Magnetic and Electronic Structure study of Fe/MgO/Fe/Co Multilayer Stack Deposited by E-Beam Evaporation
Present work investigates the magnetic and electronic structure of MgO/Fe/MgO/Fe/Co/Au multilayer stack grown on Si(100) substrates by electron beam evaporation method. X-ray diffraction study depicts polycrystalline nature of the multilayers. Results obtained from vibrating sample magnetometry (VSM) and near-edge X-ray absorption fine structure spectra (NEXAFS) at Fe & Co L- and Mg & O K-edges are applied to understand the magnetic and electronic properties of this stack and its interface properties. While the spectral features of Fe L-edge spectrum recorded by surface sensitive total electron yield (TEY) mode shows the formation of FeO at the Fe/MgO interface, the bulk sensitive total fluorescence yield (TFY), shows Fe in metallic nature. Co L-edge spectrum reveals the presence of metallic nature of cobalt in both TEY and TFY modes. Above results are well correlated with X-ray reflectometry.
pacs:68.55.-a, 61.05.cj, 68.35.-p, 81.05.Je, 75.70.Cn
Devices of multilayer structures are important in the development of spintronic devices based on giant magnetic resistance (GMR) and tunneling magnetorsistance (TMR), thermoelectric applicationsNakatani et al. (2011); Fesharaki et al. (2012); Lyle et al. (2010); Gajek et al. (2012); Walter et al. (2011); Liebing et al. (2011); Lee et al. (2012); Ikeda et al. (2008); Butler et al. (2001). TMR based devices, known as magnetic tunnel junction(MTJ), require MgO as a barrier layer and find applications in magnetic random access memories (MRAM) Lyle et al. (2010); Zhu and Park (2006). It has been shown that properties of these devices are sharply influenced by the interface composition and roughness of the multilayer stack. Till date 604% TMR has been reported Ikeda et al. (2008) which is less than the theoretically predicated value of TMR ( 1000%) Butler et al. (2001); Parkin et al. (204) and this discrepancy has been attributed to the interface properties of the heterostructures Parkin et al. (204); Timoshevskii et al. (2012). Hence, there is a need to investigate the growth and interfaces properties of Fe/MgO/Fe like structures in order to get a correlation among the interface and its bulk properties. Spectroscopic techniques like near edge X-ray absorption fine structure (NEXAFS) has emerged as a powerful tool for determination of local chemical structure. The technique is also sensitive to small strain induced variations in bond lengths Romanato et al. (1998) and efficient to detect atomic interdiffusion processes at the interfaces Boscherini et al. (2000). These aspects have been investigated by several authors in CoFeB/MgO/CoFeB structures Miyokawa et al. (2005); Saito et al. (2007); Djayaprawira et al. (2005); Rumaiz et al. (2010). Conclusion obtained from these various study lead to utilization of several methods which include insertion of pinning layers and annealing the whole structure in high vacuum for improving interface roughness and crystallinity of this multilayer stack Hindmarch et al. (2010); Hayakawa et al. (2006); Karthik et al. (2009). Besides the post deposition treatment, deposition methods also play important role in determining the properties of the multilayer stacks. Although methods like molecular beam epitaxy (MBE) and rf-sputtering are commonly employed, the most cost effective e-beam evaporation method may be an alternative tool for deposition of this type of multilayers Diao et al. (2010). The present work reports the deposition of Fe/MgO/Fe/Co/Au multilayer stack by e-beam evaporation and the local electronic structure by using NEXAFS techniques. A layer of Co was deposited above Fe layer to check the magnetic properties. Interface properties are discussed and correlated with magnetic and electronic properties.
