Calculating Vibronic Spectra with a linear algorithm based on Gaussian Boson Sampling

TL;DR

This paper introduces a scalable linear algorithm based on Gaussian Boson Sampling for simulating molecular vibronic spectra, achieving high accuracy and experimental validation for complex molecules.

Contribution

The work presents a novel, scalable Gaussian Boson Sampling algorithm for vibronic spectra simulation, validated through classical and experimental approaches on large molecules.

Findings

Achieved high fidelity $(F>0.999)$ between simulated and analytical spectra.

Successfully simulated larger molecules like naphthalene and anthracene.

Experimental results confirmed the accuracy of the simulated spectral features.

Abstract

Accurately simulating molecular vibronic spectra remains computationally challenging due to the exponential scaling of required calculations. Here, we show that employing the linear coupling model within the gaussian boson sampling framework effectively addresses this limitation. We implement the algorithm for simulating the pentacene molecule through three distinct approaches, using numerical simulation on a classical computer and experimentally using two optical setups equipped with different photon detectors (SNSPD and SPAD). High fidelity $(F>0.999)$ was achieved between the simulated Franck-Condon profiles and analytically calculated profiles obtained by enumerating all possible transitions within the linear coupling model. Furthermore, simulations were performed for larger molecular systems using 48 vibrational modes of naphthalene and 64 vibrational modes of anthracene. Comparison with experimental data confirms that the simulated spectra accurately reproduce both the positions and shapes of the measured spectral bands. A notable advantage of our algorithm is its scalability, requiring only a fixed minimal set of optical components irrespective of the size of the studied system.