Entity

Time filter

Source Type

Tunis, Tunisia

Yaghoobi A.,Electronics Laboratory | Rezazadegan-Tavakoli H.,Center for Machine Vision Research | Roning J.,University of Oulu
Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) | Year: 2015

This paper articulates the concept of affordances use as the building block of an automated video surveillance system which learns and evolves over time. It grounds its arguments on the basis of a visual attention hardware and affordances. © Springer International Publishing Switzerland 2015. Source


News Article
Site: http://news.mit.edu/topic/mitnanotech-rss.xml

A family of compounds known as perovskites, which can be made into thin films with many promising electronic and optical properties, has been a hot research topic in recent years. But although these materials could potentially be highly useful in applications such as solar cells, some limitations still hamper their efficiency and consistency. Now, a team of researchers at MIT and elsewhere say they have made significant inroads toward understanding a process for improving perovskites’ performance, by modifying the material using intense light. The new findings are being reported in the journal Nature Communications, in a paper by Samuel Stranks, a researcher at MIT; Vladimir Bulović, the Fariborz Maseeh (1990) Professor of Emerging Technology and associate dean for innovation; and eight colleagues at other institutions in the U.S. and the U.K. The work is part of a major research effort on perovskite materials being led by Stranks, within MIT’s Organic and Nanostructured Electronics Laboratory. Tiny defects in perovskite’s crystalline structure can hamper the conversion of light into electricity in a solar cell, but “what we’re finding is that there are some defects that can be healed under light,” says Stranks, who is a Marie Curie Fellow jointly at MIT and Cambridge University in the U.K. The tiny defects, called traps, can cause electrons to recombine with atoms before the electrons can reach a place in the crystal where their motion can be harnessed. Under intense illumination, the researchers found that iodide ions — atoms stripped of an electron so they carry an electric charge — migrated away from the illuminated region, and in the process apparently swept away most of the defects in that region along with them. “This is the first time this has been shown,” Stranks says, “where just under illumination, where no [electric or magnetic] field has been applied, we see this ion migration that helps to clean the film. It reduces the defect density.” While the effect had been observed before, this work is the first to show that the improvement was caused by the ions moving as a result of the illumination. This work is focused on particular types of the material, known as organic-inorganic metal halide perovskites, which are considered promising for applications including solar cells, light-emitting diodes (LEDs), lasers, and light detectors. They excel in a property called the photoluminescence quantum efficiency, which is key to maximizing the efficiency of solar cells. But in practice, the performance of different batches of these materials, or even different spots on the same film, has been highly variable and unpredictable. The new work was aimed at figuring out what caused these discrepancies and how to reduce or eliminate them. Stranks explains that “the ultimate aim is to make defect-free films,” and the resulting improvements in efficiency could also be useful for applications in light emission as well as light capture. Previous work reducing defects in thin-film perovskite materials has focused on electrical or chemical treatments, but “we find we can do the same with light,” Stranks says. One advantage of that is that the same technique used to improve the material’s properties can at the same time be used as a sensitive probe to observe and better understand the behavior of these promising materials. Another advantage of this light-based processing is it doesn’t require anything to come in physical contact with the film being treated — for example, there is no need to attach electrical contacts or to bathe the material in a chemical solution. Instead, the treatment can simply be applied by turning on the source of illumination. The process, which they call photo-induced cleaning, could be “a way forward” for the development of useful perovskite-based devices, Stranks says. The effects of the illumination tend to diminish over time, Stranks says, so “the challenge now is to maintain the effect” long enough to make it practical. Some forms of perovskites are already “looking to be commercialized by next year,” he says, and this research “raises questions that need to be addressed, but it also shows there are ways to address” the phenomena that have been limiting this material’s performance. “I think the paper provides valuable insight that is likely to help people make more efficient solar cells by figuring out how to reduce the number of iodine vacancies,” says Michael McGehee, a professor of materials science and engineering at Stanford University, who was not involved in this research. “I think it is fascinating that illuminating the perovskites improves their photoluminescence efficiency by enabling iodine to move around and eliminate iodine vacancies. ... This research does not make solar cells better, but it does greatly increase our understanding of how these complex materials function in solar cells.” In addition to Stranks and Bulović, the team included Anna Osherov of MIT, Dane deQuilettes, Daniel Graham, and David Ginger of the University of Washington, and Wei Zhang, Victor Burlakov, Tomas Leitjens, and Henry Snaith of Oxford University in the U.K. The work was supported by the European Union Seventh Framework Programme, the U.S. National Science Foundation, the Center for Excitonics, an Energy Frontier Research Center at MIT funded by the U.S. Department of Energy, and the National Institutes of Health.


