The development of future crystalline silicon (c-Si) solar cell technologies requires innovative surface passivation layers. Sulfides are a somewhat unexplored class of passivation materials, despite previous reports showing that sulfurization of the c-Si surface enhances surface passivation. Herein, we report a novel transparent passivation stack composed of ZnS/Al2O3, sequentially deposited by atomic layer deposition (ALD). This stack exhibits remarkable surface passivation, reaching a recombination current pre-factor J0 as low as 1.0 fA/cm(2) and implied open-circuit voltages iVoc > 730 mV for a wide range of deposition and annealing conditions. Capacitance-voltage measurements reveal an extremely low interface state density of approximate to 1x10(10) cm(-2) eV(-1), on par with state-of-the-art Si-based passivation layers such as SiO2 and a-Si:H, together with a moderate positive fixed charge. A close lattice match between c-Si and ZnS suggests potential epitaxial growth, which could explain the low interface state density and outstanding surface passivation, despite the observation of a polycrystalline bulk structure. These results establish ZnS as an important new material for c-Si surface passivation, with the potential to enable future innovations either as an interlayer for passivating contacts or as a dielectric passivation layer in c-Si solar cells.

The bulk lifetime (tau(bulk)) of high-quality silicon wafers is becoming ever more important as the efficiency of the solar cells increases (i.e. >27 %). As such there is a growing necessity to measure tau(bulk) directly. In this work, we showcase a non-destructive technique to infer tau(bulk) of high-quality wafers, termed the Charge Decoupling Method. This method is very simple and only requires atomic layer deposited (ALD) aluminium oxide (Al2O3)-or any other highly charged film (e.g. >10(12) q cm(-2))-and corona charging. By measuring the decay in the effective lifetime as the net negative charge decreases-due to the sequential deposition of positive corona charges-we can quantify the shape of the curve-analogous to the fill factor for a solar cell-and from this, determine the injection-dependent tau(bulk) and surface recombination velocity S. To experimentally verify the method, we apply the charge decoupling method to Al2O3 passivated Czochralski-grown 5 Omega cm, n-type, 150 mu m thick silicon wafers. From this we obtain a tau(bulk) of similar to 30 ms and a corresponding S of 0.2 cm(-1) at an injection level of similar to 10(15) cm(-3). Mathematically, we demonstrate that the charge decoupling method depends solely on the ratio of surface to bulk recombination and thus does not depend on the doping type, resistivity, charge density, wafer thickness, interface defect density and the injection-dependent bulk lifetime.

In this work, we investigate the properties of CuxCryOz thin films deposited by atomic layer deposition (ALD) over a wide compositional range. A significant increase in growth rate is observed for intermediate compositions and shown to arise from an enhancement of the CrOx deposition rate on the CuOx surface. In addition to the characteristics of the deposition process, we explore the structural and optoelectronic properties of these films for compositions ranging from copper-free chromium oxide to chromium-free copper oxide, and for various post deposition annealing temperatures (400-800 degrees C). The resulting composition, optical constants, band gap, valence band maximum and work function are determined and used to draw full band diagrams of the binary and ternary oxides. We report for the first time the experimental work function of the spinel phase CuCr2O4 (5.0 +/- 0.2 eV). Finally, the contact resistivity of the films with p-type silicon is examined to assess their potential use as hole-selective contacts for crystalline Si solar cells. The lowest contact resistivity (1.72 ohm cm2) was found for as deposited Cu0.05Cr0.30O0.65.

The next generation of commercial solar cells must deliver a lower levelized cost of energy than the incumbent, crystalline silicon. Yet silicon solar cells set a formidable challenge, with degradation as low as 0.5 %/year and record efficiency exceeding 27 %. Success against silicon thus demands a challenging combination of performance and reliability, a combination that no other technology has met at scale. This raises an interesting possibility for solar cell designs that take maximal advantage of the strengths of silicon solar cells, rather than directly compete. In this contribution we theoretically assess a less commonly considered solar cell architecture: a non-current-matched, four-terminal tandem that pairs high efficiency silicon solar cells with wide-bandgap top-cell absorbers above 1.8 eV. This design shifts a majority share of power conversion to the silicon bottom cell and thereby expands the range of top-cell materials that can deliver tandem efficiencies above 30 % while simultaneously improving resilience to top-cell degradation. These advantages depend on high silicon efficiency, yet if this is the case, the top-cell absorber is freed from having to meet a demanding combination of optical bandgap, electronic quality and reliability.

CuO is an important p-type semiconductor with a wide range of potential applications, both on its own, and as a basis for various compounds. However, the ability to deposit CuO thin films and precisely tune their properties using a low temperature process is a limiting factor. In this work, we report a new atomic layer deposition process to deposit CuO thin films using Cu(dmap)2 and O2 plasma. We achieve self-limiting deposition behaviour, enabling depositions in the 90-150 degrees C temperature window. The films are crystalline and display low surface roughness up to 150 degrees C. More importantly, we also observe significant changes in the optoelectronic properties with deposition temperature, including band gap (1.08-1.16 eV), work function (4.82-5.15 eV), valence band maximum (5.33-5.48 eV) and optical constants. The developed deposition process provides a promising route to fabricate CuO thin films with well-defined and tunable properties at low temperature, which could open doors for future semiconductor applications of copper-based compounds.

Heavily-doped, amorphous and polycrystalline silicon layers play important roles in silicon solar cell fabrication and performance. Here we demonstrate applications of time-resolved photoluminescence decay to measure recombination lifetimes in such regions, which are generally below 1µs, and difficult to measure with other techniques. Firstly, we demonstrate the measurement of Auger lifetimes in uniformly heavily-doped silicon wafers, and show the impact of surface recombination in samples with phosphorus or boron doping concentrations below 1×1019 cm-3. We also assess the possible impact of high concentrations of iron contamination on the extraction of such Auger lifetimes. We then report recombination lifetimes measured in thin deposited intrinsic amorphous silicon films, and heavily-doped polycrystalline silicon films, as commonly used in passivating contact structures. Interestingly, recombination lifetimes in intrinsic amorphous silicon films can be significantly enhanced by a hydrogenation process. By contrast, recombination lifetimes in heavily-doped polycrystalline silicon films vary with different doping profiles for samples fabricated with different deposition techniques, but are not improved by hydrogenation.

This study explores the electronic properties of industrial antimony‐doped (Sb‐doped) n‐type silicon wafers, grown using the Recharged Czochralski (RCz) method, for photovoltaic applications. We examine the resistivity distribution along the RCz grown ingots, bulk minority carrier lifetime, and bulk implied voltages at maximum power point and open‐circuit () conditions. The impact of phosphorus diffusion gettering treatments on wafer quality is evaluated, alongside a comparison of iron (Fe) gettering rates in Sb‐doped and phosphorus‐doped (P‐doped) wafers. The results show that Sb‐doped wafers grown via the RCz method demonstrate a very high material quality, with bulk lifetimes and implied voltages approaching the Auger limit in the as‐grown state, except for the final ingot near maximum power point conditions, which requires a gettering step to reach the Auger limit. Additionally, we confirm that the dopant distribution is significantly more uniform along the Sb‐doped ingots than along comparable P‐doped RCz ingots. Our findings also show that Fe gettering rates by the films for cumulative annealing at 325°C in Sb‐doped and P‐doped RCz wafers are very similar. The high‐quality and uniform doping of the Sb‐doped ingots highlight their potential for high‐efficiency silicon solar cell production, as well as possible improvements in ingot yield and cost‐effectiveness.

Carrier-selective passivating contacts in crystalline silicon (c-Si) solar cells have expanded from doped silicon films to non-silicon wide-bandgap materials to reduce parasitic absorption and production costs. Titanium oxide (TiOx) has emerged as one of the most promising materials and has achieved high performance in c-Si solar cells. In this work, TiOx is explored as a passivation interlayer in hole-selective contacts rather than conventional electron-selective contacts. Theoretical calculations and experimental results demonstrate that negative charges and shallow states in TiOx, derived from oxygen vacancies (VO), enhance surface passivation and assist hole tunneling, respectively. As a strategy to modulate VO, forming gas annealing is performed to further improve hole selectivity. By incorporating the TiOx passivation interlayer into MoOx-based c-Si solar cells, we achieve an improved efficiency and stability of the device, with the highest efficiency of 21.28%. This work advances the understanding of TiOx as a promising material to enhance hole selectivity for c-Si solar cells.

