Abstract
Mode-locking of 1.3 μm InGaAs/GaAs quantum dot lasers was discussed. Hybrid mode-locking was also achieved. It was found that the minimum pulse width of the Fourier-limited pulses was 7 ps. Low uncorrelated timing jitter below 1 ps was also found. High-frequency operation of lasers was achieved by a ridge waveguide design. The study enables the introduction of 1.3 μm quantum dot lasers as high frequency optical comb generators.
| Original language | English |
|---|---|
| Pages (from-to) | 843-845 |
| Number of pages | 3 |
| Journal | Applied Physics Letters |
| Volume | 85 |
| Issue number | 5 |
| DOIs | |
| Publication status | Published - 2 Aug 2004 |
Funding
0003-6951 1077-3118 American Institute of Physics AIP 030430APL 1.1776340 10.1063/1.1776340 L04-1238 DEVICE PHYSICS 35 GHz mode-locking of 1.3 μ m quantum dot lasers 35 GHz Mode-Locking of 1.3 μ m Quantum Dot Lasers Kuntz M. a) Fiol G. Lämmlin M. Bimberg D. Institut für Festkörperphysik, Technische Universität Berlin , Hardenbergstrasse 36, 10623 Berlin, Germany Thompson M. G. Tan K. T. Marinelli C. Penty R. V. White I. H. Engineering Department, University of Cambridge , Trumpington Street, CB2 1PZ, United Kingdom Ustinov V. M. Zhukov A. E. Shernyakov Yu. M. A.F. Ioffe Physicotechnical Institute, Russian Academy of Sciences , 194021 St. Petersburg, Russia Kovsh A. R. NSC-Nanosemiconductors GmbH , Von-Fraunhofer-Strasse 13, 44227 Dortmund, Germany a) Electronic mail: [email protected] 02 08 2004 85 5 843 845 05 03 2004 02 06 2004 2004-07-27T16:32:16 2004 American Institute of Physics 0003-6951/2004/85(5)/843/3/ $20.00 35 GHz passive mode-locking of 1.3 μ m ( InGa ) As ∕ GaAs quantum dot lasers is reported. Hybrid mode-locking was achieved at frequencies up to 20 GHz . The minimum pulse width of the Fourier-limited pulses was 7 ps with a peak power of 6 mW . Low uncorrelated timing jitter below 1 ps was found in cross correlation experiments. High-frequency operation of the lasers was eased by a ridge waveguide design that includes etching through the active layer. Development of quantum dot (QD) lasers 1,2 with 1.3 μ m emission wavelength showing very low threshold current, large T 0 , suppressed beam filamentation, and feasibility of 2.5 Gbytes ∕ s data transmission 3–6 presents a break through toward data and telecom applications. Promising dynamic properties of quantum dot lasers like large differential gain and cut-off frequency and small chirp were predicted and first reported for emission wavelengths 960 – 1160 nm . 7–9 Cut-off frequencies up to 23 GHz at room temperature at 1 μ m 10 were obtained using tunnel injection of carriers into the dots, demonstrating the potential of quantum dot lasers for directly modulated 10 Gbytes ∕ s communication systems. At 1.3 μ m no operation beyond 10 GHz was reported yet. For ultrahigh frequency applications such as time domain multiplexing, optical comb generators operating in the 10 – 40 GHz range are needed. Initial work on mode-locking of quantum dot lasers at 1.1 μ m 11 and 1.3 μ m 12 reported repetition rates of 10 and 7.4 GHz , respectively. We report here 20 GHz hybrid and 35 GHz passive mode-locking at 1.3 μ m , the wavelength relevant for datacom and metropolitan area networks, and demonstrate Fourier-limited pulses and ultralow uncorrelated timing jitter. The AlGaAs ∕ GaAs based laser structure incorporated five InGaAs quantum dot layers grown by molecular beam epitaxy and was processed into ridge waveguide emitters with a stripe width of 4 μ m . Details on the growth and cw lasing of the QD structure can be found in Ref. 13 . The ridges were dry etched through the active layer to provide strong index guiding of the optical mode and suppression of current spreading. Index guiding combined with the small stripe width ensures large coupling efficiency into optical fibers. The suppression of current spreading leads to an improvement of the electrical high-frequency characteristics of the laser diode by reducing the parasitic capacitance, as compared to shallow mesa lasers. The sides of the ridge were coated by an insulating low refractive index material. Finally, the p contacts including bond pads were fabricated. The structure of the device is shown in Fig. 1 . The samples for mode-locking were processed into two-section devices by defining a 20 μ m gap in the p -metal. The resistance between the sections is > 1 k Ω . The lasers with as-cleaved facets were mounted p -side up on a passive copper heat sink and were electrically connected to two SMA ports via stripe lines and short ( < 400 μ m ) bond wires. The samples for mode-locking had lengths of 2000 