Tag Archives: letter-pulse-tech

Brewer Science, Inc. today from SEMICON Taiwan introduced the latest additions to its industry-leading BrewerBOND® family of temporary bonding materials, as well as the first product in its new BrewerBUILD™ line of thin spin-on packaging materials. BrewerBUILD delivers an industry-first solution to address manufacturers’ evolving wafer-level packaging challenges.

The BrewerBOND T1100 and BrewerBOND C1300 series combine to create Brewer Science’s first complete, dual-layer system for temporary bonding and debonding of product wafers. The new system was developed for power, memory and chip-first fan-out devices – all of which have stringent requirements with respect to temperature, power and performance. The system can be used with either mechanical or laser debonding methods.

The BrewerBUILD material was specifically created for redistribution-layer (RDL)-first fan-out wafer-level packaging (FOWLP). Developed to meet the needs of chipmakers looking to transition from chip-first FOWLP but not yet ready to tackle 2.5D/3D packaging, the single-layer material is compatible with both wafer- and panel-level temporary bonding/debonding processes.

“As industry requirements advance, Brewer Science continues to push forward the state of the art in our materials offerings,” said Kim Arnold, executive director, Advanced Packaging Business Unit, Brewer Science Inc. “Through close collaboration with our customers, we are driving the technology forward, leveraging our R&D braintrust to create unique solutions like these that are designed to meet customers’ needs – current and future.”

Today KLA-Tencor Corporation (NASDAQ : KLAC ) announced two new defect inspection products designed to address a wide variety of integrated circuit (IC) packaging challenges. The Kronos™ 1080 system offers production-worthy, high sensitivity wafer inspection for advanced packaging, providing key information for process control and material disposition. The ICOS™ F160 system examines packages after wafers have been diced, delivering fast, accurate die sort based on detection of key defect types—including sidewall cracks, a new defect type affecting the yield of high-end packages. The two new inspection systems join KLA-Tencor’s portfolio of defect inspection, metrology and data analysis systems that help accelerate packaging yield and increase die sort accuracy.

KLA-Tencor’s new Kronos™ 1080 wafer inspection system and ICOS™ F160 die sorting and inspection system are designed to address a wide variety of IC packaging challenges.

“As chip scaling has slowed, advances in chip packaging technology have become instrumental in driving device performance,” said Oreste Donzella, Senior Vice President and Chief Marketing Officer at KLA-Tencor. “Packaged chips need to achieve simultaneous targets for device performance, power consumption, form factor and cost for a variety of device applications. As a result, packaging design has become more diverse and complex, featuring a range of 2D and 3D structures that are more densely packed and shrinking in size with every generation. At the same time, the value of the packaged chip has grown substantially, along with electronics manufacturers’ expectations for quality and reliability. To meet these expectations, packaging manufacturers, whether in the back end of a chip manufacturing fab or in an outsourced assembly and test (OSAT) facility, have demanded more sensitive, cost-effective inspection, metrology and data analysis—and more accurate identification of bad parts. Our engineering teams have developed the new Kronos 1080 and ICOS F160 systems to serve the electronics industry’s growing needs for production-worthy defect detection for a wide variety of packaging types.”

The Kronos 1080 system is designed to inspect advanced wafer-level packaging process steps, providing information on the full range of defect types for inline process control. Advanced packaging technology necessarily includes ever-smaller features, higher-density metal patterns, and multi-layer redistribution layers—all of which have increasing inspection requirements that demand innovative solutions. The Kronos system achieves its industry-leading performance through multi-mode optics and sensors and advanced defect detection algorithms. The Kronos system also introduces FlexPoint™, an advanced technology derived from KLA-Tencor’s leading inspection solutions for IC chip manufacturing. FlexPoint focuses the inspection system on key areas within the die where defects would have highest impact. Flexible wafer handling enables the inspection of high-warp wafers, frequently encountered in a package type called fan-out wafer-level packaging—an established type for mobile applications and an emerging technology for networking and high-performance computing.

After wafer-level packages are tested and diced, the ICOS F160 performs inspection and die sorting. Manufacturers of high-end packages, such as those used for mobile applications, will benefit from new capability to detect laser-groove, hairline and sidewall cracks. These cracks result from a change in the materials used to insulate the dense on-chip metal routing to facilitate increased speed and reduced power consumption. The new material is brittle, making it susceptible to cracks during wafer dicing. Sidewall cracks are notoriously difficult to detect, as they lie perpendicular to the top of the die and are not detectable using traditional visual inspection. Another major advantage of the ICOS F160 system, beneficial to many packaging types, is its flexibility: input and output modes can be wafer, tray or tape. The system is easily changed from one configuration to another. Its automatic calibrations and precision die pickup facilitate increased tool utilization in high volume manufacturing environments.

