Tag Archives: letter-wafer-tech

A research study on low noise and high-performance transistors led by Suprem Das, assistant professor of industrial and manufacturing systems engineering, in collaboration with researchers at Purdue University, was recently published by Physical Review Applied.

The study has demonstrated micro/nano-scale transistors made of two-dimensional atomic thin materials that show high performance and low noise. The devices are less than one-hundredth of the diameter of a single human hair and could be key to innovating electronics and precision sensing.

Many researchers worldwide are focusing attention on building the next generation of transistors from atomic scale “exotic” 2D materials such as molybdenum di-selenide. These materials are promising because they show high-performance transistor-action that may, in the future, replace today’s silicon electronics. However, very few of them are looking at yet another important aspect: the inherent electronic noise in this new class of materials. Electronic noise is ubiquitous to all devices and circuits and only worsens when the material becomes atomic thin.

A recent study conducted by Das’ research team has systematically shown that if one can control the layer thickness between 10 and 15-atomic thin in a transistor, the device will not only show high performance — such as turning the switch “on” — but also experience very low electronic noise. This unique finding is essential to building several enabling technologies in electronics and sensing using a number of emerging 2D materials. This research is a comprehensive effort of a previous finding, where Das’ team conducted the first study on noise in MoSe2 transistors.

A team of researchers at the New York University Tandon School of Engineering and NYU Center for Neural Science has solved a longstanding puzzle of how to build ultra-sensitive, ultra-small electrochemical sensors with homogenous and predictable properties by discovering how to engineer graphene structure on an atomic level.

Finely tuned electrochemical sensors (also referred to as electrodes) that are as small as biological cells are prized for medical diagnostics and environmental monitoring systems. Demand has spurred efforts to develop nanoengineered carbon-based electrodes, which offer unmatched electronic, thermal, and mechanical properties. Yet these efforts have long been stymied by the lack of quantitative principles to guide the precise engineering of the electrode sensitivity to biochemical molecules.

Davood Shahrjerdi, an assistant professor of electrical and computer engineering at NYU Tandon, and Roozbeh Kiani, an assistant professor of neural science and psychology at the Center for Neural Science, Faculty of Arts and Science, have revealed the relationship between various structural defects in graphene and the sensitivity of the electrodes made of it. This discovery opens the door for the precise engineering and industrial-scale production of homogeneous arrays of graphene electrodes. The researchers detail their study in a paper published today in the journal Advanced Materials.

Graphene is a single, atom-thin sheet of carbon. There is a traditional consensus that structural defects in graphene can generally enhance the sensitivity of electrodes constructed from it.  However, a firm understanding of the relationship between various structural defects and the sensitivity has long eluded researchers. This information is particularly vital for tuning the density of different defects in graphene in order to achieve a desired level of sensitivity.

“Until now, achieving a desired sensitivity effect was akin to voodoo or alchemy — oftentimes, we weren’t sure why a certain approach yielded a more or less sensitive electrode,” Shahrjerdi said. “By systematically studying the influence of various types and densities of material defects on the electrode’s sensitivity, we created a physics-based microscopic model that replaces superstition with scientific insight.”

In a surprise finding, the researchers discovered that only one group of defects in graphene’s structure — point defects — significantly impacts electrode sensitivity, which increases linearly with the average density of these defects, within a certain range. “If we optimize these point defects in number and density, we can create an electrode that is up to 20 times more sensitive than conventional electrodes,” Kiani explained.

These findings stand to impact both the fabrication of and applications for graphene-based electrodes. Today’s carbon-based electrodes are calibrated for sensitivity post-fabrication, a time-consuming process that hampers large-scale production, but the researchers’ findings will allow for the precise engineering of the sensitivity during the material synthesis, thereby enabling industrial-scale production of carbon-based electrodes with reliable and reproducible sensitivity.

Currently, carbon-based electrodes are impractical for any application that requires a dense array of sensors: The results are unreliable due to large variations of the electrode-to-electrode sensitivity within the array. These new findings will enable the use of ultra-small carbon-based electrodes with homogeneous and extraordinarily high sensitivities in next-generation neural probes and multiplexed “lab-on-a-chip” platforms for medical diagnostics and drug development, and they may replace optical methods for measuring biological samples including DNA.

