Designed Synthesis of Zeolites for Environmental and Biorenewables Catalysis

Lead Research Organisation: University College London
Department Name: Chemistry

Abstract

Abstracts are not currently available in GtR for all funded research. This is normally because the abstract was not required at the time of proposal submission, but may be because it included sensitive information such as personal details.

Publications

10 25 50

publication icon
Hoffman A (2023) A Critical Assessment on Calculating Vibrational Spectra in Nanostructured Materials in Journal of Chemical Theory and Computation

publication icon
Vamvakeros A (2020) DLSR: a solution to the parallax artefact in X-ray diffraction computed tomography data in Journal of Applied Crystallography

 
Description Operando Kerr gate Raman spectroscopy has been further developed and successfully applied to the characterisation of zeolites and supported catalysts under reaction conditions. Further insights into deactivation mechanisms due to the rate and nature of coke species has been observed. This insight has again been underpinned by the combination with theory. Outside of the work on zeolites, a series of papers have been published on using artificial intelligence to process and interpret large volumes of diffraction data allowing insight which hitherto had not been physically possible. This has also enabled the performing experiments that could not be done before.
Exploitation Route The infrastructure to perform in situ/operando Kerr gate Raman spectroscopy is now available to a wider number of users. As our design criteria for the development of zeolites with enhanced stability for the MTO process.
Sectors Chemicals

Energy

Environment

Manufacturing

including Industrial Biotechology

 
Description A self-funded PhD who joined the project and who had an interest to apply machine learning (ML) to help process large volumes of in situ spectroscopy data found themselves unable to progress with the project on joining in January 2020. As such they began working with other members of the group on X-ray imaging data (particularly useful when developing assisted ML methods for image interrogation) and successfully developed processing pipelines, software and papers. The data processing pipelines have been made publicly available and the work has been used as a basis for software commercialisation.
First Year Of Impact 2021
Sector Chemicals,Energy
Impact Types Economic

 
Description Enabling Chemical Tomography of Large Objects
Amount £142,620 (GBP)
Funding ID 106003 
Organisation Innovate UK 
Sector Public
Country United Kingdom
Start 03/2020 
End 04/2021
 
Description Phase Identification in Chemical Imaging using Artificial Intelligence
Amount £126,129 (GBP)
Funding ID 106017 
Organisation Innovate UK 
Sector Public
Country United Kingdom
Start 03/2020 
End 01/2021
 
Description The UK Catalysis Hub - 'Science': 1 - Optimising, predicting and designing new Catalysts
Amount £300,000 (GBP)
Funding ID The UK Catalysis Hub - 'Science': 1 - Optimising, predicting and designing new Catalysts 
Organisation United Kingdom Research and Innovation 
Sector Public
Country United Kingdom
Start 11/2022 
End 10/2025
 
Title Development of operando kerr gate Raman capabilities. 
Description A setup and procedure for measuring zeolite catalysts using the Kerr-gate Raman during reaction (i.e. at temperature and with gas flow) has been developed. This has been published in the methods section of the following output: I. Lezcano-Gonzalez et al. Nature Materials volume 19, pages 1081-1087(2020) 
Type Of Material Improvements to research infrastructure 
Year Produced 2020 
Provided To Others? Yes  
Impact As mentioned in the narrative, this result prompted interest in commercial companies for using/exploiting this capability. From an academic perspective, there is greater interest in using these capabilities for examining challenging catalysis problems at earlier TRL levels (i.e. 0/1).. 
URL https://www.clf.stfc.ac.uk/Pages/Kerr-Gated-Raman-Spectroscopy.aspx
 
Title A deep convolutional neural network for real-time full profile analysis of big powder diffraction data: Datasets 
Description Training and test datasets used in the manuscript entitled "A deep convolutional neural network for real-time full profile analysis of big powder diffraction data". 
Type Of Material Database/Collection of data 
Year Produced 2021 
Provided To Others? Yes  
Impact The data which form the basis of the publication entitled 'A deep convolutional neural network for real-time full profile analysis of big powder diffraction data' has been used to develop software which has now been commercialised by Finden Ltd and which has been used to improve their client service offering. 
URL https://zenodo.org/record/4664596
 
Title Data and code used for the paper: A scalable neural network architecture for self-supervised tomographic image reconstruction 
Description We present a lightweight and scalable artificial neural network architecture which is used to reconstruct a tomographic image from a given sinogram. A self-supervised learning approach is used where the network iteratively generates an image that is then converted into a sinogram using the Radon transform; this new sinogram is then compared with the sinogram from the experimental dataset using a combined mean absolute error and structural similarity index measure loss function to update the weights of the network accordingly. We demonstrate that the network is able to reconstruct images that are larger than 1024 × 1024. Furthermore, it is shown that the new network is able to reconstruct images of higher quality than conventional reconstruction algorithms, such as the filtered back projection and iterative algorithms (SART, SIRT, CGLS), when sinograms with angular undersampling are used. The network is tested with simulated data as well as experimental synchrotron X-ray micro-tomography and X-ray diffraction computed tomography data. 
Type Of Material Data analysis technique 
Year Produced 2023 
Provided To Others? Yes  
Impact The approach is now being commercialised via a start-up company Finden Ltd 
URL https://github.com/robindong3/SD2I
 
Description JM 
Organisation Johnson Matthey
Country United Kingdom 
Sector Private 
PI Contribution We are measuring model catalysts provided by JM to benchmark our novel measurement techniques in order to see if we can understand further why they work or fail.
Collaborator Contribution Samples are provided by JM for investigation. JM will also provide pathways to impact for the research that is done should it prove of commercial interest. Regular meetings, coordinated by JM, ensures mutual and beneficial progress throughout.
Impact None as of yet.
Start Year 2019