Science & Technology News (658)
Posted: October 30, 2014 02:05PM
One of the fundamental functions of computers is that they can perform computations very rapidly and accurately. Achieving perfect accuracy is not always easy though, and may be getting harder as components shrink in size. By tolerating error in a program or algorithm, computers could become more efficient and even faster, and now researchers at MIT have developed a new tool to help developers find parts of their code that can endure some error.
The system builds on a language called Rely that is able to analyze code to identify sections that can tolerate some error, without completely compromising the results. While Rely could find these sections, Chisel, the new system, is able to mark them automatically for being run on unreliable hardware. Chisel is also guaranteed to maximize energy savings. Of course there is a limit to how much inaccuracy is acceptable, which is why Chisel can also simulate results with different tolerances, according to the developers' wishes.
The researchers tested Chisel on image processing and financial analysis algorithms, with a range of unreliable hardware models, and found a power savings of 9% to 19%. Those models consider the reliability of individual operations though, but if a bunch of operations were all executed at once, the savings and performance could be many times greater.
Posted: October 30, 2014 10:08AM
Prosthetics can be amazing pieces of technology as some now replace the functionality of the lost limb, such as prosthetic hands actually capable of multiple grips. Of course, being so advanced, these prosthetics are not cheap with the average myoelectric prosthetic costing between $30,000 and $40,000. By using 3D printing though, researchers at the University of Illinois at Urbana-Champaign have brought that cost down to $270.
For years the researchers have been working on electromyographic technology needed for controlling and receiving feedback from prosthetics, though this use of 3D printing has sped up the work. The pieces of the prosthetic take 30 hours to print and another two hours to assemble, with the electronics needed to convert neural signals into movements inside of it. That is except for the electromyographic board, which is currently about the size of an audio mixing board, but will eventually be small enough to fit into the residual limb's socket. A key part of the electronics is a microcontroller with a machine learning algorithm. It is responsible for learning the inputs and correctly converting them into gestures.
While the lower cost could benefit any number of amputees, the researchers are looking at bringing these prosthetics to those in nations without access to such care. The next step for this technology is to add a feedback system, so that the user has some feeling of the force the fingers are applying, and where the hand is, without having to look at it.
Posted: October 30, 2014 07:11AM
Lasers are already pretty awesome for what they are used for every day, but some day we may see them also used to propel aircraft. Such propulsion systems have already been proposed, but now a new design may push things to a new level. As reported in The Optical Society's journal Applied Optics, researchers have found described a way to potentially help rockets exceed Mach 10 using lasers.
Previous proposals for laser propulsion use a laser ablation process. By firing a pulsed laser at the surface of a craft, the surface is heated and some of the material is burned off. This creates a plasma plume that flies off of the surface, pushing the object in the opposite direction. What the researchers have done is combined this approach with the traditional gas nozzle of a rocket, so that the plasma plume and the gas flow work together, pushing the gas to supersonic speeds, while burning less fuel. The trick is making sure the shockwaves the laser introduces travel along the nozzle's inside walls. The other laser ablation methods could choke the nozzle's inlet with their shockwaves, reducing thrust.
Of course this work is currently theoretical, but it is bringing us close to a practical use for laser-propulsion systems. Such systems would dramatically affect space and aircraft, by reducing the need for fuel, thanks to remote lasers.
Source: The Optical Society
Posted: October 29, 2014 02:07PM
Wires carrying electrical currents surround us in buildings and offices, just as we can clutch electronic devices in our hands. All of these wires, no matter their size, resist the currents the carry, which results in energy being lost as heat, but that may change in the future. Superconductivity is a special state some materials can enter that allows currents to flow without resistance, and now researchers at Brown University have found a kind of superconductivity first described fifty years ago.
The kind of superconductivity that has been studied for decades relies on Cooper pairs, which are pairs of electrons, with one having an up spin and the other a down spin. Unlike single electrons, these pairs are able to flow through a superconductor with little interference. When exposed to a strong magnetic field though, the pairs can be ripped apart, but in theory, some superconductors will instead take on a new kind of superconductivity. To study this, the researchers had to use an organic superconductor that is comprised of ultra-thin sheets stacked atop each other and, counterintuitively, raise its temperature. Normally higher temperatures destroy quantum effects, like superconductivity, but here it actually grew the window needed to observe Andreev bound states. These states are groups of like-spin electrons, forced together by a magnetic field that can flow through non-superconducting regions, without resistance
Obviously this discovery will impact the future of superconductivity, but it may also affect studies into astrophysics and spintronics. The latter is a potential replacement for electronics that could promise faster and more efficient computers.
