Recently molecular photovoltaics, such as dye sensitized cells (DSCs) and perovskite solar cells (PSCs) have emerged as credible contenders to conventional p-n junction photovoltaics. Their certified power conversion efficiency currently attains 25.5 %, exceeding that of the market leader polycrystalline silicon. This lecture covers the genesis and recent evolution of DSCs and PSCs, describing their operational principles and current performance. DSCs have meanwhile found commercial applications for ambient light harvesting and glazing producing electric power from sunlight. The scale up and pilot production of PSCs are progressing rapidly but there remain challenges that still need to be met to implement PSCs on a large commercial scale. PSCs can produce high photovoltages rendering them attractive for applications in tandem cells, e.g. with silicon and as power source for the generation of fuels from sunlight. Examples to be presented are the solar generation of hydrogen from water and the conversion of CO2 to chemical feedstocks such as ethylene, mimicking natural photosynthesis.

Organic-inorganic metal halide perovskites have emerged as attractive materials for solar cells with power-conversion efficiencies now exceeding 25%. However, challenges and opportunities remain relating to material microstructure, ionic migration and toxicity.

We have recently investigated ultrafast charge-carrier dynamics in lead-free silver-bismuth semiconductors^[1-3] which promise lower toxicity and potentially higher barriers against ion migration than their more prominent lead-halide counterparts. We examined the evolution of photoexcited charge carriers in the double perovskite Cs₂AgBiBr₆ using a combination of temperature-dependent photoluminescence, absorption and optical pump−terahertz probe spectroscopy.^[1] We observe rapid decays in terahertz photoconductivity transients that reveal an ultrafast, barrier-free localization of free carriers on the time scale of 1.0 ps to an intrinsic small polaronic state. While the initially photogenerated delocalized charge carriers show bandlike transport, the self-trapped, small polaronic state exhibits temperature-activated mobilities, allowing the mobilities of both to still exceed 1 cm²V⁻¹s⁻¹ at room temperature. Self-trapped charge carriers subsequently diffuse to color centers, causing broad emission that is strongly red-shifted from a direct band edge. Overall, our observations suggest that strong electron−phonon coupling in this material induces rapid charge-carrier localization which may inhibit the use of this material as an efficient light harvester in photovoltaic devices.

We further demonstrate the novel lead-free semiconductor Cu₂AgBiI₆ which exhibits several advantages over Cs₂AgBiBr₆, namely a low exciton binding energy of ~29 meV and a lower and direct band gap near 2.1 eV,^[2,3] making it a significantly more attractive lead-free material for photovoltaic applications. However, charge carriers in Cu2AgBiI6 are found to exhibit similarly strong charge-lattice coupling strength^[3] to that in Cs₂AgBiBr₆, suggesting a link with the presence of AgBi. Tuning such charge-lattice interactions therefore emerges as a serious challenge for this class of materials.

Finally, we show that lead halide perovskites can exhibit intrinsic quantum confinement, apparent through surprising oscillatory features in the absorption spectrum.^[4] Such materials may thus offer the sought-after target of bottom-up nanostructuring.

Halide perovskite (HaP) materials have attracted great scientific interest in the past years, which is in part because of their unique combination of properties. Specifically, these systems show various fascinating physical properties revolving around their apparent electronic-structure and optical characteristics that are concurrent with finite-temperature lattice-dynamical properties that are very unusual for technologically useful semiconductors. Here, I will present our most recent theoretical findings obtained by means of molecular dynamics based on density-functional theory of the all-inorganic HaP material CsPbBr₃. It will be shown that at finite temperature, the HaP structure favors atomic displacements perpendicular to the inorganic network, so-called transversal displacements, over longitudinal ones. The resulting high degree of transversality in HaP materials provides them with favorable optoelectronic properties at finite temperature, such as narrow energy distributions of electronic states, which has important implications for optical, transport and defect characteristics. These findings are contrasted with results obtained for the case of PbTe, a material that shares certain key properties with CsPbBr₃, but due to its structure cannot allow for transversality, and thus is shown to exhibit less favorable optoelectronic properties at finite temperature. It is concluded that the simple concept of transversality may guide and support material design strategies for alternative compounds with favorable optoelectronic properties.

