BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//wp-events-plugin.com//7.2.3.1//EN
TZID:America/New_York
X-WR-TIMEZONE:America/New_York
BEGIN:VEVENT
UID:0-7911@eng.ufl.edu
DTSTART;TZID=America/New_York:20251113T125000
DTEND;TZID=America/New_York:20251113T134000
DTSTAMP:20251201T210520Z
URL:https://www.eng.ufl.edu/news-events/events/mae-seminar-data-driven-dis
 covery-of-stable-surfaces-with-tailored-properties-via-equivariant-graph-n
 eural-n/
SUMMARY:MAE Seminar - Data-driven Discovery of Stable Surfaces with Tailore
 d Properties via Equivariant Graph Neural N
DESCRIPTION:Dear Undergraduate and Graduate Students\, Faculty\, and Staff\
 ,\nYou are invited! UF Department of Mechanical and Aerospace Engineering'
 s Seminar Series\nThis is a perfect opportunity to enjoy some coffee\, coo
 kies\, and captivating talks! These sessions feature amazing guest speaker
 s\, from academic trailblazers and industry movers to our very own faculty
  candidates showing off their expertise and fresh perspectives.\nCome for 
 the treats\, stay for the engaging discussions\, and connect with fellow M
 AE enthusiasts. Everyone is welcome!\nData-driven Discovery of Stable Surf
 aces with Tailored Properties via Equivariant Graph Neural Networks and Fo
 undational Interatomic Potentials\nNovember 13\, 2025\, at 12:50pm\nLocati
 on: MAE-A 303\nDr. Peter Schindler\nAssistant Professor at Northeastern Un
 iversity\nAbstract\nAccurately assessing the properties of materials’ su
 rfaces and their stability is critical for diverse applications such as he
 terogeneous catalysis\, electron emission technologies\, and interface eng
 ineering in semiconductors and batteries. The stability of a surface with 
 a particular Miller index and termination is governed by its cleavage or s
 urface energy. This property also governs the Wulff construction\, which d
 efines the equilibrium shapes of nanoparticles. Another crucial surface pr
 operty is the work function (i.e.\, the energy required to extract an elec
 tron from the surface) that determines the contact barrier at interfaces. 
 While first-principles calculations provide reliable predictions of these 
 surface properties\, their computational cost severely limits the explorat
 ion of the vast space of possible surfaces.\nEquivariant graph neural netw
 orks that enforce symmetry of three-dimensional Euclidean space (E3GNNs) h
 ave demonstrated accurate predictions of structure-property relationships 
 of materials and molecules. Instead of relying on invariant feature engine
 ering\, E3GNNs incorporate physical symmetries directly into the neural ne
 twork layers. The features of equivariant networks are thus more expressiv
 e and better able to capture local symmetries\, promising greater transfer
 ability and training data efficiency.\nAnother recent advance is the emerg
 ence of universal machine learning interatomic potentials (uMLIPs\, or rec
 ently termed “foundational interatomic potentials”) that promise appli
 cability across the entire periodic table and structural types. They enabl
 e rapid zero-shot predictions of materials and molecular properties that c
 an be derived from the potential energy surface or its derivative\, with l
 ittle to no need for additional training.\nIn this talk\, I will discuss (
 1) a novel E3GNN architecture that incorporates symmetry-breaking along th
 e surface normal\, for accurately predicting both the stability and work f
 unction of surfaces\, and (2) a benchmarking study of cleavage energy pred
 ictions for a wide range of state-of-the-art uMLIPs. Finally\, I will demo
 nstrate the capability of these approaches to speed up screening for stabl
 e surfaces with tailored properties by orders of magnitude.\nBiography\nDr
 . Peter Schindler is an Assistant Professor at Northeastern University lea
 ding the Data-Driven Renewables Research (D2R2) group that seeks to discov
 er novel materials for renewable energy applications using high-throughput
 \, quantum chemistry calculations\, and data-driven materials property pre
 dictions. He is a scientific advisory board member (and former senior scie
 ntist) at Aionics Inc.\, a startup that has become a recognized leader in 
 the field of battery informatics. His work has been featured on the covers
  of Advanced Materials\, ACS Energy Letters\, and Digital Discovery. Durin
 g his postdoctoral research at the University of Vienna and Stanford Unive
 rsity\, his research established renewable energy materials both computati
 onally and experimentally\, for which he was awarded the Erwin-Schrödinge
 r fellowship by the Austrian Science Fund (FWF). He received his Ph.D. in 
 physics from the University of Vienna\, working on thin-film semiconductor
  synthesis and characterization. His expertise in both experimental synthe
 sis and computational materials simulations enables him to carry out cutti
 ng-edge research at the intersection of the two fields.\nMAE Faculty Host:
  Dr. Jaeyun Moon
CATEGORIES:Events
END:VEVENT
BEGIN:VTIMEZONE
TZID:America/New_York
X-LIC-LOCATION:America/New_York
BEGIN:STANDARD
DTSTART:20251102T010000
TZOFFSETFROM:-0400
TZOFFSETTO:-0500
TZNAME:EST
END:STANDARD
END:VTIMEZONE
END:VCALENDAR