Carbon doped ZnO: Synthesis, characterization and interpretation ...

Journal of Magnetism and Magnetic Materials 329 (2013) 146–152

Contents lists available at SciVerse ScienceDirect

Journal of Magnetism and Magnetic Materials journal homepage: www.elsevier.com/locate/jmmm

Carbon doped ZnO: Synthesis, characterization and interpretation D.K. Mishra a,n, J. Mohapatra a, M.K. Sharma b,2, R. Chattarjee b, S.K. Singh c, Shikha Varma d, S.N. Behera e,f,1, Sanjeev K. Nayak g, P. Entel g a

Department of Physics, Institute of Technical Education and Research, Siksha ‘O’ Anusandhan University, Bhubaneswar 751030, Odisha, India Department of Physics, Indian Institute of Technology, IIT Delhi, Delhi 110016, India c Advanced Materials Technology Department, Institute of Minerals and Materials Technology (CSIR), Bhubaneswar 751013, Odisha, India d Institute of Physics, Schivalaya Marg, Bhubaneswar 751004, India e School of Electrical Sciences, Indian Institute of Technology, IIT Bhubaneswar, Bhubaneswar 751013, India f National Institute of Science and Technology (NIST), Berhampur 761008, Odisha, India g Faculty of Physics and CeNiDE, University of Duisburg-Essen, 47057 Duisburg, Germany b

a r t i c l e i n f o

a b s t r a c t

Article history: Received 14 August 2012 Received in revised form 22 September 2012 Available online 22 October 2012

A novel thermal plasma in-flight technique has been adopted to synthesize nanocrystalline ZnO and carbon doped nanocrystalline ZnO matrix. Transmission electron microscopy (TEM) studies on these samples show the average particle sizes to be around 32 nm for ZnO and for carbon doped ZnO. An enhancement of saturation magnetization in nanosized carbon doped ZnO matrix by a factor of 3.8 has been found in comparison to ZnO nanoparticles at room temperature. Raman measurement clearly indicates the presence of Zn–C complexes surrounded by ZnO matrix in carbon doped ZnO. This indicates that the ferromagnetic signature in carbon doped ZnO arises from the creation of defects or the development of oxy-carbon clusters, in the carbon doped ZnO system. Theoretical studies based on density functional theory also support the experimental analyses. & 2012 Elsevier B.V. All rights reserved.

Keywords: Nanostructure Plasma deposition Raman spectroscopy Photoelectron spectroscopy Magnetic Property

1. Introduction It is well known that the phenomena of spin-dependent transport in metallic ferromagnetic multilayers have revolutionized the technology in computer industry. This success is also anticipated for semiconductor spintronic devices where the main aim is to use the spin-dependent phenomena in conventional semiconductor heterostructures as well as in hybrid systems that combine magnetic and semiconducting properties [1–3]. The primary requirement to build a hybrid system is to find a reliable material that shows both semiconducting as well as ferromagnetic properties at the same time. The diluted magnetic semiconductors (DMS) are the materials of interest toward this goal [4]. The study of DMS is an emerging field of research and a lot of effort has been focused into it [5–9]. Out of many proposed materials, only few cases, like GaMnAs and EuO, have so far been observed to show reproducible DMS properties with ferromagnetism far below the room temperature (RT) [10–12]. Since a practical DMS device invariably needs ferromagnetism at

n

Corresponding author. Tel.: þ91 9778896120. E-mail address: [email protected] (D.K. Mishra). 1 Deceased. 2 Present address: Department of Applied Physics, Amity Institute of Applied Sciences, Amity University, Noida 201301, India. 0304-8853/$ - see front matter & 2012 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.jmmm.2012.09.058

