We have comparatively investigated electrically pumped random lasing (RL) actions of two metal-insulator-semiconductor structured devices using pure ZnO and Zn2TiO4-nanoparticle-incorporated ZnO films as the semiconductor components i.e. light-emitting layers, respectively. It is demonstrated that the device using the Zn2TiO4-nanoparticle-incorporated ZnO film as the light-emitting layer exhibits a much smaller threshold current for the electrically pumped RL, which is ascribed to the enhanced multiple light scattering by incorporation of Zn2TiO4 nanoparticles into ZnO film. It is believed that this work provides a strategy for developing low-threshold ZnO-based random lasers.
©2011 Optical Society of America
Random lasing (RL) in disorder media is of significance for not only fundamental research but also technological applications [1–3]. In the past four decades, great research efforts have been expended in the field of RL [4–15]. As for the experimental studies of optically pumped RL, two works are worthy to be highlighted. Lawandy et al. reported the lasing action in a laser dye dispersed in a colloidal suspension of TiO2 particles, in which the dye and TiO2 particles played roles in amplifying and light scattering, respectively . Cao et al. demonstrated RL from ZnO polycrystalline powders, which exerted the bi-function of amplifying and scattering . From the technological point of view, electrical pumping is the prerequisite for extensive application of RL. Yu et al. realized electrically pumped RL from ZnO nanoparticle clusters in SiO2 films formed by spin-coating . Subsequently, we achieved the electrically pumped RL from ZnO polycrystalline films by means of metal-insulator-semiconductor (MIS) structures [9–11]. Afterwards, several research groups reported other device structures to achieve the electrically pumped RL from ZnO polycrystalline films [12–15]. Obviously, the above-mentioned works shed light on the practical application of ZnO-based random lasers.
As is the case for conventional lasers, low-threshold is an important pursuit for random lasers. It was reported that the threshold for RL can be decreased by enhancing multiple light scattering . Therefore, enhancing the multiple light scattering in ZnO polycrystalline films is a viable strategy to develop the low-threshold electrically pumped ZnO-based random lasers. The light scattering in ZnO polycrystalline films is ascribed to the refractive-index-difference (Δnr) between the grains and the grain boundaries . In principle, larger Δnr will lead to enhanced light scattering . Although the Δnr between the grains and the grain boundaries within ZnO films could be increased by altering the fabrication conditions or post-annealing schemes for ZnO films, the resulting Δnr is still limited. In this work, second-phase particles acting as the light scatterers are deliberately incorporated into ZnO films. In this case, the growth of ZnO grains is retarded thus resulting in smaller grains. Moreover, there is Δnr between the second-phase particles and ZnO grains. Such two effects arising from the incorporation of second-phase particles lead to enhanced multiple light scattering in ZnO polycrystalline films. In order not to bring about adverse effect on the electrically pumped RL from ZnO, the second-phase particles should be with a desirably low content in a ZnO film and will not substantially absorb the near-band-edge (NBE) emission of ZnO. Although TiO2 particles are strongly scattering media, they are not desirable to be incorporated into ZnO films because the TiO2 particles will absorb the NBE emission of ZnO, due to that TiO2 possesses a somewhat narrower bandgap than ZnO. Actually, ZnO films are often annealed at relatively high temperatures (e.g. 700 °C and above) to enhance the NBE emission. In this context, the TiO2 particles cannot survive in ZnO films due to the reaction between TiO2 and ZnO. In this work, to enhance the multiple light scattering in ZnO films, Zn2TiO4 nanoparticles are introduced into ZnO films by reactive DC sputtering and subsequent annealing at 700°C. For the sake of description, ZnO films incorporated with the Zn2TiO4 nanoparticles are denoted as ZnO/Zn2TiO4 films hereafter. It has been reported that Zn2TiO4 has a wider bandgap and a larger nr than ZnO [19,20]. Consequently, the Zn2TiO4 nanoparticles incorporated into ZnO films can enhance multiple light scattering but not absorb the NBE emission of ZnO. Significantly, with the same MIS structure, the device using ZnO/Zn2TiO4 film as the semiconductor component (light-emitting layer) exhibits a much lower threshold current for the electrically pumped RL than that using ZnO film as the semiconductor component. Moreover, the former outputs higher optical power than the latter at the same injection current. In a word, the highlight of this work is to improve the performances of the electrically pumped ZnO-based random lasers through modification of materials rather than design of new device structures.