An international team of researchers has developed a technique that, for the first time, allows single-crystal hybrid perovskite materials to be integrated into electronics. Because these perovskites can be synthesized at low temperatures, the advance opens the door to new research into flexible electronics and potentially reduced manufacturing costs for electronic devices.

Hybrid perovskite materials contain both organic and inorganic components and can be synthesized from inks, making them amenable to large-area roll-to-roll fabrication. These materials are the subject of extensive research for use in solar cells, light-emitting diodes (LEDs) and photodetectors. However, there have been challenges in integrating single-crystal hybrid perovskites into more classical electronic devices, such as transistors.

Single-crystal hybrid perovskites are preferable because single-crystalline materials have more desirable properties than polycrystalline materials; polycrystalline materials contain more defects that adversely affect a material’s electronic properties.

The challenge in incorporating single-crystal hybrid perovskites into electronics stems from the fact that these macroscopic crystals, when synthesized using conventional techniques, have rough, irregular edges. This makes it difficult to integrate with other materials in such a way that the materials make the high-quality contacts necessary in electronic devices.

The researchers got around this problem by synthesizing the hybrid perovskite crystals between two laminated surfaces, essentially creating a single-crystal hybrid perovskite sandwich. The perovskite conforms to the materials above and below, resulting in a sharp interface between the materials. The substrate and superstrate, the “bread” in the sandwich, can be anything from glass slides to silicon wafers that are already embedded with electrodes – resulting in a ready-made transistor or circuit.

The researchers can further fine-tune the electrical properties of the perovskite by selecting different halides for use in the perovskite’s chemical make-up. The choice of halide determines the bandgap of the material, which affects the color appearance of the resulting semiconductor and leads to transparent and even imperceptible electronic devices when using high-bandgap perovskites.

“We have demonstrated the ability to create working field-effect transistors using single-crystal hybrid perovskite materials fabricated in ambient air,” says Aram Amassian, corresponding author of a paper on the work and an associate professor of materials science and engineering at NC State.

“That’s of interest because traditional single-crystal materials have to be manufactured in ultra-high vacuum, high-temperature environments, and often require exquisite epitaxial growth,” Amassian says. “Hybrid perovskites can be grown from solution, essentially from an ink, in ambient air at temperatures below 100 degrees Celsius. This makes them attractive from a cost and manufacturing standpoint. It also makes them compatible with flexible, plastic-based substrates, meaning that they may have applications in flexible electronics and in the internet of things (IoT).

“That said, there are still major challenges here,” Amassian says. “For example, current hybrid perovskites contain lead, which is toxic and therefore not something that’s desirable for applications like wearable electronics. However, research is ongoing to develop hybrid perovskites that don’t contain lead or that are even entirely metal-free. This is an exciting area of research, and we feel this work is a significant step forward for the device integration of these materials, leading to the development of new technological applications.”

In microelectronic devices, the bandgap is a major factor determining the electrical conductivity of the underlying materials. Substances with large bandgaps are generally insulators that do not conduct electricity well, and those with smaller bandgaps are semiconductors. A more recent class of semiconductors with ultrawide bandgaps (UWB) are capable of operating at much higher temperatures and powers than conventional small-bandgap silicon-based chips made with mature bandgap materials like silicon carbide (SiC) and gallium nitride (GaN).

In the Journal of Applied Physics, from AIP Publishing, researchers at the University of Florida, the U.S. Naval Research Laboratory and Korea University provide a detailed perspective on the properties, capabilities, current limitations and future developments for one of the most promising UWB compounds, gallium oxide (Ga2O3).

Gallium oxide possesses an extremely wide bandgap of 4.8 electron volts (eV) that dwarfs silicon’s 1.1 eV and exceeds the 3.3 eV exhibited by SiC and GaN. The difference gives Ga2O3 the ability to withstand a larger electric field than silicon, SiC and GaN can without breaking down. Furthermore, Ga2O3 handles the same amount of voltage over a shorter distance. This makes it invaluable for producing smaller, more efficient high-power transistors.

