RESEARCH in silicon photonics has accelerated in the

Transcription

1 IEEE JOURNAL OF SELECTED TOPICS IN QUANTUM ELECTRONICS, VOL. 15, NO. 3, MAY/JUNE Single-Wavelength Silicon Evanescent Lasers Alexander W. Fang, Member, IEEE, Matthew N. Sysak, Member, IEEE, BrianR.Koch, Student Member, IEEE, Richard Jones, Member, IEEE, Erica Lively, Ying-Hao Kuo, Di Liang, Student Member, IEEE, Omri Raday, and John E. Bowers, Fellow, IEEE (Invited Paper) Abstract The silicon evanescent device platform provides electrically pumped active device functionality on a low-loss siliconon-insulator waveguide platform. We present here recent research in the area of single-wavelength silicon evanescent lasers that utilize distributed feedback, distributed Bragg reflector (DBR), and sampled grating (SG) DBR laser topographies. Index Terms Semiconductor lasers, silicon-on-insulator technology. I. INTRODUCTION RESEARCH in silicon photonics has accelerated in the last decade, with advances in high-speed silicon modulators [1] [7], photodetectors [8] [12], advanced passive waveguides, and optically pumped silicon lasers [13]. Recently, new heterogeneous integration architectures have been developed, allowing for alignment-free transfer of III V materials to silicon, to fulfill the need of electrically pumped lasers for use in silicon photonic ICs [14] [16]. These demonstrations consist of microring and Fabry Perot membrane lasers coupled to silicon waveguides outside of the cavity, and Fabry Perot, racetrack resonator, and mode-locked silicon evanescent lasers (SELs), where III V materials are placed in close proximity to a silicon waveguide, such that the mode overlaps both regions. All of these lasers, aside from the microring configuration, have random longitudinal mode selection, making them unsuitable for applications that require single-wavelength operation with precise wavelength selection. Fig. 1 shows the vision of a 1 Tb/s silicon transmitter, consisting of 25 single-wavelength lasers, externally modulated at 40 Gb/s and multiplexed together to create a 1 Tb/s wavelength-division-multiplexed data stream. In this paper, we review the recent advancements in the area Manuscript received November 1, 2008; revised January 19, First published May 19, 2009; current version published June 5, This work was supported in part by the Defense Advanced Research Projects Agency (DARPA)/Microsystems Technology Office and Air Resources Laboratory (ARL) under Award W911NF and Award W911NF , and in part by Intel Corporation. A. W. Fang, E. Lively, Y.-H. Kuo, D. Liang, and J. E. Bowers are with the Department of Electrical and Computer Engineering, University of California Santa Barbara, Santa Barbara, CA USA ( awfang@ece.ucsb.edu; elively@ece.ucsb.edu; yinghao.kuo@ece.ucsb.edu; dliang@ece.ucsb.edu; bowers@ece.ucsb.edu). M. N. Sysak, B. R. Koch, R. Jones, and O. Raday are with the Photonics Technology Laboratory, Intel Corporation, Santa Clara, CA USA ( matthew.n.sysak@intel.com; brian.r.koch@intel.com; richard.jones@ intel.com; omri.raday@numonyx.com). Color versions of one or more of the figures in this paper are available online at Digital Object Identifier /JSTQE Fig. 1. Silicon 1 Tb transmitter with 25 single-wavelength SELs externally modulated at 40 Gb/s. of single-wavelength SELs. In Section II, we review the silicon evanescent device platform including the basic waveguide structure, recent developments of quantum well intermixing, and different methods of forming gratings. Sections III V review distributed feedback (DFB) [17], distributed Bragg reflector (DBR) [18], and sampled grating (SG) DBR [19] lasers demonstrated on this platform. II. SILICON EVANESCENT DEVICE PLATFORM A. Platform Waveguide Structures Fig. 2 shows the waveguide cross section for a passive silicon rib waveguide, and the silicon evanescent waveguide in panels (a) and (b), respectively. The silicon evanescent waveguide consists of III V materials placed in close proximity such that the mode exists in both the silicon and III V regions, leading to efficient coupling to passive silicon rib waveguide regions and the ability to achieve electrically pumped gain, photodetection, electroabsorption, and phase modulation. The optical mode can be manipulated by changing the effective index of the silicon region by narrowing the waveguide width or height to push the mode into the III V region. The same can be done in the III V region in order to push the mode into the silicon region. This allows the confinement factors for different devices on the photonic IC to be individually optimized. In addition, lateral tapering of the III V mesa region allows the realization of low-loss adiabatic transitions from hybrid regions to passive regions X/$ IEEE

