Machine Learning-Based Detection of Graphene Defects with Atomic Precision
- PDF / 2,551,972 Bytes
- 13 Pages / 595.276 x 790.866 pts Page_size
- 71 Downloads / 255 Views
ARTICLE
Cite as Nano-Micro Lett. (2020) 12:181 Received: 18 June 2020 Accepted: 12 August 2020 © The Author(s) 2020
https://doi.org/10.1007/s40820-020-00519-w
Machine Learning‑Based Detection of Graphene Defects with Atomic Precision Bowen Zheng1, Grace X. Gu1 * * Grace X. Gu, [email protected] Department of Mechanical Engineering, University of California, Berkeley, CA 94720, USA
1
HIGHLIGHTS • A machine learning-based approach is developed to predict the unknown defect locations by thermal vibration topographies of graphene sheets. • Two prediction strategies are developed: an atom-based method which constructs data by atom indices, and a domain-based method which constructs data by domain discretization. • Our machine learning model can achieve approximately a 90% prediction accuracy on the reserved data for testing, indicating a promising extrapolation into unseen future graphene configurations.
ABSTRACT Defects in graphene can pro-
foundly impact its extraordinary properties,
J = ||Xw−y||2 + λ||w'||2
ultimately influencing the performances of graphene-based nanodevices. Methods to detect
15
defects with atomic resolution in graphene can be technically demanding and involve complex
jR
sample preparations. An alternative approach
5
is to observe the thermal vibration properties of the graphene sheet, which reflects defect information but in an implicit fashion. Machine
J
10
5 Graphene with unknown defects
10
15 iR
Energy distribution at thermal vibration
w1
w2
Machine learning: kernel ridge regression
ed
ict
ed
Pr
al
tu
Ac
Defect prediction
learning, an emerging data-driven approach that offers solutions to learning hidden patterns from complex data, has been extensively applied in material design and discovery problems. In this paper, we propose a machine learning-based approach to detect graphene defects by discovering the hidden correlation between defect locations and thermal vibration features. Two prediction strategies are developed: an atom-based method which constructs data by atom indices, and a domain-based method which constructs data by domain discretization. Results show that while the atom-based method is capable of detecting a single-atom vacancy, the domain-based method can detect an unknown number of multiple vacancies up to atomic precision. Both methods can achieve approximately a 90% prediction accuracy on the reserved data for testing, indicating a promising extrapolation into unseen future graphene configurations. The proposed strategy offers promising solutions for the non-destructive evaluation of nanomaterials and accelerates new material discoveries. KEYWORDS Machine learning; Graphene; Defects; Molecular dynamics; Nanomaterials
Vol.:(0123456789)
13
181
Page 2 of 13
1 Introduction Graphene, due to its extraordinary electrical [1–3], thermal [4–6], and mechanical [7–10] properties, has been widely used as building blocks in high-performance nanoelectromechanical systems (NEMS) [11, 12], stretchable electronics [13, 14], supercapacitors [15, 16],
Data Loading...
