We demonstrate room temperature continuous-wave laser operation at 1.3 µm in a photonic crystal nanocavity with InAs/GaAs self-assembled quantum dots by optical pumping. By analyzing a coupled rate equation and the experimental light-light characteristic plot, we evaluate the spontaneous emission coupling factor of the laser to be ~0.22. Three-dimensional carrier confinement and a low transparent carrier density due to volume effect in a quantum dot system play important roles in the cw laser operation at room temperature as well as a high quality factor photonic crystal nanocavity.
©2006 Optical Society of America
Photonic crystal (PhC)  microcavities with very high quality factor Q can strongly confine photons in very small volumes, of the order of a cubic wavelength. The strong light-matter interaction due to the photon confinement opens up new possibilities in various fields, such as efficient optical sources [2, 3], cavity quantum electrodynamics , and photonics . In particular, PhC nanocavities are considered one of the best candidates for ultra-low threshold lasers due to their small mode-volume and high Q.
The first PhC laser was reported on InGaAsP multi-quantum well (QW) structures by Painter et al. in 1999 . PhC lasers with quantum dots (QDs) as the gain material have been more difficult devices to construct due to their inherently lower modal gain. Over the past few years, some groups have been challenging to achieve laser operation in PhC nanocavities with QDs: the first such lasing was reported at room temperature with pulsed excitation , and more recently, continuous-wave (cw) lasing at 4.5 K and up to 80 K using optical pumping has been reported . However, in spite of the great importance for practical applications, cw laser operation in PhC nanocavities at room temperature has not been reported so far with either QW or QD as the gain material, due to the strict requirements for lasing. At room temperature, the rate of nonradiative recombination is much higher than that of radiative recombination. Therefore, sufficiently high modal gain, a very high Q cavity and/or some mechanisms that reduce nonradiative recombination are required. Quantum dots are preferable to QW as gain material because of their carrier confinement and low transparent carrier density, which leads to a very low lasing threshold. However, the modal gain of a QD is much lower than that of a QW, therefore, very high Q is essential to achieve a laser operation at room temperature with cw excitation. In this paper, we report the demonstration of room temperature cw laser operation at 1.3 µm in a PhC nanocavity with InAs/GaAs self-assembled QDs as the gain material. The spontaneous emission coupling factor β was estimated by comparing experimental results with theoretical L-L curves calculated by a conventional coupled rate equation. This PhC nanocavity cw laser has, to the best of our knowledge, the smallest mode volume among the cw lasers operating at room temperature.
2. Photonic crystal nanocavity fabrication and structure
2.1 Crystal growth and structure
We investigated PhC nanocavities fabricated in the sample grown on an undoped (100)-oriented GaAs substrate by molecular beam epitaxy. First, a 300-nm-thick GaAs buffer layer was deposited on the substrate at 600 °C. Then, a 700-nm-thick Al0.7Ga0.3As sacrificial layer was grown at 600 °C. Finally, a 250-nm-thick GaAs slab layer was grown including five-stacked self-assembled InAs QD layers as the active gain material. This slab layer consisted of five periods of structures grown on a 50-nm-thick GaAs layer grown at 600 °C. Each 40-nm-thick period consisted of an InAs QD layer grown at 445 °C, a 4-nm-thick In0.16Ga0.84As strain reducing layer grown at 445 °C, and a 36-nm-thick GaAs inter-layer grown at 520 °C. The nominal thickness of the InAs QD layer was 2.85 monolayers (MLs) and the layer was grown with a growth rate of 0.0076 ML/s. The strain reducing layer was grown to shift the wavelength of the photon emission from QDs from 1.25 to 1.3 µm . The areal quantum dot density was ~2×1010 cm-2 for each QD layer.