2 536 IEEE JOURNAL OF SELECTED TOPICS IN QUANTUM ELECTRONICS, VOL. 15, NO. 3, MAY/JUNE 2009 Fig. 2. Cross-section diagrams of (a) silicon rib waveguide, and (b) silicon evanescent waveguide. Fig. 3. Overview of the QWI process used for the hybrid laser. The three bandgaps realized are numbered 1, 2, and 3. (a) Implantation of P into InP buffer with SiN x mask to preserve the as-grown bandgap. (b) Diffusion of vacancies through QWs and barriers via RTA for bandgap 2. (c) Removal of InP buffer layer to halt intermixing. (d) Diffusion of vacancies via RTA for bandgap 3. (e) Removal of InP buffer layer and InGaAsP stop etch layer. B. Quantum Well Intermixing (QWI) The active region can be modified through QWI to allow for different bandgaps throughout the device. This adds an additional degree of design freedom such that optical gain, electroabsorption, and passive sections can be placed adjacent to one another while still using a planar III V epitaxial structure and a single bonding step. The QWI process implemented in this paper is based on implant-enhanced intermixing in combination with selective removal of an InP buffer. Details of the QWI process and the as-grown hybrid laser III V base structure are shown in Fig. 3(a) (e). Intermixing begins with phosphorous implantation into a patterned InP buffer as shown in Fig. 3(a). An SiN x dielectric protects certain regions of the III V from the implant to preserve the as-grown bandgap. The implant generates vacancies that are then diffused through the quantum wells and barriers via a rapid thermal anneal (RTA) [Fig. 3(b)]. These vacancies cause atoms in the wells and barriers to interdiffuse, modifying their potential profile, and hence bandgap. To stop the intermixing, the buffer layer that contains the vacancies can be selectively removed using wet etching [Fig. 3(c)]. Additional RTA steps are then used to continue bandgap shifting in regions where the InP buffer remains [Fig. 3(d) (e)]. There are several parameters that affect the intermixing performance. These include the implant species, dose, energy, and temperature, and the RTA temperature and time. The implant conditions selected include a dose of cm 2, energy of 100 kev, and temperature of 200 C. Using these implant conditions, the RTA time and temperature was selected based on two objectives. The first was to achieve a large separation between the photoluminescence (PL) of the optical gain (implant protected) and intermixed regions. A large PL separation between these two regions minimizes the optical loss associated with the Urbach tail of the bandgap in the intermixed sections. The second objective is to minimize parasitic PL shifts where intermixing is not desired. A shift in the PL wavelength of the gain section, for example, affects control over the operating wavelength range of the laser. Fig. 4(a) shows the change in PL of the intermixed and the implant-protected quantum well regions as a function of RTA temperature after a 240 s anneal. Higher RTA temperatures have larger PL shifts due to an increase in the vacancy diffusion constant. PL shifts of up to 190 nm can be achieved with a 240 s anneal at 775 C. While these large shifts are attractive for low optical loss, the high-temperature conditions have large parasitic PL shifts in regions where intermixing is not desired (>80 nm). Reducing the anneal temperature to 725 C results in PL shifts of >110 nm with a parasitic PL shift in the implantprotected regions of <30 nm. Further reductions of the anneal temperature show a maximum PL shift of 60 nm at 675 C, which is not sufficient to minimize the Urbach tail loss in tail in passive regions. Based on achieving a large separation between the intermixed and protected regions and minimizing the shift in the laser gain section PL, an RTA temperature of 725 Cwas selected. Fig. 4(b) shows the PL shift in the quantum wells as a function of RTA time with a fixed annealing temperature of 725 C. The initial rapid shift in the PL is related to the strong atomic concentration gradient between the wells and barriers in the asgrown material. However, as the RTA time increases and the driving force in the intermixing process is reduced, the rate of change in the PL peak slows. Eventually, the rate of PL shift in the implanted regions slows enough to become equivalent

3 FANG et al.: SINGLE-WAVELENGTH SILICON EVANESCENT LASERS 537 Fig. 4. (a) Photoluminescence peak shift in implanted and implant-protected regions for a fixed RTA time of 240 s and RTA temperatures between 675 C and 775 C. (b) PL shift as a function of anneal time for III V regions that have been protected from the implant, partially intermixed with the InP buffer layer removed, and fully intermixed at 725 C. (c) Normalized PL spectra from the three bandgaps utilized in the SGDBR devices. C. Grating Structures Fig. 5 illustrates three methods for creating gratings on the silicon evanescent device platform. Passive gratings can be fabricated by etching a surface corrugation on the top surface of the silicon waveguide. Hybrid gratings can be formed by patterning a surface corrugation at the bonding interface between the III V regions and the silicon on either the silicon surface, or III V surface. The grating strength, κ, can be three to four times larger