Share with Facebook
Tweet This
Post on reddit
Share with LinkedIn
Add to CiteULike
Add to Mendeley
Add to BibSonomyAbstract
A model, by combining Maxwell’s equations with all-parameters of Sellmeier’s fitting equations and four-level rate equations, is built to investigate linear dispersive effect on the property of random lasing modes. Computed results show that the first excited modes for both dispersive and non-dispersive scattering cases have almost the same resonant frequency but the spectral intensity for dispersive case is lower than that for non-dispersive case, and there exist more modes in the whole spectra for dispersive case. Further analysis demonstrates that threshold of random lasing in dispersive case is higher than that of the non-dispersive case.
©2011 Optical Society of America
1. Introduction
Over the past few years, laser action in random media (random lasing) have attracted considerable attention [1–16]. The feedback of laser action for random lasing is not provided by an external resonator, but by scatterers which are randomly distributed in an active medium or which by themselves act as optical amplifiers. It is noting that the sharp peaks in the emission spectra of semi-conductor powders, first observed in 1999, has therefore lead to an intense debate about the nature of the lasing modes in these so-called lasers with resonant feedback. Many experimental and theoretical studies aimed at clarifying the nature of the lasing modes in disordered scattering systems with gain [3–14].
For a random lasing system different modes supported by the disordered medium have different resonant wavelengths, for which the wavelength interval between modes usually has only a few nanometers. Therefore, in the previous theoretical models of random lasing, the electric permittivity ε of the medium, to the best of our knowledge, was treated as a constant because the dispersion of the medium employed was usually quite small in optical band [3–5,8–14]. However, if the dispersive effect is strong enough, what will happen? For example, the value of ε for Al2O3 takes from 3.6 to 3.1 while the wavelength λ varies from 0.2 µm to 0.8 µm. How such a dispersion effect will influence the property of random lasing supported by the disordered medium made by Al2O3 scattering particles? In this paper, we devote to study the linear dispersive effect on the character of random lasing. To do so, we build a new model via combining Maxwell’s equations with all-parameters of Sellmeier’s fitting equations and four-level rate equations to investigate the random lasing phenomenon in the disordered system with dye as active medium and Al2O3 grains as scattering particles. Our numerical results show that the peak intensity of the first excited mode for dispersive case is lower than that for non-dispersive case, but the resonant wavelengths of the modes are nearly the same, and there are more modes in the whole spectra for dispersive case. Further analysis demonstrates that the threshold of random lasing in dispersive case is higher than that of the non-dispersive case.
2. Theoretical model
We start our analysis with constructing a one-dimension (1-D) system, which consists of two dielectric materials, as shown in Fig. 1 . The white layer with a random variable thickness a n and a dielectric constant ε1 = ε0 simulates the dye, while the black layer with a fixed thickness b = 300 nm simulates the scatters. For sake of contrast, two scattering medium models of a linear dispersive material and a non-dispersive material are selected, respectively. Note that the linear dispersive materials are characterized by Sellmeier’s equation while the permittivity of the non-dispersive material is chosen as a constant ε2. The random variable a n is described as , where a = 180 nm, w is the strength of randomness, and γ is a random value in the range [-0.5, 0.5].
In the following, we will discuss the computational theory and algorithm for gain and scattering medium, respectively.
2.1 Optical gain material
For optical gain and non-magnetic medium, the 1D time-dependent Maxwell equations read
where P gain is the polarization density component from which the amplification or gain can be obtained in z direction, and are the electric permittivity and the magnetic permeability of vacuum, respectively. is the relative electric permittivity of gain medium. For dye, is chosen as 1.For the four-level atomic system, the rate equations of the gain medium read
Set the particle population in unit volume of each level is N 4, N 3, N 2 and N 1 individually. The pumping rate from E 1 to E 4 is W p; The particles arrived E 4 transfer to E 3 quickly in the form of radiationless transition, the factor of probability is . Before the population inversed, E 3 transfer to E 2 quickly in the form of spontaneous activity emission, the factor of probability is . E 2 transfer to E 1 mostly in the form of spontaneous activity emission, the factor of probability is .The polarization Pgain obeys the following equation:
This equation links Maxwell’s equations with rate equations. is the population difference density between the populations in the lower and upper levels of the atomic transition. Amplification takes place when the external pumping mechanism produces population inversion ΔN<0. The linewidth of the atomic transition is Δωl=1/τ21+2/T 2 where the collision time T 2 is usually much smaller than the lifetime τ21. The constant κ is given by κ=6πε0 c 3/ωl 2τ21.2.2 Scattering medium
2.2.1 Linear dispersive dielectrics
In the linear dispersive scattering medium, Maxwell’s equations for the linear dispersive scattering medium read
Here, is the linear polarization density component in z direction. For simplicity, we omit hereafter the expression of x dependence, that is, we simplify (t, x) and (ω, x) to (t) and (ω), respectively. In the frequency domain, P lorentz of Eq. (4d) is defined aswhere the Fourier transform of all values is expressed with a tilde and is the linear susceptibility. Here, we assume that, using Sellmeier’s equation [17], (i = 1, 2 and 3) or dielectric constant εr(ω) is expressed bywhere ωi is the resonance frequency and B i is the strength of the ith resonance.Equations (4b), (5) and (6) lead to the following differential equations:
Next, the treatment presented in references [16] is followed, where for Sellmeier fitting equation with three resonances, ωi, the following system of differential equations is developed: In our simulation, the parameters [18] for Al2O3 in Eq. (6) are set as B 1 = 1.43134936, B 2 = 0.65054713, B 3 = 5.3414021, λ1 = 0.0726631 µm, λ2 = 0.1193242µm, and λ3 = 18.028251µm, where λi = 2πc/ωi and c is the velocity of light in vacuum. Figure 2 depicts the permittivity of Al2O3 in the visible wavelength range as calculated from Sellmeier fitting coefficients. This figure illustrates that the variation takes ε from 3.6 to 3.1 while the wavelength λ varies from 0.2 µm to 0.8 µm..
