- Email: [email protected]
Low-temperature preparation of phosphorus doped μc-Si:H thin films by low-frequency inductively coupled plasma assisted chemical vapor deposition
Thin Solid Films 520 (2012) 1724–1728
Contents lists available at SciVerse ScienceDirect
Thin Solid Films j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / t s f
Low-temperature preparation of phosphorus doped μc-Si:H thin films by low-frequency inductively coupled plasma assisted chemical vapor deposition W.S. Yan, D.Y. Wei, Y.N. Guo, S. Xu ⁎, T.M. Ong, C.C. Sern Plasma Sources and Application Center, NIE, Nanyang Technological University, 1 Nanyang Walk, 637616, Singapore
a r t i c l e
i n f o
Article history: Received 21 February 2011 Received in revised form 6 August 2011 Accepted 9 August 2011 Available online 22 August 2011 Keywords: Solar cells Microcrystalline films Thin films Silicon Phosphorus doping Inductively-coupled plasma-assisted chemical vapor deposition Raman spectroscopy Electrical properties
a b s t r a c t The phosphorus doped n-type hydrogenated microcrystalline silicon (n-μc-Si:H) thin films are prepared, at the two low substrate temperatures of room temperature and 200 °C, through a low-frequency inductively coupled plasma assisted chemical vapor deposition. The effect of the substrate temperature on the structural properties of the thin films, such as the X-ray Diffraction (XRD) patterns and the Raman spectra, is studied. The XRD measurements show that the diffraction orientations of the thin films present an obvious change when the radio frequency power is increased from 1300 W to 2300 W. The Raman spectra of the thin films deposited at room temperature unambiguously present a phase transition from the amorphous structure to microcrystalline structure whereas no structural phase transition is observed for the thin films deposited at 200 °C. The effect of the substrate temperature on the crystalline volume fraction of the thin films presents a large difference for the radio frequency power in the range of 1300 W–1700 W, while the difference becomes small when the power is increased from 1700 W to 2300 W. The deposition rate and the radio frequency power-sheet resistance curve of the thin films deposited at room temperature are obviously different from those of the thin films prepared at 200 °C. It is attributed to the joint effect of the radio frequency power and substrate temperature on the doping concentration. The electron energy distribution function of the species in the chamber is mainly distributed in a low energy range. © 2011 Elsevier B.V. All rights reserved.
1. Introduction Hydrogenated microcrystalline silicon thin films (μc-Si:H) have been receiving amounts of attentions due to their important applications in thin film solar cells and other devices such as thin film transistors and image sensor [1–8]. In comparison with doped amorphous silicon thin films, doped μc-Si:H thin films have many advantages such as high doping efficiency, high mobility, and low absorption coefficient [1]. As is known, μc-Si:H thin films are usually considered as crystallites embedded in an amorphous tissue [9]. The investigation of the structural transition from the amorphous to microcrystalline phase is interesting [10,11]. The μc-Si:H thin films are usually fabricated by the plasma enhanced chemical vapor deposition (PECVD) and hot wire chemical vapor deposition (HWCVD) [12–17]. The fabrication of typical μc-Si:H thin films is often at high substrate temperature higher than 300 °C [16,18]. Low-temperature deposition is desirable in that low temperature causes less damage to the thin films. Furthermore, high deposition rate is important for industrial applications. However, as is well known, the deposition rate of typical PECVD is usually low.
⁎ Corresponding author. Tel./fax: + 65 67903818. E-mail address: [email protected] (S. Xu). 0040-6090/$ – see front matter © 2011 Elsevier B.V. All rights reserved. doi:10.1016/j.tsf.2011.08.048
In comparison with previous reports [9,19–23], in the present study, phosphorus doped n-μc-Si:H thin films are fabricated and studied by a low-frequency inductively coupled plasma assisted chemical vapor deposition at the low substrate temperature of room temperature and 200 °C. The effect of substrate temperature on the structural properties of the thin films such as the X-ray Diffraction (XRD) patterns and the Raman spectra is studied. The crystalline volume fraction and the deposition rate of the thin films are determined. The effect of the two substrate temperatures on the deposition rate and the sheet resistance of the thin films are investigated. A real time, in-situ plasma diagnostics of Langmuir probe is used to reveal the plasma properties of the species in the chamber such as the electron energy distribution function and the electron temperature. 2. Experimental details A low-frequency (460 kHz) inductively coupled plasma assisted chemical vapor deposition (LF-ICPCVD) is used to fabricate the n-μcSi:H thin films [24–28]. The schematic diagram of the low-frequency inductively coupled plasma assisted chemical vapor deposition is shown in Fig. 1. Through the fused silica plate, the radio frequency power is transferred into the stainless steel chamber. Thus, a discharge is generated and remained in the chamber. In the present study, the shape of the chamber is a cylinder with a height of 33 cm
W.S. Yan et al. / Thin Solid Films 520 (2012) 1724–1728
RF Generator Matching unit
RF coil
Cooling water out
Gas inlet Fused silica plate Port
To vacuum
Detector Sample
Bottom plate Bias Cooling water in
Negative bias
Fig. 1. The schematic diagram of low-frequency inductively coupled plasma assisted chemical vapor deposition.
