Visible luminescence from nanocrystalline silicon films produced by plasma enhanced chemical vapor deposition Erik Edelberg, Sam Bergh, Ryan Naone, Michael Hall, and Eray S. Aydil Citation: Applied Physics Letters 68, 1415 (1996); doi: 10.1063/1.116098 View online: http://dx.doi.org/10.1063/1.116098 View Table of Contents: http://scitation.aip.org/content/aip/journal/apl/68/10?ver=pdfcov Published by the AIP Publishing Articles you may be interested in High-mobility nanocrystalline silicon thin-film transistors fabricated by plasma-enhanced chemical vapor deposition Appl. Phys. Lett. 86, 222106 (2005); 10.1063/1.1942641 Low temperature deposition of nanocrystalline silicon carbide films by plasma enhanced chemical vapor deposition and their structural and optical characterization J. Appl. Phys. 94, 5252 (2003); 10.1063/1.1609631 Effects of visible light illumination during plasma enhanced chemical vapor deposition growth on the film properties of hydrogenated amorphous silicon J. Appl. Phys. 91, 840 (2002); 10.1063/1.1421242 Visible electroluminescence from nanocrystallites of silicon films prepared by plasma enhanced chemical vapor deposition Appl. Phys. Lett. 69, 596 (1996); 10.1063/1.117918 Small‐scale high‐strength silicon carbide fibers fabricated from thin films produced by plasma‐enhanced chemical vapor deposition J. Vac. Sci. Technol. A 8, 1422 (1990); 10.1116/1.576850
This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP: 134.84.192.103 On: Thu, 23 Jul 2015 12:43:33
Visible luminescence from nanocrystalline silicon films produced by plasma enhanced chemical vapor deposition Erik Edelberg Department of Chemical Engineering, University of California Santa Barbara, Santa Barbara, California 93106
Sam Bergh Department of Electrical and Computer Engineering, University of California Santa Barbara, Santa Barbara, California 93106
Ryan Naone Materials Department, University of California Santa Barbara, Santa Barbara, California 93106
Michael Hall and Eray S. Aydila) Department of Chemical Engineering, University of California Santa Barbara, Santa Barbara, California 93106
~Received 6 December 1995; accepted for publication 2 January 1996! Thin nanocrystalline silicon ~nc-Si! films deposited by plasma enhanced chemical vapor deposition ~PECVD! exhibited room-temperature photoluminescence in the visible range of the electromagnetic spectrum. High resolution transmission electron microscopy revealed that the films are made of Si crystals with dimensions 2–15 nm. The photoluminescence spectra of the nc-Si films were similar to the spectra observed from porous silicon produced by anodization and electrochemical dissolution of crystalline Si. This similarity suggests that the luminescence mechanism of nc-Si films is similar to the mechanism of light emission from porous silicon. The ability to manufacture luminescent Si films by methods which are compatible with the current Si based technology, such as PECVD, can provide new possibilities in the realization of optoelectronic devices. © 1996 American Institute of Physics. @S0003-6951~96!03510-9#
The ability to manufacture light emitting silicon films by methods that are compatible with the current silicon based microelectronics and very large scale integrated circuit technology would provide new possibilities in the realization of optoelectronic devices. Research into light emitting forms of Si was rekindled by the discovery of porous silicon. Since its discovery in 1990 by Canham,1 porous silicon and light emission from porous silicon have been the topic of many investigations.2 Porous silicon is formed by electrochemical dissolution of crystalline silicon wafers which leaves behind an interconnected network of long and narrow pores on the wafer surface.1–3 Porous silicon fabricated in this manner exhibits strong and visible photoluminescence even at room temperature.1–7 The mechanism of light emission from porous silicon is not fully understood but several hypotheses have been put forth.2 One hypothesis is that quantum confinement of charge carriers in narrow crystalline silicon walls separating the pores is responsible for luminescence.8 Other explanations invoke the existence of luminescent surface species adsorbed on the inner pore walls as the source of light emission.9–11 These hypotheses differ in the origin of the luminescence but are all based on the unique microstructure of porous silicon which is characterized by high porosity and a crystalline skeleton with typical crystal dimensions on the order of a few nanometers. Many variations of the electrochemical dissolution method for producing porous silicon have been attempted but all variations were based on solution based ~wet! synthesis methods. A dry production method for luminescent Si a!
