Why the WTMF-3 gathering in Berlin is important
Professor Ton Backx is President of the Institute for Photonic Integration at Eindhoven University of Technology. Ton led the establishment of a national “photonics delta” in 2015. As we approach the third major international discussion forum in June 2019, Ton reflects on the achievements so far and the success of involving industry.
“So far, Integrated Photonics has developed using various materials for the chip wafers. In the Netherlands, the expertise is centred around Indium Phosphide and Silicon Nitride. Other countries like Belgium, France, Japan, the USA, and Taiwan have built a different parallel focus around Silicon photonics, which is the material used in the existing micro-electronics industry.
As the complexity of these next generation chips increases, light losses in passive light processing need to be compensated for by using light amplification and additional light sources. That puts limits on the chip's functionality, if only passive circuits can be used, because at a certain moment, all the light is gone. Active components on the chip will be needed to regenerate and amplify the light. In Europe, we believe that Indium Phosphide is going to be needed in all the scenarios because when you need hundreds of laser sources and light amplifiers, you are not going to be able to use external discrete lasers and amplifiers anymore.
Building on the shoulders of giants
Today, we take it for granted that extremely complex technologies fit in the palm of a hand. Most of that is possible today thanks to careful technology planning in the 80’s and 90’s by international groups like the International Technology Roadmap for Semiconductors.
Manufacturing of an integrated circuit requires a series of operations including chip design, mask set manufacturing, wafer manufacturing, photolithography, various wafer processing steps, metal deposition, on wafer testing, dicing, dice testing, chip packaging and chip testing. In the beginning industries developed their own manufacturing processes covering all steps. As the industry evolved, companies started making use of specialized machines built by different commercial companies to their own standards.
But this specialization made it difficult for the semiconductor industry to advance with the speed dictated by Moore’s law, since one company could not successfully mass produce a new product if essential components and systems were not available around the same time. A technology roadmap can help to solve this timing problem by signalling when a certain capability or performance will be needed. Then each supplier can target this date for their piece of the puzzle.
The ITRS grew into a trusted technology assessment network. It worked because it was always independent of any commercial interests pertaining to individual products or equipment.
We need the same approach for Photonics!
In 2016, I met with Lionel Kimerling (photo), who leads the AIM Photonics Academy at MIT, Cambridge Massachusetts. We jointly agreed that a global approach to mapping was essential, if the photonics industry was going to successfully scale up in the same way as semi-con. Shared manufacturing platforms and underlaying standards for photonic integrated circuit design are key to cost-effective, high-volume manufacturing. This is going to be needed by emerging industries like 5G Telecom, next generation datacentres and a wide variety of smart “connected” sensors, more commonly known as the Internet of Things.
But eighteen months ago, there were two very distinct photonics roadmaps.
- From Eindhoven University of Technology, we started the Joint European Platform for the Integration of Photonics Systems and Circuits. The JePPIX eco-system includes many businesses spread across process development, chip fabrication, packaging, software development, design and training. Since 2012 it has grown to a trusted network of several hundred international users. We have focussed on the highly complementary technologies of high-performance Indium Phosphide (Eindhoven area) and low-loss dielectric Si3N4 based (TriPleX) waveguide technologies developed in the Twente region of the Eastern Netherlands. The latest JePPIX roadmap was issued in 2018.
- Meanwhile, our colleagues in the USA had established the “Integrated Photonic Systems Roadmap (IPSR)”. This is a collaborative program organized by the International Electronics Manufacturing Initiative and the MIT Microphotonics Center and funded by the American Institute for Manufacturing Integrated Photonics. The 2017 IPSR roadmap can be found here.
The goal of both these roadmaps is to address technology gaps and challenges that are limiting the advancement of hardware technology for integrated photonic system manufacturing.
“We have capitalized on the extraordinary level of trust that was built up in Brabant and last year’s forum in Enschede”, continues Ton. “Since these meeting, we've been working out the details gleaned from the parallel technical working groups, which have been running since 2017. Based on that wealth of data and ideas, we streamlined the interactions to make the findings more coherent and decide what is still missing. As well as the European summer meeting, there have been discussions in Cambridge, MA fall meetings, the January winter meeting in Tokyo, Japan and most recently in Sunnyvale California.
The first draft version of a single IPSR-I roadmap is now under discussion. This has been prepared together with Berenschot, who are well-known technical strategy consultants in the Netherlands. They have been tapping in to the experiences of our colleagues at AIM Photonics Academy to ensure a collaborative process that is manageable and yet has short communications lines.”
Bringing the two plans together in Berlin
“We are in the final stages of combining the two existing roadmaps into a single coherent roadmap. If we look at next generation devices that will come on the market in the next three years, you will see several competing solutions from the separate silicon, silicon nitride and indium phosphide worlds. That is fine for the short-term.
One of the drawbacks of silicon is that you can only produce passive circuits. When the circuit complexity is relatively low, you can make use of external discrete laser sources or simple light sources integrated on chip. There are many manufacturers who can supply those discrete lasers today. But look ahead 5 years and we see a very different world.”
