The authors report on the achievement of lasing in rolled-up semiconductor microtubes at room temperature, wherein self-organized InGaAs/GaAs quantum dots are incorporated as the gain medium. The free-standing quantum dot microtubes, with a diameter of ~ 5-6 μm and wall thickness of ~ 100 nm, are formed when the coherently strained InGaAs/GaAs quantum dot heterostructure is selectively released from the GaAs substrate. The devices are characterized by an ultralow threshold (~ 4 μW) and a minimum intrinsic linewidth of ~ 0.2 – 0.3 nm at room temperature. The multiple lasing modes are analyzed using both the finite-difference time domain method and also a planar dielectric waveguide model.
©2009 Optical Society of America
The development of micro- and nanoscale lasers has been spurred by the rapid progress in optical micro- and nanocavities [1–4]. Significant advancements in strained heteroepitaxy have also enabled the achievement of superior quality self-organized quantum dot heterostructures with near-discrete density of states and large and broad gain . It has therefore been envisioned that the incorporation of quantum dots in high Q optical microcavities can lead to nanoscale lasers with potentially ultralow or near-zero threshold, temperature invariant operation, and ultrahigh-speed frequency response, providing the basic building blocks for chip-level optical communications and future quantum networking systems. In this regard, significant progress has been made in photonic crystal, microdisk, and micropillar quantum dot lasers and coherent light emitting devices [2, 6, 7]. However, most of these devices operate under low temperature and/or pulsed mode. Recently, a new type of optical cavities, with the use of rolled-up micro- and nanotubes, have been developed, which are formed when a coherently strained epitaxial heterostructure is selectively released from the host substrate [8–11]. Self-organized quantum dots embedded in the tube structures generally exhibit further enhanced luminescence efficiency, due to the reduced strain distribution [8, 12]. Photons are strongly confined around the periphery of the tube, and the resulting near-perfect overlap between the maximum optical field intensity and the active region can lead to an extremely large modal gain. More importantly, rolled-up micro- and nanotubes exhibit epitaxially smooth surface, which can greatly reduce, or eliminate carrier nonradiative recombination and optical scattering losses associated with the etched active regions of conventional III-V semiconductor optical cavities, thereby promising a new generation of micro- and nanoscale lasers with further reduced threshold and enhanced output power.
Optical resonance modes have been first observed in InAs/GaAs quantum dot microtubes at 5 K  and subsequently in SiOx/Si tubes at 300 K . More recently, we have achieved strong coherent emission from ultrathin-walled (~ 50 nm) InGaAs/GaAs quantum dot microtubes at room temperature . It has also been demonstrated that their emission characteristics, including the mode profiles, emission direction, and output coupling efficiency, can be controlled by varying the tube diameters, wall thicknesses, as well as the surface geometry using standard photolithography process [14–16]. To date, however, a rolled-up micro- or nanotube laser has not been reported.
In this context, we have achieved room temperature lasing in rolled-up microtubes under optical pumping, wherein self-organized InGaAs/GaAs quantum dots are incorporated as the gain region. The free-standing quantum dot microtube laser active region has a diameter of ~5 – 6 μm and a wall thickness of ~ 100 nm. At room temperature, the lasers exhibit an ultralow threshold power of ~ 4.0 μW and a minimum intrinsic linewidth of ~ 0.2 – 0.3 nm. The multiple lasing modes are analyzed using both the finite-difference time domain method and also a planar dielectric waveguide model.
2. Device Growth and Fabrication
Self-organized quantum dot microtube laser heterostructures were first grown on GaAs substrates by solid source molecular beam epitaxy. Illustrated in the inset of Fig. 1(a) , the laser heterostructure consists of two InGaAs/GaAs quantum dot layers, a 20 nm pseudomorphic In0.18Ga0.82As quantum well, and a 50 nm AlAs sacrificial layer. Growth conditions for the quantum dot layers were varied in order to achieve a relatively large inhomogeneous broadening without compromising the quality of the dot layers, which can provide large and broad gain. Additionally, the resulting 3-dimensional carrier confinement greatly reduces nonradiative recombination associated with the presence of surface defects. Important considerations in the design of microtube lasers include the photon confinements both around the circumference of the tube as well as along the axial direction. These requirements can be met by utilizing free-standing microtubes with a controlled surface geometry, which can lead to a variation of the effective refractive index along the tube axial direction and therefore appropriate photon confinement for laser operation. The involved fabrication process, detailed elsewhere, is also briefly described [11, 14]. A U-shaped mesa, shown in Fig. 1(a), was first defined by etching to the InGaAs layer. Quantum dot microtubes with a controlled surface geometry, illustrated in Fig. 1(b), were subsequently formed, when the underlying AlAs sacrificial layer was selectively etched. The surface geometry is directly related to the corrugations introduced at the inner edge of the U-shaped mesa. To reduce the radiative loss through the substrate, the region between the two side pieces of the U-shaped mesa was etched to ~ 1 μm before the tube formation, which further increases the air gap between the central part of the tube and the substrate, shown in Fig. 1(c). In this experiment, the fabricated quantum dot microtubes had approximately two revolutions, corresponding to a wall thickness of ~ 100 nm. The tube diameters are ~ 5 – 6 μm, predetermined by the strain of the pseudomorphic InGaAs layer and the subsequently grown quantum dot heterostructure. Self-organized quantum dots embedded in microtube structures generally exhibit stronger luminescence emission with a small (~10 meV) red shift, compared to the as grown dot layers, due to the reduced strain distribution [8, 12].