and a radius of 16 cm. The LF-ICPCVD technique has an advantage of high-efficiency gas usage. For conventional PECVD, highly diluted hydrogen is usually requested. However, low hydrogen dilution and even no hydrogen dilution are requested for LF-ICPCVD. Prior to the n-μc-Si:H thin film deposition, the chamber is evacuated to a base pressure of 5 × 10 − 5 Pa. SiH4, PH3, and H2 are used as reaction gasses to prepare n-μc-Si:H thin films on glass substrates. In the present experiment, the flows of SiH4, PH3, and H2 are set at 3, 0.8, and 10 sccm, respectively (sccm defines cubic centimeters per minute at standard temperature and pressure). The working pressure is fixed at 2.0 Pa. The radio frequency power is increased from 1300 W to 2300 W. One of the structural properties of the n-μc-Si:H thin films, i.e. the XRD patterns, is characterized by the Siemens D5005 X-ray diffractometer using CuKαirradiation (λ = 0.15406 nm). In the present measurement, the scan type is the locked coupled mode. The operating voltage and the current are 20 kV and 5 mA. The Raman spectroscopy is determined by the Renishaw 1000 micro-Raman system using a 514.5 nm laser for excitation. The grain size is calculated from the XRD data using the well-known expression. The crystalline volume fraction is determined using the deconvolution of Raman spectrum. The surface morphology of the n-μc-Si:H thin films is characterized using the JEOL JSM-6700F field emission scanning electron microscope (SEM). In the present measurement, the operating voltage is 5.0 kV. The thickness and, hence, the deposition rate of the n-μc-Si:H thin films are determined by the cross-sectional SEM measurements. The sheet resistance and the doping concentration are measured by the Ecopia HMS-3000 Hall effect measurement system at room temperature. A real time, in-situ plasma diagnostics of Langmuir probe is used to measure the plasma properties in the chamber, in particular, the electron energy distribution function and the electron temperature. In the present study, the home-made Langmuir probe is a cylindrical probe. The conducting wire inside the probe is properly shielded to prevent radio frequency interference. The probe is powered by AC (50 Hz) variable voltage by using a variable transformer.
1725
100 nm. Fig. 2 (a) shows the radio frequency power dependence of the XRD patterns of the n-μc-Si:H thin films deposited at room temperature whereas Fig. 2 (b) corresponds to the radio frequency power dependence of the XRD patterns of the thin films prepared at 200 °C. It is seen in Fig. 2 (a) that no diffraction peaks are observed when the radio frequency power is set at 1300 W. Under this condition, the structure of the thin film is amorphous, as confirmed by the Raman measurement (see the following Raman measurement section). When the radio frequency power is increased from 1700 W to 2300 W, an obvious diffraction orientation (111) appears, which is located at around 28.1°. The obvious directions (220) and (311) appear when the radio frequency power is increased to 2300 W. The Scherrer expression d = kλ/β cos(θ) is adopted to estimate the grain size, where k is a constant determined by the geometry of the crystallites (0.89 is adopted in this paper), λ (=0.154187 nm) is the wavelength of the CuKα monochromatic X-ray radiation, β is the full width at half maximum of diffraction peaks, and θ is the diffraction angle [13]. Using the Scherrer formula, we calculate that the grain size of the thin films along (111) increases from several nanometer to 23 nm when the radio frequency power is increased from 1700 W to 2300 W. In comparison with Fig. 2 (a), it is seen from Fig. 2 (b) that high radio frequency power is helpful to promoting the intensity of the diffraction patterns. No obvious peaks are observed for the radio frequency power of 1300 W. However, it is not a pure amorphous structure, as indicated by the Raman data (see the following section on the Raman measurement). It can be found that the intensity of the diffraction patterns is still not strong when radio frequency power is increased to 2300 W. It is seen in Fig. 2 (a) and (b) that the
a)
b)
3. Results and discussion 3.1. X-ray diffraction The structural properties of the n-μc-Si:H thin films are investigated by the XRD, where the thickness of the thin films is around
Fig. 2. The radio frequency power dependence of the XRD patterns of the n-μc-Si:H thin films deposited at room temperature (a) and 200 °C (b).
1726
W.S. Yan et al. / Thin Solid Films 520 (2012) 1724–1728
preferential orientation is (111) for the radio frequency power of 1700 W–2300 W. The formation of the preferential orientation should be attributed to the lowest surface energy of the n-μc-Si:H thin films in this crystal plane.
a)
3.2. Deposition rate The effect of radio frequency power on the deposition rate of the films is shown in Fig. 3, where the thin films are prepared at room temperature and 200 °C. It is found in Fig. 3 that the deposition rate of the thin films prepared at 200 °C is obviously higher than that of the thin films deposited at room temperature, indicating that increasing substrate temperature is favorable for improving the deposition rate. Furthermore, the deposition rate increases with the increasing input power. It means that increasing radio frequency power is helpful to increasing the deposition rate. For the thin films deposited at room temperature, the deposition rate presents a large difference from that of the thin films deposited at 200 °C for the radio frequency power in the range of 1300 W–1700 W, whereas the difference is reduced when the radio frequency power is increased from 1700 W to 2300 W. It shows that the effect of substrate temperature on the deposition rate is more remarkable in the low power range of 1300 W–1700 W. It is also seen from Fig. 3 that the deposition rate increases from 5 nm/min to 27 nm/min for the thin films deposited at room temperature, while the deposition rate increases from 18 nm/min to 32 nm/min for the thin films grown at 200 °C. In the present study, the largest size of the thin films can be deposited as 9cm × 9cm, and excellent homogeneity of the thin films can be demonstrated. As is known, high deposition rate is important for industrial application. It is found from Fig. 3 that the deposition rate in the present study is impressive. The maximum deposition rate reported is around 3 nm/min of the intrinsic layer by Mukhopadyay et al. using a capacitively coupled radio frequency PECVD [29]. The maximum deposition rate is obtained and at around 36 nm/min by Jadkar et al. using a HWCVD [30].
b)
Fig. 4. The radio frequency power dependence of the Raman spectra of the thin films deposited at room temperature (a) and 200 °C (b).
3.3. Raman spectroscopy The n-μc-Si:H thin films are characterized by the Raman measurements. Fig. 4 (a) shows the Raman spectra of the n-μc-Si:H thin films deposited at room temperature, while Fig. 4 (b) shows the Raman spectra of the thin films prepared at 200 °C. It is seen in Fig. 4 (a) that a phase transition from the amorphous to microcrystalline structure is clearly observed when the radio frequency power increases from 1300 W to 2300 W. The thin films present a pure amorphous structure when the radio frequency power is 1300 W.
