Characterization of High Dose Fe Implantation into p-GaN

 

 

N. Theodoropoulou1, A.F. Hebard1, S.N.G. Chu2, M.E. Overberg3, C.R. Abernathy3, S.J. Pearton3, R.G. Wilson4 and J.M. Zavada5

 

 

1 Department of Physics

University of Florida, Gainesville, FL 32611

 

2 Bell Laboratories, Lucent Technologies

Murray Hill, NJ 07974

 

3 Department of Materials Science and Engineering

University of Florida, Gainesville, FL 32611

 

4 Consultant, Stevenson Ranch, CA 95131

 

5 Army Research Office, Research Triangle Park, NC 27709

 

ABSTRACT

High concentrations (3-5at.%) of Fe were incorporated into p-GaN by direct implantation at elevated substrate temperature (350°C).  Subsequent annealing at 700°C produced ferromagnetic behavior below ~100 K for the 3at.% sample and ~50 K for the 5at.% sample.  Selected area diffraction patterns did not reveal the presence of any other phases in the Fe-implanted region.  The direct implantation process appears promising for examining the properties of magnetic semiconductors with application to magnetotransport and magneto-optical devices.

            The interest in the carrier-induced ferromagnetism in dilute magnetic compound semiconductors such as (In, Mn)As and (Ga,Mn)As has increased recently because of their potential application to new classes of devices based on spin-polarized transport or integration of magnetic, optical and electronic functions on a single chip.(1-9)  The ability to control the magnetic properties through application of an electric field (i.e. field gating to manipulate the carrier density) to the material has recently been demonstrated by Ohno et al.(10)  The Curie temperatures, TC, of (In,Mn)As(11) and (Ga,Mn)As(4,12) are relatively low (~35 and 110 K, respectively) and from practical considerations it is desirable to find materials with higher values.  Recent calculations based on a Zener model of paramagnetism predict the possibility that wide bandgap systems such as (Ga,Mn)N and (Zn,Mn)O might have TC values above room temperature.(13)  Mean-field theory models suggest that the TC’s in Mn-doped compound semiconductors would be proportional to the valence band density-of-states in the host(14)

At present, little is known about the properties of GaN heavily doped with impurities that might induce ferromagnetic behavior.  Some initial reports have appeared on microcrystalline Ga1-xMnxN with Mn contents up to x = 0.005 which exhibited ferromagnetic behavior.(15,16)  Akinaga et al.(17) reported ferromagnetic properties at <100 K in heavily Fe-doped GaN grown by low temperature (380°C) Molecular Beam Epitaxy.  We have found apparent ferromagnetic behavior in Mn-implanted p-GaN at temperatures up to ~250 K for implanted Mn concentrations of 3-5at.%.(18)  The implantation process is a simple approach for introducing magnetic ions into different host materials and could readily be used for making selected-area contact regions for injection of spin-polarized current into device structures.

In this letter we report on the properties of p-GaN implanted with Fe at doses designed to produce concentrations of 3-5at.% at the peak of the implanted profile.  Under these conditions, samples annealed at 700°C do not show any evidence of secondary phase formation (at least to the sensitivity of transmission electron microscopy and selected area diffraction pattern analysis).  Consistent with the past results on epitaxial Fe-doped GaN, ferromagnetic behavior was observed in the implanted samples.

The p-GaN samples were grown by Metal Organic Chemical Vapor Deposition on Al2O3 substrates using triethylgallium and ammonia as precursors and Cp2Mg as the Mg dopant source.  The acceptor concentration measured by capacitance-voltage was ~2x1019 cm-3, with a hole concentration of ~3x1017 cm-3 at 25°C due to the deep ionization level of the Mg (~170 meV).  The total epilayer thickness was 4mm.  Fe+ ions were implanted at an energy of 250 keV at dose of ~3-5x1016 cm-2 to produce average volume concentrations of ~3-5at.% in the top ~2000Å of the GaN.  Amorphization of the implanted region was avoided by holding the samples at ~350°C during the implantation.(19,20)  Annealing was performed at 700°C for 5 mins under flowing N2 gas with the samples face-down on another GaN wafer.  The crystalline quality of the implanted regions were examined by transmission electron microscopy (TEM) and selected area diffraction pattern (SADP) analysis, while the magnetic properties were measured in a Quantum Design MPMS SQUID magnetometer.