2.2 Photonic crystal nanocavity fabrication
The PhC nanocavity in the sample was fabricated by using electron beam lithography, with two dry etching processes and a wet etching process. The architecture of the sample, which resembles to the investigated PhC, is shown in the scanning electron micrograph of Fig. 1(a). A 150-nm-thick SiO2 layer was deposited on the surface of the sample and a 350-nm-thick electron-beam resist was spun on the SiO2 layer. This SiO2 layer functioned as a mask for the GaAs layer during the Cl2 dry etching process mentioned later. We adopted a point defect structure, called L3 defect, which consists of three missing air holes along the Γ-K direction of the triangular PhC lattice. In addition, the first and third nearest air holes at both edges of the cavity were shifted to outside the cavity to obtain higher cavity quality factor Q . We fabricated a sample with a period of the lattice a=353 nm and radius of the air hole r=0.27a. The displacement of the shifted air holes was 0.15a as depicted in Fig. 1(b). The PhC structures were patterned using an electron-beam lithography system and then transferred to the SiO2 layer by inductive coupled plasma reactive ion etching (ICP-RIE) using a CF4/O2/Ar mixture. An additional ICP-RIE process using a Cl2/Ar mixture was used to etch the air holes into the GaAs layer. Finally, the residual SiO2 mask and AlGaAs sacrificial layer were removed using an HF:H2O (1:4) acid solution to form freestanding 250-nm-thick air bridge structures. This series of processes fabricated a semiconductor based air-bridged PhC slab with an air hole array, which produces an in-plane photonic bandgap. Photons are also confined in the vertical direction due to the refractive index contrast between the slab and air.
3. Experimental results and discussions
3.1 Experimental setup
Measurements were performed with a micro-photoluminescence (µ-PL) setup at room temperature using a laser diode (λ=785 nm) as the excitation source. The pump laser beam was focused to a 4 µm diameter spot on the sample surface by a microscope objective [50x, numerical aperture=0.42], and was positioned on the PhCs using piezo-electric nanopositioners. The PL was collected by the same microscope objective. The sample emission was PL from the QDs in the excitation spot. Some of the QDs were coupled to the cavity mode and the photon emission from the QDs was enhanced. The leakage of the cavity mode and the PL from the QD ensemble out of the top of the sample was observed in this geometry.
3.2 Experimental results and discussions
Figure 2 shows the lasing spectrum observed using cw excitation light with an excitation power of 40 µW. The sharp peak observed at around 1320 nm corresponds to the strong light emission from the cavity mode, which is located at the PL peak of the ground state of the InAs QD ensemble. Single-mode operation was observed in a wide spectral range covering 60 nm, and lasing operation was observed for many PhC nanocavities fabricated in the same wafer.
The laser output power was stable with a moderate cw excitation power above the threshold. However, the cavity suffered from thermal damage and/or contamination by air when the structures were irradiated by pump beam, for a long time, with cw excitation power above ~1.5 times greater than its threshold power. Therefore, L-L curve with cw excitation showed approximately the same threshold power as that with quasi-cw excitation, but the output power gradually decreased with relatively high-power cw excitation because of the damage. For an accurate evaluation, we used 100 µs-long quasi-cw pumping with a repetition rate of 1 kHz (10% duty cycle) to avoid thermal problems in the following experiments. We note that the pump duration of 100 µs is much longer (~x 105) than the timescales of any dynamics in the system (~1 ns) except the heating. In the previous reports on PhC microcavity lasers with pulse pumping, the pump pulse durations were 10 ns  and 20 ns .
The excitation power dependence of the output power is shown in Fig. 3. It is difficult to determine the precise threshold due to gradual change of the L-L curve around the threshold. The conventional estimation of the lasing threshold, which is obtained by extrapolation of the red line to zero output power, yields an averaged threshold excitation power of ~2.5 µW at 10% duty cycle, corresponding to a cw pump power of 25 µW. The ratio of the absorbed power in the GaAs slab layer to the incident pump power can be roughly estimated by considering a surface reflectivity of ~30%, the thickness of the GaAs slab layer of 250 nm, and an approximate absorption coefficient of 10 000 cm-1. From these values, the absorbed power in the slab layer is estimated to be ~15% of the pump power. Consequently, the actual absorbed power at the threshold is ~4 µW for cw operation. This soft turn-on implies a large spontaneous emission coupling factor β. Above the threshold excitation power, linear increase of the output power is observed as indicated by the red line.
