Hydrogen Incorporation, Diffusivity and Evolution in Bulk ZnO

 

K. Ip, M.E. Overberg, Y.W. Heo, D.P. Norton and S.J. Pearton

Department of Materials Science and Engineering

University of Florida, Gainesville, FL 32611, USA

 

C.E. Stutz

AFRL/MLPS

Wright-Patterson AFB, OH 45433, USA

 

S.O. Kucheyev*, C. Jagadish and J.S. Williams

Department of Electronic Materials Engineering

Australian National University, Canberra, ACT 0200 , Australia

 

B. Luo and F. Ren

Department of Chemical Engineering

University of Florida, Gainesville, FL 32611, USA

 

D.C. Look

Semiconductor Research Center

Wright State University, Dayton, OH 45435, USA

 

J.M. Zavada

US Army Research Office

Research Triangle Park, NC 27709, USA

 

 

ABSTRACT

 

            Hydrogen is readily incorporated into bulk, single-crystal ZnO during exposure to plasmas at moderate (100-300°C) temperatures.  Incorporation depths of >25 mm were obtained in 0.5h at 300°C, producing a diffusivity of ~8x10-10 cm2/V×s at this temperature.  The activation energy for diffusion is 0.17± 0.12 eV, indicating an interstitial mechanism.  Subsequent annealing at 500-600°C is sufficient to evolve all of the hydrogen out of the ZnO, at least to the sensitivity of Secondary Ion Mass Spectrometry (<5x1015 cm-3).  The thermal stability of hydrogen retention is slightly greater when the hydrogen is incorporated by direct implantation relative to plasma exposure, due to trapping at residual damage.

Introduction

            There has been considerable recent attention paid to the properties of hydrogen in ZnO and related oxides.(1-17)  The primary interest comes from the prediction that interstitial atomic hydrogen will introduce a shallow donor state in ZnO(1-4), which is supported by some experimental data for its muonium counterpart (5,6), from electron paramagnetic resonance of bulk single-crystals(7) and from the fact that hydrogen can lead to electron conduction in other wide bandgap oxides such as SnO.(18)  It is clear that the presence of hydrogen in the growth on processing ambient can significantly affect the electrical and optical properties of ZnO (8-17), but there is little systematic data available on its diffusivity and thermal stability when introduced by different methods.

            In this paper we report on an investigation on the diffusion of hydrogen in ZnO during exposure to 2H plasmas, a comparison of the thermal stability of hydrogen incorporated by either direct implantation or by plasma exposure and finally on changes in the electrical and optical properties of the ZnO or a result of hydrogen incorporation.  The hydrogen is found to diffuse very rapidly even at temperatures of 100°C, but can be evolved from the ZnO by subsequent annealing under N2 at £600°C.

 

Experimental

            Bulk wurtzite (0001) ZnO crystals from Eagle-Picher (grade I quality) were used in all cases.  The samples were nominally undoped with as-received n-type carrier concentrations of ~1017 cm-3 and a room temperature mobility of 190 cm2/V×s.  The samples were exposed to 2H plasmas at temperatures of 100-300°C in a Plasma Therm 720 series reactor operating at 900 mTorr with 50W of 13.56MHz power.  Some of these samples were subsequently annealed at temperatures up to 600°C under flowing N2 ambients for 5 mins.  We also implanted 100 keV 2H on 1H ions at doses of 1015 – 1016 cm-2 and then used Secondary Ion Mass Spectrometry (SIMS) measurements to obtain the hydrogen or deuterium profiles as a function of post-implant annealing temperature.(19)  The electrical properties of some of the samples were examined by electrochemical capacitance-voltage (C-V) measurements using a 0.2 M NaOH/0.1 M EDTA electrolyte as the rectifying contact.  Finally, optical properties were measured using photoluminescence (PL) spectroscopy at variable temperatures, with a He-Cd laser as the excitation source.

 

Results and Discussion

            Figure 1 shows SIMS profiles of 2H in plasma exposed ZnO, for different sample temperatures during the plasma treatment.  The profiles follow those expected for diffusion from a constant or semi-infinite source, i.e.

where C(x,t) is the concentration at a distance x for diffusion time t, Co is the solid solubility and D is the diffusivity of 2H in ZnO.(20)  The incorporation depths of 2H are very large compared to those in GaN or GaAs under similar conditions, where depths of 1-2 mm are observed.(21,22)  It is clear that hydrogen must diffuse as an interstitial, with little trapping by the lattice elements or by defects or impurities.  The position of H in the lattice after immobilization has not yet been determined experimentally, but from theory the lowest energy states for H+ is at a bond-centered position forming an O-H bond, while for H2 the anti-bonding Zn site is most stable.(1)

            Using a simple estimate of the diffusivity D, from D = X2/4t, and where X is taken to be the distance at which 2H concentration has fallen to 5x1015 cm-3 in Figure 1, we can estimate the activation energy for diffusion from the data shown in Arrhenius form in Figure 2.  The extracted activation energy, Ea, is 0.17 ± 12 eV for 2H in ZnO.  Note that the absolute diffusivities of 1H would be ~40% larger because of the relationship for diffusivities of isotopes, i.e.(20)

 

The small activation energy is consistent with the notion that the atomic hydrogen diffuses in interstitial form.

