Inverting pump-probe spectroscopy for state tomography of excitonic systems

Stephan Hoyer, K. Birgitta WhaleyThe Journal of Chemical Physics 2013 · DOI · arXiv

Abstract

We propose a two-step protocol for inverting ultrafast spectroscopy experiments on a molecular aggregate to extract the time-evolution of the excited state density matrix. The first step is a deconvolution of the experimental signal to determine a pump-dependent response function. The second step inverts this response function to obtain the quantum state of the system, given a model for how the system evolves following the probe interaction. We demonstrate this inversion analytically and numerically for a dimer model system, and evaluate the feasibility of scaling it to larger molecular aggregates such as photosynthetic protein-pigment complexes. Our scheme provides a direct alternative to the approach of determining all Hamiltonian parameters and then simulating excited state dynamics.

Introduction

In ultrafast science, laser pulses are used to observe energy dynamics in molecular aggregates, such as pigment-protein complexes in photosynthesis. While techniques like 2D spectroscopy map these dynamics, directly observing the underlying quantum state—the density matrix—remains a challenge. Traditionally, researchers use "forward simulation": proposing a physical model, estimating Hamiltonian parameters, and iteratively matching simulations to experimental data. This indirect approach is prone to errors if the model is incomplete. This research explores a direct alternative: mathematically inverting spectroscopic data to reconstruct the time-evolving quantum state.

Methods

The proposed inversion protocol consists of two distinct steps:

  1. Signal Deconvolution: Experimental signals are convolutions of the system's response and laser pulse shapes. The authors utilize Tikhonov regularization to "undo" this convolution and extract a pump-dependent response function. This step is purely mathematical, requiring only pulse characterization rather than internal system physics.
  2. Quantum State Inversion: The extracted response function is mapped back to the excited-state density matrix. This requires a model for post-probe system behavior but avoids needing a model for the complex pre-probe energy transfer dynamics.
Comparison of naive and Tikhonov inversion methods for reconstructing the pump-probe response. (a) shows the reconstructed response vs. delay time, while (b) displays error maps for noisy and noise-free conditions.

The protocol was tested using numerical simulations of a dimer model (two coupled pigments) with realistic static disorder and a Markovian bath, and evaluated for scalability using the FMO complex (a 7-pigment system).

Key Results and Insights

The study highlights the necessity of robust regularization. A "naive" approach assuming impulsive (infinitely short) pulses fails to capture fast quantum beats. Tikhonov regularization effectively manages noise and finite pulse durations.

For the dimer model, the Tikhonov-based inversion significantly outperformed naive methods. Under noisy conditions, it reduced the root-mean-squared error (RMSE) by a factor of 12.5. The protocol achieved a state reconstruction fidelity of 99.9% even with realistic static disorder.

Quantitative comparison of RMSE and improvement factors for different inversion protocols under varying noise levels.

Scaling to larger systems like the FMO complex reveals information bottlenecks. While populations (diagonal terms) are robustly reconstructed, extracting coherences (off-diagonal terms) from ensemble measurements is more difficult. Experimental factors like lower temperatures or oriented samples can improve the "conditioning" of the inversion for these larger complexes.

Discussion and Future Work

The primary challenge for this method is "static disorder"—the inherent variability in bulk samples. High disorder can break the assumption that the signal is a simple product of average response and state. Additionally, "dark states" that do not interact with light remain invisible to optical probes.

Future work aims to extend this framework to 2D spectroscopy, which offers more information axes and could enable full state tomography for larger aggregates. This approach could also serve to verify quantum control experiments designed to steer energy pathways.

Summary updated

Abstract

We propose a two-step protocol for inverting ultrafast spectroscopy experiments on a molecular aggregate to extract the time-evolution of the excited state density matrix. The first step is a deconvolution of the experimental signal to determine a pump-dependent response function. The second step inverts this response function to obtain the quantum state of the system, given a model for how the system evolves following the probe interaction. We demonstrate this inversion analytically and numerically for a dimer model system, and evaluate the feasibility of scaling it to larger molecular aggregates such as photosynthetic protein-pigment complexes. Our scheme provides a direct alternative to the approach of determining all Hamiltonian parameters and then simulating excited state dynamics.
BibTeX
@article{hoyer2013inverting,
  author = {Stephan Hoyer and K. Birgitta Whaley},
  title = {{Inverting pump-probe spectroscopy for state tomography of excitonic systems}},
  journal = {The Journal of Chemical Physics},
  year = {2013},
  doi = {10.1063/1.4800800},
}
Figure