RT, the search of new materials showing RT ferromagnetism is seriously being pursued. The prediction by Dietl et al., that oxide and nitride semiconductors like ZnO and GaN can show ferromagnetism above RT when doped with Mn at sufficiently large hole concentration [7], has put lot of attention on ZnO based DMS. There is active interest not only in Mn as a dopant but also on other transition metals as dopants [13,14]. To achieve RT-ferromagnetic DMS one must deal with two main aspects of magnetism. Firstly, there must be a way to achieve local magnetic moment and secondly the local moment must interact with each other and the correlation must lead to RT ferromagnet. The first part is easy to conceive in TM doped ZnO because the unpaired d-orbitals of TM ions can give rise to local moment with magnetic moment equal to the number of unpaired electrons (a feature expected to be more prominent in semiconducting host matrix). Almost all the reports in literature consistently agree upon it. However, it is in the second aspect, i.e. how the local moments interact, that the reports do not converge to any clear conclusion. While some reports claim ferromagnetism, other reports claim paramagnetic and there are other claims of anti-ferromagnetic interactions prevailing in the system. The condition necessary to achieve ferromagnetism in ZnO is still not clear and is an active topic for research. Based on atomistic characterization techniques there are recent claims for segregated phase like TM clusters or some complex constituents as the origin

D.K. Mishra et al. / Journal of Magnetism and Magnetic Materials 329 (2013) 146–152

of ferromagnetism. It must be stressed that the degrees of reproducibility of ferromagnetic samples are rather very low. For example, Ueda et al. report that one out of ten samples (10%) developed by them showed ferromagnetic behavior [15]. This shows how subtle the search for magnetism in these materials can be and the role of intrinsic or extrinsic defects, if any, has not been thoroughly established. Added to this, there are other claims of RT ferromagnetism in pure ZnO induced from intrinsic defects [16–21]. So it is essential that one pays special care for characterizing the samples in order to avoid any biased conclusions. There are recent claims for RT ferromagnetism in carbon doped ZnO [22–27]. In line with the understanding of magnetism, as stated before, the question about the origin of magnetic moment does arise since there are no TM elements doped in the samples. There are also other reports of defect-induced ferromagnetism in pure ZnO samples, which clearly shows that the role of intrinsic defects cannot be neglected. Present study is for the case of carbon doped ZnO. In what follows, a novel technique for the synthesis of carbon doped ZnO using an inflight plasma reactor is presented. Room temperature ferromagnetism is observed in the samples prepared using this technique. Theoretical studies based on density functional theory (DFT) have been carried out in order to understand the experimental observation. The results are discussed keeping in view the possibility of induced lattice defects due to the incorporation of carbon in the ZnO matrix and their relation to the observed ferromagnetism. Apart from magnetism, this material could be promising because of its luminescence behavior and other optical properties [28–31] and also for photocatalytic activity as demonstrated for ZnO–carbon hetero-architectures [32].

2. Experimental procedure Nanocrystalline powder of ZnO and carbon doped ZnO matrices were synthesized using the thermal plasma in-flight technique. Commercially available ZnO (99.9% purity) and graphite powder with mole ratio 1:8 were mixed in an agate mortar. The mixture was fed to the thermal plasma reactor with a flow rate of 1 g/min for the synthesis of nanocrystalline carbon doped ZnO powder. Pure ZnO powder was fed to the reactor for the synthesis of nanocrystalline ZnO. The reactor assembly consists of a material or precursor feeder for feeding the raw material to the high temperature plasma zone for reaction, vaporization and dissociation in a hearth. A schematic picture of the thermal plasma reactor is shown in Fig. 1. Extended arc plasma is generated in between two graphite electrodes arranged in a vertical configuration in a sealed water-cooled stainless steel chamber. The argon plasma forming gas is inserted through the axial hole of the top electrode. The plasma vapor is then transported to the cooler zone and deposited on the inner surface of the water-cooled reactor chamber. In real operation the plasma arc is struck between the graphite electrode and the graphite crucible, which act as the cathode and anode, respectively. Plasma was generated through extended arcs by adjusting the distance between the crucible and the movable electrode. The reactor is unique in the sense that it utilizes both the non-transferred mode (when the powder travels through the plasma) and transferred mode (when the powder is at the bottom of the crucible) of the plasma. In case of non-transferred mode of plasma operation, the finer fraction of the feed particle is evaporated during in-flight processing where chemical reaction and primary particle formation occur. Subsequently, the particles grow and deposit on the cold wall of the reactor chamber or collect in a powder collection chamber. In case of transferred mode, the feed particles, which do not evaporate during the in-