2. Experimental details
Pure ZnO and Ti-containing ZnO films were prepared by reactive sputtering on <100>-oriented heavily arsenic-doped silicon slices sized in 2 × 2 cm2. Herein, one 4N pure Zn target (3 inch. in diameter) and another one inlaid with three 4N pure Ti strips were used for the sputtering. In the Ti-inlaid Zn target, the three Ti strips covered one eighth of the whole target. For the reactive sputtering, the chamber was firstly evacuated to 10−3 Pa. Then argon and oxygen gases with a flow rate ratio of 2:1 were inlet into the chamber to reach a working pressure of 8 Pa. The sputtering power was 100 W and the substrate temperature was kept at 300°C. With appropriate sputtering times, both ZnO and Ti-containing ZnO films reached a thickness of ~150 nm. The sputtered films were subsequently annealed at 700°C for 2 h in oxygen ambient. As a result, the Ti-containing ZnO films were transformed into Zn2TiO4-nanoparticle-incorporated ZnO films. Sequentially, ~50 nm thick SiO2 films were deposited on the above-mentioned two kinds of ZnO films by a sol-gel process described elsewhere . Herein, the SiO2 films were annealed at 550°C for 1 h in air. Finally, ~30 and 200 nm thick circular Au films with a diameter of 1 cm were sputtered onto SiO2 films and the backsides of silicon substrates, respectively, as the electrodes of the MIS devices.
The two kinds of ZnO films were analyzed using powder XRD (Rigaku D/max 2550-pc) with Cu Kα radiation, SEM (JEOL JSM-7100F), AFM (SPA 400, SII Nanotechnologies, Japan.) and HRTEM (JEOL JEM-2100F oxford inca). PL of ZnO films was measured at RT with an excitation source at 325 nm provided by a He-Cd laser. The RT EL spectra for ZnO-based MIS devices were measured under forward DC bias where the positive voltage was connected to the front Au electrode of the device. Herein, the spectra were acquired with an Acton spectraPro 2500i spectrometer with a lowest spectrum resolution of 0.5 Å and an accuracy of ± 2 Å. For the acquisition of EL spectra, the scanning step size was 1 Å. Moreover, the output optical powers of the MIS devices were measured using a Newport 1931-C power meter equipped with an 818-UV/DB detector (~1 cm in diameter). For this measurement, the device was in face of the detector and the distance between the two was ~2 cm. For such a measurement configuration, it was roughly estimated that only ~2% of the output optical power of the device was detected by the above-mentioned power meter.
3. Results and discussion
Figure 1(a) shows the X-ray diffraction (XRD) patterns of the films sputtered from the Zn target and the Ti-inlaid Zn target, respectively. The two films were annealed at 700°C for 2 h under O2 ambient. In each XRD pattern, there are two peaks that can be indexed to (002) and (103) planes of hexagonal ZnO. While, in the XRD pattern of the film sputtered from the Ti-inlaid Zn target, accompanying the pronounced ZnO (002) peak there is an obvious shoulder peak, which can be indexed to (311) plane of cubic Zn2TiO4. Herein, Zn2TiO4 is believed to be a product of reaction between ZnO and TiOx (x≤2) during the annealing at 700°C. Therefore, it is derived that the second-phase particles of Zn2TiO4 have been incorporated into ZnO film sputtered from the Ti-inlaid Zn target, thus forming the above-mentioned ZnO/Zn2TiO4 film. The incorporation of Zn2TiO4 particles into ZnO film will be further verified by high resolution transmission electron microscopy (HRTEM) characterization. Figures 1(b) and 1(c) show the representative field-emission scanning electron microscopy (FESEM) images for ZnO and ZnO/Zn2TiO4 films, respectively. Obviously, the grains in ZnO/Zn2TiO4 film are much smaller than those in ZnO film. It is believed that the incorporated Zn2TiO4 particles hinder the growth of ZnO crystal grains. Note that the Zn2TiO4 particles are too small to be discernable in the FESEM image. The elemental maps of Zn and Ti in ZnO/Zn2TiO4 film measured by energy dispersive X-ray spectroscopy (EDS) are shown in Figs. 1(d) and 1(e), respectively. As can be seen, the Ti element distributes extensively in the film, implying that the Zn2TiO4 particles pervade the film. Moreover, EDS analysis indicates that the molar ratio of Ti and Zn is ~1:20 in ZnO/Zn2TiO4 film.
