Ferromagnetic GeMn thin film prepared by ion implantation and ion beam induced
epitaxial crystallization annealing
C. H. Chen, H. Niu, D. C. Yan, H. H. Hsieh, C. P. Lee, and C. C. Chi
Citation: Applied Physics Letters 100, 242412 (2012); doi: 10.1063/1.4729752 View online: http://dx.doi.org/10.1063/1.4729752
View Table of Contents: http://scitation.aip.org/content/aip/journal/apl/100/24?ver=pdfcov Published by the AIP Publishing
Articles you may be interested in
Enhanced magnetic and electrical properties in amorphous Ge:Mn thin films by non-magnetic codoping J. Appl. Phys. 111, 033916 (2012); 10.1063/1.3679076
Effect of annealing on the magnetic properties of Gd focused ion beam implanted GaN Appl. Phys. Lett. 91, 072514 (2007); 10.1063/1.2770762
Ferromagnetism and colossal magnetic moment in Gd-focused ion-beam-implanted GaN Appl. Phys. Lett. 89, 062503 (2006); 10.1063/1.2267900
Ferromagnetism and anomalous Hall effect in Co x Ge 1 x Appl. Phys. Lett. 89, 042504 (2006); 10.1063/1.2236266
Carrier-induced ferromagnetism in Ge(Mn,Fe) magnetic semiconductor thin-film structures Appl. Phys. Lett. 88, 222508 (2006); 10.1063/1.2208552
Ferromagnetic GeMn thin film prepared by ion implantation and ion beam
induced epitaxial crystallization annealing
C. H. Chen,1H. Niu,2D. C. Yan,3H. H. Hsieh,4C. P. Lee,1and C. C. Chi3
1
Center for Nano Science and Technology, National Chiao Tung University, Hsin Chu 30010, Taiwan
2
Nuclear Science and Technology Development Center, National Tsing Hua University, Hsin Chu 30013, Taiwan
3
Department of Physics, National Tsing Hua University, Hsin Chu 30013, Taiwan
4
Department of Electrical Engineering, Chung Cheng Institute of Technology, Ta-shi, Tao-yuan 335, Taiwan
(Received 8 March 2012; accepted 1 June 2012; published online 15 June 2012)
Ferromagnetic GeMn was prepared by Mn implantation followed by ion beam-induced epitaxial crystallization annealing. The damage caused by Mn implantation was repaired by subsequent helium ion irradiation. Various structural analyses were performed and Mn ions were found to incorporate uniformly into the Ge lattice without the formation of any secondary phases. The remnant magnetic moment exhibited room temperature ferromagnetism. Anomalous Hall effect and field dependent magnetization were measured at the same time at room temperature indicating spin polarized free carrier transport. Additional measurement using x-ray magnetic circular dichroism also revealed that the carriers were spin-polarized.VC 2012 American Institute of Physics.
[http://dx.doi.org/10.1063/1.4729752]
Since the discovery of ferromagnetic ordering of InMnAs,1 diluted magnetic semiconductors (DMS)1–3,28,29have attracted considerable attention because of their potential applications in spintronics. GeMn, a group-IV DMS, has received considerable attention in recent years because of its compatibility with main-stream silicon technology.3–7 Over the years, several research groups have attempted to prepare ferromagnetic GeMn thin films using various methods.5,7–10 Although these samples exhibited ferromagnetic ordering,4,5,11–13,30 it was difficult to form a connection between the magnetic properties and the electronic transport properties.14
Intrinsically GeMn is less stable than GaMnAs. Because of limited solid solubility of Mn in Ge, GeMn thin films have to be prepared in a non-equilibrium way such as by low temperature-MBE growth or by ion-implantation. Even so, phase separation resulting in nano-metal cluster formation can easily happen when samples are subject to elevated tem-perature processes.9,12,13,15–17Once secondary phase clusters are formed in the Ge matrix, it is difficult to remove them by conventional post-annealing processes.
Ion implantation is a convenient method in incorporat-ing impurities into semiconductors at non-equilibrium condi-tions. However, thermal annealing process is usually needed to repair the damage in the crystal caused by implantation. For GeMn thin films, post-annealing has to be done in a very short time to prevent Mn atoms from being gathering and precipitating into clusters. Scarpulla et al.18 and Zhou et al.10,19have demonstrated the use of pulsed laser anneal-ing to form GaMnAs and GeMn thin films.
