The quantum process that decides whether a solar cell captures or wastes absorbed light now has a face. Physicists at the University of Graz in Austria, Philipps-Universität Marburg in Germany, and Forschungszentrum Jülich in Germany published findings on August 28, 2026 in Physical Review X showing that they directly filmed — for the first time — the full quantum-mechanical structure of the particle responsible for converting light into electricity, and watched it shrink by roughly 25 percent in less than 400 femtoseconds (a femtosecond is one quadrillionth of a second). The University of Graz announced the result on September 1, 2026.
The particle in question is called an exciton — a quantum-mechanical pair formed when light knocks an electron out of its normal resting state and leaves behind a positively charged vacancy, or “hole.” Electron and hole attract each other through the same Coulomb force that holds electrons in atoms, forming a neutral bound pair that carries energy but not net charge through the material. In organic solar cells, excitons are everything: they are the first product of absorbed sunlight, and the efficiency with which they travel to the interface where they can split apart into free electricity-carrying charges determines how much of that sunlight becomes current. Every exciton that self-destructs before reaching that interface is wasted energy.
Until this paper, no one had ever directly measured the full quantum-mechanical structure — the “wave function” — of an exciton in an organic semiconductor. Researchers could infer energy levels, lifetimes, and diffusion distances through indirect spectroscopic probes, but the exciton’s actual spatial shape, its internal phase structure, and how both evolved over time were accessible only through theoretical models. The new work, led by Marcel Theilen and colleagues, changes that fundamental limitation.
Before understanding what was measured, it helps to understand what a wave function is and why it matters. In quantum mechanics, a particle’s wave function is a mathematical description of the probability distribution of its possible states — where it is likely to be found, how it is moving, and in the case of complex particles like excitons, how the quantum state varies from location to location within the material. The phase of a wave function is analogous to the crest-and-trough pattern of an ocean wave — it describes the relative quantum-mechanical “orientation” of the electron-hole pair across adjacent molecules in the material, and it determines how the exciton interacts with its surroundings and whether it can coherently spread across multiple molecules or must collapse to a single one.
In organic semiconductors, this distinction — between an exciton that is delocalized (spread across several molecules) and one that is localized (confined to one) — has enormous practical consequences. Delocalized excitons can travel farther and reach charge-splitting interfaces more easily; localized ones tend to self-trap and dissipate their energy as heat. The Graz–Marburg–Jülich experiment directly measured both the spatial extent and the phase structure of a newborn exciton in a model organic semiconductor, alpha-sexithiophene (α-6T), and tracked both as the exciton aged.
The technique the team used is called femtosecond time-resolved photoemission orbital tomography (trPOT). It works in two steps separated by a precisely controlled time delay.
First, an ultrashort laser pulse — just 45 femtoseconds long — strikes a thin film of α-6T held at 24 Kelvin (−249°C or −416°F) inside an ultra-high vacuum chamber, creating an exciton. The cryogenic temperature suppresses thermal noise that would otherwise smear the quantum signal; the vacuum prevents surface contamination that would quench the exciton before it could be observed.
Second, after a variable delay, a second high-energy laser pulse at 21.7 electron-volts — generated by a nonlinear optical process called high-harmonic generation — ejects a photoelectron from the exciton. The energy and three-dimensional momentum direction of that ejected electron are recorded by a time-of-flight momentum microscope, an instrument that captures the full angular distribution of emitted electrons simultaneously. Varying the delay between pump and probe pulses yields different temporal snapshots; strung together, these form a quantum-mechanical film of the exciton’s evolution.
The critical breakthrough came from Graz: the team, led by Peter Puschnig, developed a quantitative theoretical model that translates the momentum-space images into the real-space wave function of the exciton. “Using our model, it is possible to deduce the spatial shape and the internal quantum-mechanical phase of the exciton wave function directly from measured photoelectron images,” explained Siegfried Kaidisch, a PhD student who made key contributions to the theory work.
This is a tabletop experiment — the high-energy probe pulse comes from a laboratory laser system rather than a synchrotron particle accelerator, meaning the technique can in principle be adopted by any well-equipped ultrafast laser lab. That accessibility matters: if trPOT is to become a standard characterization tool for organic semiconductor design, it cannot require large-facility access.