In 2D perovskites, as in all semiconductors, the degeneracy of the dark and bright exciton states is lifted by the exchange interaction between the electron and hole. This exciton fine structure is essential to understand the interaction of matter with light, and reflects the underlying symmetry of the system. The bright-dark splitting is also of paramount importance for light emitters which rely on the radiative recombination of excitons, since the excitons usually relax to the lowest lying dark state, which is detrimental for the device efficiency. It is expected that the enhanced Coulomb interaction in the 2D limit strongly increases the splitting of dark and bright excitonic states, however, such a splitting has never been measured directly in 2D perovskites. The only available report, based on photoluminescence (PL) studies, provides only a rough estimate of the splitting, since PL can be affected by trap states, and the complex exciton dynamics. For the particular case of the soft perovskites lattice, a direct comparison of the bright-dark splitting with the energy of phonons is crucial to understand the thermal mixing of the two excitonic states which drives radiative recombination in this material system. For the future development of 2D perovskites it is therefore of paramount importance to understand the exciton fine structure in 2D perovskites. I will discuss our optical spectroscopy measurements with an applied in-plane magnetic
field to mix the bright and dark excitonic states of (PEA)2PbI4, providing the first direct measurement of the bright-dark splitting. The induced brightening of the dark state allows us
to directly observe an enhancement of the absorption at the low-energy side of the spectrum related to the dark state. With the signature of dark state in the optical spectra we are able precisely determine the bright-dark exciton state splitting of 8 meV for (PEA)2PbI4at B= 0 T. The brightening of the dark state was also visible in photoluminescence, dominating the emission already at a moderate magnetic field of 6T. The evolution of the PL signal in the magnetic field, suggests that at low temperatures (4.5K) the exciton population is not fully thermalized due to the existence of a phonon bottle-neck, which occurs due to the specific nature of the exciton-phonon coupling in soft perovskite materials.

Lead-halide perovskite thin films have successfully been incorporated as light-absorbing and -emitting layers in a wide range of optoelectronic applications.[1,2] Cs₂AgBiBr₆ (CABB) has been proposed as a promising non-toxic alternative to lead-based perovskites. However, the indirect bandgap and low charge carrier collection efficiencies remain an obstacle for the incorporation of CABB in optoelectronic applications.[3] The limited charge carrier extraction has been ascribed to the high trap density and strong electron-phonon coupling, compared to its lead-containing analogues, resulting in fast charge carrier localization and low charge carrier mobilities. On the other hand, transient absorption experiments show a long-lived charge carrier lifetime ranging over several microseconds.[4] Moreover, at elevated temperatures mobile carriers are observed for microseconds after photoexcitation, highlighting CABB as a potential alternative for lead-halide analogues.[5]

To study the (free) charge-carrier dynamics we have performed transient absorption (TA) spectroscopy and time-resolved microwave conductivity (TRMC) experiments on CABB thin films on nanosecond to microsecond timescales. TA spectroscopy revealed that the charge carrier density decays on a 40-nanosecond timescale but a fraction of the photogenerated holes near the valence band maximum have a lifetime ranging over several microseconds. TRMC measurements showed that these long-lived carriers are, however, not mobile. Comparison of the TRMC and TA traces shows that the conductivity loss is the result of a combined effect of charge carrier loss and localization. Finally, we estimate that the charge carrier diffusion length is ca. 100 nm. This is of the same order of magnitude as the grain size, suggesting that grain boundaries are an important contributor to charge-carrier loss.

Halide double perovskites are an emerging class of materials with considerable structural and electronic diversity [1, 2, 3] and reliable stability towards heat and moisture under ambient conditions. We have recently shown that silver-pnictogen double perovskites, e.g. Cs₂AgBiBr₆, a material with promising optoelectronic properties, exhibit non-hydrogenic and strongly localized resonant excitons. This finding can be traced back to their chemical heterogeneity that leads to anisotropic effective masses and local field effects [4].

In this contribution, we systematically investigate how the electronic and optical excitations in halide double perovskites are impacted by the band edge orbital character. We use ab initio many-body perturbation theory within the GW approximation [6] and the Bethe-Salpeter equation [5] to compute excitations of a group of double perovskites with different valence and conduction band orbital character (Cs₂AgBiCl₆, Cs₂AgInCl₆, Cs₂BiInCl₆). We find that contrary to Cs₂AgBiCl₆, Cs₂AgInCl₆ and Cs₂BiInCl₆ exhibit delocalized excitonic states with low binding energy. Furthermore, carrier effective masses are highly isotropic and local field effects small, which we attribute to the orbital character at the band edges of these direct band gap semiconductors. Our results contribute to a detailed atomistic understanding of the light-matter interactions in chemically heterogeneous double perovskites and highlight the potential of these materials for light harvesting applications.