“Gallium oxide offers semiconductor manufacturers a highly applicable substrate for microelectronic devices,” said Stephen Pearton, professor of materials science and engineering at the University of Florida and an author on the paper. “The compound appears ideal for use in power distribution systems that charge electric cars or converters that move electricity into the power grid from alternative energy sources such as wind turbines.”

Pearton and his colleagues also looked at the potential for Ga2O3 as a base for metal-oxide-semiconductor field-effect transistors, better known as MOSFETs. “Traditionally, these tiny electronic switches are made from silicon for use in laptops, smart phones and other electronics,” Pearton said. “For systems like electric car charging stations, we need MOSFETs that can operate at higher power levels than silicon-based devices and that’s where gallium oxide might be the solution.”

To achieve these advanced MOSFETs, the authors determined that improved gate dielectrics are needed, along with thermal management approaches that will more effectively extract heat from the devices. Pearton concluded that Ga2O3 will not replace SiC and GaN as the as the next primary semiconductor materials after silicon, but more likely will play a role in extending the range of powers and voltages accessible to ultrawide bandgap systems.

“The most promising application might be as high-voltage rectifiers in power conditioning and distribution systems such as electric cars and photovoltaic solar systems,” he said.

An international research team, co-led by a physicist at the University of California, Riverside, has discovered a new mechanism for ultra-efficient charge and energy flow in graphene, opening up opportunities for developing new types of light-harvesting devices.

Shining light on graphene: Although graphene has been studied vigorously for more than a decade, new measurements on high-performance graphene devices have revealed yet another unusual property. In ultra-clean graphene sheets, energy can flow over great distances, giving rise to an unprecedented response to light. Credit: Max Grossnickle and QMO Labs, UC Riverside.

The researchers fabricated pristine graphene — graphene with no impurities — into different geometric shapes, connecting narrow ribbons and crosses to wide open rectangular regions. They found that when light illuminated constricted areas, such as the region where a narrow ribbon connected two wide regions, they detected a large light-induced current, or photocurrent.

The finding that pristine graphene can very efficiently convert light into electricity could lead to the development of efficient and ultrafast photodetectors — and potentially more efficient solar panels.

Graphene, a 1-atom thick sheet of carbon atoms arranged in a hexagonal lattice, has many desirable material properties, such as high current-carrying capacity and thermal conductivity. In principle, graphene can absorb light at any frequency, making it ideal material for infrared and other types of photodetection, with wide applications in bio-sensing, imaging, and night vision.

In most solar energy harvesting devices, a photocurrent arises only in the presence of a junction between two dissimilar materials, such as “p-n” junctions, the boundary between two types of semiconductor materials. The electrical current is generated in the junction region and moves through the distinct regions of the two materials.

“But in graphene, everything changes,” said Nathaniel Gabor, an associate professor of physics at UCR, who co-led the research project. “We found that photocurrents may arise in pristine graphene under a special condition in which the entire sheet of graphene is completely free of excess electronic charge. Generating the photocurrent requires no special junctions and can instead be controlled, surprisingly, by simply cutting and shaping the graphene sheet into unusual configurations, from ladder-like linear arrays of contacts, to narrowly constricted rectangles, to tapered and terraced edges.”

Pristine graphene is completely charge neutral, meaning there is no excess electronic charge in the material. When wired into a device, however, an electronic charge can be introduced by applying a voltage to a nearby metal. This voltage can induce positive charge, negative charge, or perfectly balance negative and positive charges so the graphene sheet is perfectly charge neutral.

“The light-harvesting device we fabricated is only as thick as a single atom,” Gabor said. “We could use it to engineer devices that are semi-transparent. These could be embedded in unusual environments, such as windows, or they could be combined with other more conventional light-harvesting devices to harvest excess energy that is usually not absorbed. Depending on how the edges are cut to shape, the device can give extraordinarily different signals.”

The research team reports this first observation of an entirely new physical mechanism — a photocurrent generated in charge-neutral graphene with no need for p-n junctions — in Nature Nanotechnology today.