147

flight processing fall to the bottom of the crucible and the arc is transferred to the particles, which now constitute and act as the part of the electrode. Because of the extremely high heat input into the material acting as the electrode, the material vaporizes and the vapor is then cooled to induce the formation of nanopowder. In the present investigation the samples were developed by the transferred arc mode and the nano-powder is formed almost instantly due to high temperature and high quenching rate associated with thermal plasma. Further details on the thermal plasma reactor can be found elsewhere [33]. To produce the thermal plasma, 15 kW DC power (50 V applied voltage and 300 A current) was used in the form of an extended arc. The plasma forming gas (argon) was allowed to flow into the chamber through a narrow hole in the upper electrode (cathode). The feeder powder was treated by the thermal plasma produced with argon gas with a flow rate of 1 l/min. During the preparation process the plasma power, argon gas flow rate and electrode arcing time were kept fixed. The vaporized powder got deposited in the chamber and top cover of the plasma reactor. The powder was collected and prepared for various characterization studies like X-ray diffraction (XRD) to determine the crystal structure, transmission electron microscopy (TEM) to find the particle size and crystalline ordering, energy dispersive X-ray (EDX) analysis to find the presence of different elemental impurities and micro-Raman spectroscopy to determine the structural information. Atomic absorption spectroscopic (AAS) studies were carried out to estimate the percentage of magnetic impurities (Fe, Co and Ni). X-ray photoelectron spectroscopy (XPS) measurements were performed using a VG ESCA system. This system is equipped with a dual Mg–Al anode and has a base pressure of 1.0 10 10 Torr. The system resolution is 0.9 eV. XPS measurements were carried out using the Mg Ka X-ray source with pass energy of 20 eV. The magnetization measurements were carried out using a Quantum Design Superconducting Quantum Interference Device (SQUID) magnetometer.

3. Results and discussions The XRD peaks of nanocrystalline ZnO as well as that of carbon doped ZnO are shown in Fig. 2 and compared with Joint Committee on Powder Diffraction Standards (JCPDS) data (no. 890510). All peaks found in carbon doped ZnO system correspond to the wurtzite structure. Free carbon peaks are also found in the carbon doped system. The average crystallite size determined from the X-ray analysis is on the order of 20 nm, which compares well to the particle size of 32 nm obtained from the TEM analysis. Lattice parameters of ZnO nanoparticles and C-doped ZnO nanoparticles are calculated 2 2 2 2 using the formula, 1=dhkl ¼ 4=3ððh þ hkþ k Þ=a2 Þ þðl =c2 Þ. The lattice parameter obtained for ZnO nanoparticle is a¼3.244 A˚ and c¼5.198 A˚ which is in close agreement with other reported values [22]. The lattice parameter for carbon doped ZnO matrix was ˚ The c/a obtained for estimated to be a¼3.242 A˚ and c¼5.234 A. carbon doped ZnO (1.6144) is slightly higher than those obtained for pure ZnO nanoparticles (1.6025). The TEM micrograph of the nanoparticles shown in Fig. 3 indicates that the nanoparticles of both pure ZnO and C doped ZnO are spherical in shape and size. The average particle size is approximately 32 nm. The energy dispersive X-ray data shown in Fig. 4 for C doped ZnO confirms that the nanoparticles are composed of Zn and O without the presence of any impurities. Along with Zn and O peaks, the spectrum shows the presence of Cu and C. The exact percentage of carbon incorporated in the carbon doped ZnO matrix cannot be estimated by this method because of the carbon coated Cu grid used for the TEM analysis. The atomic absorption spectroscopic analysis (AAS) reveals

148


Fig. 1. Schematic picture of DC extended arc inflight thermal plasma reactor.

Fig. 2. XRD patterns of nanocrystalline ZnO and carbon doped ZnO system.