Source: Brown University
Posted: October 29, 2014 10:25AM
Though it has been awhile, I can remember a number of times in school when a teacher could not quite align a projector's image on a wall. Imagine trying to accomplish the same task on a curved or uneven surface, instead of a flat wall or screen. While there are means of doing just that, they are flawed. Researchers at Fraunhofer IOF, however, have a new solution that not only makes it possible, but at very high speed.
Traditionally the aperture of the projector has to be closed down, to create a sharp image on a curved or uneven surface, but this reduces the brightness of the image. The Fraunhofer solution still uses a small aperture size of 0.8 mm, but employs an array of microprojectors that can overlay their images, brightening the projection. Over one hundred of these microprojectors are in each 1 cm x 1 cm element of the larger projector. Each microprojector itself consists of a micro-slide projector and projection lenses, with a single LED. Beyond allowing for projection onto non-flat surfaces, this design also makes it easier to adapt to the geometry of the surface, as the slide array can be altered, instead of the entire optical system.
Further this projection system can project about 1,000,000 images a second, which is roughly ten thousand times faster than conventional projectors. This would be useful for projecting images from high-speed cameras.
Posted: October 29, 2014 07:01AM
Lasers are pretty cool pieces of technology as they are able to produce coherent beams of photons. This had made them almost ubiquitous in many devices used every day. Thanks to researchers at Princeton University, they may also be getting a dramatic bump in power soon.
Lasers work by exciting electrons in a material, because when the electrons fall, they will emit a photon of a specific frequency. Not all of the lasing material is used in this process, which is something some researchers have been investigating. At first the research just found where the lasing first occurs, so that energy could be focused there to initiate photon emission sooner. The Princeton researchers looked past this first step though and found that some of the later patterns or modes of emission can be more efficient than the first. The reason for this is that the different modes will compete with each other for energy, but by directing the energy to the mode of interest, all of the energy can be used by it.
According to the researchers' model, by focusing on the more efficient, higher order modes, output efficiency could improve 100 to 10,000 fold. That would translate to high power lasers, typically requiring the power of a wall outlet, being run off of batteries.
Source: Princeton University
Posted: October 28, 2014 09:59AM
Some discoveries bring out a lot of skeptics and demand more and more evidence before being accepted. One example of this would be if the microbe Geobacter produces nanowires capable of transmitting electrons like a metallic wire. Researchers at the University of Massachusetts at Amherst, where the original discovery of the nanowires was made, have now provided the most significant evidence of these biological nanowires.
Microbial nanowires are very small filaments of protein some bacteria grow in order to connect with each other, sharing energy and information. Typically proteins carry electrons by having them make discrete jumps, unlike in metals where electrons are delocalized from the molecules, and can freely move. The electrons in Geobacter's filaments are delocalized though and now the Massachusetts researchers have shown this by capturing the movement of electrons with electrostatic force microscopy. This technique allows for the flow of electric charges to be imaged, and showed them travelling just as was predicted.
Beyond verifying the past research, this study could open up some interesting applications, such as using the microbes to convert waste materials into methane gas. They could also be used as environmental sensors and even biological computers, with the filaments matching the conductivity of carbon nanotubes.
Posted: October 28, 2014 06:41AM
Anyone with a smartphone has likely experienced its battery dying at an inopportune moment. Of course many things will drain the battery, but one in particular is the power amplifier, which is necessary for transmitting radio signals. Researchers at MIT formed a startup company, Eta Devices, to produce a chip to regulate the voltage of the amplifier, drastically reducing the power used.
To make sure the radio is always ready to send a signal, the power amplifier tends to be at maximum voltage all of the time. Naturally that cuts into battery life and, because half of that power may be lost as heat, it influences phones in a number of other ways too. The Eta Devices chip however, acts like a gearbox to step down the voltage, when full power is not needed. Base stations could also benefit from this new chip, raising efficiency to 70%, from an optimistic 45%. That could save a large carrier $100 million each year.
As Eta Devices is already working with manufacturers, we may see its technology deployed by the end of next year in base stations and smartphones. While some manufacturers are working on similar systems to increase efficiency, the Eta Devices technology is unique in its ability to work with the high frequencies of new communication standards, like LTE Advanced and 802.11ac.
