Open Access
Add to BibSonomy
Abstract
Short-pulse high-peak-power lasers are crucial laser sources for various applications such as non-thermal ultrafine material processing and eye-safe high-resolution remote sensing. Realizing such operation in a single semiconductor laser chip without amplifiers or external resonators is expected to contribute to the development of compact, affordable laser sources for such applications. In this paper, we demonstrate short-pulse high-peak-power photonic-crystal surface-emitting lasers based on simultaneous absorptive and radiative Q-switching. The proposed device induces an instantaneous and simultaneous decrease in both absorptive and out-of-plane radiation losses due to saturable absorption and self-evolution of the photonic band, respectively, which results in drastic Q-switching operation of the device. Based on this concept, we experimentally demonstrate short-pulse generation with 200-W-class peak power and a pulse width of < 30 ps. In addition, via pulse compression with dispersion compensation, we achieve an even higher peak power of ∼300 W with a shorter pulse width of ∼10 ps.
© 2023 Optica Publishing Group under the terms of the Optica Open Access Publishing Agreement
1. Introduction
The realization of high-peak-power (>100 W), short-pulse (<ns) operation of compact laser sources is desirable for various applications such as non-thermal material processing [1,2], laser remote sensing with high resolution [3,4], and multi-photon bio-imaging [5,6]. Currently, these applications rely on bulky laser systems, such as solid-state lasers and fiber lasers, or complex optical systems combining semiconductor lasers with external amplifiers. By replacing these systems with a single semiconductor laser chip, a significant reduction in system size and cost is expected. However, in conventional edge-emitting semiconductor lasers and vertical-cavity surface-emitting lasers, it is difficult to simultaneously achieve high beam quality and high peak power because multimode oscillation occurs when the emission area is increased for high-power operation.
To solve the above issues, photonic-crystal surface-emitting lasers (PCSELs) [7–11], which have a two-dimensional photonic crystal in the vicinity of an active layer as a laser cavity, have attracted increasing attention. Owing to the two-dimensional standing-wave resonance at a singularity point (Γ-point, etc.) of the photonic crystal, PCSELs enable coherent lasing over areas several orders of magnitude larger than those of conventional semiconductor lasers. Recently, by using the double-lattice photonic crystal, where one lattice point is shifted from another in the x- and y-directions by about a quarter of the wavelength in the material, CW lasing oscillation with an output power of ∼ 7 W was experimentally demonstrated in a PCSEL with a circular resonant diameter of 800 µm [12]. More recently, a design rule for single-mode lasing over even larger (≥ 3 mm) emission area was established based on the control of Hermitian and non-Hermitian couplings, which denote the strengths of optical coupling without and with accompanying radiation loss, respectively [13,14].
To realize sub-nanosecond short-pulse operation in PCSELs, we proposed a PCSEL with a two-dimensional (2D) arrangement of gain and saturable-absorber sections [15]. This device enables passive Q-switching due to the saturation of the absorption loss in the saturable absorbers, leading to short-pulse generation with a peak power of ∼20 W and a pulse width of ∼ 35 ps. In addition, we recently proposed another type of short-pulse PCSEL with a band-edge frequency gradation, which we called a “self-evolving PCSEL” [16]. In the latter device, the Q-switching is induced without saturable absorbers owing to carrier-induced changes of the band-edge frequency and in-plane optical absorption loss. As a result, short-pulse generation with a peak power of ∼80 W and a pulse width of <30 ps was realized. These results show that PCSELs are potentially suitable for the various short-pulse applications mentioned earlier, where even higher peak powers (>100W or even 1 kW) are desired.
In this paper, we develop photonic-crystal surface-emitting lasers capable of short-pulse operation with higher peak powers based on the instantaneous decrease in not only the absorption loss but also the out-of-plane radiation loss. In our proposed device, the absorption-loss change is induced by saturable absorbers while the radiation-loss change is realized by the introduction of rotationally symmetric self-evolving photonic crystals. This simultaneous decrease of these two types of losses induces a drastic Q-switching effect, with which we successfully demonstrate high-peak-power (200-W-class), short-pulse (< 30 ps) operation from a single emitter. We also achieve single-pulse operation with a peak power of > 100 W by driving the developed PCSEL with nanosecond-pulsed current injection. Furthermore, we experimentally demonstrate pulse compression of the developed PCSEL using dispersion compensation and realize short-pulse generation with an even higher peak power of ∼300 W and an even shorter pulse width of ∼10 ps.