Previous work by the Gabor lab showed a photocurrent in graphene results from highly excited “hot” charge carriers. When light hits graphene, high-energy electrons relax to form a population of many relatively cooler electrons, Gabor explained, which are subsequently collected as current. Even though graphene is not a semiconductor, this light-induced hot electron population can be used to generate very large currents.

“All of this behavior is due to graphene’s unique electronic structure,” he said. “In this ‘wonder material,’ light energy is efficiently converted into electronic energy, which can subsequently be transported within the material over remarkably long distances.”

He explained that, about a decade ago, pristine graphene was predicted to exhibit very unusual electronic behavior: electrons should behave like a liquid, allowing energy to be transferred through the electronic medium rather than by moving charges around physically.

“But despite this prediction, no photocurrent measurements had been done on pristine graphene devices — until now,” he said.

The new work on pristine graphene shows electronic energy travels great distances in the absence of excess electronic charge.

The research team has found evidence that the new mechanism results in a greatly enhanced photoresponse in the infrared regime with an ultrafast operation speed.

“We plan to further study this effect in a broad range of infrared and other frequencies, and measure its response speed,” said first author Qiong Ma, a postdoctoral associate in physics at the Massachusetts Institute of Technology, or MIT.

Advances in the technology of material growth allow fabricating sandwiches of materials with atomic precision. The interface between the two materials can sometimes exhibit physical phenomena which do not exist in both parent materials. For example, a magnetic interface found between two non-magnetic materials. A new discovery, published today in Nature Physics, shows a new way of controlling this emergent magnetism which may be the basis for new types of magnetic electronic devices.

Sensitive magnetic imaging detects strain tunable magnetism. Credit: Kalisky Lab

Using very sensitive magnetic probes, an international team of researchers led by Prof. Beena Kalisky, of Bar-Ilan University’s Department of Physics and Institute of Nanotechnology and Advanced Materials (BINA), has found surprising evidence that magnetism which emerges at the interfaces between non-magnetic oxide thin layers can be easily tuned by exerting tiny mechanical forces. The team also includes Prof. Lior Klein, of Bar-Ilan’s Department of Physics and BINA, and researchers from DTU (Denmark) and Stanford University (USA).

Magnetism already plays a central role in storing the increasing amount of data produced by humanity. Much of our data storage today is based on tiny magnets crammed into our memory drive. One of the promising means in the race to improve memory, in terms of quantity and speed, is the use of smaller magnets. Until today the size of memory cells can be as small as a few tens of nanometers — almost a millionth of the width of a strand of hair! Further reduction in size is challenging in three main respects: the stability of the magnetic cell, the ability to read it, and the ability to write into it without affecting its neighboring cells. This recent discovery provides a new and unexpected handle to control magnetism, thus enabling denser magnetic memory.

These oxide interfaces combine a number of interesting physical phenomena, such as two-dimensional conductance and superconductivity. “Coexistence of physical phenomena is fascinating because they do not always go hand in hand. Magnetism and superconductivity, for example, are not expected to coexist,” says Kalisky. “The magnetism we saw did not extend throughout the material but appeared in well-defined areas dominated by the structure of the materials. Surprisingly, we discovered that the strength of magnetism can be controlled by applying pressure to the material”.

Coexistence between magnetism and conductivity has great technological potential. For example, magnetic fields can affect the current flow in certain materials and, by manipulating magnetism, we can control the electrical behavior of electronic devices. An entire field called Spintronics is dedicated to this subject. The discovery that tiny mechanical pressures can effectively tune the emerging magnetism at the studied interfaces opens new and unexpected routes for developing novel oxide-based spintronic devices.

MIT researchers have invented a way to fabricate nanoscale 3-D objects of nearly any shape. They can also pattern the objects with a variety of useful materials, including metals, quantum dots, and DNA.

“It’s a way of putting nearly any kind of material into a 3-D pattern with nanoscale precision,” says Edward Boyden, an associate professor of biological engineering and of brain and cognitive sciences at MIT.

Using the new technique, the researchers can create any shape and structure they want by patterning a polymer scaffold with a laser. After attaching other useful materials to the scaffold, they shrink it, generating structures one thousandth the volume of the original.