Posted: October 27, 2014 04:29PM
The computer industry is rapidly approaching the limits of current technology as it races to make smaller and faster components. One of these limits has to do with how dense you can make the features on a chip. There is a potential way to push back this limit using polymers, and two techniques recently developed at NIST should help that happen.
Block copolymers are a type of materials that are actually two or more polymers, connected together. These materials have some interesting properties, such as the ability to self-organize into specific patterns. Testing this ability is challenging though, as the equipment and processes is both time consuming and expensive. This is still true with what the NIST researchers have developed, but this new work still has value. Specifically the researchers have produced two studies that describe how to use soft X-rays to measure the average shape of a block copolymer structure and how to use transmission electron microscope tomography to image defects within a film. Combined you can find problems in the patterns you are trying to work with.
While these processes do not offer a way to detect defects on the factory floor, they can be used for refining computer models, which can be tested rapidly and cheaply by the industry. Using block copolymer patterns, it should be possible to reach smaller feature sizes in computer chips than current technology, but only if the patterns they layout can be trusted. Those models are what will one day identify if the pattern will have flaws or not.
Posted: October 27, 2014 10:50AM
We all make mistakes and can really only hope that these mistakes are either caught, or of little consequence. That is not always the case though with some errors having great impacts on results. Researchers at the University of Massachusetts at Amherst have created a plugin for Microsoft Excel for finding potential errors in spreadsheets that could dramatically affect the final result.
CheckCell uses data analysis and program analysis to examine a spreadsheet and identify cells that either stand out as suspicious or could have a great impact. The latter is especially useful as sometimes the importance of a cell is not immediately apparent. By following how the data can affect the result, CheckCell will flag the entry to let the user know it should be verified.
The CheckCell plugin is freely available to everyone now, after being released at the Object-Oriented Programming Systems, Languages, and Applications (OOPSLA) conference. You can find it here: www.checkcell.org.
Posted: October 27, 2014 06:39AM
Lithium-ion batteries are everywhere today and in almost everything. If you need a rechargeable energy storage system with high capacity, those are what many turn to. As great as these batteries are though, more and more applications are demanding more, and thanks to researchers at Berkeley Lab, a possible replacement has gotten an important boost.
Lithium ions have a single positive charge and an obvious way to improve energy density would be to use an ion with a greater charge, like magnesium, which can have two positive charges. If it were as simple as just substituting lithium for magnesium, we would already have magnesium-ion batteries. One of the challenges facing magnesium-ion batteries has been that often, when magnesium is in a solid, it is surrounded by six neighbors. In a battery these neighbors would come from the electrolyte, and would inhibit the ion from entering an electrode. According to the Berkeley researchers' new computer model though, the problem is less severe than many thought, as only four neighbors would come in. That may not sound like that big a change, but it will make entering the electrode and delivering energy much easier.
The researchers do have an explanation for why some studies have shown the six neighbors from the electrolyte gathering around the ions. They suggest that if the electrolyte concentration is too high that it could cause the ions to take on the six neighbors like in a solid, at the electrode interface. By keeping the concentrations low enough, this would be avoided.
Source: Berkeley Lab
Posted: October 24, 2014 02:23PM
As though holograms were not cool enough already, now quantum holograms may have a potential use for storing quantum information. As published in the European Physical Journal's EPJD, researchers at St. Petersburg State University have successfully retrieved information from a quantum hologram.
Holograms are a well understood means for storing optical information but have not been used before to store quantum information. Normally they store information using light intensity, but for a quantum hologram the medium that holds the information has to be able to store a quantum superposition. Superposition is a quantum phenomenon whereby quantum objects can exist in mutually exclusive states at the same time. To actually write the information onto the medium, the researchers had to be able to shape the control field in both space and time. The control field is the reference light used to write and later read information from a hologram medium.
Eventually the researchers would like to see quantum holograms used to store quantum signals but also to transform their quantum states. This ability would be useful for quantum communication and computation.
Source: European Physical Journal
Posted: October 24, 2014 09:03AM
There are many chemical reactions we would like to be able to do, such as converting hydrocarbons into fuels, but the processes tend to require prohibitive amounts of energy. To overcome this issue, catalysts are used, which reduce the amount of energy needed at any one time, by adding intermediate steps. Researchers at Ames Laboratory have recently modelled the movement of particles through narrow channels, which these catalyzed reactions may someday occur in.