2. Lasing principle
Figure 1(a) shows the schematic of our device, in which saturable absorbers and a self-evolving photonic crystal are introduced to induce both absorptive and radiative Q-switching effects. The upper left panel of Fig. 1(a) shows a schematic cross section of the device, and the lower left panel shows the 2D arrangement of gain and saturable-absorber sections. By electrically insulating the saturable absorber section from the gain section by proton implantation, the active layer in the saturable absorber section becomes an absorber, which increases an initial cavity loss (absorption loss) for the lasing mode. As a result, the number of carriers accumulated in the gain section before lasing becomes larger than that in PCSELs without saturable absorber sections. Once the device starts to lase, the absorption loss in the saturable absorber section rapidly decreases owing to carrier generation therein, resulting in passive Q-switching. Here, we employ multiple ring-shaped saturable absorbers because they can increase the absorption losses of the fundamental and higher-order modes simultaneously, which maintains a large threshold gain difference between these modes to ensure single-mode lasing [15].
Fig. 1. Schematic structure and lasing principle of short-pulse high-peak-power photonic-crystal surface-emitting lasers (PCSELs). (a) Schematic structure of a PCSEL with saturable absorbers and a rotationally symmetric self-evolving double-lattice photonic crystal, in which the lattice constant distribution is rotationally symmetric about the center. (b) Band-edge frequency distribution before and after lasing. (c) Dependence of the in-plane wavenumber k// on the radiation constant of the PCSEL.
The right panel of Fig. 1(a) shows a schematic of the rotationally symmetric self-evolving photonic crystals for further enhancement of the Q-switching effect using out-of-plane radiation loss. Here, the lattice constant increases parabolically from the center (a0) to the edge of the gain section (a0+Δa), such that the lattice constant distribution is rotationally symmetric about the center; we note that this structure is different from that proposed in our previous work [16], where the lattice constant distribution increases monotonically along one diagonal direction to induce a change of the in-plane absorption loss [16]. The principle of the radiative Q-switching in the rotationally symmetric self-evolving photonic crystal is illustrated in Fig. 1(b) and Fig. 1(c). At the beginning of lasing [Fig. 1(b)(i)], the light is localized near the center due to the photonic mode gap formed by the parabolic lattice constant distribution. Such a localized state has a large in-plane wavenumber k// and thus a large radiation constant, as shown in the dispersion diagram of the double-lattice photonic crystal [Fig. 1(c)]. On the other hand, once lasing oscillation starts [Fig. 1(b)(ii)], carrier recombination due to stimulated emission induces a refractive index change near the center, upon which the band-edge frequency distribution “self-evolves” into a flatter one, causing the light to spatially expand. Accordingly, k// of the lasing state decreases, leading to an instantaneous reduction of the radiation constant as shown in Fig. 1(c). Consequently, the proposed device induces an instantaneous decrease in not only absorptive loss but also radiation loss, resulting in more drastic Q-switching operation than what can be achieved by decreasing only one type of loss alone.
3. Calculations
To confirm the lasing principle described above, we calculated the temporal waveforms of the proposed device using time-dependent three-dimensional coupled-wave theory [17]. The diameter of the gain section of the device is L = 800 µm. The saturable absorbers consist of three concentric ring-shaped sections with widths of 8 µm and outer diameters of 160 µm, 400 µm, and 640 µm. The lattice constant at the center of the gain section is a0 = 276 nm. Other calculation conditions are the same as those used in Ref. [16]. Figures 2(a) and 2(c) show the temporal waveforms of the output power of the device without and with the rotationally symmetric self-evolving photonic crystal (Δa = 1.0 × 10−3a0) at a continuous current injection of 20 A. Here, the device with the self-evolving photonic crystal [shown in Fig. 2(c)] yields a pulse train with a longer pulse period and a higher peak power than the device without the self-evolving photonic crystal [shown in Fig. 2(a)], indicating that the Q-switching effect is enhanced by the instantaneous decrease in absorption and radiation loss as explained in the previous section. Figures 2(b) and 2(d) show enlarged views of a single pulse in Figs. 2(a) and 2(c), respectively. Although the pulse width is slightly increased by introducing the self-evolving photonic crystal, this pulse width (on the order of 30 ps) is still considerably short. Figure 2(e) shows the normalized photon density distributions inside the photonic-crystal layer at two different timings during the pulse generation in Fig. 2(d). As shown in these figures, the light is localized near the center of the gain section at the start of lasing and spreads over the gain section at the peak of the pulse, in agreement with the lasing principle schematically illustrated in Fig. 1(b). The far-field pattern of the device with the self-evolution photonic crystal at a current injection of 20 A is shown in Fig. 2(f), where a symmetric unimodal beam with a beam divergence angle <0.2° is obtained.
