Tag: spintronics

Intel ready to sacrifice power for energy efficiency

Intel's Gordon Moore and Robert NoyceChipzilla is ready to embrace alternatives to the speed at all cost ethos which has made it so successful.

William Holt, who leads the company’s technology and manufacturing group, said this week that for chips to keep improving, Intel will soon have to start using fundamentally new technologies.

Holt pointed to two possible candidates – tunnelling transistors and spintronics. Both would require big changes in how chips are designed and manufactured, and would likely be used alongside silicon transistors.

What is important is that the technology will not offer speed benefits over silicon transistors and chips may stop getting faster. The new tech would improve the energy efficiency of chips, something important for many leading uses of computing today, such as cloud computing, mobile devices, and robotics.

“We’re going to see major transitions,” said Holt, speaking at the International Solid State Circuits Conference in San Francisco. “The new technology will be fundamentally different.”

Holt said that the status quo can only continue for two more generations, just four or five years, by which time silicon transistors will be only seven nanometres in size.

Tunnelling transistors are far from commercialization, although DARPA and industry consortium Semiconductor Research Corporation are funding research on the devices. They take advantage of quantum mechanical properties of electrons that harm the performance of conventional transistors and that have become more problematic as transistors have got smaller.

Spintronic devices are doable and could hit the market next year. They represent digital bits by switching between two different states encoded into a quantum mechanical property of particles such as electrons known as spin.

Spintronics will appear in some low-power memory chips in the next year or so, perhaps in high-powered graphics cards.

For example, Toshiba announced last year that it had developed an experimental spintronic memory array that consumed 80 percent less power than SRAM, a type of high-speed memory.

Holt claimed that continued gains in energy efficiency, not raw computing power, are most important for the things asked of computers today.

“Particularly as we look at the Internet of things, the focus will move from speed improvements to dramatic reductions in power,” Holt said. Power is a problem across the computing spectrum. The carbon footprint of data centres operated by Google, Amazon, Facebook, and other companies is growing at an alarming rate. And the chips needed to connect many more household, commercial, and industrial objects from toasters to cars to the Internet will need to draw as little power as possible to be viable,” he said.

Graphene spins like nothing else on earth

illustration courtesy Chalmers University of TechnologyScientists at Chalmers University of Technology said that large area graphene preserves electronic spin over a long period.

And that means the window is now open for the long-touted spintronics, promising faster CPUs and memory for computers.

The aptly named Saroj Dash, head of the research group at Chalmers, said: “These results will attract a lot of attention in the research community and put graphene on the map for applications in spintronic components.”

Spintronics, already used in state of the art hard drives but Dash believes there’s more exciting times ahead. “The thin carbon film is not only an excellent electrical conductor, but also theoretically has the rare ability to maintain the electronics with the spin intact,” said Dash.

That means that in spin components of the future, electrons will be able to travel several tens of micrometers with spins staying aligned, something you can’t do with aluminium or copper, Dash said.

The researchers have managed to create graphene through chemical vapour deposition. Dash said: “The CVD graphene can also be easily removed from the copper foil on which it grows and is lifted onto a silicon wafer. There are good prospects for the production of large area graphene on an industrial scale.”

So when, and if, will we see spintronic computers? On that subject Dash is a little more pessimistic. He said: “Whether spintronics can eventually fully replace semiconductor technology is an open question.”

Cambridge researchers promise breakthrough in 3D chip design

Physicists at the University of Cambridge have developed a new way of transferring data and logic signals from one layer of a 3D chip to another.

Although it is possible to transfer date using conventional circuitry, such a traditional approach is cumbersome and generates a lot of heat inside the 3D circuit. However, the physicists managed to develop a spintronic shift register that allows information to be passed between different layers of the chip, Gizmag reports.

At the moment, 3D chips appear to be the only way of overcoming limitations in chip design, posed by lithography, materials and thermal limitations. Chip designers could use the 3D approach for the next decade or two, as it will allow them to stack multiple layers of circuitry in a single package. However, until now there was no good way to transfer information between the layers. Using currently available technology, the advantages of 3D chips could be nullified, as they would eventually have to use most of their circuitry to pass information back and forth between different layers.

Professor Russell Cowburn and his team took up the challenge and came up with a vertically layered spintronics shift register to move data vertically between the layers, without the need for specialised circuitry. High density hard drives and magnetic random access memory currently use Spintronics. It relies on electronic current with parallel electron spins to pass information. Spin polarised current is sent through an ultrathin magnetic layer and the size of the current that passes depends on the orientation of the magnetic field and the spin direction. Since the current is controlled by a magnetic field, it can be used to read data.

