International Journal of Spectroscopy

International Journal of Spectroscopy / 2012 / Article
Special Issue

Spectroscopic and Structural Insight into the Size-Dependent Behavior of the Nanophase

View this Special Issue

Research Article | Open Access

Volume 2012 |Article ID 281851 | 5 pages |

Vibrational Spectroscopy of Chemical Species in Silicon and Silicon-Rich Nitride Thin Films

Academic Editor: Alexander Milekhin
Received24 Mar 2011
Revised06 Jun 2011
Accepted20 Jun 2011
Published02 Oct 2011


Vibrational properties of hydrogenated silicon-rich nitride ( ) of various stoichiometry ( ) and hydrogenated amorphous silicon ( ) films were studied using Raman spectroscopy and Fourier transform infrared spectroscopy. Furnace annealing during 5 hours in Ar ambient at and pulse laser annealing were applied to modify the structure of films. Surprisingly, after annealing with such high-thermal budget, according to the FTIR data, the nearly stoichiometric silicon nitride film contains hydrogen in the form of Si–H bonds. From analysis of the FTIR data of the Si–N bond vibrations, one can conclude that silicon nitride is partly crystallized. According to the Raman data films with hydrogen concentration 15% and lower contain mainly Si–H chemical species, and films with hydrogen concentration 30–35% contain mainly Si–H2 chemical species. Nanosecond pulse laser treatments lead to crystallization of the films and its dehydrogenization.

1. Introduction

The interest in amorphous, nano- and microcrystalline silicon films on nonrefractory inexpensive substrates and their crystallization is stimulated by demands of giant microelectronics. For example, the enlargement of sizes of flat-panel displays with active thin film transistor matrix can be described as “reverse Moor’s law.” Scientific interest in the silicon-rich nitride (SRN) films is growing due to their perspectives for application. The most important applications of SRN films are silicon-based optoelectronic devices [1], nonvolatile memory (flash-memory) devices [2], and new tandem solar cells based on silicon and silicon oxide (silicon nitride) heterostructures [3]. Silicon nanoclusters and SRN films can be applied in novel quantum-dots-based solar cells with enhanced efficiency [4].

Modern plasma-enhanced chemical vapor deposition (PECVD) technique allows the production of amorphous silicon films and SRN films with temperature of substrate lower than 100°C [5]. As reagents containing hydrogen are mainly used in PECVD, as-deposited films are hydrogenated. Hydrogen plays important role in electrical and optical properties of the films and also has influence on stability of the films. So, it is important to develop express, nondestructive spectroscopic approaches for studies of hydrogen in amorphous hydrogenated silicon (a-Si:H) films and hydrogenated SRN films. The present paper is devoted to studies of as-deposited films as well as furnace- and pulse-laser-treated films with the use of Raman spectroscopy and Fourier Transform Infrared (FTIR) spectroscopy.

2. Experiment Details

Hydrogenated SRN films (SiNx:H) of different stoichiometry were deposited on the Si (100) substrates with the use of low-frequency PECVD method. The frequency of discharge was 55 kHz, and the films were deposited from the gas mixture of ammonia (NH3) and monosilane (SiH4). The stoichiometry parameter depends on the ratio of ammonia/monosilane flows. The growth technology is described elsewhere in more details [6]. All SRN films were deposited at substrate temperature 100°C, the reactor pressure during deposition was 150 Pa, the deposition time was 16 minutes, and the plasma power was 400 W. Parameters of the films are shown in Table 1.

Number of samplesFlow ratio NH3/SiH4Thickness, nmGrowth temperature, °CStoichiometric parameter


Amorphous silicon films were deposited on glass substrates using two methods. The first method is evaporation of monocrystalline silicon target by electron beam in high vacuum and deposition onto unheated substrate. The films deposited by this method do not contain hydrogen. The second method is PECVD at various temperatures using different PECVD reactors. The parameters of the films are shown in Table 2. It is known that optical gap of a-Si:H depends on hydrogen concentration [7]. According to optical gap measurements from light transmission and reflection data, the films grown at temperature 280°C contain 10 atomic % of hydrogen (optical gap  eV); the films grown at temperature 225°C contain 15% of hydrogen (optical gap  eV); the films grown at temperature 200°C contain 30–35% of hydrogen (optical gap  eV).

of samples
Thickness, nmGrowth temperature, °CHydrogen concentration, atomic per cents


