Magnetic Properties of Fe- and Mn-Implanted SiC - UF Physics

5 downloads 0 Views 260KB Size Report
p-Type 6H-SiC substrates were implanted with Mn or Fe at doses of 3-5. 1016 cm 2 under conditions that avoided amor- phization substrate temperature. 350°C.
Electrochemical and Solid-State Letters, 4 共12兲 G119-G121 共2001兲

G119

1099-0062/2001/4共12兲/G119/3/$7.00 © The Electrochemical Society, Inc.

Magnetic Properties of Fe- and Mn-Implanted SiC N. Theodoropoulou,a A. F. Hebard,a S. N. G. Chu,b,* M. E. Overberg,c,** C. R. Abernathy,c,* S. J. Pearton,c,*, z R. G. Wilson,d and J. M. Zavadae a

Department of Physics, cDepartment of Materials Science and Engineering, University of Florida, Gainesville, Florida 32611, USA b Bell Laboratories, Lucent Technologies, Murray Hill, New Jersey 07974, USA d Consultant, Stevenson Ranch, California 95131, USA e Army Research Office, Research Triangle Park, North Carolina 27709, USA p-Type 6H-SiC substrates were implanted with Mn⫹ or Fe⫹ at doses of 3-5 ⫻ 1016 cm⫺2 under conditions that avoided amorphization 共substrate temperature ⬃350°C兲. After annealing at 700°C, the magnetic properties of the samples were examined by superconducting quantum interference device magnetometry. Both the Fe- and Mn-implanted samples showed ferromagnetic properties up to approximately 250 K for the highest doses employed. The origin of the ferromagnetism is not the formation of secondary phases involving precipitation of Fe or Mn. © 2001 The Electrochemical Society. 关DOI: 10.1149/1.1414945兴 All rights reserved. Manuscript submitted June 14, 2001; revised manuscript received August 24, 2001. Available electronically October 15, 2001.

There is currently a great deal of interest in the synthesis and characterization of dilute magnetic semiconductors, in which spinpolarized transport could be exploited in a range of new device concepts for ultralow power switches or memories. Most of the work to date has focused on the 共In, Mn兲As and 共Ga, Mn兲As materials systems, in which Mn concentrations up to ⬃5 atom % can be incorporated before the introduction of secondary phases.1-5 When such high Mn concentrations are present and the material is p-type, one observes a carrier-mediated ferromagnetism whose origin is still the subject of intense effort on both the experimental and theoretical fronts.4-9 A major drawback of these materials systems for practical applications is their relatively low Curie temperatures, T C . The highest reported values for T C are ⬃35 K for 共In, Mn兲As and 110 K for 共Ga, Mn兲As.10,11 A recent theoretical prediction12 of much larger T C values for wide bandgap dilute magnetic semiconductors such as 共Ga, Mn兲N and 共Zn, Mn兲O has motivated work in this area. There has been recent progress on the growth of 共Ga, Mn兲N microcrystallites, which are found to exhibit paramagnetic behavior due to their high background n-type conductivity.13,14 In addition, low temperature epitaxial growth of 共Ga, Fe兲N produced ferromagnetic properties below ⬃100 K under optimized conditions.15 Using techniques similar to those reported here, we have found ferromagnetism below 250 K in heavily Mn-implanted GaN when the Mn concentration is in the range 3-5 atom %.16 Implantation is an efficient way of introducing magnetic ions into a wide range of host semiconductors and the use of the technique therefore facilitates the discovery of the most promising materials combinations. In this paper, we report on the magnetic properties of p-SiC implanted with high doses of either Mn or Fe. We find that both 共Si, Mn兲C and 共Si, Fe兲C show apparent ferromagnetic behavior up to approximately 250 K with implant concentrations of ⬃5 atom %. This Curie temperature is below room temperature, but might be increased by having higher hole densities in the SiC if carriermediated processes are the cause of the ferromagnetism. Bulk 6H-SiC wafers 共Al doped兲 with a room temperature hole concentration of ⬃1017 cm⫺3 were implanted 共into the Si-face兲 with either 250 keV Mn⫹ or Fe⫹ at doses of 3-5 ⫻ 1016 cm⫺2. The samples were held at ⬃350°C during the implant step to avoid amorphization. The implant conditions were designed to produce average Mn or Fe concentrations of 3 or 5 atom % over a depth of ⬃2000 Å into the SiC. The samples were subsequently annealed at 700°C under flowing N2 with the implanted side face down on a Si