I Experimental Details
i.1 Deposition of multilayer structures
Multilayer stack of MgO/Fe/MgO/Fe/Co/Au was deposited on Si(100) substrates by e-beam evaporation method with base pressure better than Torr. Si substrates were cleaned in trichloroethylene, acetone, isopropyl alcohol and de-ionized water in order to remove chemical impurity and surface contamination. Si substrate was also dipped into hydrogen fluoride (HF) for 2-3 minute to remove any native SiO. For the deposition of MgO layers, MgO powder (purity %, Alfa-Aesar) was pressed into pellet form and evaporated. Fe, Co and Au targets were used for metal layer deposition. All these targets were kept inside the chamber, so that deposition of all the layers could be done without breaking vacuum. This is essential to avoid any contaminations of interfaces. All the layers were deposited with deposition rate of nm/s with online monitoring of thickness using quartz crystal monitor. First, MgO buffer layer was deposited on Si(100) substrate in order to prevent silicide formation at Si/Fe interface due to diffusion. MgO layer also works as a good buffer layer to grow epitaxial Fe layerButler et al. (2001). The substrate temperature was kept at C during deposition. On top of MgO buffer layer, Fe thin film was deposited at 180 C. Subsequently, MgO barrier layer was deposited at the same growth temperature. The upper Fe layer was deposited at temperature of C. Further Co layer was deposited at C. This multilayer structure was annealed at C for hr in ultra high vacuum to improve crystallinity and make interfaces sharp. After annealing a capping layer of Au was deposited at room temperature in order to prevent Co from oxidation.
i.2 Characterization Techniques
Rutherford backscattering spectroscopic (RBS) study for these multilayer structure was performed on RBS spectrometer (Inter University Accelerator Centre (IUAC), New Delhi). It is observed that all the layers exhibit good stoichimety and the detailed study has been published elsewhere Singh et al. (2012a). X-ray diffraction study on multilayer structure was carried out on a Philips X-ray diffractometer with step time of 2.5 s and step size of 0.05 was used for recording the XRD pattern. Magnetic studies on the multilayers were carried out using a vibrating sample magnetometer (VSM) at National Physical Laboratory, New Delhi. Further, electronic structure was investigated by near-edge X-ray absorption fine structure (NEXAFS) at the high energy spherical grating monochromator (HSGM) BL20A1 beamline in the National Synchrotron Radiation Research Center (NSRRC) in Taiwan. All measurements were processed in an ultra high vacuum (UHV) chamber ( Torr) at K, in (a) total electron yield (TEY, surface sensitive), measured by monitoring the total sample photocurrent, and (b) total fluorescence yield (TFY, bulk sensitive), measured with a negatively biased microchannel plate(MCP), simultaneously. Measurements in TEY and TFY modes probe the electronic structure of materials with a probing depth 4 nm and several hundred nanometers, respectively Stöhr (1996). Around the edge energy in NEXAFS spectrum, data points with an energy spacing of eV were recorded, using a sec. collection interval per point. The incoming radiation flux (I) was monitored by the total photocurrent produced in a highly transmissive Au-mesh inserted into the beam. The overall photon resolution around O K-edge was eV using m slits. After a constant background substraction, all spectra were normalized to the post-edge step height using Athena 0.0.061 Ravel and Newville (2005). X-ray reflectometry (XRR) has also been carried out for these structures and the results obtained were simulated by using GenX 2.0b reflectivity software Björck and Andersson (2007); Raju et al. (2011).
Ii Results and Discussion
ii.1 Structural and Magnetic study
Figure 1 shows the XRD pattern of Fe/MgO/Fe/Co multilayer. The peaks in the XRD pattern has been identified to come from body centered Fe [JCPDS-06-0696], rocksalt MgO [JCPDS-78-0430], and cubic close packed Au [JCPDS 04-0784]. No peak is observed corresponding to the phase of Co. Hence, it may be contemplated that Co layer is amorphous in nature. For better visualization of structure of various layer and their sequence, a schematic figure has been shown in Fig 1:inset. In Table 1 we have shown the thickness of various layers as simulated by RBS. Figure 2 shows the magnetic hysteresis curves of the multilayer stack while applying the magnetic field parallel and perpendicular to the film surface. Both these hysteresis are almost symmetric. Parallel hysteresis is very much similar to the pure bcc Fe Huang et al. (1993). It saturates only at the value of Oe and has the value of saturation magnetic moment(m) emu. The loop is also closed at both ends. The coercivity(H) and remanence(m) are Oe and emu, respectively. The hysteresis curves measured in perpendicular does not saturate upto kOe and have almost paramagnetic-like behavior after kOe. The coercivity and remanence of this stack are Oe and emu, respectively. This hysteresis exhibits two step behaviour, which is may be due to presence of insulating layer (MgO) between two ferromagnetic (FM) layers. Absence of sharp steps in this hysteresis may be due to polycrystalline nature and oxidation of Fe. This multilayer stack exhibits the in-plane anisotropy with squareness ratio m/m 1.
