Open Access
Add to BibSonomyAbstract
Detrimental surface recombination of carriers in InP-based photonic crystal nanobeams containing quantum wells is reduced by employing chemical treatment followed by silica encapsulation. Carrier lifetime is shown to recover to 2.63ns close to the bulk value. This enables us to obtain optically pumped room-temperature continuous-wave nanolasers at 1.55µm integrated onto Silicon on insulator waveguide platform with a threshold of 8µW.
© 2015 Optical Society of America
1. Introduction
Planar or 2-dimensional (2D) semiconductor photonic crystals (PhC) have been the subject of intense activity due to their ability to confine extensively the electromagnetic field (EM) [1], manipulate the dispersion [2] and enhance the nonlinearity [3].
Amongst the PhC-based devices, nanolasers have recently grabbed a lot of attention as they can unlock the deployment of on-chip optical interconnects thanks to their outstanding performance in terms of footprint, energy consumption and speed. It is remarkable that in less than 2 decades, many design concepts have been brought to fruition, but it is only recently that their use for applications is seriously envisaged thanks to advances in the tackling of the blocking issues: electrical driving [4–6], optical interfacing [7,8] and room temperature (RT) continuous-wave (CW) operation. CW operation has been problematic as these photonic nanostructures are made in thin semiconductor slabs, most of the time air-bridged, drilled at the wavelength-scale with an array of holes to provide the necessary high refractive index contrast between air and the material to obtain a quasi-extensive control over the EM field confinement [9]. Indeed, this particular material configuration comes at a high price, i.e. poor heat sinking and with large non radiative surface recombination of carriers, both highly detrimental to obtain RT CW operation. In order to overcome the hurdle of poor heat sinking, several groups have explored mainly two channels: the bonding to heat sinking substrates [10] and dielectric encapsulation [11], and the engineering of the constituting material of the PhC slab with higher thermal conductivity [12].
Regarding non radiative surface recombination of carriers, the plasma etching of the holes through the semiconductor slab during processing exposes a large surface of active material with an increased density of non-radiative recombination centres for carriers, due to the bombardment of high energy ions [13]. Moreover, as the typical dimension associated to the PhC structuration (~100nm) being well below the diffusion length of the carriers during their lifetime (~1µm), the effect of the increase of the surface recombination velocity is even more important and results in impressive carrier lifetime reductions from few ns to less than 10ps in GaAs-based PhC [14], to 200ps in InP-based PhCs [15].
Even though this can be advantageously implemented to achieve ultrafast switching [16], for laser emission, the consequences are nefarious. The large rate of non-radiative recombination participates in heating the material which implies a decrease in the radiative efficiency. Then, it also directly raises the laser threshold [17], as the injected carriers recombine before they can participate towards increasing the photon number in the cavity. This laser threshold rise leads also to further heating in the structure, possibly preventing the laser effect and at times inducing material degradation. This is, of course, particularly hampering in the CW regime.
In order to tackle this issue, two main approaches have been deployed, the first one consists in choosing active materials that are minimally affected by surface recombination like quantum dots (QDs) and quantum wires which confine carriers and suppress diffusion towards the surface [18,19]. However, QW-based active materials, which are often desirable as they exhibit higher gain and higher saturation power, are not of this type. The second approach, which suits QWs, makes use of specific processing technology as plasma [20] or chemical treatment [21] to eliminate centres of recombination.
In this paper, we demonstrate the recovery of the carrier lifetime close to the bulk level in InP-based PhC nanocavities embedding InGaAsP quantum wells (QWs) using a procedure based on chemical treatment followed by encapsulation. Thanks to this surface passivation, we demonstrate the stable RT CW laser operation around 1.55µm of InP-based nanocavities integrated onto Silicon on insulator (SOI) waveguide circuitry using optical pumping.
2. Design and fabrication of the hybrid nanolasers
The structures under investigation are InP-based PhC nanobeam cavities ensconcing 4 InGaAsP strained quantum wells designed to exhibit a high quality factor (Q>106) resonant mode close to 1.55µm where the active material gain peaks. They consist in a 505nm wide, 285nm thick ridge waveguide drilled with a row of equally sized cylindrical holes (radius~105nm) whose inter-distance is varied accordingly to [22]. As indicated in Fig. 1, the nanocavities lie on top of a ~450nm-thick low refractive index layer (bonding layer) deposited on 500nmx220nm SOI waveguides. This configuration enables the evanescent wave coupling of light between the cavities and the SOI waveguides [23]. The SOI wires are terminated with fiber grating couplers to interface the chip with cleaved SMF fibers.