An important characteristic of the lasing operation is the narrowing of the emission linewidth. Figure 4 shows PL spectra around the cavity mode measured at various averaged excitation powers. The figures in the inset are the averaged excitation powers in units of µW. The output power increases and the linewidth decreases as the excitation power is increased. The redshift of the lasing mode can be attributed to the change of the refractive index due to the sample heating.
Figures 5(a) and 5(b) show the linewidths of the lasing mode on a linear scale, and L-L curve on a logarithmic scale, at various excitation powers. The linewidths measured below 0.2 µW have a value of ~380 µeV. This relatively wide linewidth corresponds to Q=2500, which is limited by the absorption of the InAs QDs. We also fabricated PhC nanocavities with nearly the same structural parameters in another sample with only a single InAs QD layer, exhibiting much less absorption, using the same process conditions. The cavity Q of the PhC nanocavities with a single QD layer were over 10 000. Below the threshold, the linewidth is decreased as the excitation power is increased, because the absorption by QDs is reduced. Around the threshold, the phase transition from spontaneous radiation into lasing results in a pronounced kink both in the linewidth and output power, as shown in Figs. 5(a) and 5(b). Strauf et. al. reported the linewidth and intensity dependence of the pump power in cw laser operation measured at 4.5 K in ref. 8. The results in Fig. 5 are similar to their results and calculations in microcavity lasers . However, the kink in the L-L curve around the threshold is more pronounced compared with the results in ref. 8. The linewidth decreases above the threshold, reaching the measurable energy resolution limit of 35 µeV. We can approximately estimate the value of the cavity Q from the linewidth just below the threshold, where the optical loss by the absorption of QDs is very weak. The linewidth of ~50 µeV measured at just below the threshold corresponds to Q ~19 000.
In order to evaluate the β of the PhC nanocavity laser, the experimental L-L plot is compared with theoretical L-L curves calculated using rate equations. A conventional coupled rate equation model  for the carrier density N and the photon density P in the nanocavity was used:
Where R ex is the pumping rate, τ r and τ nr are radiative and nonradiative recombination times, respectively, and c/n eff=1×1010 cm/s. Γ is the confinement factor estimated to be ~0.01, and the gain function g(N) is given by g olog(N/N tr), where g o is the gain coefficient. The best fit has been found for g o=1×105 cm-1, transparent carrier density N tr=5×1014 cm-3 with τ r and τ nr being 3 ns and 150 ps, respectively. The value of τ nr=150 ps is very close to the PL lifetime of ~120 ps measured at room temperature by time resolved PL measurements. The value of τ r=3 ns is longer than the PL lifetime of ~1.5 ns measured for another single QD layer sample. However, the difference in the parameter did not largely change the value of β. The solid green curve in Fig. 5(b) is the best fitting curve for β=0.22. The value of β is the spontaneous emission rate into the lasing mode divided by the total spontaneous emission rate . Therefore, the value of β is a measure of the optical efficiency of a laser. Recently, other groups have reported large values of β ranging larger than 0.05 in microdisk lasers [13, 14] and 0.125 in PhC laser . Compared with these values, the obtained value of β=0.22 is relatively high. A very large value of β~0.85 was reported in a PhC microcavity laser with a few QDs at low temperature . If such a high value of β can be obtained at room temperature, the threshold pump power will be greatly reduced and lead to a stable cw operation well above the threshold.
It is also insightful to evaluate the approximate number of QDs contributing to the laser operation. Only the QDs spectrally overlapping the lasing mode and within the cavity can contribute to the lasing. The inhomogeneous linewidth of the ground state is ~35 meV, and assuming the homogeneous linewidth of a single QD is 10 meV at room temperature [16, 17], the fraction of QDs coupling with the lasing mode is estimated to be ~30%. The total number of QDs in the nanocavity is roughly estimated from the area of the nanocavity ~0.8 µm2, and five-stacked QD layers with the density of ~2×1010 cm-2. Hence, the number of QDs within the cavity is ~800 that of QDs contributing to the lasing is estimated to be ~240. Therefore we can conclude that QDs of the order of a hundred, not a few QDs, provide the gain for the laser operation.
