850-nm VCSEL Research Articles

Impact of Oxide Aperture on the Static and Dynamic Performance of 850-nm VCSELs

Impact of Oxide Aperture on the Static and Dynamic Performance of 850-nm VCSELs

Single-Channel 106.25 Gb/s PAM-4 and 64 Gb/s NRZ Transmission With a 33.4-GHz 850-nm VCSEL With low-RIN Characteristics

Single-Channel 106.25 Gb/s PAM-4 and 64 Gb/s NRZ Transmission With a 33.4-GHz 850-nm VCSEL With low-RIN Characteristics

Improvement of thermally induced current bifurcation in VCSEL arrays with non-uniform series resistance design

Non-uniform series resistance design of VCSEL arrays is studied to improve thermally induced current bifurcation based on an electro-opto-thermal model of VCSEL arrays. Taking an 850nm VCSEL array with 4×4 cells for example, the impact of series resistance on current bifurcation is investigated. Increasmg series resistance is helpful to enhance the critical current values of current bifurcation point (Irc) and hence delay the current bifurcation phenomenon. For VCSEL array with non-uniform series resistance, Irc is increased by 28.6% and the total output optical power is enhanced by 14.3% when compared with that of VCSEL array with uniform series resistance. Therefore, non-uniform series resistance design is a better method for delaying the current bifurcation phenomenon and enhancuig the output optical power of VCSEL arrays.

Mode Division Multiplexed 850-nm Single-Mode VCSEL Transmission Using Standard Single-Mode Fiber

Mode division multiplexing (MDM) has attracted great attention in recent years to increase the network capacity. To apply MDM in data center applications, a low-cost and low power consumption system is desired. In this letter, we utilize 850-nm single-mode VCSEL as the laser source and report mode division multiplexed transmission over a standard single-mode fiber, which is two-mode at 850 nm. Due to the small spot size and low numerical aperture, single-mode VCSEL can be coupled into single-mode fibers with low loss. The mode multiplexer and de-multiplexer devices are designed and fabricated based on fiber mode selective couplers using Hi780 fiber as the 850-nm single-mode launch and receiving fiber and standard single-mode fiber as the few-mode transmission fiber. The MDM system supports two channels, corresponding to the LP <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">01</sub> and LP <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">11</sub> mode in the few-mode transmission fiber. Detailed characterizations of the multiplexer and de-multiplexer are conducted including the insertion loss and crosstalk measurements. System transmission at 25 Gb/s shows error free performance over 1-km standard single- mode fiber for both channels. The utilization of VCSEL and standard single-mode fiber can potentially provide a low-cost solution for MDM systems in data center applications.

Development of Vertical Cavity Surface Emitting Laser Modulation for Data Communication

This paper includes some basic knowledge of VCSEL and reviews some researches focusing on development of modulation characteristics in data communication of 850nm, 980nm, 1060nm and 1100nm VCSEL, which is helpful to researchers who are studying in this field. The popular modulation in data communication VCSEL is also mentioned in this paper. There are still some difficulties in the modulation of VCSEL-based links directly, which may be solved by modulating relaxation-oscillation frequency of VCSEL in the near future.

Thermal analysis of VCSEL arrays based on first principle theory and finite element method

AlxGa1−xAs is a main epitaxial material of 850-nm VCSEL. In order to analyze lattice match and heat production, the band gap of AlxGa1−xAs was built, using the first-principles calculation theory. In addition, based on thermoelectric coupling three-dimensional (3D) finite element method, details on the surface temperature distributions of different 850-nm VCSEL arrays were analyzed. Array spacing and array form have an effect on heat dissipation of VCSEL arrays. The diamond array with 18 light emitting units shows the best thermal stability among others.