The fabrication of the structures relies on the die-to-die adhesive bonding of the InP-based heterostructure on top of a SOI waveguide circuitry using divinylsiloxane-benzocyclobutene (DVS-BCB) diluted in mesitylene (Mes). Prior to the bonding, the III-V wafer is coated with 400nm thick layer of SiO2 in order to set the distance separating the III-V and the SOI to the desired value. As shown in Fig. 2(a), the DVS-BCB layer above the waveguide is as thin as 40nm. The bonded III-V semiconductor material is then patterned using electron beam lithography with Hydrogen silsesquioxane (HSQ) followed by inductively coupled plasma etching employing HBr/O2/He chemistry. HSQ is then removed using fluorine based reactive ion etching. As can be seen on Figs. 2(b) and 2(c), the nanobeam cavities have smooth vertical sidewalls with no undercut and are aligned to the subjacent waveguide with a precision better than 30nm [24].
Fig. 2 Scanning electron microscope pictures of the fabricated sample a) right after bonding, b) and c) after nanobeam cavity patterning.
To passivate the surface of the nanolasers in order to reduce the surface recombination, we use a procedure consisting in a 2-steps chemical treatment [25]. First, H2SO4:H2O2:H2O solution at a ratio 1:8:5000 is used during 10s to gently etch the surface of the InGaAsP material (QWs and barriers) with the goal to remove the material on the side walls damaged by the bombardment of the ions during the ICP etching. The sample is then plunged for 10min in a bath of (NH4)2S solution at a 10% concentration heated at 45°C which deoxidises the structures and fills the dangling bonds at their surface with sulphur atoms and remove the recombination centres [26].
To stabilise this process in time and prevent further re-oxidation, the sample is spin-coated with DVS-BCB:Mes solution (1:10) that was calibrated to fill completely the PhC holes. Finally, a 1µm-thick silica layer is deposited using plasma enhanced chemical vapour deposition in order to encapsulate the nanolasers and increase heat sinking as shown in [11].
3. Effect of the chemical surface passivation on the emission wavelength and on the carrier lifetime
As previously indicated, one of the effects of the chemical treatment is to gently etch the material. This has an immediate impact on the resonant wavelengths of our cavities and, hence, on their emission wavelengths, as the optical response of these structures is highly sensitive to any shift in size of their geometric features. In order to calibrate this etching, we measure the emission wavelength at RT of the nanolasers before and after the passivation process without BCB filling nor SiO2 encapsulation.
We optically pump the sample at 1180nm with a laser diode modulated to obtain 10ns pulses at a 250kHz repetition rate to ensure laser emission. The light outputting the SOI wires is collected with an SMF fibre and sent to a spectrometer. The emission wavelength is measured for pump powers above the laser threshold. In order to gain statistics, the measurement is performed on nanocavities defined with different central lattice constants, which gives different resonant wavelengths. As can be seen on Fig. 3, the passivation treatment induces a blue shift of the resonant wavelengths of about 15nm. This is in concordance with some matter removal. From complementary measurements, we find the dependence of the resonant wavelength as a function of the holes radius and the nanobeam width to be dλ/dr = −3.8 and dλ/dw = 1.25. By assuming that the matter is removed uniformly at the etched interfaces, the passivation process removes ~2nm of material.
The carrier lifetime is then measured at RT on the samples prior to and after the passivation treatment. We use a pump-probe set-up where the pump laser is a pulsed source at 810nm providing 100fs pulses at 80MHz repetition rate and the probe laser, synchronous with the pump, a source delivering 150fs pulses adjustable in wavelength to match the cavities resonances (in the 1.55µm range). The pump is focused on the cavity surface whereas the probe is injected into the SOI waveguides through the grating couplers. When the sample is pumped, carriers are created leading to a change in the refractive index of the material and to a consequent change in the resonant wavelength of the cavity. After the passage of the pump, the carriers recombine so the material refractive index recovers to its original value over the duration of the carrier lifetime. Thus, by following the shift of the resonance wavelength in the probe transmission spectrum as a function of the delay between the pump and the probe, the entire carrier dynamics may be traced.