However, when the radio frequency power increases from 1500 W to 2300 W, the thin films present a microcrystalline structure. As is well known, μc-Si:H thin films are usually considered as crystallites embedded in an amorphous tissue. By the calculations, it can be found that the volume fraction of the microcrystallites in the matrix gradually increases when the radio frequency power increases from 1500 W to 2300 W. It will be detailedly discussed in Fig. 5. In contrast to Fig. 4 (a), it is seen in Fig. 4 (b) that there is no pure amorphous structure when the power increases from 1300 W to 2300 W. It means that improving substrate temperature has remarkably changed the structural properties of the thin films. In order to study the effect of substrate temperature on the crystalline volume fraction of the thin films, the crystalline volume fraction ϕCis calculated using following expression: [31,32] ϕC = Iμ ðcÞ = Iμ ðcÞ + Ia = ðI520 + I510 Þ = ðI520 + I510 + I480 Þ:
Fig. 3. The effect of substrate temperature on the deposition rate of the thin films, where the thin films are fabricated at room temperature and 200 °C.
ð1Þ
In Eq. (1), Iμ(c) and Ia refer to the integrated Raman scattered intensities assigned to the crystalline and amorphous phase, respectively. ϕC is calculated from the ratio of the area of the deconvoluted two peaks (I520 + I510) to the total area of the three peaks (I520 + I510 + I480). The effect of substrate temperature on the crystalline volume fraction is shown in Fig. 5, where the substrate temperatures are at room temperature and 200 °C. It is found in Fig. 5 that the crystalline volume fraction increases with increasing radio frequency power for the thin films deposited at room temperature and 200 °C. However, it is also found in Fig. 5 that the crystalline volume fraction of the thin films deposited at 200 °C is
W.S. Yan et al. / Thin Solid Films 520 (2012) 1724–1728
Fig. 5. The effect of substrate temperature on the crystalline volume fraction of the thin films fabricated at room temperature and 200 °C.
obviously higher than that of the thin films deposited at room temperature. It indicates that improving substrate temperature is helpful for increasing the crystalline volume fraction of the thin films. In addition, it is also seen that the change trend of the crystalline volume fraction with the radio frequency power is different for the thin films deposited at room temperature and 200 °C. The difference is large in the radio frequency power of 1300 W–1700 W. However, the difference becomes small when the radio frequency power is increased from 1700 W to 2300 W. It means that the substrate temperature obviously changes the structure of the thin films in the range of 1300 W–1700 W. Thus, it also affects the crystalline volume fraction of the thin films. When the radio frequency power is in the range of 1700 W–2300 W, the change trend of the crystalline volume fraction with the power is nearly same for the two curves. 3.4. Surface morphology and sheet resistance Fig. 6 shows the surface morphology of the thin films fabricated at the radio frequency power of 1900 W and the substrate temperature of 200 °C. It is clearly seen that the general grain size on the surface of the thin films is several tens nanometers under the present growth condition. But, some grain sizes are around 100 nm. The effect of substrate temperature on the sheet resistance of the n-μc-Si:H thin films is shown in Fig. 7, where the thin films are deposited at room
Fig. 6. The surface morphology of the thin films deposited at the radio frequency power of 1900 W and the substrate temperature of 200 °C.
1727
Fig. 7. The effect of substrate temperature on the sheet resistance of the thin films fabricated at room temperature and 200 °C.
temperature and 200 °C. For the thin films deposited at room temperature, it is found that the sheet resistance decreases from around 12000 ohm/square to around 40 ohm/square when the radio frequency power is increased from 1300 W to 2300 W. For the thin films deposited at 200 °C, the sheet resistance decreases from around 450 ohm/square to around 40 ohm/square when the radio frequency power is increased from 1300 W to 2300 W. However, the radio frequency power dependence of the sheet resistance of the thin films deposited at room temperature is remarkably different from that of the thin films prepared at 200 °C. The difference gradually reduces when the radio frequency power increases from 1700 W to 2300 W. It shows that the effect of substrate temperature on the sheet resistance is obviously weakened when the radio frequency power is increased from 1700 W to 2300 W. The change of the sheet resistance is mainly attributed to the joint impact of the radio frequency power and substrate temperature on the doping concentration of the thin films. On the one hand, the increasing radio frequency power can obviously improve the doping concentration and, thus, reduce the sheet resistance. On the other hand, for the radio frequency power in the range of 1300 W–1700 W, increasing substrate temperature also remarkably increases the doping concentration of the thin films.
3.4. Real-time plasma diagnostics The Langmuir probe measurement is of importance for the study on the plasma properties of the species in the chamber. In order to investigate the plasma properties in the chamber, a real time, in-situ plasma diagnostics of Langmuir probe is used to measure the electron energy distribution function (EEDF) of the species in the chamber, which can reveal the electron energy distribution and electron temperature in the chamber. Fig. 8 shows the radio frequency power dependence of the electron energy distribution function of the species measured. It is seen that the increasing radio frequency power leads to increasing electron energy distribution function. However, it is a remarkable fact that the electron energy is mainly distributed in the low electron energy range (lower than 32 eV). In contrast to the conventional capacitively coupled radio frequency plasma sources, the present low-frequency inductively coupled plasma assisted chemical vapor deposition works in the electromagnetic (H) mode. It has several unique properties such as high electron density, low plasma sheath potentials (several or tens of volts) and low electron temperature of a few eVs (shown in Fig. 8). These structural and electric properties of the thin films are closely related with the electromagnetic (H) mode of low-frequency inductively coupled plasma.
1728
W.S. Yan et al. / Thin Solid Films 520 (2012) 1724–1728
Acknowledgments This work is supported by the Singapore National Research Foundation.
References
Fig. 8. The effect of radio frequency power on the electron energy distribution function (EEDF) of the species in the chamber.
4. Conclusion The low-frequency inductively coupled plasma assisted chemical vapor deposition is used to fabricate and study the n-type hydrogenated microcrystalline silicon thin films. The n-μc-Si:H thin films are prepared in a wide radio frequency power range from 1300 W to 2300 W at low temperature of room temperature and 200 °C. The effect of substrate temperature on the structural and electric properties of the thin films is studied. The XRD measurements show that the diffraction orientations present an obvious change when the radio frequency power increases from 1300 W to 2300 W. The Raman spectra of the thin films deposited at room temperature show that a structural phase transition from a pure amorphous to microcrystalline when the radio frequency power is increased from 1300 W to 2300 W, while no phase transition is observed for the thin films deposited at 200 °C. The influence of the two substrate temperature on the crystalline volume fraction of the thin films is remarkably different for the radio frequency power at the range from 1300 W to 1700 W, while the difference of the influence becomes small when the radio frequency power increases from 1700 W to 2300 W. It is attributed to the effects of the substrate temperature and the radio frequency power on the structure of the thin films. High deposition rate is achieved in the present study. For the radio frequency power — sheet resistance curves measured, which is mainly attributed to joint impact of the radio frequency power and substrate temperature on the doping concentration. Through a Langmuir probe, it is found that the electron energy distribution function of the species in the chamber is mainly distributed in a low energy range, which is lower than 32 eV.
[1] R. Martins, I. Ferreira, E. Fortunato, Solid State Phenom. 44–46 (1995) 299. [2] D. Han, K. Wang, J.M. Owens, L. Gedvilas, B. Nelson, H. Habuchi, M. Tanaka, J. Appl. Phys. 93 (2003) 3776. [3] S.B. Concari, R.H. Buitrago, M.T. Gutiérrez, J.J. Gandía, J. Appl. Phys. 94 (2003) 2417. [4] K. Tao, D.X. Zhang, J.F. Zhao, L.S. Wang, H.K. Cai, Y. Sun, J. Non-Cryst. Solids 356 (2010) 299. [5] Saleh, N.H. Nichel, Thin Solid Films 427 (2003) 266. [6] P. Kumar, M. Kupich, D. Grunsky, B. Schroeder, Thin Solid Films 501 (2006) 260. [7] P.Q. Luo, Z.B. Zhou, Y.J. Li, S.Q. Lin, X.M. Dou, X.M. Cui, Microelectron. J. 39 (2008) 12. [8] T. Toyama, W. Yoshida, Y. Sobajima, H. Okamoto, J. Non-Cryst. Solids 354 (2008) 2204. [9] R. Martins, A. Macarico, I. Ferreira, R. Nune, A. Bicho, E. Fortunato, Thin Solid Films 303 (1997) 47. [10] O. Vetterl, F. Finger, R. Carius, P. Hapke, L. Houben, O. Kluth, A. Lambertz, A. Mück, B. Rech, H. Wagner, Sol. Energy Mater. Sol. Cells 62 (2000) 97. [11] U. Kroll, J. Meier, P. Torres, J. Pohl, A. Shah, J. Non-Cryst. Solids 68 (1998) 227. [12] Z. Li, W. Zhou, T. Lei, F. Luo, Y. Huang, Q. Cao, J. Alloys Compd. 457 (2009) 506. [13] C. Das, A. Dasgupta, S.C. Saha, S. Ray, J. Appl. Phys. 91 (2002) 9401. [14] H. Kakinuma, M. Mohri, M. Sakamoto, T. Tsuruoka, J. Appl. Phys. 70 (1991) 7374. [15] D. Das, M. Jana, A.K. Barua, J. Appl. Phys. 89 (2001) 3041. [16] J.K. Rath, H. Meiling, R.E.I. Schropp, Sol. Energy Mater. Sol. Cells 48 (1997) 269. [17] S. Klein, F. Finger, R. Carius, M. Stutzmann, J. Appl. Phys. 98 (2005) 024905. [18] D.V. Tsu, B.S. Chao, S.R. Qvshinsky, S. Guha, J. Yang, Appl. Phys. Lett. 71 (1997) 1317. [19] R. Martins, A. Macarico, M. Vieira, I. Ferreira, E. Fortunato, Philos. Mag. Part B 76 (3) (1997) 249. [20] S. Hamma, P.R. Cabarrocas, J. Appl. Phys. 81 (1997) 7282. [21] A.L. Donne, S. Binetti, G. Iseiia, B. Pichaud, M. Texier, M. Acciarri, S. Pizzini, Appl. Surf. Sci. 254 (2008) 2804. [22] A.M. Nardes, A.M.D. Andrade, F.J. Fonseca, E.A.T. Diranilme, E.A.T. Dirani, R. Muccillo, E.N.S. Muccillo, J. Mater. Sci. Mater. Electron. 14 (2003) 407. [23] S. Hamma, P.R. Cabarrocas, Sol. Energy Mater. Sol. Cells 69 (2001) 217. [24] S. Xu, S.Y. Huang, I. Levchenko, H.P. Zhou, D.Y. Wei, S.Q. Xiao, L.X. Xu, W.S. Yan, K. Ostrikov, Adv. Energy Mater. 1 (2011) 373. [25] K. Ostrikov, Z. Tsakadze, P.P. Rutkevych, J.D. Long, S. Xu, I. Denysenko, Contrib. Plasma Phys. 45 (2005) 514. [26] S. Xu, I. Levchenko, S.Y. Huang, K. Ostrikov, Appl. Phys. Lett. 95 (2009) 111505. [27] H.P. Zhou, D.Y. Wei, S. Xu, S.Q. Xiao, L.X. Xu, S.Y. Huang, Y.N. Guo, W.S. Yan, M. Xu, J. Appl. Phys. 110 (2011) 023517. [28] H.P. Zhou, L.X. Xu, S. Xu, S.Y. Huang, D.Y. Wei, S.Q. Xiao, W.S. Yan, M. Xu, J. Phys. D: Appl. Phys. 43 (2010) 505402. [29] S. Mukhopadyay, C. Das, S. Ray, J. Phys. D: Appl. Phys. 37 (2004) 1736. [30] S.R. Jadkar, J.V. Sali, M.G. Takwale, D.V. Musale, S.T. Kshirsagar, Thin Solid Films 395 (2001) 206. [31] C. Droz, E. Vallat-Sauvain, J. Bailat, L. Feitknecht, J. Meiter, A. Shah, Sol. Energy Mater. Sol. Cells 81 (2004) 61. [32] I. Torres, R. Barrio, J.D. Santos, N. González, J.J. Gandía, Thin Solid Films 518 (2010) 7019.
Contents lists available at SciVerse ScienceDirect
Thin Solid Films j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / t s f
Low-temperature preparation of phosphorus doped μc-Si:H thin films by low-frequency inductively coupled plasma assisted chemical vapor deposition W.S. Yan, D.Y. Wei, Y.N. Guo, S. Xu ⁎, T.M. Ong, C.C. Sern Plasma Sources and Application Center, NIE, Nanyang Technological University, 1 Nanyang Walk, 637616, Singapore
a r t i c l e
i n f o
Article history: Received 21 February 2011 Received in revised form 6 August 2011 Accepted 9 August 2011 Available online 22 August 2011 Keywords: Solar cells Microcrystalline films Thin films Silicon Phosphorus doping Inductively-coupled plasma-assisted chemical vapor deposition Raman spectroscopy Electrical properties
a b s t r a c t The phosphorus doped n-type hydrogenated microcrystalline silicon (n-μc-Si:H) thin films are prepared, at the two low substrate temperatures of room temperature and 200 °C, through a low-frequency inductively coupled plasma assisted chemical vapor deposition. The effect of the substrate temperature on the structural properties of the thin films, such as the X-ray Diffraction (XRD) patterns and the Raman spectra, is studied. The XRD measurements show that the diffraction orientations of the thin films present an obvious change when the radio frequency power is increased from 1300 W to 2300 W. The Raman spectra of the thin films deposited at room temperature unambiguously present a phase transition from the amorphous structure to microcrystalline structure whereas no structural phase transition is observed for the thin films deposited at 200 °C. The effect of the substrate temperature on the crystalline volume fraction of the thin films presents a large difference for the radio frequency power in the range of 1300 W–1700 W, while the difference becomes small when the power is increased from 1700 W to 2300 W. The deposition rate and the radio frequency power-sheet resistance curve of the thin films deposited at room temperature are obviously different from those of the thin films prepared at 200 °C. It is attributed to the joint effect of the radio frequency power and substrate temperature on the doping concentration. The electron energy distribution function of the species in the chamber is mainly distributed in a low energy range. © 2011 Elsevier B.V. All rights reserved.
1. Introduction Hydrogenated microcrystalline silicon thin films (μc-Si:H) have been receiving amounts of attentions due to their important applications in thin film solar cells and other devices such as thin film transistors and image sensor [1–8]. In comparison with doped amorphous silicon thin films, doped μc-Si:H thin films have many advantages such as high doping efficiency, high mobility, and low absorption coefficient [1]. As is known, μc-Si:H thin films are usually considered as crystallites embedded in an amorphous tissue [9]. The investigation of the structural transition from the amorphous to microcrystalline phase is interesting [10,11]. The μc-Si:H thin films are usually fabricated by the plasma enhanced chemical vapor deposition (PECVD) and hot wire chemical vapor deposition (HWCVD) [12–17]. The fabrication of typical μc-Si:H thin films is often at high substrate temperature higher than 300 °C [16,18]. Low-temperature deposition is desirable in that low temperature causes less damage to the thin films. Furthermore, high deposition rate is important for industrial applications. However, as is well known, the deposition rate of typical PECVD is usually low.
⁎ Corresponding author. Tel./fax: + 65 67903818. E-mail address: [email protected] (S. Xu). 0040-6090/$ – see front matter © 2011 Elsevier B.V. All rights reserved. doi:10.1016/j.tsf.2011.08.048
In comparison with previous reports [9,19–23], in the present study, phosphorus doped n-μc-Si:H thin films are fabricated and studied by a low-frequency inductively coupled plasma assisted chemical vapor deposition at the low substrate temperature of room temperature and 200 °C. The effect of substrate temperature on the structural properties of the thin films such as the X-ray Diffraction (XRD) patterns and the Raman spectra is studied. The crystalline volume fraction and the deposition rate of the thin films are determined. The effect of the two substrate temperatures on the deposition rate and the sheet resistance of the thin films are investigated. A real time, in-situ plasma diagnostics of Langmuir probe is used to reveal the plasma properties of the species in the chamber such as the electron energy distribution function and the electron temperature. 2. Experimental details A low-frequency (460 kHz) inductively coupled plasma assisted chemical vapor deposition (LF-ICPCVD) is used to fabricate the n-μcSi:H thin films [24–28]. The schematic diagram of the low-frequency inductively coupled plasma assisted chemical vapor deposition is shown in Fig. 1. Through the fused silica plate, the radio frequency power is transferred into the stainless steel chamber. Thus, a discharge is generated and remained in the chamber. In the present study, the shape of the chamber is a cylinder with a height of 33 cm
W.S. Yan et al. / Thin Solid Films 520 (2012) 1724–1728
RF Generator Matching unit
RF coil
Cooling water out
Gas inlet Fused silica plate Port
To vacuum
Detector Sample
Bottom plate Bias Cooling water in
Negative bias
Fig. 1. The schematic diagram of low-frequency inductively coupled plasma assisted chemical vapor deposition.