Comparative study on stained InGaAs quantum wells for high-speed optical-interconnect VCSELs

The gain-carrier characteristics of InGaAs quantum well for 980 nm high-speed, energy-efficient vertical-cavity surface-emitting lasers are investigated. We specially studied the potentially InGaAs quantum well designs can be used for the active region of energy-efficient, temperature-stable 980-nm VCSEL, which introduced a quantum well gain peak wavelength-to-cavity resonance wavelength offset to improve the dynamic performance at high operation temperature. Several candidate quantum wells are being compared in theory and measurement. We found that ∼5 nm InGaAs QW with ∼6 nm barrier thickness is suitable for the active region of high-speed optical interconnect 980 nm VCSELs, and no significant improvement in the 20% range of In content of InGaAs QWs. The results are useful for next generation green photonic device design.

Theoretical Investigation of State Bistability Between Pure- and Mixed-Mode States in a 1550-nm VCSEL Under Parallel Optical Injection

Based on the spin-flip model, state bistability (SB) between pure- and mixed-mode states in a 1550-nm vertical-cavity surface-emitting laser under parallel optical injection is theoretically investigated. The simulated results show that two types of SB can be observed when the injection light frequency ( $v_{\textrm {inj}}$ ) is smaller than the dominant mode frequency of the free-running laser ( $v_{e}$ ). For type-I SB, which occurs through fixing $v_{\textrm {inj}}$ and scanning injection power ( $P_{\textrm {inj}}$ ) along different routes, there exist two hysteresis loops in which the laser operates at pure- or mixed-mode state, depending on the variation route of $P_{\textrm {inj}}$ , and the hysteresis loop width increases sharply for increasing $\vert \Delta \nu \vert $ ( $\Delta v = v_{\textrm {inj}}- v_{e}$ ) with relatively strong $P_{\textrm {inj}}$ whereas the loop width increases slowly for relatively weak $P_{\textrm {inj}}$ , in agreement with our recent experimental report. Furthermore, type-II SB is also investigated through fixing $P_{\textrm {inj}}$ and scanning $v_{\textrm {inj}}$ along different routes. There also exist two hysteresis loops in which the laser may operate at a pure- or mixed-mode state. With an increase of $P_{\textrm {inj}}$ , the hysteresis loop width located at lower $\vert \Delta \nu \vert $ increases sharply and then decreases after reaching a maximum. However, for the hysteresis loop located at higher $\vert \Delta \nu \vert $ , the loop width gradually increases with the increase of $P_{\textrm {inj}}$ .

Investigation of single-mode vertical-cavity surface-emitting lasers with graphene-bubble dielectric DBR

An inter-cavty contact single mode 850nm VCSEL was fabricated with a graphene assisted self-assembly curved dielectric bubble Bragg mirror for the first time. Taking the advantage of graphene’s uniform low surface energy, the low cost dielectric bubble DBR (Si3N4/SiO2) was deposited on top of the graphene/half-VCSEL structure via van der Waals Force (vdWF) without using any additional spacing elements and sacrificial layer release-etch process. The continuous-wave operating VCSELs with an aperture diameter of 7μm exhibit single-mode output power of more than 1mW with a slope efficiency of 0.2W/A. The sidemode suppression ratios are >40dB. This novel modification into the lasers can also be applied to a variety of other optoelectronic devices, such as resonance photodetecter and super narrow linewidth VCSEL.

94-Gb/s 4-PAM Using an 850-nm VCSEL, Pre-Emphasis, and Receiver Equalization

We present the results from an experimental demonstration of a 100-Gb/s 4-PAM transmission over a few meter MMFs using an 850-nm VCSEL with both analog pre-emphasis and a digital offline equalizer. The effective data rate is 94 Gb/s after accounting for the forward error correction overhead. Digital equalizers of lengths 11 to 71 taps were investigated.

