Room Temperature Ferromagnetism in GaMnN and GaMnPNew Materials for Semiconductor Spintronics

 

S.J.J. Pearton(1),M.E.Overberg(!),G.T.Thaler(1),C.R.Abernathy(1),J.Kim(2),F.Ren(2),N.Theodoropoulou(3),A.F.Hebard(3) and Yun Daniel Park((4)2)

 

(1)   (1) Department of Materials Science and Engineering, University of Florida, Gainesville, FL 32611 USA

(2)   Department of Chemical Engineering,University of Florida,Gainesville,FL 32611,USA

(3)   Department of Physics,University of Florida,Gainesville,FL 32611,USA

 

(4)  Center for Strongly Correlated Materials Research, Seoul National University, Seoul 151-747 Korea

 

 

ABSTRACT

Results on the synthesis of ferromagnetic GaMnN and GaMnP by both Molecular Beam Epitaxy or implantation of Mn directly into p-GaN or GaP(C) at elevated temperatures to avoid amorphization will be described.There is a relatively broad range of growth conditions under which single-crystal,single-phase material may be obtained with significant Mn concentrations.The effects of background doping level are discussed ,along with a comparison of results on direct implantation of Fe and Ni instead of Mn.Essential requirements for utilizing the spin of the electron in device structures include the ability to achieve efficient electrical spin injection and transport of spin-polarized carriers,along with effective detection of these carriers.


INTRODUCTION

            The field of semiconductor spin transfer electronics (spintronics) seeks to exploit the spin of charge carriers in semiconductors. It is widely expected that new functionalities for electronics and photonics can be achieveded if the injection, transfer and detection of carrier spin can be controlled above room temperature.Since the magnetic properties of ferromagnetic semiconductors are a function of carrier concentration in the material in many cases, then it will be possible to have electrically or optically-controlled magnetism through field-gating of transistor structures or optical excitation to alter the carrier density.  This novel control of magnetism has already been achieved electronically and optically in an InMnAs metal-insulator semiconductor structure at low temperatures [1,2].  A number of recent reviews have covered the topics of spin injection, coherence length and magnetic properties of materials systems such as (Ga,Mn)As [3-5], (In,Mn)As [3,6] and the general areas of spin injection from metals into semiconductors and applications of the spintronic phenomena [6,7].  The current interest in magnetic semiconductors can be traced to difficulties in injecting spins from a ferromagnetic metal into a semiconductor [8-10].

         There are two major criteria for selecting the most promising materials for semiconductor spintronics.  First, the ferromagnetism should be retained to practical temperatures (i.e. >300 K).  Second, it would be a major advantage if there were already an existing technology base for the material in other applications.  Most of the work in the past has focused on (Ga,Mn)As and (In,Mn)As. However, the highest Curie temperatures reported for these materials (~110 K for (Ga,Mn)As [1] and ~35 for (In,Mn)As [6]) are too low for most practical applications. 

            The key breakthrough that focused attention on wide bandgap semiconductors as being the most promising for achieving practical ordering temperatures was the theoretical work of Dietl et al. [11].  They predicted that cubic GaN doped with ~5at.% of Mn and containing a high concentration of holes (3.5x1020 cm-3) should exhibit a Curie temperature exceeding room temperature.A schematic of the predicted ordering temperatures is shown in Figure 1.

   Two basic approaches to understanding the magnetic properties of dilute magnetic semiconductors have emerged.  The first class of approaches is based on mean-field theory. The theories that fall into this general model implicitly assume that the dilute magnetic semiconductor is a more-or-less random alloy, e.g. (Ga,Mn)N, in which Mn substitutes for one of the lattice constituents.  The second class of approaches suggests that the magnetic atoms form small (a few atoms) clusters that produce the observed ferromagnetism [12].  A difficulty in experimentally verifying the mechanism responsible for the observed magnetic properties is that depending on the growth conditions employed for growing the DMS material, it is likely that one could readily produce samples that span the entire spectrum of possibilities from single-phase random alloys to nanoclusters of the magnetic atoms to precipitates and second phase formation.  Therefore, it is necessary to decide on a case-by-case basis which mechanism is applicable.

            The mean field approach basically assumes that the ferromagnetism occurs through interactions between the local moments of the Mn atoms, which are mediated by free holes in the material.  The spin-spin coupling is also assumed to be a long-range interaction, allowing use of a mean-field approximation [13-18].  In its basic form, this model employs a virtual-crystal approximation to calculate the effective spin-density due to the Mn ion distribution.  The direct Mn-Mn interactions are antiferromagnetic so that the Curie temperature, TC, for a given material with a specific Mn concentration and hole density (derived from Mn acceptors and/or intentional shallow level acceptor doping), is determined by a competition between the ferromagnetic and anti-ferromagnetic interactions. 