the presence of magnetic impurities (Fe, Co and Ni) content upto 25–30 ppm in nanocrystalline ZnO and C doped ZnO which could not be estimated from EDX analysis. The magnetization (M) versus magnetic field (H) curve, shown in Fig. 5, shows that pure ZnO nanoparticles have ferromagnetic properties. The ferromagnetism in C doped ZnO is enhanced and is above the room temperature (RT). For pure ZnO nanoparticles, the value of saturation magnetization is estimated to be 2.7 10 2 emu/g with remnant magnetization of 3.8 10 3 emu/g and the coercivity of 114 Oe. On the other hand for the carbon doped ZnO the saturation magnetization is estimated to be 1.03 10 1 emu/g with a remnant magnetization of 2.17 10 2 emu/g and the coercivity of 220 Oe. This is 3.8 times higher than the saturation magnetization of pure ZnO nanoparticles. It must be noted that the percentage of magnetic impurities present in the sample i.e. within a limit of 25–30 ppm, could give rise to a maximum saturation magnetic moment of 4.643 10 3 emu/g if interacted ferromagnetically. This value is much less than the magnitude observed for nanocrystalline ZnO and C doped ZnO. Thus the role of magnetic impurities in the enhancement of ferromagnetic signal in C doped ZnO with respect to pure ZnO may be discarded. The magnitudes of the saturation magnetization obtained for nanocrystallline ZnO as well as for

carbon doped ZnO are found to be larger than those previously reported for the carbon doped ZnO films synthesized by the pulsed laser deposition technique [34]. Thus we may assert that intrinsic defects may be responsible for the magnetic properties obtained in our samples. It must be noted that room-temperature ferromagnetic signals from single crystal ZnO have also been reported [17], which is attributed to intrinsic defects. One of the common trends we note from literature is that with increase in carbon concentration the magnetization of the sample reduces [23–25]. The saturation magnetization obtained at 300 K is comparable to that of those reported in literature [23–25]. Fig. 6 shows the Raman spectra of nanocrystalline ZnO and carbon doped ZnO matrix. From the micro-Raman spectrum of nanocrystalline ZnO we find that the peak corresponding to the E2 (high) vibration mode, which is the band characteristic of the wurtzite phase of ZnO, depends strongly on the isotropic composition of ZnO and is centered at 437 cm 1. This is consistent with the conclusions of structural analysis obtained from the XRD spectra of ZnO. Apart from this, a distinct sharp peak together with a shoulder peak was found at 581.8 cm 1, which is identified as the 1LO (A1/E1) mode. It is claimed [35] that the Raman peak at 582 cm 1 is characteristic of nanocrystalline ZnO. Normally, Raman peaks observed in the frequency range from 570 cm 1 to 590 cm 1 are considered to be associated with structural disorder, such as oxygen vacancy, Zn interstitial and their combination, due to their strong dependence on the oxygen stoichiometry [36–38]. Therefore the observation of the sharp peak and its shoulder at 581.8 cm 1 as depicted in Fig. 6 clearly indicates the presence of defects in the specimen. It is possible that the vacancies on the surface of nanoparticles interact with nearby vacancies forming vacancy clusters, which may explain the origin of the peak associated with the 1LO mode. The other peaks at 331.37 cm 1, 512.4 cm 1 and 643.95 cm 1 are assigned to possible multi-phonon-scattering modes and the peak at 276.02 cm 1 is attributed to the activation of the normally dormant B2 modes. The peaks around 183.76 cm 1 and 385 cm 1 are related to the A1 TO modes that are activated due to the presence of defects [39,40]. It can be seen from Fig. 6 that the Raman spectrum of nanocrystalline carbon doped ZnO shows only two prominent peaks ocurring at 437 cm 1 and 579.9 cm 1 together with other


149

Fig. 3. TEM picture of nanocrystalline (a) ZnO and (b) carbon doped ZnO system.

Fig. 6. Raman spectra of nanocrystalline ZnO and nanocrystalline carbon doped ZnO. Fig. 4. EDX spectrum of nanocrystalline carbon doped ZnO matrix.

Fig. 5. Magnetization versus. magnetic field (M–H) curves at T¼ 300 K for nanocrystalline ZnO and nanocrystalline carbon doped ZnO.