for hybrid gratings with equal grating dimensions to their passive grating counterparts, since the field intensity at this interface is much stronger. Fig. 6 shows an SEM micrograph of a silicon hybrid grating. Since the mode profile differs between the hybrid waveguide and the passive grating, the use of passive silicon gratings in a laser-based device requires a taper that can introduce losses in the cavity. Even though each taper may have a very low single-pass loss of 1 db, this can lead to a round-trip cavity loss of 4 db for two tapers in a DBR structure. This leads to an increase in threshold current and a decrease in differential quantum efficiency. Hybrid gratings, on the other hand, have no modal discontinuity and only require a taper structure at the output of the laser. The hybrid grating also has the option of being used in an active grating in a DFB or as passive mirrors if quantum well intermixing is employed in DBR and SGDBR structures. The typical etch depths used in these grating structures are between 15 and 100 nm. In the case of shallow etch depths, III V hybrid gratings may have increased etch control through the use of specially designed epitaxial layer structures, such that the etch can be conducted with selective etches. In addition, processing gratings on a planar structure is simpler than etching a grating on top of a waveguide, due to resist coverage and the effects of grating processing on waveguide sidewall roughness. On the other hand, gratings etched in silicon can use standard CMOS lithography, leading to lithographically defined firstorder gratings and phase shifts within the grating. In addition, placement alignment can be avoided if all patterning is done on the silicon, leading to a simpler manufacturing process. In this paper, we look at all three type of grating architectures. to the shift in the implant-protected regions. At this point, the maximum band edge separation between different intermixed regions has been achieved, and further RTA is not useful in reducing optical loss associated with the Urbach. Based on achieving the maximum band edge separation between the intermixed and implant-protected regions in the sampled grating distributed Bragg reflector (SGDBR), an RTA time of 330 s was selected for mirror, phase, and taper regions. Fig. 4(b) also shows the shift in PL as a function of anneal time for the regions where an intermediate bandgap is desired. An RTA shift of 45 s was selected before the InP buffer is removed to achieve a 60 nm separation in the PL between these regions. Fig. 4(c) shows PL spectra from three bandgaps obtained using the intermixing process. Good uniformity of the PL fullwidth at half-maximum can be seen for all three bandgaps, indicating consistent material quality. III. DISTRIBUTED FEEDBACK SELS A. Device Topography The DFB-SELs consist of a 360-µm-long quarterwavelength-shifted hybrid silicon grating with a 238 nm periodicity, 71% duty cycle, and 25 nm etch depth. This results in a grating κ of 247 cm 1 and reflectivity peak at 1600 nm. The top of Fig. 7 shows the device layout. The DFB-SEL consists of a 14-µm-wide and 200-µm-long gain region. N contacts are placed adjacent to the III V mesa while p contacts are placed on top of the mesa. The probe pads are not drawn on this diagram for simplicity but are shown in the microscope image. 80-µm-long tapers are formed by linearly narrowing the III V mesa region above the silicon waveguide. This adiabatically transforms the mode from the hybrid waveguide to the passive silicon waveguide allowing for losses of the order of

4 538 IEEE JOURNAL OF SELECTED TOPICS IN QUANTUM ELECTRONICS, VOL. 15, NO. 3, MAY/JUNE 2009 Fig. 5. Longitudinal cross structure and mode intensity profile of a silicon hybrid grating (left panel), a III V hybrid grating (center panel), and a passive silicon grating (right panel). The lasing spectrum is taken by dicing off the right photodetector, polishing, and antireflection coating the silicon waveguide output facet. Light is collected with a lensed fiber into a spectrum analyzer with a 0.08-nm-resolution bandwidth. Fig. 9 shows the spectrum with a 10 nm span with the laser being driven at 90 ma. The laser has a lasing peak of nm and a side-mode suppression ratio of 50 db. It can be seen from the inset that the laser operates single mode over a 100 nm span. The laser line width is measured by using the delayed-self heterodyne method [23]. Fig. 10 shows the minimum measured line width at a laser output power of 1.8 mw with a convoluted Lorentzian line width of 7.16 MHz corresponding to 3.6 MHz, a typical value for commercial DFB lasers. Fig. 6. SEM image of a silicon hybrid grating. 