Fig. 2. Calculated transient response of the designed device. (a) Temporal waveform and (b) enlarged waveform of a single pulse for a reference PCSEL with only saturable absorbers (SA) at a current injection of 20 A. (c) Temporal waveform and (d) enlarged waveform of a single pulse for a PCSEL with both SA and a rotationally symmetric self-evolving photonic crystal (PC) at a current injection of 20 A. (e) Normalized photon density distributions at two different timings, labelled (1) and (2) in Fig. 2(d). (f) Far-field pattern with the SA and self-evolving photonic crystal at a current injection of 20 A.
4. Demonstration
To demonstrate high-peak-power, short-pulse lasing based on the above principle, we fabricated the designed a PCSEL with saturable absorbers and a rotationally symmetric self-evolving photonic crystal. The method of the device fabrication, including the implementation of the saturable absorbers, is detailed in Ref. [15]. The dimensions of the saturable absorber sections and the self-evolving photonic crystal in the fabricated device are the same as those used in the calculations of Fig. 2. The lattice constant in the middle of the gain section was a0 = 276 nm, resulting in a lasing wavelength of ∼940 nm.
First, to measure the temporal waveform, the laser beam was focused on a slit of a streak camera using a lens (whose focal length f = 30 mm) after being appropriately attenuated by a variable neutral density filter. Then, the peak power was estimated from the pulse width and pulse period measured by the streak camera and the average power measured by a power meter. The temporal waveforms of the fabricated device at an injection current of 20.6 A acquired by sweeping the streak camera for 5 ns and 1 ns are shown in Figs. 3(a) and 3(b), respectively. A short pulse train with a peak power of ∼185 W, a pulse width of ∼27 ps, and a repetition time interval of ∼1.1 ns was observed. This peak power was 10 times higher than that of our previous PCSEL which had only saturable absorbers [15] and more than two times higher than that of our previous PCSEL which had only a self-evolving photonic crystal [16].
Fig. 3. Experimental demonstration of short-pulse generation in a fabricated device. (a)(b) Measured temporal waveforms at an injection current of 20.6 A acquired by sweeping the streak camera for (a) 5 ns and (b) 1 ns. (c)(d) Streak camera images at an injection current of 20.6 A acquired by sweeping the streak camera for (c) 5 ns and (d) 1 ns. (e) Peak power and average power of the device as a function of the injection current. (f) Pulse width and repetition rate of the device as a function of the injection current. (g) Measured far-field beam pattern. (h) Beam divergence angle of the far-field beam pattern along the u- and v-directions as functions of the injection current. (i) Illustration of single-pulse operation via nanosecond-pulsed current injection. (j) Measured temporal waveform of single-pulse operation using a pulsed current source with a short pulse width of ∼1.3 ns and a peak current of ∼ 20.0 A.
Next, we measured the spatial-temporal evolution of the laser beam by transferring the magnified near-field pattern on the slit of the streak camera using two lenses (f = 50 mm, 75 mm). The slit of the streak camera was parallel to the u-axis of the self-evolving photonic crystal shown in Fig. 1(a). The streak camera image acquired by sweeping for 5 ns is shown in Fig. 3(c), where the streaks indicate periodic light emission throughout the gain section. Figure 3(d) shows the streak camera image acquired by sweeping for 1 ns. In each pulse, the lasing starts from the center of the device and then spreads over the gain section, in agreement with the simulated results shown in Fig. 2(e).
