Journal Search Engine
Search Advanced Search Adode Reader(link)
Download PDF Export Citaion korean bibliography PMC previewer
ISSN : 1225-0562(Print)
ISSN : 2287-7258(Online)
Korean Journal of Materials Research Vol.28 No.4 pp.208-213
DOI : https://doi.org/10.3740/MRSK.2018.28.4.208

Heat Treatment of Carbonized Photoresist Mask with Ammonia for Epitaxial Lateral Overgrowth of a-plane GaN on R-plane Sapphire

Dae-sik Kim1, Jun-hyuck Kwon1, Junggeun Jhin2, Dongjin Byun1
1Department of Materials Science & Engineering, Korea University, 145 Anam-ro, Seongbuk-gu, Seoul 02841, Republic of Korea
2LED Procurement Team, LG Innotek, 570 Hyuam-ro, Munsan-eup, Paju-si, Gyeonggi-do 10842, Republic of Korea
(*

These two authors are equally contributed as first authors.)


Corresponding author E-Mail : dbyun@korea.ac.kr (D. Byun, Korea Univ.)
November 2, 2017 March 12, 2018 March 16, 2018

Abstract


Epitaxial (1120) a-plane GaN films were grown on a (1102) R-plane sapphire substrate with photoresist (PR) masks using metal organic chemical vapor deposition (MOCVD). The PR mask with striped patterns was prepared using an ex-situ lithography process, whereas carbonization and heat treatment of the PR mask were carried out using an in-situ MOCVD. The heat treatment of the PR mask was continuously conducted in ambient H2/NH3 mixture gas at 1140 °C after carbonization by the pyrolysis in ambient H2 at 1100 °C. As the time of the heat treatment progressed, the striped patterns of the carbonized PR mask shrank. The heat treatment of the carbonized PR mask facilitated epitaxial lateral overgrowth (ELO) of a-plane GaN films without carbon contamination on the R-plane sapphire substrate. Thhe surface morphology of a-plane GaN films was investigated by scanning electron microscopy and atomic force microscopy. The structural characteristics of a-plane GaN films on an R-plane sapphire substrate were evaluated by ω-2θ high-resolution X-ray diffraction. The a-plane GaN films were characterized by X-ray photoelectron spectroscopy (XPS) to determine carbon contamination from carbonized PR masks in the GaN film bulk. After Ar+ ion etching, XPS spectra indicated that carbon contamination exists only in the surface region. Finally, the heat treatment of carbonized PR masks was used to grow high-quality a-plane GaN films without carbon contamination. This approach showed the promising potential of the ELO process by using a PR mask.



초록


    © Materials Research Society of Korea. All rights reserved.

    This is an Open-Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (http://creativecommons.org/licenses/by-nc/3.0) which permits unrestricted non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited.

    1. Introduction

    GaN and its alloys with indium or aluminum have attracted much attention for their successful application in optoelectronics and microelectronics.1-3) The majority of these GaN-based structures is grown along the [0001] c-axis of the wurtzite crystal structure. But the optoelectronic devices grown along [0001] direction suffer from strong undesirable spontaneous and piezoelectric polarization fields, which give rise to internal electrical fields.4-6) The internal electrical fields can spatially separate the electrons and holes at the opposite interface of the quantum wells and reduce the overlap of their wave functions,6,7) which cause a reduction of the recombination efficiency in light-emitting devices and red shift of the emission wavelength.4-6,8,9) One of the promising methods to eliminate polarization-induced electric field effects is to grow group-III nitride layers which are no polarization field in directions perpendicular to the c-axis, because the polar planes lie in the c-axis direction.10-12) The (1010) m-plane and (1120) a-plane of GaN are the two main facets of the no polariztion field.13,14) But the large lattice mismatch and thermal expansion coefficient difference between GaN and sapphire substrate generate a high dislocation density and even cracks in the epitaxial layer.15,16) Therefore, many studies have focused on stress and defect management techniques such as Epitaxial lateral overgrowth (ELO)17-19) to reduce the density of cracks and misfit dislocations. The ELO is a promising technique, whose value has already been demonstrated in fabrication of devices.20) Nevertheless, using the ELO process is limited due to complicated processes and mask materials, such as SixNy or SiO2 which are potential sources of impurities. Therefore, several approaches for reducing the complicated processes of conventional ELO using by mask-less or single-step ELO process have been studied.13,16,21,22)

    In this paper, we reported on the effective single-step ELO process using the photoresist (PR) masks without any carbon contamination by optimizing the heat treatment time in H2/NH3 mixture gas ambient during growth of the a-plane GaN on the (1102) R-plane sapphire substrate.