Laser ellipsometry ( nm) was applied to measure the thickness of the films. To study the states of hydrogen and vibration of the Si–N bonds, the FTIR spectroscopy was applied. The FTIR spectrometer FT-801, produced by scientific enterprise “SIMEX” (Novosibirsk, Russia), was used. The spectral range of the spectrometer is from 550 to 5500 cm−1, and the spectral resolution can be optionally changed from 4 to 0.5 cm−1. The structural properties of as-deposited and annealed films were studied using Raman spectroscopy technique. All Raman spectra were recorded in the back-scattering geometry. The 514.5 nm Ar+ laser line was used as a light source. Triple spectrometer T64000 (Horiba Jobin Yvon) with micro-Raman setup and liquid-nitrogen-cooled CCD matrix detector was used. The spectral resolution was not worse than 1 cm−1. All spectra were measured at room temperature. To avoid heating of the films, the laser beam was slightly defocused; the spot diameter was equal to about 4–6 μm; the laser power reaching the sample was within 2-3 mW. The spectra were registered in the polarization geometry. Furnace annealing of the samples in Ar ambient for 5 hours at temperature 1130°C was performed at the Institute of Electron Technology (Warsaw, Poland). The triple harmonic of YAG:Nd laser (  nm, pulse duration of 10 ns) was used for laser treatments of the as-deposited films.

3. Results and Discussion

In Figure 1 one can see Raman spectroscopy and FTIR data concerning N–H valence (stretching) vibrations in sample 611 (stoichiometry is estimated as SiN1.3:H). For convenience, the FTIR data was multiplied by 20 times. According to both Raman data and FTIR data, the position of N–H valence vibration peak is 3350 cm−1. An advantage of Raman spectroscopy method is the possibility to use substrates that are not transparent in IR region. When the substrate is not transparent, only reflection setup can be used. Using multiple internal transmission, the sensitivity of IR spectroscopy can be notably improved [8]. Figure 2 shows shift of the position of N–H valence vibration peak versus NH3 to SiH4 ratio during deposition of SiNx:H films. With changing of stoichiometry parameter (1.3, 0.95, and 0.6) the frequency also changes (3350, 3342, and 3335 cm−1) accordingly. The frequency of Si–H valence vibrations changes from 2170 to 2100 cm−1 with decrease of stoichiometry parameter from 1.3 to 0.6 [6]. This is because of influence of neighboring atoms on frequencies of valence modes of Si–H and N–H bonds. It should be noted that Ali with coworkers [9] have also observed the shift of the position of N–H valence vibration peak versus stoichiometry of SiNx:H films.

In Figure 3 one can see Raman spectroscopy and FTIR data concerning Si–H valence vibrations for SiN1.3:H film. According to the FTIR data the position of Si–H valence vibration peak is 2170 cm−1, but according to the Raman spectroscopy data the position of peak is 2187 cm−1. A mercury lamp was used for back lighting of LCD monitor of experimental setup, and parasitic light can reach input slit of spectrometer. So, one can see in the Raman spectra a doublet (leakage from mercury lamp). One can see also a peak due to Raman scattering on N2 molecules in atmosphere. Both Raman and FTIR peaks are asymmetric, so one can assume presence of Si–H and SiH2 bonds in this film. From fitting of FTIR spectrum by two Gaussians one can assume that frequency of Si–H valence vibration is 2170 cm−1 and frequency of Si–H2 valence vibration is 2260 cm−1. The difference in positions of the Raman scattering and the FTIR peaks may be due to different contribution of Si–H and SiH2 bond valence vibration in dipole moment and polarisability.