The Kronos 1080 and ICOS F160 systems are part of KLA-Tencor’s portfolio of packaging solutions designed to address inspection, metrology, data analysis and die sorting needs for a variety of IC packaging types.

Scientists at the Department of Energy’s Oak Ridge National Laboratory induced a two-dimensional material to cannibalize itself for atomic “building blocks” from which stable structures formed.

The findings, reported in Nature Communications, provide insights that may improve design of 2D materials for fast-charging energy-storage and electronic devices.

“Under our experimental conditions, titanium and carbon atoms can spontaneously form an atomically thin layer of 2D transition-metal carbide, which was never observed before,” said Xiahan Sang of ORNL.

He and ORNL’s Raymond Unocic led a team that performed in situ experiments using state-of-the-art scanning transmission electron microscopy (STEM), combined with theory-based simulations, to reveal the mechanism’s atomistic details.

“This study is about determining the atomic-level mechanisms and kinetics that are responsible for forming new structures of a 2D transition-metal carbide such that new synthesis methods can be realized for this class of materials,” Unocic added.

The starting material was a 2D ceramic called a MXene (pronounced “max een”). Unlike most ceramics, MXenes are good electrical conductors because they are made from alternating atomic layers of carbon or nitrogen sandwiched within transition metals like titanium.

The research was a project of the Fluid Interface Reactions, Structures and Transport (FIRST) Center, a DOE Energy Frontier Research Center that explores fluid-solid interface reactions that have consequences for energy transport in everyday applications. Scientists conducted experiments to synthesize and characterize advanced materials and performed theory and simulation work to explain observed structural and functional properties of the materials. New knowledge from FIRST projects provides guideposts for future studies.

The high-quality material used in these experiments was synthesized by Drexel University scientists, in the form of five-ply single-crystal monolayer flakes of MXene. The flakes were taken from a parent crystal called “MAX,” which contains a transition metal denoted by “M”; an element such as aluminum or silicon, denoted by “A”; and either a carbon or nitrogen atom, denoted by “X.” The researchers used an acidic solution to etch out the monoatomic aluminum layers, exfoliate the material and delaminate it into individual monolayers of a titanium carbide MXene (Ti3C2).

The ORNL scientists suspended a large MXene flake on a heating chip with holes drilled in it so no support material, or substrate, interfered with the flake. Under vacuum, the suspended flake was exposed to heat and irradiated with an electron beam to clean the MXene surface and fully expose the layer of titanium atoms.

MXenes are typically inert because their surfaces are covered with protective functional groups–oxygen, hydrogen and fluorine atoms that remain after acid exfoliation. After protective groups are removed, the remaining material activates. Atomic-scale defects–“vacancies” created when titanium atoms are removed during etching–are exposed on the outer ply of the monolayer. “These atomic vacancies are good initiation sites,” Sang said. “It’s favorable for titanium and carbon atoms to move from defective sites to the surface.” In an area with a defect, a pore may form when atoms migrate.

“Once those functional groups are gone, now you’re left with a bare titanium layer (and underneath, alternating carbon, titanium, carbon, titanium) that’s free to reconstruct and form new structures on top of existing structures,” Sang said.

High-resolution STEM imaging proved that atoms moved from one part of the material to another to build structures. Because the material feeds on itself, the growth mechanism is cannibalistic.

“The growth mechanism is completely supported by density functional theory and reactive molecular dynamics simulations, thus opening up future possibilities to use these theory tools to determine the experimental parameters required for synthesizing specific defect structures,” said Adri van Duin of Penn State.

Most of the time, only one additional layer [of carbon and titanium] grew on a surface. The material changed as atoms built new layers. Ti3C2 turned into Ti4C3, for example.

“These materials are efficient at ionic transport, which lends itself well to battery and supercapacitor applications,” Unocic said. “How does ionic transport change when we add more layers to nanometer-thin MXene sheets?” This question may spur future studies.

“Because MXenes containing molybdenum, niobium, vanadium, tantalum, hafnium, chromium and other metals are available, there are opportunities to make a variety of new structures containing more than three or four metal atoms in cross-section (the current limit for MXenes produced from MAX phases),” Yury Gogotsi of Drexel University added. “Those materials may show different useful properties and create an array of 2D building blocks for advancing technology.”