These tiny structures could have applications in many fields, from optics to medicine to robotics, the researchers say. The technique uses equipment that many biology and materials science labs already have, making it widely accessible for researchers who want to try it.

Boyden, who is also a member of MIT’s Media Lab, McGovern Institute for Brain Research, and Koch Institute for Integrative Cancer Research, is one of the senior authors of the paper, which appears in the Dec. 13 issue of Science. The other senior author is Adam Marblestone, a Media Lab research affiliate, and the paper’s lead authors are graduate students Daniel Oran and Samuel Rodriques.

Implosion fabrication

Existing techniques for creating nanostructures are limited in what they can accomplish. Etching patterns onto a surface with light can produce 2-D nanostructures but doesn’t work for 3-D structures. It is possible to make 3-D nanostructures by gradually adding layers on top of each other, but this process is slow and challenging. And, while methods exist that can directly 3-D print nanoscale objects, they are restricted to specialized materials like polymers and plastics, which lack the functional properties necessary for many applications. Furthermore, they can only generate self-supporting structures. (The technique can yield a solid pyramid, for example, but not a linked chain or a hollow sphere.)

To overcome these limitations, Boyden and his students decided to adapt a technique that his lab developed a few years ago for high-resolution imaging of brain tissue. This technique, known as expansion microscopy, involves embedding tissue into a hydrogel and then expanding it, allowing for high resolution imaging with a regular microscope. Hundreds of research groups in biology and medicine are now using expansion microscopy, since it enables 3-D visualization of cells and tissues with ordinary hardware.

By reversing this process, the researchers found that they could create large-scale objects embedded in expanded hydrogels and then shrink them to the nanoscale, an approach that they call “implosion fabrication.”

As they did for expansion microscopy, the researchers used a very absorbent material made of polyacrylate, commonly found in diapers, as the scaffold for their nanofabrication process. The scaffold is bathed in a solution that contains molecules of fluorescein, which attach to the scaffold when they are activated by laser light.

Using two-photon microscopy, which allows for precise targeting of points deep within a structure, the researchers attach fluorescein molecules to specific locations within the gel. The fluorescein molecules act as anchors that can bind to other types of molecules that the researchers add.

“You attach the anchors where you want with light, and later you can attach whatever you want to the anchors,” Boyden says. “It could be a quantum dot, it could be a piece of DNA, it could be a gold nanoparticle.”

“It’s a bit like film photography — a latent image is formed by exposing a sensitive material in a gel to light. Then, you can develop that latent image into a real image by attaching another material, silver, afterwards. In this way implosion fabrication can create all sorts of structures, including gradients, unconnected structures, and multimaterial patterns,” Oran says.

Once the desired molecules are attached in the right locations, the researchers shrink the entire structure by adding an acid. The acid blocks the negative charges in the polyacrylate gel so that they no longer repel each other, causing the gel to contract. Using this technique, the researchers can shrink the objects 10-fold in each dimension (for an overall 1,000-fold reduction in volume). This ability to shrink not only allows for increased resolution, but also makes it possible to assemble materials in a low-density scaffold. This enables easy access for modification, and later the material becomes a dense solid when it is shrunk.

“People have been trying to invent better equipment to make smaller nanomaterials for years, but we realized that if you just use existing systems and embed your materials in this gel, you can shrink them down to the nanoscale, without distorting the patterns,” Rodriques says.

Currently, the researchers can create objects that are around 1 cubic millimeter, patterned with a resolution of 50 nanometers. There is a tradeoff between size and resolution: If the researchers want to make larger objects, about 1 cubic centimeter, they can achieve a resolution of about 500 nanometers. However, that resolution could be improved with further refinement of the process, the researchers say.

Better optics

The MIT team is now exploring potential applications for this technology, and they anticipate that some of the earliest applications might be in optics — for example, making specialized lenses that could be used to study the fundamental properties of light. This technique might also allow for the fabrication of smaller, better lenses for applications such as cell phone cameras, microscopes, or endoscopes, the researchers say. Farther in the future, the researchers say that this approach could be used to build nanoscale electronics or robots.

“There are all kinds of things you can do with this,” Boyden says. “Democratizing nanofabrication could open up frontiers we can’t yet imagine.”

