1. Introduction

As one of the most important technologies in our daily life, laser has been extensively used in many aspects, ranging from science and technology, medical, industry, to communications [1]. The conventional lasers require precise resonator made of two-settled mirrors. The customized components and meticulous fabrications of conventional lasers make its output emission has high spatial coherence, which is adverse for the speckle-free image. However, in comparison with conventional lasers, random lasers with small-size exhibit laser-level intensity with low spatial coherence, two properties that have traditionally been mutually exclusive in conventional light sources [2]. So random laser, as a new kind of ideal light source, has brought many opportunities in the fields of full-field imaging, biological medicine, and light-emitting diodes [3]. Random laser achieve feedback through multiple scattering of light provided by the disordered structures embedded with an optically active element. In pioneering experiments, a smooth, single-peaked emission spectrum is observed, which is dubbed as random laser with incoherent feedback and is explained by the diffusion framework neglecting the interference of light in random systems [4]. However, the random laser emission spectrum with sharp peak is regarded as random laser with resonant feedback and can only be explained by framework involved with interference, such as localized modes [5–6]. Random lasers have been reported across different disordered gain-scattering systems, such as dye doped with semiconductor powders [7–10], polymer films [11], metallic nanostructures [12–14], organic dye-doped gel films [15], liquid-crystal-based media [16] and dye-infiltrated opals [17]. Besides, perovskite quantum dot as gemerging gain material with excellent luminescence properties are also used to generate random laser in form of thin film [18–20] and single crystals [21].

In recent years, metal nanoparticle is often used as an effective scatter in random-gain system due to its collective electron oscillations at the surface when it is excited by the external light. The collective electron oscillations are known as localized surface plasmon resonance (LSPR), which can be easily modified by changing the size and shape of the metal nanoparticles as well as the properties of the host medium [22]. In comparison with the conventional dielectric particles with the same-size, the metal nanoparticles have larger scattering cross section. However, conventional noble metals, such as gold and silver, also have some disadvantages. For example, they have very large magnitudes of real permittivity and quite high losses in the visible range. The nanofabrication and integration of the noble metals nanostructures are difficult, which result in the difficulty of tuning their optical properties. Recently, many researchers found that titanium nitride (TiN) is a possible candidate of noble metals for plasmonic devices due to its specialities of hard, abrasion-resistant as well as lower raw material cost in comparison with the noble metals [23]. Additionally, in the visible range, the magnitude of real permittivity of the titanium nitride is much smaller than that of the noble metal due to its smaller carrier concentration. The most important advantage of the titanium nitride is the flexible control of the optical properties by simply tuning manufacturing processes [24]. Besides, the titanium nitride has extreme thermal stability (melting point >2,700 °C), which makes it a good alternative to replace noble metals in optical metamaterials, thermal radiation engineering and thermophotovoltaic applications [25–26]. Our laboratory has initially found that titanium nitride nanoparticles have the property of enhancing random lasing [27], but there is no random lasing exploration on titanium nitride films. At present, the electrical characteristics of titanium nitride films have been deeply explored in the field of materials. However, there are relatively few experimental investigations on the optical properties of titanium nitride films.

In this paper, we report the effect of TiN film on the lasing properties of gain material for the first time. Random lasing from the Pyrromethene-597 (PM597)/Silicon and PM597/TiN/Silicon samples is measured. Results show that the scattering and the plasmon resonance (PR) of the TiN film plays a major role on the enhanced random laser behavior in the PM597/TiN/Silicon sample. Contour maps of shot-to-shot lasing spectra for PM597/TiN/Silicon sample obtained at various pump pulses and same pump energy confirms the influence of PR on the stability of lasing spectra. Simulations of the local electric field distributions for TiN film explain the fluctuation of the lasing spectra obtained at different positions of the PM597/TiN/Silicon sample. It is foreseeable that TiN films can be used as plasmon materials for random laser production. At the same time, this paper provides a means to explore the optical properties of titanium nitride films by analyzing random lasing spectra.

2. Experimental details

In the experiment, the titanium nitride (TiN) film were fabricated on the silicon substrates by Ion beam assisted deposition (IBAD) method, which including a sputtering source and an assisting source. The titanium atoms sputtered out from a 99.99% pure metal titanium target by Ar⁺ beam reacted with N₂ and deposited on the silicon substrates. The growing films were bombarded by an assisting Ar⁺ beam. In one hour, keeping the growth parameters invariant, the TiN film with a thinkness of 180 nm is achieved, which is characterized by a surface profile system (Veeco Dektak150).