Knowing the thickness of a film and effective absorption cross-sections for Si–N, Si–H, and N–H bonds, one can calculate the concentration of chemical bonds. The effective absorption cross-section eff for valence vibration of Si–N bonds is  cm2 [10], and eff for valence vibration of Si–H bonds is  cm2 [11]. According to different data, eff for valence vibration of N–H bonds is  cm2 [11] or cm2 [12]. Concentration of chemical species can be calculated as , where A is the height of absorbance peak (dimensionless parameter) and is thickness. The results are shown in Table 3. The concentration of N–H species is calculated using eff according to Lanford [11] and Stein [12]. It should be noted that Stein and Wegener [12] have used multiple internal transmissions but Lanford and Rand [11] used normal transmission geometry. We also have used normal transmission geometry. So, the eff obtained by Lanford is more preferable for analysis of our data. From Table 3 one can see that total concentration of Si–N, Si–H, and N–H bonds decreases with decreasing of stoichiometry parameter . This is because vibrations of Si–Si bonds do not change dipole momentum and are not active in FTIR. But they are active in Raman scattering. And Raman signal from amorphous silicon clusters was detected for sample number 615 [6]. The total concentration of Si–N bonds in stoichiometric Si3N4 is about 15.5·1022 cm−3 [13]. So, one can assume that annealed near stoichiometric sample number 611 (Table 3) contains not only Si–N and Si–H bonds but also some concentration of Si–Si bonds. But concentration of such bonds is not high enough to be detected by Raman spectroscopy.

Number of samplesSi–NSi–HN–H (for eff according to Lanford) 1022 cm−3N–H (for eff according to Stein) 1022 cm−3
1022 cm−31022 cm−3

611 annealed10.56bellow 0.1bellow 0.06

The FTIR spectra of as-deposited and annealed SiN1.3:H films are shown in Figure 4 in range from 500 to 3900 cm−1. Surprisingly, this nearly stoichiometric silicon nitride film still contains hydrogen after annealing with high-thermal budget, mainly in the form of the Si–H bonds. In this film, the peak, corresponding to absorption on stretching vibrations of the Si–H bonds, becomes even higher. The peak at 1175 cm−1 in spectrum of as-deposited film can be related to absorbance by bending vibrations of the N–H bonds [14]. Peculiarities at 850–900 cm−1 in spectrum of annealed film can be related to absorbance by stretching vibrations of the Si–N bonds in the α-modification of crystalline silicon nitride [15, 16]. Maybe partial crystallization of silicon nitride retards the dehydrogenization of the film. It has been detected earlier [17] that the closer is the stoichiometry parameter of the SRN films to that of stoichiometric Si3N4, the higher temperature is needed for effusion of hydrogen (dehydrogenization) from these films. However, short-time annealing at temperature 850°C [17] and annealing at 800°C for 2 hours [18] lead to total dehydrogenization of the SRN films, with the stoichiometry parameter below 1. So, as it was abovementioned, one can assume that, for the case of nearly stoichiometric silicon nitride film, furnace annealing leads to so-called hardening of silicon nitride inhibiting hydrogen outdiffusion.

Figure 5 shows Raman spectra of amorphous silicon films with different hydrogen concentrations. All films were deposited on glass substrates that are not transparent in IR region, so it was impossible to analyze them using FTIR in normal geometry. One can see peaks due to scattering on valence vibrations of Si–H bonds. In Raman spectrum of the only film deposited in vacuum (without hydrogen) this peak is absent. As one can see from Raman spectra, film deposited at temperature 200°C contains mostly Si–H2 bonds, because the position of Si–H2 peak is about 2100 cm−1. The frequency of Si–H peak is about 2000 cm−1 [19]. The films deposited at temperatures higher than 200°C contain mostly Si–H bonds. It should be noted that frequencies of stretch vibrations of Si–H bonds in silicon nitride and in a-Si:H are notably different: about 2200 cm−1 for the first case and 2000–2100 cm−1 for the second case. As it was shown by Hien et al. [20], the frequency of stretch vibrations of Si–H bonds in the case of neighbouring N and H atoms can reach 2122 cm−1. In the case of only Si neighbouring atoms this frequency is 2012 cm−1 [20]. FTIR method can be used for quantitative hydrogen measurements [21], but for transmission experiments the substrate should be transparent in IR range.

Figure 6 shows the Raman spectra of as-deposited a-Si:H film and nanosecond-pulse-annealed film. The Raman spectrum of amorphous silicon contains broad peaks at approximately 480 cm−1 appearing as a result of effective density of trafnsverse optical (TO) vibration modes [19]. Due to scattering on optical phonon modes localized in the nanocrystals, the Raman spectrum of nanocrystals is characterized by narrow peak at position between 500 and 520 cm−1. The position and the width of the peak strongly depend on size and structure of the nanocrystals according to dispersion of the localized modes [22]. The peak width is mainly determined by size dispersion of the nanocrystals. The intensity of the “nanocrystal” peak depends on the volume of nanocrystal phase. So, one can determine the volume part of nanocrystal phase: , where and are integral intensities of crystalline and amorphous peaks and σ is relation between the integral Raman cross-sections of nanocrystal Si and a-Si. According to the literature data, σ can change from 0.1 for large grain polycrystalline silicon up to 1.1 for nanocrystalline Si [23]. As one can see, the used nanosecond treatment leads to almost total crystallization of as-deposited film. According to peak position, the size of nanocrystals in our case should be about 10 nm [22]. It should be noted that no Si–H peaks were detected in nanosecond-pulse-annealed film. So, the pulse annealing can be used not only for crystallization but also for dehydrogenization.