One indicator of the experiment’s difficulty: each individual time-delay point required approximately 25 hours of continuous data collection. Mapping the exciton’s evolution at multiple delays — the equivalent of multiple frames in the quantum film — is an extraordinary logistical feat. The α-6T films were grown at Jülich by Monja Stettner (whose sample preparation was part of her PhD dissertation) and transported to Marburg under ultra-high vacuum to preserve their quality. “The precise alignment of the molecules and their targeted decoupling from the substrate are important for maintaining the exciton long enough to make its formation visible,” Stettner told the university press office.
The reconstruction produced three findings that the theoretical community had predicted but could not directly verify until now.
An exciton spread across three molecules, not one. Conventional textbook descriptions of excitons in organic molecular crystals — known as “Frenkel excitons” — treat them as tightly confined to a single molecule. The new data showed that the exciton in α-6T was coherently spread across approximately three molecular units, with an initial spatial width of about 8.8 ångströms (roughly 0.88 nanometers, or about 0.035 millionths of an inch). This puts α-6T’s excitons in an intermediate category between purely localized Frenkel excitons (single-molecule) and the large-radius Wannier-Mott excitons found in inorganic semiconductors like silicon. The PRX paper documents this finding in quantitative detail.
A characteristic quantum phase pattern. The wave function showed a near-π phase shift between adjacent molecules along the molecular stacking direction — a signature of a coherent intermolecular quantum interaction, predicted by GW/BSE many-body theory and now confirmed in the PRX paper for the first time in any organic semiconductor system. This phase structure is what makes the multi-molecule delocalization coherent rather than accidental.
Self-trapping in 400 femtoseconds. Within less than 400 femtoseconds of its birth, the exciton contracted from its initial 8.8 Å width to approximately 6.9 Å — a roughly 25 percent shrinkage — as the surrounding molecular lattice responded to its presence and tightened its grip. This process, called exciton self-trapping driven by exciton-phonon coupling, had been predicted theoretically and is believed to reduce exciton mobility and limit organic solar cell efficiency, but had never been directly observed in real time before. The team also observed a shift in the exciton’s energy over the same period, consistent with the increased binding energy that accompanies self-trapping.
“We have now succeeded for the first time in experimentally reconstructing the spatial distribution and temporal evolution of an exciton’s wave function in the very first moments of its existence,” said Peter Puschnig, Professor of Electronic Structure of Nanomaterials at the University of Graz, who led the theoretical side of the work. “The measurements show that, after its formation, the electron-hole pair extends across approximately three molecules and then shrinks by around 25 percent within the first 400 femtoseconds.” The full Puschnig quote was released through the university’s official press office.
The self-trapping process revealed here is exactly what photovoltaic engineers have long suspected but could not directly measure: a rapid localization event that reduces how far an exciton can travel before it reaches the charge-splitting interface, and therefore caps how efficiently a solar cell can convert absorbed light into current.
Standard models of exciton transport in organic solar cells — particularly those based on Förster resonance energy transfer (FRET) — assume that excitons hop incoherently from molecule to molecule as localized, single-molecule Frenkel particles. The new data shows that in α-6T, at least, the exciton begins its life delocalized across three molecules with a coherent quantum phase structure before self-trapping collapses it. This initial delocalization window — a few hundred femtoseconds — is the period during which the exciton is most mobile and most likely to reach a heterojunction before trapping. FRET-based design models that ignore this window may systematically underestimate exciton mobility and point engineers toward suboptimal junction geometries.
A separate June 2026 study from researchers at Linköping University and the University of Potsdam, published in Nature Photonics, independently found that longer exciton lifetimes improve fill factors — suggesting the field is converging on the exciton dynamics problem from multiple angles.
Alpha-sexithiophene is a model system, chosen for its well-understood structure and tractable experimental properties. The physical principles that make trPOT work — photoemission from an oriented molecular film, combined with quantitative wave-function reconstruction from momentum-space data — are broadly applicable. Puschnig identified the next target directly: “In the next step, we want to observe the separation of electrons and holes in so-called donor-acceptor systems. This process determines how efficiently light can be converted into electrical current and is therefore central to future developments in organic photovoltaics.” The donor-acceptor next steps are the primary focus of the team’s planned follow-on work.