One way to improve catalytic reactions is to increase the surface area of the catalyst, as that provides more reaction sites. To that end, researchers are investigating using porous nanoparticles, as the channels within them have a great deal of surface area. The size of these channels is important though, because the narrower they are, the more of them can exist, but if they are too narrow, molecules will not be able to pass through them. This is why the Ames researchers have created a model of mesoporous nanoparticles to simulate how molecules will behave when passing each other, in a narrow channel.
Due to the complexity of the processes involved, the researchers called upon those experienced in theoretical chemistry and applied mathematics. The hope is to optimized the size of the pores in mesoporous nanoparticles, and thereby their performance.
Source: Ames Laboratory
Posted: October 24, 2014 07:16AM
Many of the solar cells you can find today are made of ultra-pure silicon, which is a slight problem as such silicon is energy intensive and expensive to make. It is necessary though, for solar panels to efficiently generate electricity. However, researchers at the Norwegian University of Science and Technology have developed a way to create solar cells of a different geometry, made of less-pure silicon.
The new design borrows from fiber optic fabrication by coating silicon fibers in glass, and heating them both up. Once heated, the silicon will melt and the glass will soften enough that they can be stretched, forming a long, thin glass fiber with a silicon core. This process naturally purifies the silicon, be melting and re-solidifying it within the glass. These fibers can then be made into vertical rod radial-junction solar cells, with electrons in the silicon absorbing the energy from the light and being captured at the junction on the surface of the silicon. Planar solar cells, for contrast, must be made of ultra-pure silicon because of the distance the electrons must travel to reach a junction or electrode.
As this manufacturing approach is based on the technique for creating bulk silicon-core fibers, producing solar cells from dirty silicon this way would be rather cheap. The catch is that prototype has only achieved 3.6% efficiency, which is far below commercially available solar panels. Of course, that will change as the technology is further developed.
Posted: October 23, 2014 02:12PM
Traditionally a camera's zoom works by changing the distance between lenses. While this does work, it is not particularly ideal, given the size of the necessary mechanics. Adaptive zoom however uses a different approach that requires a less-bulky system, and now researchers at Sandia National Laboratories have successfully brought this technology to riflescopes.
For our eyes to focus on an object, muscles must flex a lens, causing its focal length to change. Adaptive zoom works in the same way, though the lenses are now polymers and the muscles are, in this case, piezoelectric materials. More accurately the muscles are an ultrasonic piezo motor that move a rotor and lead screw, which flex the lenses. The use of an ultrasonic motor is important as it allows the focus and zoom to be maintained, even when power is lost. When tested, this new system showed it could complete 10,000 actuations on just the power of two AA batteries.
Adaptive zoom is not a new technology, but this is the first time it has been built into a riflescope for the military with the Rapid Adaptive Zoom for Assault Rifles (RAZAR), which can change from high to low magnifications at the push of a button. The change in magnification happens quickly and does not require the user to move their hands or eyes, which could save their life in a firefight. This technology could also see use in other optical devices, such as binoculars and cell phone cameras.
Source: Sandia National Laboratories
Posted: October 23, 2014 10:24AM
What would science fiction be without guns going pew pew with laser bullets? The question can also be, what would reality be with those bolts of energy, and researchers at the Institute of Physical Chemistry of the Polish Academy of Sciences decided to find out. The researchers filmed the laser pulses traveling down a hallway, watching for how they illuminate the area.
Of course a normal camera cannot actually film a light pulse, as light travels too quickly. To overcome this issue, the camera was connected to the laser and recorded consecutive frames for different pulses, combining them into a single film. As the physics involved does not change with each pulse, the film provides all of the information as though the camera recorded just one pulse. Those pulses, by the way, were femtosecond long and 10 terawatts in power. That high power is especially important as it allowed the pulse to create a plasma fiber in the air, as it traveled. This caused the laser to self-focus, allowing it to travel much farther without degrading.
The laser itself had a frequency in the near infrared, but as it traveled through the air, it produced a white light, which is important for potential uses of the laser. White light contains many frequencies, and if this laser were used in a LIDAR system, those frequencies could be used to measure certain elements and compounds in the atmosphere.