Pyrromethene-597 (PM597) dispersed in ethanol with the concentration of 10⁻³ M was used as active material in the samples. The PM597/Silicon and PM597/TiN/Silicon samples were prepared by spin coating of the PM597 solution on the surface of the silicon substrate only and silicon substrate that supports the TiN film, respectively. The structure of PM597/TiN/Silicon sample is shown in Fig. 1(b). The surface morphology of the TiN film used in the experiment, imaged by atomic force microscopy (AFM), is depicted in Fig. 1(d). Figure 1(c) describes the absorption spectrum of TiN film.

 

Fig. 1. (a) Schematic diagram for the experimental setup of lasing measurement; (b) schematic illustration of configuration of PM597/TiN/Silicon; (c) The absorption spectra of TiN film; (d) The atomic force microscopy (AFM) micrograph of the TiN film.

Download Full Size | PDF

The samples were excited by a second-harmonic of a Q-switched Nd: YAG laser (532 nm wavelength, 10 HZ repetition rare and 8 ns pulse duration). As shown in the experimental setup (Fig. 1(a)), to record the pump energy, a polarizing beam splitter was used to split the beam into two parts, and an energy meter is used to measure the reflected beam. Through a half-wave plate, a cylindrical lens and an adjustable slit, the transmitted sub-beam was focused on the samples. We lastly used an optical fiber bundled into a spectrometer to collect emission from the sample’s surface.

3. Results and discussion

Figure 2(a) reveals the emission spectra of the random lasing for PM597/Silicon sample at various pump energies. The output emission intensity and the full width at a half maximum (FWHM) of laser peak as a function of input pumping energy intensity for the PM597/Silicon sample is shown in Fig. 2(b). A broad band centered at 603.9 nm with FWHM of 10 nm is found in Fig. 2(a) at a pump energy of 78.1 $\textrm{pJ}/{\mathrm{\mu}} {\textrm{m}^2}$, which is indicative of the amplified spontaneous emission (ASE) spectrum from the PM597 gain molecules. However, the FWHM of the emission spectrum reduces gradually as the pump energy increases. When the pump energy increases up to 82.0 $\textrm{pJ}/{\mathrm{\mu}} {\textrm{m}^2}$, a sharp peak at around 604.1 nm with a FWHM as narrow as 0.34 nm appears above the broad background spectrum. More importantly, for the pump energy larger than 82.0 $\textrm{pJ}/{\mathrm{\mu}} {\textrm{m}^2}$, the emission spectra exhibit more multiple sharp peaks with a FWHM less than 0.30 nm. A clear threshold behavior is observed in Fig. 2(b), where the emission intensity grows linearly with the pump energy below and above the threshold of 81.2 $\textrm{pJ}/{\mathrm{\mu}} {\textrm{m}^2}$. This phenomenon is a typical characteristic of laser emission. These results related to the sharp peaks suggest that the random lasing from the PM597/Silicon sample is contributed to the coherent feedback of photons. This coherent feedback may be provided by the multiple scattering of PM597 dye molecules randomly distribute on the silicon slice. Besides, defects at the interface between dye and the silicon substrate which have significantly different refractive index probably provide additional contributions for the coherent feedback [28]. Closed loop paths are formed when the photons are scattered in the sample, which act as the oscillation cavities of the random lasing [29]. When the gain reaches the loss as the increase of the pump energy, the laser oscillation occurs in the closed loop, resulting in a large number of sharp peaks in the emission spectra.

 

Fig. 2. (a) Spectra of the random lasing from PM597/Silicon sample at different pump energies; (b) The intensity and FWHM of the output laser for the PM597/Silicon sample as a function of pump energy.