Polarization Switching Characteristics of 1550-nm Vertical-Cavity Surface-Emitting Lasers Subject to Double Polarization Pulsed Injection

Based on the spin-flip model, the polarization switching (PS) characteristics of two polarization outputs in a 1550-nm vertical-cavity surface-emitting laser (1550-nm VCSEL) subject to double polarization pulsed injection (DPPI) are investigated theoretically, and the obtained results are compared with the case, in which the 1550-nm VCSEL is simultaneously injected by the pulses of orthogonal polarized light and a continuous wave (CW) parallel polarized light, which is called double optical injection with pulsed orthogonal polarization (DIPOP). The simulated results show that, compared with the DIPOP, the DPPI method is beneficial to improving the PS performance under suitable parameters. The influences of the timing difference (Δt) in the switching of two injection pulses on the PS performance of 1550-nm VCSEL subject to DPPI are analyzed, and the optimized values of At under different bias currents are determined. Furthermore, for optimized At, the dependences of the switching times on the amplitudes of two polarization pulses and the frequency detuning are given.

850-nm VCSELs With p-Type $\mathrm {{\delta }}$ -Doping in the Active Layers for Improved High-Speed and High-Temperature Performance

In this paper, we study the influence of p-type modulation doping on the dynamic/static performance of high-speed 850-nm VCSELs with highly strained multiple quantum wells. The studied device structure has a 3/ $2~\lambda $ asymmetric cavity design, which can let the internal transit time of injected carriers be as short as that of 1/ $2~\lambda $ cavity design and further improve its performance in terms of speed and output power for high single-mode operation. Our proposed VCSEL structure with p-type doping shows superior modulation speed with an output power comparable with that of the un-doped reference device under room temperature operation. Furthermore, when the operating temperature reaches 85 °C, there is a significant improvement in both the modulation speed and maximum power of the p-doped structures. According to our simulation, this can be attributed to the change in the quasi-Fermi levels of the injected carriers after the addition of p-doping in the active layers, which minimizes the electron leakage under high-temperature operation.

ESD tests on 850 nm GaAs-based VCSELs

Forward and reverse HBM, MM, CDM ESD tests have been performed on 850-nm VCSELs, together with EOS and overpower test. The physical analysis of the tested devices showed a variety of damages not easily correlated to the measured electro-optical degradations. The solution requires the detailed interpretation of the observed physical mechanism, by means of electron microscopy and device modelling.

Aging and environmental tolerance of an optical transmitter for the ATLAS Phase-I upgrade at the LHC

The dual channel Miniature optical Transmitter (MTx) is developed for the ATLAS Phase-I upgrade requiring durable performance in the Large Hadron Collider environment. The data transmission has achieved 8Gbps per channel with a custom-designed LOCld laser driver and 850nm VCSELs packaged in transmitter optical sub-assemblies (TOSAs). The performance of the MTx opto-electronics is evaluated. Accelerated aging tests of the VCSELs were conducted in a chamber at 85°C, 85% relative humidity, with TOSA and bare-die samples prepared in non-hermetic condition. Radiation tolerance of the VCSELs was investigated with 30MeV and 70MeV protons. The radiation induced effects in data transmission were investigated for light-power degradation and parameters of eye-diagrams.

Dynamics of Normal and Reverse Polarization Switching in 1550-nm VCSELs Under Single and Double Optical Injection

Polarization switching (PS) in 1550-nm VCSELs under single and double pulsed polarized optical injection is investigated experimentally and in theory. “Normal” and “reverse” PS were achieved, respectively, when the device was subject to single and double optical injection. Speed operation enhancement is observed for “reverse” PS under double polarized optical injection. Also, the minimum injection power requirements for both “normal” and “reverse” PS have been analyzed showing significant differences for the two investigated cases. Furthermore, we have also investigated numerically the influence of important system parameters, such as the normalized bias current, the injection field strength, and the linewidth enhancement factor in the optimal operation conditions for both types of PS. Overall good agreement between theory and experiments is found.

