The development of advanced photovoltaic devices, including those that might overcome the single junction efficiency limit, as well as the design of new materials, rely heavily on advanced characterization techniques. Among all the existing methods, optical ones are well suited to probe quantitatively optoelectronic properties, and luminescence-based ones feature preeminently for this purpose. We here present the use of multidimensional imaging techniques that record spatially (with up to 2 µm spatial resolution), spectrally (5 nm), and time-resolved (50 ps) luminescence images. We will discuss the benefits and challenges of looking into energy conversion systems from a multidimensional perspective. We will use some examples, mostly drawn from halide perovskite and III-V materials and device, which will help revisit questions related to efficient transport and conversion in solar cells.
Absorption coefficient rules photonic transport within solar cells and is key for optical optimization. By probing materials density of states, it offers insights in defects characterization of photovoltaic device. Moreover, obtaining this quantity locally would refine the interpretation of the photoluminescence (PL) maps. Its measurement raise challenges as far as classical spectroscopy signals depends on the whole stack structure. Also, its exponential decrease hinders its accurate determination near bandgap energy.
When describing photoluminescence, light absorption influences both the generation function - at the excitation wavelength - and the photons emission and propagation - close to the bandgap energy. We take advantage of these dependances to explore different techniques to obtain either the absorptivity or the absorption coefficient and discuss the link between these two quantities. Transient photoluminescence excitation (PLE) measurement is used to determine the local absorptivity above the bandgap with multidimensional PL imaging techniques. We assess the impact of pulsed excitation on the measurement by modelling the both transient and continuous regime. In addition, absorptivity can be obtained near the optical bandgap directly from the PL spectrum. In this frame, low temperature measurements have already allowed the determination of absorption coefficient with ultra-low values.
These different approaches are compared to reflection measurements classically used at the global scale. The limitations and application scope of the different absorption determination will be described, as well as how they relate to one another. Finally, application to PL imaging will be discussed.
Inverted perovskite solar cells passivated via organic cations exhibit superior power conversion efficiency compared to un-passivated ones. These record efficiencies have been reached thanks to the use of large organic cations to passivate the interface between the perovskite absorber and the transport layers. Here, we study the optoelectronic properties and chemical structure of interface doped perovskite solar cells where large cations, namely Cl-PEAI and F-PEAI, were incorporated at both front and rear interfaces of the absorber. The effect of the cation addition led to an increase of all the main PV characteristic, reaching PCE values up to 23.7%. We combined steady-state and time-resolved multidimensional photoluminescence imaging techniques to probe the main optoelectronic and transport properties of such devices. We thus obtained quantitative maps for physical parameters such as Quasi Fermi Level Splitting (QFLS), Urbach Energy and surface recombination rate, which proved a homogenous passivation by Cl-PEAI and F-PEAI cations over the 3D surface. For instance, the front and rear surface recombination velocities are reduced by a factor of 6 to 8 for Cl-PEAI based samples and 5 to 10 for F-PEAI based samples. In addition, we identified interfacial passivation as the main mechanism leading to a clear improvement of the Voc which increases from 1.10 to 1.16 eV. Indeed, we noticed a clear increase in terms of QFLS only after the addition of the electron transport layer whereas only an increase in the range of 0.01-0.02 eV was observed for bare perovskite thin film with the cation on top. Mapping the opto-electronical properties showed their good spatial homogeneity, By linking optical and electrical measurements we highlight the benefits of this passivation method in maximising all the main photovoltaic characteristics and in approaching inverted perovskite solar cell theoretical limit.
Perovskite-based solar cells are the subject of intense study today because of their promise in terms of high efficiency, easy and low-cost fabrication. To gain insight on the behavior of carriers inside the perovskite layer, time resolved photo luminescence (TR-PL) and time resolved fluorescence imaging (TR-FLIM) are used. However, owing to their long lifetimes (~1µs) and slow diffusion (D~1e-2cm2s-1) the acquired signals require specific care for interpretation.
In a previous work, we showed how these properties can be exploited to derive a scaling-law for the normalized time derivative of the TR-PL signal just after the laser pulse 1/tau. This scaling links the derivative to the material parameters: interface and bulk non radiative recombination, radiative recombination, and diffusion. Our previous focus showed the impact of the laser fluence on the derivative and its use to obtain among others the external radiative recombination coefficient.
In this work we extend the possibility of our previous technique to separate surface and bulk contributions using the impact of the laser wavelength on the scaling of 1/tau through its impact on the spatial distribution of photogenerated carriers. The absorption coefficient of the material at the laser wavelength plays a crucial role in the scaling. We use theoretical computations as well as drift-diffusion simulation to analyze the range of applicability of our technique. We apply experimentally our methodology on perovskite samples with a pulsed laser of varying wavelength. The aim is to determine quantitatively the bulk, front and bottom surface non radiative. We show experimental validation of the scaling on perovskite material and analyze how it can be combined with drift-diffusion simulation. We investigate both interfaces (bottom and front) by varying the illumination side and show how this technique allows for the quantitative comparison of non-radiative recombination at both interfaces. We discuss the experimental uncertainty.
