Periodic Reporting for period 1 - TEMCO (Transmission Electron Microscopy Studies of Charge-Density Waves in YBa2Cu3Oy and ErTe3)
Reporting period: 2022-10-01 to 2024-09-30
Charge-density waves (CDWs) are electronic modulations that break translational symmetry and can affect thermodynamic, optical and electronic properties. Typically coupled to a periodic lattice distortion (PLD), CDWs are characterised by parameters such as their amplitude, the wave vector, the phase, correlation lengths etc. of the electronic charge modulation. CDWs preferentially occur in low-dimensional systems with weakly coupled one-dimensional chains or in layered materials. When such layered materials host uniaxial modulations, these can locally break rotational symmetry, forming an anisotropic electronic state.
Electron microscopy is a suitable probe to study the PLD associated with CDWs with high resolution. This technique is exciting because cryogenic electron microscopy is progressively advancing to lower and lower temperatures where quantum effects like high temperature superconductivity can be readily observed.
In particular, cuprates are a class of high temperature superconductors that host CDWs. When doping the Mott-insulating antiferromagnetic parent compound of cuprates with charge carriers (holes or electrons), these materials lose their electrical resistivity as they are cooled below relatively high
superconducting critical temperatures, Tc. The carrier doping does not only lead to superconductivity but also generates a number of complex electronic phases that emerge alongside or in competition with superconductivity. The competition between the CDW and superconductivity and their occurrence in a common doping and temperature range suggest that it is the same electrons that either participate in the CDW or form superconducting Cooper pairs. Our aim is to advance the understanding of the CDW phase with electron microscopy which will push the characterisation of the CDW structure down to the atomic scale at cryogenic temperatures. This will help to improve microscopic theories that also have to explain the origin of the other electronic phases of high temperature superconductors.
Unlike other cuprates which only show short-range CDW correlations, YBa2Cu3Oy (YBCO) hosts a long-range CDW phase. Strong competition with superconductivity prevents long-range order in zero field. Conventionally, the long-range CDW is reached by suppressing superconductivity with high magnetic fields, but recently it has been shown that uniaxial pressure can stabilise the long-range CDW even in zero field.This has opened an exciting avenue to study a truly long-ranged phase that is in fierce competition with superconductivity. How exactly superconductivity and charge ordering compete is largely unknown. Information from most techniques is indirect, as, e.g. the CDW correlation lengths from X-ray diffraction are average values that are determined over the beam size of hundreds of square micrometers. Fortunately, electron microscopy allows to map CDW patches and domains down to the atomic scale in real space and thus visualise the competition of superconductivity and CDWs directly.
This motivates our project “TEMCO” which proposes a “Transmission Electron Microscopy study of Charge Ordering in cuprate superconductors”.
Our objective is to study of the short-range CDW phase of YBCO and to induce the long-range CDW phase by tensile stress. Samples with doping close to p=0.12 (y=6.60) are most suitable as they have the highest CDW onset temperature and the largest CDW amplitude. As the CDW amplitude and correlation length grow with cooling, it is of interest to reach the range of 60-70 K, just above the superconducting Tc, which is rather challenging for electron microscopy. Reaching temperatures below Tc would allow visualising the impact of superconductivity on the CDW at the nano scale. We face the challenge that the c-axis coherence length in between different CuO2 bi-layers that host CDWs is only a couple of unit cells in the short-range CDW phase. Thus signals from layers across the specimen do not necessarily add up coherently and it is unclear whether the CDW can be directly seen by TEM. It might be necessary to improve the CDW signal by tuning the competition between the CDW and superconductivity. This is why we develop a technique to apply tensile stress to the sample.
Electron microscopy is a suitable probe to study the PLD associated with CDWs with high resolution. This technique is exciting because cryogenic electron microscopy is progressively advancing to lower and lower temperatures where quantum effects like high temperature superconductivity can be readily observed.
