Light could make semiconductor computers a million times faster or even go quantum

May 10, 2018 by Katherine Mcalpine, University of Michigan
An illustration showing the “up” and “down” pseudospin states, a light pulse and the hilly energy landscape experienced by the electrons. Credit: Stefan Schlauderer, University of Regensburg

A technique to manipulate electrons with light could bring quantum computing up to room temperature.

A team of researchers in Germany and at the University of Michigan have demonstrated how can shift electrons between two different , the classic 1 and 0, in a thin sheet of semiconductor.

"Ordinary electronics are in the range of gigahertz, one billion operations per second. This method is a million times faster," said Mackillo Kira, U-M professor of electrical engineering and computer science.

He led the theoretical part of the study, to be published in the journal Nature, collaborating with physicists at the University of Marburg in Germany. The experiment was done at the University of Regensburg in Germany.

Quantum computing could solve problems that take too long on conventional computers, advancing areas such as artificial intelligence, weather forecasting and drug design. Quantum computers get their power from the way that their quantum-mechanical bits, or qubits, aren't merely 1s or 0s, but they can be mixtures— known as superpositions—of these states.

"In a classical computer, each bit configuration must be stored and processed one by one while a set of qubits can ideally store and process all configurations with one run," Kira said.

Credit: University of Michigan

This means that when you want to look at a bunch of possible solutions to a problem and find the best fit, quantum computing can get you there a lot faster.

But qubits are hard to make because quantum states are extremely fragile. The main commercial route, pursued by companies such as Intel, IBM, Microsoft and D-Wave, uses superconducting circuits—loops of wire cooled to extremely cold temperatures (-321°F or less), at which the electrons stop colliding with each other and instead form shared quantum states through a phenomenon known as coherence.

Rather than finding a way to hang onto a for a long time, the new study demonstrates a way to do the processing before the states fall apart.

"In the long run, we see a realistic chance of introducing quantum information devices that perform operations faster than a single oscillation of a lightwave," said Rupert Huber, professor of physics at the University of Regensburg, who led the experiment.  "The material is relatively easy to make, it works in air, and at just a few atoms thick, it is maximally compact."

The material is a single layer of tungsten and selenium in a honeycomb lattice. This structure produces a pair of electron states known as pseudospins. It's not the spin of the electron (and even then, physicists caution that electrons are not actually spinning), but it is a sort of angular momentum. These two pseudospins can encode the 1 and 0.

An artist’s rendering of a pulse of circularly polarized light hitting a 2-D semiconductor, putting the electrons into a pseudospin state that could store information as part of a new, faster computing technology. Credit: Stephen Alvey, Michigan Engineering

Huber's team prodded electrons into these states with quick pulses of infrared light, lasting just a few femtoseconds (quintillionths of a second). The initial pulse has its own spin, known as , that sends electrons into one pseudospin state. Then, pulses of light that don't have a spin (linearly polarized) can push the electrons from one pseudospin to the other—and back again.

By treating these states as ordinary 1 and 0, it could be possible to create a new kind of "lightwave" computer with the million-times-faster clock speeds that Kira mentioned. The first challenge along this route will be to use a train of laser pulses to "flip" the pseudospins at will.

But the electrons can also form superposition states between the two pseudospins. With a series of pulses, it should be possible to carry out calculations until the electrons fall out of their coherent state. The team showed that they could flip a qubit quickly enough to execute a string of operations—basically, it's fast enough to work in a quantum processor.

Moreover, the are constantly sending out light that makes it easy to read a without disturbing its delicate state. Clockwise circular polarization indicates one pseudospin state, counterclockwise the other.

The next steps toward will be to get two qubits going at once, near enough to one another that they interact. This could involve stacking the flat sheets of semiconductor or using nanostructuring techniques to fence off qubits within a single sheet, for example.

Explore further: Ultrashort light pulses for fast 'lightwave' computers

More information: F. Langer et al. Lightwave valleytronics in a monolayer of tungsten diselenide, Nature (2018). DOI: 10.1038/s41586-018-0013-6

Related Stories

New input for quantum simulations

January 22, 2018

An international group of researchers, including UvA physicist Michael Walter, have devised new methods to create interesting input states for quantum computations and simulations. The new methods can be used to simulate ...