* Electrochemical Society Fellow. ** Electrochemical Society Student Member. z

E-mail: [email protected]

wafer. The structural properties were examined by transmission electron microscopy 共TEM兲 and selected area diffraction pattern 共SADP兲 analysis. The magnetic properties were measured on a Quantum Design SQUID magnetometer. Figure 1 shows a cross-sectional view of the SiC implanted with 5 atom % Fe after annealing at 700°C. There is relatively light

Figure 1. Cross-sectional TEM micrographs of SiC implanted with 5 atom % Fe and annealed at 700°C.

G120

Electrochemical and Solid-State Letters, 4 共12兲 G119-G121 共2001兲

Figure 2. Magnetization curve at 10 K of SiC implanted with 5 atom % Fe and annealed at 700°C. The coercive field is about 50 G. For comparison, the saturation magnetization for Fe is 220 emu/g.

Figure 4. Temperature dependence of the difference between FC and ZFC magnetization for the 5% Mn sample.

residual damage in the form of dislocation loops to a depth of ⬃0.19 ␮m from the surface, followed by a 260 Å thick heavily defective region at the end-of-range of the Fe⫹ ions. The resultswere similar for the Mn-implanted SiC, as expected due to the nearidentical masses of 56Fe and 55Mn. Figure 2 shows the magnetization curve at 10 K for the SiC implanted with 5 atom % Fe. A strong diamagnetic contribution, which was measured at higher fields, was subtracted. The absence of a true saturation in the magnetization is a fairly common feature of dilute magnetic semiconductors such as 共In, Mn兲As and 共Ga, Mn兲As and its physical origin is not yet clear.10,11 The material implanted with 3 atom % Fe did not show any signature of ferromagnetism. Since the Curie temperature is predicted to be a strong function of both the hole concentration and magnetic ion concentration in wide bandgap dilute magnetic semiconductors,12 the lower dose samples

may be below the threshold for inducing ferromagnetism. A similar effect was observed in Mn-implanted GaN, where doses below 3 atom % did not produce ferromagnetism.16 Figure 3 shows the temperature dependence of the difference between the field-cooled 共FC兲 and zero field-cooled 共ZFC兲 magnetization for the 5 atom % Fe sample. The inset of the figure shows the raw data. The subtraction of ZFC from FC data advantageously eliminates para- and diamagnetic contributions and simultaneously indicates the presence of hysteresis if the difference is nonzero. Although ferromagnetism is the usual explanation for hysteresis, spinglass effects or superparamagnetism can also be the cause. All of these effects however are magnetic phenomena involving the ordering of spins, and it is in this sense that we refer to the hysteresis measured by the FC-ZFC data in Fig. 3 as ‘‘ferromagnetic.’’ It is clear that a remnant of a ferromagnetic contribution is present even at room temperature. The origin of ferromagnetic behavior in implanted SiC is still not clear; we did not observe secondary phases involving precipitation of Fe or formation of FeCx or FeSix compounds. Both plan-view and cross-sectional samples were examined,

Figure 3. Temperature dependence of the difference between field-cooled and zero field-cooled magnetization for the 5 atom % Fe sample. The inset shows the raw data.

Figure 5. Magnetization curve at 10 K of SiC implanted with 5 atom % Mn and annealed at 700°C.

Electrochemical and Solid-State Letters, 4 共12兲 G119-G121 共2001兲 and no evidence of second phase formation was found. While it might be expected that the solid solubility of Fe would be relatively low in SiC (⬍1017 cm⫺3), ion implantation is a nonequilibrium process and as long as the postimplant anneal temperature is kept low, then much higher concentrations of Fe can be incorporated. Moreover, the implanted region was relatively resistive; the capacitance-voltage measurements showed depletion beyond the Fe range after the implantation and the annealing process, and the hole concentration was ⬍1015 cm⫺3 in that region.17,18 Whether the remaining hole density is sufficient to induce carrier-mediated ferromagnetism in the SiC needs additional work to answer, involving substrates of different conductivity level and type. There is theory proposed in the literature that random doping effects in wide bandgap materials could be sufficient to explain the magnetism observed.19,20 We also observed ferromagnetic contributions to the magnetization of SiC implanted with 5% Mn samples, albeit at a lower level compared to Fe as a magnetic impurity. The difference between the FC and the ZFC magnetization data is shown in Fig. 4 for the sample implanted with 5 atom % Mn and annealed at 700°C. In Fig. 5, the magnetization curve of SiC共Mn兲 shows that, at 10 K, there is hysteresis in the magnetization. A coercive field of 150 G was observed. This shows the usefulness of the ion implantation technique in being able to introduce a range of different impurities into the host semiconductor and being able to screen those that are the most promising for magnetic applications. The implant process is also attractive for its ability to create selective area magnetic regions that might be employed as spin-injection contacts in device structures. In summary, high doses of Fe⫹ or Mn⫹ have been implanted into p-SiC. Both the Mn- and Fe-implanted material show a ferromagnetic contribution below ⬃250 K for Fe and Mn concentrations of ⬃5 atom %. Future work needs to focus on trying to increase the Curie temperature in the SiC doped with Fe or Mn and to understand the microscopic origin of the ferromagnetism.

