Open Access
Add to BibSonomyAbstract
Green photoluminescence (PL) from silicon microstructures fabricated by femtosecond laser in air was studied at different temperature by time-resolved spectroscopy. The PL decay profiles are well fitted by a stretched exponential function: The dependence of the decay time constant τ and of the stretching index β on PL photon energy and on the temperature is investigated. A model of transport and recombination of the carriers is introduced as a possible explanation of the stretched exponential decay. The nonradiative recombination rate of the localized carriers, which is dependent on the carrier density and influenced by the trapping site density and the temperature, is deduced to be responsible for this kind of decay.
© 2013 Optical Society of America
1. Introduction
In past decades, much effort has been devoted to achieve light-emitting devices based on silicon, in order to conveniently integrate electronic and optical functions onto the same chip. However, the nature of the indirect band gap of silicon makes it an inefficient light source. Several approaches were undertaken to overcome this difficulty, e.g., using porous silicon (PS) [1,2] or silicon nanocrystals (Si NCs) [3–5]. Different PL bands have been observed on various samples of PS and Si NCs [6–10]. A blue-green band has been attributed to surface localized states related to oxygen defects [11,12]. A red band, whose peak wavelength has a blue shift as the size of silicon nanocrystals decreases, originates from a quantum confinement effect [13,14]. Recently, photoluminescence (PL) was observed at room temperature on microstructured silicon fabricated in air by a femtosecond laser [15]. It is promising to be used for the light-emitting devices. Moreover, femtosecond laser microstructured silicon has a very high absorptance from the near ultraviolet to the near inferred, which makes it possible to be used to improve the efficiency of silicon solar cells [16]. This material exhibits advantages in the field of field-emission and photo-detection as well [17,18]. It is notable that the photo-detector based on femtosecond laser microstructured silicon has exhibited excellent performance [19]. Up to now, however, the dynamics of photo-generated carriers in microstructured silicon, which is very important information for improving the performance of the devices based on this kind of material, are still unclear, due to the complexity of the material.
In this letter, PL of the microstructured silicon was studied at different temperature by time-resolved spectroscopy that is a powerful tool to study the mechanism of PL and the dynamics of the carriers in a complex system [20], aiming to solve the problems mentioned above. The temperature dependence of PL intensity implies a wide band tail in the microstructured silicon. The PL decay profile is nonexponential, which is well fitted with a stretched exponential function. The dependences of the fitting parameters (the decay time constant and the stretching index) on temperature and on PL photon energy were determined. To explain the stretched exponential decay of PL and illustrate the dynamics of the carriers, a model of transport and recombination of the carriers is established. The effect of the trapping sites is discussed as well. This model is well consistent with the experimental results.
2. Experiments
The sample was fabricated by irradiating an n-type silicon (111) wafer in air with a train of 120 fs laser pulses at a repetition rate of 1 kHz and at a central wavelength of 800 nm. The silicon wafers, which had a resistivity of more than 1000 , were cleaned with hydrofluoric acid to remove any native oxide and then rinsed with distilled water. Microstructured silicon patches (5 mm × 5 mm) were produced by translating the silicon wafer during irradiation under a laser fluence of 10 kJ/m2 in air. Subsequently, the sample was annealed at 1300 K for 1 hour in vacuum.
PL investigations were performed with the experimental setup shown in Fig. 1. The 800 nm femtosecond laser was frequency doubled to 400 nm using a BBO crystal to excite the sample that was placed in a hot and cold stage (HCS402, INSTEC). This stage can keep the sample temperature from 90 K to 300 K. PL signals were collected by a light collection system, which contained a longpass filter to block the scattered pump light. The collected signal was focused onto the slit of a spectrograph (SpectraPro-300i, Acton Research Corporation) coupled to an ICCD camera (PicoStar HR 12, LaVision). The ICCD and the femtosecond laser system were synchronized with a precision better than 10 ps. A programmable delay generator was used to control the delay between the laser pulse and the gate signal of the ICCD. The minimum width of the gate signal is 30 ps.
Fig. 1 The schematic of the experimental setup. SHG: second harmonic generation; Filter1: blue bandpass filter; M1: reflection mirror; L1, L2, L3: lens; Filter2: longpass filter.