“By 2024 the global market will be asking for transceivers that operate at a speed of 1.6 Terabit per second. If we take that as an end-goal, we should compare it with the situation in 2019 where 0.1 Terabit per second (Tb/sec) is common and 0.4 Tb/sec is widely introduced. Today, communication links between datacentres all use optical links with single mode fibres. But the processing of the data in the data centres is done electronically. The problem is that when you go from 0.1 to 1.6 Tb/sec, the light part can cope relatively easily, but the speed of the electronics becomes a severe bottleneck."
Discussion Points from Tokyo
“The winter 2019 roadmapping meeting in Tokyo, Japan was a significant step. There the goal was to get feedback on the draft roadmap from colleagues in Japan, as well as participants from Taiwan, Singapore and China.”
“We heard from our colleagues in Japan that several parties working on next generations 400 Gb/sec solutions with silicon-based PICs have concluded that for the moment, the chips will do limited optical processing. They will make use of a single or a few external laser sources to provide light to the silicon transceiver. The systems operating in that way will not have very long-distance communication capabilities. The distance they can bridge is not much more than 2-3 kilometres.
While that may sound a lot, next generation distributed data centres require high speed transceivers that can bridge distances of 50-100 kilometres. Indium phosphide transceivers have already bridged 50 kilometres and beyond.
So, one trend coming out of the IPRS-I discussions is that next generation transceivers used for short distance links will be predominantly based on silicon-based photonics fed by an external laser while indium phosphide based high speed transceivers will be used for longer distances.”
Exponential energy challenges imply disruptive solutions
“Looking five years ahead, we foresee the need for radically different designs by 2024. This is because there is mismatch between the rapidly growing energy consumption of large datacentres and the local energy supply. New architectures are needed to solve the problems encountered. We may need to build distributed datacentres and that means we will need to design much longer distance, high speed, high performance, low latency interconnects between them.”
“There is another trend we need to consider. As the complexity of these next generation chips increases, light losses in passive light processing need to be compensated for by using light amplification and additional light sources. That puts limits on the chip's functionality, if only passive circuits can be used, because at a certain moment, all the light is gone. Active components on the chip will be needed to regenerate and amplify the light. In Europe, we believe that Indium Phosphide is going to be needed in all the scenarios because when you need hundreds of laser sources and light amplifiers, you are not going to be able to use external discrete lasers and amplifiers anymore.
Towards hybrid solutions
So, assuming you need Indium Phosphide to produce the active components needed in those circuits, the discussion now is how to do that. One option is to build active islands using Indium Phosphide on the Silicon photonics chip. Another option is an Indium Phosphide chip handling all photonics processing functions on top of a Silicon carrier. The silicon carrier part can take care of the microelectronics driver and additional signal processing part. Everything in between is possible, and I believe we need to investigate both options.
By 2021/2022 we need to have a clear structured plan of how industry is going to do this, so roll out of the second chip generation is ready by 2023/4.”
Beyond Berlin – Understanding Applications.
The goal for coming year is to strongly improve the application side of the roadmap and align the technology roadmap to these application demands.
With such a broad range of applications emerging, no-one is clear yet on what best solutions are. But it's already clear to everyone that each of these materials has its specific merits in specific applications. So, when you consider specific sensor applications requiring coverage of a wide spectrum, the properties of low-loss Silicon Nitride which material is transparent over a very wide spectrum may be the best material to be applied. That’s very attractive and hardly achievable in other materials.
In Eindhoven we have started a new Europe wide pilot-line. With JePPIX we already have an ability to design and make complex photonics chips, and check whether they do what you expect them to do. The new InPulse project enables the next step. It takes the chip from concept into scalable manufacturing. This step from concept to scalable manufacturing requires robust, predictable performance and gets you to a stage where businesses can start assembling modules and shipping them to their customers.
There is an ongoing discussion as to when six-inch Indium Phosphide wafers will be needed. We had that discussion in Japan as well. For the applications that need to be developed and manufactured up to 2021, three and four inch substrates may be sufficient. Pilot-Line initiatives like InPulse are therefore important to achieve this.’
The difficulty of course is that the way this material is grown means that going to larger wafer size means you cannot guarantee that these wafers will be produced without defects. They also need to be thicker in order to be able to handle them without breaking. Or maybe we need to migrate immediately to a hybrid type of structure between Silicon and Indium Phosphide although that technology is still in early stages of development.
Growing Investor and industry interest
Many parties are closely watching the outcome of the deliberations from our technology forum. Large private and public investments still need to be made to scale up the manufacturing volume of next generation photonics chips. Investors are naturally keen to find, select and back the winning technologies that have a proven product-market fit.
Background Articles related to World Technology Mapping Forum
Many articles have been published in 2017 and 2018 on the old PhotonDelta website about the work of the World Technology Mapping Forum. Many are still current. Here is a selection: