Skip to main content

A preparative approach to geometric effects in innovative solar cell types based on a nanocylindrical structure

Periodic Reporting for period 4 - SOLACYLIN (A preparative approach to geometric effects in innovative solar cell types based on a nanocylindrical structure)

Reporting period: 2020-03-01 to 2020-08-31

Two necessary requirements of solar cells are the efficient absorption of light and the subsequent efficient collection of the charge carriers generated. In planar devices, these two physical phenomena impose contradicting constraints on the only one experimentally variable geometric parameter, the thickness of the semiconductor layer. Thus, planar systems necessitate that materials parameters be exploited in the photovoltaic optimization, as well. In c-Si cells, light absorption is maximized by a large layer thickness (>100 µm), and a material of extreme purity maximizes carrier lifetimes and diffusion distances. In thin film cells, most prominently of the chalcopyrite family ('CIGS'), carrier collection is maximized by a small layer thickness (<100 nm), whereas a material of extraordinarily large light absorption coefficient is used.
These outstanding materials and their energy payback times put a hurdle on the widespread adoption of photovoltaics as a primary energy source. Unexpensive materials containing exclusively earth-abundant elements and processed without high-vacuum techniques cannot compete with the quality of c-Si and CIGS. Therefore, the optimization of photovoltaic devices based on them must exploit an additional experimentally adjustable parameter: Conceptually, this is the lateral size of structures elongated perpendicularly to the plane of the bulk device. In principle, the length of such structures can be adjusted to the material's optical density and their thicknesses to the diffusion lengths of the carriers. In reality, systems prepared on these principles to date are usually disordered and have not reached the efficiencies of planar types of cells.
Their disordered geometry prevents systematic investigation of the device physics, given that no well-defined geometric parameters can be addressed. Such a systematic investigation is possible if the semiconductor junction is organized in cylindrical nanostructures of tunable length and diameter. Before the start of the SOLACYLIN project, however, a direct, systematic and exhaustive investigation of the geometric effects on photovoltaic properties in solar cells of this nanorod structure had been missing. This may be related to the difficulties of preparing large parallel arrays of cylinders combining two semiconductors in a geometry that is well defined, tunable, and homogeneous over each sample.
"SOLACYLIN has developed preparative methods for making solar cells out of alternative (inexpensive, abundant, non-toxic) materials in a novel geometry consisting of parallel nanocylinders in ordered arrays, each of which combines three semiconductors arranged coaxially.
One piece of work has focused on the nanoporous matrices to be used as templates for the cylinders. We use ""anodization"", an electrochemical procedure used industrially for the surface treatment of some metals. In SOLACYLIN, we have explored simple methods for generating anodized arrays of pores that display a high degree of order and a homogeneous geometry from the very beginning. This contrasts with existing methods, in which either the pores start out disordered and only become ordered later, or expensive manufacturing techniques (lithography) are used to define the order preliminarily. In particular, we have established how to generate arrays of short (0.1 to 3 µm) pores directly on a transparent conductive substrate in a manner such that the materials deposited inside them are in direct electrical contact to the underlying substrate.
In another line of research, we have developed methods allowing for the homogeneous, conformal coating of such elongated pores based on surface chemical reactions, a strategy called ""atomic layer deposition"" (ALD). We have optimized methods for depositing thin silica films by ALD, into which we mix in small amounts of either aluminum or antimony. Upon thermal reduction using lithium vapor, the material is converted to amorphous silicon with either p or n doping as needed in photovoltaics. Further ALD work has been pursued towards controlling the surface reactions from 'precursors' dissolved in liquids, instead of bringing them from the gas phase in a vacuum chamber. This evolution has rendered novel types of ALD chemistry possible, and thereby, allowed for the ALD coatings of materials which to date have not been accessible. The 'solution ALD' (sALD) approach has enabled us to deposit materials inaccessible by traditional ALD, such as a solid hydride, a polymer, a metal-organic framework, and ionic solids, some of them with outstanding purity and crystallinity at room temperature. In particular, we have been able to generate hybrid perovskites of the type that is relevant to photovoltaics, both lead-based and antimony-based.
We have prepared functional perovskite solar cell prototypes in various geometries -- planar films, colloidal nanoparticles, and coaxial nanocylinders. We have proposed a material that is directly accessible by sALD and which could replace the lead-based perovskites used currently, thereby circumventing the two major impediments of lead-based perovskites, namely their toxicity and their instability in air. In parallel to this, we have also developed an ""extremely thin absorber"" (ETA) solar cell materials system based on antimony sulfide as the light absorber. The deposition of individual layers by ALD has enabled us to find by a systematic approach that the optimal thickness of antimony sulfide is 60 nm in a planar configuration. We have identified ZnSas an interfacial layer which provides proper adhesion and antirecombination barrier properties, with an optimized thickness of 0.6 nm. We have transferred this interface engineering to coaxial nanoyclindrical geometry, and varied the length of the cylinders as well as the thickness of each layer. We have explored a number of materials as alternatives to the classical semiconductors, in particular MoS2, HfS2, SnO2, and V2O5."
Firstly, the ability to generate parallel arrays of nanopores directly as a layer on a substrate, and with ordered from the beginning, will enable their use in a much greater variety of applications. Integrating them into devices, including on flexible substrates, is now possible for a variety of purposes, in particular photovoltaics but not limited to it -- batteries and sensors are among the other fields which could profit of this development.
Secondly, the novel ALD chemistry that we have advanced renders the coating method relevant to a range of novel fields, from protective organic coatings on displays to inorganic and hybrid semiconductors. The ALD community has reacted enthusiastically to the invention of sALD, which is now being adopted by several other groups worldwide. In particular, we predict that it will become of particular interest for the development of future embedded devices based on printed electronics. In essence, ALD including sALD has become a method for printing semiconductors with atomic precision.
Thirdly, our solar cells have established the importance of interfaces and interfacial layers in photovoltaic stacks built from alternative semiconductors. We have opened the door to a strategy in which not the absolute purity of outstanding semiconductors is in the focal point but the fundamental understanding and control of the interfaces between semiconductors.