Category Archives: Semiconductors

By Yoichiro Ando

Artificial intelligence (AI) is on the verge of transforming entire industries as it gears up to power semiconductor industry innovation and growth, thrusting the technology to front and center at SEMICON Japan 2019, December 12-14 at the Tokyo Big Sight (Tokyo International Exhibition Center).

A number of Japanese startups are on leading edge of AI innovation in machine and deep learning. One is Preferred Networks Inc., a company that applies cutting-edge deep learning technology to Internet of Things (IoT) applications across transportation, manufacturing and healthcare.

In his opening day keynote At SEMICON Japan 2019, Toru Nishikawa, president and CEO of Preferred Networks, Inc., will highlight the latest developments and promise of using deep learning for industrial applications. Nishikawa will unpack how AI companies jockeying for competitive advantage will win by harnessing technologies to process massive amounts of data efficiently and quickly.

Following is look at Preferred Networks, Inc. and five other Japanese startups that are driving AI innovation.

Within Japan’s world of AI, machine learning, and deep dearning, Preferred Networks is likely the most well-known Japanese company. The parent company, Preferred Infrastructure, was founded in March 2006 by Toru Nishikawa and Daisuke Okanohara, who focused on search engine development before turning to machine learning and establishing Preferred Networks to commercialize the technology.

Preferred Networks established itself as one of the world’s top providers of machine learning technology with the development of Chainer – an open source deep learning framework that has been offered free of charge since June 2015 and was released before TensorFlow, Google’s renowned Deep Learning framework.

Established in 2012, ABEJA is thought to be Japan’s first venture company to specialize in deep learning. ABEJA’s core technology is its AI platform ABEJA Platform. Based on this platform, the company offers various solutions to more than 100 client companies. ABEJA also offers ABEJA Insight, a specialized package service for the retail and distribution, manufacturing, and infrastructure industries.

Data analytics provider BrainPad Inc. was the first Japanese AI venture listed on the Tokyo Stock Exchange. Established in 2004, before the advent of big data, BrainPad Inc. cultivated a vision of analyzing vast amounts of data in increase the competitiveness of Japanese companies.

LeapMind Inc. aims to offer deep learning technology that uses fewer computing resources and draws less power. Both are important capabilities since deep learning requires considerable computing resources to perform image and speech recognition. The company’s answer to this deep learning challenge is a small form factor FPGA with low power consumption.

In April 2018, LeapMind started offering the tool DeLTA-Lite to support model construction for Deep Learning. The tool simplifies the development of deep learning design models, eliminating the need for model design, hardware, and software expertise.

Hacarus Inc.’s HACARUS-X AI technology, which combines sparse modeling and machine learning technology, features low power consumption and small devices such as FPGAs. In collaboration with semiconductor trading company PALTEK, Hacarus is integrating HACARUS-X algorithms with Xilinx’s FPGA Zynq UltraScale + MPSoC. Both companies area also implementing HACARUS-X algorithms in a box computer.

Sparse modeling is gaining attention as a modeling method by which humans can understand the judgment process of AI by extracting features from a small amount of learning data.

With expertise in life science fields such as medical and biology and image processing technology, LPixel, Inc. develops image analysis systems with original algorithms and machine learning techniques. It has developed a cloud-based AI image analysis platform and an AI medical image diagnosis support technology that streamlines the review of large amounts of research data and detects image fraud in research papers and other documents for the medical and biology fields, freeing researchers to devote more time to their core work.

Yoichiro Ando is a marketing director at SEMI Japan. 

Cabot Microelectronics Corporation (Nasdaq: CCMP), today announced that it has completed its previously announced acquisition of KMG Chemicals, Inc.  As a result of the acquisition, KMG has become a wholly owned subsidiary of Cabot Microelectronics.  Under the terms of the definitive agreement, each share of KMG common stock was converted into the right to receive $55.65 in cash and 0.2000 of a share of Cabot Microelectronics common stock, without interest and with cash paid in lieu of any fractional shares.

The acquisition will extend and strengthen Cabot Microelectronics’ position as one of the leading suppliers of consumable materials to the semiconductor industry.  Additionally, the combined company will be a leading global provider of performance products and services for improving pipeline operations and optimizing throughput.  The transaction is expected to be significantly accretive to Cabot Microelectronics’ free cash flow and adjusted earnings per share in year one, excluding any acquisition-related costs.