3. Results and discussion
The inset [Fig. 2(b)] shows the SEM (scanning electron microscopy) image of the surface of the annealed sample. After femtosecond laser ablation in air, micro-cones with a height of some tens of microns are seen on the silicon surface. The cones are covered by smaller microstructures that contain silicon nanocrystals with an oxide layer [15]. This kind of generation of nanoparticles and nanostructures is very common during femtosecond laser ablation [21–23]. The removed material in the ablated plume can redeposit on the surface of the sample due to confinement of the ambient gas [24], thus forming the nanostructures.
Fig. 2 Temporal integral of time-resolved PL spectra of microstructured silicon obtained at 90 K and 300 K, respectively. The inset (a) shows the temperature dependence of the PL intensity of the annealed sample at peak wavelength. The solid line represents a theoretical fit. The inset (b) shows the SEM image of the annealed sample.
The PL spectra shown in Fig. 2 are obtained by temporally integrating the time-resolved spectra measured at 90 K and 300 K respectively. The peak wavelength, which is approximately at 530 nm for both annealed sample and unannealed sample, does not vary with sample temperature. This green band is due to oxygen-related defects that locate at the Si/SiO2 interface [15] and can greatly affect the properties of the silicon nanocrystals [25,26]. This can be confirmed by the disappearance of the green band after removing the oxide layer on the sample’s surface. As shown in Fig. 2, the PL intensity of the annealed sample is higher than that of the unannealed sample due to the decrease of the trapping sites (nonradiative recombination centers) after annealing. Hence, we paid more attention to the annealed sample in this study.
As seen in Fig. 2, the PL intensity is much higher at 90 K than that at 300 K. To determine the temperature effect, the peak intensity of the annealed sample is plotted as a function of temperature, as shown in Fig. 2(a). From 90 K to 300K, the PL intensity decreases with temperature. This is due to the competition between the radiative and nonradiative recombination. At low temperature, the photogenerated carriers are more easily localized in the surface states that lie in the band tails and are related with the oxygen-related defects at Si/SiO2 interface, causing a stronger optical radiation. As the temperature increases, lots of carriers can be reemitted from the surface states to the mobility edge and diffuse to the trapping sites, and then recombine nonradiatively, leading to a decrease of the optical emission. The dependence of PL intensity on temperature can be described by [6]:
Here B and I0 are constants, i.e., independent of temperature, and T0 denotes the degree of the system disorder, related to the tail width of the density of localized states [27]. From the fit we obtain T0~163 K (i.e. kBT0~14 meV). This points to a wide band tail in the microstructured silicon.
Figures 3(a) and 3(c) show the time-resolved PL spectra of the annealed sample at 90 K and 300 K. The PL decay profiles at different wavelengths are extracted from them, as shown in Figs. 3(b) and 3(d). It is evident that the decay processes are temperature dependent and that the PL decay profiles are nonexponential. The decay profiles are well fitted by a stretched exponential function [28]:
where τ is a time constant, and is the stretching index. The stretched exponential function is well known to describe the PL decay and transport properties of disordered systems [29–31]. The stretching index is a measure for the degree of disorder in the material.Fig. 3 Time-resolved PL spectra of the annealed sample measured at (a) 300 K and (c) 90 K. The decay profiles at different wavelengths are obtained at (b) 300 K and (d) 90 K. The solid lines are the fits with the stretched exponential function.
As seen in Figs. 3(b) and 3(d), the PL decay profiles at different wavelengths differ for 300 K and 90 K. The fitting results are listed in Table 1.
Table 1. Values of τ and β Obtained by Fitting PL Decay Data with Eq. (2) at Different Wavelengths at 300 K and 90 K.
At a certain temperature, the time constant τ decreases with wavelength while β nearly keeps constant. This suggests that the surface states with different energy levels have different lifetimes. The invariable β proves that the mechanisms causing the stretched exponential decay are similar at different wavelengths. We will introduce a model to explain the mechanisms in more detail in next paragraph. It is notable that the values of τ and β are larger at 90 K than that at 300 K. To determine the dependence of τ and β on temperature, the values of τ and β for the wavelength of 530 nm are plotted as a function of temperature, as shown in Fig. 4. The value of β varies only a little from 90 K to 240 K, while visibly decreases for temperatures above 240 K. The value of τ decreases with temperature, which can be explained by an enhancement of the nonradiative recombination at higher temperature. The temperature dependence of the time constant τ can be described by the classical expression [32]:
where τ0 and s are constants, and E denotes the activation energy. Applying Eq. (3) yields an activation energy of E~98.6 meV.Fig. 4 Dependences of the time constant and the stretching index on temperature for the wavelength of 530 nm. The data of the time constant are fitted by using Eq. (3) and the solid curve through the data of the stretching index serves only to guide the eye.
