Pitt professor researches quantum computer

Finding a specific gene in the human genome is like looking up a number in a phone book —… Finding a specific gene in the human genome is like looking up a number in a phone book — except that the phone book has no names and is arranged in no particular order.

To assist in genome-mapping projects and other complex computing problems that fascinate researchers across academia, Pitt professor Gurudev Dutt is working to build a quantum-computing machine that can sort large amounts of information quickly.

That work has earned him a pair of prestigious research awards and represents the cutting edge of quantum-computing technology. In February, Dutt received the 2010 Alfred P Sloan Foundation Fellow award, and in 2009 he was selected for the National Science Foundation CAREER Award.

“This is a very prestigious award. For me, personally one of the most rewarding things about being in the field is the knowledge that my peers have recognized that this field is exciting and could have an impact on scientific knowledge,” Dutt said.

Dutt conducts his research in his lab in the Old Engineering Hall along with graduate research assistants Naufer Nusran and Michael Beran and undergraduates Lori Stover and Nicholas Volker.

Their goal is to find a means of computing that can find and sort information exponentially faster and more securely than classical computers. To do this, they are developing a way to store and transmit information in the form of “qubits,” or electrons with a particular spin, rather than in standard bits.

Standard computers use bits, or codes of zeros and ones, to store information. Using qubits presents an advantage because they can store information in zeros, ones and any “superposition” of zero and one, greatly increasing the amount of memory and computing power available.

Using vast quantities of qubits, a quantum computers can have the power to simultaneously explore entire databases, reducing the time it takes to locate specific information. The technology, still in it infant state, also has security and encryption implications.

It was about 30 years ago that the military first began contracting quantum-computing research, Dutt said, as a means to code and crack information.

“One of the big things with quantum computers,” Beran said, “is that some people can use algorithms for factoring really big numbers into all of their primes [thus allowing them to decode information]. Primes and algorithms form the basis for a lot of cryptography, which is how sensitive information is encrypted over the Internet.”

The idea is that if qubits are used to store information, others will not be able to break them down into simple ones and zeros and intercept information.

“If we can develop a quantum network, then information can be sent in absolutely secure networks and no one can eavesdrop,” Dutt said.

The reason for this is that qubits don’t necessarily represent either zero or one, but can represent both zero and one. The numbers zero and one correspond to the spin of the electron — if it’s zero, it spins up, and if it’s one, it spins down.

“But the real reason as to why qubits are appealing is that they can represent both zeros and ones,” Beran said. “It turns out that something that spins up and down at the same time is plausible. That’s how superposition plays into our computers.”

Beran explained how an electron is put into superposition and how the process allows computers to work faster.

“We use sample diamonds that have a defect. A diamond is carbon and the defect is nitrogen, which is a vacancy in the diamond where a bunch of electrons exist,” Beran said. “The first thing we do is make the electronic state of the electron zero.”

The researchers shine a green laser on a diamond for “a fraction of a fraction of a second,” Beran said, and that makes the electrons spin up.

To make the electron spin down, a current is sent through a copper wire on top of the diamond. The oscillating magnetic field interacts with the zero state and the electron “becomes one,” spinning down.

“We work with extremely small times, but if we figure out the amount of time it takes to put the electron to one, and half this time we can get it to a superposition state, somewhere in between zero and one,” Beran said.

With this part of the research complete, the team now focuses on creating quantum networks. Dutt explained that the team is currently working on building a protocol that allows people to exchange information through quantum networks.

“We will use an electron to encode information onto a photon, which is a package of light that can transmit information over a fiber link to another computer,” Dutt said.

All of the applications of this upcoming technology are theoretical, Dutt explained. There are other researchers involved in the software side of quantum computers to develop uses.

“We are at the level that the [first computer scientists] were at 62 years ago when computers filled up rooms,” Dutt said. “Now that technology can all be found in the palm of your hand in a calculator or cellphone.”