The microstructure, crystallinity and composition of ZnO/Zn2TiO4 film have been further characterized by HRTEM and EDS analysis. A typical HRTEM image of ZnO/Zn2TiO4 film grown on a silicon substrate is displayed in Fig. 2 . As can be seen, there is a ~5 nm thick amorphous layer, which is generally believed to be of SiOx (x≤2), between the silicon substrate and the film. Moreover, the HRTEM image illustrates the grain-separated polycrystalline structure of ZnO/Zn2TiO4 film. For example, the areas denoted (1) and (2) exhibit the lattice fringes of ZnO, as shown in Fig. 2(a) (1)-(2). With a scrutiny of the HRTEM image, the areas denoted (3) and (4) are found to display lattice fringes different to those of ZnO. As revealed in Fig. 2(a) (3)-(4), the areas (3) and (4) actually correspond to the particles of Zn2TiO4. It should be mentioned that the Zn2TiO4 particles can be only discernible in the HRTEM image because their sizes are generally smaller than 20 nm. The EDS spectrum acquired around area (5) is shown in Fig. 2(a) (5), where the elements of Zn, Ti and O existing in the film are revealed. Besides, the signals of Cu and Si in the EDS spectrum come from the Cu grid used for TEM observation and the above-mentioned SiOx layer.
Moreover, note the HRTEM image that multiple lattice fringes are interwoven in quite a few areas. In order to decouple the lattice fringes corresponding to ZnO and Zn2TiO4, respectively, we derived the fast Fourier transform (FFT) pattern, as shown in Fig. 2(b), from the HRTEM image. Then, addressing the spots indexed as ZnO (002), (103) and (101) planes, a successive FFT was performed to reproduce the lattice fringes corresponding to the above-mentioned planes of ZnO, which is demonstrated in Fig. 2(c). Likely, the lattice fringes for Zn2TiO4 (311) plane, as shown in Fig. 2(d), was derived. As can be derived from the comparison between Figs. 2(c) and 2(d), overall, the Zn2TiO4 nanoparticles distribute around the ZnO grain.
Figure 3(a) shows the room-temperature (RT) photoluminescence (PL) spectra for ZnO and ZnO/Zn2TiO4 films. The two PL spectra are quite similar in shape and exhibit emission peaks at nearly the same wavelengths. This indicates that the incorporation of Zn2TiO4 nanoparticles hardly affects the luminescence of ZnO. Figures 3(b) and 3(c) illustrate the evolution of RT electroluminescence (EL) spectra with increasing forward bias voltage/current for the two MIS devices with ZnO and ZnO/Zn2TiO4 films as the light-emitting layers, respectively. All the EL spectra feature a number of discrete sharp peaks with line-width less than 5 Å. Overall, the number of sharp peaks and the integrated emission intensity increase with the forward current. According to our previous reports [9–11], the discrete sharp peaks in the EL spectra can be ascribed to the electrically pumped RL from ZnO. Moreover, Figs. 3(b) and 3(c) show that the MIS device with the light-emitting ZnO/Zn2TiO4 film can be electrically pumped into RL at much smaller injection current. This fact can be further verified in Fig. 3(d), which shows the dependences of the detected output power on the injection current for the two MIS devices as mentioned above. For each device, the detected output power increases more rapidly with increasing injection current above a threshold, which is characteristics of lasing action. Strikingly, the RL threshold current for the MIS device with the light-emitting ZnO/Zn2TiO4 film is only ~6 mA, much smaller than that (~18 mA) for the counterpart with the light-emitting ZnO film. Moreover, at the same injection current, the MIS device with the light-emitting ZnO/Zn2TiO4 film outputs a higher optical power. Accordingly, in terms of the electrically pumped RL, ZnO/Zn2TiO4 film is superior to ZnO one. In the following, the mechanism underlying the results as mentioned above will be tentatively elucidated.