Download Full Size | PDF

The evolution of emission spectra as a function of pump energy for PM597/TiN/Silicon sample is presented in Fig. 3(a). It only exhibits a broad ASE spectrum peaked at 602.5 nm with a FWHM of 8.5 nm when the pump energy is 64.6 $\textrm{pJ}/{\mathrm{\mu}} {\textrm{m}^2}$. The emission spectrum collapses to a narrow emission with a sharp peak centered around 602.6 nm and the FWHM of 0.23 nm when the pump energy is increased to 65.5 $\textrm{pJ}/{\mathrm{\mu}} {\textrm{m}^2}$. More and more sharp peaks emerge above the broad band as the pump energy is further increased, implying the occurrence of the coherent random lasing. Except the multiple scattering induced by the local inhomogeneity of gain medium, the scattering from the rough surface of the TiN film is also responsible for the coherent feedback. In addition, the plasmon resonance (PR) of the TiN film plays an indispensable effect on the random lasing. An overlap is found between the absorption spectrum of TiN film (see Fig. 1(c)) and the photoluminescence spectrum of PM597 (see Ref. [29]), which provides an important clue to the existence of energy transfer between TiN and PM597. This means that the dye molecules can transfer radiation energy to TiN. At the same time, TiN can directly absorb the pumping energy. These energies not only directly lead to the PR enhancement of TiN, but also promote the reabsorption of dye molecules, and finally promote random lasing [30]. So it is noted that the emission spectra in Fig. 2(a) and Fig. 3(a) have a big difference from each other, where the spectrum from the latter sample exhibits more sharp peaks. The intensity and the FWHM of the emission spectra as a function the pump energy for the PM597/TiN/Silicon sample is plotted in Fig. 3(b), where it shows a threshold of 64.6 $\textrm{pJ}/{\mathrm{\mu}} {\textrm{m}^2}$. The data provided in Fig. 2(b) and Fig. 3(b) are all consistent with random laser theories, which directly indicate that TiN lowers the threshold of random lasing.

 

Fig. 3. (a) Emission spectra versus pump energy for PM597/TiN/Silicon sample; (b) The intensity and FWHM of the emission spectra for the PM597/TiN/Silicon sample as a function of pump energy.

Download Full Size | PDF

As we know, the local electric field at the surface of TiN film is dependent on the incident electric field, the surface morphology of the film and the dielectric properties of the surrounding medium. The local electric field is low at the low pump energy, which has a weak influence on the lasing wavelength. However, larger local electric field is induced when the pump energy is high enough, which has a stronger effect on the lasing wavelength. Thus, we can see in Fig. 3(a) that the main emission wavelengths are stable when the pump energies are low (lower than 68.8 $\textrm{pJ}/{\mathrm{\mu}} {\textrm{m}^2}$), while they have small shifts with increasing the pump energy (larger than 68.8 $\textrm{pJ}/{\mathrm{\mu}} {\textrm{m}^2}$).

Figure 4(a) depicts the contour map of shot-to-shot lasing spectra at various pump pulses while keeping the same pump position and the pump energy of 69.8 $\textrm{pJ}/{\mathrm{\mu}} {\textrm{m}^2}$for the PM597/TiN/Silicon sample. Photobleaching effect of the random lasing is observed in Fig. 4(a) under the excitation of the multiple pump shots, however, the lasing main speaks are immobile. This attributes to the local electric stability arise from the PR effect of TiN film. Single shot lasing spectra obtained upon different pump positions and the same pump energy are depicted in Fig. 4(b), where the spectra are different from each other. This is owing to the different local field enhancement on the different positions of the TiN film. For an isolated particle on the surface of TiN film, as shown in Fig. 1(d), the field around it is dependent on its shape, composition and the dielectric properties of the surrounding medium. This field can be expressed as [31]

 

Fig. 4. (a) Contour map of shot-to-shot lasing spectra obtained at various pump pulses while keeping the same pump energy and pump position for the PM597/Silicon and PM597/TiN/Silicon. (b) Single shot lasing spectra obtained upon different pump positions and the same pump energy. (c) Spatial distribution of the electric intensity for an TiN film (similar to figure1(d)) with the thickness of 180 nm at the excitation wavelength of 532 nm.

Download Full Size | PDF

4. Conclusion

In summary, we presented a novel approach to enhance random lasing of dye molecules (PM597) based on TiN film. First, random laser emission from PM597/Silicon was observed, which is attributed to the multiple scattering of photons induced by PM597 dye molecules randomly distribute on the silicon slice. Second, enhanced random lasing (in comparison with that of the PM597/Silicon) from PM597/TiN/Silicon sample was found. This is closely related to the PR of the TiN film, which increases the energy transfer between TiN and PM597. In addition, the rough structure on the surface of the TiN film serves an efficient scattering medium, which increases the multiple scattering probability of the photon that involved in the random modes. Third, an approach to tune the random lasing, which is based on the change of pump position, was demonstrated. The demonstrated plasmon-enhanced random laser emission assisted by the TiN shows that TiN is a good alternative to replace noble metals in random lasing field, which is of guiding significance for studying plasmonic random lasers.