Dr Robert Hoye, FIMMM CEng CSci

The Energy Materials and Devices Group is an interdisciplinary team working on electronic materials for clean energy conversion, with applications including photovoltaics and light-emitting diodes. Our research spans from creating new fundamental insights into carrier-matter interactions through to applying this understanding to create high-performing devices from new materials processed using scalable methods.

The group specialise in developing solution- and vapour-based techniques for growing complex materials, understanding the carrier dynamics in the materials through spectroscopic methods, as well as creating novel structures to realise efficient devices.

A particularly unique aspect of our work is the application of our skills to create new classes of semiconductors that can tolerate defects. This can lead to the next generation of energy devices that can achieve high performance when made by low-cost methods.

More details on the three main areas of research are given below.

Background and Motivation

PV materials absorb light to generate electrons and holes. These need to be extracted but can be irreversibly lost via defects. Defects introduce energy levels (traps) in the band gap that annihilate carriers through Shockley-Read-Hall (SRH) recombination. Carefully manufacturing the semiconductor to reduce the defect density is one way to reduce the SRH recombination rate so that photo-generated carriers live long enough to be extracted. A lifetime of at least 1–100 ns is needed for PVs. But SRH also depends on the trap energy level in the band gap. Defects that are close to the band-edge (i.e., shallow) give lower SRH rates and is one way to enable defect tolerance, where a long lifetime is achieved despite a high defect density.

Methylammonium lead iodide perovskite has recently been found by serendipity to be defect tolerant. Made by cheap solution processing, these materials have a high trap density of 10¹⁵–10¹⁶ cm^-3 (cf. 10⁸ cm^-3 for silicon), but still achieve long SRH lifetimes >100 ns, with devices already outperforming multicrystalline silicon. Defect tolerance has also been found in other lead-halide perovskites, as well as copper indium diselenide. However, safety concerns over the soluble toxic lead content in perovskites may limit their deployment, whereas indium has high cost and limited supply. It is therefore critical to significantly expand upon the classes of defect-tolerant materials well beyond the limited range currently available to fulfil the urgent need for efficient, low-cost PV.

Recent Results

Recently, it has been proposed that ns² compounds could replicate the defect tolerance of the lead-halide perovskites. These compounds have a partially-oxidised heavy metal cation with a lone pair of valence electrons (e.g., Bi³⁺, In⁺ or Sb³⁺). This is because (1) the valence s²electrons can hybridise with the anion p orbitals to replicate the perovskite electronic structure, (2) there is large spin-orbit coupling, resulting in disperse bands and shallower defects, and (3) the large electron cloud leads to high polarizability, resulting in a high dielectric constant to Coulombically screen charged defects. Point (1) is especially important. In traditional materials (e.g., silicon), when atomic orbitals overlap, bonding-antibonding orbital pairs form across the band gap. When these bonds are broken due to defects, the dangling bonds form close to where the original atomic orbitals were, leading to deep defects. By contrast, lead-halide perovskites form bonding-antibonding orbital pairs in the valence band, as well as across the band gap, and dangling bonds then form defects in the valence band or close to the band-edges, resulting in most defects being shallow rather than deep.

One of the materials we predicted using these design rules is bismuth oxyiodide (BiOI). We showed the material to be air-stable and to have a lifetime promising for photovoltaics. We developed an all-inorganic device structure, from which we achieved an external quantum efficiency of 80% at 450 nm wavelength, exceeding previous reports of bismuth-based absorbers.

Key literature

Robert L. Z. Hoye, et al. Strongly Enhanced Photovoltaic Performance and Defect Physics of Air‐Stable Bismuth Oxyiodide (BiOI). Advanced Materials, 2017, 29, 1702176.