.  Numerous refinements of this approach have appeared recently, taking into account the effects of positional disorder [15,16], indirect exchange interactions [17], spatial inhomogeneities and free-carrier spin polarization [15,18].

         Initial reports of the energy level of Mn in GaN show it is very deep in the gap, Ev + 1.4 eV [19], and thus would be an ineffective dopant under most conditions.  Some strategies for enhancing the hole concentration do exist, such as co-doping both acceptors and donors to reduce self-compensation effects [20] or the use of selectively-doped AlGaN/GaN superlattices in which there is transfer of free holes from Mg acceptors in the AlGaN barriers to the GaN quantum wells

            A further issue that needs additional exploration in the theories is the role of electrons, rather than holes, in stabilizing the ferromagnetism in DMS materials.  All of the reports of ferromagnetism in (Ga,Mn)N, for example, occur for material that is actually n-type.  Since the material has to be grown at relatively low temperatures to avoid Mn precipitation and therefore only Molecular Beam Epitaxy (MBE) can be used, there is always the possibility of unintentional n-type doping from nitrogen vacancies, residual lattice defects or impurities such as oxygen.  Therefore stoichiometry effects, crystal defects or unintentional impurities may control the final conductivity, rather than Mn or the intentionally-introduced acceptor dopants. 

            While most of the theoretical work for DMS materials has focused on the use of Mn as the magnetic dopant, there has been some progress on identifying other transition metal atoms that may be effective.  Figure 2 shows the predicted stability of ferromagnetic states in GaN doped with different 3d transition metal atoms [21].  The results are based on a local spin-density approximation which assumed that Ga atoms were randomly substituted with the magnetic atoms and did not take into account any additional carrier doping effects.  In this study it was found that (Ga,V)N and (Ga,Cr)N showed stable ferromagnetism for all transition metal concentrations whereas Fe, Co or Ni doping produced spin-glass ground states [42].  For the case of Mn, the ferromagnetic state was the lowest energy state for concentrations up to ~20%, whereas the spin-glass state became the most stable at higher Mn concentrations.

Magnetic Properties of (Ga,Mn)N

            The first reports of the magnetic properties of (Ga,Mn)N involved bulk microcrystallites grown at high temperatures (~1200°C), but while percent levels of Mn were incorporated, the samples exhibited paramagnetic behavior [22].  By sharp contrast, in epitaxial GaN layers grown on sapphire substrates and then subjected to solid state diffusion of Mn at temperatures from 250-800°C for various periods, clear signatures of room temperature ferromagnetism were observed [23,24]. The Curie temperature was found to be in the range 220-370K, depending on the diffusion conditions.  The use of ion implantation to introduce the Mn produced lower magnetic ordering temperatures [25]. A key question is whether the resulting material is indeed an alloy of (Ga,Mn)N or whether it remains as GaN with clusters, precipitates or second phases that are responsible for the observed magnetic properties.

            In (Ga,Mn)N films grown by MBE at temperatures between 580-720°C with Mn contents of 6-9 at.%, magnetization (M) versus magnetic field (H) curves showed clear hysteresis at 300K, with coercivities of 52-85 Oe and residual magnetizations of 0.08-0.77 emu/g at this temperature [26].  Figure 3 shows the temperature dependence of the magnetization for a sample with 9 at.% Mn, yielding an estimated TC of 940K using a mean field approximation.  Note that while the electrical properties of the samples were not measured, they were almost certainly n-type..  Room temperature ferromagnetism in n-type (Ga,Mn)N grown by MBE has also been reported by Thaler et al. [27]. Figure 4 shows M-T and M-H data as a function of Mn content in the GaMnN. In that case, strenuous efforts were made to exclude any possible contribution from the sample holder in the superconducting quantum interference device (SQUID) magnetometer or other spurious effects.  It is also worthwhile to point out that for the studies of (Ga,Mn)N showing ferromagnetic ordering by magnetization measurements, a number of materials characterization techniques did not show the presence of any second ferromagnetic phases within detectable limits.  In addition, the values of the measured coercivities are relatively small.

            Other reports have also recently appeared on the magnetic properties of GaN doped with other transition metal impurities.  For initially p-type samples directly implanted with either Fe or Ni, ferromagnetism was observed at temperatures of ~200K [28] and 50K [29], respectively.Figure 5 (top) shows the M-T data from Fe-implanted p-GaN,while the selected area diffraction pattern from the single-phase material is shown at bottom.Similarly,(Ga,Fe)N films grown by MBE showed Curie temperatures of £100K, with EXAFS data showing that the majority of the Fe was substitutional on Ga sites [30].  (Ga,Cr)N layers grown in a similar fashion at 700°C on sapphire substrates showed single-phase behavior, clear hysteresis and saturation of magnetization at 300K and a Curie temperature exceeding 400K [31].