In this study, we prepared ferromagnetic GeMn films using ion beam-induced epitaxial crystallization (IBIEC) annealing after Mn implantation.20,21,31The GeMn films pre-pared this way are extremely stable without any evidence of phase separation. The ferromagnetic properties were meas-ured at room temperature. The anomalous Hall effect was observed at the same time and correlated very well with the magnetization measurement.
In IBIEC annealing, light ions are used to irradiate a pre-viously implanted crystal to repair the damage.21Part of the energy from the incident ions is transferred by nuclear stop-ping power to the dislocated host atoms to restore the crystal-linity. In this work, we choose helium ions, which are light enough to go deep and at the same time to minimize the sur-face layer damage, which is usually a problem when heavy ion beam irradiation is used.
Mn-implantation was performed on p-type (001) Ge wafers, which were doped with Ga at a concentration of 1.1 1018cm3. To form a uniform Mn distribution in the implanted region, we implanted Mnþ ions with multiple energies of 70, 120, 170, 250, and 350 keV. Several samples were prepared with a total implanted dose varied from 5.5 1014 to 2.2 1016cm2, which corresponds to a Mn atomic percentage of 0.125% to 5.0% in the implanted region. The substrate was maintained at room temperature during implantation to prevent secondary phase formation. IBIEC treatment was then carried out using He ions at a beam current of 0.4 lA cm2with an energy of 350 keV for 2 h. During IBIEC treatment, the substrate was kept at a con-stant temperature of 523 K. After IBIEC, the samples were slightly etched in a NaOHþ H2O2solution to remove any
possible oxide layer that may have formed during the im-plantation processes. Magneto-transport properties were measured using a Hall bar structure with the magnetic field applied perpendicularly to the sample surface. To understand the electronic structure of the prepared GeMn thin films, we used x-ray circular dichroism32,33 (XMCD), with measure-ment taken at the National Synchrotron Radiation Research Center, to analyze possible crystal formation of GeMn and the spin polarization of its carriers. The XMCD was per-formed at 78 K.
The structure of GeMn and the distribution of Mn were analyzed by double crystal x-ray diffraction (DCXRD) and SIMS. The CuKa1x-ray (0.1542 nm) was selected by double
Si(111) single crystals. The XRD spectra of the GeMn thin
film around the (004) peak before and after IBIEC annealing are displayed in Fig. 1. Only one broadened peak was observed for the as-implanted sample indicating the Ge crys-tal had been damaged during Mn implantation. After IBIEC annealing, however, an additional shoulder peak appeared on the smaller angle side of the Ge (004) peak. This is a clear indication of the formation of a compressively strained GeMn film. The IBIEC treatment has caused the damaged Ge to recrystallize. But, because the smaller Mn atoms sit on the substitutional sites, the GeMn film is compressively strained. We have also checked Ge(224) and Ge(311) peaks, and similar shoulder peaks were observed. The formation of the compressively strained GeMn film was also confirmed by a separate study using XAFS.22 To make sure there was no secondary phase formation, we carefully scanned the XRD spectrum, even in logarithmical scale, using grazing incident x-ray diffraction. All the IBIEC annealed GeMn samples showed highly crystallized structure without the for-mation of secondary phases.9,13,16,23We have also performed a cross-sectional TEM study of a sample before and after IBIEC treatment (see supplementary material for high reso-lution TEM pictures34). Figure2shows the TEM picture of IBIEC annealed sample. The clear selected area diffraction (SAD) points shows that the amorphous region caused by Mn implant was repaired and the crystal lattice was restored.
Time-of-flight secondary ion mass spectrometry (SIMS) was used to determine Mn distribution in Ge shows the measured Mn profiles for samples with various Mn doses before and after IBIEC treatment (also see supplementary material for SIMS profiles34). The SIMS profiles do not actually change after IBIEC and are generally very similar to one another. This implies that the Mn atoms stayed where they were implanted during the IBIEC induced recrystalliza-tion process and the formarecrystalliza-tion of the GeMn alloy, clearly indicating no Mn segregation took place. This is quite differ-ent from the results obtained using convdiffer-entional thermal annealing techniques,24,25 where serious secondary phase precipitation usually takes place. The actual total Mn implanted doses calculated from the SIMS profiles were found to be in good agreement with intended implanted doses.