“I am pleased to announce that we have completed the KMG transaction.  We welcome KMG’s employees to our team and look forward to our journey together towards becoming the premier global provider of semiconductor and specialty materials.  We believe that our employees, customers and shareholders will benefit from this transaction as we become a stronger company, focused on providing high-performing and innovative solutions to our customers,” said David Li, President and CEO of Cabot Microelectronics.  “KMG’s industry-leading electronic materials business will expand our CMP product offerings with high-purity solutions used throughout the semiconductor manufacturing process.  We are also excited about the addition of KMG’s performance materials businesses to our portfolio which will allow us to expand our participation into new markets including the attractive, high-growth pipeline performance segment.”

In connection with the acquisition, Cabot Microelectronics borrowed $1.065 billion under a new senior secured term loan facility, the proceeds of which were used to finance in part the cash portion of the merger consideration, to repay KMG’s existing indebtedness and to pay fees and expenses related to the acquisition.  Cabot Microelectronics issued approximately 3.2 million shares of common stock to holders of KMG common stock for the stock portion of the merger consideration.

FormFactor, Inc. (NASDAQ: FORM), an electrical test and measurement supplier to the semiconductor industry, announced that it has been recognized for supplier excellence by SK hynix, a manufacturer of DRAM, Flash and CMOS image sensor technology. FormFactor supplies SK hynix with advanced wafer probe cards for high bandwidth memory (HBM) and high-throughput, one-touchdown DRAM testing, as well as engineering probe systems that enable measurements in extreme environments to support cutting edge semiconductor devices.

“We are proud to be recognized as an outstanding supplier by SK hynix, and especially honored to be selected for our leadership in Technology Innovation,” said FormFactor’s CEO, Mike Slessor. “With a relationship of more than a decade, we continue to deliver innovative technologies to help them achieve their test objectives from their analytical labs to production test in their fabs. These innovations include high-parallelism, high-speed wafer probe test technologies that deliver on the quality and reliability that our customers demand. We’ve also provided probe system technology that pushes next generation devices to new limits in extreme conditions. These examples demonstrate FormFactor’s unique ability to support customers’ most demanding applications across the spectrum of electrical test and measurement.”

The SK hynix awards are given to a select few suppliers who demonstrate commitment to exceptional performance in support of the company’s development and manufacturing objectives.

In its annual customer survey earlier this year, FormFactor was recognized by VLSIresearch for its high customer satisfaction. “FormFactor’s high rankings reflect the company’s close partnerships with its customers,” said G. Dan Hutcheson, CEO of VLSIresearch. “Technical leadership, partnering and trust are all key to achieving consistently high results. These factors combined with excellent support and quality help FormFactor to stand out year after year in its categories.”

Micron Technology, Inc., (Nasdaq: MU) today announced at Electronica 2018 that it will collaborate with the BMW Group to further advance the development of automotive memory solutions used in vehicles. Memory and storage are key components in accelerating the intelligence and user experience of next-generation systems in vehicles, including in-cabin infotainment as well as advanced driver-assistance systems (ADAS) technology, which together play an important role in making self-driving autonomous cars a reality.

Micron and the BMW Group will intensify their existing efforts toward testing and development of automotive memory solutions at Micron’s state-of-the-art lab in Munich, Germany. Using the Test Automation Framework of the BMW Group as a car emulator platform, the two companies will work together to define and validate memory and storage solutions for next-generation platforms. The collaborative effort will leverage Micron’s memory and storage technology expertise, along with its broad portfolio of DRAM, NAND, and NOR technologies, including LPDRAM, e.MMC, UFS and SSD storage solutions.

As a proven memory partner for automotive manufacturers, Micron recognizes the importance of validating and testing new automotive memory technologies for robustness and reliability before releasing them into the market. Micron’s customer lab expertise in developing innovative automotive memory technologies will enable the BMW Group to raise the quality of the driving experience in automobiles of the future.

“The incorporation of new features and capabilities in advanced in-vehicle infotainment (IVI) and ADAS, such as voice recognition, hand gesturing and image recognition, are driving an explosive growth in both volatile and nonvolatile memory embedded in vehicles, accelerating intelligence at the edge,” said Giorgio Scuro, vice president of Micron’s automotive division. “Micron has a long-standing record working with automotive industry partners, and this joint initiative with the BMW Group is a testament to our expertise in bringing innovative automotive memory technologies to market.”

As a leading memory partner with more than 25 years of experience, Micron provides advanced automotive memory solutions that meet stringent quality, reliability and compliance requirements. Micron’s broad portfolio of volatile and nonvolatile memory products are optimized for automotive and supported by a formal product longevity program.

