Gain characterization and passive modelocking of electrically pumped VECSELs

Abstract

Linear and nonlinear gain characterization of electrically pumped vertical external cavity surface emitting lasers (EP-VECSELs) is presented with spectrally resolved measurements of the gain and with gain saturation measurements of two EP-VECSEL samples with different field enhancement in the quantum-well gain layers. The spectral bandwidth, small-signal gain and saturation fluence of the devices are compared. Using the sample with the larger bandwidth, we have demonstrated the shortest pulses generated from a passively modelocked EP-VECSEL to date. With a low-saturation-fluence SESAM for passive modelocking we have achieved 9.5-ps pulses with 7.6 mW average output power at a repetition rate of 1.4 GHz. With a higher output coupler transmission the pulse duration was increased to 31 ps with an average output power of 13.6 mW. The pulses were chirped mainly due to the group delay dispersion (GDD) introduced by the intermediate DBR, which compensates the optical loss in the structure.

Novel VECSEL for short‐wave infrared Raman spectroscopy applications

“…This achieves a combination of high continuous‐wave (CW) output power, owing to a large gain area, and a low‐divergence near‐diffraction‐limited beam, owing to the transverse mode control of the extended cavity. The advantages of VECSELs over edge‐emitting semiconductor lasers are circular output beam, easy array fabrication, small size, and low cost at mass production . Their cost advantages stem from their vertical nature, which allows on‐wafer testing, easy extension to an extended cavity, and a better match with conventional semi‐conductor equipment.…”

Absorber and gain chip optimization to improve performance from a passively modelocked electrically pumped vertical external cavity surface emitting laser

“…31 In 2012, we presented sub-10-ps pulses in collaboration with Philips Technologie GmbH. 32 Shortly after, pulses as short as 6.3 ps with an average output power of 6.2 mW or a peak power of more than 1 W were demonstrated with chips designed, grown, and fabricated at ETH. 29 The previous results indicated that a better trade-off between the electrical and optical properties of the gain chip combined with an improved SESAM could potentially generate both shorter pulse durations and higher average output powers.…”

“…29 The previous results indicated that a better trade-off between the electrical and optical properties of the gain chip combined with an improved SESAM could potentially generate both shorter pulse durations and higher average output powers. [29][30][31][32] Based on our prior experience, we used an improved SESAM, specifically designed for EP-VECSELs, to obtain modelocking with the new gain chip 33 from Philips Technologie GmbH, optimized for fundamental-transverse mode at high output power (up to 90 mW). As a result, we set benchmarks to the peak and average output power, pulse duration and repetition rate from modelocked EP-VECSELs: pulses as short as 2.5 ps, average output power up to 53.2 mW, and repetition rates of up to 18.2 GHz were obtained with a SESAM modelocked EP-VECSEL in different cavity configurations.…”

“…36 We use the latest generation EP-VECSEL gain chip fabricated by Philips Technologie GmbH based on the concept and design presented in 2014 33 and similar to the ones previously used for modelocking. 32 The basic structure chosen for this gain chip is referred to as a bottom emitter, meaning that the light is emitted through the substrate of the semiconductor layer stack. This processing scheme allows for bonding the bottom mirror directly onto the heat sink and thus enables an efficient thermal management, which is crucial for high-power operation.…”

“…After processing, the EP-VECSEL gain chip consists of the following elements (listed from bottom to top): An AlN heat spreader; a bottom contact with a diameter (BCD) of 60 lm; an AlGaAs 37-pair p-doped distributed Bragg reflector (p-DBR) serving as bottom mirror; the active region consisting of 3 InGaAs quantum wells (QWs) embedded in GaAs; an intermediate n-doped DBR with 11 pairs; an oxide aperture with a diameter of 100 lm; the substrate thinned down to 100 lm; a top ring electrode and a single-layer dielectric anti-reflection (AR) coating. The two major improvements (compared to the Philips chips previously presented) 32 can be summarized as follows. First, an optimized doping scheme is implemented leading to a better trade-off between electrical and optical losses.…”