Talks

Optimizing leapover lengths of Lévy flights with resetting

Large Deviations Beyond the Kibble-Zurek Mechanism

Crossing a quantum phase transition in finite time leads to the formation of excitations, such as topological defects, since the dynamics necessarily fails to be adiabatic near the critical point. The average number of excitations is well described by the celebrated Kibble-Zurek (KZ) mechanism, predicting a universal scaling law with the quench time. Recently, the scope of the KZ paradigm has been expanded, enabling the prediction of quantities beyond averages, such as the full counting statistics of defects. In this talk, I will present some results that clarify the role of universality in beyond-KZ physics, by borrowing tools from Large Deviations Theory. Using the transverse-field Ising model as test bed, I will show how the rate function obeys a universal scaling relation with the quench time. Then, I will expand the result to classical phase transitions, using few additional assumptions on the way defects form. I will finally argue how these theoretical predictions are already testable in current quantum simulators and annealers. FB, M. Beau, J. Yang, A. Gambassi and A. del Campo, Phys. Rev. Lett. 131, 230401 (2023)

CondMat for Dummies: tba

Chemomechanical theory of active solids

IMPRS Seminar: Connecting Ultrafast Spectra to Ultrafast Dynamics

Ultrafast laser pulses allow scientists to watch the wave packet motion of a molecule or material following photoexcitation with incredible time resolution. However, no ultrafast experiment can take a perfect snapshot of the molecular wave packet. Instead, spectroscopic experiments record a lossy projection of the molecular wave packet. For this reason, computer simulation has become an essential partner of ultrafast experiment. Computer simulations can provide a detailed, if approximate, picture of the motions following photoexcitation, explicitly including all nuclear and electronic degrees of freedom. Working in collaboration with ultrafast spectroscopists, our group has developed novel strategies for simulating and interpreting ultrafast spectra, with an eye toward definitively assigning spectral features to specific molecular motions. In this talk, I will present two stories of such work. First, in collaboration with the Allison group at Stony Brook University, we have developed a novel computational strategy for the direct simulation of ultrafast transient absorption spectra. We will demonstrate how this method may be used to assign spectral features to individual molecular motions in two molecules that undergo excited state proton transfer. The second story will present our work with the group of Warren Beck at Michigan State University interpreting the two-dimensional electronic spectrum of colloidal CdSe quantum dots undergoing hot carrier cooling, a process that limits the efficiencies of solar cells. The Beck group’s experiments indicate the existence of coherences between core electronic excitations and ligand vibrational states during relaxation. By identifying and analyzing conical intersections between potential energy surfaces of quantum dots, we are able to assign the spectral features to specific motions and identify the nature of the observed vibronic coherences.

IMPRS Seminar: Tracing the fastest chemical reactions with X-ray photons: Focusing on proton transfer