Simple is beautiful in quantum computing

November 15, 2017

Quantum computing could solve problems impossible for today's supercomputers. The challenge for this new form of computing is processing the quantum bits (qubits) that represent data. A qubit can be made by controlling the ...

A new kind of quantum computer

November 6, 2017

Quantum mechanics incorporates some very non-intuitive properties of matter. Quantum superposition, for example, allows an atom to be simultaneously in two different states with its spin axis pointed both up and down, or ...

New silicon structure opens the gate to quantum computers

December 12, 2017

In a major step toward making a quantum computer using everyday materials, a team led by researchers at Princeton University has constructed a key piece of silicon hardware capable of controlling quantum behavior between ...

Recommended for you

Ytterbium: The quantum memory of tomorrow

July 23, 2018

Quantum communication and cryptography are the future of high-security communication. But many challenges lie ahead before a worldwide quantum network can be set up, including propagating the quantum signal over long distances. ...

Physicists demonstrate new method to make single photons

July 23, 2018

Scientists need individual photons for quantum cryptography and quantum computers. Leiden physicists have now experimentally demonstrated a new production method. Publication in Physical Review Letters on July 23rd.

Uncovering the interplay between two famous quantum effects

July 23, 2018

The Casimir force and superconductivity are two well-known quantum effects. These phenomena have been thoroughly studied separately, but what happens when these effects are combined in a single experiment? Now, Delft University ...

Researchers report success with complex quantum states

July 23, 2018

Scientists from the Niels Bohr Institute at the University of Copenhagen have, for the first time, succeeded in producing, controlling and understanding complex quantum states based on two electron spins connected to a superconductor. ...

Current noises of Majorana fermions

July 23, 2018

Majorana fermions are particles that are their own antiparticles. In condensed matter physics, zero-energy Majorana fermions obey non-abelian statistics, and can be used in fault-tolerant topological quantum computation. ...


Adjust slider to filter visible comments by rank

Display comments: newest first

3.7 / 5 (3) May 10, 2018
You mean like "Photon Induced Electric Field Poling ?, like the patents granted many 18 years ago ?

Like Quantum Holographic Optical Data Storage that is non-volatile, has spintronics capability, has quantum communications capability, etc.
3 / 5 (2) May 10, 2018
Anything more than 15 years old gets regularly reinvented
1 / 5 (3) May 10, 2018
Anything more than 15 years old gets regularly reinvented

Really, and can you teach a 50 year pioneer of many technologies he already know's about ?

Does that mean after 15 years anyone can steal personal IP property ?

I guess fusion research with $ 100's of billions and 50 years, if also me too technology ?
3 / 5 (2) May 11, 2018
It seems Moore's law has a new lease on life. Perhaps we can define 10 years as the time that fusion energy will be successful, the US economy will crash, the world will end and Moore's law will have run its course.
3.7 / 5 (3) May 11, 2018
The disparity of electron reaction velocity and photon velocity

Manipulating electrons at absolute zero prodding them with laser light your restricted in the reactions of the electron which does not physically move at the speed of light plus the fact computers at absolute zero won't make much headway in the commercial world of computers, tablets and mobile phones attached to a cryogenic device.

As these computers have to operate at room temperature to have practical applications I can't see the point of carrying out these experiments at absolute zero now the proof of concept has been proved. But it's still not possible to square the photon and electrons reaction time where in the final interface with the human operator all these photonic signals are converted to conventional circuits with their processors to realise their calculations in the macro world as we input the data where conventional circuits slows the possible advantages to a snail's pace.
not rated yet May 11, 2018
I chose this article to beat the bongos for Richard Feynman 100th birthday, after all he was the first to talk about quantum computing. I find it quite sad that there is not much emphasys on the web for this event... this man realy deserve to be celebrated. Happy birthday Dick Feynman :-)
not rated yet May 11, 2018
I have a patent on a system of this type, which was granted 20 years ago. I utilized an image converter to address multiple electrical sensors simultaneously that are withing a square area. This is faster than today's parallel processors. Unfortunately, the project was evidently abandoned after I retired.

Please sign in to add a comment. Registration is free, and takes less than a minute. Read more

Click here to reset your password.
Sign in to get notified via email when new comments are made.