The magnetic properties of Mn-implanted samples were measured using a superconducting quantum interference de-vice magnetometer (SQUID). Figure 3shows the measured magnetization as a function of magnetic field at 5 K and 300 K for a 5% Mn sample. The diamagnetic background was subtracted from the data. Clear ferromagnetic behavior was observed. Elongated hysteresis loops are also visible in the figure. Figure 4 shows the remnant magnetic moment, measured after a 1 T-field was applied and then removed at low temperature, as a function of temperature. These results indicate that the IBIEC-annealed Mn-implanted samples have ferromagnetism above room temperature.
One of the most important criteria for DMS to be used in spintronics applications is to have a spin polarized trans-port property. The ferromagnetic behavior of the material has to be able to mediate the formation of spin polarization for the free carriers. In the past, however, such spin polarized transport property often fails to go hand in hand with the fer-romagnetic behavior of the material. In this study, we studied the transport property of our samples together with their magnetic properties. We performed Hall measurement to look for anomalous spin polarized transport behavior. Anom-alous Hall effect (AHE) was clearly observed. Figure 5
shows the measured Hall resistance (at room temperature) as a function of the magnetic field from1 T to 1 T for a 2% Mn sample. To correlate the AHE behavior with the
FIG. 1. DCXRD x-2h scan revealing the formation of GeMn at the low angle side of Ge (004) peak after IBIEC annealing.
FIG. 2. The TEM images of IBIEC-annealed Ge:Mn sample, the inset is the SAD pattern which is selected from the regrowth area.
FIG. 3. Field dependent magnetization for a 5% Mn sample at 5 K and 300 K. The magnetization, expressed in number of Bohr magneton per Mn ion, was obtained by dividing the measured magnetization by the total dose of the implanted Mn atoms. The saturated moment is 1.58 lBat 5 K and
1.03 lBat 300 K, and magnetization reached saturation around 0.15 T.
magnetization of the material, the measured field dependent magnetization is shown in the same figure. It should be men-tioned that the background of the normal Hall effect due to the p-type doped Ge substrate was subtracted from the meas-ured data to get the AHE signal. The field dependent magnet-ization and the AHE behavior correlate with each other perfectly. They have similar saturation behavior and saturate at the same field strength (þ0.15 T). The AHE behavior was observed from 20 K to 300 K (the lowest temperature of 20 K was due to instrumentation limitation). The fact that the AHE correlates with the ferromagnetic behavior clearly indi-cates that the spin polarized carriers that are responsible for current transport are directly related to the ferromagnetic property of the material.
The ferromagnetism of GeMn has been postulated by RKKY model26as a result of magnetic moment coupling of Mn atoms mediated by holes. We have performed XMCD to study the electronic structure of doped Mn ions and the ori-gin of ferromagnetism. The 2p-XAS and XMCD signals of the samples were measured either with total electron yield (TEY) or total fluorescence yield (TFY). Here the energy re-solution of synchrotron radiation was set to 0.4 eV. Figure6
shows the TFY of XMCD for the 2p-3d transitions of Mn. The measurement was carried out at 78 K for the sample
with 4% Mn. The difference in the XAS spectra at 2p1/2and
2p2/3obtained by excitations from opposite circularly
polar-ized light indicates that the origin of ferromagnetism is indeed from the unpaired d shell electrons of Mn atoms. The characteristics of the 2p-XAS spectrum were also used to assess the electronic valence state of Mn ions in the Ge ma-trix. From the 2p3/2 peak positions in TEY and the sharp
peaks of TEY (see inset) we may infer that the electronic va-lence state was 2þ. This state also indicates that Mn ions are on the substitutional sites in the Ge lattice. These results pre-sented above also provide evidence that the Mn ions serve as acceptors in Ge and the free holes are spin polarized to carry the magnetic property of the material.