At ORNL’s Center for Nanophase Materials Sciences (CNMS), Yu Xie, Weiwei Sun and Paul Kent performed first-principles theory calculations to explain why these materials grew layer by layer instead of forming alternate structures, such as squares. Xufan Li and Kai Xiao helped understand the growth mechanism, which minimizes surface energy to stabilize atomic configurations. Penn State scientists conducted large-scale dynamical reactive force field simulations showing how atoms rearranged on surfaces, confirming defect structures and their evolution as observed in experiments.

The researchers hope the new knowledge will help others grow advanced materials and generate useful nanoscale structures.

They are among the thinnest structures on earth: “two dimensional materials” are crystals which consist of only one or a few layers of atoms. They often display unusual properties, promising many new applications in opto-electronics and energy technology. One of these materials is 2D-molybdenum sulphide, an atomically thin layer of molybdenum and sulphur atoms.

The production of such ultra-thin crystals is difficult. The crystallisation process depends on many different factors. In the past, different techniques have yielded quite diverse results, but the reasons for this could not be accurately explained. Thanks to a new method developed by research teams at TU Wien, the University of Vienna and Joanneum Research in Styria, for the first time ever it is now possible to observe the crystallisation process directly under the electron microscope. The method has now been presented in the scientific journal ACS Nano.

At first, the atoms are randomly distributed, after being manipulated with the electron beam, they form crystal structures (right). Credit: TU Wien

From gas to crystal

“Molybdenum sulphide can be used in transparent and flexible solar cells or for sustainably generating hydrogen for energy storage”, says the lead author of the study, Bernhard C. Bayer from the Institute of Materials Chemistry at TU Wien. “In order to do this, however, high-quality crystals must be grown under controlled conditions.”

Usually this is done by starting out with atoms in gaseous form and then condensing them on a surface in a random and unstructured way. In a second step, the atoms are arranged in regular crystal form – through heating, for example. “The diverse chemical reactions during the crystallisation process are, however, still unclear, which makes it very difficult to develop better production methods for 2D materials of this kind”, Bayer states.

Thanks to a new method, however, it should now be possible to accurately study the details of the crystallisation process. “This means it is no longer necessary to experiment through trial and error, but thanks to a deeper understanding of the processes, we can say for certain how to obtain the desired product”, Bayer adds.

Graphene as a substrate

First, molybdenum and sulphur are placed randomly on a membrane made of graphene. Graphene is probably the best known of the 2D materials – a crystal with a thickness of only one atom layer consisting of carbon atoms arranged in a honeycomb lattice. The randomly arranged molybdenum and sulphur atoms are then manipulated in the electron microscope with a fine electron beam. The same electron beam can be used simultaneously to image the process and to initiate the crystallization process.

That way it has now become possible for the first time to directly observe how the atoms move and rearrange during the growth of the material with a thickness of only two atomic layers. “In doing so, we can see that the most thermodynamically stable configuration doesn’t necessarily always have to be the final state”, Bayer says. Different crystal arrangements compete with one another, transform into each other and replace one another. “Therefore, it is now clear why earlier investigations had such varying results. We are dealing with a complex, dynamic process.” The new findings will help to adapt the structure of the 2D materials more precisely to application requirements in future by interfering with the rearrangement processes in a targeted manner.

Most current displays do not always accurately represent the world’s colors as we perceive them by eye, instead only representing roughly 70% of them. To make better displays with true colors commonly available, researchers have focused their efforts on light-emitting nanoparticles. Such nanoparticles can also be used in medical research to light up and keep track of drugs when developing and testing new medicines in the body. However, the metal these light-emitting nanoparticles are based on, namely cadmium, is highly toxic, which limits its applications in medical research and in consumer products–many countries may soon introduce bans on toxic nanoparticles.

These are structures of silver indium sulfide/gallium sulfide core/shell quantum dots and pictures of the core/shell quantum dots under room light. Credit: Osaka University

It is therefore vital to create non-toxic versions of these nanoparticles that have similar properties: they must produce very clean colors and must do so in a very energy-efficient way. So far researchers have succeeded in creating non-toxic nanoparticles that emit light in an efficient manner by creating semiconductors with three types of elements in them, for example, silver, indium, and sulfur (in the form of silver indium disulfide (AgInS2)). However, the colors they emit are not pure enough–and many researchers declared that it would be impossible for such nanoparticles to ever emit pure colors.