Scientists at Nagoya Institute of Technology (NITech) and collaborating universities in Japan have gained new insights into the mechanisms behind degradation of a semiconductor material that is used in electronic devices. By highlighting the specific science behind how the material degrades, they are making way for potential discoveries that may prevent the performance degradation of the material.

The study was published in the Journal of Applied Physics in September of 2018. The scientists used Silicon Carbide (SiC) material for the experiment. SiC is becoming a more popular alternative to standard semiconductor materials for electronic devices. The study is based on a specific type of SiC material that is characteristic for its structure, or 4H-SiC. This material was exposed to both photoluminescence as well as various temperatures as a means to create specific kinds of deformations that lead to the degradation of SiC-based devices. The scientists were able to observe how these deformations actually take place on an atomic level.

“We quantified the speed at which electric charge particles move in regions of 4H-SiC material where the atomic structure has been defected. This will usher discoveries of ways to suppress degradation of SiC-based devices such as power electronic systems,” states Dr. Masashi Kato, an associate professor at the Frontier Research Institute for Materials Science in NITech.

In order to better understand the actual mechanism behind atomic deformation that lead to degradations, the researchers used photoluminescence to induce movement of electric charge particles and measured the speeds at which that took place. They looked for specific factors that may limit particle movement, including the material that was used.

They also tested the effects of increasing temperature, specifically looking to see if higher temperatures will increase or decrease rate of deformation.

According to Dr. Kato, the presence of a particular kind of atomic deformation that causes the material degrade is particularly problematic for SiC-based power devices. “While a particular SiC-based device is in operation, the atoms of the material deform, which leads to degradation. The process by which these atoms deform is not clear yet. What is known, however, is that movement of electric charge within the material as well as areas where the material has become defect already contribute to the aforementioned atomic deformation,” he states.

So far similar experiments have been conducted in the past by other researchers, the results that have been reported are not consistent. Here, the result of experiments with photoluminescence indicates that the carrier recombination in single Shockley stacking faults (1SSFs) and at partial dislocations (PDs) is faster than that in regions without 1SSFs in 4H-SiC. Such fast recombination will induce the degradation of the device with 1SSFs. In addition, 1SSF expansion velocity also increases with temperature increase.

As such, they pave the way for research that will revolve around the slowing of SiC-based devices degradation. This, in turn, could potentially result in higher quality and more durable devices.

Along those lines, the authors state that their future research endeavors will focus on finding out ways to prevent SiC-based devices from degrading as well as creating devices that will not wear down over time.

Micron Technology, Inc., (Nasdaq: MU) today announced that its GDDR6 memory, Micron’s fastest and most powerful graphics memory, will be the high-performance memory of choice supporting Achronix’s next-generation stand-alone FPGA products built on TSMC 7nm process technology. GDDR6 is optimized for a variety of demanding applications, including machine learning, that require multi-terabit memory bandwidth and will enable Achronix to offer FPGAs at less than half the cost of FPGAs with comparable memory solutions.

Achronix’s high-performance FPGAs, combined with GDDR6 memory, are the industry’s highest-bandwidth memory solution for accelerating machine learning workloads in data center and automotive applications.

This new joint solution addresses many of the inherent challenges in deep neural networks, including storing large data sets, weight parameters and activations in memory. The underlying hardware needs to store, process and rapidly move data between the processor and memory. In addition, it needs to be programmable to allow more efficient implementations for constantly changing machine learning algorithms. Achronix’s next-generation FPGAs have been optimized to process machine learning workloads and currently are the only FPGAs that offer support for GDDR6 memory.

“From GPUs and beyond, Micron delivers high-performance memory solutions that meet the needs of today’s most demanding applications, including artificial intelligence (AI) and machine learning — most recently demonstrated by achieving throughput of up to 16 Gb/s on our GDDR6 solutions,” said Andreas Schlapka, director of Micron’s networking segment. “In addition to offering increased performance, Micron has developed an ecosystem to support companies like Achronix whose FPGAs with GDDR6 will enable rapid creation of designs. This, in effect, translates into faster time to market for customers using this powerful new memory technology.”

Achronix’s next-generation FPGAs include up to eight hardened GDDR6 memory interfaces that provide customers the flexibility to choose from multiple memory configurations for their end application. Customers can use from one to eight GDDR6 memory devices, which can offer over 4 Tb/s memory bandwidth and from 8Gb to 128Gb density. This type of flexibility allows customers to optimize cost and power for their application, which is not currently possible with alternate high-bandwidth memory solutions like HBM2.