            Figure 3 shows SIMS profiles of a ZnO sample exposed to a 2H plasma of 0.5h at 200°C, then annealed for 5 mins under N2 at different temperatures.  There is significant loss of 2H even after a short anneal at 400°C, with virtually all of it evolved out of the crystal by 500°C.  This is in sharp contrast to 2H in GaN, where much higher temperatures (³800°C) are needed to evolve the deuterium out of the sample.(21,22)

            To compare this data to the thermal stability of 2H incorporated by direct implantation(17), Figure 4 shows the percentage of 2H remaining (measured by SIMS) as a function of annealing temperature for incorporation by either plasma exposure or implantation.  The 2H is slightly more thermally stable in the latter case, most likely due to trapping at residual damage in the ZnO carried by the nuclear stopping process.  Lavrov et al.(23) have identified two hydrogen-related defects in ZnO, by using local vibrational mode spectroscopy.  The H-I center consists of a hydrogen atom at the bond centered site, while the H-II center contains two inequivalent hydrogen atoms bound primarily to two oxygen atoms.(23)

                Figure 5 shows donor concentration profiles in the ZnO before and after plasma exposure and following subsequent annealing.  The 2H plasma treatment causes an increase in donor concentration, consistent with past reports.(9)  In that case, the effect was attributed to hydrogen passivation of compensating acceptor impurities present in the as-grown ZnO epitaxial layers.(9)  An alternative explanation is that the hydrogen induces a donor state and thereby increases the free electron concentration.(1)  Subsequent annealing reduces the carrier density to slightly below the initial value in the as-received ZnO, which may indicate that it contained hydrogen as a result of the growth process.  We emphasize that the n-type conductivity probably arises from multiple impurity sources(24-26) and we cannot unambiguously assign all of the changes to hydrogen.

            Figure 6 shows the PL spectrum from a plasma treated sample as a function of measurement temperature.  The sample shows strong band-edge luminescence and a small deep-level band (~2.6 eV).  Past reports have shown that the efficiency of band-edge emission was increased by plasma hydrogenation of various types of ZnO(10), but that the degree of improvement depended on the impurity and defect concentration in the original samples.(10,12)  We did not observe any significant difference in the intensity or shape of the PL spectra as a result of plasma hydrogenation of our samples.

            More detail on the measurement temperature dependence of the bandedge and deep-level emissions from the plasma deuterated ZnO are shown in Figure 7.  As expected and as reported previously(12), the bandedge intensity increases significantly as the temperature is lowered and the deep level emission is quenched.  The overall intensity of the plasma treated ZnO remains much higher than the material hydrogenated by direct implantation of protons or deuterons.  Figure 8 shows 300K PL spectra from ZnO after 2H implantation at a dose of 10+15+ cm-3, followed by annealing at different temperatures.(17) The implantation step severely degrades the bandedge intensity, and even annealing at 700°C where all of the  2H has been evolved from the ZnO leave the intensity about 2 orders of magnitude lower than in the implanted material.(17)

 

Summary and Conclusions

            Hydrogen is found to exhibit a very rapid diffusion in ZnO when incorporated by plasma exposure, with D of 8.7x10-10 cm2/VS at 300°C.  The activation energy for diffusion is indicative of interstitial motion.  All of the plasma-incorporated hydrogen is removed from the ZnO by annealing at ³500°C.  When the hydrogen is incorporated by direct implantation, the thermal stability is somewhat higher, due to trapping at residual damage.  The free electron concentration increases after plasma hydrogenation, consistent with the small ionization energy predicted for H in ZnO(1) and the experimentally measured energy of 60 ± 10 meV for muonium in ZnO.(5)  The electrical activity and rapid diffusivity of H or ZnO must be taken into account when designing device fabrication processes such as deposition of dielectrics using SiH4 as a precursor or dry etching involving use of CH4/H2/Ar plasmas since these could lead to significant changes in near-surface conductivity.

 

Acknowledgments

            The work at UF is partially supported by ARO DAAO 190210420 and NSF (DMR0101438 and CTS 994473).