Photoluminescence is an ubiquitous technique featuring in fields ranging from bio-diagnostics, organic electronics to perovskite solar cells. This accessible characterization tool provides researchers a window into the optoelectronic properties of a semiconductor. What many don’t realize, however, is the extent to which the optical environment plays a role in the recorded spectrum, especially in thin films. Effects extrinsic to the emitting dipole obscure the intrinsic spectrum, making interpretation of experimental spectra non-trivial.

In our work we have developed a model that takes into account the effect of self-absorption and thin film interference of emitted light [1]. We demonstrate the impact of these on the photoluminescence lineshape of an organic semiconductor as a function of film thickness and incident angle. The experimentally determined spectral characteristics are found to differ as much as 60% from their intrinsic material value. The differences in lineshape are calculated to be especially sensitive to film thickness.

We experimentally verified the model by reversely applying it to a measured spectrum. Further, we retrieve the intrinsic spectrum of two non-fullerene acceptor films, allowing for a correct analysis of the impact of their two different fabrication methods on the optoelectronics.

Finally, we provide a chart for the general experimentalist to assess the extent of extrinsic influences on the measured spectrum as function of film thickness, bandgap and Stokes shift.

Taken together, our work aims to bridge the gap between experimentalist and optical modeler to aid in the correct analysis of spectra. We find there is a clear need to take into account the influence of extrinsic effects on the measured spectrum, and provide estimates on the relative impact for various film parameters.

Excitons dominate the optics of atomically-thin transition metal dichalcogenides and 2D van der Waals heterostructures. Interlayer 2D excitons, with an electron and a hole residing in different layers, form rapidly in heterostructures either via direct charge transfer or via Coulomb interactions that exchange energy between layers. Here I will discuss our recent work on the light-matter interaction on quasi-1D van der Waals heterostructures consisting of C/BN/MoS₂ core/shell/skin nanotubes. I will describe how strong intertube excitonic coupling effects between excitons in CNTs and MoS₂ NTs are evident under pulsed infrared excitation of excitons in the semiconducting CNTs. We observed a rapid (sub-picosecond) excitonic response in the visible range from the MoS₂ skin following IR excitation, which we attribute to intertube biexcitons mediated by dipole-dipole Coulomb interactions in the coherent regime. On longer (>100ps) timescales hole transfer from the CNT core to the MoS₂ skin further modified the MoS₂'s absorption. Our direct demonstration of intertube excitonic interactions and charge transfer in 1D van der Waals heterostructures suggests future applications in infrared and visible optoelectronics using these radial heterojunctions.

Theoretical predictions of excited-state phenomena in emerging semiconductor systems can lead to a better understanding of nanoscale energy conversion mechanisms and of the excitonic properties and dynamical effects dominating these processes. In this talk, I will discuss the connection between exciton dynamics, lifetimes, and scattering to exciton dispersion. I will present our recent computational study within many-body perturbation theory, showing that nonanalytical discontinuities are arising from the exchange scattering of electron−hole pairs. These discontinuities are material-specific and stem from its symmetry and dimensionality, with jump discontinuities occurring in 3D and different orders of removable discontinuities in 2D and 1D. I will discuss a general theory to explain these unique features in the exciton bandstructure, combining ab initio and model Hamiltonian approaches. These features are directly connected to ultrafast ballistic transport and can be used to approximate exciton radiative lifetimes and diffusion coefficients, in good correspondence with recent experimental observations.

I will describe a new suite of ultrafast optical microscopies capable of monitoring exciton and phonon dynamics on femtosecond and nanometer spatiotemporal scales in semiconductors and metals. I will focus on two-dimensional van der Waals materials whose properties are radically modified by strong interactions between light, excitons and phonons. Specifically, we find that in semiconductors whose excitons strongly hybridize with photons (forming exciton-polaritons whose wavefunctions are macroscopically delocalized), both polariton-polariton and polariton-exciton interactions overwhelmingly dictate energy flow and the semiconductors’ response to external electromagnetic fields. I will also show that acoustic phonons that strongly couple to localized excitons can be used to coherently manipulate the latter over macroscopic length scales. Directly imaging these quasiparticle dynamics in real space and time allows us to identify the key inter-particle interactions responsible for governing how materials respond to light and how light can be used to steer electronic transport and induce exotic phases of matter.