Fig. 2 Relative electric permittivity of Alumina (Al2O3) assuming no absorption as a function of wavelength in the visible range.
This system of three equations above is then solved to obtain P i (i = 1, 2, 3) which is used to calculate Ez:
2.2.2 Non-dispersive dielectrics
For the 1D time-dependent Maxwell equations and for a non-dispersive and non-magnetic medium, we have
where ε2 is a fixed constant.2.2.3 The computational algorithm
We now briefly summarize the computational algorithm.
For the optical gain medium, at time step n + 1, we first implement Eq. (3) to update P gain. Here, the use of explicit second-order finite-differences centered at time-step n requires only knowledge of E z at n. Next, we apply Eq. (1a) to update E z to time-step n + 1. Next, we apply Eq. (2b)–(2d) to update N i (i = 2, 3, 4)to time-step n + 1. Next, N 0 are calculated by using the conservation of electron populations. Finally we update H to time-step n + 3/2 by applying the Maxwell-Faraday law Eq. (1b).
For the dispersive scattering medium, applying explicit second-order finite-differences centered at time-step n for Eqs. (8), (9), and (10), this system can be solved to update P 1, P 2, and P 3 at time step n + 1. Next, with the updated values P 1, P 2, and P 3, we can update D z from Eq. (4a). Next, we apply Eq. (11) to update E z. Finally we update H to time-step n + 3/2 by applying Eq. (4b).
For the non-dispersive scattering medium, first we apply Eq. (12) to update E z. Next, we update H to time-step n + 3/2 by applying Eq. (13).
The values of those parameters in above equations that will be used in simulating the active part in the following numerical calculations are taken as: T 2 = 2.14 × 10−14 s, τ21 = 5 × 10−12 s, τ32 = 1 × 10−10 s, τ43 = 1 × 10−13 s, NT = = 3.313 × 1024 /m3, and = 6 × 1014 Hz (λl = 500 nm). When pumping is provided over the whole system, the electromagnetic fields can be calculated. In order to model such an open system, a Liao absorbing layer [19] is used to absorb the outward wave. The space and time increments have been chosen to be Δx = 10 nm and Δt = 1.67 × 10−17s, respectively. The pulse response is recorded during a time window of length Tw = 6 × 10−12 s at all nodes in the system and Fourier transformed in order to obtain the intensity spectrum.
3. Numerical results
We start our analysis with calculating spectral intensities vs pumping rates in the case of dispersive scattering medium, as shown in Fig. 3 . When pumping rate is relatively low, there exist many discrete peaks, each denoting a mode supported by disordered medium, as shown in Fig. 3(a). Note that three peaks are indicated by their central wavelengths λ0(497.5 nm), λ1(491.2 nm), and λ2(504.1 nm), respectively. When pumping rate increase to a special value (W p = 1 × 109 s−1), the mode λ0 dominates the whole spectra and its width becomes quitebroader than those at lower Wp, as shown in Fig. 3(b). This suggests that the mode λ0 is perhaps the first excited mode. With pumping rates further increasing, more modes are also excited and their widths become narrower and narrower, as shown in Fig. 3(c), 3(d), 3(e), and 3(f). And we can see that there are obvious mode competitions in whole spectrum.