Now, researchers from Osaka University have proven that it is possible by fabricating semiconductor nanoparticles containing silver indium disulfide and adding a shell around them consisting of a semiconductor material made of two different elements, gallium and sulfur. The team was able to reproducibly create these shell-covered nanoparticles that are both energy efficient and emit vivid, clean colors. The team have recently published their research in the Nature journal NPG Asia Materials.

“We synthesized non-toxic nanoparticles in the normal way: mix all ingredients together and heat them up. The results were not fantastic, but by tweaking the synthesis conditions and modifying the nanoparticle cores and the shells we enclosed them in, we were able to achieve fantastic efficiencies and very pure colors,” study coauthor Susumu Kuwabata says.

Enclosing nanoparticles in semiconductor shells in nothing new, but the shells that are currently used have rigidly arranged atoms inside them, whereas the new particles are made of a more chaotic material without such a rigid structure.

“The silver indium disulfide particles emitted purer colors after the coating with gallium sulfide. On top of that, the shell parts in microscopic images were totally amorphous. We think the less rigid nature of the shell material played an important part in that–it was more adaptable and therefore able to take on more energetically favorable conformations,” first author Taro Uematsu says.

The team’s results demonstrate that it is possible to create cadmium-free, non-toxic nanoparticles with very good color-emitting properties by using amorphous shells around the nanoparticle cores.

Rudolph Technologies, Inc. (NYSE: RTEC) today announced its new Dragonfly™ G2 platform, which incorporates many of the benefits of the Firefly™ system onto the Dragonfly platform, including higher sensitivity and throughput and the proprietary Clearfind™ Technology. The new system increases the options for advanced packaging customers to meet their wafer-based application challenges on a single platform. To date, customer evaluations have reported throughput increases greater than 50 percent over the first-generation Dragonfly system. The new Dragonfly G2 systems are scheduled to begin shipment in the latter part of the fourth quarter and will be highlighted at the SEMICON® Taiwan trade show September 5-7 in Rudolph’s booth N686-4F.

The Dragonfly G2 system achieves significant throughput and productivity increases using proprietary camera technology combined with stage speed and accuracy. Additionally, its modular architecture permits plug-and-play configurability of Rudolph’s technologies such as Truebump™ Technology, for more accurate bump height measurement, and Clearfind Technology, for non-visual residue detection. Streamlined software algorithms contribute to the faster throughput and enable the system to handle increasing bump counts, which have already exceeded 80 million bumps per wafer.

“Advanced packaging processes are evolving rapidly, with larger packages, shrinking features, and higher counts of smaller bumps on every wafer, and the Dragonfly G2 system is designed to meet these new challenges,” said Tim Kryman, senior director of corporate marketing at Rudolph Technologies. “At the same time, its increased throughput reduces cost-of-ownership and its configurable modular design lets one system do the work of two. Based on the positive feedback from customers’ beta testing we are expecting strong demand for this latest evolution of our technology. We expect the Dragonfly G2 system to meet our customers’ future inspection needs as increasing demands for higher quality products are driving more data with greater integrity and faster throughput to meet the growing volumes of consumer and auto electronics products.”

“An important driver for Rudolph Technologies is to increase our pace of innovation to ensure we are anticipating our customers’ roadmaps,” added Mike Goodrich, vice president and general manager of Rudolph’s Process Control Group. “We were very pleased to be able to demonstrate that commitment with the release of this Dragonfly G2 system. Not only have we significantly improved throughput and imaging capability, but we have also provided the powerful Clearfind Technology to make a compelling, no compromise, advanced packaging process control system.”

The Dragonfly G2 system can be ordered now with shipments expected to begin in Q4. First-generation Dragonfly systems can be retrofitted on-site with a second-generation upgrade kit.

Scientists at Tokyo Institute of Technology designed a new type of molecular wire doped with organometallic ruthenium to achieve unprecedentedly higher conductance than earlier molecular wires. The origin of high conductance in these wires is fundamentally different from similar molecular devices and suggests a potential strategy for developing highly conducting “doped” molecular wires.

The proposed wire is ‘doped’ with a ruthenium unit that enhances its conductance to unprecedented levels compared with previously reported similar molecular wires. Credit: Journal of the American Chemical Society

Since their conception, researchers have tried to shrink electronic devices to unprecedented sizes, even to the point of fabricating them from a few molecules. Molecular wires are one of the building blocks of such minuscule contraptions, and many researchers have been developing strategies to synthesize highly conductive, stable wires from carefully designed molecules.