3. Device Characterization
Rolled-up InGaAs/GaAs quantum dot microtube lasers were characterized using micro-photoluminescence spectroscopy at room temperature. The devices were optically excited by a continuous wave He-Ne laser (λ = 632.8 nm) through a microscope objective. Emission from the microtube was collected by the same objective and analyzed using a high-resolution spectrometer and an InGaAs detector with lock-in amplification. The emission spectrum measured at an absorbed pump power of ~ 3 μW (below threshold) is shown in the inset of Fig. 2(a) , which is characterized by a sequence of sharp peaks superimposed on a broad quantum dot emission spectrum in the wavelength range of 1.1 – 1.3 μm. The observed optical modes, separated by ~ 20 meV, are identified to be the azimuthal modes, which satisfy the phase matching for resonance around the tube circumference. The associated azimuthal mode numbers (m) are denoted. With the increase of pump power, the peak intensity increases drastically. Illustrated in Fig. 2(a) are the optical resonance modes associated with the quantum dot ground state transitions measured at a pump power well above the threshold. It is seen that the quantum dot background emission becomes essentially negligible, compared to the peak intensity. The dominant lasing wavelengths are 1193.6, 1216.5, and 1240.7 nm, respectively.
Variations of the peak intensity versus pump power were also measured. Plotted in Fig. 2(b) is the integrated intensity of the emission peak at 1240.7 nm (m = 37) as a function of the pump power, which is the estimated power absorbed by the device. The solid curve is used as a guide to the experimental data points. An extremely low threshold (~ 4 μW) was estimated. Other lasing modes also exhibit similar threshold behavior. The achievement of such a low threshold is attributed to the use of superior quality quantum dot layers in the active region as well as the very small optical loss in the cavity, due to the epitaxially smooth tube surface. From detailed experimental and theoretical analysis, it has been confirmed that coherent emission from the microtube device takes place primarily at the inside edge of the tube [17, 18]. In this experiment, the microtube was formed such that the inside edge was positioned away from the top of the tube, where light emission was collected. Consequently, only a very small portion of the emitted photons were detected when the device was pumped above threshold. This is believed to be the primary reason for the observed “soft” threshold behavior. The variation of the spectral linewidth versus pump power for the same lasing mode was shown in the inset of Fig. 2(b). The linewidth decreases from ~ 0.6 – 0.8 nm to approximately 0.4 – 0.5 nm with the increase of the pump power, due to the increase of temporal coherence . The reduction of the spectral linewidth is in excellent agreement with the measured light-light characteristics, clearly confirming the achievement of lasing in rolled-up InGaAs/GaAs quantum dot microtubes. A small increase of the spectral linewidth with the increase of excitation power is seen, potentially due to heating effect. Also illustrated in Fig. 2(b) is a detailed view of the optical resonance mode at ~ 1240.7 nm above threshold. It is seen that the mode is highly asymmetric. In rolled-up tube structures, due to the spiral symmetry associated with the presence of inside and outside edges, each azimuthal resonance mode is broken into two nondegenerate ones . By fitting the peak using two Lorentzians, we derived an intrinsic spectral linewidth of ~ 0.2 – 0.3 nm, which is largely limited by the resolution of the spectrometer used in this experiment.