The development of laser technologies has opened the way to femtochemistry – tracing chemical reactions with a femtosecond laser pulse. The Nobel Prize for this work was already awarded in 1999. However, the typical time resolution was only possible within hundreds of femtoseconds, which is not enough to capture the fastest chemical processes. The ultimate time limit for chemical transformations, i.e. changing atomic nuclei, is only becoming possible with recent developments in high harmonic generation (HHG) or X-ray free electron lasers. In my presentation, I will discuss theoretical calculations needed for the interpretation of very fast non-adiabatic processes observed with the reaction microscope approach: one or two pulses excited the molecule and the ions formed are traced in coincidence. As a first example, I will focus on ultrafast acidity in a water dimer. The ionization of hydrogen-bonded systems gives rise to highly acidic cationic species, which then relax via Proton transfer (PT), clearly within the sense of Brønsted–Lowry acid theory. In particular, the water molecule forms a radical cation H2O·+, which acts as an extremely strong acid. I will present the calculation of the dynamics in different electronic states and compare them with recent experiments.[1] This work complements the different types of experiments in the liquid phase. [2] Next, I will focus on the more complicated dynamics of the doubly ionized pyrrole-water complex. Here, we either observe the process in the energy domain, using a single electron pulse or we interpret the time-resolved experiment, starting with a strong field ionization followed.[3] Finally, I will briefly discuss a more complicated X-ray pump/X-ray probe experiment on a water dimer. Here, the induced dynamics interfere with spontaneous Auger decay which makes the simulations even more challenging. [4] [1] Schnorr K., Belina M., et al., Direct tracking of ultrafast proton transfer in water dimers, Sci. Adv., 2023, 9, eadg7864, doi: 10.1126/sciadv.adg7864 [2] Loh Z.-H., et al., Observation of the fastest chemical processes in the radiolysis of water, Science, 2020, 367, 179-182 [3] Zhou J., Wu L., Belina M., et al., Ultrafast proton transfer between water and pyrrole induced by localized double ionization esses in the radiolysis of water, in preparation [4] Jahnke T., Belina M., et al., Inner-Shell-Ionization-Induced Femtosecond Structural Dynamics of Water Dimer Imaged at an X-Ray Free-Electron, in preparation The development of laser technologies has opened the way to femtochemistry – tracing chemical reactions with a femtosecond laser pulse. The Nobel Prize for this work was already awarded in 1999. However, the typical time resolution was only possible within hundreds of femtoseconds, which is not enough to capture the fastest chemical processes. The ultimate time limit for chemical transformations, i.e. changing atomic nuclei, is only becoming possible with recent developments in high harmonic generation (HHG) or X-ray free electron lasers. In my presentation, I will discuss theoretical calculations needed for the interpretation of very fast non-adiabatic processes observed with the reaction microscope approach: one or two pulses excited the molecule and the ions formed are traced in coincidence. As a first example, I will focus on ultrafast acidity in a water dimer. The ionization of hydrogen-bonded systems gives rise to highly acidic cationic species, which then relax via Proton transfer (PT), clearly within the sense of Brønsted–Lowry acid theory. In particular, the water molecule forms a radical cation H2O·+, which acts as an extremely strong acid. I will present the calculation of the dynamics in different electronic states and compare them with recent experiments.[1] This work complements the different types of experiments in the liquid phase. [2] Next, I will focus on the more complicated dynamics of the doubly ionized pyrrole-water complex. Here, we either observe the process in the energy domain, using a single electron pulse or we interpret the time-resolved experiment, starting with a strong field ionization followed.[3] Finally, I will briefly discuss a more complicated X-ray pump/X-ray probe experiment on a water dimer. Here, the induced dynamics interfere with spontaneous Auger decay which makes the simulations even more challenging. [4]

Colloquium

Title t. b. a.

Title t. b. a.

Title t. b. a.

QDS

Colloquium: Lighting up Superconductivity

t.b.a.

Colloquium

QDS

t.b.a.

Title t. b. a.

QDS: Can quantum computing enhance machine learning and, if yes, how?

I will begin by demonstrating that the answer to the first question in the title is yes [1], in principle. I will then discuss if the quantum advantage of quantum machine learning can be exploited in practice. To discuss how to build optimal quantum machine learning models, I will describe our recent work [2-3] on applications of classical Bayesian machine learning for quantum predictions by extrapolation. In particular, I will show that machine learning models can be designed to learn from observables in one quantum phase and make predictions of phase transitions as well as system properties in other phases. I will also show that machine learning models can be designed to learn from data in a lower-dimensional Hilbert space to make predictions for quantum systems living in higher-dimensional Hilbert spaces. I will then demonstrate that the same Bayesian algorithm can be extended to design gate sequences of a quantum computer that produce performant quantum kernels for data-starved classification tasks [4]. [1] J. Jäger and R. V. Krems, Universal expressiveness of variational quantum classifiers and quantum kernels for support vector machines, Nature Communications 14, 576 (2023) [2] R. A. Vargas-Hernandez, J. Sous, M. Berciu, and R. V. Krems, Extrapolating quantum observables with machine learning: Inferring multiple phase transitions from properties of a single phase, Physical Review Letters 121, 255702 (2018) [3] P. Kairon, J. Sous, M. Berciu and R. V. Krems, Extrapolation of polaron properties to low phonon frequencies by Bayesian machine learning, Phys. Rev. B 109, 144523 (2024). [4] E. Torabian and R. V. Krems, Compositional optimization of quantum circuits for quantum kernels of support vector machines, Physical Review Research 5, 013211 (2023)

IMPRS Seminar

IMPRS Seminar

Colloquium