ii.2 Near Edge X-ray Absorption Fine Structure (NEXAFS) study
Figure 3 shows the Fe L-edge NEXAFS spectra for the multilayer system measured in TEY and TFY mode along with FeO as a reference. NEXAFS spectra recorded for Fe L-edge in TEY mode show the peaks at 703, 705, 716 and 718 eV. The Fe L-edge spectrum arises due to Fe 2 core level, which in effect of spin-orbit coupling, give rise to degenerate state 2 and 2 showing multiplets centered on and eV. These octahedral crystal fields lifts the degeneracy of the 2 and 2 levels so that two levels with t and e symmetry are created, as indicated by the two structures at about and eV and at and eV. These structures are indicative of Fe oxidation state and generally observed in Fe-based oxide systems Wilke et al. (2001). Since, TEY mode is surface sensitive, hence it may be contemplated that MgO/Fe interfaces in multilayers have oxidized Fe at interface. The oxidation of upper Fe layer cannot expect in the present case because this layer is capped by Co and Au layer on top. The bulk sensitive TFY mode Fe L-edge spectrum exhibits peaks at and eV and indicates that Fe is not bonded with oxygen. This shows the metallic nature of Fe. Mg K-edge spectrum exhibits peaks around the position , , and eV (Fig. 4(b) that are associated primarily with Mg 1 to 3 states Yoshida et al. (1995) and similar to MgO spectrum.
Figure 5 shows O K-edge NEXAFS spectra in TEY and TFY mode of the multilayer along with MgO for comparison. Out of these, TFY spectrum consists of pre-edge structure around eV and these structures generally due to the excitation to the localized bound state and consistent with the previous reports Singh et al. (2012b). The observed spectral features in this spectrum, around , , and eV are very much similar to the spectrum of MgO thin films reported by Luchs et al. Luches et al. (2004). A visual inspection shows large difference in the TEY and TFY spectra of the samples. In TEY spectrum, we observe peaks around , , , , eV. These features are very much similar to ferrites Abbate et al. (1992); Regan et al. (2001), where the spectra originate from transitions into unoccupied states with O 2 character hybridized with metal ions. This is in accordance with the formation of Fe-O at the Fe/MgO and MgO/Fe interfaces. Fig. 4(a) shows the Co L-edge spectrum of Fe/MgO/Fe/Co/Au structure with the peaks occurring at and eV. Spectral feature is very much analogues to the metallic Co Li et al. (Nov). The spectra have the same intensity in both the TEY and TFY mode. This indicate that Co atom was not bonded with Fe atom consistent with RBS study Singh et al. (2012a), however the possibility of diffusion of Co metal into Fe cannot be neglected. The results obtained from NEXAFS spectra confirms the assumption of possible oxidation of Fe layers at the interface.
ii.3 X-Ray Reflectivity(XRR) study
Above results obtained from VSM and NEXAFS analysis provide strong evidence for oxidation at Fe/MgO/Fe interface and further characterization of thickness and interface was carried out by using XRR. Figure 6 shows XRR pattern for multilayer and the fitted curves. Parameters used to simulate the experimental data are given in Table 1. Data from the Table shows that the thickness of layers is different from that are obtained from RBS study. Besides this formation of Fe-Co alloy also expected at Fe/Co interface. This corroborates results obtained from the RBS analysis discussed elsewhere Singh et al. (2012a). The barrier layer of multilayer is nm, which is very large compared to conventional MTJ stack. Due to this large thickness, de-coupling between two FM layers is expected and different coercivities of upper and lower FM layers should be clearly observed but this effect could not be observed in the hysteresis curve of multilayer. It is expected that coercivity is also affected by the polycrystalline nature of various layers and oxidation at interfaces inhibiting the presence of magnetic switching between these two layers. Coercivity of any magnetic thin film depends on microstructure of film, thickness and also the orientation of film growth Li et al. (Nov); Harrell (2001); Li et al. (1998). A detailed investigation on Fe film shows that due to presence of magnetocrystalline anisotropy, coercivity is different for different plane Boubeta et al. (2003). Hence, it may be contemplated that due to same nature of growth both the Fe film have almost same value of coercivity.
Fe/MgO/Fe/Co/Au multilayer stacks were deposited by e-beam evaporation method in ultra high vacuum. Magnetization of this stack is similar to the pure bcc Fe, even though a thick MgO barrier is present between the ferromagnetic layers. NEXAFS study performed on this stack shows the oxidization of Fe/MgO interface, which is also confirmed by XRR. Metallic nature of Fe (bulk) and Co layers were observed by using NEXAFS study.