We have further investigated the polarization properties of the coherent emission. As defined in the inset of Fig. 3 , for photons circulating around the periphery of the tube, electric fields of the TE and TM modes are parallel and normal to the tube surface, respectively. The polarization measurements were performed by inserting a linear polarizer in the optical beam path. The microtube and polarizer were carefully aligned such that 0° and 90° correspond to TE and TM polarizations, respectively. The peak intensity was then recorded by varying the polarization angle. Plotted in Fig. 3 is the intensity of the lasing mode at 1240.7 nm as a function of the polarization angle. It is seen that the laser emission is primarily TE polarized. This observation is also consistent with recent theoretical and experimental studies that only TE optical modes, with an electric field parallel to the tube surface, can be supported by a rolled-up microtube ring resonator with a relatively thin wall (~ 40 – 200 nm) [9, 18, 20].
The confined optical modes in rolled-up microtube lasers were studied using the finite-difference time domain method. Shown in Fig. 4(a) are the simulated azimuthal mode profiles for photons (m = 37) confined in a rolled-up tube with a diameter of ~5.6 μm and wall thickness of ~ 100 nm. It is seen that coherent emission from rolled-up microtubes is predominantly determined by the photon scattering occurred at the inside edge [17, 18]. The calculated Q-factor is > 14,000, which is primarily limited by the optical scattering at the inside and outside edges and, in practice, any irregularities on the surface of the tube as well. This unique phenomenon is enormously important for achieving micro- and nanoscale lasers with controlled emission direction and output efficiency that are generally difficult to realize using photonic crystal, microdisk, and toroidal based optical cavities.
The optical resonance modes in rolled-up microtubes are also strongly influenced by the presence of surface corrugations [11, 14–16]. In the simplest case, we can approximate the microtube structure as a planar dielectric waveguide with periodic boundary conditions. By solving the Helmholtz equation for photons propagating through the waveguide, we can then derive axial field distributions for a given microtube surface geometry. Detailed analysis is described elsewhere . The first two axial field distributions associated with each azimuthal optical mode confined in the rolled-up microtubes are schematically shown in Fig. 4(b), which explains the observed higher order modes near the dominant azimuthal modes in the emission spectrum (Fig. 2(a)). Based on this consideration, the optical modes at wavelengths of 1191.4, 1193.6, 1214.2, 1216.5, 1238.3, 1240.7 nm, shown in Fig. 2(a), are denoted as TE39,1, TE39,0, TE38,1, TE38,0, TE37,1, and TE37,0 modes, with the first and second subscripts representing the azimuthal and axial mode numbers, respectively. It is also evident that control of the lasing modes can possibly be achieved in rolled-up micro- and nanotube lasers by varying the tube surface geometry. The mode competition amongst various azimuthal modes may not be significant, since these modes are separated by ~ 20 meV, which are larger than the homogeneous linewidth of a single dot (10 – 15 meV) at room temperature. However, strong mode competition for the various axial modes associated with the same azimuthal mode number is expected to occur, due to their small (2 – 6 meV) separation in energy . This is evidenced by the presence of a dominant mode associated with each azimuthal mode number, with other axial modes essentially suppressed, shown in Fig. 2(a).
The modification of spontaneous emission by the optical cavity is described by the Purcell factor. The upper limit for the Purcell factor can be estimated using F=3λ3Q/4π2n3Vmode, where λ is the operation wavelength, n is the refractive index, and Vmode is the mode volume of the cavity . In this experiment, the maximum intrinsic Q-factor measured under low pump power is ~ 3,500, limited by the absorption of the quantum dots. Therefore, the maximum Purcell factor is estimated to be ~ 12. Larger Purcell factors can be achieved in rolled-up nanotubes with much smaller diameters and consequently greatly reduced mode volumes. Work is currently in progress in achieving lasing in rolled-up quantum dot nanotubes as well as electrically injected devices.
In summary, we have demonstrated lasing in rolled-up InGaAs/GaAs quantum dot microtubes under optical pumping at room temperature. An ultralow threshold of ~ 4 μW was achieved in tubes with a diameter of ~ 5 – 6 μm and wall thickness of ~ 100 nm. The observed lasing modes are determined by both the tube diameter as well as the tube surface geometry. Such novel micro- and nanotube lasers may emerge as a new class of nanophotonic devices for applications in chip-level optical interconnects, single photon generation, and chemical and biochemical sensors.
The authors wish to gratefully thank P. Bhattacharya at the Univ. of Michigan for providing some of the quantum dot samples. This work is being supported by the Natural Sciences and Engineering Research Council of Canada (Grant No. 355628-08) and Fonds de recherché sur la nature et les technologies (Grant No. 2009-NC-125840).
