y. Nakamura, Thermal stability of nonequilibrium crystalline Fe–Ag alloys produced by vapor quenching, Acta Metall. 37 (1989) 1135–1142. [26] B.D. Cullity, C.D. Graham, Introduction to Magnetic Materials, 2nd ed., John Wiley & Sons Inc., New Jersey, 2009. [27] S. Chikazumi, Physics of Magnetism, Wiley, New York, 1964, pp. 247.