Perovskites are a growing family of photovoltaic materials that have created a vast amount of research since their recent discovery. Due to their tunable electronic and optical properties, they are considered state-of-the art materials for solar cell and laser devices. Up to now, most work has focused on polycrystalline thin films A holy grail in this field is the development of ultrathin large-size single crystalline nanosheets which have low defect density and superior properties than their polycrystalline counterparts. “Looking at silicon as the exemplar material, it is clear that the availability of large single crystals with low point defect and dislocation densities is the key enabler for most of our ICT technologies. For perovskites, we need to have a source of large-area ultrathin materials that are compatible with standard lithography for optoelectronic device production”, says Prof. John Donegan, the leader of the semiconductor photonics group and this project.
In spite of prior reports on the fabrication of perovskite nanosheets, the synthetic control over this material is still quite limited compared with other semiconductors. Several methods have been explored to synthesize bulk perovskite single crystals but the thickness is also large, which need further processing for practical device applications. Ultrathin single crystals are also reported but the sizes are less than 1 mm. Here in this project we have tackled this challenge. The work was carried out under the AMBER industry research programme in the CRANN insitute in Trinity College Dublin in collaboration with the Thermal Management Research Group, of Nokia Bell Labs.
“In an attempt to make PbBr2 nanowires, we unintentionally obtained these shiny crystals in solution at room temperature. When we drop-cast these nanosheets and measured the thickness of them, to our surprise, they are only 100 nm or less in thickness, however, the lateral size can go up to millimeter in the first try”, says Jian-Yao Zheng, the former postdoctoral researcher in CRANN & AMBER and the first author of the paper. In order to get centimeter-size nanosheets, we carefully control the concentration and the temperature of the system, then you can see the growth of large colorful nanosheets. The growth occurred at very low temperature (about 40 ℃), the solution can be heated up in an oven or by an infrared lamp. Therefore, we can observe the in-situ growth of crystals in real time under an optical microscope..
The growth of nanosheets is highly anisotropic, with an aspect ratio reaching 100,000. In collaborating with Prof. Stefano Sanvito, the director of CRANN as well as the expert in molecular dynamics theory, we find the very different formation energy of different crystal facets can lead to the preferential two dimensional growth, resulting in the formation of nanosheets. It is noteworthy that (1) the preparation is low-cost, fast, with environment friendly with simple solvents (ethanol and acetone) and all the chemicals can be recycled, we don’t need bulky equipment such as chemical vapor deposition (CVD) system; (2) The as-prepared crystals are free-standing and are found floating in the solvent, you can scoop them up easily onto any substrate. These lead halide nanosheets are much more stable than perovskite materials which are sensitive to heat, moisture, and oxygen. Lead halide is immune to these environmental factors and can be stored for a long time without degradation. They can be converted to perovskite whenever necessary to incorporate into optoelectronic devices.
These PbBr2 nanosheets are single crystalline. However, following the conversion to perovskite, we cannot be sure about the crystalline nature of the converted material. We do not have convincing evidence that the materials remain single crystal, which still need much more rigorous experimental work. X-ray diffraction (XRD) results show clear and sharp peaks with both lead halide and perovskite nanosheets. We did transmission electron microscopy (TEM) and selected-area electron diffraction (SAED) characterization on these converted nanosheets. The perovskite nanosheets are not stable on exposure to the electron beam, they suffer from the degradation, therefore, we are still not quite sure if our nanosheets are single-crystalline before exposing them to the intense electron beam. Further effort can be made on either improving the conversion process or the extensive study of the crystal nature of these samples after conversion.
Once fabricated, we are very interested in the nanolasers and photodetectors made from these nanoscale perovskite materials. By utilizing focused ion beam (FIB) milling in our Advanced Microscopy Laboratory (AML), we are able to make nice patterns and arrays out from these large nanosheets. Dr. Hugh G. Manning and Prof. John J. Boland from CRANN helped with the characterization of photoconducting behavior of these samples. All of these devices are working with 100% device yield, indicating the high quality of the perovskite materials.
Last but not least, can we make even larger nanosheets? The answer is yes. Firstly, the precursors should be continuously supplied during the growth. Secondly, the solvent (acetone and ethanol) is volatile and suffer from the loss during the synthesis, resulting in the change of concentration of precursor, which is very important to control the growth equilibrium. An improved apparatus will facilitate the growth of nanosheets up to wafer-scale while the thickness remains at nanoscale, which should be extremely interesting both in fundamental and practical sense and offer a novel, facile and scalable strategy to make perovskite optoelectronic devices, this is our ambitious hypothesis and we hope it will be realized in the near future.
Figure: The PbBr2 nanosheets form easily in solution and grow over time to be 1 cm or more in diameter and less than 100 nm in thickness. Both photodiode and laser devices were fabricated once the PbBr2 was converted to perovskite materials.
Please sign in or register for FREE
If you are a registered user on Nature Portfolio Engineering Community, please sign in