News Article
Site: http://news.mit.edu/topic/mitmaterials-science-rss.xml

A longstanding problem in designing nanoscale electromechanical switches is the tendency for metal-to-metal contacts to stick together, locking the switch in an “on” position. MIT electrical engineering graduate student Farnaz Niroui has found a way to exploit that tendency to create electrodes with nanometer-thin separations. By designing a cantilever that can collapse and permanently adhere onto a support structure during the fabrication process, Niroui's process leaves a controllable nanoscale gap between the cantilever and electrodes neighboring the point of adhesion. Niroui, who works in Professor Vladimir Bulović’s Organic and Nanostructured Electronics Laboratory (ONE Lab), presented her most recent findings Jan. 20 at the IEEE Micro Electro Mechanical Systems (MEMS) Conference in Portugal. MIT collaborators include professors Jeffrey Lang in electrical engineering and Timothy M. Swager in chemistry. Their paper is titled, “Controlled Fabrication of Nanoscale Gaps Using Stiction.” Stiction, as permanent adhesion is called, is a very important challenge in electromechanical systems and often results in device failure. Niroui turned stiction to her advantage by using a support structure to make nanoscale gaps. "Initially the cantilever is fabricated with a relatively larger gap which is easier to fabricate, but then we modulate the surface adhesion forces to be able to cause a collapse between the cantilever and the support. As the cantilever collapses, this gap reduces to width much smaller than patterned," she explains. "We can get sub-10-nanometer gaps," she says. "It's controllable because by choosing the design of the cantilever, controlling its mechanical properties and the placement of the other electrodes, we can get gaps that are different in size. This is useful not only for our application, which is in tunneling electromechanical switches, but as well for molecular electronics and contact-based electromechanical switches. It’s a general approach to develop nanoscale gaps.” Niroui's latest work builds on her earlier work showing a design for a squeezable switch — or "squitch" — which fills the narrow gap between contacts with an organic molecular layer that can be compressed tightly enough to allow current to tunnel, or flow, from one electrode to another without direct contact — the "on" position — but that will spring back to open a gap wide enough that current cannot flow between electrodes — the "off" position. The softer the filler material is, the less voltage is needed to compress it. The goal is a low-power switch with repeatable abrupt switching behavior that can complement or replace conventional transistors. Niroui designed, fabricated, tested, and characterized the cantilevered switch in which one electrode is fixed and the other moveable with the switching gap filled with a molecular layer. She presented her initial findings at the IEEE MEMS Conference in San Francisco last year in a paper titled, "Nanoelectromechanical Tunneling Switches Based on Self-Assembled Molecular Layers." "We're working right now on alternative designs to achieve an optimized switching performance," Niroui says. "For me, one of the interesting aspects of the project is the fact that devices are designed in very small dimensions," Niroui adds, noting that the tunneling gap between the electrodes is only a few nanometers. She uses scanning electron microscopy at the MIT Center for Materials Science and Engineering to image the gold-coated electrode structures and the nanogaps, while using electrical measurements to verify the effect of the presence of the molecules in the switching gap. Building her switch on a silicon/silcon-oxide base, Niroui added a top layer of PMMA, a polymer that is sensitive to electron beams. She then used electron beam lithography to pattern the device structure and wash away the excess PMMA. She used a thermal evaporator to coat the switch structure with gold. Gold was the material of choice because it enables the thiolated molecules to self-assemble in the gap, the final assembly step. For the initial tunneling current demonstration, Niroui used an off-the-shelf molecule in the gap between electrodes. Work is continuing with collaborators in Swager's chemistry lab to synthesize new molecules with optimal mechanical properties to optimize the switching performance. "Our project uses this design to have two metal electrodes with a single layer of molecules in the middle," Niroui explains. "We use self-assembly of molecules that allows the gap to be fabricated very small. By choosing the molecule and its properties such as the molecular length, we can control the gap thickness very precisely in the few-nanometer regime. The reason we want the gap small is that it allows us to reduce the switching voltage. The smaller the gap, the smaller the switching voltage and the less energy you are going to consume to switch on and off your device, which is very desirable." The molecules filling the gap act as tiny springs. When an electrostatic force is applied, the electrodes compress the filler, squishing all the molecules. "These molecules are going to prevent the two metals coming into contact. At the same time the compressed layer is going to provide a restoring force, so it's going to avoid the typical sticking problem, permanent adhesion between the two electrodes, that is otherwise very common in electromechanical systems," she says. Tunneling electromechanical switches work by controlling the gap between two metal electrodes that never come into direct contact. "You always will have a gap between the two electrodes. Because of the gap, the current that you modulate is the tunneling current," Niroui says. Niroui tested a version of her original device without a molecular gap filler and the two electrodes immediately stuck together. By filling the gap, current-voltage tests showed characteristics that were reproducible and repeatable, so the devices didn't short. "By comparing to theoretical models, we observe that we get some compression of the molecules, and we extract mechanical properties of molecules that match what is reported experimentally in the literature," she says. While the device established proof of concept, improvements are needed in the filler material for practical use. Niroui, 26, is from Toronto, Canada, and received her bachelor's in nanotechnology engineering at the University of Waterloo. She received a master's in electrical engineering at MIT in 2013. She hopes to complete her doctoral work in 2016.


