Porotech: A different kind of GaN

UK start-up, Porotech, is set to commercialise porous GaN, opening the door to high performance optoelectronic devices with unexpected properties, reports Rebecca Pool.

A relatively unknown material from a newly minted UK start-up is set to makes waves in the rapidly growing GaN market.

With a host of novel properties that open the door to high performance optoelectronic devices, 'porous GaN' is attracting more and more interest in compound semiconductor circles. And thanks to £1.5 million in investment funds, University of Cambridge spin-off, Poro Technologies - or Porotech - is on course to be one of the first companies to bring this mesoporous version of GaN to market

Porous GaN can be regarded as a semiconductor composite of solid GaN and air. As Professor Rachel Oliver, Co-founder and Chief Scientific Officer of Porotech, and Director of the Cambridge Centre for Gallium Nitride puts it: “Porous GaN is basically GaN with holes in it that are a few tens of nanometres across.”

“With porous GaN we can engineer a wide range of material properties... and offer a new material platform to build semiconductor devices on,” she adds.

Poro Technologies co-founders (left to right): Dr Tongtong Zhu (CEO), Professor Rachel Oliver (CSO), Dr Yingjun Liu (CTO)

The Porotech team creates the nanoscale porosity in GaN wafers using electrochemical etching. The etch is conductivity selective and responds differently to the material depending on its doping density. Porosity is created in doped layers while undoped layers are left undamaged, allowing complex 3D nanostructures to be created.

According to Oliver, the etchant flows to and from the doped layers via the many nanometre-scale channel-like defects - dislocations - that exist within any GaN wafer.

“Even your best quality GaN wafer will still have around 105 dislocations per square centimetre,” she explains. “So the etchant will flow down a dislocation and when it hits the doped layer will etch it very quickly to create the porosity before continuing down that channel to the next doped layer.”

“We can take an entire wafer and using this conductivity selective etching mechanism to create GaN [structures] with a whole new set of properties that haven't been available before,” she adds. “It's very cool from a commercial perspective.”

Cross-sectional SEM image of a porous/non-porous Distributed Bragg Reflector structure.

Indeed, both Oliver and Porotech chief executive, and co-founder, Dr Tongtong Zhu, are certain their porous GaN fabrication process lends itself to commercial exploitation. The method has been tried and tested wafers up to eight inches in diameter and Zhu reckons it will seamlessly scale to even larger wafer sizes.

“The first thing that inspired us to pursue this process commercially is that it works with wafers,” says Zhu. “What's more, during the process we preserve the surface quality and the integrity of the materials so anyone can take the porous GaN wafer and insert it into their production processes without any disruption.”

Devices to go

Oliver, Zhu and colleagues have already fabricated several components and device prototypes. They have demonstrated highly reflective Distributed Bragg Reflectors (DBRs) - or 'Poro Mirrors' - based on wafers comprising alternating layers of solid and nanoporous GaN. InGaN LEDs were then grown on these epi-ready DBR pseudo-substrates that were some 25% more energy efficient than standard LEDs.

“This DBR substrate is fully compatible with GaN so a customer doesn't need to process anything differently during LED fabrication but can still get a brighter and more efficient LED,” points out Zhu.

Seed round investment is now being used to develop a pilot plant in Cambridge, UK, and Porotech's first products.

Porotech has also joined forces with other institutions and start-ups to fabricate thermal sensors that used porous GaN layers as on-chip thermal insulation. And in a novel technology twist, the team has filled the pores in a porous GaN wafer with a halide perovskite to create an optoelectronic material with longer-lasting luminescence.

“These luminescent perovskites are useful in solar cells and LEDs but degrade very quickly,” says Oliver. “We can slow this degradation down with porous GaN.”

Excitingly, novel porous GaN devices could be reaching commercial markets sooner rather than later. Despite the coronavirus pandemic, the Porotech pilot plant is currently being built at Cambridge, and Oliver, Zhu and colleagues are transferring production process from the laboratory to the plant.

“The plant will be operational before June this year and we're going to start with small-scale production to show that our wafers can be produced in volume,” says Zhu. “We're currently preparing tens of wafers for customer trials and eventually intend to demonstrate that we can produce thousands of wafers a year.”

The company is also working with foundry services, wafer producers and integrated device manufacturers, and hopes to license out technology in the coming years. “We've been working to get the company off the ground for several years and we aren't going to stop now,” says Oliver. “There has been no other time when GaN is rising so quickly and we have to keep up our momentum.”