4. Conclusion

In conclusion, the FTIR method was applied for determination of concentrations of chemical species in silicon-rich nitride films containing amorphous silicon nanoclusters. It was observed that nearly stoichiometric hydrogenated silicon nitride film contains hydrogen mainly in the form of the Si–H bonds even after 5 hours of annealing at temperature 1130°C. According to Raman spectroscopy data, nanosecond pulse laser treatment leads to crystallization and dehydrogenization of a-Si:H films with initial concentration of hydrogen up to 35%.


The authors are thankful to Dr. A. A. Popov (Yaroslavl Department of FTI RAS) and A. Kh. Antonenko for providing of as-deposited films, Professor A. Misiuk for furnace annealing, and Dr. S. A. Kochubei for help in laser annealing. They are pleased to thank directorate of Scientific and Educational Center “Nanosystems and Advanced Materials,” Novosibirsk State University.


  1. N. M. Park, C. J. Choi, T. Y. Seong, and S. J. Park, “Quantum confinement in amorphous silicon quantum dots embedded in silicon nitride,” Physical Review Letters, vol. 86, no. 7, pp. 1355–1357, 2001. View at: Publisher Site | Google Scholar
  2. M. Dai, K. Chen, X. Huang et al., “Formation and charging effect of Si nanocrystals in a-SiNx/a-Si/a-SiNx structures,” Journal of Applied Physics, vol. 95, no. 2, pp. 640–645, 2004. View at: Publisher Site | Google Scholar
  3. C. W. Jiang and M. A. Green, “Silicon quantum dot superlattices: modeling of energy bands, densities of states, and mobilities for silicon tandem solar cell applications,” Journal of Applied Physics, vol. 99, no. 11, Article ID 114902, pp. 1–7, 2006. View at: Publisher Site | Google Scholar
  4. A. J. Nozik, “Quantum dot solar cells,” Physica E, vol. 14, no. 1-2, pp. 115–120, 2002. View at: Publisher Site | Google Scholar
  5. C. Doughty, D. C. Knick, J. B. Bailey, and J. E. Spencer, “Silicon nitride films deposited at substrate temperatures <100°C in a permanent magnet electron cyclotron resonance plasma,” Journal of Vacuum Science and Technology A, vol. 17, no. 5, pp. 2612–2618, 1999. View at: Google Scholar
  6. T. T. Korchagina, D. V. Marin, V. A. Volodin, A. A. Popov, and M. Vergnat, “Structure and optical properties of SiNx:H films with Si nanoclusters produced by low-frequency plasma-enhanced chemical vapor deposition,” Semiconductors, vol. 43, no. 11, pp. 1514–1519, 2009. View at: Publisher Site | Google Scholar
  7. J. D. Joannopoulos and G. Lucovsky, The Physics of Hydrogenated Amorphous Silicon, vol. I, Springer, Berlin, Germany, 1984.
  8. A. Milekhin, M. Friedrich, K. Hiller, M. Wiemer, T. Gessner, and D. R. T. Zahn, “Characterization of low-temperature wafer bonding by infrared spectroscopy,” Journal of Vacuum Science and Technology B, vol. 18, no. 3, pp. 1392–1396, 2000. View at: Publisher Site | Google Scholar
  9. S. Ali, M. Gharghi, S. Sivoththaman, and K. Zeaiter, “Properties and characterization of low-temperature amorphous PECVD silicon nitride films for solar cell passivation,” Journal of Materials Science, vol. 40, no. 6, pp. 1469–1473, 2005. View at: Publisher Site | Google Scholar
  10. V. I. Belyi, L. L. Vasilyeva, A. S. Ginkover et al., Silicon Nitride in Electronics, Elsevier, New York, NY, USA, 1988.
  11. W. A. Lanford and M. J. Rand, “The hydrogen content of plasma-deposited silicon nitride,” Journal of Applied Physics, vol. 49, no. 4, pp. 2473–2477, 1978. View at: Publisher Site | Google Scholar