    2. Experimental

    The R-plane sapphire substrate, 2-inch, was prepared by using PR (AZ 1512) to substitute for mask materials such as SixNy or SiO2 in conventional ELO process. As shown Fig. 1(a-c), the stripes of patterns aligned parallel to primary orientation flat (45° from c-axis projection onto R-plane)23) were fabricated by using standard ex-situ photolithography. Field-emission scanning electron microscopy (FE-SEM) image presented in Fig. 1(d) shows the striped patterns with 3.4-μm wide opening region and 12-μm wide PR mask region on the R-plane sapphire substrate. As shown the process time chart in Fig. 2, the PR mask on the R-plane sapphire substrate were carbonized by pyrolysis at 1100 °C in the H2 ambient for 5 minutes using by metal organic chemical vapor deposition (MOCVD). The heat treatment of carbonized PR masks was continuously conducted at 1140 °C in the H2/NH3 mixture gas ambient for different process times: 1 minute, 5 minutes, and 10 minutes after carbonization process. The main growth of the a-plane GaN was carried out at 1140 °C for 60 minutes. The a-plane GaN films were grown on R-plane sapphire substrate (squares of dimensions 10 mm × 10 mm) by a home-MOCVD system consisting of a horizontal quartz reactor and silicon carbide susceptor. The trimethylgallium (TMGa) was used as a gallium (Ga) precursors and temperature of TMGa precursor was maintained at −10 °C to control the excess vapor pressure. The TMGa was directly flowed into a horizontal quartz reactor with hydrogen (H2, 99.9999 % purity) carrier gas of 8.0 × 10−3 slm by mass flow controller (MFC). The H2 of 1.5 slm by MFC was supplied as an ambient gas in order to maintain the H2 ambient in the reactor. The ammonia (NH3, 99.9999 % purity) of 1.2 slm by MFC was used reactant gas as a nitrogen (N) source. The a-plane GaN was deposited on the R-plane sapphire substrate under a V/III ratio of 2875. The reactor pressure was maintained at 85 Torr during the all processes. The morphology of the surface was analyzed by optical microscopy (OLYMPUS® BX51M) operating in Nomarski interference contrast mode and FE-SEM (Hitachi S-4300) operated at 15 kV. The crystal quality of a-plane GaN was evaluated by X-ray diffraction (XRD; Bede D1) with double axis diffractometer using Cu-Kα radiation at 1.540598 Å. The roughness of a-plane GaN was investigated by atomic force microscopy (AFM; Veeco Dimension 3000) operating in tapping mode. The GaN (sample B) film’s composition was investigated by X-ray photoelectron spectroscopy (XPS; Thermo scientific MultiLab 2000) with a monochromatized Al-Kα line at 1486.6 eV. The XPS spectra were calibrated using the C1s signal at 284.6 eV.24)

    3. Results and Discussion

    The FE-SEM images presented in Fig. 3(a-c) clearly reveal the morphologies of PR masks on R-plane sapphire substrate after heat treatment in H2/NH3 mixture gas for different process times: (a) 1 minute, (b) 5 minutes, and (c) 10 minutes. As shown in Fig. 3(a), the striped patterns that have same opening region and PR mask region shown in Fig. 1(d) still exist on the substrate surface after the heat treatment for 1 minute. The size of the opening region and the PR mask region was reversed (i.e. the size of the opening region is smaller than that of the PR mask region.) after the heat treatment for 5 minutes shown in Fig. 3(b). After the heat treatment for 10 minutes shown in Fig. 3(c), it was difficult to observe the existence of the striped patterns by the resolution of FE-SEM. Consequentially the size of PR mask shrank as time progresses of the heat treatment by reaction between carbon and NH3. The reaction of carbon or carbon residues with NH3 lead to hydrogen cyanide (HCN) synthesis:25-27)