Figure 1 (top) shows some cross-sectional TEM micrographs of the 3at.% Fe-implanted GaN at different magnifications.  There is residual disorder in the form of dislocation loops extending throughout the implanted region, with an increase intensity toward the end-or-range of the Fe.  The total damage depth is ~0.22mm, which correlates fairly well with the Fe incorporation depth obtained from Transport-of-ions-in-Matter (TRIM) simulations.  The bottom of Figure 1 shows the SADP from the implanted region.  There are no obvious extra sports from secondary phase formation and only the diffraction from the GaN hexagonal crystal structure is observed.  This is similar to the case of SiC single crystals implanted with Fe under the same conditions, where we also did not observe secondary phases after annealing.(21)  Our results are consistent with the previous MBE-grown GaN (Fe) results in which ~1at.% of Fe was incorporated substitutionally on the Ga site and no other phases such as FexN precipitates were observed.(17)

Figure 2 shows the magnetization curve at 10 K for the 3at.% Fe-implanted GaN annealed at 700°C, with the inset showing an expanded view of the lower field data.  From the difference in magnetization for field-cooled versus zero field-cooled samples, we could observe a ferromagnetic contribution present in the films below ~100 K.  Our previous results on Mn-implanted GaN found a ferromagnetic contribution present below ~250 K but in that case platelet-type regions were observed which were suggested to be GaxMn1-xN.  The origin of the magnetic behavior in the Fe-implanted GaN is less clear, since the effective hole concentration is certainly less than the 2x1017 cm-3 present in the starting sample due to residual implant damage.   Theory suggests much higher hole densities (³1020 cm-3) are necessary for carrier-mediated ferromagnetism.(13)

Figure 3 shows the hysteresis loop at 10 K from the 5at.% Fe-implanted GaN.  The coercivity at this temperature was ~ 150 G.  The Fe-implantation clearly induces a major change in the magnetic properties of the GaN, because the unimplanted material is paramagnetic.  Figure 4 (inset) shows the magnetization as a function of temperature for both field-cooled and zero field-cooled conditions, whereas the main part of the Figure shows the difference in magnetization between the two conditions.  The ferromagnetic contribution is present at < 50 K in these samples.  The decrease in apparent Curie temperature relative to the 3at.% Fe-implanted GaN may be due to a lower hole concentration as a result of more residual implant-induced donor levels.  Past work on the electrical effects of implant damage in GaN have established its donor nature.(22)

Since the origin of the ferromagnetism in Fe-implanted GaN is still not clear, it is difficult to compare the TC values observed experimentally with those predicted by theory.(13)  In particular, the hole concentrations in implanted material with residual damage are many orders of magnitude lower than assumed in the theoretical predictions.  Hall measurements on our samples showed the hole concentration to be 1016 £cm-3.  The other question is how high levels of Fe incorporation would affect the hole concentration, even in epitaxially grown material.  If the Fe has a deep level energy state in the gap with a significant capture cross-section for holes, then at room temperature most of the holes in the GaN would be trapped at these centers.  Since the TC is expected to be a strong function of both the Fe ion concentration and the hole density, it would be necessary to have both as high as possible.  At the very least, ion implantation of different impurities into GaN is an efficient method for examining their effectiveness in creating ferromagnetism and can be used as a way of  screening the various impurities and finding the most likely ones to succeed during epitaxial growth.

In summary, high dose implantation of Fe into p-GaN produced ferromagnetism below ~175 K.  This behavior is consistent with previous results on epitaxial GaN(Fe).  Future work should focus on the improvement in Curie temperature by increasing both Fe solubility and the hole concentration in the GaN.

