Cold Ion Group @ CQT

LATEST RESULTS

Single proton cooled by distant ions

01.09.2021

Laser-cooled ions have been used to substantially lower the temperature of a proton located several centimetres away. This technique could be useful in ultra precise measurements of the properties of antimatter particles.

Nature 596, 490-491 (2021)

This will open new possibilities to couple distant quantum systems using conducting wires

Fig 1: How good a classifier works depends on how well it can predict the class of a point within a boundary. In the above three figures each colour corresponds to a class determined by the quantum machine. If the color falls within its boundary, the prediction is correct. The boundaries chosen to test the machine are considered to be hard to solve classically. In each case the quantum machine is accurate to more than 95% when trained with more than 4 layers. This is way better than a neural network in terms of the gate counts as it doesn't need sigmoid evaluation which is resource hungry.

A single atom noise probe operating beyond the Heisenberg limit

08.01.2020

According to the Heisenberg uncertainty principle, the energy or frequency uncertainty of a measurement can be at the best inversely proportional to the observation time (TT). The observation time in an experiment using a quantum mechanical probe is ultimately limited by the coherence time of the probe....

DETAILS

Single atom energy-conversion device with a quantum load

01.05.2020

We report on a single atom energy-conversion quantum device operating as an engine, or a refrigerator, coupled to a quantum load. The “working fluid” consists of the two optical levels of an ion, while the load is one of its vibrational modes, cooled down to the quantum regime. We explore two important differences with classical engines: (i) the presence of a strong generic coupling interaction between engine and load which can induce correlations between them; (ii) the use of non-thermal baths. We examine the ergotropy of the load, which indicates the maximum amount of energy of the load extractable using solely unitary processes. We show that ergotropy rises with the number of engine cycles despite an increase in the entropy of the load. The increase of ergotropy of the load points to the possibility of using the phonon distribution of a single atom as a form of quantum battery.

Quantum Machine Learning: quantum classifier

01.07.2021

Data classification is a ubiquitous problem, currently addressed by classical machine learning. Alternative quantum approach based on classical support vector classifier is promising but with limited applicability. Here, we experimentally demonstrate a general quantum classifier to be as competitive as a neural network, yet far more resource and energy efficient than it, in a resource limited setting. Current solutions to large data classification is resource intensive and hence our contribution leveraging on the complexity of Hilbert space, will make a paradigm shift on the development of quantum classifier for large data.

Defect generation and dynamics during quenching in finite size homogeneous ion chains

01.02.2020

An equally spaced linear chain of ions provides a test-bed for studying the defect formation in a finite size 1D system. In particular, defect formation related to topological phase transition from a linear configuration to a zig-zag one is of interest here. A semi-empirical expression provides an excellent agreement to the numerical results. The non-adiabatic transition between the chain and zig-zag topologies for a finite size system of 30 ions shows clear distinction from non-uniformly distributed ion chain. Thus the underlying Homogeneous Kibble-Zurek model can be tested in presently accessible ion trap experiments. Furthermore, our study indicates collective defect behaviour appearing through the correlation length measurements.

DETAILS