As performance and power consumption of modern micro-chips are increasingly limited by electrical on-chip interconnects, all-optical interconnect systems promise data transmission at speed of light and wavelength- division multiplexing. To realize complex networks, active devices, like lasers, need to be integrated on Si. III-Vs are excellent candidates for optical devices, however, their integration on Si is challenging due to a significant lattice and thermal mismatch. Template-assisted selective epitaxy (TASE) was recently developed by our group, allowing for the selective growth of III-Vs from a small Si seed in a confined oxide template. In this work, we extend TASE towards optical devices and demonstrate the monolithic integration of InGaAs lasers via a novel approach using a virtual substrate (VS) in a two-step templated growth. First, μm2 sized 60 nm thick InGaAs VSs are grown by MOCVD using TASE on SOI. Subsequently, 500 nm oxide are deposited onto the VS and patterned in arbitrary shapes like disks, and rings. In a second InGaAs growth, the defined vertical cavities are filled. The investigated structures have diameters of 1.7 μm, thicknesses of 0.5 µm and total cavity volumes of 0.5 λ30. Photoluminescence spectroscopy reveals a broad spontaneous emission peak around 1.1 μm (FWHM = 150 nm) that increases linearly with pump power for low excitation powers (<< 2.6 pJ/pulse). Above excitation threshold, a strong emission peak emerges at 1.1 μm (FWHM = 7 nm). The Input-Output curve (log- log, T = 10 K) exhibits the characteristic S-shape which constitutes a strong indication for the lasing operation. The onset of the lasing threshold is observed up to 200 K with a characteristic temperature of T0 = 192 K.
Terahertz quantum cascade lasers (THz QCL) are a very promising source for efficient frequency comb generation at terahertz frequencies. They do not only provide an output power of the order of milliwatts but are also covering a large spectral bandwidth. Octave spanning devices have recently been reported by our group. They provide a very low intrinsic dispersion due to the flat gain curve and the flat losses of the resonator. This allows frequency comb operation up to more than 600 GHz bandwidth with standard broadband metal-metal waveguide Fabry-Pérot QCLs. Frequency combs at terahertz frequencies are especially interesting for spectroscopic applications employing the powerful dual-comb setup. Such a setup requires a fast detector which is difficult to get with a sufficient sensitivity at terahertz frequencies. We present here an alternative approach, which does not need a fast detector but rather uses one of the two THz QCL frequency combs as an ultrafast multiheterodyne detector integrating local oscillator (LO) and detector in one single device. Two laser ridges are fabricated on the same chip at a distance of 500 um. Part of the light from the sample laser is coupled into the LO laser via the metallic ground plane. The downconverted multiheterodyne beatnote can be measured through the laser power supply line with a bias Tee. The obtained dual-comb covers a bandwidth of 630 GHz with a central frequency of 2.5 THz. The frequency comb spacing was analysed using frequency counting techniques revealing an accuracy down to _frep=fcarrier 10^(-12) at the carrier frequency of 2.5 THz.
We develop a spectroscopy platform for industrial applications based on semiconductor quantum cascade laser (QCL)
frequency combs. The platform’s key features will be an unmatched combination of bandwidth of 100 cm-1, resolution of
100 kHz, speed of ten to hundreds of μs as well as size and robustness, opening doors to beforehand unreachable
markets. The sensor can be built extremely compact and robust since the laser source is an all-electrically pumped
semiconductor optical frequency comb and no mechanical elements are required. However, the parallel acquisition of
dual-comb spectrometers comes at the price of enormous data-rates. For system scalability, robustness and optical
simplicity we use free-running QCL combs. Therefore no complicated optical locking mechanisms are required. To
reach high signal-to-noise ratios, we develop an algorithm, which is based on combination of coherent and non-coherent
averaging. This algorithm is specifically optimized for free-running and small footprint, therefore high-repetition rate,
comb sources. As a consequence, our system generates data-rates of up to 3.2 GB/sec. These data-rates need to be
reduced by several orders of magnitude in real-time in order to be useful for spectral fitting algorithms.
We present the development of a data-treatment solution, which reaches a single-channel throughput of 22% using a
standard laptop-computer. Using a state-of-the art desktop computer, the throughput is increased to 43%. This is
combined with a data-acquisition board to a stand-alone data processing unit, allowing real-time industrial process
observation and continuous averaging to achieve highest signal fidelity.
Classical methods for modeling electromagnetic scattering from the topography of lithographic reticles must place
a high premium on fast computation, and toward that end they apply pre-stored perturbations (e.g. the so-called
boundary layers) to feature edges in order to approximate the impact of finite-thickness mask films. Though
approximate, these methods involve E&M calculations with vector fields, and so employ edge-field corrections
that are different for edges oriented parallel or perpendicular to the vector field. As a result these methods entail a
requirement for two separate aerial image simulations using orthogonal source polarizations in order to represent
unpolarized illumination. This imposes a minimum 2X runtime penalty relative to baseline thin-mask (TMA)
simulations, since the known method for combining the effect of both polarizations into one single set of imaging
TCCs applies only to thin-mask calculations. More severe performance penalties are common in so-called sparse
imaging methodologies when topographic effects are included, since the separated treatment of feature edges and
the internal area of the features can increase the number of memory lookups required.
In this paper an isotropic field perturbation approach is evaluated, in which an isotropic edge field correction,
common to all edge orientations, mimics the effect of the true parallel and perpendicular edge field perturbations
when the mask is illuminated with unpolarized light, as well as in certain cases of polarized illumination. The
isofield is not an ad hoc empirical correction but rather an accurate approximation in the limit of modest departures
from scalar TMA. More specifically, we show that the isofield model accounts for vector imaging effects with full
accuracy in the TMA terms, and in an approximate way in the electromagnetic edge-field terms that becomes
accurate when the polarization dependence of the TMA terms is small. We will show with comparison to more
rigorous electromagnetic models and simulations, as well as against wafer measurements that the accuracy loss
relative to classic polarized EMF correction approach is within a small percentage on mask blanks where the
electromagnetic edge field perturbation terms are small relative to the TMA term. Methodology to extend these
models into the subwavelength diffraction regime will be discussed.