Author to whom correspondence should be addressed.
would be ideal for integrating this material into the existing very large scale integrated ~VLSI! circuit manufacturing technology. The ability to manufacture luminescent Si films by methods that are compatible with the mature Si microelectronics technology can provide new possibilities in optoelectronics and allow for monolithic integration of Si technology with optical signal processing. In this letter, we report on observation of room-temperature visible photoluminescence from thin nanocrystalline Si films produced by plasma enhanced chemical vapor deposition. Luminescence has been previously observed from ultrafine Si particles synthesized by homogeneous nucleation of reactive SiH4 fragments in a SiH4 /H2 plasma or by magnetron sputtering.12,13 Nanocrystalline films, intended for applications in photovoltaics, have also been deposited previously.14 –17 However, to our knowledge, this letter is the first report of visible photoluminescence from thin nc-Si films produced by plasma enhanced chemical vapor deposition. Nanocrystalline silicon films were deposited using a helical resonator plasma enhanced chemical vapor deposition reactor. The apparatus has been previously described in detail.18,19 The reactor is a six-way stainless steel cross pumped by a turbomolecular pump. A plasma source and a substrate platen are attached to two opposite ports of the cross. The gas discharge is maintained in a 5 cm diam Pyrex tube using a helical resonator plasma source. Substrates are positioned below the plasma tube on a substrate platen and maintained at 230 °C during the deposition. Argon and H2 gases are introduced directly into the plasma while SiH4 diluted in argon carrier gas ~0.93%! is fed '1 cm above the substrate platen through a gas dispersion ring. The plasma
This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP: Appl. Phys. Lett. 68 (10), 4 March 1996 0003-6951/96/68(10)/1415/3/$10.00 © 1996 American Institute of Physics 1415 134.84.192.103 On: Thu, 23 Jul 2015 12:43:33
TABLE I. Plasma operating conditions during PECVD of luminescent nc-Si films. Parameter
Value
SiH4 gas flow rate H2 gas flow rate Total Ar gas flow rate Pressure rf power Substrate temperature Deposition rate
0.45 sccm 20 sccm 55 sccm 25 mTorr 100 W
[email protected] MHz! 230 °C 15 Å/min
operating conditions during the nc-Si film deposition are listed in Table I. Typically, 0.2–0.5 mm thick films were deposited on c-Si and characterized using Raman spectroscopy, transmission infrared spectroscopy, transmission electron microscopy, and spectroscopic ellipsometry. The films were also examined for photoluminescence using the 458 nm line excitation of an Ar ion laser. The continuous wave laser output was modulated using a chopper to enable lock-in detection. After passing through a filter to block the laser light, luminescence was detected using a 0.25 m monochromator and a photomultiplier tube. The photoluminescence ~PL! spectra of a 3800 Å thick nc-Si film on a c-Si substrate is displayed in Fig. 1; both room temperature and 27 K spectra are shown. The light emission at 27 K could be seen by the naked eye through a filter. We could not detect any luminescence from a silicon wafer without the nc-Si film. Since a-Si:H exhibits an optical gap at '1.7 eV we also attempted to detect luminescence from an a-Si:H that is of similar thickness ~'0.4 mm!. However, luminescence from the a-Si:H film was ;0 in the range and scale shown in Fig. 1. This proves that the luminescence displayed in Fig. 1 is originating from the nc-Si film and not from the substrate or any amorphous material that may exist between the crystals or at the c-Si/nc-Si film interface. The spectra recorded at 27 K exhibits two broad peaks centered at 530 nm ~2.34 eV! and 680 nm ~1.82 eV!. Both of these peaks are at energies well above the band gap ~'1.2 eV at 27 K! energy for c-Si which has an indirect band gap and is therefore not expected to luminesce in the visible range. The shape of the photoluminescence spectrum shown
FIG. 1. Photoluminescence ~PL! spectra of a 0.38 mm thick nc-Si film deposited under the conditions listed in Table I.