In our experiment, the pump power is chosen to be well below the laser threshold to induce a blue shift of the resonant wavelength of about 1nm. This way, stimulated emission is absent during carrier recombination. The wavelength shift of the resonances of unpassivated and passivated nanocavities is plotted on Fig. 4 (a) as a function of the pump probe delay. As can be clearly seen, the resonance recovers its initial wavelength for a much greater pump-probe delay for the passivated cavity indicating a large improvement of the carrier lifetime. In both cases, the measurements can be well fitted with single exponential decays meaning that, for the chosen pump power, non-radiative recombination (rate proportional to the carrier density N) dominates the radiative ones (rate proportional to N2). Consequently, we obtain the non-radiative carrier lifetime, τnr, to be 370ps for the untreated sample and 2.63ns for the passivated ones, showing an improvement of about one order of magnitude. We calculate the associated surface recombination velocities, vs, by using the following relation:
with V and As are respectively the volume and area of the active volume. For the unpassivated sample, we obtain vs = 2.84x104cm.s−1, which is of the same order of magnitude than that of InGaAs QWs found in [25] (vs = 4.5x104cm.s−1). For the passivated structures, we find vs = 4x103cm.s−1 which is more than a 3-fold improvement compared to best value obtained in [25] (vs = 1.2x104cm.s−1).Fig. 4 a) Pump-probe measurement of the cavity resonance wavelength shift for passivated (blue diamonds) and unpassivated (red circles) samples. The dotted lines indicate the fits of the experimental datas with single exponential decay function giving a non-radiative carrier lifetime, respectively for passivated and unpassivated, of 2.63ns and 0.37ns. b) Time resolved photoluminescence of the QWs measured on unpatterned InP-based slab. The dotted line shows the single exponential decay fit of the data with a decay time of 4.1ns.
The lifetimes are further compared to the one obtained for the wafer-bonded unpatterned material. The latter is deduced from the time-resolved measurement of QW luminescence under pulsed pumping (see Fig. 4(b)) performed using a fast superconducting single photon detector. By fitting the experimental data with, again, a single exponential decay, we obtain a non-radiative carrier lifetime of 4.1ns. This allows us to demonstrate that our passivation treatment enables a substantial recovery of the quality of the active material close to the level of unpatterned InP-based slab.
4. Demonstration of CW RT laser emission
Following the temporal lifetime measurements, the samples are explored at room temperature by detecting the light emitted under CW optical pumping at 1.18µm. The pump laser diode is focused down on the cavities surface to a spot size of about 5µm using a 10x long working distance microscope objective. The emitted light is collected via the grating couplers using SMF and sent to a spectrometer equipped with a cooled array of InGaAs detectors.
CW laser operation is only observed for the samples which underwent the passivation treatment. The samples were continuously pumped for days and no degradation was observed. Moreover, samples passivated a month prior to optical characterisation were still exhibiting CW laser behaviour. Of course, systematic reliability tests of our nanolasers still have to be performed to assess the lasting time of our passivation process.
A typical light-light characteristic giving the emitted power as a function of the pump power in the CW regime is represented on Fig. 5(a) in log-log scale. The measurements are performed on a cavity with a central lattice constant of 365nm and a radius of the holes of r = 115nm such that the emission wavelength is around 1565nm. We observe the typical S-shaped curve attesting to the transition from spontaneous to stimulated emission for an absorbed pump power around 8μW (inflection point). As can be also seen on Fig. 5(a), the emission linewidth narrows down as the pump power is increased and takes on values under the spectrometer resolution (0.1nm) above the laser threshold. In order to evaluate the impact of the carrier lifetime increase on the laser threshold, we characterise, in the pulsed regime, unpassivated and passivated cavities resonant at the same wavelength in order to keep the material gain constant. As can be seen on Fig. 5(b), we measure a threshold reduction of a factor 2.5 which is in agreement with a rate equation modelling [17] of quantum well laser considering a spontaneous emission factor β = 0.01 which is typical for this type of laser.
Fig. 5 a) CW operation: measured emission power (red dots) and linewidth (black triangles) as a function of the pump power for a PhC nanobeam cavity with a0 = 365nm and r = 115nm. b) Pulsed operation: Emission power as a function of the pump power for unpassivated and passivated PhC nanolasers.