We have estimated the percentage of Mn atoms that are ferromagnetic in our samples. The saturated magnetic moment divided by the total number of implanted Mn atoms gives the average magnetic moment per atom. Since a fully magnetized Mn atom has 2.5lB,
27
the ratio of the average magnetic moment per atom and 2.5lB gives the percentage
of ferromagnetic Mn atoms. For the 5% Mn sample, the esti-mated magnetization ratio is 63%. This number is signifi-cantly higher than what has been reported for Mn-implanted Ge films with conventional annealing techniques.10,17,19,25
In summary, we have prepared ferromagnetic GeMn films using Mn implantation and IBIEC annealing. He irradi-ation was found to be very effective in removing damage caused by ion-implantation. Several structural analyses were carried out and showed that the Mn ions were uniformly incorporated into the Ge lattice without the formation of sec-ondary phases. The ferromagnetic property along with the anomalous Hall effect was measured at room temperature. The 2p-XAS measurements showed that Mn ions were incor-porated in the substitutional sites of the Ge lattice with 2þ state. The correlation of the magnetic behavior and the trans-port behavior, along with the XMCD result, provides direct evidence that the spin polarized transport is directly related to the ferromagnetic property of the GeMn film. With our IBIEC technique, a very high magnetization ratio of 63% for the Mn ions was obtained. This study showed that IBIEC is a very valuable tool in repairing damages for thin films pre-pared in a non-equilibrium process. It is particularly useful for the fabrication of DMS thin films, where the magnetic
FIG. 4. Temperature dependence of the remnant magnetization for samples with Mn concentration: 0.125%, 2%, and 5%.
FIG. 5. Comparison between magnetization moment (emu) and Hall resist-ance (Rxy) curves with magnetic fields from1 T to 1 T. The signal is
scaled for visibility.
FIG. 6. Experimental Mn 2p-XAS spectra of the GeMn0.04measured at
78 K, the MCD signal can be obtained from the difference in the spectra. The inset shows the total electron yields of the 2p-XAS spectrum.
impurities need to be incorporated into the host material beyond the solid solubility.
This work was supported by the National Science Coun-cil under Grant Nos. NSC-99-2120-M-009-009, NSC-100-2120-M-009-005, and NSC-100-2221-E-007-124.
1
H. Munekata, H. Ohno, S. von Molnar, A. Segmu¨ller, L. L. Chang, and L. Esaki,Phys. Rev. Lett.63(17), 1849 (1989).
2H. Ohno, A. Shen, F. Matsukura, A. Oiwa, A. Endo, S. Katsumoto, and Y.
Iye,Appl. Phys. Lett.69(3), 363 (1996).
3
Y. D. Park, A. T. Hanbicki, S. C. Erwin, C. S. Hellberg, J. M. Sullivan, J. E. Mattson, T. F. Ambrose, A. Wilson, G. Spanos, and B. T. Jonker, Sci-ence295(5555), 651 (2002).
4S. Cho, S. Choi, S. C. Hong, Y. Kim, J. B. Ketterson, B.-J. Kim, Y. C.
Kim, and J.-H. Jung,Phys. Rev. B66(3), 033303 (2002).
5
F. Tsui, L. He, L. Ma, A. Tkachuk, Y. S. Chu, K. Nakajima, and T. Chi-kyow,Phys. Rev. Lett.91(17), 177203 (2003).
6D. Bougeard, N. Sircar, S. Ahlers, V. Lang, G. Abstreiter, A. Trampert, J.
M. LeBeau, S. Stemmer, D. W. Saxey, and A. Cerezo,Nano Lett.9(11), 3743 (2009).
7M. Jamet, A. Barski, T. Devillers, V. Poydenot, R. Dujardin, P.
Bayle-Guillemaud, J. Rothman, E. Bellet-Amalric, A. Marty, J. Cibert, R. Mattana, and S. Tatarenko,Nat. Mater.5(8), 653 (2006).
8
J. Ye, Y. Jiang, Q. Liu, Y. Sun, Z. Pan, and S. Wei,J. Phys.: Conf. Series
190(1), 012104 (2009).
9L. Ottaviano, M. Passacantando, S. Picozzi, A. Continenza, R. Gunnella,
A. Verna, G. Bihlmayer, G. Impellizzeri, and F. Priolo,Appl. Phys. Lett.
88(6), 061907 (2006).
10S. Zhou, D. Bu¨rger, A. Mu¨cklich, C. Baumgart, W. Skorupa, C. Timm, P.