Fig. 3 The spectral intensity in arbitrary units versus the wavelength in the case of the dispersive scattering medium, as shown in Fig. 1 at (a) WP = 1 × 108 s−1, (b) WP = 1 × 109 s−1, (c) WP = 1 × 1010 s−1, (d) WP = 1 × 1011 s−1, (e) WP = 1 × 1012 s−1, and (f) WP = 1 × 1013 s−1.
To explore the gain threshold behavior on random lasing in the case of dispersive scattering medium, numerical calculations are performed at different pump rates from which we can obtain the curves of the peak intensity and the spectral width vs the pump rate, as shown in Fig. 4 . According to the traditional method, the pump thresholds for the three modes can be measured from the intensity curves in Fig. 4(a) as WI 0 = 1.1 × 1010 s−1, WI 1 = 2 × 1010 s−1, and WI 2 = 4 × 1010 s−1, thus indicating different modes have different pump thresholds. Note that the mode λ0 has the minimum lasing threshold, that is to say the mode λ0 is the first excited mode. This is due to the fact that the central wavelength of the mode λ0 is very near the transition wavelength of the active medium.
Fig. 4 The plot of the peak intensity and spectral width of the lasing modes vs the pump rate W p in the case of the dispersive scattering medium.(a) The peak intensity for the three indicated modes, and the lasing threshold measured from the plots are W I0 = 1.1 × 1010 s−1, W I1 = 2 × 1010 s−1, and W I2 = 4 × 1010 s−1. (b) the peak intensity and spectral width for the mode λ0; and (c) the spectral width for the three indicated modes, and the lasing threshold measured from the plots are W w0 = 5 × 109 s−1, W w1 = 0.9 × 1010 s−1, and W w2 = 1.1 × 1010 s−1.
As can be seen from Fig. 4(b), within a pump regime near the pump threshold, a jump occurs for the spectral width for the mode λ0. The peak value of the jump appears at the point that is very near the threshold of the mode. A method was proposed to determine the lasing threshold for a 2D random laser based on the spectral width [6]. That is, the lasing threshold is defined as such a pumping energy at which the spectral width becomes half of its maximal value. This method was used to analyze the threshold gain behavior for random lasing in a 2D disordered medium [10]. Based on this method and the width curves in Fig. 4(c), the thresholds are measured as W w0 = 5 × 109 s−1, W w1 = 0.9 × 1010 s−1, and W w2 = 1.1 × 1010 s−1. Obviously, the results from the two methods are consistent.
Compared with dispersive scattering case, we restart to calculate the non-dispersive random lasing by the use of fixed constant ε2 = 3.1487, which is obtained from Sellmeier Eq. (6) at λ0(497.54 nm). The calculated spectral intensities vs pumping rates are plotted in Fig. 5 . Here two peaks are indicated by their central wavelengths λ0 ´(497.3 nm) and λ1 ´(493.1 nm), respectively. And the pump thresholds for the two modes via the above two methods can be measured from the intensity and width curves in Fig. 6(a) as WI 0 = 4 × 109 s−1 and WI 1 = 4 × 1010 s−1, and from the width curves in Fig. 6(c) as W'W 0 = 3 × 109 s−1, and = 0.9 × 1010 s−1, respectively. The two methods indicate that the mode is the first excited mode in the non-dispersive scattering case. It is quoting that the first excited modes for both dispersive and non-dispersive cases are only slightly different, whereas threshold of random lasing in dispersive scattering case is higher than that in the non-dispersive case via the two measured methods. Further note that, as shown in Fig. 3 and Fig. 5, the excited modes in the case of non-dispersive scattering case are fewer than that in the case of dispersive scattering case.As discussed above, we now know that, although the linear dispersion in optical domain is relatively small, the linear dispersion for scattering nanoparticles has a strong effect on modes of random lasing, which lead to richer lasing modes. In order to check whether the results above are universal or not, we calculate five different disordered structures for Al2O3, ZnO and TiO2, as shown in Table 1 . One can see that all thresholds of random lasing in dispersive scattering case are higher than that in the non-dispersive case for three scattering medium. And the numbers of the spectral spikes for dispersive medium are more than that for non-dispersive medium. All the above results demonstrate that the conclusion we have obtained is universal.
Fig. 5 The spectral intensity in arbitrary units versus the wavelength in the case of the non-dispersive scattering medium. shown in Fig. 1 at (a) WP = 1 × 108 s−1, (b) WP = 1 × 109 s−1, (c) WP = 1 × 1010 s−1, (d) WP = 1 × 1011 s−1, (e) WP = 1 × 1012 s−1, and (f) WP = 1 × 1013 s−1.