Beyond photovoltaics, the technique is directly applicable to biological light-harvesting complexes (where exciton transport governs photosynthetic efficiency), quantum-coherent excitonic devices, and hybrid organic-two-dimensional-material systems — any setting where the real-space quantum structure of a correlated electron-hole pair matters.
The research is the most significant milestone to date in the EU-funded Orbital Cinema project, an ERC Synergy Grant worth approximately €11.35 million (approximately $13.2 million USD at September 2026 exchange rates) that runs through June 2029. The project, which funds four research groups in Austria and Germany, aims to produce the equivalent of slow-motion video footage of electrons moving in quantum-mechanical orbitals with sub-femtosecond time resolution — and the exciton film published this week is precisely that ambition realized for the first time.
The photoemission orbital tomography technique itself was pioneered in 2009, when Puschnig and colleagues published a landmark paper in Science demonstrating that molecular orbital densities could be reconstructed directly from photoemission data. The extension to time-resolved measurements was demonstrated in 2021, and the current paper represents the first application to delocalized, multi-molecular excitons with full phase resolution. The theoretical framework developed for this experiment — validated against the most sophisticated available quantum chemistry calculations — is now ready to be applied to the broader class of materials that will determine how much of the world’s rooftop solar energy is actually captured.
Currency conversions in this article are approximate and based on exchange rates as of September 1, 2026.
An exciton is a short-lived quantum particle formed when light knocks an electron out of its resting state in a semiconductor, leaving a positively charged “hole” behind. Electron and hole attract each other, forming a neutral bound pair that carries energy through the material. In an organic solar cell, sunlight first creates excitons; those excitons must then travel to a specially designed interface where they can split into free charges that generate current. If an exciton self-destructs before reaching that interface — either by radiating energy as light or by becoming trapped in the lattice — that portion of absorbed sunlight is lost. The efficiency of an organic solar cell is therefore heavily determined by how far excitons travel and how quickly they self-trap. Until now, the quantum physics of how excitons form and self-trap could only be modeled, not directly filmed. For a primer on exciton basics and organic semiconductors, Ossila provides a detailed introduction.
A regular camera detects photons of visible light reflecting off a surface. Photoemission orbital tomography (trPOT) works at quantum scales and femtosecond timescales by firing a high-energy laser pulse at a material, ejecting electrons whose direction and energy encode information about the quantum-mechanical state of the particle they came from. A sophisticated detector called a time-of-flight momentum microscope captures the full three-dimensional distribution of those ejected electrons simultaneously. A theoretical model then reverses the calculation, reconstructing the spatial shape and quantum phase of the exciton wave function from the electron distribution data. Each “frame” of the resulting quantum film took approximately 25 hours of continuous data collection. The Wikipedia article on photoemission orbital tomography covers the technique’s history and foundations in further detail.
Standard models of exciton energy transport in organic solar cells — particularly those based on Förster resonance energy transfer — treat excitons as localized, single-molecule particles hopping randomly between molecules. The new data shows that in alpha-sexithiophene, newly born excitons are coherently spread across roughly three molecules before self-trapping collapses them. During that brief delocalized phase — a few hundred femtoseconds — an exciton may travel significantly farther than localized models predict, suggesting that FRET-based design rules for donor-acceptor junction distances may be systematically too conservative in materials with strong π-orbital overlap. Directly measuring this delocalized phase for different organic semiconductor candidates may help identify materials whose excitons stay delocalized long enough to improve charge-generation efficiency. The Theilen et al. preprint provides technical detail on the wave function reconstruction methodology and the implications for energy transport models.
Yes — trPOT requires only that a material can be deposited as a thin, ordered molecular film and that a high-energy ultrashort laser pulse (generated by high-harmonic generation) can be applied to eject photoelectrons. These are conditions met by a broad class of organic semiconductors, two-dimensional materials such as transition metal dichalcogenides, and hybrid organic-inorganic systems. The same research team at Graz has already extended the photoemission tomography framework to periodic 2D systems in theoretical work published in late 2025. The next planned experiments involve directly observing how electron-hole pairs separate at donor-acceptor interfaces — the critical step that determines organic solar cell efficiency.
ⓒ 2026 TECHTIMES.com All rights reserved. Do not reproduce without permission.

source