This study reports on the electronic properties of industrial phosphorus-doped n-type silicon ingots for photovoltaic applications grown using the Recharged Czochralski method. The electronic quality is assessed via carrier lifetime measurements, both directly on the ingots and on passivated wafers, and via implied open-circuit (iVOC), and implied maximum power point (iV(MPP)) voltages. The wafers are studied in the as-grown state, and after various high temperature steps, including Tabula Rasa, phosphorus diffusion gettering, and boron diffusion. The material exhibited very high bulk quality, with bulk lifetimes up to 8 ms at an injection level of 5 x 10(14) cm(-3), and with iVOC (1-sun) values up to 750 mV, prior to any high temperature processing. A Tabula Rasa step did not significantly improve the wafer quality, indicating a low presence of oxygen-related defects in this material, consistent with the low interstitial oxygen content of below 5 x 10(17) cm(-3). However, phosphorus diffusion gettering improved the wafer quality, especially towards the tail end of each ingot, and at lower injection levels near maximum power point. Phosphorus diffusion gettering increased the iV(OC) (1-sun) of the wafers by around 5 mV, approaching the Auger limit. Additionally, a boron diffusion step had minimal impact on the bulk lifetimes. Overall, our findings suggest that these RCz-grown n-type wafers exhibit very high quality, approaching the Auger limit near open-circuit, and are well-suited for high-efficiency solar cells without the need for additional high-temperature processing.

The need to increase transparency in existing passivating contacts for crystalline silicon solar cells has motivated the development of transparent contacts based on transition metal oxides (TMOs). Among hole‐selective materials, molybdenum oxide (MoO x ) has achieved the greatest success so far. However, despite providing low contact resistivity, MoO x relies on an intrinsic hydrogenated amorphous silicon (a‐Si:H(i)) interlayer to achieve high levels of surface passivation and thus high open‐circuit voltage at a device level, partially defeating the objective of improved transparency. Herein, we report unprecedented performance for a‐Si:H‐free MoO x ‐based contacts by employing an alternative passivating interlayer based on a well‐engineered chlorine‐containing Al‐alloyed titanium oxide/titanium dioxide (Al y TiO x /TiO 2 )stack. The resulting Al y TiO x /TiO 2 /MoO x stack achieved record levels of passivation, reaching J 0 values as low as 16 fA cm −2 , closer to values reported for a‐Si:H‐based contacts, while maintaining lower contact resistivity, well below 100 mΩ cm −2 . Additionally, the stack presents improved transparency compared to a‐Si:H‐based contacts, with gains in short‐circuit current density of at least 0.8 mA cm −2 . The work pushes the performance of hole‐selective passivating contacts based on TMOs to new levels, enabling a record efficiency of 22.53% for cells with fully transparent hole‐selective passivating contacts. This work serves as an important stepping stone toward low‐thermal‐budget, simple manufacturing of high‐efficiency solar cells.

Passivating contacts based on transition metal oxides are of great interest for applications in crystalline silicon (c‐Si) solar cells due to their improved optical transparency and potential cost reduction. In this work, the contact resistivity and passivation for thermally evaporated Cu 2 O are investigated and optimized, with and without an Al 2 O 3 interlayer, as a hole‐selective contact to c‐Si. Additionally, we implement an Al y TiO x /TiO 2 stack as a novel passivating tunnel interlayer for hole‐selective contacts, achieving an implied open‐circuit voltage iV oc of 630 mV and a record‐low J 0 of 212 fA cm −2 while maintaining a contact resistivity ρ c of 62 mΩ cm 2 . A record‐low ρ c of 8 mΩ cm 2 for Cu 2 O‐based contacts is also demonstrated at the expense of passivation. The addition of the interlayer resulted in a 2% absolute improvement in the efficiency of proof‐of‐concept c‐Si cells with full‐area rear Cu 2 O contacts, reaching 19.1%.The demonstration of this novel interlayer stack provides new avenues to improve the performance also of other hole‐selective passivating contacts.

Our research unveils strategies for developing exceptional TiOx-based passivating contacts, potentially replacing traditional Si-based ones with highly stable, transparent alternatives.