peaks with relatively lower intensity. The peak around 437 cm 1 corresponds to the E2 (high) of ZnO which is the characteristic signature of the wurtzite structure, as already mentioned. The second peak observed at 579.9 cm 1 is of higher intensity as compared to that of the E1 (LO) mode of pure ZnO nanocrystallites. The peak in pure ZnO at 581.8 cm 1 has already been attributed to intrinsic defects. A small lowering in wavenumber of only 0.9 cm 1 observed between the two peaks may be due to the zinc carbon bonds. It must be noted that in nanocrystalline ZnO this peak is lower in intensity as compared to the E2 (high) peak. Considering the fact that the preparation procedure involves about eight times the carbon as compared to ZnO, such abundance of ZnC cannot be ruled out. Inspite of the strong presence of ZnC structures in the ZnO matrix, the associated enhancement of the magnetization in the carbon doped ZnO sample cannot be attributed to ZnC or the segregated carbon present in the matrix. This conclusion stems from the observation of low magnetization in ZnC powder prepared by the same technique. The wide survey scan of the X-ray photoelectron spectrum of nanocrystalline carbon doped ZnO sample is shown in Fig. 7a. Apart from the main peaks corresponding to Zn, O and C, the O auger peaks around 766.14 eV, 742.19 eV, 553.13 eV and the Zn auger peaks around 486.73 eV, 469.10 eV, 430.16 eV, 355.7 eV, 330.45 eV, 265.49 eV and 244.14 eV are observed in the survey scan. Besides these, no other peaks corresponding to the core

150


d- or f-levels are detected in the XPS spectrum within the detection limit of 0.1–1 atomic percentage. XPS spectrum of C1s is shown in Fig. 7b. The 1s carbon peak is fitted with four component Gaussians: the first at 283.62 eV corresponds to the carbon bound to Zn, the second at 284.75 eV is assigned to the free graphitic carbon peak, the third at 285.97 eV is assigned to the zinc oxy-carbide complex and fourth at higher binding energy 287.36 eV is attributed to carbon oxygen bonds (i.e. C–O and CQO) [41]. The Zn 2p spectra shown in Fig. 7c contain a doublet whose binding energies are 1022.02 eV and 1045.15 eV, assigned as Zn2p3/2 and Zn2p1/2 lines, respectively. The binding energy difference between both the lines is 23.13 eV and is within the standard reference value of ZnO. The binding energies and the binding energy difference show that Zn ions are in þ2 oxidation state. The deconvolution of Zn2p3/2 peak gives rise to two components at around 1021.9 eV and 1022.5 eV The former is ascribed to Zn–O–C in carbon doped ZnO matrix whereas the latter is ascribed to Zn–O bond in ZnO. The O1s peak of the XPS spectra is deconvoluted to understand the oxidation state of oxygen ion. The deconvolution of the O1s peak, shown in Fig. 7d, reveals three components at 530.06 eV, 530.84 eV and 532.32 eV [42]. The lower binding energy component at 530.06 eV is attributed to the wurtzite structure of hexagonal Zn2 þ ion of the metal oxide [43]. The middle binding energy component in the spectrum at 530.84 eV is associated with the oxy-carbon clusters and the higher binding energy component at 532.32 eV is attributed to chemisorbed oxygen,

possibly from hydroxyl species. In literature [43], the binding energy associated with the middle peak at 530.84 eV is attributed to the presence of oxygen ions in the oxygen deficient regions within the matrix of ZnO indicating an environment with O vacancies. This illustrates the presence of oxygen deficiency in the carbon doped ZnO nanoparticles.

4. Density-functional theory calculations In order to understand the origin of magnetic properties, the electronic structure of carbon doped ZnO has been studied with the density functional theory within the plane-wave formalism using the Vienna ab-initio simulation package (VASP) [44,45]. The generalized gradient approximation (GGA) [46] is used as the exchange-correlation functional. It is well known that GGA is not sufficient in reproducing one of the important properties of semiconductors, namely the band gap; as a consequence the estimation of the energy of impurity states with respect to the Fermi energy is usually not correct. Nonetheless, GGA is a powerful tool to explore the magnetic properties of a system and has been successful in predicting the complicated magnetic properties of many materials, especially those of metals and their alloys. The lattice constants used in these calculations are the minimum energy lattice parameters obtained from GGA [47], which are comparable to the lattice parameters obtained from XRD analysis stated above. The atomic positions in the supercells have been

Fig. 7. (a) Wide survey X-ray photoelectron spectrum of nanocrystalline carbon doped ZnO. (b) C1s core level spectrum of nanocrystalline carbon doped ZnO. (c) Zn 2p core level spectrum of nanocrystalline carbon doped ZnO. (d) O1s core level spectrum of nanocrystalline carbon doped ZnO.


