1. Introduction

Over the past decades, ultrafast lasers have been employed to study nonlinear processes [1], optical breakdown [2], surface micromachining [3], and this technique has been used in a broad range of applications, from waveguide fabrication to cell ablation [4]. Femtosecond lasers have attracted considerable interest for the physicochemical phenomena occurring within an ultra-short duration and the microscopically modification to various materials in recent years [5]. Various luminescence phenomena have been observed during and after the femtosecond laser irradiation. Multi-photon excited up-conversion (UC) luminescence [6,7], long-persistent luminescence (LPL) [8,9], photochemical spectral hole burning memory [10], space-selective permanent photo-reduction [11,12] and other important phenomena [13] have been advocated in rare earths ions doped materials via femtosecond laser pumping for a variety of applications in high-density optical storage, three dimensional display, optical communication, and medical imaging. The LPL evoked tremendous interests [14, 15], since it can be widely utilized in optical memory devices, photonic devices, display devices, and so on. Qiu et al. [16] suggested a mechanism of the LPL consisting of a thermo-stimulated recombination of holes and electrons at traps induced by the laser irradiation at room temperature. It is worth noting that various defects can be induced in glasses after the irradiation of a femtosecond laser, however, the formation and stabilization of defects were not clarified. More importantly, the application of those permanent defects induced via femtosecond laser irradiation should be carefully studied, which could provide new possibilities in the production of optical materials and devices with novel functions.

In this paper, we report the observation of the UC LPL in a Sm³⁺-doped 80B₂O₃-5Na₂O-15Al₂O₃ glass under femtosecond laser irradiation. It is demonstrated that the metastable trapping centers are induced under the femtosecond laser pumping with the thermo-luminescence (TL) technique, which contributes to the LPL. Furthermore, some permanent trapping centers can be refilled with ultraviolet (UV) excitation as experimentally observed, and it suggests a rewritten process could be realized with UV irradiation. The mechanism of the phenomena is illustrated and discussed in details.

2. Experimental

The chemical composition of the glass sample prepared was 80B₂O₃-5Na₂O-15Al₂O₃-0.1Sm₂O₃ (mol%). Reagent grade B₂O₃, Na₂CO₃, Al₂O₃ and Sm₂O₃ were used as the starting materials. Approximately 10g batch was mixed and melted in Pt crucibles in an electric furnace at 1400°C for 2 h in an ambient atmosphere. The melt was then quenched to obtain transparent glass. The glass was cut and polished for optical measurements.

A regeneratively amplified 800nm Ti: sapphire laser system with 1 kHz repetition rate and approximately 120fs pulse duration was used as an irradiation source. The laser beam was focused on the surface of the sample through optical lens in order to obtain higher power density. The focal spot being monitored in situ by using the confocal microscope with CCD. By choosing appropriate objective lens or optical lens and adjusting the power density of laser beam, the focal spot dimension can be varied down to few microns. The focal length is 100 mm, the focal spot size is 8mm*30um, and the power density is 12mJ/cm². The fluorescence spectra excited with femtosecond laser were measured in orthogonal configuration by a spectrophotometer of ZOLIX SBP300. The scanning rate of this spectrophotometer was 100nm/min. The photoluminescence spectra were measured on a Hitachi F-7000 fluorescence spectrophotometer with a Xe lamp as an excitation source. The absorption spectrum was measured on Hitachi UV-4100 spectrophotometer. In addition, LPL decay curves were measured with a PR305 long afterglow instrument. All the measurements were carried out at room temperature. The TL curves were measured with a FJ-427ATL meter (Beijing Nuclear Instrument Factory). Prior to the measurements, powder samples were first exposed to radiation by UV light (365nm) for about 20 min, then heated from room temperature to 524K with a rate of 1K/s.

3. Results and discussion

Figure 1(a) presents the emission spectra of the Sm³⁺-doped glass obtained separately by using an excitation wavelength of 402 nm from a Xenon lamp and 804 nm of the femtosecond laser focused by an optical lens. Under the excitation at 402 nm, the characteristic red emissions of Sm³⁺ located at 560, 596 and 650 nm are observed. The excitation spectrum of the glass monitored at 596 nm shown in Fig. 1(a) reveals that all the excitation peaks can be assigned to the 4f-4f transitions of Sm³⁺. The strongest excitation peak at 402 nm is assigned to the ⁶H_5/2→⁶P_3/2 transition of Sm³⁺ [17]. The same appearance of the emission spectra is recorded under the excitation of the focused femtosecond laser at 804 nm as shown in Fig. 1(a), which indicates that the characteristic emission originated from the same luminescence centers via an UC process.

