Imagine that you run a business. Researchers in your R&D department are working day and night on a device that could change the world. But suddenly they face a serious obstacle: for the machine to work as it should, a material with very specific properties must be used for its construction. And that material does not exist.

So that it, all is lost and the path of progress blocked once and for all? Not exactly. The required material can simply be designed and manufactured. But the road to achieving this is quite complicated.

"It`s not that someone can come to a lab, mix two components and hope that they will get a new material that, for example, expands by 5 mm when temperature rises, and shortens when it drops" - Dr. Mariusz Semczuk from the Faculty of Physics, University of Warsaw says in interview with PAP.

Before a new material can be produced, it`s a good idea to run computer simulations to avoid costly errors. They can help design the material that will meet your expectations. Improving such simulations is the objective of Dr. Semczuk`s research.

The scientist explains that in many cases the necessary calculations are so complicated that the computing power of classic computers is not enough to carry them out. Researchers have high hopes for quantum computers - machines that would use the laws of quantum mechanics for computing. This would significantly increase their valuable computing power.

Fully functioning quantum computers are still only in the planning phase, but Dr. Semczuk`s scientific interests concentrate around them.

"I want to take the first step: prepare a platform, bases on which it will be possible to develop quantum simulators" - he explains. "These devices can be compared to a computer with enormous power, designed to perform one specific task, for example create weather simulations for a given area" - he says. The task of quantum simulators is to calculate the quantum properties of materials.

Cold atoms are serious candidates for use in simulators and - later - in quantum computers. But the question is: can simulators using cold atoms operate efficiently and with little human interference? "Modern atom cooling methods are very complicated. This is a problem from the point of view of possible commercial applications - I suspect that companies would not be willing to employ teams of physicists just to keep their computers ready" - he says.

"In my project, I would like to significantly reduce the time needed to prepare atoms in a quantum simulator, the so-called initialisation" - says Dr. Semczuk. "The standard procedure leading to obtaining the result by the simulator consists of several stages, and the initialisation of these atoms takes 5-10 seconds - while the measurement process itself often takes much less than 500 milliseconds! Remember that for this laboratory to be useful, such measurements must be made hundreds or thousands of times. In this context, these few seconds at the beginning are a waste of time" - he emphasises.

To better illustrate the role of the long initialisation time in the quantum simulator, Dr. Semczuk proposes to imagine that you need to check something on your computer - but the computer has been turned off. "So in order to get access to it, you need to start the computer, open the file, take a look at it, and then turn off the computer. Looking at the file might only take five seconds, but you have to wait for the computer to start, and that can take a few minutes" - he says.

"I expect that I will be able to reduce preparation time to 500 milliseconds - which is already comparable with the duration of subsequent simulations" - adds Semczuk.

But what about atoms that are already cooled? Scientists plan to put them in an optical network. It consists of intersecting laser beams that interact with each other - and individual atoms can be placed in the so-called interference maxima (grid meshes). "It turns out that the atom in a single mesh of an optical network behaves like an electron in crystals: from a mathematical point of view, its behaviour is the same, so it is an artificial crystal".

If the designed material is supposed to be a crystal, we could check its properties before actually making it.

"In the case of real crystals, we are limited by certain specific physical properties of atoms and the interactions they impose. Take distances between atoms. Yes, they can be modified, but it is definitely not a trivial thing. Which is why it`s a good idea to check how the material will behave after changing one of its parameters - and that`s what simulations in optical networks can be used for. It is easy change a parameter in a simulation, for example by changing the angle of intersection of beams that form the optical network" - the researcher concludes.

PAP - Science in Poland

author: Katarzyna Florencka

kflo/ ekr/ kap/

tr. RL