We next investigated the properties of the laser beam as functions of the injection current. Figure 3(e) shows the measured peak power (blue line) and average power (black line) as functions of the injection current. The peak power was observed to be one-to-two orders of magnitude larger than the average power at the same injection current. The measured peak power was slightly smaller than the calculated peak power shown in Fig. 2(c) owing to optical absorption in the GaAs substrate, which was not considered in the simulation. Figure 3(f) shows the measured pulse width (blue line) and repetition rate (black line) as functions of the injection current. The pulse width was nearly constant (< 30 ps) regardless of the injection current. On the other hand, the repetition rate increased with the injection current, because the gain recovery time after each pulsation became shorter as the injection current increases.
We also measured the far-field beam pattern, which is shown in Fig. 3(g). The far-field beam pattern was unimodal, in agreement with the one calculated in Fig. 2(f). Figure 3(h) shows the beam divergence angles of the far-field beam pattern along the u- and v-directions as functions of the injection current. Note that the beam divergence angles were evaluated at 1/e2 of the maximum value in each direction. Narrow beam divergence angles of ∼ 0.15° were observed in both directions, corresponding to a beam quality factor M2 ∼ 1.75, and these divergence angles were nearly constant regardless of the injection current. The peak brightness, evaluated using the measured peak power and beam quality M2 at 20.6 A, was ∼ 6.8 GWcm−2sr−1, which is the highest reported value among all the Q-switched semiconductor lasers without amplifiers.
It should be noted that the repetition rate of the pulse train of the above device was on the order of GHz, which makes the device potentially suitable for applications such as highly efficient fine material processing via ablation cooling [18,19] and real-time nonlinear laser microscopy [20]. On the other hand, single short-pulse generation or short-pulse trains with an electrically controllable repetition rate on the order of kHz∼MHz is also desired in many other applications such as eye-safe, high-resolution LiDAR. Therefore, to achieve single pulse operation, we drove the above device using a pulsed current source whose pulse width was comparable to the repetition period of the device, and we extracted only the first pulse as shown in Fig. 3(i). Figure 3(j) shows the temporal waveform measured by the streak camera when driven by a pulsed current source with a short pulse width of ∼1.3 ns and a peak current of ∼20.0 A. Here, the peak power of the laser pulse was estimated from the pulse energy measured by an energy meter and the integral of the temporal waveform measured by the streak camera. As shown in Fig. 3(j), a single optical pulse with a peak power of ∼148 W and a pulse width of ∼27 ps was obtained. This peak power was lower than those shown in Fig. 3(a) due to the finite rise time (∼1 ns) of the pulsed current. It is expected that injecting pulsed current with a shorter pulse width and a higher amplitude will result in higher peak-power single-pulse generation.
Finally, we demonstrate the generation of even shorter-width, higher-power pulses via pulse compression using a dispersion-compensation medium. The devices discussed above exhibited large wavelength chirping during pulsed oscillation due to a refractive index change caused by carrier consumption and resultant band-edge evolution [16]. By compensating the dispersion of such large wavelength chirps, even shorter pulses with even higher peak powers can be generated. Figure 4(a) shows the experimental setup for pulse compression. A chirped Bragg grating (CBG, dispersion 60 ps/nm, center wavelength 937 nm, bandwidth 8.5 nm) is used as the dispersion-compensation medium [21]. The laser beam collimated with a lens (f = 100 mm) was injected into the CBG, and the reflected beam was focused on a slit of a streak camera using another lens (f = 30 mm). The pulse waveforms before and after CBG injection are shown in Figs. 4(b) and (c), respectively. As shown in the figures, we observed a pulse width of ∼10 ps and a peak power of ∼ 300 W. The enhancement ratio of the peak power before and after pulse compression (∼ 1.5 times) was smaller than the compression ratio of the pulse width (∼3 times). This difference of ratios was partially due to the lower-than-unity reflectance of the CBG and also due to insufficient dispersion compensation caused by a mismatch between the center wavelength of the CBG and the lasing wavelength. By appropriately adjusting these CBG parameters, the realization of short pulses with 500-W-class peak power is expected. Furthermore, by increasing the carrier-induced refractive-index change coefficient of the active layer as well as the magnitude of the parabolic frequency gradation, the wavelength change (chirping) during pulse generation can be further increased, enabling much shorter Fourier-limited pulse width (∼sub picosecond) after pulse compression.
