Irea Mosquera

PhD researcher working on energy materials at @ImpMaterials

[Irea Mosquera @ireaml]

@nielsen_rs @lonepair Thank you Rasmus!

[Irea Mosquera @ireaml]

Point defects are often modelled at 0 K instead of the material synthesis or operating temperatures. But how accurate is this approximation? Explored with machine learning force fields in https://t.co/WjE9G0QEB7 - @lonepair & Johan

[Seán Kavanagh @Kavanagh_Sean_]

@ireaml's comprehensive study of pre-trained universal MLFFs for defect structure searching with ShakeNBreak is now published open-access here.

Many useful takeaways and insights to challenges for MLFFs with defects!
https://t.co/o7ucVKSgXJ

[Seán Kavanagh @Kavanagh_Sean_]

Our comprehensive i̶n̶-̶h̶o̶u̶s̶e̶ defect modelling python package 𝙙𝙤𝙥𝙚𝙙 is now fully live!⚛️

𝙙𝙤𝙥𝙚𝙙 implements the defect
simulation workflow in an efficient and user-friendly, yet powerful and fully-flexible, manner.

See 🧵 for features (1/n)
https://t.co/gdGKcMKd0b

[Kasper Tolborg @KasperTolborg]

📢Two open positions in my group!

Looking for a Postdoc or a PhD in computational materials chemistry? I have two open positions in my group at Aalborg University, Denmark.

Postdoc: https://t.co/xwMhQ4lUEG

PhD: https://t.co/u4oVCKO6Ub

Please share with anyone interested!

[Irea Mosquera @ireaml]

@Debmalya_UMN @prashungorai @Kavanagh_Sean_ @alexganose @lonepair We haven't tried defect clusters, but we expect the approach to also work for them! If targeting large/complex clusters, it might require more training configurations. There was a recent study about it: https://t.co/88DP8IRFZ1

[Irea Mosquera @ireaml]

Defect structure searching can be computationally expensive💸, especially for high-throughput studies or materials with complex PESs. Can we accelerate it with universal machine learning force fields? Explored in https://t.co/dbfHDwzKtu @Kavanagh_Sean_ @alexganose @lonepair

[Chem Soc Rev @ChemSocRev]

Don't miss this #OpenAccess Tutorial review by @ireaml, @lonepair, Seán Kavanagh, Johan Klarbring and Kasper Tolborg

@ImpMaterials

"Imperfections are not 0 K: free energy of point defects in crystals"

Part of our Pioneering Investigators collection

🔗 https://t.co/oFtnOcprcE

[Irea Mosquera @ireaml]

Defect calculations are often performed at 0 K - yet finite temperature effects can be important!🌡️ Explored in our Tutorial Review targeting the main contributions to defect free energies - with @Kavanagh_Sean_ @KasperTolborg Johan Klarbring & @lonepair https://t.co/P7Lxcs4opI

[Volker Deringer @vl_deringer]

Materials design theme continues strong at @RoySocChem #MC16 - among many highlights on day 2 here’s @HamishY_Sci @kieslichgroup @ireaml 😀

[David Scanlon @scanlond81]

.@ireaml presenting a masterclass in how to model defects correctly using computational techniques at #MC16. @UCLChemistry @ImpMaterials @ucl_arc @lonepair @Kavanagh_Sean_ @tyc_london

[Cecilia Clementi @CecClementi]

#PsikCecam23 An intense week of exciting science with many great people. Thanks to my co-organizers @MicheleCeriotti, Gabor Csanyi, and Lixin Sun.

[Zhenzhu Li @Zhenzhu__Li]

A great chance to come and join @ImpMaterials @tyc_london at Imperial College London to work on materials informatics. There are two new postdoctoral positions available in the @lonepair group! https://t.co/N8s5qNjRvF!
Feel free to contact!

[Seán Kavanagh @Kavanagh_Sean_]

Finally online!🤩
@ireaml's tour de force on symmetry-breaking and energy-lowering reconstructions at defects across many materials/chemistries🧪
(missed by the standard supercell approach!!)

You won't find this if you don't go looking!
➡️ ShakeNBreak🔍
https://t.co/ZHAUT9hSzl

[JOSS @JOSS_TheOJ]

Just published in JOSS: 'ShakeNBreak: Navigating the defect configurational landscape' https://t.co/Wm8gErR9M3

[Seán Kavanagh @Kavanagh_Sean_]

Another presentation, another GIF...
My favourite thing about being in 2 groups is getting to do twice as many presentations! 🤥
Today's GIF from the @lonepair group meeting featured the ShakeNBreak CLI and its ease-of-use 👨‍💻

https://t.co/Yv1llpEApH @ireaml @scanlond81

[Seán Kavanagh @Kavanagh_Sean_]

I had way too much fun messing around with PowerPoint while procrastinating making my slides for the @scanlond81 SMTG group meeting last week 😅

Getting the ground-state defect structure is important!

https://t.co/XmmVbcPVXp
@ireaml @lonepair

[Seán Kavanagh @Kavanagh_Sean_]

Code to implement available here: https://t.co/68BdsxJCNm
Compatible with VASP, CASTEP, FHI-AIMs, Quantum Espresso and CP2K, and takes only one or two lines of code to run! 🏃
Docs and tutorials incoming! 👷

[Aron @lonepair]

The implication of this study is that a large fraction of the computational defect literature needs to be revisited. Bold work by @ireaml & @Kavanagh_Sean_