 

ACKNOWLEDGMENTS

            The work at UF is partially supported by NSF (DMR0101438, Vern Hess), while that of RGW is partially supported by ARO.

REFERENCES

1.                  G.A. Prinz, Science 282 1660 (1998).

2.                  J. de Boeck and G. Borghs, Phys. World 12 27 (1999).

3.                  H. Ohno, Science 281 951 (1998).

4.                  T. Diehl, A. Haury and Y. Merle d’Aubigne, Phys. Rev. B. 55 R3347 (1997).

5.                  B. Lee, T. Jungwirth and A.H. MacDonald, Phys. Rev. B 61 15606 (2000).

6.                  D.D. Awschalom and R.K. Kawakami, Nature 408 923 (2000)).

7.                  D.D. Awschalom and J.M. Kikkawa, Physics Today 52 33 (1999).

8.                  B.T. Jonker, Y.D. Park, B.R. Bennet, H.D. Cheong, G. Kioseoglou and A. Petrou, Phys. Rev. B 62 8120 (2000).

9.                  Y.D. Park, B.T. Jonker, B.R. Bennet, G. Itskos, M. Furis, G. Kioseoglou and A. Petrou, Appl. Phys. Lett. 77 3989 (2000).

10.              H. Ohno, D. Chibu, F. Matsukura, T. Damiya, E. Abe, T. Dietl, Y. Ohno and K. Ohtani, Nature 408 944 (2000).

11.              H. Ohno, H. Minehata, T. Penney, S. von Molnar and L.L. Chang, Phys. Rev. Lett. 68 2664 (1992).

12.              T. Hayashi, M. Tanaka, K. Seto, T. Nishinaga and K. Ando, Appl. Phys. Lett. 71 1825 (1997).

13.              T. Dietl, H. Ohno, F. Matsukura, J. Cibert and D. Ferrand, Science 287 1619 (2000).

14.              J. Schliemann, J. Konig, H.-H. Lin and A.H. MacDonald, Appl. Phys. Lett. 78 1550 (2001).

15.              W. Gebicki, J. Strzeszewski, G. Kamler, T. Szyszko and S. Podsliadlo, Appl. Phys. Lett. 76 3870 (2000).

16.              M. Zajac, R. Doradzinski, J. Gosk, J. Szczyko, M. Lefeld-Sosnowska, M. Kaminska, A. Twardowski, M. Palczewska, E. Grzanka and W. Gebicki, Phys. Lett. 78 1276 (2001).

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

18.              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 (2001).

19.              S.O. Kucheyev, J.S. Williams, C. Jagadish, J. Zou and G. Li, Phys. Rev. B62 7510 (2000).

20.              J.S. Williams, Mat. Sci. Eng. A253 8 (1998).

21.              N. Theodoropoulou, A.F. Hebard, S.N.G. Chu, M.E. Overberg, C.R. Abernathy, S.J. Pearton and R.G. Wilson (unpublished).

22.              S.J. Pearton, J.C. Zolper, R.J. Shul and F. Ren, J. Appl. Phys. (Applied Physics Reviews) 86 1 (1999).

FIGURE CAPTIONS

 

Figure 1.         Cross-section TEM micrographs from GaN implanted with 3x1016 cm-2 Fe+, after annealing at 700°C (top and center) and selected area diffraction pattern from the implanted region (bottom).

 

Figure 2.         Magnetization curve at 10 K of GaN implanted with 3x1016 cm-2 Fe+ and annealed at 700°C.  The inset shows the low field data.

 

Figure 3.         Hysteresis loop at 10 K from GaN implanted with 5x1016 cm-2 Fe+ and annealed at 700°C.

 

Figure 4.         Temperature dependence of the difference in magnetization between field-cooled (B = 0.1T) and zero field-cooled conditions.  The inset shows the raw data.