Aerospace & Defense Technology - October 2024 - 32

Tech Briefs
Using Artificial Intelligence to Find the Polymers of
the Future
Georgia Tech researchers are using artificial intelligence to accelerate materials discovery.
Georgia Tech University, Atlanta, GA
N
ylon, Teflon, Kevlar. These are just a
few familiar polymers - large-molecule
chemical compounds - that have
changed the world. From Teflon-coated
frying pans to 3D printing, polymers are
vital to creating the systems that make
the world function better.
Finding the next groundbreaking
polymer is always a challenge, but now
Georgia Tech researchers are using artificial
intelligence (AI) to shape and
a
Pool of new
polymers to be
considered
b
Catalogue of available monomes
Diols + dihalides
Diacids + diols
Diamines + diacids
Dianhydrides + diamines
Cyclic olefins
Select, reject,
and transform
n
The polyVERSE
Polyesters
Polyesters
Polyamides
Polyimides
ROMP-based polymers
transform the future of the field. Rampi
Ramprasad's group develops and adapts
AI algorithms to accelerate materials
discovery.
This summer, two papers published
in the Nature family of journals highlight
the significant advancements
and success stories emerging from
years of AI-driven polymer informatics
research. The first, featured in Nature
Reviews Materials, showcases recent
ML-based property
prediction
Screen polymers for
examination
breakthroughs in polymer design across
critical and contemporary application
domains: energy storage, filtration
technologies, and recyclable plastics.
The second, published in Nature Communications,
focuses on the use of AI
algorithms to discover a subclass of
polymers for electrostatic energy storage,
with the designed materials undergoing
successful laboratory synthesis
and testing.
Chemical systhesis &
characterization
c
d
Multitask graph neural networks
Example dataset
Structure Properties
Structure 1 P1
I, ..., P1
Structure 2 P2
L
I, ..., P2
Structure M PM
L
Graphing, finerprinting, learning and prediction
Structure Graph
Structure 1 G1
Structure 2 G2
Fingerprint
F1
1, F1
F2
I, ..., PM
L
Structure M GM
The learning problem
Structure Properties
Structure X ?
Learning
(i.e., GNN)
Structure-property model
2, ... F1
1, F2
FM
N
2, ... F2
1, FM
N
P2
2, ... FM
N
Learning
(i.e., MLP)
Â
ƒ(Gi
) = Pi
1, Pi
2, ...,Pi
L
PM
Screening
Properties
P1
I, ..., P1
I, ..., P2
L
L
400
200
I, ..., PM
L
5
2
3
4 2
This study introduces the polyVERSE ( " polymers designed by Virtually-Executed Rule-Based Synthesis Experiments " ) paradigm (Fig. 1a), showcasing its success in
achieving these four attributes in the context of high-temperature dielectric search. In this AI-driven approach, polymers are generated from commercially-available
monomers using an expert system (Fig. 1b) and properties are estimated using multitask graph neural networks (Fig. 1c). These property estimations are used
to guide the selection (screening) of promising polymers from the larger population (Fig. 1d). (Image: Georgia Tech University)
32
mobilityengineeringtech.com
Aerospace & Defense Technology, October 2024
4
3
6
Temperature (°C)
Dielectric
constant
Band
gap (eV)

https://mobilityengineeringtech.com

Aerospace & Defense Technology - October 2024

Table of Contents for the Digital Edition of Aerospace & Defense Technology - October 2024