Many research labs are already stocked with the equipment required for this kind of fabrication. “With a laser you can already find in many biology labs, you can scan a pattern, then deposit metals, semiconductors, or DNA, and then shrink it down,” Boyden says.

ClassOne Technology, global supplier of wet processing equipment for ≤200mm semiconductor manufacturing, announced the sale of three more tools to the Ferdinand-Braun-Institut, Leibniz-Institut für Höchstfrequenztechnik (FBH) in Berlin, Germany. In recent weeks, FBH ordered ClassOne’s high-performance 8-chamber Solstice® electroplating system. Now FBH is ordering three additional wet processing tools from ClassOne, including a refurbished Semitool®Spray Solvent Tool (SST), a Semitool Spray Acid Tool (SAT) and a Trident Spin Rinse Dryer (SRD). FBH is one of the world’s leading research institutes and a producer of III-V compound semiconductors, known for prototyping advanced microwave and optoelectronic devices for communications, energy, health, mobility, and more.

“We’ve put our trust in ClassOne for our entire wet processing line,” said Olaf Krüger, Head of FBH’s Process Technology Department. “They understand the special requirements of compound semiconductor manufacturers like us. ClassOne provides state-of-the-art automated processing tools for 100mm and smaller wafers that allow for high process reproducibility during R&D of novel compound semiconductor devices and ensure compatibility to industrial standards. Plus, they’ve put together a support operation right here in Europe to provide us with everything we need.”

“We see these follow-on orders from FBH as a real vote of confidence, and one that we value highly,” said ClassOne’s CEO Byron Exarcos. “Europe is an important market for us. Which is why we’ve invested heavily to build a world-class customer support structure here, including a strong, experienced process engineering team and a seasoned field service and support group that’s able to cover everything our customers might need, up to and including an extensive inventory of spare parts. So, it’s extremely gratifying to see our European customers really recognizing and making use of our local capabilities.”

“The investment is also paying off in terms of the market share gains we’re seeing,” said Roland Seitz, Director of ClassOne’s European Operations. “European customers continue to tell me ClassOne has moved into leading-supplier position for plating and wet processing equipment for 200mm wafers and below. It’s been the result of high-performance equipment combined with strong customer support and affordable prices. Going forward, the European sales of Solstice plating systems and associated tools are growing at a very rapid pace.”

At Intel “Architecture Day,” top executives, architects and fellows revealed next-generation technologies and discussed progress on a strategy to power an expanding universe of data-intensive workloads for PCs and other smart consumer devices, high-speed networks, ubiquitous artificial intelligence (AI), specialized cloud data centers and autonomous vehicles.

Intel demonstrated a range of 10nm-based systems in development for PCs, data centers and networking, and previewed other technologies targeted at an expanded range of workloads.

The company also shared its technical strategy focused on six engineering segments where significant investments and innovation are being pursued to drive leaps forward in technology and user experience. They include: advanced manufacturing processes and packaging; new architectures to speed-up specialized tasks like AI and graphics; super-fast memory; interconnects; embedded security features; and common software to unify and simplify programming for developers across Intel’s compute roadmap.

Together these technologies lay the foundation for a more diverse era of computing in an expanded addressable market opportunity of more than $300 billion by 2022.

Intel Architecture Day Highlights:

Industry-First 3D Stacking of Logic Chips: Intel demonstrated a new 3D packaging technology, called “Foveros,” which for the first time brings the benefits of 3D stacking to enable logic-on-logic integration.

Foveros paves the way for devices and systems combining high-performance, high-density and low-power silicon process technologies. Foveros is expected to extend die stacking beyond traditional passive interposers and stacked memory to high-performance logic, such as CPU, graphics and AI processors for the first time.

The technology provides tremendous flexibility as designers seek to “mix and match” technology IP blocks with various memory and I/O elements in new device form factors. It will allow products to be broken up into smaller “chiplets,” where I/O, SRAM and power delivery circuits can be fabricated in a base die and high-performance logic chiplets are stacked on top.

