1. Introduction

In the past decades, random lasing, originally proposed by Letokhov in 1968, has been extensively investigated both in experiment and theory for its intrinsic interest and practical applications. So far, random lasing action has been found in many kinds of media such as inorganic semiconductors, organic molecules, biological systems and cold atoms, which will benefit the development of laser devices [1–4]. Quite different from conventional lasing, random lasing necessitates no mirror cavities for supporting essential feedback in the laser system. The random lasing process involves disorder-induced multiple scattering of light in the random medium. When the optical scattering forms a close-loop and the gain exceeds the loss, random lasing can be achieved. The medium for the occurrence of random lasing should include high-gain materials and efficient light scattering centers. Among these kinds of lasing materials, one of the most well-known materials is ZnO due to its unique optical properties and diversified morphologies of nanostructures.

ZnO, with a wide bandgap of 3.37 eV and a high exciton binding energy of 60 meV, acts as a glittering star for making ultraviolet light-emitting diodes and laser devices at room temperature. In the field of laser, to achieve low-threshold stimulated emission and more efficient radiative recombination, ZnO possessing the large exciton binding energy and diversified nanostructures will be one of the top choices [5]. According to different types of resonators, the lasing action of ZnO nano/micro structures can be realized by Fabry-Perot cavities, whispering-gallery-mode cavities and random cavities. Two well-faceted hexagonal end planes and six surrounding facets of the nano/micro structure ZnO act as natural reflective mirrors which are usually utilized to support an optical gain in a laser system. In addition, a high refractive index of ZnO (2.45) will cause more efficient total internal reflection in the nano/micro structure ZnO. With these advantages, ultraviolet Fabry-Perot, whispering-gallery-mode and ransom lasings from ZnO nanostructures have been successfully demonstrated [6–15].

In this study, we provide a novel method to fabricate self-organized comb-like ZnO　nanostructures. To the best of our knowledge, for most published reports ZnO nanocombs were synthesized at the temperature of 800~1450°C, which is much higher than the endurable temperature of indium tin oxide glass [11,16–19]. Therefore, the high growth temperature of ZnO nanocombs will be an obstacle on the development of electrically pumped ZnO laser. The growth temperature in our method is around 620°C and ZnO nanocombs can be successfully fabricated on indium tin oxide glass.　Additionally, we have investigated the lasing action from an ensemble of self-organized ZnO nanocombs. Our results reveal that Fabry-Perot cavity resonance provided by branches of nanocombs can be used to enhance random lasing action, which enables the occurrence of lasing action at low threshold pump intensity. Quite different from other ZnO-based random lasing spectra which usually include a wide background spectrum (spontaneous emission) and many sharp lasing peaks (stimulated emission), the lasing spectrum from our ZnO nanocombs exhibits a small number of sharp peaks with a narrow background spectrum having a full width at half maximum (FWHM) of about 5 nm due to the effect of Fabry-Perot cavity resonance. We therefore believe that the simultaneous occurrence of Fabry-Perot resonance and random laser action could open up a route for the creation of new optoelectronic devices with high efficiency.

2. Experiment

A Zn seed layer was deposited on one half side of an indium tin oxide glass substrate by using the sputtering system (JFC-1600, JEOL, Tokyo, Japan). The thickness of the seed layer was about 150 nm. We then used hydrothermal (HT) method to grow ZnO nanowires on the Zn-coated indium tin oxide glass substrate by mixing 0.05 M zinc nitrate hexahydrate Zn(NO₃)₂.6H₂O and 0.05 M hexamethylenetetramine C₆H₁₂N₄ (HMT) aqueous solution at 90°C for 4 hours.

To fabricate ZnO nanocombs by using vapor-solid (VS) growth mechanism, the as-prepared HT-ZnO nanowires/indium tin oxide glass substrate was put on the top of alumina boat loaded with high purity Zn powder (99.99%) and later the whole alumina boat was placed at the center of a tube furnace. Subsequently, the reaction chamber was evacuated and kept at a pressure of 10 Torr when argon and oxygen with high purity of 99.9% were introduced into the reaction chamber at a flow rate of 200 sccm and 5 sccm, respectively. In addition, the growth temperature was maintained at 620°C and dwell time was one hour. The procedure of fabricating ZnO nanocombs has been shown in Fig. 1 .