The decay of PL of the microstructured silicon is well described by the stretched exponential function. Many efforts have been made to find out the physics behind the exponential decay [29, 33]. However, the physical mechanism is unclear yet. For this sake, we introduce a model of transport and recombination of the carriers as a possible explanation for the stretched exponential decay. When a pump laser pulse impacts upon the sample surface, large amounts of carriers are excited in the cores of the silicon nanocrystals, and then quickly transmit to the Si/SiO2 interface and are localized in the surface states. The localized carriers can recombine radiatively, or be thermally reemitted from the localized states. The reemitted carriers may diffuse to the neighboring trapping sites and recombine nonradiatively, or be re-localized by the surface states. During the PL process, the radiative recombination and the total nonradiative recombination compete with each other. The dynamics of reemission, diffusion, re-localizing, trapping, and nonradiative recombination of the localized carriers, which is dependent on the carrier density and influenced by the trapping site density and the temperature, determines the nonexponential decay of PL.
By using a diffusion model, we estimate that the carriers transmit from the silicon core (several nanometers) to the Si/SiO2 interface on the time scale of tens to hundreds of picosecond, which is much faster than the decay of PL. Hence, we approximately consider the decay of PL with an initial density of the localized carriers. The density of the localized carriers can be described as:
where Asr is the radiative recombination rate, which is a constant. Asnr is the nonradiative recombination rate, which is determined by the dynamics of the localized carriers and should be a function of the carrier density. We assume that Asnr can be described as Here, the coefficient Asnr0 and the exponent b are constant. Hence, the optical emission intensity can be described by:where denotes the photon energy. We estimate the initial density (N0) of the excited carriers at ~4 × 1018 cm−3 by considering the incident laser fluence of 0.4 mJ/cm2 (used in our experiment) and a measured absorptivity of 0.9 for the laser at 400 nm. The decay profiles of PL at 530 nm for the annealed sample are fitted by Eq. (5) combined with Eq. (4), as shown in Fig. 5. The experimental results are perfectly consistent with this model. This hints that the stretched exponential decay of PL is possibly caused by the nonradiative recombination rate that is dependent on the carrier density.Fig. 5 The decay profiles of the PL (annealed sample) at 530 nm under 300 K and 90 K respectively. The points represent the experimental data and the solid lines denote the fits with Eq. (5). The fitting parameters are listed in the inset table.
As seen in the inset table of Fig. 5, the coefficient Asnr0 is larger at higher temperature. It means that the nonradiative recombination is more intense at higher temperature due to a stronger reemission of localized carriers and an enhancement of nonradiative recombination, leading to a smaller β. Besides dependence on sample’s temperature, the decay profile is influenced by the density of trapping sites as well. We obtain Asnr0~6.9 × 10−10 and β~0.42 for an unannealed sample at 300 K, while they are 2.4 × 10−10 and for the annealed sample respectively. This is consistent with the fact that the trapping sites remarkably decrease after high-temperature annealing. As a result, it is more difficult for the reemitted carriers to diffuse to the trapping site and nonradiatively recombine, which suppresses the nonradiative recombination.
4. Summary
In summary, PL of silicon microstructures fabricated by femtosecond laser in air has been studied at different temperature by time-resolved spectroscopy. The stretched exponential decay of PL implies a complex dynamics of photo-generated carriers in microstructured silicon. The intensity and the decay time of PL decrease with temperature, suggesting a competition between radiative and nonradiative recombination. This competition is governed by the carrier dynamics. A model of transport and recombination of the carriers is introduced to illustrate the carrier dynamics and explain the stretched exponential decay. The nonradiative recombination rate, which is determined by a complex process of reemission, diffusion, re-localizing, trapping, and nonradiative recombination of the localized carriers, is dependent on the carrier density and influenced by the trapping site density and the temperature. This is deduced to be responsible for the stretched exponential decay. The effect of the trapping sites should be similar in other photoelectric devices based on the microstructured silicon. Our results are helpful to analyze the dynamics of the carriers in these devices as well.
Acknowledgments
This work was supported by the National Basic Research Program of China (2012CB934201 and 2013CB328702), the 111 Project (B07013), and the National Natural Science Foundation of China (11074129).