1.2 db per taper and reflections of the order of [20]. Two electrically pumped tapers are placed on both ends of this gain region, giving way for a small amount of optical gain. Silicon evanescent photodetectors are placed on both sides of the laser in order to enable on-chip testing of the DFB-SEL performance. The photodetectors are 240 µm long including the two 80-µm-long tapers. The detector to the right is placed 400 µm away in order to allow room for dicing and polishing for off-chip spectral tests. B. Device Performance The light-current (L I) characteristics of the DFB-SEL are measured on chip by collecting light out of both sides of the laser with integrated silicon evanescent photodetectors. To determine the laser power output, we assume 100% internal quantum efficiency of the photodetectors [21] in order to conservatively assess the laser performance. It can be seen from Fig. 8 that at 10 C, the lasing threshold is 25 ma, with a maximum output power of 5.4 mw. This corresponds to a threshold current density of 1.4 KA/cm 2, which is slightly lower than the 2 and 1.7 KA/cm 2 seen in previously demonstrated Fabry Perot [14] and Racetrack [22] SELs, respectively. The maximum lasing temperature is 50 C. The secondary y-axis of Fig. 8 shows the voltage current curve with a laser turn ON of 1.8 V. The laser has a 13 Ω device series resistance. This value scales appropriately with the 4.5 Ω resistance measured on 800-µm-long Fabry Perot lasers with similar III V mesa dimensions. IV. DISTRIBUTED BRAGG REFLECTOR SELS A. Device Topography The device topography (Fig. 11) consists of two passive Bragg reflector mirrors placed 600 µm apart to form an optical cavity. The gratings have an etch depth and duty cycle of 25 nm and 75 %, respectively, leading to a grating strength, κ,of80cm 1. The back and front mirror lengths are 300 and 100 µm, resulting in power reflectivity of 97% and 44%, respectively. A 440-µmlong silicon evanescent gain region and two 80-µm-long tapers are placed inside the cavity. The tapers are electrically driven in parallel with the gain region in order to minimize absorption. B. Continuous-Wave (CW) Device Performance The CW laser output power is measured with an integrating sphere at the front mirror of the laser. The front mirror output power current voltage (L I V ) characteristic is shown in Fig. 12. The device has a lasing threshold of 65 ma, a maximum front mirror output power of 11 mw, leading to a differential quantum efficiency of 15%. The taper transmission loss can have a significant impact on the threshold current, and therefore, it affects many important laser characteristics such as wall plug efficiency and resonance frequency. If we use our estimations of the material and laser properties, calculations show that the taper loss of 1.2 db increases the threshold current by a factor of 2 due to the accumulated loss through four taper transitions in one round-trip through the cavity. We estimate that a reduction in the single-pass taper losses to 0.5 db would reduce this factor to 1.2. The laser operates up to a stage temperature of 45 C. The kinks in the L I are from mode hopping and will be discussed later. The device has a lasing turn-on voltage of

5 FANG et al.: SINGLE-WAVELENGTH SILICON EVANESCENT LASERS 539 Fig. 7. (top) DFB-SEL device layout. (bottom) Microscope image of DFB-SEL and integrated silicon evanescent photodetectors. Fig. 8. L I V curve for stage temperatures of 10 Cto50 C. Fig. 9. DFB lasing spectrum at 90 ma injection current with a 50 db sidemode extinction ratio. (inset) The lasing spectrum over a 100 nm span showing single-mode lasing. 2.6 V and a series resistance of 11.5 Ω. The lasing spectrum is shown in Fig. 13 with a lasing peak at nm when driven at 200 ma. The device has a free spectral range of 0.47 nm, which corresponds to a group index of 3.86 based on the sum of the physical cavity length and mirror penetration depths of 61 and 42 µm. The side-mode suppression ratio is 50 db. Fig. 14 shows the lasing spectrum as a function of drive current along with the corresponding L I curve. Note that the output power in this case is fiber-coupled, which is 5dBlower than the total output power measured earlier in Fig. 12. It can be seen that as the device heats with larger current injection, the lasing mode moves to longer wavelengths due to the thermooptic effect in the cavity. When the mode moves far enough from the reflection peak, a longitudinal mode hop to another mode occurs. The mode hopping appears to be chaotic between stable Fig. 10. power. Delayed-self heterodyned line width trace at 1.8 mw laser output