    Eq1.gif

    Cacace and Wolf28) have investigated the reaction of nucleogenic carbon atoms with NH3 and have demonstrated that methyleneamine, 1, is a primary product. The methyleneamine is undoubtedly formed by insertion of carbon into an NH bond of ammonia to generate aminomethylene, 2, which rearranges to 1.29) Fig. 4(a-i) show Nomarski optical contrast micrographs and FE-SEM images of a-plane GaN on R-plane sapphire for different heat treatment times: 1 minute (Sample A), 5 minutes (Sample B), and 10 minute (Sample C). Fig. 4(a-c) show a Nomarski optical contrast micrograph of coalesced ELO a-plane GaN formed with XAl2O3 axis oriented stripes. The sample B in Fig. 4(b), (e) and (h) showed no V-defects on the surface, and a fine mirror-like flat surface was obtained. Whereas both samples A and C had rough surface and did not sufficiently merge. Fig. 3(d-i) are cross-sectional FE-SEM images of a 6.2-um thick a-plane GaN. As the main growth time increased, a-plane GaN grew on the opening regions and covered the PR mask region by lateral growth. In this process, full coalescence takes place while minimizing the threading dislocations with a smooth and flat surface. The observed SEM behavior confirms the previous reports30) showing that too much heat treatment time must be proscribed. We assume that the accumulated compressive strain energy due to the large lattice mismatch between the carbonitriding layer and the sapphire substrate serves as a driving force to nucleate and grow protrusions. Consequently, the change observed in the time dependence of the heat treatment is closely related to the change of the surface morphology in the carbonitriding layer, i.e. from a flat surface to one with high-intensity protrusions. The structural characteristics of the as-grown a-plane GaN with respect to different heat treatment times were evaluated by XRD. As shown in Fig. 5, the 2θ scans of the sample A, B, and C exhibited several peaks that were assigned to the diffractions from (1102), and (2204) crystal planes of the R-plane sapphire substrate, from the (1120) of aplane GaN. No other diffraction from GaN was observed within the detection limits of this technique. All of this indicated that the GaN films were uniquely (1120) aplane oriented as the a-plane GaN films grown by MOCVD. The (1120) peak is observed clearly at 57.76°, which indicates that the a-plane GaN is highly oriented along the a-axis and has single crystalline character. The full width at half maximum value of GaN peaks are 880 arcsec (sample A), 647 arcsec (sample B), and 718 arcsec (sample C), respectively. This indicates that the crystalline quality of the a-plane GaN improves by optimizing the heat treatment time. The surface roughness of the samples A, B, and C are shown more clearly in the AFM images (50 um × 50 um) in Fig. 6(a-c). The samples A and C were not completely covered by the a-plane GaN film. In contrast, for the same growth conditions, the aplane GaN layer of the sample B completely covered the substrate and presented a smoother surface morphology with no pits or other types of deformation as shown in Fig. 6(b). It was obvious that the surface morphology of a-plane GaN was very much related to the heat treatment time of R-plane sapphire using the PR mask. The root mean square roughness values for samples A, B, and C were 115.0 nm, 8.1 nm, and 21.6 nm, respectively. The chemical composition of a-plane GaN (sample B) was investigated by XPS as shown in Fig. 7(a-d). The sample B was etched with an Ar+ ions beam to remove the atmospheric compounds from surface. Fig. 7(a) shows the peaks observed at binding energies that are characteristics for Ga. The Ga2p1/2 and Ga2p3/2 spectra were observed with positions at 1144.4 (after etching: 1144.2 eV) and 1117.5 eV (after etching: 1117.3 eV), respectively that validated with the values reported for binding energy of Ga-N.24,31) The N1s spectrum in Fig. 7(b) was fitted using three subpeaks located at 396.8 (after etching: 396.9 eV), 395.2 (after etching: 395.1 eV), and 392.5 eV (after etching: 392.4 eV) and thus assigned as N-Ga and two Auger Ga peaks, respectively. As shown Fig. 7(c) and (d), the C1s and O1s peaks centered at 284.6 eV and at 531.1 eV, respectively. Both C1s and O1s peaks were drastically reduced after Ar+ etching, thus indicating that the carbon and oxygen are mainly due to atmospheric exposure. Despite the use of the carbonized PR mask, aplane GaN was grown on the R-plane sapphire substrate without carbon contamination using by heat treatment in the H2/NH3 mixture gas ambient. This mean that the heat treatment is an effective process to prevent carbon contamination from carbon residue of carbonized PR mask.