Professor Cowburn’s team replicated the approach by using ultrathin layers of various materials to demonstrate how spintronics could be used to transfer data between 3D circuitry layers. They used a 2nm magnetic cobalt-iron-boron layer and a ruthenium layer, less than a single nanometer in thickness. The use of ultrathin films would allow developers to create very thin spin-up and spin-down domains, about 6nm in width.

However, a lab demonstration revealed the rapidly flipping magnetic fields are not good for super-large-scale integrated circuitry. The researchers hope to improve the technique by moving magnetically encoded data by flowing spin-polarised currents from layer to layer, causing the entire domain structure to climb within the stack. The lab demonstrator is hard wired and it can only send information upward. 


IBM claims spintronics breakthrough

IBM researchers are claiming a breakthrough in the development of spintronics based memory and storage devices.

Scientists at the IBM Research and the Solid State Physics Laboratory at ETH Zurich have succeeded in extending the ‘spin’ lifetime of electrons, opening the door for higher capcity and increased efficiency spintronics-based devices.

With spintronics, electron spins can be used to represent data, with a shift in orientation corresponding to either a one or a zero dictated by a magnetic field.  This differs from current methods which rely on the electrical charge of the electrons.

Keeping information encoded in the spin of an electron has been difficult, but the ETH researchers have succeeded in extending the spin lifetime of an electron by 30 fold. This means an increase to 1.1 nanoseconds, which is the same amount of time it takes for a 1GHz processor to cycle.

The scientists were able to do this by locking the electrons into a ‘persistent spin helix’, allowing them to monitor a regular pattern of movement, rather than the erratic spin that would usuallybe observed.   By following the waltz-like movement of the electrons using ultra short lasers it was possible to track the spin of the electron for longer, which the researchers say is a big step towards an electronically programmable spin-based transistor.

Devices using spintronics are still a long way off. Various firms such as Hitachi have also been developing spintronics technology, but there are stumbling blocks. Researchers have had some success with spintronics at room temperature, the IBM researchers were  unable to perform the extended spin tracking at higher temperatures. In this case, the research was only possible at 40 Kelvin, a rather chilly -233 degrees celcius.

OLEDs get cheaper and brighter with a 'spin'

Scientists have put a new spin on organic light emitting diode (OLED) technology which could offer brighter, cheaper, and more environmentally sound screens.

The ‘spintronic’ OLEDs differ to the ones that are used in panels such as smartphone screens. Big telly vendors have also been bleating about launching large OLED screens for some time now, though these are expected to command dizzying price tags upon their eventual release.

University of Utah physicists reckon that they have cracked a cheaper way to produce the OLED, using the ‘spin’ of electrons to record information.

This technique is being looked at for the future of computer chips too, and developments with both have been made possible through the invention of a ‘spin valve’.

This was then modified over a number of years to create a valve that, rather than controlling electron flow, was able to produce light, opening the door to developing OLEDs.

The team made advances with the previous method by changing the material used in the organic layer of the spin valve, making the light more efficient.

They also added a material to allow negatively charged electrons to be injected into the valve at the same time as a positively charged one.  This meant that more light could be generated by the device.

Creating fully working TVs based on the spintronics method may be some time off, according to the scientists.

At the moment it is only possible to use the device at a rather chilly minus 28 degrees, and they can only create one colour so far – orange.  Fine for the Dutch football team and the 90’s ‘Tango Man’ but otherwise unappealing.

The scientists say that they are working one producing red and blue over the next couple of years, with white spin OLEDs in the future too.

As for spin OLED sets hitting the shops, the team reckons it could be another five years before the necessary developments are made.

Graphene nanoribbons could extend Moore's Law by 10 years

Researchers are developing a technique with everyone’s favourite wonder material, graphene, that could potentially see Moore’s Law extended by 10 years.

The method involves standing atom thin sheets of graphene on a substrate to make high density walls for use on conventional electronic chips.  The same techniques could even be used for ongoing work into spintronics as an alternative to current chip production.

A paper produced by scientists at Rice University and Hong Kong Polytechnic University shows that certain materials such as diamond or nickel can be used to bind strips of graphene nanoribbon.

With the material used in such small quantities to hold the nanoribbons upright, on an atomic scale it is possible for graphene to still exhibit its Nobel prize-winning qualities.

Working with nanoribbons at this size to create field effect transistor, it should be possible to create massive amounts of transistors on a microchip.

