Descrizione del progetto
Membrane ZIF ad alte prestazioni
Mentre le emissioni di CO2 distruggono l’ambiente, i metodi economici per separare il composto dalle relative miscele di gas sono considerati una delle maggiori sfide ambientali del nostro secolo. Un’alternativa ai metodi attuali sono le separazioni basate su membrana. Il progetto SmartDeZIgn, finanziato dall’UE, fornisce un nuovo metodo per la progettazione di strutture di imidazolo zeolitico (zeolitic imidazolate frameworks, ZIF) per membrane selettive di CO2. Gli esperti mirano a sviluppare uno strumento computazionale basato su metodi di apprendimento automatico che selezionerà tutti i metalli e i linker adatti in combinazione con le centinaia di topologie ZIF disponibili.
Obiettivo
With CO2 emissions being an eminent threat of unprecedent global impact, cheap ways to separate it from related gas mixtures are regarded as one of the biggest environmental challenges of our century. One alternative to the current methods is membrane-based separations. However, with today’s available materials, membranes are trapped in an upper boundary permeability-selectivity performance, below the target values of industry related applications. Zeolitic-imidazolate frameworks (ZIFs) can lead to the development of membranes with high performance due to their functionalization that alters their separation performance. They haven’t achieved the status of game changer materials, though, due to limited knowledge of the structural modification-separation performance correlation. Although there are indications that replacement of the organic linker or the metal in ZIFs, affects considerably the diffusivity and separation of gases, no systematic investigation has been carried towards this direction.
I propose a novel method for the design of ZIFs of unprecedented selectivity for CO2 urgent separations: H2/CO2, CO2/N2 and CO2/CH4. The design will be based on the substitution of the organic linker and/or the metal centers of ZIFs. I will develop a computational tool based on machine learning methods which will screen all the suitable metals/linkers in combination with the hundreds of available ZIF topologies. The algorithm’s goal will be to find the missing correlation between these replacements and their impact on the separation efficiency of ZIFs. To achieve this, and contrary to the current screening machine learning-based methods, which focus solely on “static” host-guest interactions (sorption), my algorithm will take into account also the diffusivity (the governing mechanism in membrane-based separations), by adopting realistic structural flexibility response. This will facilitate the design of the optimum material for the three separations.
Campo scientifico
Programma(i)
Argomento(i)
Meccanismo di finanziamento
MSCA-IF-EF-RI - RI – Reintegration panelCoordinatore
15341 Agia Paraskevi
Grecia