Tahmida N. Huq, Lana C. Lee, ..., and Robert L. Z. Hoye, Electronic Structure and Optoelectronic Properties of Bismuth Oxyiodide Robust Against Percent-Level Iodine-, Oxygen-, and Bismuth-Related Surface Defects. Advanced Functional Materials, 2020, 1909983. Early View, DOI: 10.1002/adfm.201909983

Growth from solution

We have developed routes to synthesise exotic functional thin films by solution-based methods. One example is Cs₂AgBiBr₆ double perovskite. By controlling the nucleation rate of film formation and the post-annealing temperature, we were able to achieve phase-pure films of this quaternary compound, which gave long lifetimes exceeding a microsecond. We have also worked with LMU-Munich (Dr. Lakshmi Polavarapu, Prof. Dr. Jochen Feldmann, Dr. Alex Urban) to grow perovskite nano platelets with precise control over the number of layers. Through quantum confinement, these nano platelets blue-shift the energy of the excitons. We were able to achieve blue perovskite LEDs with colour-pure, ultra-sharp electroluminescence at 464 nm wavelength.

Key Literature

Robert L. Z. Hoye, May-Ling Lai, et al., Identifying and Reducing Interfacial Losses to Enhance Color-Pure Electroluminescence in Blue-Emitting Perovskite Nanoplatelet Light-Emitting Diodes. ACS Energy Letters, 2019, 4(5), 1181.

Robert L. Z. Hoye, et al., Fundamental Carrier Lifetime Exceeding 1 µs in Cs₂AgBiBr₆ Double Perovskite. Advanced Materials Interfaces, 2018, 5(15), 1800464.

Atmospheric Pressure Chemical Vapour Deposition

Atmospheric pressure chemical vapour deposition (AP-CVD) is a vapour-based technique we use to grow functional oxide thin films. The reactor is comprised of a gas manifold, in which the vapour-based metal precursor, oxidant and inert gases are introduced via separate channels. From these header channels, the precursor gases are vertically introduced to the substrate. The substrate is oscillated underneath these gas channels, such that the precursors chemisorb and react to form oxide layers. By repeatedly oscillating the substrate underneath the gas manifold, the oxide thickness can be controlled. We have grown a wide range of n- and p-type oxides and we find that the films have a similar uniformity and density as atomic layer deposited oxide films. Additionally, the AP-CVD films are conformal to high aspect ratio nanostructures. Importantly, we are able to grow these films two orders of magnitude faster than atomic layer deposition without the need for a vacuum chamber. This makes the process appealing for growing oxides at scale, as well as growing oxides over thermally-sensitive semiconductors. We have applied the oxides grown (e.g., ZnO, NiO) as charge transport layers in solar cells, charge injector layers in light-emitting diodes, as well as the active layer in thin film transistors. Through AP-CVD, we can tune the properties of our films over a wide range through the growth temperature, oxidant used and alloying our oxides with other elements. For example, we tuned the electron affinity of ZnO through Mg alloying, and this led to improved open-circuit voltage in solar cells, as well as reduced turn-on voltage in perovskite light-emitting diodes.

Key Literature

Robert L. Z. Hoye, et al., Improved Open‐Circuit Voltage in ZnO–PbSe Quantum Dot Solar Cells by Understanding and Reducing Losses Arising from the ZnO Conduction Band Tail. Advanced Energy Materials, 2014, 4(8), 1301544.

Robert L. Z. Hoye, et al., Synthesis and modeling of uniform complex metal oxides by close-proximity atmospheric pressure chemical vapor deposition. ACS Applied Materials & Interfaces, 2015, 7(20), 10684

Vapour-Based Growth

We have developed physical and chemical vapour-based methods to grow new solar absobers. These are scalable methods comprised of a horizontal tube heated with a two-zone furnace. The precursor (e.g., BiI₃ for growing BiI₃ by physical vapour transport or BiOI by chemical vapour transport) is vapourised in the first zone and is carried to the second zone by an inert gas. In chemical vapour transport or chemical vapour deposition, we introduce oxygen gas to grow oxyhalides. We have used these growth systems to synthesise halide and oxyhalide thin films for investigation in photovoltaics, as well as single crystals to understand the fundamental properties of the materials.

Key Literature

Robert L. Z. Hoye, et al., Strongly Enhanced Photovoltaic Performance and Defect Physics of Air‐Stable Bismuth Oxyiodide (BiOI). Advanced Materials, 2017, 29(36), 1702176.