Temperature-Stable, Energy-Efficient, and High-Bit Rate Oxide-Confined 980-nm VCSELs for Optical Interconnects

We report 980-nm VCSELs that achieve temperature-stable, energy-efficient, and high-bit rate data transmission concurrently. Oxide-aperture-dependent static characteristics and high-speed modulation properties are analyzed at room temperature and also at high temperature. It is demonstrated that VCSELs with oxide-aperture diameters smaller than ∼5 μm are most suitable to achieve energy-efficient, temperature-stable, and high-bit rate operation concurrently. We demonstrate error-free data transmission (defined as a bit error ratio < 1 × 10 −12) at 38 Gb/s from 25 to 85 °C without any change of working point and modulation condition by using VCSELs with oxide-aperture diameters smaller than 5 μm. For VCSELs smaller than ∼7 μm, 42 Gb/s error-free data transmission at 25 °C is achieved. Record low energy dissipation at 85 °C is achieved with our ∼3 μm oxide-aperture diameter VCSELs. At room temperature, only 145, 147, and 217 fJ of dissipated heat energy per bit is needed for 35, 38, and 42 Gb/s error-free data transmission. Each of these bit rates is a record low heat dissipation for 980-nm VCSELs. Our VCSELs are especially well suited for very short reach (<2 m) optical interconnects in high-performance computers, for board-to-board and chip-to-chip input/output interfaces and data buses, and for on-chip integrated photonics.

Crosstalk Characteristics and Performance of VCSEL Array for Multicore Fiber Interconnects

This paper investigates the crosstalk characteristics of a dense two-dimensional 6-channel 850-nm VCSEL array designed for multimode multicore fiber (MCF) interconnects. Interchannel optical, thermal, and electrical crosstalk is measured and found to not affect channel integrity and receiver sensitivity when transmitting data over an MCF with a fiber fan-out. We demonstrate error-free data transmission (bit-error-ratio <10−12) up to 40 Gb/s/channel over an MCF, which shows promise for an aggregate capacity of 240 Gb/s over a single fiber in a future link.

Sensitivity and Limit of Detection of biosensors based on ring resonators

In this work, we present a study of the Sensitivity (S) and Limit of Detection (LOD) of microring based photonic biosensors as a function of the waveguide composition and dimensions. The target is Aflatoxin, which is a toxin of major concern for south Europe dairy industry. The sensing device is based on an array of multiple SiON microring resonators, fiber-coupled to both an 850nm VCSEL and a silicon photodetectors, packaged with a microfluidic circuit. Volumetric sensing with glucose–water solutions of various concentrations yields a best sensitivity of 112nm/RIU. To link these results to the Limit of Detection of the sensors, we also measured the noise of our experimental readout system and then calculated the LOD of our sensors. We found a best value of LOD of 1.6×10−6RIU (referred to volumetric sensing). Finally, we detected Aflatoxin in solutions of various concentrations (down to 1.58nM) by functionalized sensors. The functionalization is based on a wet silanization and specific DNA-aptamer binding on the chip. Reproducibility and re-usability of the sensor are investigated by several chemical treatments. Optimum procedure allows multiple sequential measurements with a good endurance. This work was supported by the FP7 EU project “Symphony” (Grant agreement no.: 610580).

Low-voltage, high-speed and compact electro-absorption modulator laterally integrated with 980-nm VCSEL.

We present a compact electro-absorption slow-light modulator laterally-integrated with a 980-nm VCSEL. We figured out the small signal modulation response for different modulator lengths. While the 3-dB small-signal modulation bandwidth of conventional directly modulated VCSELs on the same epi-wafer structure was limited below 10 GHz, we obtained a modulation bandwidth over 21 GHz for a 30 µm long modulator. We also demonstrated large signal modulation up to 25 Gbps with a low driving voltage below 600 mV(pp) and an extinction ratio of 4 dB for the modulator length of 50 µm. Prospects of much higher speed (> 40 Gbps) were examined with reducing the size of the modulator. Also, the tapered waveguide coupling structure enables "quasi-unidirectional coupling", which reduces the optical feedback in a VCSEL from the integrated modulator.

Corrections to “Impact of the Quantum Well Gain-to-Cavity Etalon Wavelength Offset on the High Temperature Performance of High Bit Rate 980-nm VCSELs” [Aug 14 613-621

In the above paper (ibid., vol. 50, no. 8, pp. 613-621, Aug. 2014), the sixth author was misidentified. His name should have read James A. Lott in the byline and in his biography.