relaxed to minimize the inter-atomic forces until the total energy change is with a tolerance of 10 7 eV. The energy tolerance of self-consistent cycles is set to 10 8 eV. Out of various possibilities of substitutional doping of carbon in ZnO, i.e. carbon on Zn site (CZn) and carbon on O site (CO), it is found that CO leads to a magnetic moment of 2 mB/C in the supercell. From the site-projected density of states (DOS) shown in Fig. 8, the spin-polarization is observed to be due to the carbon atoms, where the carbon majority and minority spin-states split completely due to exchange-splitting and the Fermi level (here Fermi level is defined as the highest occupied level) passes through the minority spin-states. Since the spin-polarization is that of the p-electrons, it is rather extended in real space (compared, for example, to 3d-states) and from the spin-density (r ¼ rm rk) analysis the spin polarization has been found on nearest neighbor O atoms with CO site as the origin. Slight polarization of Zn atoms within the sphere of influence has also been observed. The studies on formation energies show that CO and CZn have similar magnitude of formation energies and are negative (suggesting that the defects are thermodynamically favorable) only for carbon-rich and zinc-rich conditions. In oxygen-rich conditions, CZn is much more favorable than CO. Since CZn develops no spin polarization in oxygen-rich conditions, such samples are not expected to develop magnetic moments. The situation in carbonpoor conditions does not favor any substitutional doping of carbon in ZnO either. Having spin polarization as stated above does not guarantee stable collective behavior of spins. To study any spin–spin interaction, the system with two carbon atoms substituting the O sites in a large supercell has been modeled. Owing to the anisotropic nature of the wurtzite crystal lattice along the hexagonal plane and in the direction perpendicular to the hexagonal plane (for example between lattice planes (1100) and (0001)), two large supercells of 6 2 2 and 2 2 6 times the primitive wurtzite lattice has been used in our studies. Details of these studies are published elsewhere [48]. The energy difference between ferromagnetic and anti-ferromagnetic orientation of carbon spins varies non-monotonically with carbon atom separation. Although at some carbon–carbon separations we find the energy difference to be slightly above the scale of room temperature, this should not be taken as hint for room temperature ferromagnetism. The main point to note here is that the total energy from

151

ferromagnetic phase must be compared with the paramagnetic phase to determine the critical temperature. Moreover, the DFT calculations do not take temperature in its formalism and as a result the value of critical temperature is always an overestimation. From total energy considerations we find a tendency for formation of carbon-dimers in ZnO, which hints towards phase segregation. Carbon clusters so formed do not have spin polarization. All these reasonings suggest that the experimental data cannot be explained from the model of carbon substituted at the oxygen site in ZnO. Theoretical calculation also shows that Zn-vacancy in ZnO lattice also leads to spin-polarized solution with magnetic moment of 2 mB per Zn-vacancy. This may explain why both pure ZnO nanocrystallite and carbon doped ZnO nanocrystallite show magnetic signal in SQUID measurements. However, a clear and convincing identification of the defects responsible for enhancement of magnetization in carbon doped ZnO samples has so far not been successful. The investigations on role of other defects are currently being pursued.

5. Conclusions An enhancement of ferromagnetism in carbon doped ZnO nanostructures as compared to that of pure ZnO has been observed experimentally through measurements carried out using a SQUID magnetometer. The micro-Raman studies indicate the presence of lattice defects and Zn–C type of complexes in the nanocrystalline carbon doped ZnO matrix. The formations of lattice defects and oxy-carbon clusters are expected to be responsible for ferromagnetism in this system. The theoretical studies show that carbon substituting the oxygen site in ZnO leads to a magnetic moment. From the formation energy calculations we find that carbon substituting the oxygen in ZnO lattice is thermodynamically rather unfavorable for common environmental conditions in experiment. We conclude from our studies that even if carbon substitution at O site (CO) in ZnO leads to spin polarization, the explanation for the observation of ferromagnetism in real samples may be more complicated and intrinsic defects have been found to play a significant role.

Acknowledgment Three of us (DKM, SKS and JM) are grateful to Director, IMMT (CSIR), Bhubaneswar, for providing research support and to DST for providing SQUID facility to IIT-Delhi under Project RP01993, where the magnetic measurements are carried out. One of us, SKN, acknowledges the cooperation of Dr. Andreas Ney, Faculty of Physics, University of Duisburg-Essen, for looking into the experimental results and making useful suggestions. References

Fig. 8. Density of states (DOS) of carbon substituting oxygen (CO) in ZnO. The DOS is plotted for lower scale to show the DOS contribution of carbon more clearly. We observe that spin–split DOS near Fermi energy (zero in the energy scale) gives rise to magnetic moment in the system.

ˇ ´ , J. Fabin, S.D. Sarma, Reviews of Modern Physics 76 (2004) 323. [1] I. Zutic [2] D.D. Awschalom, N. Samarath, D. Loss, Semiconductor Spintronics and Quantum Computation, Springer, Berlin, 2002. [3] H. Ohno, Science 281 (1998) 951. [4] M. Jain, Diluted Magnetic Semiconductors, World Scientific, Singapore, 1991. [5] A.H. MacDonald, P. Schiffer, N. Samarth, Nature Materials 4 (2005) 195. [6] D.D. Awschalom, M.E. Flatte´, Nature Physics 3 (2007) 153. [7] T. Dietl, H. Ohno, F. Matsukura, J. Cibert, D. Ferrand, Science 287 (2000) 1019. [8] S.A. Wolf, D.D. Awschalom, R.A. Buhrman, J.M. Daughton, S. von Molna´r, M.L. Roukes, A.Y. Chtchelkanova, D.M. Treger, Science 294 (2001) 1488. ¨ [9] H. Munekata, H. Ohno, S. von Molnar, A. Segmuller, L.L. Chang, L. Esaki, Physical Review Letters 63 (1989) 1849. [10] B.T. Matthias, R.M. Bozorth, J.H. Van Vleck, Physical Review Letters 7 (1961) 160. [11] H. Ohno, A. Shen, F. Matsukura, A. Oiwa, A. Endo, S. Katsumoto, Y. Iye, Applied Physics Letters 69 (1996) 363.

152


[12] K.Y. Wang, R.P. Campion, K.W. Edmonds, M. Sawicki, T. Dietl, C.T. Foxon, B. Gallagher, AIP Conference Proceedings 772 (2005) 333. [13] R. Janisch, P. Gopal, N.A. Spaldin, Journal of Physics: Condensed Matter 17 (2005) R657. [14] Q. Wang, Q. Sun, P. Jena, Y. Kawazoe, Physical Review B 79 (2009) 115407. [15] K. Ueda, H. Tabata, T. Kawai, Applied Physics Letters 79 (2001) 988. [16] A. Sundaresan, R. Bhargavi, N. Rangarajan, U. Siddesh, C.N.R. Rao, Physical Review B 74 (2006) 161306. [17] D.K. Mishra, P. Kumar, M.K. Sharma, J. Das, S.K. Singh, B.K. Roul, S. Varma, R. Chatterjee, V.V. Srinivasu, D. Kanjilal, Physica B 405 (2010) 2659. [18] Q. Wang, Q. Sun, G. Chen, Y. Kawazoe, P. Jena, Physical Review B 77 (2008) 205411. [19] X. Zhang, Y.H. Cheng, Y. Li, H. Liu, X. Zuo, G.H. Wen, L. Li, R.K. Zheng, S.P. Ringer, Physical Review B 80 (2009) 174427. [20] Y.W. Ma, J. Ding, D.C. Qi, J.B. Yi, H.M. Fan, H. Gong, A.T.S. Wee, A. Rusydi, Applied Physics Letters 95 (2009) 072501. [21] M. Khalid, M. Ziese, A. Setzer, P. Esquinazi, M. Lorenz, H. Hochmuth, M. Grundmann, D. Spemann, T. Butz, G. Brayer, W. Anwand, G. Fisher, W.A. Adeagbo, W. Hergert, A. Ernst, Physical Review B 80 (2009) 035331. [22] H. Pan, J.B. Yi, L. Shen, R.Q. Wu, J.H. Yang, J.Y. Lin, Y.P. Feng, J. Ding, L.H. Van, J.H Yin, Physical Review Letters 99 (2007) 127201. [23] X.–L. Li, J.–F. Guo, Z.–Y. Quan, X.–H. Xu, G.A. Gehring, IEEE Transactions on Magnetics 46 (2010) 1382. ¨ [24] S. Zhou, Q. Xu, K. Potzger, G. Talut, R. Grotzschel, J. Fassbender, M. Vinnichenko, J. Grenzer, M. Helm, H. Hochmuth, M. Lorenz, M Grundann, H. Schmidt, Applied Physics Letters 93 (2008) 232507. [25] X.J. Ye, C.S. Liu, W. Zhong, H.A. Song, C.T. Au, Y.W. Du, Physics Letters A 374 (2010) 496. [26] H. Peng, H.J. Xiang, S.–H. Wei, S.–S. Li, J.–B. Xia, J. Li, Physical Review Letters 102 (2009) 17201. [27] B.J. Nagare, S. Chacko, D.G. Kanhere, Journal of Physical Chemistry A 114 (2010) 2689. [28] Z.G. Wang, X.T. Zu, S.Z. Yang, L.M. Wang, Journal of Materials Science 41 (2006) 3729.

[29] V.A. Nikitenko, Journal of Applied Spectroscopy 57 (1992) 367. [30] A.B. Djurisic, Y.H. Leung, Small 2 (2006) 944. [31] B.K. Meyer, H. Alves, D.M. Hofmann, W. Kriegseis, D. Forster, F. Bertram, J. Christen, M. Strabburg, M. Dworzak, U. Haboeck, A.V. Rodina, Physica. Status. Solidi. B 241 (2004) 231. [32] J. Mu, C. Shao, Z. Guo, Z. Zhang, M. Zhang, P. Zhang, B. Chen, Y. Liu, Year 2011 ACS Applied Materials and Interfaces 3 (2011) 590. [33] S.K. Singh, A.K. Sahoo, Indian Patent 393/DEL/2006. [34] J.B. Yi, L. Shen, H. Pan, L.H. Van, S. Thongmee, J.F. Hu, Y.W. Ma, J. Ding, Y.P. Feng, Journal of Applied Physics 105 (2009) 07C513. [35] K.A. Alim, V.A. Fonoberov, M. Shamsa, A.A. Balandin, Journal of Applied Physics 97 (2005) 124313. ˜ ez, L. Artu´s, J. Jimeńez, B. Wang, [36] R. Cusco´, E. Alarcoń-Llado´, J. Ibań M.J. Callahan, Physical Review B 75 (2007) 165202. [37] S. Jeong, J. Kim, B. Lee, Journal of Physics D: Applied Physics 36 (2003) 2017. [38] Z.Q. Chen, A. Kawasuso, Y. Xu, H. Naramoto, X.L. Yuan, T. Sekiguchi, R. Suzuki, T. Ohdaira, Journal of Applied Physics 97 (2005) 013528. ¨ zgur, ¨ O ¨ Y.I. Alivov, C. Liu, A. Teke, M.A. Reshchikov, S. Do˘gan, V. Avrutin, [39] U. S.J. Cho, H. Morkoc- , Journal of Applied Physics 98 (2005) 041301. [40] F.J. Manjoń, B. Mari, J. Serrano, A.H. Romero, Journal of Applied Physics 97 (2005) 053516. [41] C.D. Wagner, W.M. Riggs, L.E. Davis, in: J.F. Moulder, G.E. Muilenberg (Eds.), Handbook of X-ray Photoelectron Spectroscopy, Parkin-Elmer, 1979. [42] M. Chen, X. Wang, Y.H. Yu, Z.L. Pei, X.D. Bai, C. Sun, R.F. Huang, L.S. Wen, Applied Surface Science 158 (2000) 134. [43] J.C.C. Fan, J.B. Goodenough, Journal of Applied Physics 48 (1977) 3524. ¨ [44] G. Kresse, J. Furthmuller, Computational Materials Science 6 (1996) 15. ¨ [45] G. Kresse, J. Furthmuller, Physical Review B 54 (1996) 11169. [46] J.P. Perdew, in: P. Ziesche, H. Eschrig (Eds.), Electronic Structure of Solids, 91st ed.,Akademie Verlag, Berlin, 1991. [47] S.K. Nayak, M. Ogura, A. Hucht, H. Akai, P. Entel, Journal of Physics: Condensed Matter 21 (2009) 064238. [48] S.K. Nayak, M.E. Gruner, S. Sakong, S. Sil, P. Kratzer, S.N. Behera, P. Entel, Physical Review B 86 (2012) 054441.