Posted: October 23, 2014 06:29AM
A device's LCD display can easily be one of its greatest power drains, thanks to the backlight needed to actually show the image. This is not the only power draw though, as the pixels themselves require energy to maintain an image. Bi-stable displays however, can eliminate this requirement, and now researchers at the Hong Kong University of Science and Technology have created a bi-stable display that can also show 3D images, as reported in The Optical Society's Optics Letters journal.
Liquid crystal displays operate by twisting the liquid crystals within each pixel, causing different amounts of polarized light to escape. Normally the twisting is controlled by an electric field, but in bi-stable displays the crystals are set in place using a flash of light. This removes the constant power draw and the electrodes to deliver the energy, allowing the displays to be thinner and more efficient. The Hong Kong researchers went a step further with their design though, by building three zones into the display with different polarizations. With the proper glasses and filters, the displayed image will be 3D to the viewer.
Before you prepare to throw out your current monitor or television, the display is far from commercially viable. Currently it is only greyscale and has a refresh rate too low to show video. The technology could still see use with ereaders and possibly in credit cards, as an added security measure.
Source: The Optical Society
Posted: October 22, 2014 02:17PM
Batteries are a big deal and we keep creating more technologies that rely on them. Naturally we also want the batteries to last longer than they currently do. One way to potentially achieve this is with lithium anode batteries, and researchers at Oak Ridge National Laboratory have made a discovery that could help bring those batteries out of the lab.
In theory a lithium anode battery would offer the highest known theoretical capacity, when used with an aqueous electrolyte. The problem here is that lithium and water react violently together, so something must keep them separated in the battery. What the ORNL researchers have done is identified a new separator called LLZO, which is a cubic garnet material. Unlike some other separators, LLZO is able to withstand very extreme alkaline environments, even staying stable at a pH value over 14.
This great stability can indirectly increase some battery's energy density, as some have turned to diluting the aqueous solution to keep other separators from failing. By removing the need to dilute, the battery can be smaller, bumping its energy density up.
Source: Oak Ridge National Laboratory
Posted: October 22, 2014 11:29AM
Everybody wants to get their data faster, whether it is a game or video, the sooner it can be on their screen, the better. Of course achieving higher speeds is not easy and special technologies may be required. Researchers at Chalmers University of Technology have recently set a new record for transmission rate of 40 Gbps when using signals at the high frequency of 140 GHz.
To accommodate the increased bandwidth users' demand, researchers are trying to use signals of higher frequencies. The higher the frequency, the more data a signal can carry, but it is harder to send a 100 GHz or higher signal with enough power to be detected. The Chalmers researchers were able to achieve this in their laboratory though, by wielding semiconducting circuits made of indium phosphide, and will soon be discussing their work at a conference.
The uses for such high speed, wireless data transmission are obviously numerous and include such things as transmitting live cultural and sporting events to screens in high resolution, and making connections between server rooms and even within our homes, without the need for more cables. The researchers believe that in a few years they may even be able to achieve their goal of 100 Gbps wireless transmission.
Posted: October 22, 2014 06:47AM
Some say we are in the era of big data, where massive collections of information are gathered and shared across the planet. Some of the most massive collections come from and go to research facilities, which is why unbelievably fast networks are built between them. Soon new connections will be built between the United States and Europe, delivering a capacity 340 Gbps, and the project will be managed by Berkeley Lab.
Perhaps the prime example of why this extension is needed is the Large Hadron Collider, as it can produce 30 petabytes of data a year and that value may go up over time (one petabyte is a thousand times larger than a terabyte). To be of any use, that data must be shared with other facilities for processing and study, which this new link will help with. Here in the US it will connect with the Energy Sciences Network, for distribution to US laboratories and universities, while in Europe it will connect to the GÉANT network organization with 100 Gbps links.
The goal is to have the network extension in production by January, to take advantage of the LHC current downtime for upgrades.
Source: Berkeley Lab
Posted: October 21, 2014 02:16PM
Many believe that there will be a new renaissance of sorts in the near future, thanks to 3D printing. Additive manufacturing in general opens up some interesting doors at it allows for the precise construction of objects, with potentially little waste. Now researchers at Oak Ridge National Laboratory have found a way to control the crystal structure of a material using an additive manufacturing system.
To achieve this control, the researchers had to precisely manage the electron beam from an ARCAM electron beam melting system. This system works by fusing layers of metal powder together using an electron beam. In this case, it was a nickel-based part they were creating.