A team of researchers from Tokyo Institute of Technology, including Yuya Tanaka, designed a novel molecular wire in the form of a metal electrode-molecule-metal electrode (MMM) junction including a polyyne, an organic chain-like molecule, “doped” with a ruthenium-based unit Ru(dppe)2. The proposed design, featured in the cover of the Journal of the American Chemical Society, is based on engineering the energy levels of the conducting orbitals of the atoms of the wire, considering the characteristics of gold electrodes.

Using scanning tunneling microscopy, the team confirmed that the conductance of these molecular wires was equal to or higher than those of previously reported organic molecular wires, including similar wires “doped” with iron units. Motivated by these results, the researchers then went on to investigate the origin of the proposed wire’s superior conductance. They found that the observed conducting properties were fundamentally different from previously reported similar MMM junctions and were derived from orbital splitting. In other words, orbital splitting induces changes in the original electron orbitals of the atoms to define a new “hybrid” orbital facilitating electron transfer between the metal electrodes and the wire molecules. According to Tanaka, “such orbital splitting behavior has rarely been reported for any other MMM junction”.

Since a narrow gap between the highest (HOMO) and lowest (LUMO) occupied molecular orbitals is a crucial factor for enhancing conductance of molecular wires, the proposed synthesis protocol adopts a new technique to exploit this knowledge, as Tanaka adds “The present study reveals a new strategy to realize molecular wires with an extremely narrow HOMO?LUMO gap via MMM junction formation.”

This explanation for the fundamentally different conducting properties of the proposed wires facilitate the strategic development of novel molecular components, which could be the building blocks of future minuscule electronic devices.

Cymer, a manufacturer of excimer lasers used in semiconductor manufacturing, today announced the first customer installation of its XLR 860ix light source, which is expected to be used in the production of chips at advanced logic and memory nodes.

The XLR 860ix is a deep-ultraviolet (DUV) light source based on an Argon Fluoride excimer laser. The first customer installation was completed this month, and the XLR 860ix was paired up with ASML’s latest lithography system, the TWINSCAN NXT:2000i, for which the source was qualified earlier this year.

“The XLR 860ix, through improvements in high-speed controls and redesigned, on-board bandwidth metrology, reduces variations in bandwidth by a factor of two compared to its predecessor. This is an important achievement, since these variations contribute to errors in critical dimension (CD) uniformity, which in turn affects image quality and ultimately manufacturing yields. The improvement in the spectral stability of the light has been verified by our customers using early-access versions of the XLR 860ix, which gives us confidence that this light source will help to improve CD uniformity when used in the production of advanced ICs,” said Cymer Vice President Product Marketing Patrick O’Keeffe.

Ahead of the full release of the XLR860ix, Cymer made the key technologies available to customers in an early access program by upgrading existing light sources. Four such upgrades have been completed, and seven additional sources are planned to be upgraded by the end of the year. A total of seven customers are participating in the early access program. In response to strong customer demand, Cymer has rapidly shifted its production capacity to the XLR 860ix model for all future ArF immersion shipments.

“In addition to the early access program, we have also offered one of the key new technologies of the XLR 860ix as an upgrade to previous light source models. This upgrade, which extends the lifetime of the optics and the chamber, increases the time between service intervals by 33%, and thus allows our customers to better utilize their lithography systems and expose thousands of additional wafers per tool, per year. Our customers have aggressively taken advantage of this upgrade, with more than 400 upgrades completed within the past year. The majority of the upgraded systems are exceeding the targeted service intervals,” O’Keeffe said.

GOWIN Semiconductor Corp., an innovator of programmable logic devices, announces their development of RISC-V Microprocessor IP implemented in their current ARORA® Family GW-2A FPGA products.  In addition, GOWIN launches an Industry Early Adopter Program to kickstart engineering design activity.  The Industry EAP includes: an FPGA programming bit-file reference design with RISC-V Microprocessor core, AHB & APB Bus, Memory Control & Peripherals, as well as the GW-2A development board for a complete, ready to use solution.

RISC-V is a free and open ISA enabling a new era of processor innovation through open standard collaboration. The RISC-V ISA has been designed with small, fast, and low-power real-world implementations in mind without over-architecting for a particular microarchitecture style. The instruction set also has a substantial body of supporting software for a comprehensive design ecosystem.

“GOWIN’s FPGA solutions showcase the growing adoption of RISC-V around the world. It’s exciting to see how the RISC-V ecosystem is maturing as more companies design innovative implementations based on the free and open ISA,” said Rick O’Connor, executive director of the non-profit RISC-V Foundation.

