Surface Science 606 (2012) L79–L81
Contents lists available at SciVerse ScienceDirect
Surface Science journal homepage: www.elsevier.com/locate/susc
Surface Science Letters
Evidence for preferential reactivity of the atomic oxygen with hydrogenated diamond (111) facets Sh. Michaelson, R. Akhvlediani, L. Tkach, A. Hoffman ⁎ Schulich Faculty of Chemistry, Technion, Israel Institute of Technology, Haifa 32000, Israel
a r t i c l e
i n f o
Article history: Received 16 April 2012 Accepted 14 May 2012 Available online 18 May 2012 Keywords: Surface vibrational spectroscopy HR-EELS Oxygen bonding conﬁguration Hydrogen surface bonding
a b s t r a c t Chemical bonding conﬁguration of the adsorbed oxygen on diamond polycrystalline hydrogenated surface was investigated by high resolution electron energy loss spectroscopy (HR-EELS). Hot ﬁlament chemical vapor deposited diamond ﬁlms with sub-micron grain size have been exposed in-situ to thermally activated atomic oxygen (AO) and annealed in ultra-high vacuum in the range of 600–900 °C. HR-EELS features comparison of as-deposited and AO exposed diamond surface as a function of thermal annealing suggests that AO preferentially adsorbs on hydrogenated (111) facets keeping hydrogenated (100) ones intact. © 2012 Elsevier B.V. All rights reserved.
1. Introduction Oxygen plays a key role in the quality and defect density of diamond grown by chemical vapor deposition (CVD). The electronic properties of diamond surfaces are strongly affected and determined by surface impurities, in particular, by hydrogen, oxygen and water molecules [1–4]. Thus, the well-known p-type surface conductivity and negative electron afﬁnity of diamond surfaces are determined by proper hydrogen surface termination followed by exposure of the surfaces to oxygen containing species (such as H2O) [5,6]. During the last two decades, molecular and activated oxygen adsorption and desorption phenomena on bare, partially and fully hydrogenated diamond surfaces were investigated both experimentally [7–23] and theoretically [24–27]. The majority of this work was performed onto well deﬁned (111), (100) and (110) surfaces, while much less efforts were invested into more common polycrystalline diamond ﬁlms. The two main low energy diamond surface orientations forming the crystal faces of well-faceted polycrystalline CVD diamond ﬁlms are the (100) and (111) surfaces. The obvious obstacle in careful investigation of the oxygen adsorption phenomena on the surface of poly-crystalline ﬁlms is the presence of different crystallographic orientations, defects, grain boundaries etc. On the other hand, solely poly-crystalline ﬁlms can provide the difference in physicochemical reactivity, thermal stability and adsorption/desorption phenomena of different crystallographic facets, subjected exactly to the same external conditions.
⁎ Corresponding author. E-mail address: [email protected]
(A. Hoffman). 0039-6028/$ – see front matter © 2012 Elsevier B.V. All rights reserved. doi:10.1016/j.susc.2012.05.008
In the present work we study the chemical state of fully and partially hydrogenated polycrystalline diamond ﬁlm surfaces in-situ exposed to activated oxygen (AO) and systematically monitor the C, O and H bonding conﬁguration by high resolution electron energy loss spectroscopy (HR-EELS). Polycrystalline diamond ﬁlms were deposited by Hot Filament (HF) CVD in a home-made reactor, using a 1/99 methane/hydrogen ratio, pressure of 50 Torr, Rhenium ﬁlament temperature of 2000 °C, and substrate temperature of 800 °C. Scanning electronic microscopy, Raman, X-ray photoelectron and HR-EEL spectroscopies were applied to characterize the morphology, phase composition, chemical bonding and vibrational signatures of the ﬁlm's surfaces [28,29]. In the present research, samples were immediately transferred to ultra-high vacuum (UHV) chamber following the deposition and the total ambient exposure time was b5 min. Samples were annealed in-situ in the temperature range of 600–900 °C followed by oxidation. Oxidation was carried out by introduction of oxygen gas at the partial pressure of l × 10 − 5 Torr via Ir ﬁlament (heated at 1200 °C) for 30 min. The Ir ﬁlament was positioned ~5 cm from the ﬁlm's surface. Under these circumstances it is expected that thermal activation of the molecules results in ≈1% dissociation total AO exposure dose of 180 L. Annealing and oxygen exposure steps were monitored by HR-EELS using a Delta 0.5 spectrometer (VSI-SPECS) consisting of a double monochromator and a single analyzer housed in an UHV system with base pressure of 1 × 10 − 9 Torr. All spectra were obtained at room temperature in the specular geometry with an incident angle of 55° from the surface normal, incident electron energy of 4 eV and a full width half maximum (FWHM) of the elastic peak lower than 12 meV. HR-EELS data was analyzed by curve ﬁtting using the XPSpeak4.1 program.