The GW approximation to the self-energy and the corresponding Bethe-Salpeter equation (BSE) together provide a state-of-the-art computational framework for the study of charged and neutral excitations in molecules and materials. However, their computational cost can be prohibitive for the description of certain electronic phenomena or large systems. In this talk, I'll describe two new directions that aim to lower the cost of GW/BSE calculations. The first direction addresses full-frequency implementations of GW/BSE including dynamical screening. I'll show that the requisite frequency-dependent eigenvalue problems of GW and BSE calculations can be exactly recast as frequency-independent eigenvalue problems in an enlarged space. Combined with the density fitting approximation to electron repulsion integrals, this reformulation leads to reduced scaling implementations of both methods. Moreover, it allows us to quantitatively study double excitations, including their energies and wavefunction character. In a second direction, I'll consider GW/BSE calculations on very large systems, with thousands of electrons or more. Here, we have used a simplifying approximation to the electron repulsion integrals with the same structure as in methods based on tensor hypercontraction. This approximation leads to GW/BSE calculations with storage and execution times that scale quadratically and cubically with the system size, respectively, and typically exhibit errors of only about 0.5 eV with respect to fully ab initio calculations. These methods can be applied to study systems containing thousands of electrons in only a few hours on commodity hardware.

Photoinduced charge transfer in van der Waals heterostructures occurs on ultrafast timescales of order 100 fs, despite the weak interlayer coupling and momentum mismatch. However, little is understood about the microscopic mechanism behind it and role of the lattice in mediating this process. Here, we use ultrafast electron diffraction to directly visualize lattice dynamics in photoexcited heterostructures of WSe2/WS2 monolayers. Following selective optical excitation of WSe2, we measure surprisingly concurrent heating of both WSe2 and WS2 layers on ~1 ps timescales, two orders of magnitude faster than would be expected from the thermal phononic coupling between the layers alone. Using first-principles density functional theory calculations, we identify a fast channel, involving an electronic state hybridized across the heterostructure, for interlayer phonon-assisted transfer of photoexcited electrons between the two layers. Our calculations demonstrate that, via this channel, phonons are emitted in both layers on femtosecond timescales, consistent with the simultaneous lattice heating observed experimentally. Taken together, our work indicates strong electron-phonon coupling via layer-hybridized electronic states – a possible novel route to control thermal transport across van der Waals heterostructures.

The fundamental band gap is a decisive observable when characterizing energy converting materials. Being able to predict it reliably at affordable numerical cost is a prerequisite for the large-scale computational search for new materials. Density Functional Theory (DFT) is often the method of choice for calculating a material's electronic structure. The band gap, however, has been a notoriously difficult observable for DFT. Inexpensive exchange-correlation approximations such as LDA and GGAs systematically underestimate the gap because they lack a derivative discontinuity. Hybrid functionals often yield more realistic results, but at a sharply increased computational cost. In any case, many-body perturbation theory methods, such as the GW approach, have often been required for a reliable prediction of the gap, adding a further layer of methodological complexity and cost. We here show that band gaps can be predicted accurately within DFT, without any empirical parameters, from the TASK meta-generalized gradient approximation (meta-GGA) functional. The latter has been derived from first principles by fulfilling known constraints and by taking into account the derivative discontinuity. Unlike some other meta-GGAs, the TASK functional leads to well converging calculations and predicts band gaps with reasonable accuracy for a large range of systems at the same order of computational expense as a GGA calculation.

The concept of optimal tuning of range-separated hybrid functionals has become an important tool for overcoming the fundamental gap problem and the charge transfer excitation problem in molecular systems. Here, this concept is extended to the solid state by introducing dielectric screening into the functional form. This approach, couched rigorously within the generalized Kohn-Sham formalism of density functional theory, can produce quantitatively the same one and two quasi-particle excitation picture given by many-body perturbation theory (MBPT), without any empiricism. Specifically, for covalent/ionic semiconductors and insulators, accurate band structures and optical absorption spectra, which agree well with those obtained from MBPT, are obtained. For molecular solids, the approach predicts the correct gap renormalization - even from single molecule calculations if a polarizable continuum model is used in an electrostatically consistent manner – and also predicts absorption spectra well.

Exciton transport and separation processes across organic-inorganic interfaces are key ingredients in emerging applications in energy conversion. In particular, heterostructures consisting of molecular crystals are widely explored. While advances in experimental methods allow direct observation and detection of exciton properties across such junctions, a detailed understanding of the exciton nature and its relation to the interface structure and composition is still largely lacking. In this talk I will present a computational assessment of the many-body interactions dominating the excitonic nature at the interface between the perylene diimide (PDI) molecular crystal and an Au substrate, using many-body perturbation theory within the GW and Bethe-Salpeter equation (BSE) approach. We study the effect of structural modifications on the  dielectric screening, manifested through changes in the quasiparticle and excitation energy gaps, as well as the electron-hole binding.  Our findings suggest a close look into local and non-local interactions dominating the excitation energies and the exciton binding and nature in the examined PDI molecular crystals, and shed light on the change in exciton nature and properties upon their adsorption on an inorganic, weakly-interacting substrate, offering structural design principles for excitonic tuning at organic-inorganic interfaces.