6 540 IEEE JOURNAL OF SELECTED TOPICS IN QUANTUM ELECTRONICS, VOL. 15, NO. 3, MAY/JUNE 2009 Fig. 11. DBR-SEL top-view topographical structure (top) and microscope image (bottom). Fig. 12. DBR-SEL L I V curve for various temperatures measured out of the front mirror. mode positions and the reason for the exact hopping pattern we observed is unknown at this time. C. Direct Modulation We measured the modulation characteristics of this device by using a bias-t to drive the laser simultaneously with a dc bias current and an RF signal while measuring the electrooptic (EO) response on a photodetector. Fig. 15 shows the photodetected EO response of the laser combined with all connected components under small-signal modulation ( 10 dbm). S 11 measurements indicate that the electrical contact geometry is not limiting the performance of the device, so reflected power has not been factored out of these curves. A 2 pf device capacitance was extracted from the S 11 measurement, resulting in an RC-limited bandwidth of 7 GHz. Fig. 16(inset) shows the resonance frequency versus the square root of dc drive current above threshold, which has a roughly linear dependence as expected. Under higher modulation powers, the resonance peak becomes significantly dampened. The 3 db electrical bandwidth at 105 ma is 2.5 GHz. Fig. 13. Optical spectrum of the DBR-SEL driven at 200 ma. We directly modulated the laser biased at 105 ma dc current with a 2.5 Gb/s, pseudo-random bit sequence (PRBS) electrical signal with 20 mw of RF power. The resulting eye diagram is shown in Fig. 16(a). The extinction ratio is 8.7 db and the fiber-coupled output power is 0.7 mw when cooled to 18 C. Although the output power and modulation bandwidth increases at higher dc currents, the peak-to-peak (P P) current swing corresponds to a lower P P output power swing at higher current biases (Fig. 12). This leads to lower extinction ratios at higher current biases unless larger RF modulation powers are used. For example, Fig. 16(b) shows an eye diagram at 4 Gb/s that can be obtained with a dc bias of 135 ma and an RF power of 39 mw. However, in this case, the extinction ratio is closer to 5.5 db. Improving the laser design to decrease the threshold current and increase the differential gain is expected

7 FANG et al.: SINGLE-WAVELENGTH SILICON EVANESCENT LASERS 541 Fig. 14. Fiber-coupled L I curve and spectrum versus current at a stage temperature of 18 C. Fig. 16. Eye diagrams of a directly modulated DBR-SEL at (a) 2.5 Gb/s and (b) 4 Gb/s. 0.3 µm. The buried oxide layer is 1 µm thick. The measured κ and loss of the III V gratings was 165 and 69 cm 1, respectively, for a 2-µm-wide Si waveguide and 100 nm etch depth (into the III V). Fig. 15. Photodetected frequency response of the DFB-SEL for three different bias currents with a stage temperature of 18 C and (inset) plot of resonance frequency versus the square root of current above threshold. to significantly improve the modulation bandwidth in future devices. V. SAMPLED GRATING DBR SELS A. Device Topography The SGDBR utilizes quantum well intermixing [24], [25] in order to maintain a hybrid waveguide inside the cavity of the laser, while minimizing the absorption over the tunable phase and mirror sections. The adiabatic taper transition can then be placed outside the cavity, leading to only a single-pass power loss, as opposed to a four-pass power loss if two tapers were placed inside the cavity. The SG-DBR consists of five electrically isolated sections. These sections include a 650-µm-long back mirror, 80-µm-long phase section, 550-µm-long gain region, a 270-µm-long front mirror, and a 100-µm-long taper. The back mirror has fourteen 7.6-µm-wide grating bursts and a 46.4-µm sampling period (Fig. 17). The front mirror has five 5.2-µm-wide grating bursts with a sampling period of 52.4 µm. Each section of the laser is separated by a 10-µm-wide proton-implanted region for electrical isolation. The measured resistance between neighboring sections was >10 kω. The silicon epitaxial layer beneath the III V mesa is 0.4 µm thick and the waveguide is 1.0 µm thick with an etch depth of B. Device Performance The SGDBR laser CW L I characteristics are shown in Fig. 18. The optical power is measured using an integrating sphere to minimize fiber coupling uncertainty. CW lasing is achieved up to 30 C with gain section resistance of 14 Ω, output power up to 1.0 mw, and a threshold current of 48 ma. The low output power is a result of the high loss in the etched gratings. The dips in the L I characteristics shown in Fig. 18 are a result of temperature-induced cavity mode hops [24] as shown by the lasing spectra from the SGDBR as a function of gain current in Fig. 19. As the bias current is increased, the temperature of the gain section increases resulting in the cavity modes shifting to longer wavelengths. As the lasing mode moves across the mirror stop band, it eventually comes into competition with a neighboring cavity mode at a lower wavelength and a mode hop occurs. Overlaid output spectra from the SGDBR with current injection into both front and back mirrors are shown in Fig. 20. Tuning over three cavity modes can be achieved between 1501 and 1514 nm with side-mode suppression > 30 db. The limited tuning range in Fig. 20 comes from several factors, including a small quantum well confinement factor (3%) and a large device thermal impedance since the SGDBR is fabricated using silicon-on-insulator material. Increased confinement factor, as suggested by [27] would increase the tuning range. The increase in amplified spontaneous emission at lower wavelengths in Fig. 8 is a result of reduced loss in the front mirror when current is injected for tuning. The tuning can be improved