“With more than 14 years’ experience in developing high-performance FPGAs, Achronix is the first FPGA company to support GDDR6 memory and deliver multi-terabit memory bandwidth at the lowest cost for data center, blockchain, networking and automotive applications that require the highest-performance programmable platform,” said Manoj Roge, vice president of strategic planning and business development at Achronix Semiconductor Corporation. “Achronix is excited to work closely with Micron and other ecosystem partners to accelerate time to market for GDDR6-based solutions for our customers’ most demanding workloads and applications.”

Micron works closely with partners like Achronix to accelerate engineering efforts to build robust models and toolsets and deliver board layout validation. Through this ecosystem approach, Micron delivers high-bandwidth memory technology that provides a path for engineers to incorporate GDDR6 in designs and bring bandwidth-intensive applications to market.

By Emir Demircan

Joining distinguished speakers from the European Commission, industry, academia and Member States, Laith Altimime, SEMI Europe president, will keynote on “European Competitiveness in the Context of the Global Digital Economy” on 20 November at the European Forum for Electronic Components and Systems (EFECS) in Lisbon, Portugal.

Players across the European electronics manufacturing value chain will gather 20-22 November, 2018, at EFECS to share the industry’s vision and set the future direction of technology innovation. Themed “Our Digital Future,” this year’s forum focuses on how rapid innovation in electronics components and systems-based applications are shaping Europe’s digital future. Start-ups, SMEs, research institutes, academia, large and medium enterprises and public authorities will learn about new collaboration initiatives and the latest developments in European funding instruments while offering their expectations for future funding programmes.

Organized by AENEAS, ARTEMIS-IA, EPoSS, ECSEL Joint Undertaking and the European Commission, in association with EUREKA, EFECS will also highlight the impact and results of various European funding instruments.

For more information about the event, please click here.

Emir Demircan senior manager for advocacy and public policy at SEMI Europe. 

EV Group (EVG), a supplier of wafer bonding and lithography equipment for the MEMS, nanotechnology and semiconductor markets, today announced that IHP – Innovations for High Performance Microelectronics (IHP), a German research institute for silicon-based systems, highest-frequency integrated circuits, and technologies for wireless and broadband communication, has purchased an EVG® ComBond® automated high-vacuum wafer bonding system for use in developing next-generation wireless and broadband communication devices.

The EVG ComBond features micron-level wafer-to-wafer alignment accuracy and room-temperature covalent bonding, which enables a wide variety of substrate and interconnect combinations for producing advanced engineered substrates, next-generation MEMS and power devices, stacked solar cells, and high-performance logic and “beyond CMOS” devices. The ability to conduct oxide-free aluminum-to-aluminum (Al-Al) direct bonding at low temperature is a unique capability of the EVG ComBond platform, and is among the new bonding applications that IHP will explore with the system.

The EVG ComBond® features micron-level wafer-to-wafer alignment accuracy and room-temperature covalent bonding, which enables a wide variety of substrate and interconnect combinations.

Covalent bonding enables wafer-level packaging and heterogeneous integration

Heterogeneous integration through wafer-level-packaging (WLP) — where multiple semiconductor components with different design nodes, sizes or materials are combined into a single package at the wafer level — is key to extending the semiconductor technology roadmap. Metal and hybrid wafer bonding are key process technologies for WLP and heterogeneous integration due to their ability to enable ultra-fine pitch interconnections between the stacked devices or components. The continuous drive to higher performance and functionality of these integrated systems requires constant reductions in the dimensions and pitch of the interconnects — which in turn drives the need for tighter wafer bond alignment accuracy.

In addition, for certain WLP applications, Al-Al direct bonding is a promising new method of metal-based bonding due to aluminum’s low cost coupled with its high thermal and electrical conductivities. However, conventional Al-Al thermo-compression bonding requires high temperatures and bond forces to provide reliable bonding interfaces — making it incompatible with heterogeneous integration efforts.

According to Paul Lindner, executive technology director at EV Group, “Combining different materials and device components into a single package has taken on greater importance in adding performance and value to electronic devices. The EVG ComBond facilitates the bonding of nearly ‘anything on anything’ in wafer form. This provides our customers with a powerful solution for researching new material combinations for future semiconductor devices. Its micron-level alignment capability also makes the EVG ComBond uniquely suited for use in high-volume manufacturing of emerging heterogeneous integration device designs.”