and a radius of 16 cm. The LF-ICPCVD technique has an advantage of high-efficiency gas usage. For conventional PECVD, highly diluted hydrogen is usually requested. However, low hydrogen dilution and even no hydrogen dilution are requested for LF-ICPCVD. Prior to the n-μc-Si:H thin film deposition, the chamber is evacuated to a base pressure of 5 × 10 − 5 Pa. SiH4, PH3, and H2 are used as reaction gasses to prepare n-μc-Si:H thin films on glass substrates. In the present experiment, the flows of SiH4, PH3, and H2 are set at 3, 0.8, and 10 sccm, respectively (sccm defines cubic centimeters per minute at standard temperature and pressure). The working pressure is fixed at 2.0 Pa. The radio frequency power is increased from 1300 W to 2300 W. One of the structural properties of the n-μc-Si:H thin films, i.e. the XRD patterns, is characterized by the Siemens D5005 X-ray diffractometer using CuKαirradiation (λ = 0.15406 nm). In the present measurement, the scan type is the locked coupled mode. The operating voltage and the current are 20 kV and 5 mA. The Raman spectroscopy is determined by the Renishaw 1000 micro-Raman system using a 514.5 nm laser for excitation. The grain size is calculated from the XRD data using the well-known expression. The crystalline volume fraction is determined using the deconvolution of Raman spectrum. The surface morphology of the n-μc-Si:H thin films is characterized using the JEOL JSM-6700F field emission scanning electron microscope (SEM). In the present measurement, the operating voltage is 5.0 kV. The thickness and, hence, the deposition rate of the n-μc-Si:H thin films are determined by the cross-sectional SEM measurements. The sheet resistance and the doping concentration are measured by the Ecopia HMS-3000 Hall effect measurement system at room temperature. A real time, in-situ plasma diagnostics of Langmuir probe is used to measure the plasma properties in the chamber, in particular, the electron energy distribution function and the electron temperature. In the present study, the home-made Langmuir probe is a cylindrical probe. The conducting wire inside the probe is properly shielded to prevent radio frequency interference. The probe is powered by AC (50 Hz) variable voltage by using a variable transformer.
1725
100 nm. Fig. 2 (a) shows the radio frequency power dependence of the XRD patterns of the n-μc-Si:H thin films deposited at room temperature whereas Fig. 2 (b) corresponds to the radio frequency power dependence of the XRD patterns of the thin films prepared at 200 °C. It is seen in Fig. 2 (a) that no diffraction peaks are observed when the radio frequency power is set at 1300 W. Under this condition, the structure of the thin film is amorphous, as confirmed by the Raman measurement (see the following Raman measurement section). When the radio frequency power is increased from 1700 W to 2300 W, an obvious diffraction orientation (111) appears, which is located at around 28.1°. The obvious directions (220) and (311) appear when the radio frequency power is increased to 2300 W. The Scherrer expression d = kλ/β cos(θ) is adopted to estimate the grain size, where k is a constant determined by the geometry of the crystallites (0.89 is adopted in this paper), λ (=0.154187 nm) is the wavelength of the CuKα monochromatic X-ray radiation, β is the full width at half maximum of diffraction peaks, and θ is the diffraction angle [13]. Using the Scherrer formula, we calculate that the grain size of the thin films along (111) increases from several nanometer to 23 nm when the radio frequency power is increased from 1700 W to 2300 W. In comparison with Fig. 2 (a), it is seen from Fig. 2 (b) that high radio frequency power is helpful to promoting the intensity of the diffraction patterns. No obvious peaks are observed for the radio frequency power of 1300 W. However, it is not a pure amorphous structure, as indicated by the Raman data (see the following section on the Raman measurement). It can be found that the intensity of the diffraction patterns is still not strong when radio frequency power is increased to 2300 W. It is seen in Fig. 2 (a) and (b) that the
a)
b)
3. Results and discussion 3.1. X-ray diffraction The structural properties of the n-μc-Si:H thin films are investigated by the XRD, where the thickness of the thin films is around
Fig. 2. The radio frequency power dependence of the XRD patterns of the n-μc-Si:H thin films deposited at room temperature (a) and 200 °C (b).
1726
W.S. Yan et al. / Thin Solid Films 520 (2012) 1724–1728
preferential orientation is (111) for the radio frequency power of 1700 W–2300 W. The formation of the preferential orientation should be attributed to the lowest surface energy of the n-μc-Si:H thin films in this crystal plane.
a)
3.2. Deposition rate The effect of radio frequency power on the deposition rate of the films is shown in Fig. 3, where the thin films are prepared at room temperature and 200 °C. It is found in Fig. 3 that the deposition rate of the thin films prepared at 200 °C is obviously higher than that of the thin films deposited at room temperature, indicating that increasing substrate temperature is favorable for improving the deposition rate. Furthermore, the deposition rate increases with the increasing input power. It means that increasing radio frequency power is helpful to increasing the deposition rate. For the thin films deposited at room temperature, the deposition rate presents a large difference from that of the thin films deposited at 200 °C for the radio frequency power in the range of 1300 W–1700 W, whereas the difference is reduced when the radio frequency power is increased from 1700 W to 2300 W. It shows that the effect of substrate temperature on the deposition rate is more remarkable in the low power range of 1300 W–1700 W. It is also seen from Fig. 3 that the deposition rate increases from 5 nm/min to 27 nm/min for the thin films deposited at room temperature, while the deposition rate increases from 18 nm/min to 32 nm/min for the thin films grown at 200 °C. In the present study, the largest size of the thin films can be deposited as 9cm × 9cm, and excellent homogeneity of the thin films can be demonstrated. As is known, high deposition rate is important for industrial application. It is found from Fig. 3 that the deposition rate in the present study is impressive. The maximum deposition rate reported is around 3 nm/min of the intrinsic layer by Mukhopadyay et al. using a capacitively coupled radio frequency PECVD [29]. The maximum deposition rate is obtained and at around 36 nm/min by Jadkar et al. using a HWCVD [30].
b)
Fig. 4. The radio frequency power dependence of the Raman spectra of the thin films deposited at room temperature (a) and 200 °C (b).
3.3. Raman spectroscopy The n-μc-Si:H thin films are characterized by the Raman measurements. Fig. 4 (a) shows the Raman spectra of the n-μc-Si:H thin films deposited at room temperature, while Fig. 4 (b) shows the Raman spectra of the thin films prepared at 200 °C. It is seen in Fig. 4 (a) that a phase transition from the amorphous to microcrystalline structure is clearly observed when the radio frequency power increases from 1300 W to 2300 W. The thin films present a pure amorphous structure when the radio frequency power is 1300 W.