Intel expects to launch a range of products using Foveros beginning in the second half of 2019. The first Foveros product will combine a high-performance 10nm compute-stacked chiplet with a low-power 22FFL base die. It will enable the combination of world-class performance and power efficiency in a small form factor.

Foveros is the next leap forward following Intel’s breakthrough Embedded Multi-die Interconnect Bridge (EMIB) 2D packaging technology, introduced in 2018.

New Sunny Cove CPU Architecture: Intel introduced Sunny Cove, Intel’s next-generation CPU microarchitecture designed to increase performance per clock and power efficiency for general purpose computing tasks, and includes new features to accelerate special purpose computing tasks like AI and cryptography. Sunny Cove will be the basis for Intel’s next-generation server (Intel® Xeon®) and client (Intel® Core™) processors later next year.

Sunny Cove enables reduced latency and high throughput, as well as offers much greater parallelism that is expected to improve experiences from gaming to media to data-centric applications.

Next-Generation Graphics: Intel unveiled new Gen11 integrated graphics with 64 enhanced execution units, more than double previous Intel Gen9 graphics (24 EUs), designed to break the 1 TFLOPS barrier. The new integrated graphics will be delivered in 10nm-based processors beginning in 2019.

The new integrated graphics architecture is expected to double the computing performance-per-clock compared to Intel Gen9 graphics. With >1 TFLOPS performance capability, this architecture is designed to increase game playability. At the event, Intel showed Gen11 graphics nearly doubling the performance of a popular photo recognition application when compared to Intel’s Gen9 graphics. Gen11 graphics is expected to also feature an advanced media encoder and decoder, supporting 4K video streams and 8K content creation in constrained power envelopes. Gen11 will also feature Intel® Adaptive Sync technology enabling smooth frame rates for gaming.

Intel also reaffirmed its plan to introduce a discrete graphics processor by 2020.

“One API” Software: Intel announced the “One API” project to simplify the programming of diverse computing engines across CPU, GPU, FPGA, AI and other accelerators. The project includes a comprehensive and unified portfolio of developer tools for mapping software to the hardware that can best accelerate the code. A public project release is expected to be available in 2019.

Memory and Storage: Intel discussed updates on Intel® Optane™ technology and the products based upon that technology. Intel® Optane™ DC persistent memory is a new product that converges memory-like performance with the data persistence and large capacity of storage. The revolutionary technology brings more data closer to the CPU for faster processing of bigger data sets like those used in AI and large databases. Its large capacity and data persistence reduces the need to make time-consuming trips to storage, which can improve workload performance. Intel Optane DC persistent memory delivers cache line (64B) reads to the CPU. On average, the average idle read latency with Optane persistent memory is expected to be about 350 nanoseconds when applications direct the read operation to Optane persistent memory, or when the requested data is not cached in DRAM. For scale, an Optane DC SSD has an average idle read latency of about 10,000 nanoseconds (10 microseconds), a remarkable improvement.2 In cases where requested data is in DRAM, either cached by the CPU’s memory controller or directed by the application, memory sub-system responsiveness is expected to be identical to DRAM (<100 nanoseconds).

The company also showed how SSDs based on Intel’s 1 Terabit QLC NAND die move more bulk data from HDDs to SSDs, allowing faster access to that data.

The combination of Intel Optane SSDs with QLC NAND SSDs will enable lower latency access to data used most frequently. Taken together, these platform and memory advances complete the memory and storage hierarchy providing the right set of choices for systems and applications.

Deep Learning Reference Stack: Intel is releasing the Deep Learning Reference Stack, an integrated, highly-performant open source stack optimized for Intel® Xeon® Scalable platforms. This open source community release is part of our effort to ensure AI developers have easy access to all of the features and functionality of the Intel platforms. The Deep Learning Reference Stack is highly-tuned and built for cloud native environments. With this release, Intel is enabling developers to quickly prototype by reducing the complexity associated with integrating multiple software components, while still giving users the flexibility to customize their solutions.

A new collaborative study led by a research team at the Department of Energy’s Pacific Northwest National Laboratory and University of California, Los Angeles could provide engineers new design rules for creating microelectronics, membranes, and tissues, and open up better production methods for new materials. At the same time, the research, published in the journal Science, helps uphold a scientific theory that has remained unproven for over a century.