Optical gain is the first condition for RL. Therefore, it is necessary to clarify how the optical gain is achieved in ZnO-based MIS device. The energy band structure of ZnO-based MIS device applied with sufficiently high forward bias is schematically shown in Fig. 4(a) . As can be seen, the energy band of ZnO bends downward in the region adjacent to SiO2 layer (denoted as ‘region L’ in Fig. 4(a), herein, the word ‘L’ is the abbreviation of ‘lasing’), where electrons accumulate. Consequently, in this region the concentration of electrons is very high so that the quasi- Fermi-level of electrons (EFn) enters into the conduction band of ZnO. On the other hand, considerable amount of electrons in the valence band of ZnO are driven into the electron traps in the SiO2 layer by the electric field, thus resulting in an equivalent amount of holes in the valence band of ZnO. It should be mentioned that the sol-gel derived SiO2 layer in this work was annealed at a considerably low temperature (550°C). In this case, such a SiO2 layer is considerably defective, that is, numerous electron traps exist in the SiO2 layer, which favors for the injection of holes into ZnO. Under sufficiently large forward bias, the concentration of holes in the ‘region L’ becomes sufficiently high so that the quasi-Fermi-level of holes (EFp) is close to (case 1) or even below (case 2) the valence band edge (Ev) of ZnO. Therefore, as shown in Fig. 4(b), EFn - EFp > Eg (bandgap energy of ZnO) is satisfied for stimulated emission and therefore optical gain in the ‘region L’. It should be pointed out that the optical gain factor (g) is proportional to the value of EFn – EFp., which increases with the injection current . Figure 4(c) shows the schematic diagram for the increase of maximum optical gain factor (gmax) in the gain spectrum with the injection current.
Multiple light scattering is another necessary condition for RL. Both ZnO and ZnO/Zn2TiO4 films prepared in this work are polycrystalline in nature. Therefore, light is naturally subjected to multiple scattering in the course of propagation within the above-mentioned films. By comparison, the multiple light scattering in ZnO/Zn2TiO4 film is stronger than that in ZnO film. This is ascribed to the following two reasons. First, as mentioned above, the grains in ZnO/Zn2TiO4 film are much smaller. Accordingly, the light is much more frequently scattered during the propagation within ZnO/Zn2TiO4 film. Second, due to the incorporation of Zn2TiO4 nanoparticles, the spatial nr fluctuation in ZnO/Zn2TiO4 film is much more remarkable so that the light scattering is enhanced. It should be noted that, as described above, the stimulated emission in the MIS devices only occurs in the ‘region L’, which is actually an accumulation layer of electrons (majority carriers in ZnO). Generally, the accumulation layer of the majority carriers in semiconductor can be regarded as a quasi 2D system. Accordingly, the stimulated light in ZnO-based MIS devices is subjected to multiple scattering in a quasi 2D system. Apalkov et al. have theoretically suggested that in a 2D system the random variations of nr could lead by chance to waveguiding structures, thus forming resonators. Moreover, the coherent multiple light scattering in the resonators can provide trapping . Obviously, the trapping of light in an amplifying medium will continuously pick up optical gain.