Amorphous thin-film TiOx prepared via atomic layer deposition (ALD) has been identified as one of the most promising materials for use in transparent passivating contacts in high-efficiency and low-cost crystalline silicon (c-Si) solar cells. As highlighted in this work, the passivation performance of ALD TiOx layers strongly depends on the metal precursor used, with films prepared using TiCl4 recently showing the best results. However, a full understanding of how such films achieve their high level of surface passivation has not yet been demonstrated. This study provides a clear demonstration that a key part of this passivation mechanism is due to chlorine (Cl) accumulation at the Si surface. This mechanism is demonstrated to be quite general in nature by showing how 2 nm of ALD TiOx (TiCl4 + H2O) can be applied as a capping layer for either ZnO or Al2O3 interlayers to dramatically reduce silicon surface recombination. Cl depth profiles obtained using secondary ion mass spectrometry confirm the presence of Cl extending through the depth of the interlayers with a peak at the silicon interface. Remarkably, this diffusion of Cl is observed following low-temperature (75 °C) deposition of the TiOx capping layer, without any subsequent thermal treatment. Contrary to earlier studies that treated residual Cl in ALD films as a general contamination issue, these findings reveal unequivocally that chlorine plays a crucial role in Si surface passivation and can be classed as an effective passivation element, similar to hydrogen in its ability to passivate Si dangling bonds. The outcomes of this research emphasize the importance of residual chlorine in enhancing the passivation of buried interfaces and provide additional motivation for employing metal chloride precursors for silicon surface passivation applications.

III–V semiconductors are among the highest performing materials for solar energy conversion devices. Exposing III–V semiconductors to a hydrogen plasma can improve optoelectronic properties and is a critical step in fabricating efficient InP solar cells. However, there is a limited understanding of the changes induced by hydrogen plasma exposure to the surface and in the bulk of III–V semiconductors. Herein, it is demonstrated that a 19.3% efficient p‐InP solar cell with a TiO 2 electron selective contact layer can be achieved by exposing the InP substrate to hydrogen plasma. Detailed investigations employing ultraviolet photoelectron spectroscopy and capacitance–voltage measurement unveil that the hydrogen plasma exposure on p‐InP leads to charge carrier polarity inversion in the near‐surface region (charge inversion layer) while simultaneously reducing the carrier concentration (charge‐depleted layer) in the bulk. The study provides important insights into the impact of hydrogen plasma exposures on InP which may lead to more efficient optoelectronic devices such as solar cells, photodetectors, light‐emitting diodes, and photoelectrochemical cells.

We investigate the potential of ultra-thin HfO2 films grown by atomic layer deposition for passivating contacts to silicon focusing on variations in film thickness and post-deposition annealing temperature. A peak in passivation quality - as assessed by carrier lifetime measurements - is reported for 2.2 nm thick films annealed at 475 degrees C, for which a surface recombination velocity <1 cm/s is determined. For films <2.2 nm thick, there is a marked decrease in passivation quality. X-ray diffraction highlights a change from crystallised monoclinic to amorphous HfO2 as film thickness decreases from 12 nm to 2.2 nm. Kelvin probe results indicate that as-deposited 2.2-12 nm films have similar effective work functions, although the work function of 1 nm films is considerably lower. Upon post-deposition annealing in vacuum, all films exhibit a reduction in effective work function at temperatures coincident with the onset of passivation in air-annealed samples. An initial investigation into the contact resistivity in a passivating contact structure utilizing HfO2 reveals a strong post-deposition annealing temperature dependence, with the lowest resistance achieved below 375 degrees C, followed by a decrease in performance as temperature increases towards the optimal temperature for passivation (475 degrees C). Limitations of the contact structure used are discussed.

Passivating contact technologies are essential for fabricating high‐efficiency crystalline silicon (c‐Si) solar cells, and their application and incorporation into manufacturing lines has ranked as a hot topic of research. Generally, ideal passivating contacts should combine excellent electrical contact, outstanding surface passivation, and high optical transparency. However, addressing all these criteria concurrently is challenging since it is unlikely for any single material to exhibit both efficient carrier transport and surface‐defect passivation while demonstrating negligible parasitic absorption. In this work, several earth‐abundant, wide‐bandgap materials are combined to engineer high‐quality transparent electron‐selective passivating contact structures capable of overcoming these obstacles. A highly transparent Al y TiO x /ZnO/TiO 2 stack with a total thickness of 3 nm, prepared by atomic layer deposition, is shown to provide a close‐to‐ideal passivating contact to Si surfaces by enabling dual functions of remarkable silicon surface passivation (with an effective minority carrier lifetime of 12.3 ms, an implied open‐circuit voltage of 730 mV, and a surface recombination current density prefactor of 2.6 fA cm −2 ), combined with efficient carrier transport with a very low contact resistivity of 3.4 mΩ cm 2 . These results demonstrate that low‐cost silicon interface‐engineering strategiesbased on transition metal oxides can push c‐Si solar cell performance to its theoretical limits.