EVG’s breakthrough ComBond wafer activation technology and high-vacuum handling and processing allow the formation of covalent bonds at room or low temperature for fabricating engineered substrates and device structures. The EVG ComBond facilitates the bonding of heterogeneous materials with different lattice constants and coefficients of thermal expansion (CTE) as well as the formation of electrically conductive bond interfaces through a unique oxide-removal process. The EVG ComBond maintains a high-vacuum and oxide-free environment throughout the entire bonding process, enabling low-temperature bonding of metals, such as aluminum, that re-oxidize quickly in ambient environments. Void-free and particle-free bond interfaces and excellent bond strength can be achieved for all material combinations.

GLOBALFOUNDRIES and indie Semiconductor today announced the release of a new generation of customized microcontrollers on GF’s 55nm Low Power Extended (55LPx) automotive-qualified platform, which includes embedded non-volatile memory (SuperFlash®) technology. indie Semiconductor’s new Nigel products are based on ARM Cortex-M4 microcontroller cores, capable of supporting advanced functionalities in IoT, medical and automotive markets. indie Semi is already shipping products, manufactured on GF’s 55LPx process, to automotive customers in volume.

indie’s custom microcontrollers integrate in a single device mixed-signal functionality for sensing, processing, controlling and communicating. GF’s 55LPx platform, with SST’s SuperFlash® memory technology, enables the use of high-density memory and high-performance processing combined with mixed-signal functions in indie’s Nigel M4 controllers, delivering a highly integrated automotive solution at 55nm node.

“indie’s Nigel controller is designed to support high performance computing for automotive system architectures,” said Paul Hollingworth, executive vice president of sales and marketing at indie Semiconductor. “As automotive system requirements get more complex, our customers need solutions to perform complex processing while combining multiple functions into a single chip to minimize size and weight. We chose GF’s automotive-qualified 55LPx platform for its combination of density, performance and cost.”

“GF is pleased to be working with indie Semiconductor, a leader in state-of-the-art SoC technology,” said Rajesh Nair, vice president of mainstream offering management at GF. “indie Semiconductor joins our rapidly growing client base for GF’s 55LPx platform, which offers a combination of superior low-power logic, embedded non-volatile memory, extensive IP, and superior reliability for consumer, industrial and automotive grade 1 applications.”

The 55LPx RF-enabled platform provides a fast path-to-product solution that includes silicon-qualified RF IP and Silicon Storage Technology’s (SST) highly reliable embedded SuperFlash® memory. The platform is in volume production on GF’s 300mm line in Singapore. In addition to Nigel, indie Semiconductor is currently developing several products on the technology, many of which are for automotive applications.

Process design kits and an extensive offering of silicon proven IP are available now. For more information on GF’s mainstream CMOS solutions, contact your GF sales representative or go to globalfoundries.com.

Micron Technology, Inc. today announced the company has joined CERN openlab, a unique public-private partnership, by signing a three-year agreement. Under the agreement, Micron will provide CERN with advanced next-generation memory solutions to further machine learning capabilities for high-energy physics experiments at the laboratory. Micron’s memory solutions that combine neural network capabilities will be tested in the data-acquisition systems of experiments at CERN.

High-energy physics scientists are looking to deploy leading-edge technologies that can support their experiments’ computing and data processing requirements. Memory plays a vital role in accelerating intelligence by processing vast amounts of data, helping researchers gain valuable insights from data generated by high-energy physics experiments.

As part of the work with CERN, Micron will develop and introduce a specially designed Micron memory solution that will be tested by researchers at CERN for use in rapidly combing through the vast amount of data generated by experiments. The project will feature FPGA-based boards with Micron’s most advanced high-performance memory combined with an advanced neural network technology developed in collaboration between Micron and FWDNXT, a provider of deep learning and AI solutions.

“Micron is committed to pushing the limits of innovation by providing high-performance memory and storage solutions to solve the world’s greatest computing and data processing challenges in data analytics and machine learning,” said Steve Pawlowski, vice president of advanced computing solutions at Micron Technology. “We’re proud to work with CERN to deliver machine learning capabilities that will enable high-energy physics scientists to make advances in their science and research experiments.”

“CERN collaborates openly with both the public and private sector, and working with technology partners like Micron helps ensure that members of the research community have access to the advanced computing technologies needed to carry out our groundbreaking work,” said Maria Girone, CTO at CERN openlab. “It is critical to the success of the Large Hadron Collider that we are able to examine the petabytes of data generated in a fast and intelligent manner that enables us to unlock new scientific discoveries. These latest-generation memory solutions from Micron and machine learning solutions from FWDNXT offer significant potential in terms of enabling us to process more data at higher speeds.”

Micron will demonstrate its high-performance memory solutions running FWDNXT’s Machine Learning SDK at SC18, November 12-15, in Dallas, Texas.