Magnetic Properties of (Ga,Mn)P

            While mean-field theories predict relatively low Curie temperatures (<110K) for (Ga,Mn)P , recent experiments show ferromagnetism above 300K [32,33]. Figure 6 shows a hysteresis loop at 300K from an MBE-grown GaMnP sample. In other respects, the magnetic behavior of the (Ga,Mn)P was consistent with mean-field predictions.  For example, the magnetization versus temperature plots showed a more classical concave shape  than observed with many DMS materials.  In addition, the Curie temperature was strongly influenced by the carrier density and type in the material, with highly p-type samples showing much higher values than n-type or undoped samples.  Finally, the Curie temperature increased with Mn concentration up to ~6 at.% and decreased at higher concentrations.No secondary phases or clusters could be detected by transmission electron microscopy, x-ray diffraction or selected area diffraction patterns.  Similar results were achieved in samples in which the Mn was incorporated during MBE growth or directly implanted with Mn.

            GaP is a particularly attractive host material for spintronic applications because it is almost lattice-matched to Si.  One can therefore envision integration of (Ga,Mn)P spintronic magnetic sensors or data storage elements to form fast non-volatile Magnetic Random Access Memories (MRAM).  Although it has an indirect bandgap, it can be made to luminescence through addition of isoelectric dopants such as nitrogen or else one could employ the direct bandgap ternary InGaP, which is lattice matched to GaAs.  The quaternary InGaAlP materials system is used for visible light-emitting diodes, laser diodes, heterojunction bipolar transistors and high electron mobility transistors.  An immediate application of the DMS counterparts to the component binary and ternary materials in this system would be to add spin functionality to all of these devices.  A further advantage to the wide bandgap phosphides is that they exhibit room temperature ferromagnetism even for relatively high growth temperatures during MBE. 

Acknowledgments     

            The work at UF was partially supported by NSF-DMR 0101438, while the work at SNU was partially supported by KOSEF and Samsung Electronics Endowment through CSCMR and by the Seoul National University Research Foundation.


References

1.      Ohno, H. Chiba, D., Matsukura, F., Omiya, T., Abe, E., Dietl, T., Ohno, Y. and Ohtani, K., Nature 408, 944 (2000).

2.      Oiwa, A., Mitsumori, Y., Moriya, R., Slupinski, T. and Munekata. H., Phys. Rev. Lett. 88, 137202 (2002).

3.      Ohno, H., J. Vac. Sci. Technol. B18, 2039 (2000).

4.      Wolf, S.A., Awschalom, D.D., Buhrman, R.A., Daughton, J.M., von Molnar, S., Roukes, M.L., Chtchelkanova, A.Y. and Treger, D.M., Science 294, 1488 (2001).

5.      Das Sarma, S., American Scientist 89, 516 (2001).

6.      Ohno, H., Matsukura, F. and Ohno, Y.,JSAP International 5, 4 (2002).

7.      Awschalom, D.D. and Kikkawa, J.M., Science 287, 473 (2000).

8.      Hammar, P.R., Bennett, B.R., Yang and M.J., Johnson, Mark, Phys. Rev. Lett. 83, 203 (1999).

9.      Schmidt, G., Ferrand, D., Molenkamp, L.W., Filip, A.T. and van Wees, B.J., Phys. Rev. B 62, R4793 (2000).

10.  Rashba, E. I.,Phys. Rev. B 62, R16267 (2000).

11.  Dietl, T., Ohno, H., Matsukura, F., Cibert, J. and Ferrand, D., Science 287, 1019 (2000).

12.  Van Schilfgaarde, M. and Myrasov, O.N., Phys. Rev. B63, 233205 (2001).

13.  Dietl, T., Ohno, H. and Matsukura, F., Phys. Rev. B63, 195205 (2001).

14.  Jungwirth, T., Atkinson, W.A., Lee, B.H. and MacDonald, A.H.,Phys. Rev. B59, 9818 (1999).

15.  Berciu, M. and Bhatt, R.N., Phys. Rev. Lett. 87, 108203 (2001).

16.  Bhatt, R.N., Berciu, M., Kennett, M.D. and Wan, X, J. Superconductivity: Incorporating Novel Magnetism 15, 71 (2002).