References and links
1. H. Altug, D. Englund, and J. Vuckovic, “Ultrafast photonic crystal nanocavity laser,” Nat. Phys. 2(7), 484–488 (2006). [CrossRef]
2. Q. Song, H. Cao, S. T. Ho, and G. S. Solomon, “Near-IR subwavelength microdisk lasers,” Appl. Phys. Lett. 94(6), 061109 (2009). [CrossRef]
3. S. Reitzenstein, A. Bazhenov, A. Gorbunov, C. Hofmann, S. Munch, A. Loffler, M. Kamp, J. Reithmaier, V. Kulakovskii, and A. Forchel, “Lasing in high-Q quantum-dot micropillar cavities,” Appl. Phys. Lett. 89(5), 051107 (2006). [CrossRef]
4. P. Pauzauskie and P. Yang, “Nanowire photonics,” Mater. Today 9(10), 36–45 (2006). [CrossRef]
5. P. Bhattacharya and Z. Mi, “Quantum-dot optoelectronic devices,” Proc. IEEE 95(9), 1723–1740 (2007). [CrossRef]
6. S. Ates, S. Ulrich, P. Michler, S. Reitzenstein, A. Loffler, and A. Forchel, “Coherence properties of high-beta elliptical semiconductor micropillar lasers,” Appl. Phys. Lett. 90(16), 161111 (2007). [CrossRef]
7. S. Chakravarty, P. Bhattacharya, and Z. Mi, “Electrically injected quantum-dot photonic crystal microcavity light-emitting arrays with air-bridge contacts,” IEEE Photon. Technol. Lett. 18(24), 2665–2667 (2006). [CrossRef]
8. S. Mendach, R. Songmuang, S. Kiravittaya, A. Rastelli, M. Benyoucef, and O. Schmidt, “Light emission and wave guiding of quantum dots in a tube,” Appl. Phys. Lett. 88(11), 111120 (2006). [CrossRef]
10. V. Prinz, V. Seleznev, A. Gutakovsky, A. Chehovskiy, V. Preobrazhenskii, M. Putyato, and T. Gavrilova, “Free-standing and overgrown InGaAs/GaAs nanotubes, nanohelices and their arrays,” Phys. E 6(1-4), 828–831 (2000). [CrossRef]
11. S. Vicknesh, F. Li, and Z. Mi, “Optical microcavities on Si formed by self-assembled InGaAs/GaAs quantum dot microtubes,” Appl. Phys. Lett. 94(8), 081101 (2009). [CrossRef]
12. M. Hosoda, Y. Kishimoto, M. Sato, S. Nashima, K. Kubota, S. Saravanan, P. Vaccaro, T. Aida, and N. Ohtani, “Quantum-well microtube constructed from a freestanding thin quantum-well layer,” Appl. Phys. Lett. 83(5), 1017–1019 (2003). [CrossRef]
13. R. Songmuang, A. Rastelli, S. Mendach, and O. Schmidt, “SiOx/Si radial superlattices and microtube optical ring resonators,” Appl. Phys. Lett. 90(9), 091905 (2007). [CrossRef]
15. Ch. Strelow, H. Rehberg, C. M. Schultz, H. Welsch, Ch. Heyn, D. Heitmann, and T. Kipp, “Optical microcavities formed by semiconductor microtubes using a bottlelike geometry,” Phys. Rev. Lett. 101(12), 127403 (2008). [CrossRef] [PubMed]
16. F. Li, S. Vicknesh, and Z. Mi, “Optical modes in InGaAs/GaAs quantum dot microtube ring resonators at room temperature,” Electron. Lett. 45(12), 645–U20 (2009). [CrossRef]
17. C. Strelow, H. Rehberg, C. Schultz, H. Welsch, C. Heyn, D. Heitmann, and T. Kipp, “Spatial emission characteristics of a semiconductor microtube ring resonator,” Phys. E 40(6), 1836–1839 (2008). [CrossRef]
18. M. Hosoda and T. Shigaki, “Degeneracy breaking of optical resonance modes in rolled-up spiral microtubes,” Appl. Phys. Lett. 90(18), 181107 (2007). [CrossRef]
19. J. Scheuer, W. M. J. Green, G. A. DeRose, and A. Yariv, “Lasing from a circular Bragg nanocavity with an ultrasmall modal volume,” Appl. Phys. Lett. 86(25), 251101 (2005). [CrossRef]
20. G. S. Huang, S. Kiravittaya, V. A. Bolaños Quiñones, F. Ding, M. Benyoucef, A. Rastelli, Y. F. Mei, and O. G. Schmidt, “Optical properties of rolled-up tubular microcavities from shaped nanomembranes,” Appl. Phys. Lett. 94(14), 141901 (2009). [CrossRef]