8 542 IEEE JOURNAL OF SELECTED TOPICS IN QUANTUM ELECTRONICS, VOL. 15, NO. 3, MAY/JUNE 2009 Fig. 17. Cut away of a hybrid silicon SGDBR shown with four front mirror and back mirror grating bursts. Proton implantation is used for electrical isolation between various laser sections. The active (PL = 1520 nm) and passive (PL = 1440 nm) bandgaps are labeled 1 and 2, respectively. Fig. 18. Measured laser output power and voltage as a function of drive current for various state temperatures. Fig. 20. Spectra from SGDBR tuned to three supermodes. Gain section bias is 140 ma with backside temperature of 25 C. Front and rear mirror tuning currents for lasing at 1501, , and 1514 nm are 20 and 0 ma, 0 and 0 ma, and 0 and 20 ma, respectively. TABLE I SUMMARY OF SINGLE-WAVELENGTH SEL PERFORMANCE Fig. 19. Contour map of fiber-coupled SGDBR spectra as a function of gain current. Stage temperature is held fixed at 25 C and front and rear mirrors are unbiased. by increasing the number of quantum wells and introducing a tuning layer into the III V base structure [28]. SELs with a single-wavelength output using DFB, DBR, and SGDBR topographies have been reviewed here. The key device parameters are shown in Table I. The design tradeoffs between the DFB and DBR topography can clearly be seen, where the DFB has lower thresholds, due to the high kappa hybrid grating leading to shorter device lengths, and the absence of an active passive taper inside the cavity. Thresholds below 10 ma should be possible. The DBR, on the other hand, can be designed to be longer, while still achieving a higher differential quantum efficiency over the DFB such that maximum powers of 11 mw were demonstrated and 30 mw should be possible. The SGDBRs utilized quantum well intermixing in grating regions, allowing for tapers to be placed outside the cavity, allowing for lower thresholds than the DBR design, but suffer lower output powers due to higher losses in the passive hybrid gratings. With improvements in the design, a tuning range of 100 nm in the SGDBRs should be possible. The demonstrations of these three types of single-wavelength lasers provide multiple methods to create laser sources for the 1 Tb/s transmitter shown in Fig. 1. The major elements of this transmitter, the AWG, the high-speed modulator, and the single-wavelength laser, have all been demonstrated and can be integrated to realize the vision of creating a 1 Tb/s silicon transmitter. ACKNOWLEDGMENT The authors would like to thank D. Blumenthal, L. Coldren, H. Park, A. Ramaswamy, L. Johansson, M. Haney, J. Shah, and W. Chang for insightful discussions and B. Kim, and H.-W. Chen for help with device fabrication.

9 FANG et al.: SINGLE-WAVELENGTH SILICON EVANESCENT LASERS 543 REFERENCES [1] A. Liu, L. Liao, D. Rubin, J. Basak, H. Nguyen, Y. Chetrit, R. Cohen, N. Izhaky, and M. Paniccia, High- speed silicon modulator for future VLSI interconnect, presented at the Integr. Photon. Nanophoton. Res. Appl., OSA Tech. Dig. (CD), Optical Society of America, 2007, Paper IMD3. [2] A. S. Liu, R. Jones, L. Liao, D. Samara-Rubio, D. Rubin, O. Cohen, R. Nicolaescu, and M. Paniccia, A high-speed silicon optical modulator based on a metal-oxide semiconductor capacitor, Nature, vol. 427, pp , [3] D. Marris-Morini, X. Le Roux, L. Vivien, E. Cassan, D. Pascal, M. Halbwax, S. Maine, S. Laval, J. M. Fedeli, and J. F. Damlencourt, Optical modulation by carrier depletion in a silicon PIN diode, Opt. Exp., vol. 14, pp , [4] Y. Q. Jiang, W. Jiang, L. L. Gu, X. N. Chen, and R. T. Chen, 80-micron interaction length silicon photonic crystal waveguide modulator, Appl. Phys. Lett., vol. 87, pp , [5] Q. Xu, B. Schmidt, S. Pradhan, and M. Lipson, Micrometre-scale silicon electrooptic modulator, Nature, vol. 435, pp , [6] R. S. Jacobsen, K. N. Andersen, P. I. Borel, J. Page-Pedersen, L. H. Frandsen, O. Hansen, M. Kristensen, A. V. Lavrinenko, G. Moulin, H. Ou, C. Peucheret, B. Zsidri, and A. Bjarklev, Strained silicon as a new electrooptical material, Nature, vol. 441, pp , [7] J. E. Roth, O. Fidaner, R. K. Schaevitz, Y.-H. Kuo, T. I. Kamins, J. S. Harris, Jr., and D. A. B. Miller, Optical modulator on silicon employing germanium quantum wells, Opt. Exp., vol. 15, pp , [8] D. Ahn, C.-Y. Hong, J. Liu, W. Giziewicz, M. Beals, L. C. Kimerling, J. Michel, J. Chen, and F. X. Kärtner, High performance, waveguide integrated Ge photodetectors, Opt. Exp., vol. 15, pp , [9] J. Liu, D. Ahn, C.