For the cw laser operation at room temperature, sufficient modal gain and high-Q cavity are essential. We point out that three-dimensional carrier confinement, which is one of the important characteristics of a QD system, plays a very important role in the cw laser operation at room temperature. The PhC nanocavity consists of air hole arrays, thus any photocarriers diffuse and easily be captured at surface states at the material-air boundaries especially at room temperature. Therefore, it is very important to reduce surface recombination in the system. Ryu et al. reported the effect of nonradiative recombination in InGaAsP-based two-dimensional PhC slab structures . They claimed that the surface recombination process is the main nonradiative recombination process in the structures at low power excitation. In GaAs-based structures, the surface recombination velocity is higher than that in InP-based structures , thus, the suppression of the carrier diffusion is very important. Actually, the value of τ r/τ nr=20 obtained by the fitting indicates that approximately 95% of photocarries are nonradiatively recombined. The diffusion of the captured carriers in QDs is suppressed in the QD system compared with other systems . Therefore, this carrier confinement effect, which reduces nonradiative recombination, contributes to the cw laser operation at room temperature. We also point out the small volume effect of the QD system probably has an advantage in the laser operation in air-bridge-type PhC nanocavity structure. The transparent carrier density of PhC nanocavity lasers with QDs as the active materials is much smaller compared with the PhC nanocavity lasers with QWs  by the ~3 order of magnitude due to small volume of QD. This remarkably low transparent carrier density requires much less excitation power, which avoids excessive sample heating and enables the cw lasing at room temperature. However, a nanocavity with a very high Q and β, which enables lasing with small number of QDs, is necessary to achieve stable cw laser operation well above the threshold at room temperature. It may be another solution to adopt some other structures with more efficient thermal dissipation [21, 22]. The cavity resonant excitation technique, which excites QDs in PhC nanocavities very efficiently, will also be useful to reduce the sample heating . These improvements and techniques will be able to realize a highly efficient nano-laser with an ultra-low threshold.
We have experimentally demonstrated cw laser operation at 1.3 µm in a PhC nanocavity with InAs/GaAs self-assembled QDs by optical pumping. The L-L curve shows a soft turn-on of lasing resulting in a relatively high spontaneous emission coupling factor of 0.22. The average lasing threshold power is ~2.5 µW at 10% duty cycle, corresponding to an actual absorbed power of ~4 µW at the threshold for cw operation. The linewidth of the lasing mode decreases by more than a factor of ten and reaches the measurable resolution limit of ~35 µeV. Two important characteristics of QD system of three-dimensional carrier confinement and small volume effect contribute to the cw laser operation at room temperature, but also critical is a high quality factor PhC nanocavity.
The authors would like to thank S. Kako, Dr. T. Nakaoka, and Dr. D. A. Redfern for useful discussions. This work was supported by the Focused Research and Development Project for the Realization of the World’s Most Advanced IT Nation, Ministry of Education, Culture, Sports, Science and Technology.