In particular, cuprates are a class of high temperature superconductors that host CDWs. When doping the Mott-insulating antiferromagnetic parent compound of cuprates with charge carriers (holes or electrons), these materials lose their electrical resistivity as they are cooled below relatively high
superconducting critical temperatures, Tc. The carrier doping does not only lead to superconductivity but also generates a number of complex electronic phases that emerge alongside or in competition with superconductivity. The competition between the CDW and superconductivity and their occurrence in a common doping and temperature range suggest that it is the same electrons that either participate in the CDW or form superconducting Cooper pairs. Our aim is to advance the understanding of the CDW phase with electron microscopy which will push the characterisation of the CDW structure down to the atomic scale at cryogenic temperatures. This will help to improve microscopic theories that also have to explain the origin of the other electronic phases of high temperature superconductors.
Unlike other cuprates which only show short-range CDW correlations, YBa2Cu3Oy (YBCO) hosts a long-range CDW phase. Strong competition with superconductivity prevents long-range order in zero field. Conventionally, the long-range CDW is reached by suppressing superconductivity with high magnetic fields, but recently it has been shown that uniaxial pressure can stabilise the long-range CDW even in zero field.This has opened an exciting avenue to study a truly long-ranged phase that is in fierce competition with superconductivity. How exactly superconductivity and charge ordering compete is largely unknown. Information from most techniques is indirect, as, e.g. the CDW correlation lengths from X-ray diffraction are average values that are determined over the beam size of hundreds of square micrometers. Fortunately, electron microscopy allows to map CDW patches and domains down to the atomic scale in real space and thus visualise the competition of superconductivity and CDWs directly.
This motivates our project “TEMCO” which proposes a “Transmission Electron Microscopy study of Charge Ordering in cuprate superconductors”.
Our objective is to study of the short-range CDW phase of YBCO and to induce the long-range CDW phase by tensile stress. Samples with doping close to p=0.12 (y=6.60) are most suitable as they have the highest CDW onset temperature and the largest CDW amplitude. As the CDW amplitude and correlation length grow with cooling, it is of interest to reach the range of 60-70 K, just above the superconducting Tc, which is rather challenging for electron microscopy. Reaching temperatures below Tc would allow visualising the impact of superconductivity on the CDW at the nano scale. We face the challenge that the c-axis coherence length in between different CuO2 bi-layers that host CDWs is only a couple of unit cells in the short-range CDW phase. Thus signals from layers across the specimen do not necessarily add up coherently and it is unclear whether the CDW can be directly seen by TEM. It might be necessary to improve the CDW signal by tuning the competition between the CDW and superconductivity. This is why we develop a technique to apply tensile stress to the sample.
To cool down samples to temperatures as low as 60-70 K the Ropers group has started a cooperation with the startup CondenZero which develops a dedicated TEM holder that is cooled by liquid helium. Liquid helium is stored in a dewar which is positioned on an actively damped stage to minimise vibrations that affect the stability of the holder and which would otherwise impede measurements with high spatial resolution. At the end of the funding period the development of the holder had not been finished, so all experiments were performed using a commercial liquid nitrogen cooled holder at temperatures above 100 K. The short-range CDW of YBCO is faint at this temperature and e.g. to observe it by x-ray diffraction one needs to turn to synchrotron sources such as the European Synchrotron Radiation Facility (ESRF). Unsurprisingly, we could not observe the short-range CDW above 100 K.
To strengthen the CDW and to potentially induce the long-range CDW phase with an order of magnitude more intense CDW diffraction peaks, we developed an in-situ strain technique. For this a sample is cut out from a bulk YBCO single crystal using a gallium Focused Ion Beam (Ga-FIB). Conventionally, a free-standing sample of 10-20 microns in size is thinned down until it becomes transparent to the electron beam of the TEM. Here, instead, the sample is cut out and platinum-welded into the empty window of a regular silicon membrane, in line with a copper rod (see figure). Upon cooling, tensile stress is produced by differential thermal contraction because the silicon window shrinks very little compared to the YBCO sample and the long copper rod. The largest tensile strain is produced in the thinned section of the YBCO sample. Although the strain cannot be tuned in-situ, in principle, very large strains can be achieved. The largest strain at about 104 K was estimated to be around 2.0 - 2.2 %. This is the largest tensile strain that has been reported in a bulk YBCO crystal.