Hrizi H.,Electronics Laboratory | Sboui N.,Electronics Laboratory
Applied Computational Electromagnetics Society Journal | Year: 2012

The wave iterative method is a numerical method used to model electromagnetic circuits. It is based on the concept of waves in the place of electromagnetic fields. To study the electronic circuits having complex structures, this method requires much time. We propose in this article to improve this method by using techniques of image processing. That's why the structure of the studied circuit is considered as an image. The objective is to reduce computing time by reducing dimensions of the calculation matrices. The reduced matrices are built containing only the important part of the information. Our goal is to prove that the most important zones in the structure are located in the contour with small steps in the vicinity of the contour. © 2012 ACES. Source


Boulejfen N.,University of Calgary | Harguem A.,Electronics Laboratory | Hammi O.,University of Calgary | Ghannouchi F.M.,University of Calgary | Gharsallah A.,Electronics Laboratory
IET Microwaves, Antennas and Propagation | Year: 2010

Measurement of non-linearity effects in wireless transmitters had been always difficult under complex digitally modulated signals. On the other hand, blind time domain simulation of these effects is generally time consuming and insufficient. In this study, the authors present new frequency domain closed-form formulas for predicting the spectral regrowth in wireless communication systems/subsystems with fifth-order non-linearity exhibiting memory effects. Particularly, the suggested formulas estimate the power spectral density, as well as the undesirable adjacent-channel and co-channel emissions of the modelled device, under phase-aligned and randomly phase-modulated multitone excitations. In addition, they are able to predict the correlated and uncorrelated distortion powers separately. The efficiency and robustness of the proposed approach has been demonstrated by predicting the output of a laterally diffused metal oxide semiconductor Doherty-based power amplifier. The obtained results have been compared to measurements and those of time-domain simulators and have revealed good accuracy and time efficiency. © 2010 The Institution of Engineering and Technology. Source

Discover hidden collaborations