Many properties of a material are influenced by its microstructure, so this work could have some very powerful implications. Microelectronics to jet engine components could all have their characteristics tailored to achieve the performance desired.
Source: Oak Ridge National Laboratory
Posted: October 21, 2014 09:44AM
Though perhaps not as lauded as some science fiction technologies, tractor beams are still somewhat common in the genre. They are also among the sci-fi technologies that can or do exist, in one form or another. Researchers at the Australian National University have recently developed a laser tractor beam that operates on a different principle than most, allowing it to actually reverse its effects.
Normally a laser-based tractor beam uses the momentum of photons to move the particles they target. This design however uses the laser beam to selectively heat the target particle. That heat then causes air molecules to warm and move away from the surface, forcing the particle to recoil. By moving the hotspot on the particle, the researchers are able to change how the particle moves, potentially having to reverse direction.
As this design uses only a single laser, it is somewhat versatile and could find some interesting applications. These include capturing particles, including pollution, from the air or retrieving other particles for sampling purposes.
Source: Australian National University
Posted: October 21, 2014 05:43AM
For many of us, the Internet is likely always available, except during an outage. To better understand the Internet and recognize outages, researchers at the University of Southern California tracked Internet usage across the planet for two months. From this data, they created a map that shows the Internet sleeping, as it were.
To collect the data, the research pinged 3.7 million IP address blocks, which would be around 950 million individual addresses, every 11 minutes for two months. This allowed them to establish a baseline for Internet usage and then track its use through the day. In the most developed areas, such as the United States and Europe, usage appeared constant as the routers we use to connect are always on. In less developed and wealthy nations however, usage will fluctuate.
The plan for this research is to help scientists and policymakers understand the normal operation of the Internet. With this information, normal downtimes may not be misidentified as outages and be able to predict how networking policies can influence network usage.
Posted: October 20, 2014 02:12PM
Superconductivity is an interesting phenomenon that could have many applications from power lines to computer circuits. Exactly how a superconducting circuit would take shape though is still being determined. Researchers at MIT have recently developed and tested one circuit design that could make superconducting circuits cheaper and easier to produce.
Josephon junctions are a kind of superconducting circuit that have been able to reach 770 GHz, but while they do have great potential, they are hard to make and hard to work with. The new design from MIT is called a nanocryotron, or nTron, and would be much simpler to create as it consists of nanowires forming a T shape. Where the base meets the crossbar though, the base tapers down to a point. When an electrical current is applied to the base, the point causes electrons to bounce off of each other, producing enough heat to warm the crossbar past its critical temperature. This ends the conductivity of the crossbar, until it is cooled again, and thus acts as an electrical switch. This design does have limitations though, such as likely never surpassing 1 GHz, but being simpler to create, working with currents similar to those in other hardware, and an impedance that matches that of mRAM may still make it very useful.
As superconductors can carry a current without resistance, they can work with a percent the energy needed by modern electronics. Of course the liquid-helium cooling system would cut into some of that savings, but it could still be a net improvement.
Posted: October 20, 2014 09:16AM
Light emitting diodes for home lighting have been growing in popularity of late, thanks to their efficiency and long lives. One problem they possess though is cost, and that problem may get worse in the future. Rare earth elements are needed for LEDs and many other technologies, and as the supply shrinks, costs will increase, unless someone does something about it.
As reported in the Journal of the American Chemical Society, researchers have developed an alternative to the rare earth elements in LEDs that is instead based on copper iodide. This material is much more abundant and can still be tuned to produce a variety of colors, including white. The white light the researchers were able to make was also warmer than that typical white LEDs emit.
Source: American Chemical Society
Posted: October 20, 2014 05:56AM
It can be weird to think of the number of technologies that future generations may never encounter, and one of these will likely be the typewriter. With computers and word processors, it would seem like typewriters no longer serve a purpose. With some imagination and creativity though, a student at the University of Plymouth has resurrected a typewriter and connected it to the Internet.
It was not an easy task, as some parts had to be built and supplied by a German company, but it was a fun project, according to the 22-year-old student. The typewriter, called Dico, can tell when someone is approaching it using ultrasound, and will then connect to a random person on an Internet chat service. Using micro-controllers and solenoids, messages can be received and typed onto the paper to be read, though sometimes it will be encrypted. This was intentional, to cause the user to think about online security.
The student would like to refine Dico some more and may even build another, so the pair can talk to each other. This is not the only time he has converted older technology into something new and interesting, as he gave the 1950s short wave radio the ability to create small poems from people's brain activity.