5. Conclusion
In conclusion, we unambiguously demonstrated using time resolved pump-probe experiments that the highly detrimental surface recombination of carriers occurring in InP-based active photonic crystals can largely be subdued by chemical treatment with a (NH4)2S solution followed by silica encapsulation. The increase in the carrier lifetime from 370ps to 2.63ns allowed us to obtain stable CW operation of QW nanobeam PhC lasers with a low threshold of 8µW. These nanolasers were fabricated on and integrated to SOI waveguides with the view to extend III-V/SOI hybrid platforms towards nanophotonics. This combination of III-V based nanodevices together with SOI is indeed one of the enabler of the “More than Moore Revolution” promised by the convergence of photonics and microelectronics on a single chip.
Acknowledgments
We acknowledge for funding the FP7 European Projects COPERNICUS (249012) and PhoxTroT (318240) as well as the French ANR jeunes chercheurs PROWOC.
References and links
1. Y. Akahane, T. Asano, B. S. Song, and S. Noda, “High-Q photonic nanocavity in a two-dimensional photonic crystal,” Nature 425(6961), 944–947 (2003). [CrossRef] [PubMed]
2. P. Colman, S. Combrié, G. Lehoucq, and A. De Rossi, “Control of dispersion in photonic crystal waveguides using group symmetry theory,” Opt. Express 20(12), 13108–13114 (2012). [CrossRef] [PubMed]
3. K. Nozaki, T. Tanabe, A. Shinya, S. Matsuo, T. Sato, H. Taniyama, and M. Notomi, “Sub-femtojoule all-optical switching using a photonic crystal nanocavity,” Nat. Photonics 4(7), 477–483 (2010). [CrossRef]
4. H.-G. Park, S.-H. Kim, S.-H. Kwon, Y.-G. Ju, J.-K. Yang, J.-H. Baek, S.-B. Kim, and Y.-H. Lee, “Electrically driven single-cell photonic crystal laser,” Science 305(5689), 1444–1447 (2004). [CrossRef] [PubMed]
5. S. Matsuo, K. Takeda, T. Sato, M. Notomi, A. Shinya, K. Nozaki, H. Taniyama, K. Hasebe, and T. Kakitsuka, “Room-temperature continuous-wave operation of lateral current injection wavelength-scale embedded active-region photonic-crystal laser,” Opt. Express 20(4), 3773–3780 (2012). [CrossRef] [PubMed]
6. B. Ellis, M. A. Mayer, G. Shambat, T. Sarmiento, J. Harris, E. E. Haller, and J. Vučković, “Ultralow-threshold electrically pumped quantum-dot photonic-crystal nanocavity laser,” Nat. Photonics 5(5), 297–300 (2011). [CrossRef]
7. K. Nozaki, H. Watanabe, and T. Baba, “Photonic crystal nanolaser monolithically integrated with passive waveguide for effective light extraction,” Appl. Phys. Lett. 92(2), 021108 (2008). [CrossRef]
8. Y. Halioua, A. Bazin, P. Monnier, T. J. Karle, I. Sagnes, G. Roelkens, D. Van Thourhout, F. Raineri, and R. Raj, “III-V photonic crystal wire cavity laser on silicon wafer,” J. Opt. Soc. Am. B 27(10), 2146–2150 (2010). [CrossRef]