References and links
3. W.-H. Chang, W.-Y. Chen, H.-S. Chang, T.-P. Hsieh, J.-I. Chyi, and T.-M. Hsu, “Efficient single-photon sources based on low-density quantum dots in photonic-crystal nanocavities,” Phys. Rev. Lett. 96, 117401-1–117401-4 (2006). [CrossRef]
4. T. Yoshie, A. Scherer, J. Hendrickson, G. Khitrova, H.M. Gibbs, G. Rupper, C. Ell, O.B. Shchekin, and D.G. Deppe, “Vacuum Rabi splitting with a single quantum dot in a photonic crystal nanocavity,” Nature 432, 200–203 (2004). [CrossRef] [PubMed]
7. T. Yoshie, O.B. Shchekin, H. Chen, D.G. Deppe, and A. Scherer, “Quantum dot photonic crystal lasers,” Electron. Lett. 38, 967–968 (2002). [CrossRef]
8. S. Strauf, K. Hennessy, M.T. Rakher, Y.-S. Choi, A. Badolato, L.C. Andreani, E.L. Hu, P.M. Petroff, and D. Bouwmeester, “Self-tuned quantum dot gain in photonic crystal lasers,” Phys. Rev. Lett. 96, 127404-1–127404-4 (2006). [CrossRef]
9. K. Mukai, Y. Nakata, K. Ohtsubo, M. Sugawara, N. Yokoyama, and H. Ishikawa, “1.3-µm CW lasing of InGaAs-GaAs quantum dots at room temperature with a threshold current of 8 mA,” IEEE Photon. Technol. Lett. 11, 1205–1207 (1999). [CrossRef]
11. G. Björk and Y. Yamamoto, “On the linewidth of microcavity lasers,” Appl. Phys. Lett. 60, 304–306 (1992). [CrossRef]
12. G. Björk and Y. Yamamoto, “Analysis of semiconductor microcavity lasers using rate equations,” IEEE J. Quantum Electron. 35, 2386–2396 (1991). [CrossRef]
13. R.E. Slusher, A.F.J. Levi, U. Mohideen, S.L. McCall, S.J. Pearton, and R.A. Logan, “Threshold characteristics of semiconductor microdisk lasers,” Appl. Phys. Lett. 63, 1310–1312 (1993). [CrossRef]
14. L.Z. Zhang and E. Hu, “Lasing from InGaAs quantum dots in an injection microdisk,” Appl. Phys. Lett. 82, 319–321 (2003). [CrossRef]
15. H.Y. Ryu, M. Notomi, E. Kuramoti, and T. Segawa, “Large spontaneous emission factor (>0.1) in the photonic crystal monopole-mode laser,” Appl. Phys. Lett. 84, 1067–1069 (2004). [CrossRef]
16. A. Sakamoto and M. Sugawara, “Theoretical calculation of lasing spectra of quantum-dot lasers: Effect on homogeneous broadening of optical gain,” IEEE Photon. Technol. Lett. 12, 107–109 (2000). [CrossRef]
17. M. Bayer and A. Forchel, “Temperature dependence of the exciton homogeneous linewidth in In0.60Ga0.40As/GaAs self-assembled quantum dots,” Phys. Rev. B 65, 041308-1–041308-4 (2002).
18. H.-Y. Ryu, J.-K. Hwang, D.-S. Song, I.-Y. Han, and Y.-H. Lee, “Effect of nonradiative recombination on light emitting properties of two-dimensional photonic crystal slab structures,” Appl. Phys. Lett. 78, 1174–1176 (2001). [CrossRef]
19. D. D. Nolte, “Surface recombination, free-carrier saturation, and dangling bonds in InP and GaAs,” Sol.-Stat. Electron. 33, 295–298 (1990). [CrossRef]
20. K. Kounoike, M. Yamaguchi, M. Fujita, T. Asano, J. Nakanishi, and S. Noda, “Investigation of spontaneous emission from quantum dots embedded in two-dimensional photonic-crystal slab,” Electron. Lett. 41, 1402–1403 (2005). [CrossRef]
21. J.-K. Hwang, H.-Y. Ryu, D.-S. Song, I.-Y. Han, H.-W. Song, H.-K. Park, and Y.-H. Lee, “Roomtemperature triangular-lattice two-dimensional photonic band gap lasers operating at 1.54 µm,” Appl. Phys. Lett. 76, 2982–2984 (2000). [CrossRef]
22. C. Monat, C. Seassal, X. Letartre, P. Viktorovitch, P. Regreny, M. Gendry, P. Rojo-Romeo, G. Hollinger, E. Jalaguier, S. Pocas, and B. Aspar, “InP 2D photonic crystal microlasers on silicon wafer: room temperature operation at 1.55 µm,” Electron. Lett. 37, 764–766 (2001). [CrossRef]
23. M. Nomura, S. Iwamoto, T. Nakaoka, S. Ishida, and Y. Arakawa, “Localized excitation of InGaAs quantum dots by utilizing a photonic crystal nanocavity,” Appl. Phys. Lett. 88, 141108-1–141108-3 (2006). [CrossRef]