Threshold reduction and yield improvement of semiconductor nanowire lasers via processing-related end-facet optimization

Abstract

For nanowire lasers, end-facets matter; a rigorous statistical study demonstrates that short ultrasound or PDMA transfer provides optimized lasing performance.

Holistic Nanowire Laser Characterization as a Route to Optimal Design

“…Before characterisation, the NWs from these arrays were transferred onto a Si substrate via ultrasonication and drop casting. [ 31 ] The result was a large population of NWs (>5000) in the plane of the substrate, which is ideal for detailed optical characterisation.…”

A needle in a needlestack: exploiting functional inhomogeneity for optimized nanowire lasing

“…These approaches are known to produce high yield lasing. 29 Briefly, type 1 wires have an 80 nm GaAs core, eight Al x Ga (1−x) As/GaAs quantum wells (with x=0.42 and a nominal thickness of 3.5±1.5 nm), and a final 5 nm GaAs capping layer. 26,30 These wires had typical final dimensions of 3.4±0.4 µm length and 460 ± 10 nm diameter.…”

“…Here, we propose that, far from being a hindrance, the inherent spread in material and geometric properties arising from bottom-up growth can in fact be exploited to provide a statistically relevant understanding of lasing. To achieve this, we have developed an automated machine-vision enabled data acquisition approach to scale characterization to the > 10 3 nanowire level, 26 used computational approaches typically used in astrophysics to perform markerfree multi-modal characterization, 28 and begun to use statistical methods from the life sciences 29 to systematically optimize nanowire lasers in the presence of inhomogeneity. In this communication, we describe the experimental and statistical approaches used to characterize the dominant loss mechanism in silicon integrable nanolasers based on a multiple-quantum-well gain region.…”

Characterization, Selection, and Microassembly of Nanowire Laser Systems

“…Future scalable integration of NW devices will have two major requirements, (1) that as-grown NWs can be removed from their growth structure with high yield and (2) that large sets of these devices can be easily characterised before integration to ensure performance matching the application. The transfer of NWs from their growth substrate to a host wafer usually includes a fracturing stage where devices are physically 'snapped' at some point along their length, producing one of the two reflective facets required for lasing 12 . This mechanical process produces popu-lations of devices across a sample with variations in length and facet quality that in turn affect lasing threshold and modal spectrum.…”

“…The transfer of NWs from their growth substrate to a host wafer usually includes a fracturing stage where devices are physically "snapped" at some point along their length, producing one of the two reflective facets required for lasing. 12 This mechanical process produces populations of devices across a sample with variations in length and facet quality that in turn affect lasing threshold and modal spectrum. Therefore, in order to progress toward systems incorporating NW laser sources, a scalable approach is required to map the individual device performance and allow subsequent integration of particular devices with high-accuracy spatial positioning.…”