SU RF A LE CE TT S ER CIE S NC E
S. Michaelson et al. / Surface Science 606 (2012) L79–L81
Fig. 1. (a)–(d) General HR-EEL spectra of as-deposited diamond surfaces annealed to temperatures of 600–900 °C. (e)–(h) Peak ﬁtting and detailed view of the C\H stretching region. Note the splitting for at least three different modes at 352, 361, and 377 meV, associated to hydrogen bonding to (111), (100) surfaces and (sp2) C\H, respectively. The ﬁtting peak number and their exact position were derived from best ﬁtting parameters. The inset shows typical SEM picture of poly-crystalline diamond sample, where triangle shapes correspond to (111) oriented facets, and squares correspond to (100) ones.
Fig. 1(a)–(d) shows the HR-EEL spectra recorded on the 600–900 °C annealed polycrystalline diamond ﬁlm surface. In each case UHV in-situ annealing was performed for ~1 min keeping vacuum below 5 × 10− 8 Torr, cooled to room temperature and followed by HR-EELS measurements. Spectra recorded at TA = 600–800 °C are dominated by characteristic hydrogenated diamond features described elsewhere [30,31]. Shortly, the vibrational mode at ~155 meV is a superposition of a C\C stretch and a C\H bending vibrations. The modes at ~300 and ~450 meV energy losses are pure C\C vibrations; most likely the ﬁrst and second overtones (or multiple losses) of the diamond optical phonon positioned at ~150 meV . The peak centered at ~360 meV
energy loss is attributed to C\H stretching mode, while the mode at ~510 meV is a coupling of this C\H stretch mode (~360 meV) and the 155 meV band. TA = 900 °C results in decrease of hydrogen peaks intensity alongside with appearance of C_C dimer peak at ~90 meV, which can be associated to partial desorption of hydrogen atoms from the ﬁlm surface [32–35]. Fig. 1(e)–(h) shows detailed peak ﬁtting of the C\H stretching region centered at ~360 meV. C\H stretching region is dominated by three main contributions, namely: C(111)\H centered at 352 meV [36,37], C(100)\H centered at 362 meV  and sp2 hybridized carbon\H mode centered at 375 meV . The relative intensities of these contributions change depending on the thermal history of the sample. Weakly bonded non-diamond hydrocarbon contaminations positioned on the ﬁlm surface underwent desorption at TA ≤600 °C , giving rise to increase of 362 meV contribution (its relative area increases from 28% at 600 °C to 45% at 700 °C, Fig. 1(f)). In pass from TA = 700 °C to TA = 800 °C the relative contribution of C(111)\H mode decreases from 27% to 22% and nearly disappears after TA = 900 °C, suggesting lower thermal stability of hydrogen bonded to diamond (111) facets. Thus, signiﬁcant hydrogen desorption starts at TA > 800 °C and results in the formation of highly reactive nonsaturated low hybridized carbon bonds, which increase reactivity toward oxygen atoms, discussed below. Fig. 2(a)–(d) shows the general HR-EEL spectra recorded after annealing and RT exposure of the sample surface to AO. The characteristic oxygen related peaks are those positioned at 220 meV (carbonyl, C_O) and 435 meV (hydroxyl, C\O\H stretching) [16–18]. Non-diamond hydrocarbons readily desorb at TA > 600 °C and it is expected that following desorption of the weakly bonded hydrocarbon, remaining unsaturated carbon bonds react with activated oxygen and give rise to the partial oxygen coverage. TA = 800 and 900 °C results in more vigorous hydrogen desorption (preferentially from (111) facets), associated with creation of additional chemically reactive surface states. It is suggested, that AO reacts with these states via formation of carboxyl C_O and hydroxyl C\OH bonds, evidenced by HR-EEL peaks at 220 meV and 435 meV. The presence of these features as well as decreasing the intensity of the ﬁrst diamond phonon overtone at 300 meV suggests that the ﬁlm surface is covered by non-diamond species. These results suggest that oxygen
Fig. 2. (a)–(d) General HR-EEL spectra of diamond surfaces annealed to temperatures of 600–900 °C followed by exposure to AO at room temperature. (e)–(h). Peak ﬁtting and detailed view of the C\H stretching region. (i)–(j) Comparison of the C\H stretching peak of (i) as-deposited and 700 °C annealed poly-crystalline diamond surface and (j) 700 °C annealed surface followed by in-situ AO exposure. Apparent disappearance of the (111)C–H associated mode can be observed in spectrum (j).