FIG. 2. High resolution transmission electron micrograph of the nc-Si film whose PL spectra are shown in Fig. 1.
in Fig. 1 and the location of the peaks is very similar to the PL spectra of porous Si recently published by Zhao et al. who also observed two peaks at 530 and 680 nm. The PL spectra are also similar to the PL spectra collected from porous silicon samples produced and examined by a number of other investigators.4,20–23 This similarity in PL spectra suggests that the mechanism of light emission from nc-Si films and porous silicon is the same. High resolution cross-section transmission electron micrograph of the nc-Si film ~Fig. 2! shows a distribution of crystal sizes. The Si crystals are small with dimensions ranging from 2 to 15 nm. Using spectroscopic ellipsometry we found that the films can contain at the most 20%–30% voids, a much smaller fraction than porous silicon. In addition, we could not see any interconnected pores in the TEM images such as those shown in Ref. 3. Thus, the PECVD nc-Si films are structurally different than porous silicon. It is unlikely that the nc-Si films contain enough pore and surface area to have a significant density of luminescent species such as silicon hydrides and siloxanes adsorbed on the inner surfaces of the pores. Transmission infrared spectra of the films showed less than 5% SiH incorporated into the growing film. The magnitude of absorption by SiH2 was even smaller. Although Si–O stretches were detected in the infrared spectra, the magnitude of the SiO band was consistent with a thin native oxide on the surface. These results will be discussed in detail in a future publication. Based on these results we discounted the possibility that the luminescence is originating from polysilanes @ (SiH2 ) n # or siloxene derivatives as proposed by some investigators.2 One structural characteristic which the porous silicon and nc-Si films have in common is the existence of small crystalline Si regions. Quantum confinement effects in the
This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP: 1416 Appl. Phys. Lett., Vol. 68, No. 10, 4 March 1996 Edelberg et al. 134.84.192.103 On: Thu, 23 Jul 2015 12:43:33
crystalline walls separating the pores in porous silicon have been suggested as the origin of luminescence.1,2,8 There have been many experiments and theoretical studies which show that if made sufficiently small, crystalline silicon can exhibit visible luminescence at room temperature.8,12,13,24 For example, Furukawa and Miyasato12 observed visible photoluminescence from ultrafine silicon particles that are 2–5 nm in diameter. Most recently Komoda et al.25 discovered photoluminescence at room temperature from microcrystalline Si precipitates in SiO2 formed by ion implantation. They showed that the luminescence was originating from Si crystals, with diameters in the range 2–15 nm, embedded in SiO2 . Similarly, the dimensions of the silicon crystals in our PECVD thin films are small enough that quantum confinement effects may be responsible for their luminescence. Indeed, employing a simple electron-hole pair confinement model developed by Brus,24 we calculate that the emission peaking at 680 nm corresponds to the lowest energy of an electron-hole pair confined in a spherical Si crystal with 4.6 nm diam, a value consistent with the crystal dimensions observed in Fig. 2. Since there is a