References and links
1. L. T. Canham, “Silicon quantum wire array fabrication by electrochemical and chemical dissolution of wafer,” Appl. Phys. Lett. 57(10), 1046–1048 (1990). [CrossRef]
2. V. Lehmann and U. Gosele, “Porous silicon formation: A quantum wire effect,” Appl. Phys. Lett. 58(8), 856–858 (1991). [CrossRef]
3. T. Shimizu-lwayama, S. Nakao, and K. Saitoh, “Visible photoluminescence in Si+-implanted thermal oxide films on crystalline Si,” Appl. Phys. Lett. 65(14), 1814–1816 (1994). [CrossRef]
4. R. J. Walters, G. I. Bourianoff, and H. A. Atwater, “Field-effect electroluminescence in silicon nanocrystals,” Nat. Mater. 4(2), 143–146 (2005). [CrossRef] [PubMed]
5. K. Žídek, F. Trojánek, P. Malý, L. Ondič, I. Pelant, K. Dohnalová, L. Šiller, R. Little, and B. R. Horrocks, “Femtosecond luminescence spectroscopy of core states in silicon nanocrystals,” Opt. Express 18(24), 25241–25249 (2010). [CrossRef] [PubMed]
6. X. Chen, D. Uttamchandani, C. Trager-Cowan, and K. P. O’Donnell, “Luminescence from porous silicon,” Semicond. Sci. Technol. 8(1), 92–96 (1993). [CrossRef]
7. M. Zhu, Y. Han, R. B. Wehrspohn, C. Godet, R. Etemadi, and D. Ballutaud, “The origin of visible photoluminescence from silicon oxide thin films prepared by dual-plasma chemical vapor deposition,” J. Appl. Phys. 83(10), 5386–5393 (1998). [CrossRef]
8. T. Schmidt, A. I. Chizhik, A. M. Chizhik, K. Potrick, A. J. Meixner, and F. Huisken, “Radiative exciton recombination and defect luminescence observed in single silicon nanocrystals,” Phys. Rev. B 86(12), 125302 (2012). [CrossRef]
9. K. Kůsová, O. Cibulka, K. Dohnalová, I. Pelant, J. Valenta, A. Fucíková, K. Zídek, J. Lang, J. Englich, P. Matejka, P. Stepánek, and S. Bakardjieva, “Brightly luminescent organically capped silicon nanocrystals fabricated at room temperature and atmospheric pressure,” ACS Nano 4(8), 4495–4504 (2010). [CrossRef] [PubMed]
10. D. S. English, L. E. Pell, Z. Yu, P. F. Barbara, and B. A. Korgel, “Size tunable visible luminescence from individual organic monolayer stabilized silicon nanocrytal quantum dots,” Nano Lett. 2(7), 681–685 (2002). [CrossRef]
11. L. Tsybeskov, J. V. Vandyshev, and P. M. Fauchet, “Blue emission in porous silicon: Oxygen-related photoluminescence,” Phys. Rev. B Condens. Matter 49(11), 7821–7824 (1994). [CrossRef] [PubMed]
12. H. Tamura, M. Ruckschloss, T. Wirschem, and S. Veprek, “Origin of the green/blue luminescence from nanocrystalline silicon,” Appl. Phys. Lett. 65(12), 1537–1539 (1994). [CrossRef]
13. G. Ledoux, J. Gong, and F. Huisken, “Effect of passivation and aging on the photoluminescence of silicon nanocrystals,” Appl. Phys. Lett. 79(24), 4028–4030 (2001). [CrossRef]
14. V. Vinciguerra, G. Franzo, F. Priolo, F. Iacona, and C. Spinella, “Quantum confinement and recombination dynamics in silicon nanocrystals embedded in Si/SiO2 superlattices,” J. Appl. Phys. 87(11), 8165–8173 (2000). [CrossRef]
15. C. Wu, C. H. Crouch, L. Zhao, and E. Mazur, “Visible luminescence from silicon surfaces microstructured in air,” Appl. Phys. Lett. 81(11), 1999–2001 (2002). [CrossRef]
16. C. Wu, C. H. Crouch, L. Zhao, J. E. Carey, R. Younkin, J. A. Levinson, E. Mazur, R. M. Farrell, P. Gothoskar, and A. Karger, “Near-unity below-band-gap absorption by microstructured silicon,” Appl. Phys. Lett. 78(13), 1850–1852 (2001). [CrossRef]
17. M. Y. Shen, C. H. Crouch, J. E. Carey, R. Younkin, E. Mazur, M. Sheehy, and C. M. Friend, “Formation of regular arrays of silicon microspikes by femtosecond laser irradiation through a mask,” Appl. Phys. Lett. 82(11), 1715–1717 (2003). [CrossRef]