Just as children follow a rule to line up single file after recess, some materials use an underlying rule to assemble on surfaces one row at a time, according to the study done at PNNL, the University of Washington, UCLA, and elsewhere.

Nucleation — that first formation step — is pervasive in ordered structures across nature and technology, from cloud droplets to rock candy. Yet despite some predictions made in the 1870s by the American scientist J. Willard Gibbs, researchers are still debating how this basic process happens.

The new study verifies Gibbs’ theory for materials that form row by row. Led by UW graduate student Jiajun Chen, working at PNNL, the research uncovers the underlying mechanism, which fills in a fundamental knowledge gap and opens new pathways in materials science.

Chen used small protein fragments called peptides that show specificity, or unique belonging, to a material surface. The UCLA collaborators have been identifying and using such material-specific peptides as control agents to force nanomaterials to grow into certain shapes, such as those desired in catalytic reactions or semiconductor devices. The research team made the discovery while investigating how a particular peptide — one with a strong binding affinity for molybdenum disulfide — interacts with the material.

“It was complete serendipity,” said PNNL materials scientist James De Yoreo, co-corresponding author of the paper and Chen’s doctoral advisor. “We didn’t expect the peptides to assemble into their own highly ordered structures.”

That possibly happened because “this peptide was identified from a molecular evolution process,” adds co-corresponding author Yu Huang, a materials scientist at UCLA. “It appears nature does find its way to minimize energy consumption and to work wonders.”

Row by row

The transformation of liquid water into solid ice requires the creation of a solid-liquid interface. According to Gibbs’ classical nucleation theory, although turning the water into ice saves energy, creating the interface costs energy. The tricky part is the initial start — that’s when the surface area of the new particle of ice is large compared to its volume, so it costs more energy to make an ice particle than is saved.

Gibbs’ theory predicts that if the materials can grow in one dimension, meaning row by row, no such energy penalty would exist. Then the materials can avoid what scientists call the nucleation barrier and are free to self-assemble.

There has been recent controversy over the theory of nucleation. Some researchers have found evidence that the fundamental process is actually more complex than that proposed in Gibbs’ model.

But “this study shows there are certainly cases where Gibbs’ theory works well,” De Yoreo said.

Previous studies had already shown that some organic molecules, including peptides like the ones in the Science paper, can self-assemble on surfaces. But at PNNL, De Yoreo and his team dug deeper and found a way to understand how molecular interactions with materials impact their nucleation and growth.

They exposed the peptide solution to fresh surfaces of a molybdenum disulfide substrate, measuring the interactions with atomic force microscopy. Then they compared the measurements with molecular dynamics simulations.

De Yoreo and his team determined that even in the earliest stages, the peptides bound to the material one row at a time, barrier-free, just as Gibbs’ theory predicts.

The atomic force microscopy high imaging speed allowed the researchers to see the rows just as they were forming. The results showed the rows were ordered right from the start and grew at the same speed regardless of their size — a key piece of evidence. They also formed new rows as soon as enough peptide was in the solution for existing rows to grow; that would only happen if row formation is barrier-free.

Better control

This row by row process provides clues for the design of 2D materials. Currently, to form certain shapes, designers sometimes need to put systems far out of equilibrium, or balance. That is difficult to control, said De Yoreo.

“But in 1D, the difficulty of getting things to form in an ordered structure goes away,” he added. “Then you can operate right near equilibrium and still grow these structures without losing control of the system.”

It could change assembly pathways for those engineering microelectronics or even bodily tissues.

Huang’s team at UCLA has demonstrated new opportunities for devices based on 2D materials assembled through interactions in solution. But she said the current manual processes used to construct such materials have limitations, including scale-up capabilities.

“Now with the new understanding,” said Huang, “we can start to exploit the specific interactions between molecules and 2D materials for automatous assembly processes.”

The next step, said De Yoreo, is to make artificial molecules that have the same properties as the peptides studied in the new paper — only more robust.

At PNNL, he and his team are looking at stable peptoids, which are as easy to synthesize as peptides but can better handle the temperatures and chemicals used in the processes to construct the desired materials.