However, when the radio frequency power increases from 1500 W to 2300 W, the thin films present a microcrystalline structure. As is well known, μc-Si:H thin films are usually considered as crystallites embedded in an amorphous tissue. By the calculations, it can be found that the volume fraction of the microcrystallites in the matrix gradually increases when the radio frequency power increases from 1500 W to 2300 W. It will be detailedly discussed in Fig. 5. In contrast to Fig. 4 (a), it is seen in Fig. 4 (b) that there is no pure amorphous structure when the power increases from 1300 W to 2300 W. It means that improving substrate temperature has remarkably changed the structural properties of the thin films. In order to study the effect of substrate temperature on the crystalline volume fraction of the thin films, the crystalline volume fraction ϕCis calculated using following expression: [31,32] ϕC = Iμ ðcÞ = Iμ ðcÞ + Ia = ðI520 + I510 Þ = ðI520 + I510 + I480 Þ:
Fig. 3. The effect of substrate temperature on the deposition rate of the thin films, where the thin films are fabricated at room temperature and 200 °C.
ð1Þ
In Eq. (1), Iμ(c) and Ia refer to the integrated Raman scattered intensities assigned to the crystalline and amorphous phase, respectively. ϕC is calculated from the ratio of the area of the deconvoluted two peaks (I520 + I510) to the total area of the three peaks (I520 + I510 + I480). The effect of substrate temperature on the crystalline volume fraction is shown in Fig. 5, where the substrate temperatures are at room temperature and 200 °C. It is found in Fig. 5 that the crystalline volume fraction increases with increasing radio frequency power for the thin films deposited at room temperature and 200 °C. However, it is also found in Fig. 5 that the crystalline volume fraction of the thin films deposited at 200 °C is
W.S. Yan et al. / Thin Solid Films 520 (2012) 1724–1728
Fig. 5. The effect of substrate temperature on the crystalline volume fraction of the thin films fabricated at room temperature and 200 °C.
obviously higher than that of the thin films deposited at room temperature. It indicates that improving substrate temperature is helpful for increasing the crystalline volume fraction of the thin films. In addition, it is also seen that the change trend of the crystalline volume fraction with the radio frequency power is different for the thin films deposited at room temperature and 200 °C. The difference is large in the radio frequency power of 1300 W–1700 W. However, the difference becomes small when the radio frequency power is increased from 1700 W to 2300 W. It means that the substrate temperature obviously changes the structure of the thin films in the range of 1300 W–1700 W. Thus, it also affects the crystalline volume fraction of the thin films. When the radio frequency power is in the range of 1700 W–2300 W, the change trend of the crystalline volume fraction with the power is nearly same for the two curves. 3.4. Surface morphology and sheet resistance Fig. 6 shows the surface morphology of the thin films fabricated at the radio frequency power of 1900 W and the substrate temperature of 200 °C. It is clearly seen that the general grain size on the surface of the thin films is several tens nanometers under the present growth condition. But, some grain sizes are around 100 nm. The effect of substrate temperature on the sheet resistance of the n-μc-Si:H thin films is shown in Fig. 7, where the thin films are deposited at room
Fig. 6. The surface morphology of the thin films deposited at the radio frequency power of 1900 W and the substrate temperature of 200 °C.
1727
Fig. 7. The effect of substrate temperature on the sheet resistance of the thin films fabricated at room temperature and 200 °C.
temperature and 200 °C. For the thin films deposited at room temperature, it is found that the sheet resistance decreases from around 12000 ohm/square to around 40 ohm/square when the radio frequency power is increased from 1300 W to 2300 W. For the thin films deposited at 200 °C, the sheet resistance decreases from around 450 ohm/square to around 40 ohm/square when the radio frequency power is increased from 1300 W to 2300 W. However, the radio frequency power dependence of the sheet resistance of the thin films deposited at room temperature is remarkably different from that of the thin films prepared at 200 °C. The difference gradually reduces when the radio frequency power increases from 1700 W to 2300 W. It shows that the effect of substrate temperature on the sheet resistance is obviously weakened when the radio frequency power is increased from 1700 W to 2300 W. The change of the sheet resistance is mainly attributed to the joint impact of the radio frequency power and substrate temperature on the doping concentration of the thin films. On the one hand, the increasing radio frequency power can obviously improve the doping concentration and, thus, reduce the sheet resistance. On the other hand, for the radio frequency power in the range of 1300 W–1700 W, increasing substrate temperature also remarkably increases the doping concentration of the thin films.
3.4. Real-time plasma diagnostics The Langmuir probe measurement is of importance for the study on the plasma properties of the species in the chamber. In order to investigate the plasma properties in the chamber, a real time, in-situ plasma diagnostics of Langmuir probe is used to measure the electron energy distribution function (EEDF) of the species in the chamber, which can reveal the electron energy distribution and electron temperature in the chamber. Fig. 8 shows the radio frequency power dependence of the electron energy distribution function of the species measured. It is seen that the increasing radio frequency power leads to increasing electron energy distribution function. However, it is a remarkable fact that the electron energy is mainly distributed in the low electron energy range (lower than 32 eV). In contrast to the conventional capacitively coupled radio frequency plasma sources, the present low-frequency inductively coupled plasma assisted chemical vapor deposition works in the electromagnetic (H) mode. It has several unique properties such as high electron density, low plasma sheath potentials (several or tens of volts) and low electron temperature of a few eVs (shown in Fig. 8). These structural and electric properties of the thin films are closely related with the electromagnetic (H) mode of low-frequency inductively coupled plasma.
1728
W.S. Yan et al. / Thin Solid Films 520 (2012) 1724–1728
Acknowledgments This work is supported by the Singapore National Research Foundation.
References
Fig. 8. The effect of radio frequency power on the electron energy distribution function (EEDF) of the species in the chamber.