These results have been presented at international conferences (Gordon Research Conference on Ultrafast Phenomena in Cooperative Systems 2024 in Lucca and Superstripes Conference 2024 in Ischia).
It is not well understood how the sample preparation using the Ga-FIB affects the sample. During the cutting, thinning and Pt-welding of the sample heat is produced. This heat could lead to oxygen loss and a reduced oxygen content (y) of the YBa2Cu3Oy sample. If a significant fraction is lost, the samples become non-superconducting and no longer host CDWs. Preliminary TEM measurements confirm the presence of oxygen but a quantitative estimation has not been attempted. As part of the cooperation with the Moll group at the Max Planck Institute for Structure and Dynamics of Matter in Hamburg, we have also prepared a YBCO sample that was attached to a cooling stage during the FIB fabrication process. This should significantly reduce any oxygen loss.
To strengthen the CDW and to potentially induce the long-range CDW phase with an order of magnitude more intense CDW diffraction peaks, we developed an in-situ strain technique. For this a sample is cut out from a bulk YBCO single crystal using a gallium Focused Ion Beam (Ga-FIB). Conventionally, a free-standing sample of 10-20 microns in size is thinned down until it becomes transparent to the electron beam of the TEM. Here, instead, the sample is cut out and platinum-welded into the empty window of a regular silicon membrane, in line with a copper rod (see figure). Upon cooling, tensile stress is produced by differential thermal contraction because the silicon window shrinks very little compared to the YBCO sample and the long copper rod. The largest tensile strain is produced in the thinned section of the YBCO sample. Although the strain cannot be tuned in-situ, in principle, very large strains can be achieved. The largest strain at about 104 K was estimated to be around 2.0 - 2.2 %. This is the largest tensile strain that has been reported in a bulk YBCO crystal.
These results have been presented at international conferences (Gordon Research Conference on Ultrafast Phenomena in Cooperative Systems 2024 in Lucca and Superstripes Conference 2024 in Ischia).
It is not well understood how the sample preparation using the Ga-FIB affects the sample. During the cutting, thinning and Pt-welding of the sample heat is produced. This heat could lead to oxygen loss and a reduced oxygen content (y) of the YBa2Cu3Oy sample. If a significant fraction is lost, the samples become non-superconducting and no longer host CDWs. Preliminary TEM measurements confirm the presence of oxygen but a quantitative estimation has not been attempted. As part of the cooperation with the Moll group at the Max Planck Institute for Structure and Dynamics of Matter in Hamburg, we have also prepared a YBCO sample that was attached to a cooling stage during the FIB fabrication process. This should significantly reduce any oxygen loss.
The sample preparation and validation of large tensile strain upon cooling provides a new platform to explore strain-enhanced CDWs using the electron microscope. Currently, the prepared samples await the critical tests a lower temperature around 60-70 K, where long-range CDW have previously been observed under uniaxial compression. Once the helium cooling by the startup CondenZero is installed, the Ropers group will be able to test whether CDW become visible with further cooling. If successful, this will be the first electron microscopy experiments showing the CDW of YBCO. In addition, it will provide the opportunity to perform diffraction experiments in the ultrafast transmission electron microscope (UTEM) of the Ropers group. Until now, ultrafst diffraction experiments have only been performed at specially equipped synchrotron facilities using X-rays. The long-range CDW has not been studied using time-resolved techniques, yet. High-resolution experiments could also be performed by collaborators at the Keimer and van Aken groups of the Max Planck Institute for Solid State Research in Stuttgart. Such experiments would visualise the CDW of YBCO in real space for the first time, as scanning tunneling experiments cannot be performed on the cleaved surface of convalently bonded YBCO.