References

1.         C.C. van de Walle, Phys. Rev. Lett. 85 1012 (2000).

2.         C.C van de Walle, Phys. Stal. Solidi B229 221 (2002).

3.         C. Kilic and Z. Zunger, Appl. Phys. Lett. 81 73 (2002).

4.         C.C. van de Walle, Physica B 308-310 899 (2001).

5.         S.J.F. Cox, E.A. Davis, S.P. Cottrell, P.J.C. King, J.S. Lord, J.M. Gil, H.V. Alberto, R.C. Vilao, DJ.P. Duarte, N.A .de Campos A. Weidinger, R.L. Lichti and S.J.C. Irving, Phys. Rev. Lett. 86 2601 (2001).

6.         S.J.F. Cox, E.A. Davis, P.J.C. King, J.M. Gil, H.V. Alberto, R.C. Vilao, J.P. Duarte, N.A. de Campos and R.L. Lichti, J. Phys. C 13 9001 (2001).

7.         D.M. Hofmann, A. Hofstaetter, F. Leiter, H. Zhou, F. Henecker, B.K. Meyer, S.B. Schmidt and P.G. Baranov, Phys. Rev. Lett. 88 045504 (2002).

8.         S.J. Baik, J.H. Jang, C.H. Lee, W.Y. Cho and K.S. Lim, Appl. Phys. Lett. 70 3516 (1997).

9.         B. Theys, V. Sallet, F. Jomard, A. Lusson, J.-F. Rommeluere, and Z. Teukam, J. Appl. Phys. 91 3922 (2002).

10.     N. Ohashi, T. Ishigaki, N. Okada, T. Sekiguchi, I. Sakaguchi and H. Haneda, Appl. Phys. Lett. 80 2869 (2002).

11.     V. Bogatu, A. Goldenbaum, A. Many and Y. Goldstein, Phys. Stal. Solidi B212 89 (1999).

12.     T. Sekiguchi, N. Ohashi and Y. Terada, Jap. J. Appl. Phys. 36 L289 (1997).

13.     S. Tüzemen, G. Xiong, J. Wilkinson, B. Mischuck, K.B. Ucer and R.T. Williams

14.     C.S. Han, J. Jun and H. Kim, Appl. Surf. Sci. 175/176 567 (2001).

15.     Y. Natsume and H. Sakata, J. Mater. Sci. Materials in Electronics 12 87 (2001).

16.     Y.-S. Kang, H.Y. Kim and J.Y. Lee, J. Electrochem. Sci. 147 4625 (2000).

17.     K.Ip,M.E.Overberg,Y.W.Heo,D.P.Norton,S.J.Pearton,S.O.Kucheyev,C.Jagadish,J.S.Williams,R.G.Wilson and J.M. Zavada, Appl. Phys. Lett.(Nov 18 issue,2002).

18.     S. Samson and C.G. Fonstad, J. Appl. Phys. 44 4618 (1973).

19.     SIMS performed at Charles Evans and Associates using a Cameca system.

20.     R.J. Borg and C.J. Dienes, An Introduction to Solid State Diffusion (Academic Press, Boston. 1988).

21.     R.G. Wilson, S.J. Pearton, C.R. Abernathy and J.M. Zavada, J. Vac. Sci. Technol. A13 719 (1995).

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

23.     E.V. Lavrov, J. Weber, F. Borrnet, C.G. Van de Walle and R. Helbig, Phys. Rev. B (in press).

24.     D.G. Look, Mat. Sci. Eng. B80 383 (2001).

25.     D.G. Look, J.W. Hemsky and J.R. Sizelove, Phys. Rev. Lett. 82 2552 (1999).

26.     D.C. Look, D.C. Reynolds, J.W. Hemsky, R.L. Jones and J.R. Sizelove, Appl. Phys. Lett. 75 811 (1999).

Figure Captions

 

Figure 1.   SIMS profiles of 2H in ZnO exposed to deuterium plasmas for 0.5h at different temperatures.

 

Figure 2.   Arrhenius plot of diffusivity for 2H in ZnO.

 

Figure 3.   SIMS profiles of 2H in ZnO exposed to deuterium plasma for 0.5h at 200°C and then annealed at 400°C or 500°C for 5 mins.

 

Figure 4.   Percentage of retained 2H incorporated by direct implantation or plasma exposure, as a function of subsequent annealing temperature.  The inset shows the data on a log scale.

 

Figure 5.   Donor concentration profiles in ZnO before and after plasma exposure and after subsequent annealing.

 

Figure 6.   PL spectra from 2H plasma exposed ZnO.

 

Figure 7.   Detailed bandedge and deep level emission PL spectra from 2H plasma exposed ZnO.

 

Figure 8.   300K PL spectra from 2H implanted ZnO, as a function of subsequent anneal temperature.


 
 


 
 

 

 




 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 


* Present address, Lawrence Livermore National Laboratory, Livermore, CA