and 1130 μ m corresponding to the round trip frequencies of 20 and 35 GHz , respectively. All measurements were carried out at room temperature ( 297 K ) . From a series of devices with different length we estimated internal losses of 5 cm − 1 , an internal quantum efficiency of 76 % , and a threshold current density of 95 A ∕ cm 2 at infinite length. These data are comparable to those of a similar series of lasers that have been processed with shallow mesas, thus indicating that the deep etching of the mesa has not created any detectable defects within the active layer. The far field asymmetry decreased upon deep etching from an ellipticity of 10 – 3.5 yielding an increase of coupling efficiency into single mode fibers by a factor of 2.5. Prior to the mode-locking measurements we investigated the small-signal modulation performance of the lasers. Using a HP 8722C network analyzer combined with a 50 GHz photodetector both the S 11 parameter (reflection) and the S 12 parameter (transmission) of the laser were measured. The reflection measurement shows a − 3 dB bandwidth of 4 GHz , which is a factor of 3 larger than for the shallow mesa samples. The improvement is due to a large reduction of the parasitic diffusion capacitance along with an only moderate increase of the device resistance. The − 3 dB bandwidth of 4 GHz is still low compared to typical mode-locking frequencies between 10 and 40 GHz , so we must be aware that most of the rf power applied for hybrid mode-locking will be reflected by the sample. The differential gain at an output power of 10 mW derived from the transmission measurement is g ′ = 2.3 × 10 − 15 cm 2 , a value comparable to our previous measurements. 14 The transmission shows a − 3 dB bandwidth of 5 GHz , with only little resonance enhancement. 5 Gbytes ∕ s open eye patterns have been demonstrated with these devices and will be published elsewhere. 15 The two-section devices for mode-locking consisted of a long gain and a short absorber section operated at reverse bias levels between 0 and − 3 V . The section length were 1500 ∕ 500 μ m for the 2000 - μ m - and 980 ∕ 150 μ m for the 1130 - μ m - long devices, respectively. Due to the absorber section the threshold for both devices increased by a factor of 3. Time-domain measurements were carried out with a self-assembled autocorrelator. Figure 2 shows a typical autocorrelation measurement from the 2000 μ m device. It shows both the autocorrelation (middle peak) and the cross correlation (side peaks) of the pulses. The autocorrelation trace was deconvoluted assuming a Gaussian pulse shape. The dashed curve shows the calculated pulse shape with a full width at half maximum (FWHM) of 12 ps . This value is in good agreement with the Fourier limit ( Δ τ Δ ν = 0.44 ) of 13 ps estimated from the spectral FWHM of 180 pm . The inset of Fig. 2 shows the spectrum of the mode-locked laser centered at a wavelength of 1286 nm . In order to characterize the dependence of the pulses on parameters like reverse bias, gain current, and rf power, we did arrays of autocorrelation scans with a fully automated setup. The 2000 μ m laser was hybridly∕passively mode-locked at currents between 30 and 70 mA and reverse bias voltages between − 2.5 and 0 V . For hybrid mode-locking we applied a rf power of 20 dBm to the absorber section. The large rf power was necessary in order to compensate for the strong damping of the electrical signal on its way to the laser diode as mentioned earlier. The mode-locking frequency was 20.27 GHz . The results for passive mode-locking are depicted in Fig. 3 : The diagram shows the different regimes of device operation versus reverse bias and gain current. The center region, which corresponds to mode-locking operation, shows the grey scale coded deconvoluted FWHM pulse width of the autocorrelation traces. With increasing reverse bias, the onset of lasing shifted to larger currents, due to the increasing absorption within the waveguide. The onset of lasing occurred abruptly as mode-locking, we observed no transition region. With increasing current, the pulses became broader, until we observed a cw offset, i.e., incomplete mode-locking. At even higher currents, we observed a transition region with all kinds of pulse patterns (e.g., self pulsation) until all intensity fluctuations flattened out to cw lasing. The pulse width ranged from 8 to 18 ps . As expected, the shortest pulses occurred at large reverse bias voltages, where it was possible to go to larger gain currents with the laser still capable of mode-locking. The minimum pulse width was limited by the maximum reverse bias voltage we estimated