With a few billion able to fit on a chip currently, the team are talking up mind boggling numbers.

According to Rice theoretical physicist Boris Yakobson reckons that it will be possible to pack 100 trillion FETs onto a one centimetre square chip.

This would in effect help keep Moore’s Law going by another decade or so.

Yakobson says that he spoke to Moore himself during the infancy of developments into nanotechnology.  Apparently Moore saw silicon wafers in terms of real estate, and Yakobson will now, to extend the metaphor, use upright architecture to go from “ranch-style houses in Texas to skyscraper condos in Hong Kong”.

According to the research of Yakobson and his team, which is clearly in a very early stage, the naoribbons would be able to stand easily at a 90-degree angle. As this is its preferred state minimal energy would be required.

 The walls could then be grown as a close as 7/10ths of a nanometre, which would mean that graphene properties would remain.  This could all be done on a silicon, silicon oxide or other materials.

Two types of nanribbon can be made, of the zigzag or armchair variety.   Zigzag are magnetic  would be useful as they cause electrons to spin off in opposing directions, an important factor in allowing spin to be measured.

Armchair nanoribbons on the other hand can become semiconductors, and can create a large band gap which is essential for creating transistors.

In both cases, the electronic properties of the walls can be tuned by changing their height.

If all goes to plan with development Yakobson reckons that subnanometre electronics coul be on the cards.



Researchers develop room-temperature spintronics

The possible transition to a new generation of electronic devices based on ‘spintronics’ has been given a boost with a method which would let the technology function at room temperature.

A team of researchers at Aarhus University reckon that it has made a massive leap forward in making spintronics a viable technology sooner, rather than later. 

With current computer electronics being produced at ever lower processes scientists are looking at spintronics as a method to enable faster and more efficient processors in the future.  While such a move is not vital in the short term, spintronics has been in developmental stages in many labs.

But now the scientists believe they have brought forward the use of spintronics significantly by devising a method to make it work under normal conditions.

This is down to a topological insulator material bismuth selenide which can help to regulate the flow of electrons used to record a rotational ‘spin’.  Because the effect is so small it is necessary for the electrons to travel large distances in order to work.

This means that it has been possible to make spintronics-based transistors work closer to absolute zero, around -270°C.  This makes the use of the technology in current electronic devices a tad tricky.

But with the discoveries made with bismuth selenide, it has apparently been possible to generate an effect from electrons which is over a hundred times stronger.  Crucially this means that the spin can occur effectively at room temperature.

This opens the door for spin field effect transistors that work at on even a nanoscale, while being viable for normal use.

The discovery stemmed from examining the properties of topological insulator bismuth selenide.  A topological insulator is a material that is able to insulate on the inside while conducting on the surface. 

Topological insulators have been the basis of other endeavors into spintonics in the past, with properties that mean that heat development is reduced meaning lower energy consumption. 

Now the researchers reckon that materials are finally catching up with theoretical ideas for development of spintronics technologies.

While it may still be some time before spintronics devices are on the shelves of your local electronics shop, the “rapid development” with topological insulators means that the technology could soon be within reach.

TSMC eyes seven nanometre process

Taiwan Semiconductor Manufacturing Co (TSMC) has confirmed that pilot production of its 20 nanometre process will begin in the second half of 2012.

The firm’s senior vice president in charge of R&D, S.Y. Chiang announced that its first foray into 450 millimetre wafer production would occur late next year, following on from the 28 nm pilot production in the second quarter of this year.

Chiang also noted that TSMC would begin its volume production phase of the 28nm technology towards the end of 2011, producing more than twice as many chips as the 40 nm process, with IC designs being finalised this year.

All of which fits in with Moore’s Law, stating that the numbers of transistors will roughly double every eighteen months since the 1965 invention of integrated circuits. But semiconductor expert Malcolm Penn said  that TSMC’s announcement about the new process is “like saying Christmas will be happening in December this year”.

With regards to future adherence to Moore’s Law, Chiang estimates that the rule will remain applicable to the semi industry until at least when the  seven nanometre process comes into production.

Chiang also predicted that the room for cooperation between both silicon foundries and chip assemblers will increase as the size of single chips and system chips continues to be scaled down.

Penn agrees that the shrinking of process size will continue up until the seven nanometre threshold, where the way that the physics involved in the circuits begins to change.

“At this point it becomes a different ball game,” Penn said, “with molecular level circuits and talk of single atom transistors as the physics change from here on out.”