9. E. Yablonovitch, “Photonic crystals: what’s in a name,” Opt. Photonics News 33(3), 1–2 (2007).
10. M. Bagheri, M. H. Shih, Z.-J. Wei, S. J. Choi, J. D. O’Brien, P. D. Dapkus, and W. K. Marshall, “Linewidth and modulation response of two-dimensional microcavity photonic crystal lattice defect lasers,” IEEE Photonics Technol. Lett. 18(10), 1161–1163 (2006). [CrossRef]
11. A. Bazin, P. Monnier, X. Lafosse, G. Beaudoin, R. Braive, I. Sagnes, R. Raj, and F. Raineri, “Thermal management in hybrid InP/silicon photonic crystal nanobeam laser,” Opt. Express 22(9), 10570–10578 (2014). [CrossRef] [PubMed]
12. S. Matsuo, A. Shinya, T. Kakitsuka, K. Nozaki, T. Segawa, T. Sato, Y. Kawaguchi, and M. Notomi, “High-speed ultracompact buried heterostructure photonic-crystal laser with 13 fJ of energy consumed per bit transmitted,” Nat. Photonics 4(9), 648–654 (2010). [CrossRef]
13. K. Kash, P. Grabbe, R. E. Nahory, A. Scherer, A. Weaver, and C. Caneau, “Dynamics of photoexcited carriers in micron‐size InP‐InGaAsP etched microstructures probed by picosecond photoluminescence spectroscopy,” Appl. Phys. Lett. 53(22), 2214 (1988). [CrossRef]
14. A. M. Yacomotti, F. Raineri, G. Vecchi, I. Sagnes, M. Strassner, L. Le Gratiet, R. Raj, and A. Levenson, “Ultrafast nonlinear response around 1.5 µm in 2D AlGaAs/AlOx photonic crystal,” Appl. Phys. B 81(2-3), 333–336 (2005). [CrossRef]
15. F. Raineri, C. Cojocaru, P. Monnier, A. Levenson, R. Raj, C. Seassal, X. Letartre, and P. Viktorovitch, “Ultrafast dynamics of the third-order nonlinear response in a two-dimensional InP-based photonic crystal,” Appl. Phys. Lett. 85(11), 1880 (2004). [CrossRef]
16. A. Bazin, K. Lengle, M. Gay, P. Monnier, L. Bramerie, R. Braive, G. Beaudoin, I. Sagnes, R. Raj, and F. Raineri, “Ultrafast all-optical switching and error-free 10 Gbit/s wavelength conversion in hybrid InP-silicon on insulator nanocavities using surface quantum wells,” Appl. Phys. Lett. 104(1), 011102 (2014). [CrossRef]
17. L. A. Coldren and S. W. Corzine, Diode Lasers and Photonic Integrated Circuits (Wiley, 2012).
18. M. Nomura, S. Iwamoto, K. Watanabe, N. Kumagai, Y. Nakata, S. Ishida, and Y. Arakawa, “Room temperature continuous-wave lasing in photonic crystal nanocavity,” Opt. Express 14(13), 6308–6315 (2006). [CrossRef] [PubMed]
19. L. J. Martínez, B. Alén, I. Prieto, D. Fuster, L. González, Y. González, M. L. Dotor, and P. A. Postigo, “Room temperature continuous wave operation in a photonic crystal microcavity laser with a single layer of InAs/InP self-assembled quantum wires,” Opt. Express 17(17), 14993–15000 (2009). [CrossRef] [PubMed]
20. H. Ichikawa, K. Inoshita, and T. Baba, “Reduction in surface recombination of GaInAsP microcolumns by CH4 plasma irradiation,” Appl. Phys. Lett. 78(15), 2119–2121 (2001). [CrossRef]
21. D. Englund, H. Altug, and J. Vuckovic, “Low-threshold surface-passivated photonic crystal nanocavity laser,” Appl. Phys. Lett. 91(7), 071124 (2007). [CrossRef]
22. A. Bazin, R. Raj, and F. Raineri, “Design of silica encapsulated high-q photonic crystal nanobeam cavity,” J. Lightwave Technol. 32(5), 952–958 (2014). [CrossRef]
23. Y. Halioua, A. Bazin, P. Monnier, T. J. Karle, G. Roelkens, I. Sagnes, R. Raj, and F. Raineri, “Hybrid III-V semiconductor/silicon nanolaser,” Opt. Express 19(10), 9221–9231 (2011). [CrossRef] [PubMed]
24. T. J. Karle, Y. Halioua, F. Raineri, P. Monnier, R. Braive, L. Le Gratiet, G. Beaudoin, I. Sagnes, G. Roelkens, F. van Laere, D. Van Thourhout, and R. Raj, “Heterogeneous integration and precise alignment of InP-based photonic crystal lasers to complementary metal-oxide semiconductor fabricated silicon-on-insulator wire waveguides,” J. Appl. Phys. 107(6), 063103 (2010). [CrossRef]
25. M. Boroditsky, I. Gontijo, M. Jackson, R. Vrijen, E. Yablonovitch, T. Krauss, C.-C. Cheng, A. Scherer, R. Bhat, and M. Krames, “Surface recombination measurements on III–V candidate materials for nanostructure light-emitting diodes,” J. Phys. 87, 3497–3504 (2000).
26. H. Oigawa, J.-F. Fan, Y. Nannichi, H. Sugahara, and M. Oshima, “Universal passivation effect of (NH4)2Sx treatment on the surface of III-V compound semiconductors,” J. Appl. Phys. 30(2), L322–L325 (1991). [CrossRef]