distribution of crystal sizes and shapes, the bandwidth of the luminescence spectra shown in Fig. 1 is expected and attributed to confinement in different size crystallites. In summary, luminescent nanocrystalline silicon films have been deposited using plasma enhanced chemical vapor deposition through SiH4 diluted heavily in H2 and Ar. The similarity between the photoluminescence spectra from nc-Si films deposited by PECVD and porous silicon suggests that these two materials have a similar feature which may be responsible for the mechanism of light emission. Our results support the hypothesis that this common feature is the small crystal size in both porous silicon and nc-Si deposited by PECVD. This work was funded by the National Science Foundation National Young Investigator Award ~ECS 9457758! and
made use of the Materials Research Laboratory central facilities under NSF Award DMR-9123048. M.H. was partially supported by a Camille and Henry Dreyfus Foundation Dreyfus-Boissevain undergraduate scholarship. L. T. Canham, Appl. Phys. Lett. 57, 1046 ~1990!. D. J. Lockwood, Solid State Commun. 92, 101 ~1994!. 3 N. Sato, K. Sakaguchi, K. Yamagata, Y. Fujiyama, and T. Yonehara, J. Electrochem. Soc. 142, 3116 ~1995!. 4 C. S. Chang and J. T. Lue, Thin Solid Films 259, 275 ~1995!. 5 X. Zhao, O. Schoenfeld, Y. Aoyagi, and T. Sugano, Appl. Phys. Lett. 65, 1290 ~1994!. 6 Y. M. Weng, Zh. N. Fan, and X. F. Zong, Appl. Phys. Lett. 63, 168 ~1993!. 7 S. D. Russell, W. B. Dubbelday, P. Georgief, R. L. Shimabukuro, and P. R. de la Houssaye, J. Appl. Phys. 76, 6012 ~1994!. 8 T. Takagahara and K. Takeda, Phys. Rev. B 46, 15578 ~1992!. 9 M. Stutzmann, M. S. Brandt, E. Bustarret, H. D. Fuchs, M. Rosenbauer, A. Hopner, and J. Weber, J. Non-Cryst. Solids 164–166, 931 ~1993!. 10 P. Deak, M. Rosenbauer, M. Stutzmann, J. Weber, and M. S. Brandt, Phys. Rev. Lett. 69, 2531 ~1992!. 11 S, M. Prokes, O. J. Glembocki, V. M. Bermudez, R. Kaplan, L. E. Friedersdorf, and P. C. Searson, Phys. Rev. B 45, 13788 ~1992!. 12 S. Furukawa and T. Miyasato, Jpn. J. Appl. Phys. 27, L2207 ~1988!. 13 H. Takagi, H. Ogawa, Y. Yamazaki, A. Ishizaki, and T. Nakagiri, Appl. Phys. Lett. 56, 2379 ~1990!. 14 I. Solomon, B. Drevillon, H. Shirai, and N. Layadi, J. Non-Cryst. Solids 164–166, 989 ~1993!. 15 J. J. Boland and G. N. Parsons, Science 256, 1304 ~1992!. 16 M. Otobe and S. Oda, J. Non-Cryst. Solids 164–166, 993 ~1993!. 17 H. V. Nguyen, I. An, R. W. Collins, Y. Lu, M. Wakagi, and C. R. Wronski, Appl. Phys. Lett. 65, 3335 ~1994!. 18 S. C. Deshmukh and E. S. Aydil, J. Vac. Sci. Technol. A 13, 2355 ~1995!. 19 S. M. Han and E. S. Aydil, J. Vac. Sci. Technol. ~to be published!. 20 R. Tsu, H. Shen, and M. Dutta, Appl. Phys. Lett. 60, 112 ~1992!. 21 J. Zeman, M. Zigone, G. L. J. A. Rikken, and G. Martinez, Solid State Commun. 96, 503 ~1995!. 22 Y. M. Weng, Zh. N. Fan, and X. F. Zong, Appl. Phys. Lett. 78, 3520 ~1995!. 23 N. Kuroda, Y. Matsuda, S. Nakajima, and I. Taketsu, J. Appl. Phys. 78, 3520 ~1995!. 24 L. E. Brus, J. Chem. Phys. 80, 4403 ~1984!. 25 T. Komoda, J. Kelly, F. Cristiano, A. Nejim, P. L. F. Hemment, K. P. Homewood, R. Gwilliam, J. E. Mynard, and B. J. Seally, Nucl. Instrum. Methods B 96, 387 ~1995!. 1 2
This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP: Appl. Phys. Lett., Vol. 68, No. 10, 4 March 1996 Edelberg et al. 1417 134.84.192.103 On: Thu, 23 Jul 2015 12:43:33