RL results from concurrence of the optical gain and the random resonators. As mentioned above, the formation of random resonators is pertaining to the multiple light scattering. It is thus understandable that the optical gain necessary for RL is remarkably dependent on the strength of multiple light scattering. J. Fallert et al. found that in the case of optically pumped RL from ZnO powders the extended modes related to modest light scattering were dominant in the spectral region where the gain was highest, and the localized modes resulting from strong light scattering prevailed in the low-gain regions . Apalkov et al. quantitatively predicted the decrease of the threshold for RL upon adding disorder to a medium . Note that the addition of disorder will enhance light scattering in the medium. Very recently, we have proved that the optical gain for the electrically pumped RL from ZnO films is decreased by the enhanced light scattering . In view of the above-mentioned results, it is reasonable that the optical gain for RL is smaller for ZnO/Zn2TiO4 film than for ZnO film because ZnO/Zn2TiO4 film has a stronger light scattering capability, as has been derived earlier. Furthermore, according to Fig. 4(c), it can be deduced that the threshold current for the electrically pumped RL from the MIS device using the light-emitting ZnO/Zn2TiO4 film is smaller. Actually, this deduction has been experimentally verified in Fig. 3(d).
In summary, we have comparatively investigated the electrically pumped RL actions of the two MIS devices using ZnO/Zn2TiO4 and ZnO films as the light-emitting layers, respectively. It is demonstrated that the MIS device with the light-emitting ZnO/Zn2TiO4 film exhibits a much smaller threshold current for the electrically pumped RL. The incorporation of Zn2TiO4 nanoparticles into ZnO film leads to much smaller ZnO grain sizes. Moreover, it introduces larger spatial nr fluctuation throughout the film. Such two effects result in the enhanced multiple scattering of the stimulated light emitted from ZnO. In this context, the threshold optical gain for RL is decreased. In turn, the threshold current for the electrically pumped RL from ZnO/Zn2TiO4 film becomes smaller. In a word, this work has provided a case in developing low-threshold ZnO-based random lasers through modification of materials.
The authors would like to thank the financial supports from Zhejiang provincial Natural Science Fund (No. R4090055), “973 Program” (No. 2007CB613403), National Natural Science Foundation of China (NSFC) (No. 60906024), and the foundation of 2008DFR50250 of MOST.
References and links
1. D. S. Wiersma, “The physics and applications of random lasers,” Nat. Phys. 4(5), 359–367 (2008). [CrossRef]
2. H. Cao, “Lasing in random media,” Waves Random Media 13(3), R1–R39 (2003). [CrossRef]
3. H. Cao, “Review on latest developments in random lasers with coherent feedback,” J. Phys. Math. Gen. 38(49), 10497–10535 (2005). [CrossRef]