Cu2O is an important p-type semiconductor material with applications in thin-film transistors, photovoltaics, and water splitting. For such applications, pinhole-free and uniform thin films are desirable, thus making atomic layer deposition (ALD) the ideal fabrication technique. However, existing ALD Cu precursors suffer from various problems, including limited thermal stability, fluorination, or narrow temperature windows. Additionally, some processes result in CuO films instead of Cu2O. Therefore, it is important to explore alternative precursors and processes for ALD of Cu2O thin films. In this work, we report the successful deposition of Cu2O using copper acetylacetonate as a precursor and a combination of water and oxygen as reactants at 200 °C. Saturation of the deposition rate with precursor and reactant dose time was observed, indicating self-limiting behavior, with a saturated growth-per-cycle of 0.07 Å. The Cu2O film was polycrystalline and uniform (RMS roughness ∼2 nm), with a direct forbidden bandgap of 2.07 eV and a direct allowed bandgap of 2.60 eV.

Improving the passivation of contacts in silicon solar cells is crucial for reaching high‐efficiency devices. Herein, the impact of the contact work function on the obtained passivation is examined and quantified using a novel method—quasi‐steady‐state photoluminescence—which provides access to the surface saturation current density after metallization ( J 0s,m ). The obtained J 0s,m indicates an improvement of the surface passivation when contacts with high work function are applied onto Si wafers passivated with aluminum oxide, regardless of the wafer doping type. This improvement is mainly due to the amplification of the imbalance between the electron and hole concentrations near the Si interface. The passivation quality is reduced when using contacts with low work function in which the recombination rate increases via the charge‐assisted carrier population control. Herein, the vital importance of selecting suitable metals to minimize contact recombination in high‐efficiency solar cells is pointed.

Impedance spectroscopy (IS) is a powerful characterization technique that is commonly applied to organic, perovskite, and thin-film solar cells. However, it has not been widely applied to solar cells based on crystalline silicon (c-Si), which is by far the most relevant commercial technology, and particularly not to modern, high-efficiency silicon devices. In this work, we demonstrate the application of the IS technique to a 21.25% efficient c-Si solar cell featuring SiOx/poly-Si rear passivating contacts. This type of cell architecture is structurally similar to that of current high-efficiency industrial devices. The investigated cell was measured over a wide range of frequencies under illuminated open-circuit conditions and under different DC biases in darkness. The resistive and capacitive components associated with the p+–n junction and at n+-poly–n low–high junction, which cannot be resolved by standard DC measurements, are readily distinguished by the IS method. These parameters allowed for the determination of junction time constants and lifetimes. We find that the lifetimes derived from IS measurements performed under open-circuit illuminated conditions are in excellent agreement with the carrier recombination lifetime under illumination. Our findings demonstrate that IS is a promising technique to explore various dynamic properties of high-efficiency c-Si solar cells.

Characterisation and optimization of next-generation silicon solar cell concepts rely on an accurate knowledge of intrinsic charge carrier recombination in crystalline silicon. Reports of measured lifetimes exceeding the previous accepted parameterisation of intrinsic recombination indicate an overestimation of this recombination in certain injection regimes and hence the need for revision. In this work, twelve high-quality silicon sample sets covering a wide doping range are fabricated using state-of-the-art processing routes in order to permit an accurate assessment of intrinsic recombination based on wafer thickness variation. Special care is taken to mitigate extrinsic recombination due to bulk contamination or at the wafer surfaces. The combination of the high-quality samples with refined sample characterisation and lifetime measurements enables a much higher level of accuracy to be achieved compared to previous studies. We observe that reabsorption of luminescence photons inside the sample must be accounted for to achieve a precise description of radiative recombination. With this effect taken into account, we extract the lifetime limitation due to Auger recombination. We find that the extracted Auger recombination rate can accurately be parameterized using a physically motivated equation based on Coulomb-enhanced Auger recombination for all doping and injection conditions relevant for silicon-based photovoltaics. The improved accuracy of data description obtained with the model suggests that our new parameterisation is more consistent with the actual recombination process than previous models. Due to notable changes in Auger recombination predicted for moderate injection, we further revise the fundamental limiting power conversion efficiency for a single-junction crystalline silicon solar cell to 29.4%, which is within 0.1%abs compared to other recent assessments.