here to be not harmful for the device. It does not present a real lower limit. For hybrid mode-locking, we observed quite similar results except for a slight increase of the mode-locking regime toward larger currents ( + 5 mA ) . The pulse widths did not change significantly, which indicated that, as expected from the S 11 measurements, most of the rf power did not enter the device and thus did not contribute to absorption modulation. We characterized the influence of the rf power on the mode-locking frequency tuning range, i.e., the range of rf frequencies wherein the mode-locking frequency is pinned to the rf frequency. The maximum tuning range was 90 MHz ( 0.5 % of the locking frequency). Besides pulse shape, width, and locking range we determined the timing and amplitude jitter of the pulses. Comparison of autocorrelation and cross correlation 16 allowed us to estimate the uncorrelated jitter to be less than 1 ps . However, we expect the main jitter contribution to be correlated jitter. Further investigations are required to clarify this question. The 1130 - μ m -long laser was passively mode-locked at currents between 40 and 70 mA and reverse bias voltages between − 3 and 0 V . The mode-locking frequency was 35 GHz . It was not possible to drive this laser in a hybrid mode-locking scheme due to insufficient rf power coupled into the absorber section. Figure 4 shows both the autocorrelation and the cross-correlation measurement of the laser pulses. The minimum pulse width we achieved with this device was 7 ps , which is again in agreement with the Fourier transform limit. The uncorrelated jitter estimated from the cross correlation was less than 2 ps . The peak power from one facet of the mode-locked laser was 6 mW . Figure 5 shows the reverse bias versus current scan of the autocorrelation traces and the corresponding FWHM pulse widths. In comparison to the results on the longer device shown in Fig. 3 we observed a much smaller region of mode-locking. This is due to the shorter absorber section of 180 μ m which saturates already at low gain currents. We estimate the saturation absorption to be comparable to the saturation gain, i.e., 15 cm − 1 . The trade-off between absorber, gain, and total length establishes the current pulse width of limitation 7 ps . The ultimate limitation for the minimum pulse width of mode-locked lasers is the width of the gain spectrum. As we are yet deploying only 2 % of the intrinsic spectral width of the quantum dot gain medium, there is still plenty of room for reducing the pulse width to levels in the range or below 1 ps . Stacking of more QD layers and the insertion of an ion implanted absorber for stronger absorption are the means to achieve this goal. In conclusion we presented hybrid and passive mode-locking of quantum dot lasers at 1.3 μ m with 20 and 35 GHz repetition rate showing Fourier limited pulses. The minimum pulse width we achieved was 7 ps . The maximum locking range of the hybridly mode-locked laser was 0.5 % of the locking frequency. Uncorrelated jitter was below 1 ps at 20 GHz . Our results present a large step forward to introduce 1.3 μ m QD lasers as high frequency optical comb generators in data and telecom systems. Parts of this work are funded by the German Ministry for Education and Research (Contract No. 1BC913) and the European Union in the framework of the IST-Dotcom. We are indebted to R. Kaiser and U. Niggebrügge, Heinrich-Hertz-Institut Berlin, for helpful discussions and assistance with dry etching, respectively. FIG. 1. Schematic cross section of a deeply etched quantum dot laser diode. FIG. 2. Autocorrelation trace and deconvoluted pulse shape of a passively mode-locked quantum dot laser at 20.2 GHz repetition rate; the inset shows the corresponding spectrum. FIG. 3. Plot of pulse width dependence on reverse bias voltage and gain current for a passively mode-locked QD laser at 20.2 GHz . Three modes of operation can be distinguished: no lasing, mode-locking, and incomplete mode-locking. The grey scale coded pulse FWHM shows an increase with both gain current and voltage. FIG. 4. Autocorrelation trace of a passively mode-locked quantum dot laser at 35 GHz repetition rate; the inset shows the corresponding spectrum. FIG. 5. Plot of pulse width dependence on reverse bias voltage and gain current for a passively mode-locked QD laser at 35 GHz . Three modes of operation can be distinguished: no lasing, mode-locking (grey squares), and incomplete mode-locking. The grey scale coded pulse FWHM shows an increase with both gain current and voltage.
ASJC Scopus subject areas
- Physics and Astronomy (miscellaneous)