Imprimir Republish


Three steps forward

Physicists from Minas and Sao Paulo improve data handling and transmission in quantum computers

In three recent studies, researchers from the states of Minas Gerais and São Paulo have presented theoretical and experimental contributions that should help in the development of a special type of computer that has been circulating through the minds of physicists for three decades, even since the chemist Charles Bennett, from the major computer company IBM, demonstrated that it was possible to make use of atomic particle characteristics for processing information. This is the quantum computer, so named as it works according to the laws of quantum mechanics, an area of physics that studies the phenomena of the world of atoms and molecules.

The result of the most immediate practical application came through the work of the physicist José Maria Villas-Bôas, an ex-student of professor Nelson Studart from the Federal University of São Carlos (UFSCar), in the interior of Sao Paulo, who is currently carrying out his post-doctorate studies at the University of Ohio, United States. Villas-Bôas has discovered a simple solution for the errors in one of the nanoscopic systems, at the level of one millionth of a millimeter, most quoted to make up part of the processor of these future computers: the quantum  dots; pyramids or hemispheres billions of times smaller than the point of a needle, created upon semi-conductor material.

The look of quantum computers is not yet known, but physicists believe that the main  change will occur in the structure of the processor and in the form of dealing with the information technology units, named bits. In classical computers the processors are silicon chips the size of a coin, holding up to 400 million transistors. When the computer executes a command, the transistor allows or blocks the passage of electricity and the information is codified in a two-number system, zero or one. Substituting transmitters, quantum computers must make use of dozens or hundreds of quantum dots; atoms or packets of light (photons). And there will be advantages. While the transistor deals with one piece of information at a time, in an exclusive relationship, the quantum processor works simultaneously with innumerable physical states symbolized by infinite combinations of the probability of being zero or one: for example, 99% chance of being zero and 1% of being one or 42% of being zero and 58% of being one. This is the quantic unit of information: the quantic bit or qubit.

In order to carry out calculations, the physicists attribute arbitrary values to the properties of the atomic particles, such as the vibrational plane of the electric field of photons in a laser. An example helps one to understand. The electric field of photons oscillating in the vertical plane, somewhat like a rope being pulled up and down vertically by children, can be determined and corresponds to the ‘zero’ state, while oscillating in the horizontal direction corresponds to the ‘one’ state. In accordance with a property in the world of particles called superposition of quantum states, the photons can vibrate in infinite directions at the same time. This property guarantees to the quantum processor an unequalled agility to deal with different information at the same time, and in theory, elevate to infinity the processing capacity of a bunch of atoms.

Currently there are at least two proposals for the use of quantum dots points in order to carry out logical operations. Firstly, imprisoning a single particle of negative electrical charge (an electron) in the interior of these nanoscopic structures, and thereafter attempting to control the rotational direction of this electron with the help of electromagnetic fields. However, the apparently more visible alternative is to bombard the quantum dots with rapid pulses from a laser whose photons vibrate with more energy than the electron.

In this interaction, the laser transfers energy to the electron, which jumps out of the region in which it is found to another that is more energetic in the interior of the quantum dots, structures of between 2 to 50 nanometers. As a consequence, the region previously occupied by the electron remains empty and has a positive charge – the stable combination of the excited electron with the vacant region (hole) makes up a state that the physicists call exciton. If at this time the laser should hit the excited electron, the electrically negative charged particle returns to the region of least energy of the quantum dots and the grouping returns to its original or fundamental state.

It was this possibility of creating these distinct states – one fundamental and the other excited – that led physicists to propose the quantum dots as an alternative to the current processor. However, there are difficulties. As the quantity of electrical current generated by a single electron is low, the bombardment with the laser needs to be repeated various times until a measurable current is produced. And it is during this phase that the problems pop up. Artur Zrenner, from the University of Paderborn, in Germany, has verified that this repeated bombardment produces interference that impedes the precise reading of the codified information of the quantum dot’s energy state and described this impediment in an article published in Nature during 2002. In comparison with the macroscopic world, it is as if one needs to look many times at a person to know if he is wearing a hat, but with each look a cloud of smoke forms in front of our eyes, impeding us from seeing clearly.

Faced with this result, Villas-Bôas and the physicists Sergio Ulloa and Alexander Govorov, both from the University of Ohio, began to search for explanations for this undesirable interference, similar to the background noise that occurs in FM radio reception when one passes through a city area full of broadcasting equipment. And they found it in the origin of the quantum dots: within the fine layer upon which these structures are formed. Made from the same semiconductor material as the quantum dots – a mixture of gallium arsenide and indium arsenide –, this layer shows regions in which excited electrons, with more energy than in the interior of the quantum dots, can come about, affecting the intensity of the produced electric current, as described by Villas-Bôas, Ulloa and Govorov in Physical Review Letters of 11th February.