G121

Acknowledgments The work at the University of Florida is partially supported by NSF DMR-0101438 共S.J.P.兲 and DMR-9705224 共A.F.H.兲. The work of R.G.W. is partially supported by ARO. The University of Florida assisted in meeting the publication costs of this article.

References 1. G. A. Prinz, Science, 282, 1660 共1998兲. 2. H. Ohno, H. Munekata, T. Penney, S. von Molnar, and L. L. Chang, Phys. Rev. Lett., 68, 2664 共1992兲. 3. D. D. Awschalom and R. K. Kawakami, Nature, 408, 923 共2000兲. 4. H. Ohno, D. Chibu, F. Matsukura, T. Omiya, E. Abe, T. Dietl, Y. Ohno, and K. Ohtani, Nature, 408, 944 共2000兲. 5. D. D. Awschalom and J. M. Kikkawa, Phys. Today, 52, 33 共1999兲. 6. B. T. Jonker, Y. D. Park, B. R. Bennet, H. D. Cheong, G. Kioseoglou, and A. Petrou, Phys. Rev. B, 62, 8180 共2000兲. 7. Y. D. Park, B. T. Jonker, B. R. Bennet, G. Itskos, M. Furis, G. Kioseoglou, and A. Petrou, Appl. Phys. Lett., 77, 3989 共2000兲. 8. T. Dietl, A. Haurry, and Y. M. d’Aubigne, Phys. Rev. B, 55, R3347 共1997兲. 9. J. Konig, H. H. Lin, and A. H. MacDonald, Phys. Rev. Lett., 84, 5628 共2000兲. 10. J. de Boeck and G. Borghs, Phys. World, 12, 27 共1999兲. 11. H. Ohno, Science, 281, 951 共1998兲. 12. T. Dietl, H. Ohno, F. Matsakura, J. Cibert, and D. Ferrand, Science, 289, 1019 共2000兲. 13. W. Gebicki, J. Strzeszewski, G. Kamler, T. Szczyko, and S. Podiado, Appl. Phys. Lett., 76, 3870 共2000兲. 14. M. Zajac, R. Doradzinski, J. Gosk, J. Szczyko, M. Lefeld-Sosnowska, M. Kaminska, A. Twardowski, M. Palczewska, E. Grzanka, and W. Gebicki, Appl. Phys. Lett., 78, 1276 共2001兲. 15. H. Akinaga, S. Nemeth, J. de Boeck, L. Nistor, H. Bender, G. Borghs, H. Otuchi, and M. Oshima, Appl. Phys. Lett., 77, 4377 共2000兲. 16. N. Theodoropoulou, A. F. Hebard, M. E. Overberg, C. R. Abernathy, S. J. Pearton, S. N. G. Chu, and R. G. Wilson, Appl. Phys. Lett., 78, 3475 共2001兲. 17. M. A. Capano, R. Santhakumar, R. Venugopal, M. R. Melloch, and J. A. Cooper, Jr., J. Electron. Mater., 29, 210 共2000兲. 18. M. A. Capano, J. A. Cooper, Jr., M. R. Melloch, A. Saxler, and W. C. Mitchel, J. Appl. Phys., 87, 8773 共2000兲. 19. R. N. Bhatt and M. Berciu, Phys. Rev. Lett., In press. 20. R. N. Bhatt and X. Wan, Int. J. Mod. Phys. C, 10, 1459 共1999兲.