Two-dimensional (2D) semiconductors are primed as excellent optoelectronic materials to realize a variety of photonic devices which rely on the absorption of light and the consequent transient properties of photo-generated charges and excitons. Surprisingly, very little is known about the changes in the 2D material's refractive index upon excitation. The associated optical phase changes when light propagates through the 2D material, can be beneficial or undesired depending on the application at hand, but clearly require proper quantification. Measuring optical phase modulation of dilute 2D materials is however not trivial with common ultrafast methods. In this work, we first demonstrate that 2D colloidal CdSe quantum wells, a useful model system, can modulate the phase of light across a broad spectrum using an experimental ultrafast interferometry method. Next, we proceed to develop a toolbox to calculate the time-dependent refractive index of colloidal 2D materials from more widely available broadband transient absorption data using a modified effective medium algorithm. We confirm the experiments and show that the pronounced room temperature excitonic features found in 2D materials result in broadband, ultrafast and sizable phase modulation, even extending sub-band gap to the near-infrared where modulation is associated with well defined intraband transitions.

Light-matter interaction between a molecular transition and a confined electromagnetic field of an absorber inside a resonator can modify the optical and electrical properties of molecules, which is relevant to improve solar devices.[1] When strong coupling regime is reached, two new hybrid states separated in energy, so-called polaritonic states, are formed instead of independent eigenstates, with the energy separation between them being proportional to the coupling strength, which is known as Rabi splitting. This modification of the energy spectra of the system, which have been already demonstrated in many configurations, offers new possibilities for controlled impact on various fundamental properties of coupled matter (rate of chemical reactions, conductivity of organic semiconductors).[2] 

Recently, Subphthalocyanines (SubPcs)-based molecules have been integrated in a polaritonic organic solar cell that behaves as an optical resonator in order to modify device’s absorption onset and tune optoelectronic properties of the devices.[3] Therefore, understanding how the different electronic transitions of an absorber can couple to the resonant modes of a solar polaritonic device is an important issue that deserves to be addressed.

In view of this, we investigate how two electronically different transitions, namely charge transfer (CT)-band and excitonic Q-band, of a subphthalocyanine derivative (F₁₂-SubPc-TCBD-aniline) within an optical cavity, respond to variations in the resonant modes of such cavity. By coupling the cavity resonance with one of the two electronic transitions we observe different coupling regimes, which give rise to very different spectral and directional light harvesting features. Also, by modelling the system under analysis, we can discriminate the productive absorption occurring in the SubPc layer from the parasitic one due to the presence of metallic films in the structure, which allow us to estimate the potential optical losses that may occur in a light harvesting device devised as an optical resonator. [4]

Photoelectrochemical water splitting presents an attractive strategy to produce hydrogen gas as an alternative clean fuel in an environmentally benign and sustainable manner. The key component of a photoelectrochemical cell is a semiconductor electrode (photoelectrode) that absorbs solar light to generate, separate, and transport charge carriers to the semiconductor/electrolyte interface to participate in desired chemical reactions. The electron-hole separation and interfacial charge transfer of the photoelectrode are considerably affected by the interfacial energetics between the photoelectrode and the electrolyte and/or between the photoelectrode and the buffer, protection, or catalyst layers; hence, the interfacial properties of a photoelectrode are as important as the bulk properties of the photoelectrode.

To date, strategies for altering the atomic arrangement at the photoelectrode surface that do not involve extrinsic doping have mainly involved changing the semiconductor surface facets. However, for ternary oxide photoelectrodes with a formula of AxByOz, there exist numerous ways to terminate the surface even for the same facet. For example, the surface can be terminated with A-O or B-O, and the surface A:B ratio may be different from the bulk A:B ratio. In fact, if not grown as single crystals, AxByOz photoelectrodes can have an A-rich or B-rich surface depending on the synthesis method, which can affect their photoelectrochemical properties. However, despite being important and ubiquitous, the effects of surface termination/composition on a ternary oxide photoelectrode have not been systematically studied, and the atomic origin of their effects on interfacial energetics and photoelectrochemical properties have not been elucidated.

In this presentation, we will discuss the effects of surface termination/composition on the interfacial energetics and photoelectrochemical properties of photoanodes using BiVO4 as an example. We will compare epitaxially grown BiVO4photoelectrodes with V-rich and Bi-rich (010) exposed facets and demonstrate that the surface Bi:V ratio has a considerable effect on the surface energetics and photocurrent generation of BiVO4 even for the same (010) facet.