Fig. 4. Pulse compression via dispersion compensation. (a) Schematic of the experimental setup. (b)(c) Measured temporal waveform at an injection current of 20 A (b) before and (c) after pulse compression.
5. Conclusion
We have demonstrated a short-pulse, high-peak-power photonic-crystal surface-emitting laser based on simultaneous absorptive and radiative Q-switching. In our proposed device, an instantaneous decrease of absorption loss decrease has been realized using saturable absorbers, while an instantaneous decrease of radiation loss has been realized using a rotationally symmetric self-evolving photonic crystal. Based on this principle, we have experimentally realized the generation of laser pulses with short widths of < 30 ps and high peak powers of ∼200 W, which is 10 times higher than that obtained using only saturable absorbers [15], and more than two times higher than that obtained using a self-evolving photonic crystal without saturable absorbers [16]. In addition, toward various applications such as eye-safe, high-precision light detection and ranging, we have achieved single-pulse operation with a peak power of ∼150 W by driving the device with nanosecond-pulsed current injection. Furthermore, toward the realization of high-efficiency, high-precision non-thermal laser processing [22], we have experimentally realized the generation of laser pulses with even shorter pulse widths (∼10 ps) and higher peak powers (∼300 W) via pulse compression using a dispersion-compensation medium. We expect that our short-pulse, high-peak-power, high-beam-quality single-chip semiconductor laser will significantly simplify the optical systems used in various applications and will contribute to the realization of a future smart society.
Funding
Japan Society for the Promotion of Science (20H02655, 22H04915).
Disclosures
The authors declare no conflicts of interest.
Data availability
Data underlying the results presented in this paper are not publicly available at this time but may be obtained from the authors upon reasonable request.
References
1. S. Nolte, C. Momma, H. Jacobs, A. Tünnermann, B. N. Chichkov, B. Wellegehausen, and H. Welling, “Ablation of metals by ultrashort laser pulses,” J. Opt. Soc. Am. B 14(10), 2716–2722 (1997). [CrossRef]
2. S. Mishra and V. Yadava, “Laser beam micromachining (LBMM)–a review,” Opt. Lasers Eng. 73, 89–122 (2015). [CrossRef]
3. B. Kaldvee, A. Ehn, J. Bood, and M. Aldén, “Development of a picosecond lidar system for large-scale combustion diagnostics,” Appl. Opt. 48(4), B65–B72 (2009). [CrossRef]
4. A. Velten, W. Thomas, G. Otkrist, V. Ashok, M. G. Bawendi, and R. Raskar, “Recovering three-dimensional shape around a corner using ultrafast time-of-flight imaging,” Nat. Commun. 3(1), 745 (2012). [CrossRef]
5. W. Denk, J. H. Strickler, and W. W. Webb, “Two-photon laser scanning fluorescence microscopy,” Science 248(4951), 73–76 (1990). [CrossRef]
6. M. Kuramoto, N. Kitajima, H. Guo, Y. Furushima, M. Ikeda, and H. Yokoyama, “Two-photon fluorescence bioimaging with an all-semiconductor laser picosecond pulse source,” Opt. Lett. 32(18), 2726–2728 (2007). [CrossRef]
7. M. Imada, S. Noda, A. Chutinan, and T. Tokuda, “Coherent two-dimensional lasing action in surface-emitting laser with triangular-lattice photonic crystal structure,” Appl. Phys. Lett. 75(3), 316–318 (1999). [CrossRef]