    4. Conclusion

    In this paper, we report an effective heat treatment of carbonized PR mask for the ELO process of a-plane GaN on R-plane sapphire substrate. Using the PR mask in ELO process has several advantages,13) however there is critical risk that is the carbon contamination from carbon residues into the thin film’s bulk. The heat treatment in the H2/NH3 mixture gas ambient lead to the reaction of carbon or carbon residues with NH3 and thus carbon contamination is prevented from carbonized PR mask by HCN synthesis. The optimized heat treatment process of the carbonized PR mask were found to be an effective method to improve the morphology and the structural of the as-grown a-plane GaN films. Overall, this approach showed the promising potential of the ELO process by using PR mask.

    Figure

    MRSK-28-208_F1.gif

    (a) Coordinated systems in the sapphire crystal, and the Rplane, (b) a sapphire wafer in the R-plane, and the coordinate system in the FEA simulations,23) (c) PR mask patterns aligned parallel to primary orientation flat (45o from c-axis projection onto R-plane), and (d) FE-SEM image of the striped patterns with PR mask on the R-plane sapphire substrate.

    MRSK-28-208_F2.gif

    The process time chart of the corresponding MOCVD system for a-plane GaN ELO growth on R-plane sapphire.

    MRSK-28-208_F3.gif

    FE-SEM images of PR mask after heat treatment at 1140 oC in H2/NH3 mixture gas for: (a) 1 min, (b) 5 min, and (c) 10 min.

    MRSK-28-208_F4.gif

    (a-c) Nomarski optical contrast micrographs (top view) and (d-i) FE-SEM images (cross sectional view) of a-plane GaN on Rplane sapphire for different heat treatment times: (a), (d) and (g): 1 min, (b), (e) and (h): 5 min, (c), (f) and (i): 10 min.

    MRSK-28-208_F5.gif

    XRD 2θ scan profile of a-plane GaN grown on R-plane sapphire substrate for different heat treatment times: (blue line) 1 min, (red line) 5 min, and (orange line): 10 min.

    MRSK-28-208_F6.gif

    AFM images of a-plane GaN grown on R-plane sapphire substrate for different heat treatment times: (a) 1 min, (b) 5 min, (c) 10 min.

    MRSK-28-208_F7.gif

    (a-d) XPS photoelectron peaks of a-plane GaN (sample B) grown on R-plane sapphire substrate. (a) Ga 2p, (b) N1s, (c) C1s, and (d) O1s HR-XPS scans. Upper spectra: before Ar+ ions etching, bottom spectra: after Ar+ ions etching.

