A semiconductor injection laser that differs in a fundamental way from diode lasers has been demonstrated. It is built out of quantum semiconductor structures that were grown by molecular beam epitaxy and designed by band structure engineering. Electrons streaming down a potential staircase sequentially emit photons at the steps. The steps consist of coupled quantum wells in which population inversion between discrete conduction band excited states is achieved by control of tunneling. A strong narrowing of the emission spectrum, above threshold, provides direct evidence of laser action at a wavelength of 4.2 micrometers with peak powers in excess of 8 milliwatts in pulsed operation. In quantum cascade lasers, the wavelength, entirely determined by quantum confinement, can be tailored from the mid-infrared to the submillimeter wave region in the same heterostructure material.

Direct bandgap group IV materials may thus represent a pathway towards the monolithic integration of Si-photonic circuitry and CMOS technology.Although a group IV direct bandgap material has not been demonstrated yet, silicon photonics using CMOS-compatible processes has made great progress through the development of Si-based waveguides 12 , photodetectors 13 and modulators 14 . The thus emerging technology is rapidly expanding the landscape of photonics applications towards tele-and data communication as well as sensing from the infrared to the mid infrared wavelength range 15-17 . Today's light sources of such systems are lasers made from direct bandgap group III-V materials operated off-or on-chip which requires fibre coupling or heterogeneous integration, for example by wafer bonding 3 , contact printing 4,5 or direct growth 6,7 , respectively. Hence, a laser source made of a direct bandgap group IV material would further boost lab-on-a-chip and trace gas sensing 15 as well as optical interconnects 18 by enabling monolithic integration. In this context, Ge plays a prominent role since the conduction band minimum at the -point of the Brillouin-zone (referred to as -valley) is 3 located only approx. 140 meV above the fourfold degenerate indirect L-valley. To compensate for this energy difference and thus form a laser gain medium, heavy n-type doping of slightly tensile strained Ge has been proposed 19 . Later, laser action has been reported for optically 20 and electrically pumped Ge 21 doped to approx. 1 and 4×10 19 cm -3 , respectively. However, pump-probe measurements of similarly doped and strained material did not show evidence for net gain 22 , and in spite of numerous attempts, researchers failed to substantiate above results up to today. Other investigated concepts concern the engineering of the Ge band structure towards a direct bandgap semiconductor using micromechanicallystressed Ge nanomembranes 9 or silicon nitride (Si 3 N 4 ) stressor layers 23 . Very recently, Süess et al. 10 presented a stressor-free technique which enables the introduction of more than 5.7 % 24 uniaxial tensile strain in Ge µ-bridges via selective wet under-etching of a pre-stressedlayer. An alternative technique in order to achieve direct bandgap material is to incorporate Sn atoms into a Ge lattice, which primarily reduces the gap at the -point. At a sufficiently high fraction of Sn, the energy of the -valley decreases below that of the L-valley. This indirect-to-direct transition for relaxed GeSn binaries has been predicted to occur at about 20 % Sn by Jenkins et al. 25 , but more recent calculations indicate much lower required Sn concentrations in the range of 6.5-11.0 % 26,27 . A major challenge for the realization of such GeSn alloys is the low (< 1 %) equilibrium solubility of Sn in Ge 28 and the large lattice mismatch of about 15 % between Ge and -Sn. For GeSn grown on Ge substrates, this mismatch induces biaxial compressive strain causing a shift of the  and L-valley crossover towards higher Sn concentrations ...

