This paper presents micro-machining results obtained by combining a short-pulse 10 ns green laser with the water jet-guided laser technology.The short pulse duration offers significant advantages for a wide range of applications.Experimental results of semiconductor grooving, edge isolation and P1 fabrication step of thin film solar cells, as well as silicon wafer dot marking, are presented.The study compares the results obtained with short and long pulse lasers.

The water jet-guided laser technology has found a broad range of applications in the micromachining field. Its principle is to focus a laser beam into a hair-thin, low-pressure water jet, onto the sample. This hybrid system prevents heat damage to the material by cooling the cutting edges in between the laser pulses; simultaneously the water jet removes the laser generated molten material. Contamination is avoided thanks to a thin water film covering the wafer surface during the cutting process. The perfectly cylindrical water jet yields highly parallel kerf walls.In this paper experimental results from thin film solar cell grooving steps P1 and P3 are presented. P1 grooving steps on Mo and TCO yielded 40 µm wide clean cuts without burrs or cracks. Former P3 step scribing tests were performed using long-pulse lasers with pulse durations ranging from 100 ns to 1 ms. We introduce recent thin film solar cell grooving results obtained using a short pulse (10 ns) Q-switched frequency doubled laser with 15 W maximum output power. Shorter pulses allow for more precise depth control and enable successful layer removal in a single pass. The lower TCO layer remains untouched and the isolation is excellent.

Multiple-wavelength laser arrays at 1.55 mu m are key components of wavelength division multiplexing (WDM) systems for increased bandwidth. Vertical cavity surface-emitting lasers (VCSELs) grown on GaAs substrates outperform their InP counterparts in several points. We summarize the current challenges to realize continuous-wave (CW) GalnNAsSb VCSELs on GaAs with 1.55 mu m emission wavelength and explain the work in progress to realize CW GaInNAsSb VCSELs. Finally, we detail two techniques to realize GalnNAsSb multiple-wavelength VCSEL arrays at 1.55 mu m. The first technique involves the incorporation of a photonic crystal into the upper mirror. Simulation results for GaAs-based VCSEL arrays at 1.55 mu m are shown. The second technique uses non-uniform molecular beam epitaxy (MBE). We have successfully demonstrated 1x6 resonant cavity light-emitting diode arrays at 850 nm using this technique, with wavelength spacing of 0.4 nm between devices and present these results.

Quantum cascade lasers based on planar quantum wells have emerged as a leading candidate for infrared laser sources. However, these lasers are ultimately limited by phonon emission, and exhibit useful optical gain only for the tranverse magnetic polarization. Quantum dot (QD) gain material to replace the planar gain regions is very attractive because the unipolar approach can then lead to both a phonon bottleneck, and surface emission. However, tunneling phenomenon is quite different for unipolar QD injection, and designs that follow the now standard approaches based on planar quantum wells are known to have unfavorable tunneling characteristics. In this paper we present a new device design based on QDs that can lead to important advantages for realizing high performance unipolar injection infrared lasers. The new quantum dot cascade laser design is based on controlling electron tunneling in the different quantum dimensional systems, from zero-dimensional to two-dimensional, to both block as well as enhance tunneling in a gain stage so as to obtain the population inversion necessary for infrared gain. This new device, the quantum dot cascade laser (QDCL), can operate with a phonon bottleneck, and therefore can exhibit improved high temperature performance in contrast to planar heterostructure unipolar devices. In addition, the zero-dimensional confinement can also provide transverse electric polarization in the radiation field, and therefore surface emission. Epitaxial growth experiments based on self-organized quantum dots to realize the new QDCL approach are presented and discussed.