How Transparent Oxides Gain Some Color: Discovery of a CeNiO3 Reduced Bandgap Phase As an Absorber for Photovoltaics.

In this work, we describe the formation of a reduced bandgap CeNiO3 phase, which, to our knowledge, has not been previously reported, and we show how it is utilized as an absorber layer in a photovoltaic cell. The CeNiO3 phase is prepared by a combinatorial materials science approach, where a library containing a continuous compositional spread of Ce xNi1- xO y is formed by pulsed laser deposition (PLD); a method that has not been used in the past to form Ce-Ni-O materials. The library displays a reduced bandgap throughout, calculated to be 1.48-1.77 eV, compared to the starting materials, CeO2 and NiO, which each have a bandgap of ∼3.3 eV. The materials library is further analyzed by X-ray diffraction to determine a new crystalline phase. By searching and comparing to the Materials Project database, the reduced bandgap CeNiO3 phase is realized. The CeNiO3 reduced bandgap phase is implemented as the absorber layer in a solar cell and photovoltages up to 550 mV are achieved. The solar cells are also measured by surface photovoltage spectroscopy, which shows that the source of the photovoltaic activity is the reduced bandgap CeNiO3 phase, making it a viable material for solar energy.

Process-Function Data Mining for the Discovery of Solid-State Iron-Oxide PV.

Data mining tools have been known to be useful for analyzing large material data sets generated by high-throughput methods. Typically, the descriptors used for the analysis are structural descriptors, which can be difficult to obtain and to tune according to the results of the analysis. In this Research Article, we show the use of deposition process parameters as descriptors for analysis of a photovoltaics data set. To create a data set, solar cell libraries were fabricated using iron oxide as the absorber layer deposited using different deposition parameters, and the photovoltaic performance was measured. The data was then used to build models using genetic programing and stepwise regression. These models showed which deposition parameters should be used to get photovoltaic cells with higher performance. The iron oxide library fabricated based on the model predictions showed a higher performance than any of the previous libraries, which demonstrates that deposition process parameters can be used to model photovoltaic performance and lead to higher performing cells. This is a promising technique toward using data mining tools for discovery and fabrication of high performance photovoltaic materials.

A combined computational and experimental investigation of Mg doped α-Fe2O3.

In the current work, pristine α-Fe2O3 metal oxide was doped with Mg in an attempt to modulate its electronic properties. To this end, we employed an experimental high throughput strategy, including scanning XRD and optical spectroscopy, which were complimented by atomistic density functional theory (DFT) calculations. The combined study reveals that at Mg/Fe atomic ratios up to ∼1/3, the bandgaps of the hematite-Mg composite materials are similar to that of the pure material. The observed bandgaps are rationalized by electronic band structure and density of states calculations. Additional rationale for the similar bandgaps in pure and doped hematite is provided by topological Bader charge analyses, which indicate that the Mg and Fe ions in the hematite matrix have similar partial atomic charges. Nonetheless, the small charge density difference between the Mg and Fe ions induces a slight spin polarization on both oxygen and Fe ions, resulting in changes in the band edges. Further charge density analyses, using charge density maps and chemical-bonding analyses with the crystal orbital Hamiltonian population scheme, indicate that Mg forms ionic bonds with the neighboring oxygen atoms. This change from iron-oxygen covalent bonds to a more ionic nature for magnesium-oxygen bonds is probably responsible for the reduction observed in the computed bulk modulus of α-Mg(0.17)Fe(1.83)O3 (193 GPa) compared to α-Fe2O3 (202 GPa).

Open circuit potential build-up in perovskite solar cells from dark conditions to 1 sun.

The high open-circuit potential (Voc) achieved by perovskite solar cells (PSCs) is one of the keys to their success. The Voc analysis is essential to understand their working mechanisms. A large number of CH3NH3PbI3-xClx PSCs were fabricated on single large-area substrates and their Voc dependencies on illumination intensity, I0, were measured showing three distinctive regions. Similar results obtained in Al2O3 based PSCs relate the effect to the compact TiO2 rather than the mesoporous oxide. We propose that two working mechanisms control the Voc in PSCs. The rise of Voc at low I0 is determined by the employed semiconductor n-type contact (TiO2 or MgO coated TiO2). In contrast, at I0 close to AM1.5G, the employed oxide does not affect the achieved voltage. Thus, a change of regime from an oxide-dominated EFn (as in the dye sensitized solar cells) to an EFn, directly determined by the CH3NH3PbI3-xClx absorber is suggested.

Utilizing pulsed laser deposition lateral inhomogeneity as a tool in combinatorial material science.

