A simple method for producing what scientists have reported are record levels of strain in single-crystal silicon could lead to the ability to integrate photonics more fully with silicon structures. The method was demonstrated by a research team at the University of Surrey, which developed a one-step procedure for generating strain in semiconductors to modify the semiconductor’s band structure. Silicon semiconductors have demonstrated a limited ability to produce light due to their indirect band gap, which means that electrons cannot easily lose energy by producing photons. In direct band gap compound semiconductors, electrons lose energy easily, resulting in photon production. However, compound semiconductors are expensive to make. The ability to produce optically efficient, direct band gap semiconductors on a silicon-based platform could strengthen and advance integrated photonics. The single-step process is an ion implantation technique for generating strain in suspended membranes. It works in a way that is similar to tightening a drum skin to provide control over the degree of amorphization in and around the single-crystal membranes. The density of the material, and the tension in the neighboring crystalline regions as a result, is controlled by amorphization. Using the ion implantation method, the researchers achieved strain levels never previously reached in mesoscopic defect-free, crystalline silicon structures; the team demonstrated up to 3.1% biaxial tensile strain and 8.5% uniaxial strain in silicon, based on micro-Raman spectroscopic measurement. Even larger strains could be realized, the researchers believe, by varying the type of implant and by exploiting the underlying crystal direction. “Seeing the wrinkles annihilation and the flattening of the membranes in real time was astonishing,” researcher Mateus Masteghin said. “This new technique promises to be highly disruptive to the field of photonics, and I am looking forward to continuing developing new devices based on this proposed technique.” The team will use the ion implantation technique to generate stable, high-tensile, strained germanium layers that have the potential to create record-level optical efficiencies. Using the germanium layers and an all group-IV-based system, the researchers will demonstrate optical gain and lasing in photonic crystal nanocavities in the mid-infrared (mid-IR) wavelength. An experiment conducted by a research team at the University of Surrey helped to advance understanding of the ion beam implantation effect on thin single-crystal membranes, by creating an analogy, with a liquid droplet placed on a thin elastic membrane in which its weight creates a downward bending (bowing) of the film underneath the droplet, followed by the appearance of radial wrinkles. Courtesy of the University of Surrey. A project grant from Engineering and Physical Sciences Research Council (EPSRC) New Horizons will support this work. Germanium can be grown on silicon and has a band gap that is close to being direct. It has been theoretically predicted that by straining a germanium crystal by >2% (tensile), it will become a direct band gap semiconductor. If the ion implantation process is successfully transferred to germanium, it could enable germanium lasers to be integrated with silicon. In the group-IV semiconductor germanium, an indirect-to-direct transition in the electronic bandgap occurs at much lower strains than silicon. The combination of electronic- and photonic-band structure and strain engineering could contribute to the development of lasers on silicon with the potential to scale up. New optoelectronic devices for communications systems and infrared (IR) sensors for adding sophisticated applications such as smoke alarms, carbon monoxide sensors, and smartphones are among the technologies the work could ultimately yield. Germanium lasers could also address overheating — a potential risk to silicon-based computer systems. These lasers could be used in place of expensive III-V devices to overcome issues related to overheating. Additionally, the flexible, scalable, easy-to-control ion implantation process opens the way to high-mobility, complementary metal-oxide-semiconductor devices and easy fabrication of direct band gap germanium for silicon-compatible optoelectronics. “What excites me about this is the simplicity of the method and that it can easily be transferred to production methods,” senior research fellow David Cox said. “It will be exciting to see if this can have as significant an impact on Group-IV semiconductor photonics as Alf Adam’s long-standing legacy on the development of the strained-layer III-V-based quantum-well lasers.” The research was published in Physical Review Materials (www.doi.org/10.1103/PhysRevMaterials.5.124603).