4. V. M. Markushev, V. F. Zolin, and Ch. M. Briskina, “Powder laser,” Zh. Prikl. Spektrosk. 45, 847–850 (1986).
5. C. Gouedard, D. Husson, C. Sauteret, F. Auzel, and A. Migus, “Generation of spatially incoherent short pulses in laser-pumped neodymium stoichiometric crystals and powders,” J. Opt. Soc. Am. B 10(12), 2358–2362 (1993). [CrossRef]
6. 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]
7. H. Cao, Y. G. Zhao, S. T. Ho, E. W. Seelig, Q. H. Wang, and R. P. H. Chang, “Random laser action in semiconductor power,” Phys. Rev. Lett. 82(11), 2278–2281 (1999). [CrossRef]
8. E. S. P. Leong and S. F. Yu, “UV random lasing action in p-SiC(4H)/i-ZnO-SiO2 nanocomposite/n-ZnO:Al heterojunction diodes,” Adv. Mater. (Deerfield Beach Fla.) 18(13), 1685–1688 (2006). [CrossRef]
9. X. Y. Ma, P. L. Chen, D. S. Li, Y. Y. Zhang, and D. R. Yang, “Electrically pumped ZnO film ultraviolet random lasers on silicon substrate,” Appl. Phys. Lett. 91(25), 251109 (2007). [CrossRef]
10. X. Y. Ma, J. W. Pan, P. L. Chen, D. S. Li, H. Zhang, Y. Yang, and D. R. Yang, “Room temperature electrically pumped ultraviolet random lasing from ZnO nanorod arrays on Si,” Opt. Express 17(16), 14426–14433 (2009). [CrossRef] [PubMed]
11. Y. Tian, X. Y. Ma, P. L. Chen, Y. Y. Zhang, and D. R. Yang, “Electrically pumped wavelength-tunable ultraviolet random lasing from Mg(x)Zn(1-x)O films on Si,” Opt. Express 18(10), 10668–10673 (2010). [CrossRef] [PubMed]
12. S. Chu, M. Olmedo, Z. Yang, J. Y. Kong, and J. L. Liu, “Electrically pumped ultraviolet ZnO diode lasers on Si,” Appl. Phys. Lett. 93(18), 181106 (2008). [CrossRef]
13. H. Zhu, C. X. Shan, J. Y. Zhang, Z. Z. Zhang, B. H. Li, D. X. Zhao, B. Yao, D. Z. Shen, X. W. Fan, Z. K. Tang, X. H. Hou, and K. L. Choy, “Low-threshold electrically pumped random lasers,” Adv. Mater. (Deerfield Beach Fla.) 22(16), 1877–1881 (2010). [CrossRef]
14. H. K. Liang, S. F. Yu, and H. Y. Yang, “Directional and controllable edge-emitting ZnO ultraviolet random laser diodes,” Appl. Phys. Lett. 96(10), 101116 (2010). [CrossRef]
15. H. K. Liang, S. F. Yu, and H. Y. Yang, “ZnO random laser diode arrays for stable single-mode operation at high power,” Appl. Phys. Lett. 97(24), 241107 (2010). [CrossRef]
17. H. C. Ong, J. Y. Dai, A. S. K. Li, G. T. Du, R. P. H. Chang, and S. T. Ho, “Effect of a microstructure on the formation of self-assembled laser cavities in polycrystalline ZnO,” J. Appl. Phys. 90(3), 1663–1665 (2001). [CrossRef]
18. F. J. P. Schuurmans, M. Megens, D. Vanmaekelbergh, and A. Lagendijk, “Light scattering near the localization transition in macroporous GaP networks,” Phys. Rev. Lett. 83(11), 2183–2186 (1999). [CrossRef]
19. S. A. Mayén-Hernández, G. Torres-Delgado, R. Castanedo-Pérez, M. G. Villarreal, A. Cruz-Orea, J. G. M. Alvarez, and O. Zelaya-Angel, “Optical and structural properties of ZnO + Zn2TiO4 thin films prepared by the sol–gel method,” J. Mater. Sci. Mater. Electron. 18(11), 1127–1130 (2007). [CrossRef]
20. M. Ocana, W. P. Hsu, and E. Matijevic, “Preparation and properties of uniform-coated colloidal particles. 6. Titania on zinc oxide,” Langmuir 7(12), 2911–2916 (1991). [CrossRef]
21. P. L. Chen, X. Y. Ma, D. S. Li, Y. Y. Zhang, and D. R. Yang, “347 nm ultraviolet electroluminescence from MgxZn1-xO-based light emitting devices,” Appl. Phys. Lett. 90(25), 251115 (2007). [CrossRef]
22. T. Suhara, Semiconductor Laser Fundamentals (Marcel Dekker, Inc., 2004), equation (3.16) in Chap. 3.
23. J. Fallert, R. J. B. Dietz, J. Sartor, D. Schneider, C. Klingshirn, and H. Kalt, “Co-existence of strongly and weakly localized random laser modes,” Nat. Photonics 3(5), 279–282 (2009). [CrossRef]
24. Y. Tian, X. Y. Ma, L. Jin, and D. R. Yang, “Electrically pumped ultraviolet random lasing from ZnO films: Compensation between optical gain and light scattering,” Appl. Phys. Lett. 97(25), 251115 (2010). [CrossRef]