Pulsed laser deposition (PLD) is widely used in combinatorial material science, as it enables rapid fabrication of different composite materials. Nevertheless, this method was usually limited to small substrates, since PLD deposition on large substrate areas results in severe lateral inhomogeneity. A few technical solutions for this problem have been suggested, including the use of different designs of masks, which were meant to prevent inhomogeneity in the thickness, density, and oxidation state of a layer, while only the composition is allowed to be changed. In this study, a possible way to take advantage of the large scale deposition inhomogeneity is demonstrated, choosing an iron oxide PLD-deposited library with continuous compositional spread (CCS) as a model system. An Fe2O3-Nb2O5 library was fabricated using PLD, without any mask between the targets and the substrate. The library was measured using high-throughput scanners for electrical, structural, and optical properties. A decrease in electrical resistivity that is several orders of magnitude lower than pure α-Fe2O3 was achieved at ∼20% Nb-O (measured at 47 and 267 °C) but only at points that are distanced from the center of the PLD plasma plume. Using hierarchical clustering analysis, we show that the PLD inhomogeneity can be used as an additional degree of freedom, helping, in this case, to achieve iron oxide with much lower resistivity.

Data Mining and Machine Learning Tools for Combinatorial Material Science of All-Oxide Photovoltaic Cells.

Growth in energy demands, coupled with the need for clean energy, are likely to make solar cells an important part of future energy resources. In particular, cells entirely made of metal oxides (MOs) have the potential to provide clean and affordable energy if their power conversion efficiencies are improved. Such improvements require the development of new MOs which could benefit from combining combinatorial material sciences for producing solar cells libraries with data mining tools to direct synthesis efforts. In this work we developed a data mining workflow and applied it to the analysis of two recently reported solar cell libraries based on Titanium and Copper oxides. Our results demonstrate that QSAR models with good prediction statistics for multiple solar cells properties could be developed and that these models highlight important factors affecting these properties in accord with experimental findings. The resulting models are therefore suitable for designing better solar cells.

Four-point probe electrical resistivity scanning system for large area conductivity and activation energy mapping.

The electrical properties of metal oxides play a crucial role in the development of new photovoltaic (PV) systems. Here we demonstrate a general approach for the determination and analysis of these properties in thin films of new metal oxide based PV materials. A high throughput electrical scanning system, which facilitates temperature dependent measurements at different atmospheres for highly resistive samples, was designed and constructed. The instrument is capable of determining conductivity and activation energy values for relatively large sample areas, of about 72 × 72 mm(2), with the implementation of geometrical correction factors. The efficiency of our scanning system was tested using two different samples of CuO and commercially available Fluorine doped tin oxide coated glass substrates. Our high throughput tool was able to identify the electrical properties of both resistive metal oxide thin film samples with high precision and accuracy. The scanning system enabled us to gain insight into transport mechanisms with novel compositions and to use those insights to make smart choices when choosing materials for our multilayer thin film all oxide photovoltaic cells.

Quantum efficiency and bandgap analysis for combinatorial photovoltaics: sorting activity of Cu-O compounds in all-oxide device libraries.

All-oxide-based photovoltaics (PVs) encompass the potential for extremely low cost solar cells, provided they can obtain an order of magnitude improvement in their power conversion efficiencies. To achieve this goal, we perform a combinatorial materials study of metal oxide based light absorbers, charge transporters, junctions between them, and PV devices. Here we report the development of a combinatorial internal quantum efficiency (IQE) method. IQE measures the efficiency associated with the charge separation and collection processes, and thus is a proxy for PV activity of materials once placed into devices, discarding optical properties that cause uncontrolled light harvesting. The IQE is supported by high-throughput techniques for bandgap fitting, composition analysis, and thickness mapping, which are also crucial parameters for the combinatorial investigation cycle of photovoltaics. As a model system we use a library of 169 solar cells with a varying thickness of sprayed titanium dioxide (TiO2) as the window layer, and covarying thickness and composition of binary compounds of copper oxides (Cu-O) as the light absorber, fabricated by Pulsed Laser Deposition (PLD). The analysis on the combinatorial devices shows the correlation between compositions and bandgap, and their effect on PV activity within several device configurations. The analysis suggests that the presence of Cu4O3 plays a significant role in the PV activity of binary Cu-O compounds.

Interpretation of optoelectronic transient and charge extraction measurements in dye-sensitized solar cells.

Tools that assess the limitations of dye sensitized solar cells (DSSCs) made with new materials are critical for progress. Measuring the transient electrical signals (voltage or current) after optically perturbing a DSSC is an approach which can give information about electron concentration, transport and recombination. Here we describe the theory and practice of this class of optoelectronic measurements, illustrated with numerous examples. The measurements are interpreted with the multiple trapping continuum model which describes electrons in a semiconductor with an exponential distribution of trapping states. We review standard small perturbation photocurrent and photovoltage transients, and introduce the photovoltage time of flight measurement which allows the simultaneous derivation of both effective diffusion and recombination coefficients. We then consider the utility of large perturbation measurements such as charge extraction and the current interrupt technique for finding the internal charge and voltage within a device. Combining these measurements allows differences between DSSCs to be understood in terms such as electron collection efficiency, semiconductor conduction band edge shifts and recombination kinetics.

New insight into the regeneration kinetics of organic dye sensitised solar cells.

The order of regeneration for DSCs based on two organic dyes has been investigated by transient absorption spectroscopy on devices under operating conditions and determined to be 2nd order in iodide. The results shed light on the mechanism and limits to the regeneration rate relative to oxidation potential.