-Y. Hong, M. Beals, L. C. Kimerling, J. Michel, J. Chen, F. X. Kärtner, A. Pomerene, D. Carothers, C. Hill, J. Beattie, K. Tu, Y. Chen, S. Patel, M. Rasras, A. White, and D. Gill, Waveguide-integrated Ge photodetectors on Si for electronic and photonic integration, presented at the Integr. Photon. Nanophotonics Res. Appl. (IPNRA), Salt Lake City, UT, 2007, Paper ITuE2. [10] G. Masini, G. Capellini, J. Witzens, and C. Gunn, A four-channel, 10 Gbps monolithic optical receiver in 130nm CMOS with integrated Ge waveguide photodetectors, presented at the Opt. Fiber Commun. Conf. (OFC), Anaheim, CA, 2007, Paper PDP 31. [11] A. O. Splett, T. Zinke, B. Schueppert, K. Petermann, H. Kibbel, H. Herzog, and H. Presting, Integrated optoelectronic waveguide detectors in SiGe for optical communications, Proc. SPIE, vol. 2550, pp , [12] M. W. Geis, S. J. Spector, M. E. Grein, R. T. Schulein, J. U. Yoon, D. M. Lennon, S. Deneault, F. Gan, F. X. Kaertner, and T. M. Lyszczarz, CMOS-compatible all-si high-speed waveguide photodiodes with high responsivity in near-infrared communication band, IEEE Photon. Technol. Lett., vol. 19, no. 3, pp , Feb [13] H. Rong, A continuous-wave Raman silicon laser, Nature, vol. 433, pp , [14] A. W. Fang, H. Park, O. Cohen, R. Jones, M. J. Paniccia, and J. E. Bowers, Electrically pumped hybrid AlGaInAs-silicon evanescent laser, Opt. Exp., vol. 14, pp , [15] J. Van Campenhout, P. R. Romeo, P. Regreny, C. Seassal, D. Van Thourhout, S. Verstuyft, L. Di Cioccio, J.-M. Fedeli, C. Lagahe, and R. Baets, Electrically pumped InP-based microdisk lasers integrated with a nanophotonic silicon-on-insulator waveguide circuit, Opt. Exp.,vol.15, pp , [16] T. Maruyama, T. Okumura, S. Sakamoto, K. Miura, Y. Nishimoto, and S. Arai, GaInAsP/InP membrane BH-DFB lasers directly bonded on SOI substrate, Opt. Exp., vol. 14, pp , [17] A. W. Fang, E. Lively, Y.-H. Kuo, D. Liang, and J. E. Bowers, Distributed feedback silicon evanescent laser, presented at the OFC/NFOEC, 2008, Paper PDP15. [18] A. W. Fang, B. R. Koch, R. Jones, E. Lively, D. Liang, Y.-H. Kuo, and J. E. Bowers, A distributed Bragg reflector silicon evanescent laser, IEEE Photon. Technol. Lett., vol. 20, no. 20, pp , Oct. 15, [19] M. N. Sysak, J. O. Anthes, J. E. Bowers, O. Raday, and R. Jones, Integration of hybrid silicon lasers and electroabsorption modulators, Opt. Exp., vol. 16, pp , [20] H. Park, Y.-H. Kuo, A. W. Fang, R. Jones, O. Cohen, M. J. Pannicia, and J. E. Bowers, A hybrid AlGaInAs silicon evanescent preamplifier and photodetector, Opt. Exp., vol. 15, no. 21, pp , [21] H. Park, A. W. Fang, R. Jones, O. Cohen, O. Raday, M. N. Sysak, M. J. Paniccia, and J. E. Bowers, A hybrid AlGaInAs-silicon evanescent waveguide photodetector, Opt. Exp., vol. 15, pp , [22] A. W. Fang, R. Jones, H. Park, O. Cohen, O. Raday, M. J. Paniccia, and J. E. Bowers, Integrated AlGaInAs-silicon evanescent race track laser and photodetector, Opt. Exp., vol. 15, pp , [23] D. Derrickson, Fiber Optic Test and Measurement. Englewood Cliffs, NJ: Prentice-Hall, 1998, pp [24] J. Marsh, Quantum well intermixing, Semicond. Sci. Technol., vol. 8, pp , [25] E. Skogen, J. Barton, S. Denbaars, and L. Coldren, A quantum-wellintermixing process for wavelength-agile photonic integrated circuits, IEEE J. Select. Topics Quantum Electron., vol. 8, no. 4, pp , Jul./Aug [26] V. Jayaraman, Z. Chuang, and L. A. Coldren, Theory, design, and performance of extended tuning range semiconductor lasers with sampled gratings, IEEE J. Quantum Electron., vol. 29,no.6,pp ,Jun [27] A. Yariv and X. Sun, Supermode Si/III V hybrid lasers, optical amplifiers and modulators: A proposal and analysis, Opt. Exp., vol. 15, pp , [28] M.-C. Amann, S. Illek, C. Schanen, and W. Thulke, Tunable twin-guide laser: A novel laser diode with improved tuning performance, Appl. Phys. Lett., vol. 54, pp , Alexander W. Fang (S 05) received the B.S. degree in electrical engineering with minors in physics and mathematics from San Jose State University, San Jose, CA, in 2003, and the M.S. and Ph.D. degrees in electrical engineering from the University of California, Santa Barbara, in 2005 and 2008, respectively. He is the Co-Founder of Aurrion, LLC. His current research interests include the heterogeneous integration of III V materials with silicon for in-plane lasers. He is the author or coauthor of more than 60 journal and conference papers. Dr. Fang is a member of the Optical Society of America (OSA). America (OSA). Matthew N. Sysak (M 03) was born in Smithtown, NY, in He received the B.S. degree in chemical engineering from Pennsylvania State University, Philadelphia, in 1998, and the M.S. and Ph.D. degrees from the University of California, Santa Barbara, in 2001 and 2005, respectively. Currently, he is with Photonics Technology Laboratories, Intel Corporation, Santa Clara, CA. His current research interests include III V and silicon integration for novel optoelectronic devices. Dr. Sysak is a member of the Optical Society of Brian R. Koch (S 06) received the B.S. degree in physics from the College of William and Mary, Williamsburg, VA, in 2003, and the M.S. and Ph.D. degrees in electrical and computer engineering from the University of California, Santa Barbara, in 2004 and 2008, respectively. He is currently an Intern at Intel s Photonics Technology Laboratory, in Santa Clara, CA, where he is enaged in testing and designing hybrid silicon evanescent lasers. His dissertation was focused on modelocked lasers specifically designed for use in photonic ICs, for applications in optical clock recovery and optical signal regeneration.