In photovoltaic devices, thermalization of hot carriers generated by high energy photons is one of the major loss mechanisms, which limits the power conversion efficiency of solar cells. Hot carrier solar cells are proposed to increase the efficiency of this technology by suppressing phonon-mediated thermalization channels and extracting hot carriers isentropically. Therefore, designing hot carrier absorbers, which can inhibit electron-phonon interactions and provide conditions for the re-absorption of the energy of non-equilibrium phonons by (hot) carriers, is of significant importance in such devices. As a result, it is essential to understand hot carrier relaxation mechanisms via phonon-mediated pathways in the system. In this work, the properties of photo-generated hot carriers in an InGaAs multi-quantum well structure are studied via steady-state photoluminescence spectroscopy at various lattice temperatures and excitation powers. It is observed that by considering the contribution of thermalized power above the absorber band edge, it is possible to evaluate hot carrier thermalization mechanisms via determining the thermalization coefficient of the sample. It is seen that at lower lattice temperatures, the temperature difference between hot carriers and the lattice reduces, which is consistent with the increase of the quasi-Fermi level splitting for a given thermalized power at lower lattice temperatures. Finally, the spectral linewidth broadening of multiple optical transitions in the QW structure as a function of the thermalized power is investigated.
Short time carrier dynamics of transient photoluminescence decays contain valuable information on the optoelectronic properties of photovoltaic materials. We perform a theoretical analysis on short time dynamics to provide scaling laws for the time derivative of the transient photoluminescence signal as a function of both laser excitation power and wavelength . This innovative approach allowed us to extract in a simple and effective manner the external radiative recombination rate and was tested on different absorbers such as state-of-the-art triple cation mixed halide perovskite and III-V materials. Moreover, by coupling this analysis with the fitting of the whole PL decay, we have quantified different transport parameters and precisely estimated their uncertainties.
Analyzing the photoluminescence (PL) maps of semiconductors complementarily in time and wavelength allows to derive their key optoelectronic and transport properties. Up to now, separate acquisitions along time or wavelength had to be acquired for time and wavelength so that a comprehensive study of the dynamics was out of reach. We developed a 4D imaging set-up that allows the simultaneous acquisition of spectral and temporal luminescence intensity with micrometric spatial resolution under the exact same experimental conditions. This novel set-up relies on single pixel imaging, an approach that enables the reconstruction of the spatial information recorded from a higher resolution non-imaging detector. The sample PL signal is spatially modulated with different patterns by a digital micro-mirror device1. We make use of this technique for the first time with a streak camera as a detector, allowing to record the PL intensity decays and spectrum for each pixel with very high temporal (<100ps) and spectral resolutions (<1nm). A patent application has been filled. We demonstrate the use of this setup by characterizing III-V samples. We observe the spatial variations of a red shift occurring during the short time of the decay.
After around 10 years of research, perovskite cells are one of the most promising technologies in the field of photovoltaic. Despite this, there are still hurdles to overcome, such as stability and inhomogeneities in upscale processes, before it can be commercialized. It is therefore necessary to find a protocol to provide a quality assessment of the perovskite cells and identify the type of defects present inside. Electroluminescence (EL) imaging is an ideal candidate to meet these requirements, as it allows defect detection through the application of a voltage/current. We performed electroluminescence characterization on perovskite cells and modules fabricated at IPVF and showing efficiencies between 14 and 18%. We observed unusual behaviors such as transient phenomena, cell extinction or alternating cell luminescence. To better understand these phenomena we modeled our modules using LTSpice. We managed to reproduce these experimental behaviors by varying parameters such as shunt resistance or recombination rate and observing their effect on modeled EL intensity. Moreover, we can also identify which type of defect is predominant according to the applied voltage. Thanks to this work, we determined a precise protocol to link certain electroluminescent behaviors of the module to a specific physical parameter failure and to their potentially related synthesis defect. To go further, we are working on a 2D version of the model that will allow us to a better understanding of the effect of local inhomogeneities inside a larger cell in larger modules.
We expose here a new method to map various optoelectronic parameters of solar cells from absolutely calibrated voltage dependent electroluminescence imaging. The absolute calibration is derived from radiometric analysis of the setup coupled to a collection model of the minority carriers on the one hand and to the reciprocity existing between electroluminescence and quantum efficiency of the device on the other hand. The method is illustrated on a classical Al-BSF cast-mono silicon solar cell for which we map with good accuracy the diffusion length, dark recombination current, local voltage and lumped series resistance and analyze the results.