 

Fig. 1 Excitation spectrum of the Sm³⁺-doped glass monitored at 596 nm and the emission spectra obtained separately by using an excitation wavelength of 402 nm from a Xenon lamp and 804 nm of the femtosecond laser (a); Absorption spectrum of the glass (b); Up-conversion luminescence intensity of the ⁶H_5/2→⁶P_3/2 transition of Sm³⁺ ions in the glass as a function of the femtosecond laser pump power (c); Photograph of the emission state of Sm³⁺ under the excitation of the femtosecond laser (d).

Download Full Size | PDF

Figure 1(b) shows the absorption spectrum of the glass. Except the obvious absorption peak located at 402 nm, there is no conspicuous absorption in the range of 500-850 nm, including the energy of 804 nm, as shown in Fig. 1(b). Generally, the main UC mechanisms can be classified as energy transfer, excited-state absorption, cooperative up-conversion, and photon avalanche [18]. Since there is no linear absorption band at 804 nm in the absorption spectrum of the sample, the UC luminescence is not due to above mentioned mechanisms [19]. In our case, the two-photon absorption simultaneous can be considered to be responsible for the UC emission of Sm³⁺. The log-log relationship of the pumping powder of the femtosecond laser and the emission intensity of this glass is shown in Fig. 1(c). The slope coefficient of the fitted line is 1.957. Therefore, the dependence of the fluorescence intensity of Sm³⁺ on the pump power reveals that a two-photon absorption process dominates in the conversion of infrared radiation to the visible emission. It theoretically verifies the occurrence of the two-photon UC process. The photograph of the emission state of Sm³⁺ under the excitation of the femtosecond laser is displayed in Fig. 1(d).

After the irradiation by the focused femtosecond laser, a bright red LPL, from the path traversed by the focal point of the laser is observed inside the sample, which can be seen in the dark with naked eyes after the removal of the activating laser. Figure 2 shows the LPL decay curves of the femtosecond laser-induced phosphorescence with varying laser power. With increasing power densities of the irradiation of the focused ultra-short pulsed laser, the LPL decay time of the Sm³⁺-doped glass is significantly prolonged. We hypothesize that traps are introduced or created in the glass with the femtosecond laser irradiation, at the same time, free electrons and holes are created through a multi-photon excited processes and then trapped by the trapping centers. The electron- or hole- trapping centers increase with enhancing laser power. Therefore, the optimized femtosecond laser-induced LPL with increasing laser power is manifested as the photographs in Fig. 2. The upper right inset of Fig. 2 depicts the afterglow spectrum of the Sm³⁺-doped glass recorded at 1 min after the stoppage of the 804 nm femtosecond laser excitation, which exhibits similarity appearance as the photoluminescence and UC luminescence, which is indicative of the roles of the same emission centers as Sm³⁺ ions played.

 

Fig. 2 Decay curves of the femtosecond laser-induced LPL of the Sm³⁺-doped glass with varying laser power. The inset shows the photograph of the LPL in the glass 5 mins after the removal of the exciting laser with different power (left) and the afterglow spectrum of the glass recorded at 1 min after the stoppage of femtosecond laser excitation (right).