10 544 IEEE JOURNAL OF SELECTED TOPICS IN QUANTUM ELECTRONICS, VOL. 15, NO. 3, MAY/JUNE 2009 Richard Jones (M 05) received the B.Sc. degree in physics from the Imperial College, London University, London, U.K., in 1993, the M.Sc. degree in microwaves and optoelectronics from the University College, London University, in 1994, and the Ph.D. degree in physics from the Imperial College, London University, in He has been a Senior Optical Researcher in the Photonic Technology Laboratory, Intel Corporation, Santa Clara, CA, since His current research interests in silicon photonics include nonlinear optics, and active optical components for communication systems and optical sensing. Dr. Jones is a member of the Optical Society of America (OSA) and the American Chemical Society (ACS). Omri Raday received the B.S. degree in physics and the M.Sc. degree in applied physics from the Hebrew University, Jerusalem, Israel, in 2002 and 2006, respectively. He is currently a Researcher in the Photonics Technology Laboratory (formerly Numonyx), Intel Corporation, Santa Clara, CA. Since 2005, he has been involved in process development in the fields of photolithography, reactive ion etching (RIE), plasma enhanced chemical vapor deposition (PECVD), and thin films metrology. His current research interests include CMOS-compatible processing development of optical devices. Erica Lively received the B.S. degree from the University of Idaho, Moscow, in 2005, and the M.S. degree from the University of California, Santa Barbara, in 2007, both in electrical engineering. She is currently working toward the Ph.D. degree at the University of California. Her current research interests include the fabrication and simulation of metamaterials, slow light devices, and nanophotonic devices. She is also interested in the societal implications of science and technology. Ying-Hao Kuo was born in Taiwan. He received the B.S. degree in physics from the National Taiwan University, Taipei, Taiwan, in 1997, and the Ph.D. degree in electrical engineering from the University of Southern California, Los Angeles, in His thesis project focused on electrooptical polymer devices and optical trimming using photobleaching. He spent a year as Intern at Intel Corporation, Santa Clara, CA, after graduation working on silicon photonics in the area of Raman lasers and wavelength conversion in silicon waveguides. John E. Bowers (F 93) received the M.S. and Ph.D. degrees from Stanford University, Stanford, CA. He was with AT&T Bell Laboratories and Honeywell before joining the University of California, Santa Barbara (UCSB). He is the Director of the Institute for Energy Efficiency, and a Professor in the Department of Electrical Engineering at UCSB. He is also the Co-Founder of Terabit Technology, Calient Networks, and Aurrion. He is the author or coauthor of more than six book chapters, 400 journal papers, and 600 conference papers. He is the holder of 48 patents. His current research interests include optoelectronic devices and optical networking. Prof. Bowers is a member of the National Academy of Engineering (NAE), and a Fellow of the Optical Society of America and the American Physical Society. He is the recipient of the IEEE Lasers and Electro-Optics Society (LEOS) William Streifer Award and the South Coast Business and Technology Entrepreneur of the Year Award. Di Liang (S 02) received the B.S. degree in optical engineering from Zhejiang University, Hangzhou, China, in 2002, and the M.S. and Ph.D. degrees in electrical engineering from the University of Notre Dame, Notre Dame, IN, in 2004 and 2006, respectively. He has authored or coauthored over 40 journal and conference papers. His current research interests include Si photonics, III V semiconductor diode lasers and photodiodes, and hybrid integration techniques and microelectronic fabrication.