Spectroscopic Characterisation

The key properties of new solar absorbers that determine their suitability for photovoltaics are: 1) the band gap and absorption coefficient, 2) minority-carrier lifetime, 3) photoluminescence quantum yield and 4) electronic structure. We have access to a suite of facilities to evaluate these properties. These include time-correlated single photon counting to measure the photoluminescence lifetime, as well as photoluminescence mapping techniques with a confocal microscope. We also collaborate with the Optoelectronics Group (Cavendish Laboratory, Cambridge) to understand the photophysics of our materials by transient absorption spectroscopy. One of the key parameters influencing the minority-carrier lifetime is the defects present. One of the techniques we use to evaluate the defects present is photothermal deflection spectroscopy, with which we can measure the sub-bandgap absorbance. To evaluate the electronic structure, we have photoemission spectroscopy. For example, through detailed X-ray and ultraviolet photoemission spectroscopy measurements, we identified that one of the factors limiting the performance of BiOI solar cells was the downwards band bending of the absorber material with the NiO_x hole transport layer.

Key Literature

Tahmida N. Huq, Lana C. Lee, ..., and Robert L. Z. Hoye, Electronic structure and optoelectronic properties of bismuth oxyiodide robust against percent-level iodine-, oxygen-, and bismuth-related surface defects. Advanced Functional Materials, 2020, 1909983. Early View, DOI: 10.1002/adfm.201909983

Jeremy Poindexter, Robert L. Z. Hoye, et al., High tolerance to iron contamination in lead halide perovskite solar cells. ACS Nano, 2017, 11(7), 7101.

Thin-Film Solar Cells

We work on developing thin film solar cells based on emerging materials. In the past few years, we have worked with colloidal quantum dot absorbers, polymers, cuprous oxide, as well as ns²compounds predicted to be defect-tolerant. We have developed a wide range of charge transport layers for optimal carrier extraction and to control non-radiative losses at the interfaces. Our work heavily focusses on understanding the processes limiting device performance, such as through optical loss analyses, transient photovoltage measurements, intensity-dependent device measurements and white-light biased external quantum efficiency measurements. In numerous cases, we have devised strategies to achieve improved performance over the state-of-the-art.

Key Literature

Robert L. Z. Hoye, et al. Strongly Enhanced Photovoltaic Performance and Defect Physics of Air‐Stable Bismuth Oxyiodide (BiOI). Advanced Materials, 2017, 29, 1702176.

Tandem Solar Cells

Tandem photovoltaics combine multiple solar cells to increase the fraction of the solar spectrum converted to electrical energy, allowing single-junction device efficiency limits to be overcome. We have developed tandems between metal-halide perovskite top-cells (visible light absorbers) and silicon bottom cells (near-infrared light absorbers) in a monolithic configuration, achieving high performance and stable device operation. Our focus in this area is on the interfaces with the recombination contact (which couples together the top and bottom cells) and transparent top electrode.

Key Literature

Kevin A. Bush, Axel F. Palmstrom, et al., 23.6�fficient monolithic perovskite/silicon tandem solar cells with improved stability. Nature Energy, 2017, 2, 17009.

Robert L. Z. Hoye, et al., Developing a robust recombination contact to realize monolithic perovskite tandems with industrially common p-type silicon solar cells. IEEE Journal of Photovoltaics, 2018, 8, 1023.

Light-Emitting Diodes

Light-emitting diodes are the inverse of photovoltaics: Electrons and holes are injected to an active layer to achieve electroluminescence. Over the past few years, we have worked primarily on metal-halide perovskite emitters. We have developed device structures that reduce non-radiative losses and more effectively confine electrons and holes within the emissive layer, to give colour-pure electroluminescence. A particular advantage of metal-halide perovskites is that their low levels of disorder result in sharper electroluminescence peaks than conventional inorganic and organic materials, making the perovskites particularly appealing for ultrahigh definition display applications.

Key Literature

Robert L. Z. Hoye, May-Ling Lai, et al., Identifying and Reducing Interfacial Losses to Enhance Color-pure Electroluminescence in Blue-emitting Perovskite Nanoplatelet Light-emitting Diodes. ACS Energy Letters, 2019, 4(5), 1181.

Robert L. Z. Hoye, et al., Enhanced performance in fluorene‐free organometal halide perovskite light‐emitting diodes using tunable, low electron affinity oxide electron injectors. Advanced Materials, 2015, 27, 1414.