4. Conclusion The low-frequency inductively coupled plasma assisted chemical vapor deposition is used to fabricate and study the n-type hydrogenated microcrystalline silicon thin films. The n-μc-Si:H thin films are prepared in a wide radio frequency power range from 1300 W to 2300 W at low temperature of room temperature and 200 °C. The effect of substrate temperature on the structural and electric properties of the thin films is studied. The XRD measurements show that the diffraction orientations present an obvious change when the radio frequency power increases from 1300 W to 2300 W. The Raman spectra of the thin films deposited at room temperature show that a structural phase transition from a pure amorphous to microcrystalline when the radio frequency power is increased from 1300 W to 2300 W, while no phase transition is observed for the thin films deposited at 200 °C. The influence of the two substrate temperature on the crystalline volume fraction of the thin films is remarkably different for the radio frequency power at the range from 1300 W to 1700 W, while the difference of the influence becomes small when the radio frequency power increases from 1700 W to 2300 W. It is attributed to the effects of the substrate temperature and the radio frequency power on the structure of the thin films. High deposition rate is achieved in the present study. For the radio frequency power — sheet resistance curves measured, which is mainly attributed to joint impact of the radio frequency power and substrate temperature on the doping concentration. Through a Langmuir probe, it is found that the electron energy distribution function of the species in the chamber is mainly distributed in a low energy range, which is lower than 32 eV.
[1] R. Martins, I. Ferreira, E. Fortunato, Solid State Phenom. 44–46 (1995) 299. [2] D. Han, K. Wang, J.M. Owens, L. Gedvilas, B. Nelson, H. Habuchi, M. Tanaka, J. Appl. Phys. 93 (2003) 3776. [3] S.B. Concari, R.H. Buitrago, M.T. Gutiérrez, J.J. Gandía, J. Appl. Phys. 94 (2003) 2417. [4] K. Tao, D.X. Zhang, J.F. Zhao, L.S. Wang, H.K. Cai, Y. Sun, J. Non-Cryst. Solids 356 (2010) 299. [5] Saleh, N.H. Nichel, Thin Solid Films 427 (2003) 266. [6] P. Kumar, M. Kupich, D. Grunsky, B. Schroeder, Thin Solid Films 501 (2006) 260. [7] P.Q. Luo, Z.B. Zhou, Y.J. Li, S.Q. Lin, X.M. Dou, X.M. Cui, Microelectron. J. 39 (2008) 12. [8] T. Toyama, W. Yoshida, Y. Sobajima, H. Okamoto, J. Non-Cryst. Solids 354 (2008) 2204. [9] R. Martins, A. Macarico, I. Ferreira, R. Nune, A. Bicho, E. Fortunato, Thin Solid Films 303 (1997) 47. [10] O. Vetterl, F. Finger, R. Carius, P. Hapke, L. Houben, O. Kluth, A. Lambertz, A. Mück, B. Rech, H. Wagner, Sol. Energy Mater. Sol. Cells 62 (2000) 97. [11] U. Kroll, J. Meier, P. Torres, J. Pohl, A. Shah, J. Non-Cryst. Solids 68 (1998) 227. [12] Z. Li, W. Zhou, T. Lei, F. Luo, Y. Huang, Q. Cao, J. Alloys Compd. 457 (2009) 506. [13] C. Das, A. Dasgupta, S.C. Saha, S. Ray, J. Appl. Phys. 91 (2002) 9401. [14] H. Kakinuma, M. Mohri, M. Sakamoto, T. Tsuruoka, J. Appl. Phys. 70 (1991) 7374. [15] D. Das, M. Jana, A.K. Barua, J. Appl. Phys. 89 (2001) 3041. [16] J.K. Rath, H. Meiling, R.E.I. Schropp, Sol. Energy Mater. Sol. Cells 48 (1997) 269. [17] S. Klein, F. Finger, R. Carius, M. Stutzmann, J. Appl. Phys. 98 (2005) 024905. [18] D.V. Tsu, B.S. Chao, S.R. Qvshinsky, S. Guha, J. Yang, Appl. Phys. Lett. 71 (1997) 1317. [19] R. Martins, A. Macarico, M. Vieira, I. Ferreira, E. Fortunato, Philos. Mag. Part B 76 (3) (1997) 249. [20] S. Hamma, P.R. Cabarrocas, J. Appl. Phys. 81 (1997) 7282. [21] A.L. Donne, S. Binetti, G. Iseiia, B. Pichaud, M. Texier, M. Acciarri, S. Pizzini, Appl. Surf. Sci. 254 (2008) 2804. [22] A.M. Nardes, A.M.D. Andrade, F.J. Fonseca, E.A.T. Diranilme, E.A.T. Dirani, R. Muccillo, E.N.S. Muccillo, J. Mater. Sci. Mater. Electron. 14 (2003) 407. [23] S. Hamma, P.R. Cabarrocas, Sol. Energy Mater. Sol. Cells 69 (2001) 217. [24] S. Xu, S.Y. Huang, I. Levchenko, H.P. Zhou, D.Y. Wei, S.Q. Xiao, L.X. Xu, W.S. Yan, K. Ostrikov, Adv. Energy Mater. 1 (2011) 373. [25] K. Ostrikov, Z. Tsakadze, P.P. Rutkevych, J.D. Long, S. Xu, I. Denysenko, Contrib. Plasma Phys. 45 (2005) 514. [26] S. Xu, I. Levchenko, S.Y. Huang, K. Ostrikov, Appl. Phys. Lett. 95 (2009) 111505. [27] H.P. Zhou, D.Y. Wei, S. Xu, S.Q. Xiao, L.X. Xu, S.Y. Huang, Y.N. Guo, W.S. Yan, M. Xu, J. Appl. Phys. 110 (2011) 023517. [28] H.P. Zhou, L.X. Xu, S. Xu, S.Y. Huang, D.Y. Wei, S.Q. Xiao, W.S. Yan, M. Xu, J. Phys. D: Appl. Phys. 43 (2010) 505402. [29] S. Mukhopadyay, C. Das, S. Ray, J. Phys. D: Appl. Phys. 37 (2004) 1736. [30] S.R. Jadkar, J.V. Sali, M.G. Takwale, D.V. Musale, S.T. Kshirsagar, Thin Solid Films 395 (2001) 206. [31] C. Droz, E. Vallat-Sauvain, J. Bailat, L. Feitknecht, J. Meiter, A. Shah, Sol. Energy Mater. Sol. Cells 81 (2004) 61. [32] I. Torres, R. Barrio, J.D. Santos, N. González, J.J. Gandía, Thin Solid Films 518 (2010) 7019.