> Jean-Luc Ledys, Picogiga General Manager

Jean-Luc Ledys, joined Picogiga in 2004 from Gemplus ( now Gemalto) Engineer, he is graduated from Institut National des Sciences Appliquées (INSA) in Rennes, France. Jean luc Ledys hold several positions as process engineer , technichal manager and organization manager at Matra-Harris, MHS et Temic Semiconductor before joining Gemplus International, smart card manufacturing, where he oversaw industrial coordination and supervised smart card and components technologies R&D.

Can you remind our readers your strategy and objectives …

Picogiga’s objective is to provide innovative materials solutions to III-V challenges, especially for new markets such as GaN with high growth potential. With over 20 years in the business, our history as an innovator gives us an in-depth understanding of customer needs. In April 2003, Picogiga International joined the Soitec Group. As such, Picogiga is now uniquely positioned in the industry: we are a world leader in MBE, and the only III-V materials company that can also offer innovative engineered substrate solutions such as SopSiC™ (silicon on polysilicon on silicon carbide), tailored to specific application challenges for high power, high frequency devices. Founded in 1985, Picogiga was the world’s first company to offer commercial molecular beam epitaxy technology (MBE) for gallium arsenide (GaAs). The company is now a world-leading supplier of epitaxial compound semiconductor materials for microwave and RF applications. Thanks to strong and steady R&D investment, the company has kept a leadership position in this domain. Now we are ready with the latest generations of GaN solutions for a broader range of HEMTs, as they become the new cornerstones of advanced wireless infrastructure and RF applications.

 

…and key technologies

Smart Cut™ technology, Soitec’s de facto industry standard for manufacturing engineered substrates (best known for SOI), is a tremendous tool for us to leverage in new generations of compound materials. It enables us to design new solutions that marry III-V compounds with engineered substrates – solutions that address cost, performance, quality and new applications. With proven experience growing epitaxial layers on GaAs and other compound semiconductor substrates using MBE technology, Picogiga International is now targeting new markets requiring specific composite materials. These include gallium nitride (GaN) and silicon carbide (SiC), and indium phosphide (InP), as well as GaAs, all of which are being pursued for high-frequency, high-power applications (e.g., mobile telecommunications, local networks and radar detection) and optoelectronic devices (e.g., detection systems, fiber optics networks, LEDs and lasers) compound semiconductors. Basic capabilities include: -Installed Capacity greater than 70000 wafers per year in 2, 3, 4 and 6 inches - Phosporus doping capability - Carbon doping - Solid and Gas source - 8 MBE reactors - development and production of application-specific engineered substrates using Smart CutTM technology

 

Could you describe the way your customers use your products? For epiwafers...

Picogiga offers a full line of MBE epiwafers for wireless/RF, telecom, radar & satellite communications, optoelectronics, military, automotive and lighting applications, as well as a growing line of engineered substrate solutions tailored to specific application needs. Epiwafers Here are some examples of how our customers use our MBE epiwafers on the illustration.

 

Could you describe the way your customers use your products? For Engineered substrates?

Engineered substrates There are three basic criteria for the starting substrate in compound semiconductors: 1. it should be a good enough lattice match to grow the subsequent epitaxial layer; 2. it should be able to sufficiently evacuate the heat of the final devices; 3. and it should be appropriately priced for the target market. Finding single bulk materials that meet all three of these prerequisites has long been a III-V stumbling block – especially for mass-market applications. Enter substrate engineering, using Soitec’s Smart Cut™ technology. Best known as the key to Soitec’s success in silicon-on-insulator (SOI) wafers, Smart Cut in fact is a kind of a wafer engineering “tool box” that can be applied to a variety of materials, slicing off a thin layer of material from one wafer and bonding it onto a wafer of a different material. Our substrate solutions leveraging Smart Cut technology are multilayer wafer structures that independently optimize the top seed layer for epitaxial growth, and the back side for thermomechanical properties. In addition to decoupling the exigencies of epitaxy and heat evacuation, Smart Cut-based wafer engineering enables us to devise innovative substrates that give our customers better solutions for cost-competitive end markets. For GaN HEMTs for RF, the Smart Cut-enabled options include SiC-on-insulator (SiCOI), GaN-on-insulator (GaNOI), SiC on poly-crystalline SiC (SiCopSiC ) and silicon-on-polycrystalline-SiC (SopSiC).

 

for SopSiC ?