adsorption occurs on the isolated carbon dangling bonds produced, on partially hydrogenated surfaces, from the breaking of the π-bonding of paired dangling bonds. Fig. 2(e)–(h) shows detailed peak ﬁtting of the C\H stretching region centered at ~360 meV. Peak ﬁtting shows that oxygen exposed surface is dominated mainly by two C\H contributions, namely, (100)C\H and (sp 2)C\H, while the intensity of (111)C–H mode is strongly suppressed. In Fig. 2(i) and (j) we compare 700 °C annealed surface of as-deposited ﬁlm and followed exposure to AO. These graphs deﬁnitely show that the H terminated (100) surface is more stable under exposure to AO than the H terminated (111) surface. Peak ﬁtting procedure reveals 28% contribution of the 352 meV component ((111)C\H mode) to the total C\H peak area for the as deposited sample, while this component was completely absent following oxidation (Fig. 2(j)). Our results strongly suggest that AO preferentially reacts with the H terminated (111) surface, most probably resulting in the O/H exchange reaction, keeping hydrogenated (100) surface intact. Below we suggest possible explanation of the observed phenomenon. A bulk terminated diamond (100) surface contains two dangling bonds per surface atom which can lower their energy considerably by forming mutual bonds . Bobrov et al. showed  that molecular oxygen has been found to be easily adsorbed on the partially hydrogenated diamond C(100)–(2 × 1):H surfaces, whereas the fully hydrogenated C(100) surfaces are completely inert to molecular oxygen . Pehrsson et al.  also found that the hydrogenated (100) surfaces were largely inert to O2 at 850 b Tsub b 950 °C, but they reacted rapidly with O2 at higher temperatures. They showed that thermal oxidation of diamond with O2 requires preliminary loss of the surface hydrogen. On the other hand, (111) surface is the natural cleaving plane of diamond and shows one dangling bond per surface atom in its bulk terminated form. Loh et al.  had investigated the interaction of atomic O and D with (111)oriented diamond. Their experimental results suggest that the exchange of O with preadsorbed D (or vice versa) could proceed on the C(111) surface readily. We propose that this kind of O/H exchange reaction can proceed on the partially hydrogenated polycrystalline diamond (111) facets heated at moderated temperatures. Our present results clearly reveal the preferential stability of (100)C\H bonding and therefore oxygen adsorption is hindered by fully hydrogenated (100) surface up to elevated temperatures. Finally, it should be noted, that similar phenomenon was recently observed by us during interaction of polycrystalline diamond surface with hyper-thermal oxygen atoms of 5–7.5 eV energy . In this case we found that (111)-oriented facets become rough, whereas the (100) facets remain mostly smooth, showing evidence of etching only at their edges. 2. Summary In the present work we investigated the interaction of activated oxygen with polycrystalline hydrogenated diamond surface by HREELS. Our results strongly suggest that oxygen adsorption is strongly correlated with thermal induced hydrogen desorption: oxygen atoms react with low hybridized and highly reactive carbon atoms giving
CE N IE SC S R CE E A TT RF LE SU
S. Michaelson et al. / Surface Science 606 (2012) L79–L81