Acknowledgements.JPS is thankful to Prof. J. S. Moodera, MIT USA for fruitful discussion during the fabrication of multilayer stack. Authors are also thankful to Mr. S.R. Abhilash for providing experimental support during the e-beam deposition of stack. Synchrotron experiments at NSRRC, Taiwan are supported by the KIST (Grant No. 2V02631) and the Korean Synchrotron User Association (KOSUA).
- Nakatani et al. (2011) T. M. Nakatani, S. Mitani, T. Furubayashi, and K. Hono, Appl. Phys. Lett. 99, 182505 (pages 3) (2011).
- Fesharaki et al. (2012) M. J. Fesharaki, L. Péter, T. Schucknecht, D. Rafaja, J. Dégi, L. Pogány, K. Neuróhr, E. Széles, G. Nabiyouni, and I. Bakonyi, J. Electrochem. Soc 159, D162 (2012).
- Lyle et al. (2010) A. Lyle, J. Harms, S. Patil, X. Yao, D. J. Lilja, and J.-P. Wang, Appl. Phys. Lett. 97, 152504 (pages 3) (2010).
- Gajek et al. (2012) M. Gajek, J. J. Nowak, J. Z. Sun, P. L. Trouilloud, E. J. O’Sullivan, D. W. Abraham, M. C. Gaidis, G. Hu, S. Brown, Y. Zhu, et al., Appl. Phys. Lett. 100, 132408 (pages 3) (2012).
- Walter et al. (2011) M. Walter, J. Walowski, V. Zbarsky, M. Münzenberg, M. Schäfers, D. Ebke, G. Reiss, A. Thomas, P. Peretzki, M. Seibt, et al., Nature Mater. 10, 742 (2011).
- Liebing et al. (2011) N. Liebing, S. Serrano-Guisan, K. Rott, G. Reiss, J. Langer, B. Ocker, and H. W. Schumacher, Phys. Rev. Lett. 107, 177201 (2011).
- Lee et al. (2012) H.-M. Lee, Y.-C. Lee, H.-H. Chen, L. Horng, J.-C. Wu, C.-M. Lee, T.-H. Wu, and G. Chern, SPIN 02, 1230002 (2012).
- Ikeda et al. (2008) S. Ikeda, J. Hayakawa, Y. Ashizawa, Y. M. Lee, K. Miura, H. Hasegawa, M. Tsunoda, F. Matsukura, and H. Ohno, Appl. Phys. Lett. 93, 082508 (pages 3) (2008).
- Butler et al. (2001) W. H. Butler, X.-G. Zhang, T. C. Schulthess, and J. M. MacLaren, Phys. Rev. B 63, 054416 (2001).
- Zhu and Park (2006) J.-G. J. Zhu and C. Park, Mater. Today 9, 36 (2006), ISSN 1369-7021.
- Parkin et al. (204) S. S. P. Parkin, C. Kaiser, A. Panchula, P. M. Rice, B. Hughes, M. Samant, and S.-H. Yang, Nat. Mater. 3, 862 (204).
- Timoshevskii et al. (2012) V. Timoshevskii, Y. Hu, É. Marcotte, and H. Guo, ArXiv e-prints (2012), eprint 1212.1374.
- Romanato et al. (1998) F. Romanato, D. De Salvador, M. Berti, A. Drigo, M. Natali, M. Tormen, G. Rossetto, S. Pascarelli, F. Boscherini, C. Lamberti, et al., Phys. Rev. B 57, 14619 (1998).
- Boscherini et al. (2000) F. Boscherini, G. Capellini, L. D. Gaspare, F. Rosei, N. Motta, and S. Mobilio, Appl. Phys. Lett. 76, 682 (2000).
- Miyokawa et al. (2005) K. Miyokawa, S. Saito, T. Katayama, T. Saito, T. Kamino, K. Hanashima, Y. Suzuki, K. Mamiya, T. Koide, and S. Yuasa, Jpn. J. Appl. Phys. 44, L9 (2005).
- Saito et al. (2007) T. Saito, T. Katayama, T. Ishikawa, M. Yamamoto, D. Asakura, and T. Koide, Appl. Phys. Lett. 91, 262502 (pages 3) (2007).