8. S. Riechel, C. Kallinger, U. Lemmer, and J. Feldmann, “A nearly diffraction limited surface emitting conjugated polymer laser utilizing a two-dimensional photonic band structure,” Appl. Phys. Lett. 77(15), 2310–2312 (2000). [CrossRef]
9. M. Imada, A. Chutinan, S. Noda, and M. Mochizuki, “Multidirectionally distributed feedback photonic crystal lasers,” Phys. Rev. B 65(19), 195306 (2002). [CrossRef]
10. K. Hirose, Y. Liang, Y. Kurosaka, A. Watanabe, T. Sugiyama, and S. Noda, “Watt-class high-power, high-beam-quality photonic-crystal lasers,” Nat. Photonics 8(5), 406–411 (2014). [CrossRef]
11. S. Noda, K. Kitamura, T. Okino, D. Yasuda, and Y. Tanaka, “Photonic-Crystal Surface-Emitting Lasers: Review and Introduction of Modulated-Photonic Crystals,” IEEE J. Sel. Top. Quantum Electron. 23(6), 1–7 (2017). [CrossRef]
12. M. Yoshida, M. De Zoysa, K. Ishizaki, Y. Tanaka, M. Kawasaki, R. Hatsuda, B. Song, J. Gelleta, and S. Noda, “Double-lattice photonic-crystal resonators enabling high-brightness semiconductor lasers with symmetric narrow-divergence beams,” Nat. Mater. 18(2), 121–128 (2019). [CrossRef]
13. T. Inoue, M. Yoshida, J. Gelleta, K. Izumi, K. Yoshida, K. Ishizaki, M. De Zoysa, and S. Noda, “General recipe to realize photonic-crystal surface-emitting lasers with 100-W-to-1-kW single-mode operation,” Nat. Commun. 13(1), 3262 (2022). [CrossRef]
14. M. Yoshida, S. Katsuno, T. Inoue, J. Gelleta, K. Izumi, M. De Zoysa, K. Ishizaki, and S. Noda, “High-brightness scalable continuous-wave single-mode photonic-crystal laser,” Nature 618(7966), 727–732 (2023). [CrossRef]
15. R. Morita, T. Inoue, M. De Zoysa, K. Ishizaki, and S. Noda, “Photonic-crystal lasers with two-dimensionally arranged gain and loss sections for high-peak-power short-pulse operation,” Nat. Photonics 15(4), 311–318 (2021). [CrossRef]
16. T. Inoue, R. Morita, K. Nigo, M. Yoshida, M. De Zoysa, K. Ishizaki, and S. Noda, “Self-evolving photonic crystals for ultrafast photonics,” Nat. Commun. 14(1), 50 (2023). [CrossRef]
17. T. Inoue, R. Morita, M. Yoshida, M. De Zoysa, Y. Tanaka, and S. Noda, “Comprehensive analysis of photonic-crystal surface-emitting lasers via time-dependent three-dimensional coupled-wave theory,” Phys. Rev. B 99(3), 035308 (2019). [CrossRef]
18. C. Kerse, H. Kalaycıoğlu, P. Elahi, B. Çetin, D. K. Kesim, Ö. Akçaalan, S. Yavaş, M. D. Aşık, B. Öktem, H. Hoogland, and R. Holzwarth, “Ablation-cooled material removal with ultrafast bursts of pulses,” Nature 537(7618), 84–88 (2016). [CrossRef]
19. W. Hu, Y. C. Shin, and G. King, “Modeling of multi-burst mode pico-second laser ablation for improved material removal rate,” Appl. Phys. A 98(2), 407–415 (2010). [CrossRef]
20. S. W. Chu, T. M. Liu, C. K. Sun, C. Y. Lin, and H. J. Tsai, “Real-time second-harmonic-generation microscopy based on a 2-GHz repetition rate Ti: sapphire laser,” Opt. Express 11(8), 933–938 (2003). [CrossRef]
21. L. B. Glebov, V. Smirnov, E. Rotari, I. Cohanoschi, L. Glebova, O. V. Smolski, J. Lumeau, C. Lantigua, and A. Glebov, “Volume-chirped Bragg gratings: monolithic components for stretching and compression of ultrashort laser pulses,” Opt. Eng. 53(5), 051514 (2014). [CrossRef]
22. B. N. Chichkov, C. Momma, S. Nolte, F. von Alvensleben, and A. Tünnermann, “Femtosecond, picosecond and nanosecond laser ablation of solids,” Appl. Phys. A 63(2), 109–115 (1996). [CrossRef]