rise to high surface retention. In addition, detailed HR-EELS analysis reveals the preferential adsorption of oxygen at the H terminated (111) diamond facets, leaving H terminated (100) surface intact. Similar phenomenon was observed recently for interaction of polycrystalline diamond surface with hyperthermal oxygen atoms. This research project was carried out with the ﬁnancial support of the Israeli Academy of Science (RG 945) and the Technion Fund for promotion of research. References                                        
F. Maier, M. Riedel, B. Mantel, J. Ristein, L. Ley, Phys. Rev. Lett. 85 (2000) 3472. J. Ristein, Appl. Phys. A 82 (2006) 377. J. Ristein, Surf. Sci. 600 (2006) 3677. J. Ristein, Science 313 (2006) 1057. M. Riedel, J. Ristein, L. Ley, Phys. Rev. B 69 (2004) 125338. C.E. Nebel, Science 318 (2007) 1391. F.K. de Theije, M.F. Reedijk, J. Arsic, W.J.P. van Enckevort, E. Vlieg, Phys. Rev. B 64 (2001) 085403. Z. Shpilman, I. Gouzman, E. Grossman, R. Akhvlediani, A. Hoffman, Phys. Status Solidi A 205 (2008) 2130. M.Z. Hossain, T. Kubo, T. Aruga, N. Takagi, T. Tsuno, N. Fujimori, M. Nishijima, Surf. Sci. 436 (1999) 63. A. Laikhtman, A. Lafosse, Y. Le Coat, R. Azria, A. Hoffman, J. Chem. Phys. 119 (2003) 1794. X.F. Wang, M. Hasegawa, K. Tsugawa, A.R. Ruslinda, H. Kawarada, Diam. Relat. Mater. 24 (2012) 146. S. Ghodbane, T. Haensel, Y. Cofﬁnier, S. Szunerits, D. Steinmüller-Nethl, R. Boukherroub, S.I.U. Ahmed, J.A. Schaefer, Langmuir 26 (2010) 18798. A. Laikhtman, A. Hoffman, Surf. Sci. 522 (2003) L1. K. Bobrov, H. Shechter, A. Hoffman, M. Folman, Appl. Surf. Sci. 196 (2002) 173. K. Bobrov, G. Comtet, L. Hellner, G. Dujardin, A. Hoffman, Appl. Phys. Lett. 85 (2004) 296. Z. Shpilman, I. Gouzman, E. Grossman, R. Akhvlediani, A. Hoffman, J. Appl. Phys. 102 (2007) 114914. P.E. Pehrsson, T.W. Mercer, Surf. Sci. 460 (2000) 74. P.E. Pehrsson, T.W. Mercer, Surf. Sci. 460 (2000) 49. K.P. Loh, X.N. Xie, S.W. Yang, J.C. Zheng, J. Phys. Chem. B 106 (2002) 5230. B.L. Mackey, J.N. Russell Jr., J.E. Crowell, P.E. Pehrsson, B.D. Thoms, J.E. Butler, J. Phys. Chem. B 105 (2001) 3803. Z. Shpilman, I. Gouzman, E. Grossman, L. Shen, T.K. Minton, A. Hoffman, Appl. Phys. Lett. 95 (2009) 174106. K.P. Loh, X.N. Xie, Y.H. Lim, E.J. Teo, J.C. Zheng, T. Ando, Surf. Sci. 505 (2002) 93. P.E. Pehrsson, T.W. Mercer, J.A. Chaney, Surf. Sci. 497 (2002) 13. S. Skokov, B. Weiner, M. Frenklach, Phys. Rev. B 49 (1994) 11374. H. Tamura, H. Zhou, K. Sugisako, Y. Yokoi, S. Takami, M. Kubo, K. Teraishi, A. Miyamoto, A. Imamura, N. Mikka, Phys. Rev. B 61 (2000) 11025. X.M. Zheng, P.V. Smith, Surf. Sci. 262 (1992) 219. S. Skokov, B. Weiner, M. Frenklach, Phys. Rev. B 55 (1997) 1895. R. Akhvlediani, I. Lior, S. Michaelson, A. Hoffman, Diam. Relat. Mater. 11 (2002) 545. O. Ternyak, S. Michaelson, L. Tkach, R. Akhvlediani, A. Hoffman, Phys. Status Solidi A 204 (2007) 2839. S. Michaelson, Y. Lifshitz, A. Hoffman, Diam. Relat. Mater. 16 (2007) 855. S. Michaelson, A. Hoffman, Y. Lifshitz, Appl. Phys. Lett. 89 (2006) 223112. S. Michaelson, O. Ternyak, A. Hoffman, O.A. Williams, D.M. Gruen, Appl. Phys. Lett. 91 (2007) 103104. R.E. Thomas, R.A. Rudder, R.J. Markunas, J. Vac. Sci. Technol., A 10 (1992) 2451. C. Su, J.C. Lin, Surf. Sci. 406 (1998) 149. B.D. Thoms, P.E. Pehrsson, J.E. Butler, J. Appl. Phys. 75 (1994) 1804. T. Aizawa, T. Ando, M. Kamo, Y. Sato, Phys. Rev. B 48 (1993) 18348. T. Aizawa, T. Ando, K. Yamamoto, M. Kamo, Y. Sato, Diam. Relat. Mater. 4 (1995) 600. B.D. Thoms, J.E. Butler, Surf. Sci. 328 (1995) 291. S. Michaelson, O. Ternyak, R. Akhvlediani, A. Lafosse, M. Bertin, R. Azria, A. Hoffman, Phys. Status Solidi A 204 (2007) 2909. Z. Shpilman, I. Gouzman, E. Grossman, L. Shen, T.K. Minton, J.T. Paci, G.C. Schatz, R. Akhvlediani, A. Hoffman, J. Phys. Chem. C 114 (2010) 18996.