  12. H. J. Stein and H. A. R. Wegener, “Chemically bound hydrogen in CVD Si3N4: dependence on NH /SiH ratio and on annealing,” Journal of the Electrochemical Society, vol. 124, no. 6, pp. 908–912, 1977. View at: Google Scholar
  13. V. A. Gritsenko, “Atomic structure of the amorphous nonstoichiometric silicon oxides and nitrides,” Physics-Uspekhi, vol. 51, no. 7, pp. 699–711, 2008. View at: Publisher Site | Google Scholar
  14. D. V. Tsu, G. Lucovsky, and M. J. Mantini, “Local atomic structure in thin films of silicon nitride and silicon diimide produced by remote plasma-enhanced chemical-vapor deposition,” Physical Review B, vol. 33, no. 10, pp. 7069–7081, 1986. View at: Publisher Site | Google Scholar
  15. Y. N. Volgin and Y. I. Ukhanov, “Vibration spectra of silicon nitride,” Optics and Spectroscopy, vol. 38, pp. 412–415, 1975. View at: Google Scholar
  16. N. Wada, S. A. Solin, J. Wong, and S. Prochazka, “Raman and IR absorption spectroscopic studies on α, β, and amorphous Si3N4,” Journal of Non-Crystalline Solids, vol. 43, no. 1, pp. 7–11, 1981. View at: Google Scholar
  17. M. Molinari, H. Rinnert, and M. Vergnat, “Evolution with the annealing treatments of the photoluminescence mechanisms in a-SiNx:H alloys prepared by reactive evaporation,” Journal of Applied Physics, vol. 101, no. 12, Article ID 123532, pp. 1–8, 2007. View at: Publisher Site | Google Scholar
  18. G. Scardera, T. Puzzer, G. Conibeer, and M. A. Green, “Fourier transform infrared spectroscopy of annealed silicon-rich silicon nitride thin films,” Journal of Applied Physics, vol. 104, no. 10, Article ID 104310, pp. 1–5, 2008. View at: Publisher Site | Google Scholar
  19. M. H. Brodsky, M. Cardona, and J. J. Cuomo, “Infrared and Raman spectra of the silicon-hydrogen bonds in amorphous silicon prepared by glow discharge and sputtering,” Physical Review B, vol. 16, no. 8, pp. 3556–3571, 1977. View at: Publisher Site | Google Scholar
  20. T. T. T. Hien, C. Ishizaki, and K. Ishizaki, “SiHx absorbance bands in Si3N4 powder surfaces analyzed by diffuse reflectance infrared fourier transform spectroscopy,” Journal of the Ceramic Society of Japan, vol. 112, no. 1301, pp. 1–5, 2004. View at: Google Scholar
  21. D. M. Goldie and S. K. Persheyev, “Quantitative hydrogen measurements in PECVD and HWCVD a-Si:H using FTIR spectroscopy,” Journal of Materials Science, vol. 41, no. 16, pp. 5287–5291, 2006. View at: Publisher Site | Google Scholar
  22. V. Paillard, P. Puech, M. A. Laguna, R. Carles, B. Kohn, and F. Huisken, “Improved one-phonon confinement model for an accurate size determination of silicon nanocrystals,” Journal of Applied Physics, vol. 86, no. 4, pp. 1921–1925, 1999. View at: Google Scholar
  23. E. Bustarret, M. A. Hachicha, and M. Brunel, “Experimental determination of the nanocrystalline volume fraction in silicon thin films from Raman spectroscopy,” Applied Physics Letters, vol. 52, no. 20, pp. 1675–1677, 1988. View at: Publisher Site | Google Scholar

Copyright © 2012 Kirill O. Bugaev et al. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

More related articles

7233 Views | 4416 Downloads | 12 Citations
 PDF  Download Citation  Citation
 Download other formatsMore
 Order printed copiesOrder

Related articles

We are committed to sharing findings related to COVID-19 as quickly and safely as possible. Any author submitting a COVID-19 paper should notify us at to ensure their research is fast-tracked and made available on a preprint server as soon as possible. We will be providing unlimited waivers of publication charges for accepted articles related to COVID-19. Sign up here as a reviewer to help fast-track new submissions.