Source: University of Plymouth
Posted: October 17, 2014 12:31PM
Mirrors are fairly common and useful objects, but do you know how they work? When light, electromagnetic radiation strikes a mirror, it interacts with the electrical component, causing the reflection to have the reverse electric field. Though not possible naturally, metamaterials can allow for magnetic mirrors that reflect the magnetic component, and, as reported in The Optical Society's Optica journal, a new design has been developed.
The reason metamaterials are required for a magnetic mirror has to do with the fact that magnetic fields interact with charged particles, but light has no charge. The first metamaterials to achieve magnetic mirroring only worked on long microwave frequencies. More recent designs using 'fish-scale' shaped metallic components allow for shorter wavelengths, but have considerable signal loss and are very dependent on the components' particular shapes. The new design addresses these issues by using a 2D array of non-metallic dielectric resonators. These resonators interact strongly with the magnetic component of light and, being made out of tellurium, have a low signal loss compared to their metallic counterparts.
Among the reasons magnetic mirrors would be useful is that traditional mirrors reversing the electric components of light, causes that component to cancel out, at the surface. This prevents nanoscale antennas and quantum dots from interacting with the light. By preserving the electric component, a magnetic mirror would allow for the maximum absorption of the light, and new technologies.
Source: The Optical Society
Posted: October 17, 2014 05:42AM
More many modern technologies, thin is in, but this example may be a bit extreme. Researchers at Columbia University and the Georgia Institute of Technology have discovered a 2D material that exhibits the piezoelectric effect. This effect links mechanical deformation to the creation of electrical voltages.
Molybdenum disulfide (MoS2) is part of a family of materials known as transition metal dichalcogenides, which are 2D semiconductors. In theory, they can all be piezoelectric because they all break central symmetry, but this is the first time one has been successfully tested. While MoS2 did demonstrate the effect though, it does not always. As a bulk material, it does not, but even when working with atomically-thin sheets, it does not always. There must be an odd number of layers to generate any electricity, because with an even number the layers' output voltages will actually cancel each other out. Also electricity was only generated when the material was deformed in a specific direction.
Even with all of these conditions, this is still an important discovery, as it suggests other piezoelectric materials that can be just atoms thick. Being so thin, the materials will be naturally flexible, making it possible to use these generators to power wearable devices.
Source: Columbia University
Posted: October 16, 2014 05:26AM
We may not think about it when we turn a light on, but every light source only produces so many photons in a second. Depending on the application, this emission speed can be critical, such as in telecommunications where 10 billion photons need to be produced in the blink of an eye. Typical LEDs only produce 10 million photons in that much time, but researchers at Duke University have found a way to close the gap.
The Purcell effect describes how emissions from fluorescent molecules can be increased when the molecules are near an intense light source. To create that light source, the researchers sandwiched the molecules between a 75 n silver nanocube and a thin film of gold. When the cube size and gap are tuned properly, the researchers realized a 1000 fold increase in fluorescence speed for the LED.
That record-setting speed may be surpassed before long though, as the researchers believe that aligning the molecules could magnify the effect. Regardless, fast light sources like this could prove invaluable for telecommunications and even quantum cryptography.
Source: Duke University
Posted: October 15, 2014 03:41PM
Teams around the world are working on technologies to enable quantum computers, and some teams are right next to each other. Two teams at the University of New South Wales have independently developed two quantum bit, or qubit systems, and have even set some records with them.
The fundamental units in quantum computers are qubits, which are particles put into a superposition. Superposition is a phenomenon that allows something to exist in multiple, exclusionary states at the same time, and from this quantum computers would be able to run algorithms far different than conventional computers. Superposition is fragile though, so qubits must be designed for great accuracy, and to achieve that here, both teams used pieces of pure silicon-28, which is nonmagnetic, unlike the mix of isotopes in modern electronics. One team put a phosphorus atom in their silicon, actually creating two qubits (the nucleus and the electron) while the other created an artificial atom that is very similar to MOSFETs, a kind of silicon transistor.
The ability to build a qubit in and from silicon is of great importance as it could allow current manufacturing methods and tools to construct quantum computers in the future. The phosphorus-atom qubit design also managed to survive for 35 seconds, which is impressive for any solid-state qubit, but is actually the record for silicon qubits.
Source: University of New South Wales