Optical frequency combs act as rulers in the frequency domain and have opened new avenues in many fields such as fundamental time metrology, spectroscopy and frequency synthesis. In particular, spectroscopy by means of optical frequency combs has surpassed the precision and speed of Fourier spectrometers. Such a spectroscopy technique is especially relevant for the mid-infrared range, where the fundamental rotational-vibrational bands of most light molecules are found. Most mid-infrared comb sources are based on down-conversion of near-infrared, mode-locked, ultrafast lasers using nonlinear crystals. Their use in frequency comb spectroscopy applications has resulted in an unequalled combination of spectral coverage, resolution and sensitivity. Another means of comb generation is pumping an ultrahigh-quality factor microresonator with a continuous-wave laser. However, these combs depend on a chain of optical components, which limits their use. Therefore, to widen the spectroscopic applications of such mid-infrared combs, a more direct and compact generation scheme, using electrical injection, is preferable. Here we present a compact, broadband, semiconductor frequency comb generator that operates in the mid-infrared. We demonstrate that the modes of a continuous-wave, free-running, broadband quantum cascade laser are phase-locked. Combining mode proliferation based on four-wave mixing with gain provided by the quantum cascade laser leads to a phase relation similar to that of a frequency-modulated laser. The comb centre carrier wavelength is 7 micrometres. We identify a narrow drive current range with intermode beat linewidths narrower than 10 hertz. We find comb bandwidths of 4.4 per cent with an intermode stability of less than or equal to 200 hertz. The intermode beat can be varied over a frequency range of 65 kilohertz by radio-frequency injection. The large gain bandwidth and independent control over the carrier frequency offset and the mode spacing open the way to broadband, compact, all-solid-state mid-infrared spectrometers.

Artificial cavity photon resonators with ultrastrong light-matter interactions are attracting interest both in semiconductor and superconducting systems, due to the possibility of manipulating the cavity quantum electrodynamic ground state with controllable physical properties. We report here experiments showing ultrastrong light-matter coupling in a terahertz metamaterial where the cyclotron transition of a high mobility two-dimensional electron gas is coupled to the photonic modes of an array of electronic split-ring resonators. We observe a normalized coupling ratio Ω ωc = 0.58 between the vacuum Rabi frequency Ω and the cyclotron frequency ω c . Our system appears to be scalable in frequency and could be brought to the microwave spectral range with the potential of strongly controlling the magnetotransport properties of a highmobility 2DEG.

High-power and highly directional semiconductor cylinder-lasers based on a new optical resonator with deformed cross-section are reported. In the favorable directions of the far-field, a power increase of up to three orders of magnitude over the conventional circularly symmetric lasers was obtained. A "bowtie"-shaped resonance is responsible for the improved performance of the lasers in the higher range of deformations, in contrast to "whispering-gallery"-type modes of circular and weakly deformed lasers. This new resonator design, although demonstrated here in mid-infrared quantum-cascade lasers, should be applicable to any laser based on semiconductors or other high-refractive index materials.Lasers consist of two basic components. First, the active material in which light of a certain wavelength range is generated from an external energy source, such as electric current; second, the laser resonator, which contains the active material, provides feedback for the stimulated emission of light. The resonator largely influences the special features of the emitted light: power, beam directionality, and spectral properties, as well as the laser's physical features such as size and shape. Semiconductor lasers are the most widely used and versatile class of lasers. Their most common resonators are FabryPerot cavities, in which two cleaved semiconductor crystal planes act as parallel mirrors, reflecting the light back and forth through the active material.There have been many attempts to improve resonator properties. In particular, an increase of the reflectivity of the resonator mirrors is highly desirable. This allows low thresholds for the onset of laser action and a smaller volume of active material with concomitant moderate energy requirements and the ability to pack the lasers in a small space. * To whom correspondence should be addressed; email:fc@lucent One excellent example is the development of microdisk semiconductor lasers (1). These lasers exploit total internal reflection of light to achieve a mirror reflectivity near unity. Micro-disk, -cylinder or -droplet lasers form a class of lasers based on circularly symmetric resonators, which lase on "whispering-gallery modes" of the electromagnetic field (2,3,4). In such a mode light circulates around the curved inner boundary of the resonator, reflecting from the walls of the resonator with an angle of incidence always greater than the critical angle for total internal reflection, thus remaining trapped inside the resonator. There are only minute losses of light caused by evanescent leakage (tunneling) and scattering from surface roughness. This principle allowed the fabrication of the world's smallest lasers (2). Besides possible future applications in optical computing and networking, micro-lasers are of strong interest for research problems of cavity quantum electrodynamics, such as resonatorenhanced spontaneous emission and threshold-less lasers (5). Small resonators may also serve as model systems for the study of wave phenomena in mesoscopic systems, parti...