The GOWIN Arora® Family GW-2A FPGAs offer best-in-class performance at an effective cost. With abundant logic, high performance DSP resources, and high speed I/O, the family is optimized for co-processing of computation tasks while hosting the RISC-V Microprocessor soft core. The Arora® family is also the first FPGA with embedded DRAM in the industry, allowing customers to design without using up I/O for external memory.

GOWIN also announces today the appointment of Edge Electronics as their US National Distributor and EBBM, Inc as their East Coast Manufacturers Representative.  “Demand is high for our innovative FPGA products serving the low to mid density logic element markets,” said Scott Casper, Director of Sales for GOWIN’s Americas Region, “The need for the right channel partners is necessary for our growth.  We are excited to be working with Edge and EBBM as we continue our Americas expansion plan.”

“GOWIN Semiconductor is a natural fit alongside our semiconductor and LCD solutions product offerings, both of which are geared toward serving North American industrial, medical and automotive OEM markets among others,” says Michael Pollina, Edge Electronics’ VP Operations & Procurement. “GOWIN’s collection of development tools in tandem with Edge’s engineering team will make it simple for customers to transition existing designs or start new projects with low power, space-efficient and cost-effective FPGA solutions.”

“We are thrilled to be working with GOWIN, one of the Silicon60 Most Remarkable Global Technology Startups,” said Chief Executive Officer of EBBM, Inc., Alex Angelou. “EBBM, Inc has been helping custom logic architects quickly articulate their design and beat their competition to market for 14 years. GOWIN’s One-Stop Complete Solution, including DVK, EDA, IP, is a perfect match to bring the competitive advantage to more companies,” said EBBM, Inc. Chief Strategy Officer Ken Cheo.

Semiconducting heterostructures have been key to the development of electronics and opto-electronics. Many applications in the infrared and terahertz frequency range exploit transitions, called intersubband transitions, between quantized states in semiconductor quantum wells. These intraband transitions exhibit very large oscillator strengths, close to unity. Their discovery in III-V semiconductor heterostructures depicted a huge impact within the condensed matter physics community and triggered the development of quantum well infrared photodetectors as well as quantum cascade lasers.

Schematic illustration of charge carriers confined within a TMD flake comprising different thicknesses. Charge carriers in the ground state (blue) can be excited upon resonant light excitation to a higher state (pink). Credit: ICFO/Fabien Vialla

Quantum wells of the highest quality are typically fabricated by molecular beam epitaxy (sequential growth of crystalline layers), which is a well-established technique. However, it poses two major limitations: Lattice-matching is required, restricting the freedom in materials to choose from, and the thermal growth causes atomic diffusion and increases interface roughness.

2D materials can overcome these limitations since they naturally form a quantum well with atomically sharp interfaces. They provide defect free and atomically sharp interfaces, enabling the formation of ideal QWs, free of diffusive inhomogeneities. They do not require epitaxial growth on a matching substrate and can therefore be easily isolated and coupled to other electronic systems such as Si CMOS or optical systems such as cavities and waveguides.

Surprisingly enough, intersubband transitions in few-layer 2D materials had never been studied before, neither experimentally nor theoretically. Thus, in a recent study published in Nature Nanotechnology, ICFO researchers Peter Schmidt, Fabien Vialla, Mathieu Massicotte, Klaas-Jan Tielrooij, Gabriele Navickaite, led by ICREA Prof at ICFO Frank Koppens, in collaboration with the Institut Lumière Matière – CNRS, Technical University of Denmark, Max Planck Institute for the Structure and Dynamics of Matter, CIC nanoGUNE, and the National Graphene Institute, report on the first theoretical calculations and first experimental observation of inter-sub-band transitions in quantum wells of few-layer semiconducting 2D materials (TMDs).

In their experiment, the team of researchers applied scattering scanning near-field optical microscopy (s-SNOM) as an innovative approach for spectral absorption measurements with a spatial resolution below 20 nm. They exfoliated TMDs, which comprisedterraces of different layer thicknesses over lateral sizes of about a few micrometers. They directly observed the intersubband resonances for these different quantum well thicknesses within a single device. They also electrostatically tuned the charge carrier density and demonstrated intersubband absorption in both the valence and conduction band. These observations were complemented and supported with detailed theoretical calculations revealing many-body and non-local effects.

The results of this study pave the way towards an unexplored field in this new class of materials and offer a first glimpse of the physics and technology enabled by intersubband transitions in 2D materials, such as infrared detectors, sources, and lasers with the potential for compact integration with Si CMOS.