17.  Litvinov, V.I. and Dugnev, V.K., Phys. Rev. Lett. 86, 5593 (2001).

18.  Konig, J., Lin, H.H. and MacDonald, A.H.,Phys. Rev. Lett. 84, 5628 (2001).

19.  Korortiev, R.Y., Gregie, J.M. and Wessels, B.W.,Phys. Lett. 80, 1731 (2002).

20.  Katayama-Yoshida, H., Kato, R. and Yamamoto, T.,  J. Cryst. Growth 231, 438 (2001).

21.  Sato, K. and Katayama-Yoshida, H., Jap. J. Appl. Phys. 40, L485 (2001).

22.  Zajac, M., Gosk, J., Kaminska, M. Twardowski, A., Szyszko, T. and Podliasko, S., Appl. Phys. Lett. 79, 2432 (2001).

23.  Reed, M.L., Ritums, M.K. Stadelmaier, H.H., Reed, M.J., Parker, C.A., Bedair, S.M. and El-Masry, N.A., Mater. Lett. 51, 500 (2001).

24.  Reed, M.L., El-Masry, N.A., Stadelmaier, H.H., Ritums, M.E., Reed, N.J., Parker, C.A., Roberts, J.C. and Bedair, S.M. ,Appl. Phys. Lett. 79, 3473 (2001).

25.  Theodoropoulou, N.A., Hebard, A.F., Overberg, M.E., Abernathy, C.R., Pearton, S.J., Chu, S.N.G. and Wilson, R.G.,Appl. Phys. Lett. 78, 3475 (2001).

26.  Sonoda, S., Shimizu, S., Sasaki, T., Yamamoto, Y. and Hori, H., J. Cryst. Growth (in press).

27.  Thaler, G.T., Overberg, M.E., Gila, B., Frazier, R., Abernathy, C.R., Pearton, S.J., Lee, J.S., Lee, S.Y., Park, Y.D., Khim, Z.G., Kim, J. and Ren, F.,Appl. Phys. Lett. (in press).

28.  Theodoropoulou, N.A., Hebard, A.F., Chu, S.N.G., Overberg, M.E., Abernathy, C.R., Pearton, S.J., Wilson, R.G. and Zavada, J.M.,Appl. Phys. Lett. 79, 3452 (2001).

29.  Pearton, S.J., Overberg, M.E., Thaler, G., Abernathy, C.R., Theodoropoulou, N., Hebard, A.F., Chu, S.N.G., Wilson, R.G., Zavada, J.M., Polyakov, A.Y., Osinsky, A. and Park, Y.D.,  J. Vac. Sci. Technol. A20, 583 (2002).

30.  Akinaga, H., Nemeth, S., De Boeck, J., Nistor, L., Bender, H., Borghs, G., Ofuchi, H. and Oshima, M.,Appl. Phys. Lett. 77, 4377 (2000).

31.  Hashimoto, M., Zhou, Y.-Z., Kanamura, M. and Asahi, H.,  Solid-State Commun. (in press).

32.  Theodoropoulou, N., Hebard, A.F., Overberg, M.E., Abernathy, C.R., Pearton, S.J., Chu, S.N.G. and Wilson, R.G.,Phys. Rev. Lett. (in press).

33.  Overberg, M.E., Gila, B.P., Thaler, G.T., Abernathy, C.R., Pearton, S.J., Theodoropoulou, N., McCarthy, K.T., Arnason, S.B., Hebard, A.F., Chu, S.N.G., Wilson, R.G., Zavada, J.M. and Park, Y.D., J. Vac. Sci. Technol. B (May/June 2002).

 

 

Figure Captions

Figure 1.         Predicted Curie temperatures as a function of bandgap 

 

Figure 2.         Predicted stability of the ferromagnetic states of different transition metal (TM) atoms in GaN as a function of transition metal concentration.  The vertical axis represents the energy difference between the ferromagnetic and spin glass states for each metal atom (after ref. 21).

Figure 3.       Magnetization versus temperature for (Ga,Mn)N sample grown by MBE with ~9 at.% Mn.  The extrapolation of the curve is based on a mean-field approximation (after ref.26 ).

 

Figure 4.  (a) B-H at 300 K from MBE-grown (Ga,Mn)N with ~9 at.% Mn (closed circles).  The data from the sapphire substrate is shown as open circles.  (b) Magnetization as a function of temperature for the MBE-grown (Ga,Mn)N with ~9 at.% Mn.  (c)  B-H at 300 K from MBE-grown (Ga,Mn)N with various Mn concentrations.

Figure 5.         Magnetization versus temperature for the difference in field-cooled and zero field-cooled signals.  The sample was p-GaN implanted with 3x1016 cm-2 Fe+ and annealed at 700°C (top).  SADP from same sample is also shown (bottom).

Figure 6.         Magnetization loops for MBE-grown (Ga,Mn)P at 300 K