18. J. E. Carey, C. H. Crouch, M. Y. Shen, and E. Mazur, “Visible and near-infrared responsivity of femtosecond-laser microstructured silicon photodiodes,” Opt. Lett. 30(14), 1773–1775 (2005). [CrossRef] [PubMed]
19. Z. H. Huang, J. E. Carey, M. G. Liu, X. Y. Guo, E. Mazur, and J. C. Campbell, “Microstructured silicon photodetector,” Appl. Phys. Lett. 89(3), 033506 (2006). [CrossRef]
20. Q. Wu, S. Guo, Y. Ma, F. Gao, C. Yang, M. Yang, X. Yu, X. Zhang, R. A. Rupp, and J. Xu, “Optical refocusing three-dimensional wide-field fluorescence lifetime imaging microscopy,” Opt. Express 20(2), 960–965 (2012). [CrossRef] [PubMed]
21. A. Menéndez-Manjón, S. Barcikowski, G. A. Shafeev, V. I. Mazhukin, and B. N. Chichkov, “Influence of beam intensity profile on the aerodynamic particle size distributions generated by femtosecond laser ablation,” Laser Part. Beams 28(01), 45–52 (2010). [CrossRef]
22. Y. L. Wang, C. Chen, X. C. Ding, L. Z. Chu, Z. C. Deng, W. H. Liang, J. Z. Chen, and G. S. Fu, “Nucleation and growth of nanoparticles during pulsed laser deposition in an ambient gas,” Laser Part. Beams 29(01), 105–111 (2011). [CrossRef]
23. S. Manickam, K. Venkatakrishnan, B. Tan, and V. Venkataramanan, “Study of silicon nanofibrous structure formed by femtosecond laser irradiation in air,” Opt. Express 17(16), 13869–13874 (2009). [CrossRef] [PubMed]
24. Z. Chen, Q. Wu, M. Yang, B. Tang, J. Yao, R. A. Rupp, Y. Cao, and J. Xu, “Generation and evolution of plasma during femtosecond laser ablation of silicon in different ambient gases,” Laser Part. Beams (to be published).
25. M. V. Wolkin, J. Jorne, P. M. Fauchet, G. Allan, and C. Delerue, “Electronic states and luminescence in porous silicon quantum dots: The role of oxygen,” Phys. Rev. Lett. 82(1), 197–200 (1999). [CrossRef]
26. J. Martin, F. Cichos, F. Huisken, and C. von Borczyskowski, “Electron-phonon coupling and localization of excitons in single silicon nanocrystals,” Nano Lett. 8(2), 656–660 (2008). [CrossRef] [PubMed]
27. R. W. Collins, M. A. Paesler, and W. Paul, “The temperature dependence of photoluminescence in a-Si: H alloys,” Solid State Commun. 34(10), 833–836 (1980). [CrossRef]
28. R. Kohlrausch, “Nachtrag ueber die elastische Nachwirkung beim Cocon-und Glasfaden, und die hygroskopische Eigenschaft des ersteren,” Ann. Phys. (Leipzig) 12, 393–399 (1847).
29. B. Sturman, E. Podivilov, and M. Gorkunov, “Origin of stretched exponential relaxation for hopping-transport models,” Phys. Rev. Lett. 91(17), 176602 (2003). [CrossRef] [PubMed]
30. T. Bartel, M. Dworzak, M. Strassburg, A. Hoffmann, A. Strittmatter, and D. Bimberg, “Recombination dynamics of localized excitons in InGaN quantum dots,” Appl. Phys. Lett. 85(11), 1946–1948 (2004). [CrossRef]
31. M. Dovrat, Y. Goshen, J. Jedrzejewski, I. Balberg, and A. Sa’ar, “Radiative versus nonradiative decay processes in silicon nanocrystals probed by time-resolved photoluminescence spectroscopy,” Phys. Rev. B 69(15), 155311 (2004). [CrossRef]
32. S. E. Paje and J. Llopis, “Photoluminescence decay and time-resolved spectroscopy of cubic yttria-stabilized zirconia,” Appl. Phys., A Mater. Sci. Process. 59(6), 569–574 (1994). [CrossRef]
33. F. Sangghaleh, B. Bruhn, T. Schmidt, and J. Linnros, “Exciton lifetime measurements on single silicon quantum dots,” Nanotechnology 24(22), 225204 (2013). [CrossRef] [PubMed]