 

Fig. 1 Schematic illustration of the procedure for the fabrication of ZnO nanocombs. Inset: SEM image of ZnO nanocombs on the interface between VS-ZnO nanowires and HT-ZnO nanowires.

Download Full Size | PDF

The morphology of self-organized ZnO nanocombs was characterized by scanning electron microscopy (SEM) (JSM 6500, JEOL). The length of each nanocomb was a few tens of micrometers and nanowire branches with the length of about micrometers are uniformly distributed at one side of the stem. The cathodoluminescence mapping images were carried out on the same SEM instrument equipped with Gatan-Mono-CL3 operating at 10 kV. To investigate random lasing action, an ensemble of ZnO nanocombs is optically excited by a Q-switched Nd: yttrium aluminum garnet laser (266 nm, 3–5 ns pulse, 10 Hz) focused to a beam size about 200 μm in diameter. All measurements were performed at room temperature.

3. Results and discussions

According to the SEM image shown in Fig. 1, ZnO nanocombs mainly emerge on the interface between HT-ZnO nanowires and VS-ZnO nanowires. Although the exact reason is not clear, it might be related to the locally chaotic airflow resulting from the abruptly prominent HT-ZnO nanowires. The locally chaotic airflow may facilitate the steps of the formation of ZnO nanocombs [18]. It should be noted that there are no ZnO nanocombs found when pure indium tin oxide glass rather than HT-ZnO nanowires/indium tin oxide glass is adopted as a collecting substrate.

Figures 2 (a) and (b) show the evolution of the emission spectra of an ensemble of ZnO nanocombs with around 1.66 μm nanowire branches with increasing pump energy. At low excitation energy, a single broad spontaneous emission spectrum can be found. As the pump energy increases, the emission spectrum becomes narrower and several sharp peaks with linewidth less than 0.3 nm emerge. The occurrence of stimulated emission and the lasing threshold can be obtained by the analysis of the dependence of the emission intensity on the pump energy. As shown in Fig. 3 , there is an abrupt change of slope indicating the occurrence of stimulated emission with the lasing threshold of 69 μJ.

 

Fig. 2 Photoluminescence (PL) spectra of an ensemble of ZnO nanocombs with the average width of about 1.66 μm under (a) lower pump energy and (b) higher pump energy. (c) Three successive measurements of PL spectra of the same ensemble of ZnO nanocombs under pump energy of 140 μJ. PL spectra of (d) VS-ZnO nanowires and (e) another ensemble of ZnO nanocombs with the average width of about 2.47 μm under various pump energy.

Download Full Size | PDF

 

Fig. 3 Emission peak intensity as a function of pump energy.

Download Full Size | PDF

Figure 2 (c) shows a series of emission spectra taken at three successive measurements for the ensemble of ZnO nanocombs illuminated with 266 nm pulsed laser of 140 μJ. We can clearly see that the position, number and intensity of sharp peaks change randomly. According to the definition of random lasing, lasing is achieved when specific frequency of light is multiply scattered and amplified by stimulated emission. It is believed that in a random cavity, close-loop varies every moment and thus causes different lasing peak position and intensity. The random change of the sharp emission spikes in Fig. 2 (c) from one excitation to another supports that the lasing behavior here belongs to random lasing action. When the focused 266 nm laser is moved to an ensemble of VS-ZnO nanowires on the same substrate, there is no lasing behavior. We only observe a single broad spontaneous emission spectrum with a peak at around 385 nm and a FWHM of about 15 nm as shown in Fig. 2 (d).