SopSiC SopSiC in particular is proving to be a very effective solution, bridging the void between low-cost, low-power GaN on silicon, and high-cost, high-power SiC for GaN HEMT devices. As such, SopSiC is designed to provide cost-efficient substrate solutions for advanced high-power devices used in wireless (RF) communication systems such as radar, satellite communications and base stations. SopSiC gives our customers a significantly better performing solution than silicon—and a considerably less expensive solution than SiC. In terms of dollar/watt, SopSiC is an extremely attractive solution. The European HYPHEN project recently announced excellent initial material characterization results of GaN on SopSiC. Material characterization results have shown that all the critical performance factors (crystal quality, mobility, surface morphology and so forth) of GaN on composite substrate materials are equal to or even better than the current industry standard materials for both metallic organic CVD (MOCVD) and molecular beam epitaxy (MBE). The new composite substrates also demonstrated superior results in terms of pilot production yield and repeatability. According to the preliminary results, the epitaxy of GaN HEMT on SopSiC composite substrates is more reliable than on conventional silicon substrates. SopSiC as a substrate for GaN growth also has the advantage of being substantially cheaper and better suited to high-volumes than bulk SiC substrates. SopSiC marks the first industrialized compound epiwafer product combining both Smart Cut and MBE technologies. The SopSiC structure includes: • a high resistivity (1-1-1) silicon top layer, a good “seed” match on which to grow GaN using MBE or MOCVD technology; • an insulating buried oxide layer, which helps ensure low distortion in RF applications; • a bottom layer of polysilicon-carbide, which is available in larger diameters and is substantially cheaper than SiC but better at heat evacuation than silicon. SopSiC samples for customers are now available in 3” and 4” diameters. It can be scaled up to 6” or even 8” versions, enabling the compound industry to leverage the same economies of scale found on the silicon side of the business.

 

what are your main markets?

Gallium nitride holds terrific promise for a number of high-growth markets. GaN transistors can handle both high power and high frequencies (in the 1-20GHz range) with minimal distortion, so they could eventually replace GaAs and LDMOS in a range of wireless (RF) communication systems such as radar, satellite communications and base stations. GaN is also a leading candidate for discrete power devices (Schottky diodes or power switches) used in power conversion for everything from hybrid cars to laptop computers. The challenge is to make it economically viable. The starting substrate is the key. GaN HEMTs can be built on monocrystalline SiC substrates to handle high-end performance, or on silicon for the low-end. But, the high cost of SiC substrates is hampering the broader commercialization of GaN-based RF devices, while silicon's low-cost platform suffers from inferior thermal conductivity. Sapphire as a mid-range alternative suffers from undesirable thermal properties and a significant lattice mismatch with GaN. While GaN on both silicon and silicon carbide is part of our existing epiwafer product line for the low and high-ends respectively, Picogiga is uniquely positioned in the industry to deliver a radically different approach based on substrate engineering to solve the GaN dilemma. To see how this could be useful, consider the potential GaN market for RF applications. In this case, the ideal substrate should be electrically insulating, contribute little RF loss, and provide a seed layer that is sufficient to grow a high-quality active layer. It should also ensure high thermal conductivity, to evacuate the heat generated by the transistor. Furthermore, for GaN HEMTs to compete with existing technologies that cover that broad, mid-range of applications including radar (X and S band), satellite communications, 3G mobile communications infrastructures, and WiMAX base stations, special attention must be given to thermal management (to ensure long-term reliability and enable the device to maintain its maximum output power) – and to cost. This mid-range is where our customers are seeing a significant opportunity for leveraging the SopSiC product, as it enables the most advantageous dollar/watt.

 

Any need of VC funding for expansion?

We have the advantage of the solid backing – both technological and financial – of the Soitec Group, of which we became part in 2003. Soitec is the world leader in silicon-on-insulator (SOI) material and other engineered substrates.

 

What are Picogiga’s next steps?

Customers are currently qualifying SopSiC, which began sampling earlier this year. Response has been very positive. We will begin ramping to full production later this year. III-V RF R&D The current focus of R&D for III-V Picogiga™ products targets three RF markets: 1. Gallium Nitride (GaN) – Dedicated to high power RF applications, GaN has been under development at Picogiga since the mid-1990’s. Based on a patent filed by CRHEA and exclusively licensed to Picogiga in 2004, GaN on Si is our basic offering for cost-effective mid-frequency range solutions (up to 10GHz). In 2005, we demonstrated the world’s first single-crystal, thin-film GaN-on-insulator substrate—representing a critical step forward in enabling the development of high-performance blue and white light-emitting diodes (LEDs), as well as for improving current and future device performance in radio-frequency (RF) and discrete power applications. Smart Cut™ substrates (Si on pSiC, SiC on Insulator, SiC on pSiC or GaN/pSiC) also appear very promising for further improving performance on silicon substrates. 2. Indium Phosphide (InP) – antimonides based HBT for 40 to 150 GHz applications (such as fiber optic networks or high speed connections). 3. GaAs HBT: an InGaP HBT product family has been developed using MBE in order to avoid the issues related to MOVPE (long term reliability and wake-up instabilities linked to H2 content of the layers). Applications include mobile handsets, radar power devices and wireless RF transactions (such as remote tolls). R&D for Optoelectronic Applications Combining our expertise in epitaxy of III-V materials and the Soitec Smart Cut™ technology to develop specific Picogiga™ multilayer substrates, an R&D program has been started targeting material requirements of future optoelectronic devices. New solutions integrating Smart Cut™ transfer and bonding technology, as well as applications enabling de-bonding of substrate layers, are currently underway.