Time-resolved fluorescence imaging (TR-FLIM) allows to obtain time-resolved photoluminescence maps with a micrometric resolution. We showed last year that wide-field illumination with TR-FLIM allows to observe lateral transport to a more defective zone. This study will be focused on point and structured illumination, extending the breadth of this characterization technique to bare & homogeneous absorbers. We explore this new possibility by presenting applications for a wide range of semi-conductors. By treating the datacubes with algorithms solving 2D/3D drift-diffusion equations, we derive optoelectronic properties representative for lateral diffusion and local recombination properties.
This talk aims at introducing time-resolved fluorescence imaging (TR-FLIM) as an optical characterization method for optoelectronic devices. It allows to obtain time-resolved photoluminescence maps with a temporal resolution of 500 ps and a micrometric spatial resolution. A first experiment under wide-field illumination on a GaAs-based solar cell is presented as a proof of concept. Thanks to a model including diffusion and recombination of minority charge carriers, we could fit the experimental photoluminescence (PL) transients and decorrelate key optoelectronic properties for the considered device. For various fluence levels, we could determine a constant bulk lifetime τn = 75 ns, a constant effective diffusion length Leff = 190 μm, and an injection-dependent contact recombination velocity Sn, which is explained by the saturation of interface states. The wide-field illumination notably avoids lateral diffusion artefacts leading to a significant underestimation of τn. TR-FLIM also has a noteworthy interest for optoelectronic materials showing heterogeneous properties, as organic-inorganic halide perovskite. With the same set-up, we could investigate various chemical compositions for this semi-conductor, and highlight the need for another self-consistent model linking TRPL transients with physical properties, as no clear definition of a lifetime appears. However, the crucial role of the perovskite/TiO2 can be underlined, in particular for the sample containing Cesium, as well as the probable role of charge carrier in-depth diffusion following a pulsed excitation. As a conclusion, TR-FLIM appears as a versatile characterization method and we open the gate to further studies of other optoelectronic devices with this set-up.
In the present study, we develop a contactless optical characterization tool that quantifies and maps the trapping defects density within a thin film photovoltaic device. This is achieved by probing time-resolved photoluminescence and numerically reconstructing the experimental decays under several excitation conditions. The values of defects density in different Cu(In,Ga)Se2 solar cells were extracted and linked to photovoltaic performances such as the open-circuit voltage. In the second part of the work, the authors established a micrometric map of the trapping defects density. This revealed areas within the thin film CIGS solar cell with low photovoltaic performance and high trapping defects density. This proves that the developed tool can be used to qualify and quantify the buffer layer/absorber interface properties. The final part of the work was dedicated to finding the origin of the spatial fluctuations of the thin film transport properties. To do so, we started by establishing a micrometric map of the absolute quasi-Fermi levels splitting within the same CIGS solar cell, using the hyperspectral imager. A correlation is obtained between the map of quasi-Fermi levels splitting of and the map of the trapping defects density. The latter is found to be the origin of the frequently observed spatial fluctuations of thin film materials properties.
This study aims to provide an innovative insight on polycrystalline solar cells characterization. Accurate and complete information on the material's performance is achieved by probing micrometric fluctuations of its charge carriers' transport properties which might influence the global device’s performance[1][2]. Results on microcrystalline Cu(In,Ga)Se2 solar cells absorbers[3] exhibited an initial fast decay followed by a slower one . Short decay lifetimes varying between 0.4 ns and 1.8 ns, were found to be linked to recombination centers, whereas longer decay lifetimes fluctuating between 3ns and 8ns, were associated with the presence of shallow emission traps. By varying the excitation wavelength from 850nm to 450nm excitation, the authors observed a hysteresis phenomenon regarding the behavior of TRPL decays as a function of the value order of the excitation wavelength. This is related to the activation of metastable defects located at the absorber/buffer interface.
CIGS is a material showing high efficiencies in photovoltaic devices although numerous questions remain about its physical mechanisms. Among them is the influence of the polycrystalline nature on global efficiencies. In order to study the spatial fluctuations of the optoelectronic parameters, two original setups are developed. Firstly a Hyperspectral Imager is used to investigate spectrally resolved luminescence images. It is also possible to calibrate it in absolute values, which means that the signal is measured in number of photons. From photoluminescence measurement we deduce maps of the quasi-Fermi level splitting with variation of 30 meV. These results are compared with a more common confocal microscope, which shows that the carrier transport has to be taken into account for the interpretation of these experiments. Using electroluminescence and reciprocity relations, we calculate map of the External Quantum Efficiency with the Hyperspectral Imager. For this investigation a second setup is developed, using Light Beam Induced Current with different excitation wavelengths. Therefore mapping of the diffusion lengths is possible, exhibiting a distribution around 1.09 μm with standard deviation of 0.10 μm.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
INSTITUTIONAL Select your institution to access the SPIE Digital Library.
PERSONAL Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.