It is believed that ZnO nanocombs have several advantages over ZnO nanowires. First, according to the SEM image in Fig. 4 , the distribution of ZnO nanocombs is much more disorderly than that of VS-ZnO nanowires. Second, the nanowire branches of ZnO naoncombs have a better Fabry-Perot resonant efficiency than VS-ZnO nanowires. While the resonant cavity is in direct contact with the substrate, greater index matching will lead to higher cavity loss due to low reflection occurring at the ZnO/substrate end facet [10]. As we can see from the SEM images, the nanowire branches of ZnO nanocombs are surrounded by air, and the two ZnO/air end facets provide higher reflection and result in a better Fabry-Perot resonant efficiency. Third, for ZnO nanocombs, Fabry-Perot cavities emit the resonant light in all directions due to various orientation of nanowire branches for various ZnO nanocombs. In this way, the resonant light can be amplified again by random lasing mechanism (multiple scattering of light between ZnO nanocombs). However, for VS-ZnO nnaowires, Fabry-Perot cavities emit the resonant light along a limited direction (almost perpendicular to the substrate) and thus the possibility for amplifying the resonant light by random lasing mechanism is quite low. Therefore, it is believed that the specific geometry and random distribution of ZnO nanocombs can promote Fabry-Perot resonance and random lasing actions.

 

Fig. 4 (a) SEM image, (b) cathodoluminescence mapping image and (c) enlargement of cathodoluminescenc mapping image of ZnO nanocombs.

Download Full Size | PDF

Further examining the random lasing spectra in Fig. 2 (b), it is found that the background emission centered at 387 nm exhibits a narrow FWHM of about 5 nm. This narrow linewidth of background emission has never been found in the published reports on the random lasing spectra arising from ZnO nanostructures [13–15,20–24]. In order to exactly understand the mechanism in our lasing system, we prepare another ensemble of ZnO nanocombs with about 2.47 μm nanowire branches in length. Quite interestingly, the background emission peak redshifts to 390 nm and possesses a similar narrow FWHM of about 3 nm as we can see from Fig. 2 (e).

To understand the above interesting behavior of random lasing, we consider the possibility that Fabry-Perot resonance and random lasing occur simultaneously. The nanowire branch of ZnO nonocomb forms a natural optical cavity, which consists of well-organized two-end facets. The Fabry-Perot resonance mode can be described by the equation:

The resonant mode spacing for the cavity length of 1.66 μm (or 2.47 μm) is around 18 nm (or 13 nm). Therefore, it can be deduced that only one single Fabry-Perot mode can exist in the ultraviolet photoluminescence range of ZnO nanomaterials for each of the two cases.

To further support the occurrence of Fabry-Perot resonance, we have performed cathodoluminescence mapping images as shown in Fig. 4. Figure 4 (a) shows the SEM image of an ensemble of ZnO nanocombs grown by VS method, and Figs. 4 (b) and (c) show the corresponding cathodoluminescence mapping images. The emission light of 387 nm is selected as the mapping wavelength. From Figs. 4 (b) and (c), it is found that the very pronounced bright spots occur on the two ends of each nanowire branch of ZnO nanocombs, which means the light of 387 nm is amplified by the cavity of the nanowire branch and escapes from two end facets of the cavity.

Finally, we stress the possible process that Fabry-Perot resonance can be used to enhance random lasing action. It is believed that the ultraviolet light is amplified by Fabry-Perot cavity inside the nanowire branch of ZnO nanocombs and then escapes from two end facets of the cavity. Afterward the light is once again amplified by random cavities formed between random distributed ZnO nanocombs. Random lasing process involves disorder-induced multiple scattering of light in the random medium. When the optical scattering forms a close-loop and the gain exceeds the loss, random lasing can be achieved. Additionally, it necessitates enough scattering photons in the random gain medium to meet the condition that gain exceeds the loss. Fabry-Perot cavity can amplify the ultraviolet light and thus increase the photon density in the ambience of ZnO nanocombs. Therefore, Fabry-Perot cavity will benefit the random lasing action and equivalently reduce the threshold pump energy.

4. Conclusions

In conclusion, we have provided a useful and reproducible method to fabricate ZnO nanocombs and investigated the lasing action. Our results demonstrate that the occurrence of Fabry-Perot resonance can be used to enhance the random lasing action with a reduced threshold condition. Besides, the random lasing spectra possess sharp lasing peaks with a FWHM of less than 0.3 nm and a narrow background emission with a FWHM of about 5 nm. It is therefore believed that the unique lasing behavior of ZnO nanocombs caused by the combination of Fabry-Perot resonance and random laser action shown here may provide a new route for the development of highly efficient light emitting devices.