Fig. 6 The plot of the peak intensity and spectral width of the lasing modes vs the pump rate W p in the case of the non-dispersive scattering medium.(a) The peak intensity for the two indicated modes, and the lasing threshold measured from the plots are = 4 × 109 s−1, and = 5 × 1011 s−1. (b) the peak intensity and spectral width for the mode ; and (c) the spectral width for the three indicated modes, and the lasing threshold measured from the plots are = 3 × 109s−1, and = 0.9 × 1010s−1.
Table 1. The Thresholds (WI0, WI0´), the Numbers of the Spectral Spikes (Nums)
4. Conclusion
This work reports a model to reveal linear dispersive effect on modes of random lasing. The computed results show that dispersion leads to more modes in the spectra and higher thresholds.
Acknowledgments
The National Natural Science Foundation of China under Grant No. 10876010, and No. 60778003, and the Foundation Research Funds for the Central Universities under Grant No. 2010MS041 have supported this research.
References and links
1. N. M. Lawandy, R. M. Balachandran, A. S. L. Gomes, and E. Sauvain, “Laser action in strongly scattering media,” Nature 368(6470), 436–438 (1994). [CrossRef]
2. H. Cao, Y. Zhao, S. Ho, E. Seelig, Q. Wang, and R. Chang, “Random laser action in semiconductor powder,” Phys. Rev. Lett. 82(11), 2278–2281 (1999). [CrossRef]
3. X. Jiang and C. M. Soukoulis, “Time dependent theory for random lasers,” Phys. Rev. Lett. 85(1), 70–73 (2000). [CrossRef] [PubMed]
4. P. Sebbah and C. Vanneste, “Random laser in the localized regime,” Phys. Rev. B 66(14), 144202 (2002). [CrossRef]
5. C. M. Soukoulis, X. Jiang, J. Y. Xu, and H. Cao, “Dynamic response and relaxation oscillations in random lasers,” Phys. Rev. B 65(4), 041103–041106 (2002). [CrossRef]
6. T. Ito and M. Tomita, “Polarization-dependent laser action in a two-dimensional random medium,” Phys. Rev. E Stat. Nonlin. Soft Matter Phys. 66(2), 027601–027604 (2002). [CrossRef] [PubMed]
7. S. Mujumdar, M. Ricci, R. Torre, and D. S. Wiersma, “Amplified extended modes in random lasers,” Phys. Rev. Lett. 93(5), 053903–053907 (2004). [CrossRef] [PubMed]
8. X. Jiang, S. Feng, C. M. Soukoulis, J. Zi, J. D. Joannopoulos, and H. Cao, “Coupling, competition, and stability of modes in random lasers,” Phys. Rev. B 69(10), 104202 (2004). [CrossRef]
9. J. S. Liu, H. Lu, and C. Wang, “Spectral time evolution of quasistate modes in two-dimensional random media,” Acta Phys. Sin. 54, 3116 (2005) (in Chinese).
10. J. S. Liu and Z. Xiong, “Theoretical investigation on the threshold property of localized modes based on spectral width in two-dimensional random media,” Opt. Commun. 268(2), 294–299 (2006). [CrossRef]
11. J. S. Liu, H. Wang, and Z. Xiong, “Origin of light localization from orientational disorder in one and two-dimensional random media with uniaxial scatterers,” Phys. Rev. B 73(19), 195110 (2006). [CrossRef]
12. C. Wang and J. S. Liu, “Polarization dependence of lasing modes in two-dimensional random lasers,” Phys. Lett. A 353(2-3), 269–272 (2006). [CrossRef]
13. C. Vanneste, P. Sebbah, and H. Cao, “Lasing with resonant feedback in weakly scattering random systems,” Phys. Rev. Lett. 98(14), 143902 (2007). [CrossRef] [PubMed]
14. H. E. Türeci, L. Ge, S. Rotter, and A. D. Stone, “Strong interactions in multimode random lasers,” Science 320(5876), 643–646 (2008). [CrossRef] [PubMed]
15. D. S. Wiersma, “The physics and applications of random lasers,” Nat. Phys. 4(5), 359–367 (2008). [CrossRef]
16. O. Zaitsev and L. Deych, “Recent developments in the theory of multimode random lasers,” J. Opt. 12(2), 024001–024013 (2010). [CrossRef]
17. R. M. Joseph and A. Taflove, “FDTD Maxwell’s equations modes for nonlinear electrodynamics and optics,” IEEE Trans. Antenn. Propag. 45(3), 364–374 (1997). [CrossRef]
18. M. J. Weber, CRC Handbook of Optical Materials (CRC Press, 2003).
19. A. Taflove and S. C. Hagness, Computational Electrodynamics: The Finite-Difference Time-Domain Method, 3rd ed. (Artech House, 2005).