- Djayaprawira et al. (2005) D. D. Djayaprawira, K. Tsunekawa, M. Nagai, H. Maehara, S. Yamagata, N. Watanabe, S. Yuasa, Y. Suzuki, and K. Ando, Appl. Phys. Lett. 86, 092502 (pages 3) (2005).
- Rumaiz et al. (2010) A. K. Rumaiz, J. C. Woicik, W. G. Wang, J. Jordan-Sweet, G. H. Jaffari, C. Ni, J. Q. Xiao, and C. L. Chien, Appl. Phys. Lett. 96, 112502 (pages 3) (2010).
- Hindmarch et al. (2010) A. T. Hindmarch, K. J. Dempsey, D. Ciudad, E. Negusse, D. A. Arena, and C. H. Marrows, Appl. Phys. Lett. 96, 092501 (pages 3) (2010).
- Hayakawa et al. (2006) J. Hayakawa, S. Ikeda, Y. M. Lee, F. Matsukura, and H. Ohno, Appl. Phys. Lett. 89, 232510 (pages 3) (2006).
- Karthik et al. (2009) S. V. Karthik, Y. K. Takahashi, T. Ohkubo, K. Hono, S. Ikeda, and H. Ohno, J. Appl. Phys. 106, 023920 (pages 6) (2009).
- Diao et al. (2010) Z. Diao, J. F. Feng, H. Kurt, G. Feng, and J. M. D. Coey, Appl. Phys. Lett. 96, 202506 (pages 3) (2010).
- Singh et al. (2012a) J. P. Singh, M. Raju, K. Asokan, J. Prakash, D. Kabiraj, S. R. Abhilash, S. Chaudhary, and D. Kanjilal (AIP, 2012a), vol. 1447(1), pp. 749–750.
- Stöhr (1996) J. Stöhr, NEXAFS Spectroscopy (Springer, Berlin, 1996).
- Ravel and Newville (2005) B. Ravel and M. Newville, Journal of Synchrotron Radiation 12, 537 (2005).
- Björck and Andersson (2007) M. Björck and G. Andersson, J. Appl. Cryst. 40, 1174 (2007).
- Raju et al. (2011) M. Raju, S. Chaudhary, and D. K. Pandya, Appl. Phys. Lett. 98, 212506 (pages 3) (2011).
- Huang et al. (1993) Y. Y. Huang, C. Liu, and G. P. Felcher, Phys. Rev. B 47, 183 (1993).
- Wilke et al. (2001) M. Wilke, F. Farges, P. Petit, G. Brown, and F. Martin, Am. Minral. 86, 714 (2001).
- Yoshida et al. (1995) T. Yoshida, T. Tanaka, H. Yoshida, T. Funabiki, S. Yoshida, and T. Murata, J. Phys. Chem. 99, 10890 (1995).
- Singh et al. (2012b) J. P. Singh, I. Sulania, J. Prakash, S. Gautam, K. H. Chae, D. Kanjilal, and K. Asokan, Adv. Mat. Lett. 3, 112 (2012b).
- Luches et al. (2004) P. Luches, S. D’Addato, S. Valeri, E. Groppo, C. Prestipino, C. Lamberti, and F. Boscherini, Phys. Rev. B 69, 045412 (2004).
- Abbate et al. (1992) M. Abbate, F. M. F. de Groot, J. C. Fuggle, A. Fujimori, O. Strebel, F. Lopez, M. Domke, G. Kaindl, G. A. Sawatzky, M. Takano, et al., Phys. Rev. B 46, 4511 (1992).
- Regan et al. (2001) T. J. Regan, H. Ohldag, C. Stamm, F. Nolting, J. Lüning, J. Stöhr, and R. L. White, Phys. Rev. B 64, 214422 (2001).
- Li et al. (Nov) J. Li, R. Sinclair, S. Rosenblum, and H. Hayashi, Magnetics, IEEE Transactions on 30, 4050 (Nov).
- Harrell (2001) J. Harrell, IEEE Transactions on Magnetics 37, 7049 (2001).
- Li et al. (1998) X. W. Li, A. Gupta, G. Xiao, and G. Q. Gong, J. Appl. Phys. 83, 7049 (1998).
- Boubeta et al. (2003) C. M. Boubeta, J. L. Costa-Kramer, and A. Cebollada, J. Phys. Cond. Matter. 15, R1123 (2003).