This is where new developments in graphene based circuitry, spintronics and other exotic developments which are occurring in labs at early stages of research come into play.

Such technologies are a good while from becoming viable in production on a large level, but Penn believes we have around ten to fifteen years before the 7nm process is reached.

Up until then, though, Penn believes that progression should remain steady, with the step on the horizon being 16 or 14 nm, “depending on what your marketing department decides as they are both pretty much exactly the same.”

Hitachi, other scientists, change semi industry with spintronics

A huge number of boffins has come up with a way to control and measure spin current in semiconductors using a method similar to electrical current.

Scientists at Hitachi Cambridge Laboratory of Hitachi Europe along with the University of Cambridge, University of Nottingham, the Institute of Physics of the Academy of Sciences  and Charles University of the Czech Republic, and the Texas A&M University in the US, have developed the new technique using gallium-arsenide semiconductor material.

Electronic devices such as information processing, semiconductor, storage and power devices which have been the driving force in the rapid advancement of industry, social infrastructure, lifestyle and science in the 20th century, have previously been based on detecting the basic attribute of the “charge” of an electron.  

However, according to Joerg Wunderlich, who heads up the team of researchers at Hitachi Cambridge Laboratory, this hasn’t moved on until now.

He told TechEye that the new science and technology field of spintronics is based on the other basic attribute of an electron. This is the elementary magnetic moment, which is otherwise known as the “spin.”. This is an area which has attracted high expectations as it is believed that this could push forward a new way  for low-power consuming electronics, hybrid electric-magnetic systems and completely new technologies.  

Joerg Wunderlich told us: “The research into the new spinning began in 2005, when scientists realised they were able to measure separately an up and down spin at an extremely low temperature of -269C using a gallium-arsenide semiconductor, a non-magnetic material.”

He added that in 2009, they were able to use the same gallium arsenide semiconductor at a temperature of -53C, by measuring the flow of spin polarised current over a distance of a few microns.  

“In the current development, the up or down spin was controlled by a gate voltage, and the ON/OFF operation as a transistor verified,” he added.

“In this development, a circularly polarised light 4 was used to generate pure spin current in the semiconductor. However this is just the start. In the future when spin-injection technology for ferromagnetic material is developed, the all solid spintronics devices will be achieved.”

He added that further technology will also help researchers to use a solid device which can control and detect the polarisation of the light. This means that this can be used to open the way for even larger capacity information transmission systems.

Wunderlich said that the technology, which is still in its “early stages” will be used by Hitachi as well as other companies.

Magnetic vortices allow faster data writing and processing

Scientists have developed a method of setting a lattice of magnetic vortices in motion using a very weak electronic current –  that could offer significantly faster data writing and processing.

The experiment by scientists at the Technische Universitaet Muenchen (TUM) and the Universitaet zu Koeln signalled a vast improvement on previous attempts to observe coupling between electric current and magnetic structure through measurements at the research neutron source FRM II in Garching, with a current a million times weaker than in earlier studies.

According to a report published in Science magazine, researchers have been concentrating on how magnetic information can be directly written onto media through the use of electric current.

The problem with doing this has been that it would require extremely high currents, the side effects of which would be impossible to reign in even with nanostructures.

Just over a year ago Professor Christian Pfleiderer alongside his team at TUM had come across a new magnetic structure in a crystal of mangalese silicon – a lattice of magnetic vortices, writes Science Daily.

When Pfleiderer’s team sent electric current through the manganese silicon, using neutrons from FRM II they observed a twist in the magnetic vortex lattice which they were initially unable to explain.  Of even more interest than this was the discovery of the magnetic lattice itself.

The scientists made further measurements at the MIRA instrument of the neutron source FRM II to find out why the lattice would twist when a current was applied.  At first glance the calculations seemed to contradict earlier observations.

“The magnetic structure twists because the direction of the electric current is deflected extremely efficiently by quantum mechanical effects,” said Pfleiderer.

According to Pfleiderer, when an electron flies through a magnetic vortex the electron’s spin reacts to the vortex.  It is in this way that the electric current exerts a force on the magnetic vortices which will eventually begin to flow.

After further measurements the team were able to establish that the lattice of magnetic vortices exhibited interesting properties with relevant to the study of nanotechnology, particularly with regards to data storage systems.

The magnetic vortices are considered to be very stable, while still being very weakly anchored in the material.  This means that even very weak electrical currents can lead to movement, which the researchers believe can lead to the writing and processing of data much faster and more efficiently than has been possible before.