    Table

    Reference

    1. S. Nakamura, M. Senoh and T. Mukai, Jpn. J. Appl. Phys., 30, L1708 (1991).
    2. M.A. Khan, A. Bhattarai, J.N. Kuznia and D.T. Oison, Appl. Phys. Lett., 63, 1214 (1993).
    3. H. Morkoc, S. Strite, G.B. Gao, M.E. Lin, B. Sverdlov and M. Burns, J. Appl. Phys., 76, 1363 (1994).
    4. J.S. Im, H. Kollmer, J. Off, A. Sohmer, F. Scholz and A. Hangleiter, Phys. Rev. B, 57, R9435 (1998).
    5. R. Langer, J. Simon, V. Ortiz, N.T. Pelekanos, A. Barski, R. Andre and M. Godlewski, Appl. Phys. Lett., 74, 3827 (1999).
    6. P. Lefebvre, A. Morel, M. Gallart, T. Taliercio, J. Allegre, B. Gil, H. Mathieu, B. Damilano, N. Grandjean and J. Massies, Appl. Phys. Lett., 78, 1252 (2001).
    7. T. Takeuchi, C. Wetzel, S. Yamaguchi, H. Sakai, H. Amano, I. Akasaki, Y. Kaneko, S. Nakagawa, Y. Yamaoka and N. Yamada, Appl. Phys. Lett., 73, 1691 (1998).
    8. F. Bernardini and V. Fiorentini, Phys. Rev. B, 57, R9427 (1998).
    9. R. Cingolani, A. Botchkarev, H. Tang, H. Morkoc, G. Traetta, G. Coli, M. Lomascolo, A.D. Carlo, F.D. Sala and P. Lugli, Phys. Rev. B, 61, 2711 (2000).
    10. P. Waltereit, O. Brandt, A. Trampert, H.T. Grahn, J. Menniger, M. Ramsteiner, M. Reiche and K.H. Ploog, Nature, 406, 865 (2000).
    11. Y.J. Sun, O. Brandt, S. Cronenberg, S. Dhar, H.T. Grahn, K.H. Ploog, P. Waltereit and J.S. Speck, Phys. Rev. B, 67, 041306 (2003).
    12. M.D. Craven, S.H. Lim, F. Wu, J.S. Speck and S.P. Denbaars, Appl. Phys. Lett., 81, 469 (2002).
    13. S. Jang, D. Lee, J-H. Kwon, S-I. Kim, S.Y. Yim, J. Lee, J.H. Park and D. Byun, Jpn. J. Appl. Phys., 52, 115501 (2012).
    14. C. Roder, S. Einfeldt, S. Figge, T. Paskova, D. Hommel, P.P. Paskov, B. Monemar, U. Behn, B.A. Haskell, P.T. Fini and S. Nakamura, J. Appl. Phys., 100, 103511 (2006).
    15. B.H. Kang, J.E. Lee, D-S. Kim, S. Bae, S. Jung, J. Park, J. Jhin and D. Byun, J. Nanosci. Nanotechnol., 16, 11563 (2016).
    16. C-M. Lee, B.H. Kang, D-S. Kim and D. Byun, Korean. J. Mater. Res., 24, 645 (2014).
    17. S-I. Kim, B. Kim, S. Jang, A-Y. Kim, J. Park and D. Byun, J. Cryst. Growth, 326, 200 (2011).
    18. D-S. Kim, W.S. Jeong, H. Ko, J-S. Lee and D. Byun, Thin Solid Films, 641, 2 (2017).
    19. D-S. Kim, C-M. Lee, W.S. Jeong, S.H. Cho, J. Jhin and D. Byun, J. Nanosci. Nanotechnol., 16, 11575 (2016).
    20. C.H. Chiu, H.H. Yen, C.L. Chao, Z.Y. Li, P. Yu, H.C. Kuo, T.C. Lu, S.C. Wang, K.M. Lau and S.J. Cheng, Appl. Phys. Lett., 93, 081108 (2008).
    21. J.A. Smart, E.M. Chumbes, A.T. Schremer and J.R. Shealy, Appl. Phys. Lett., 75, 3820 (1999).
    22. A. Strittmatter, S. Rodt, L. ReiAYmann, D. Bimberg, H. Schroder, E. Obermeier, T. Riemann, J. Christen and A. Krost, Appl. Phys. Lett., 78, 727 (2001).
    23. T. Vodenitcharova, L.C. Zhang, I. Zarudi, Y. Yin, H. Domyo, T. Ho and M. Sato, J. Mater. Process. Technol., 194, 52 (2007).
    24. G. Guzman, M. Herrera, R. Silva, G.C. Vasquez and D. Maestre, Semicond. Sci. Technol., 31, 055006 (2016).
    25. T.K. Sherwood and R.O. Maak, Ind. Eng. Chem. Fundam., 1, 111 (1962).
    26. F.K. Dijen and J. Pluijmakers, J. Eur. Ceram. Soc., 5, 385 (1989).
    27. G.C. Lars, S. Felix, A-P. Frank, P. Vivien, K. Jesper, B. Thomas and N.K. Jens, Angew. Chem. Int. Ed., 50, 4601 (2011).
    28. F. Cacace and A.P. Wolf, J. Am. Chem. Soc., 87, 5301 (1965).
    29. S.B. Philip, M.W. Daniel and M. Paul, J. Am. Chem. Soc., 105, 488 (1983).
    30. K. Uchida, A. Watanabe, F. Yano, M. Kouguchi, T. Tanaka and S. Minagawa, J. Appl. Phys., 79, 3487 (1996).
    31. V. Thakur and S.M. Shivaprasad, Appl. Surf. Sci., 327, 389 (2015).