Oesterlin, M. Helm, and H. Schmidt,Phys. Rev. B81(16), 165204 (2010).
11
S. Sugahara, K. L. Lee, S. Yada, and M. Tanaka,Jpn. J. Appl. Phys.44, L1426 (2005).
12L. Ottaviano, M. Passacantando, A. Verna, R. Gunnella, E. Principi, A. Di
Cicco, G. Impellizzeri, and F. Priolo, J. Appl. Phys. 100(6), 063528 (2006).
13
S. Zhou,Appl. Phys. Lett.95(19), 192505 (2009).
14S. Zhou and H. Schmidt,Materials3(12), 5054 (2010).
15Y. Wang,Appl. Phys. Lett.
92(10), 101913 (2008); D. D. Dung, W. Feng, Y. Shin, and S. Cho,J. Appl. Phys.109(7), 07C310 (2011).
16
I. T. Yoon, C. J. Park, S. W. Lee, T. W. Kang, D. W. Koh, and D. J. Fu,
Solid-State Electronics52(6), 871 (2008).
17W. Yin, L. He, M. C. Dolph, J. Lu, R. Hull, and S. A. Wolf,J. Appl. Phys.
108(9), 093919 (2010).
18
M. A. Scarpulla, U. Daud, K. M. Yu, O. Monteiro, Z. Liliental-Weber, D. Zakharov, W. Walukiewicz, and O. D. Dubon,Phys. B: Condens. Matter
340–342(0), 908 (2003).
19
S. Zhou, D. Burger, W. Skorupa, P. Oesterlin, M. Helm, and H. Schmidt,
Appl. Phys. Lett.96(20), 202105 (2010).
20C. H. Chen, H. Niu, H. H. Hsieh, C. Y. Cheng, D. C. Yan, C. C. Chi, J. J.
Kai, and S. C. Wu,J. Magn. Magn. Mater.321(9), 1130 (2009).
21
S. T. Johnson, J. S. Williams, E. Nygren, and R. G. Elliman,J. Appl. Phys.
64(11), 6567 (1988).
22C. H. Chen, H. Niu, H. H. Hsieh, J. W. Chiou, and C. P. Lee, “XAFS
Study on Mn-implanted Ge treated by IBIEC” (unpublished).
23
J. Chen, Kang L. Wang, and K. Galatsis,Appl. Phys. Lett.90(1), 012501 (2007).
24A. Verna, L. Ottaviano, M. Passacantando, S. Santucci, P. Picozzi, F.
D’orazio, F. Lucari, M. De Biase, R. Gunnella, M. Berti, A. Gasparotto, G. Impellizzeri, and F. Priolo,Phys. Rev. B74(8), 085204 (2006).
25
M. C. Dolph, T. Kim, W. Yin, D. Recht, W. Fan, J. Yu, M. J. Aziz, J. Lu, and S. A. Wolf,J. Appl. Phys.109(9), 093917 (2011).
26M. J. Caldero´n and S. Das Sarma,Ann. Phys.
322(11), 2618 (2007).
27
T. C. Schulthess and W. H. Butler,J. Appl. Phys.89(11), 7021 (2001).
28
F. Matsukura, A. Oiwa, A. Shen, Y. Sugawara, N. Akiba, T. Kuroiwa, H. Ohno, A. Endo, S. Katsumoto, and Y. Iye,Appl. Surf. Sci.113–114, 178 (1997).
29
H. Ohno,Science281(5379), 951 (1998).
30
O. Riss, A. Gerber, I. Ya Korenblit, A. Suslov, M. Passacantando, and L. Ottaviano,Phys. Rev. B79(24), 241202 (2009).
31C. P. Z. Werner, M. Barlak, J. Piekoszewski, A. Korman, R. Heller, W.
Szymczyk, and K. Bochen´ska, Nukleonika 56(1), 4 (2011).
32
C. T. Chen, F. Sette, Y. Ma, and S. Modesti,Phys. Rev. B42(11), 7262 (1990).
33
G. van der Laan and B. T. Thole,Phys. Rev. B43(16), 13401 (1991).
34
See supplementary material at http://dx.doi.org/10.1063/1.4729752 for microstructure and magnetic properties evidenced by HRTEM, SIMS, and SQUID of as-implanted and IBIEC annealed samples.