How to get round the problem? Simple: just bombard the quantum dots with laser pulses that are less intense and more prolonged, the researchers propose. And the use of less intense pulses reduces the probability of generating excited electrons with a higher energy than in the layer below the quantum dots. And it seems to work. “Last year, Artur Zrenner spoke with me after I had presented this study at the Quantum Dot Conference, in Canada”, Villas-Bôas tells. “Even without knowing about my study, he had redone the experiments with longer laser pulses and obtained better results, but did not know how to explain the results.”

Symmetrical pathways
In parallel with the progress on the prototypes for a quantic processor, physicists from the Federal University of Minas Gerais (UFMG) and the State University of Campinas (Unicamp) have demonstrated two other relevant advances: they have found ways of increasing the capacity of the processing and transmission of a quantum computer’s information.
At the UFMG, Sebastião Pádua, Leonardo Neves, Gustavo Lima and Carlos Monken have developed and tested an ingenious strategy that allows for an increase in the quantity of information associated with each quantum dot .In collaboration with José Aguirre and Carlos Saavedra, from the University of Conception, in Chile, Pádua’s team associated the information to another inert property of photons: the path traced out by these light packets.

There is no magic. And with a little imagination one can understand the team’s proposal. On passing through a special type of crystal, the laser beam is transformed into two twin photon beams, which spread themselves in different directions, with symmetrical angles in relation to the initial trajectory. An integral property of quantum physics called quantum entanglement guarantees that two distinct and separate particles – or even two groupings of particles in the case of twin beams – will react in a predetermined manner when one of them receives a stimulus.

Pádua’s team directed each one of the twin beams to a different screen, at 20 centimeters from the crystal, and with four very narrow slits, of 0.09 millimeters. When producing the twin photon beams the physicists programmed them to comply with the following demand: on leaving the crystal the light packets would pass through symmetrically opposite slits. Thus if a photon of the right were to pass through the highest of the four slits, then the one on the left would by obligation pass through the lowest of the left screen. As well as the codified information in the polarization plane, now it is possible to sum other four pieces of information linked to the pathways that the photon might take.

And the greater the number of slits, the more the information that will be coupled to the twin beams. Experiments with screens of 4 and 8 slits, described by the Chilean and the Minas Gerais team in Physical Review Letters of March 18th, show that the strategy is viable and the rate of getting it right is high: at least 96%. The calculations indicate that it is possible to obtain good results with up to 10 slits.

One can argue that screens with slits are not the best material for integrating a quantum processor. But what is interesting is the working principle. “Imagine that in the place of slits we had optical fibers”, proposes Pádua. “This simple substitution would allow us to transport more information using fewer light pulses.”

A single package
The author of the third contribution is the physicist Gustavo Rigolin, from Unicamp. Taking into consideration the particularities of the quantum entanglement, he proposed a way out of one of the quantum computing bottlenecks: the transmission of information. There is no point in having a super efficient processor capable of carrying out in seconds calculations that would take billions of years in a classical computer if the results have to be transferred one by one to the location where they will be stored.

Almost twenty years after having revealed the possibility of using atomic particles to carry out calculations, in 1993 Charles Bennett identified a surprising property of quantum physics: teleportation, the capacity to transmit the characteristics of an atomic particle to another at a distance. Until recently the efficiency of teleportation was low, because one could only manage to transmit the characteristics of a single particle at a time. In an article in Physical Review A, Rigolin proposed procedures that would allow sending simultaneously innumerable quantum states of a group of particles to another.

Imagine that you want to transfer the information from one hundred electrons located at the Praça de Sé Cathedral in the center of Sao Paulo, to another one hundred electrons at Candelária, in the central region of Rio de Janeiro. Rigolin discovered that one could only manage to transmit the characteristics of the particles from São Paulo to the Carioca particles if a further one hundred intermediary electrons were to be available. On interlacing the intermediary particles with the São Paulo particles, both then share the same characteristics. Next, the intermediary particles function as a quantum bridge or quantum  canal and they transfer their properties to the electronsin Rio de Janeiro. As well as increasing the capacity of simultaneously transmitting information, this model allows for the correction of eventual errors in the transmitted information and for the creation of more efficient safety codes, which would denounce any attempt at intercepting the information.