Download Full Size | PDF

Figure 3 shows the TL spectra of the glass with different delay times (2h and 12h) after ceasing the femtosecond laser irradiation (black and red curves). Significant TL signs are detected from 300 to 600K temperature regions after the femtosecond laser pumping, which experimentally indicates that laser-induced trap centers with different depth are created in this sample. Perhaps we can say that there are three types trapping centers (T₁, T₂, T₃) at least present in the samples, located at 348, 430 and 495K, respectively. These TL peaks exhibit different decay tendency along with the delay time of TL measurement. In general, the TL peaks close to (or above) the room temperature are expected to be essential to the LPL. Therefore, the LPL is mainly attributed to the contribution of traps T₁ and T₂. With the delay time increases, the intensity of peak T₁ decreases rapidly, and the band of which almost disappears after 2h delay. Moreover, the intensity of peak T₂ decreases by more than half after 12h delay, which could well explain the red LPL could be perceived with naked eyes after the removal of the activating laser 12 h later. For the deepest traps corresponding to the high-temperature band (T₃), the carriers trapped by T₃ are difficult to be released at room temperature to yield the red LPL. However, the deep traps detected in these samples indicate that the Sm³⁺-doped glass would provide potential application as optical storage materials, for the deep traps precluded the thermal release of the intercepted carriers at room temperature. Moreover, after the sample heated at 524K for 20 min to completely empty the traps, no significant TL sign could be recorded as the green curve in Fig. 3. Then, it is interesting to find that the defects of T₁, T₂, and T₃ could be reproduced with different degrees under UV irradiation as the blue curve in Fig. 3 shows, which was recorded under UV irradiation for 15 min after the traps emptied. Despite the quantity of carries captured by these traps is unfavorable, the phenomena undoubted indicates that parts of electron- or hole- trapping centers introduced by femtosecond laser are preserved and could stabilize the carriers under UV excitation. The results demonstrate a possibility of the rewritten processes of the carriers recapture could be realized with the UV energy, and the regenerated LPL is obtained in this glass. The decay curve of the LPL is recorded and shown in the inset of Fig. 3.

 

Fig. 3 TL curves of the glass with different delay times (2h and 12h) after ceasing the femtosecond laser irradiation (black and red curves), and the glass after heat-treated at 250°C for 20 min and irradiated with UV light (green and blue curves). The inset is the decay curve of the LPL after UV irradiated.

Download Full Size | PDF

On the basis of the above-mentioned results, a possible mechanism is proposed to explain the generation of the UC and the red LPL in 80B₂O₃-5Na₂O-15Al₂O₃-0.1Sm₂O₃ glass with femtosecond laser radiation, as shown in Fig. 4, which should be treated as a qualitative analysis. Pumped with focused 804 nm femtosecond laser, the electrons of Sm³⁺ ions can be directly promoted from ground states to upper excited states by absorbing two-photons simultaneously. The electrons non-radiatively relax to the lowest ⁴G_5/2 level, and then radiatively to terminal state, creating the characteristic emission of Sm³⁺ in the glass. Furthermore, electron-trapping oxygen deficit and the hole-trapping centers are inevitable introduced by the femtosecond pulse laser. We assume that several kinds of defect centers (electron and hole trapping centers) are induced after laser irradiation, but only those defect centers, which can be released by thermal energy around room temperature, contribute to the LPL. Following the irradiation of the focused femtosecond laser, active electrons and holes are created in the glass through multi-photon ionization, Joule heating, and collisional ionization processes [20]. Holes are trapped by non-bridging oxygen ions as well as by tetrahedral coordinated boron atoms, while some of the electrons are trapped by the quasi-F centers as in irradiated sodium borate glasses. In addition, a part of the energy, which is due to the recombination of electrons and holes, may be lost because of multi-phonon relaxation without any emission. Furthermore, it is worth noting that some of the metastable state of defects maintained, and captured the carriers under UV excitation. Consequently, the TL peaks located at T₁, T₂ and T₃ could be detected after UV irradiated, and the red LPL is observed once again after ceasing the UV excitation. The quantitative research on the mechanism involved in the LPL in the glass is under investigation.

 

Fig. 4 Possible schematic of the up-conversion emission and LPL mechanism in the Sm³⁺-doped glass

Download Full Size | PDF

4. Conclusion

After we scan with a focused 804 nm femtosecond pulsed laser, the path traversed by the focal point of the laser in the crystal emits bright red LPL that can be seen with the naked eye after the removal of the activating laser. TL spectra of the samples show that defect centers have formed after the femtosecond laser irradiation. It is suggested that a mechanism of the LPL consists of a thermo-stimulated recombination of holes and electrons at traps induced by the laser irradiation at room temperature. More importantly, the defects retained in the glass could be refilling under UV irradiation, which has been experimentally verified with TL the first time. These findings open the door to studies of applications related to optics storage assisted by femtosecond pulsed laser and UV energy.