All-Oxide Photovoltaics.

Recently, a new field in photovoltaics (PV) has emerged, focusing on solar cells that are entirely based on metal oxide semiconductors. The all-oxide PV approach is very attractive due to the chemical stability, nontoxicity, and abundance of many metal oxides that potentially allow manufacturing under ambient conditions. Already today, metal oxides (MOs) are widely used as components in PV cells such as transparent conducting front electrodes or electron-transport layers, while only very few MOs have been used as light absorbers. In this Perspective, we review recent developments of all-oxide PV systems, which until today were mostly based on Cu2O as an absorber. Furthermore, ferroelectric BiFeO3-based PV systems are discussed, which have recently attracted considerable attention. The performance of all-oxide PV cells is discussed in terms of general PV principles, and directions for progress are proposed, pointing toward the development of novel metal oxide semiconductors using combinatorial methods.

Simulation and measurement of complete dye sensitised solar cells: including the influence of trapping, electrolyte, oxidised dyes and light intensity on steady state and transient device behaviour.

A numerical model of the dye sensitised solar cell (DSSC) is used to assess the importance of different loss pathways under various operational conditions. Based on our current understanding, the simulation describes the processes of injection, regeneration, recombination and transport of electrons, oxidised dye molecules and electrolyte within complete devices to give both time dependent and independent descriptions of performance. The results indicate that the flux of electrons lost from the nanocrystalline TiO(2) film is typically at least twice as large under conditions equivalent to 1 sun relative to dark conditions at matched TiO(2) charge concentration. This is in agreement with experimental observations (Barnes et al. Phys. Chem. Chem. Phys. [DOI: 10.1039/c0cp01855d]). The simulated difference in recombination flux is shown to be due to variation in the concentration profile of electron accepting species in the TiO(2) pores between light and dark conditions and to recombination to oxidised dyes in the light. The model is able to easily incorporate non-ideal behaviour of a cell such as the variation of open circuit potential with light intensity and non-first order recombination of conduction band electrons. The time dependent simulations, described by the multiple trapping model of electron transport and recombination, show good agreement with both small and large transient photocurrent and photovoltage measurements at open circuit, including photovoltage rise measurements. The simulation of photovoltage rise also suggests the possibility of assessing the interfacial resistance between the TiO(2) and substrate. When cells with a short diffusion length relative to film thickness were modelled, the simulated small perturbation photocurrent transients at short circuit (but not open circuit) yielded significantly higher effective diffusion coefficients than expected from the mean concentration of electrons and the electrolyte in the cell. This implies that transient measurements can overestimate the electron diffusion length in cells which have a low collection efficiency. The model should provide a useful general framework for exploring new cell descriptions, architectures and other factors influencing device performance.

Factors controlling charge recombination under dark and light conditions in dye sensitised solar cells.

A simple and powerful approach for assessing the recombination losses in dye sensitised solar cells (DSSCs) across the current voltage curve (j-V) as a function of TiO(2) electron concentration (n) is demonstrated. The total flux of electrons recombining with iodine species in the electrolyte and oxidised dye molecules can be thought of as a recombination current density, defined as j(rec) = j(inj)-j where j(inj) is the current of electrons injected from optically excited dye states and j is the current density collected at cell voltage (V). The electron concentration at any given operating conditions is determined by charge extraction. This allows comparison of factors influencing electron recombination rates at matched n. We show that j(rec) is typically 2-3 times higher under 1 sun equivalent illumination (j(inj) > 0) relative to dark (j(inj) = 0) conditions. This difference was increased by increasing light intensity, electrolyte iodine concentration and electrolyte solvent viscosity. The difference was reduced by increasing the electrolyte iodide concentration and increasing the temperature. These results allowed us to verify a numerical model of complete operational cells (Barnes et al., Phys. Chem. Chem. Phys., DOI: 10.1039/c0cp01554g) and to relate the differences in j(rec) to physical processes in the devices. The difference between j(rec) in the light and dark can be explained by two factors: (1) an increase in the concentration of electron acceptor species (I(3)(-) and/or I(2)) when current is flowing under illumination relative to dark conditions where the current is flowing in the opposite direction, and (2) a non-trivial contribution from electron recombination to oxidised dye molecules under light conditions. More generally, the technique helps to assign the observed relationship between the components, processing and performance of DSSCs to more fundamental physical processes.

Re-evaluation of recombination losses in dye-sensitized cells: the failure of dynamic relaxation methods to correctly predict diffusion length in nanoporous photoelectrodes.

Photocurrents generated by thick, strongly absorbing, dye-sensitized cells were reduced when the electrolyte iodine concentration was increased. Electron diffusion lengths measured using common transient techniques (L(n)) were at least two times higher than diffusion lengths measured at steady state (L(IPCE)). Charge collection efficiency calculated using L(n) seriously overpredicted photocurrent, while L(IPCE) correctly predicted photocurrent. This has implications for optimizing cell design.