Convolutional neural networks are constructed and validated for the crystal structure classification of simple binary salts such as the alkali halides. The inputs of the neural network classifiers are the local bond orientational order parameters of Steinhardt, Nelson, and Ronchetti [Phys. Rev. B 28, 784 (1983)], which are derived solely from the relative positions of atoms surrounding a central reference atom. This choice of input gives classifiers that are invariant to density, increasing their transferability. The neural networks are trained and validated on millions of data points generated from a large set of molecular dynamics (MD) simulations of model alkali halides in nine bulk phases (liquid, rock salt, wurtzite, CsCl, 5-5, sphalerite, NiAs, AntiNiAs, and β-BeO) across a range of temperatures. One-dimensional time convolution is employed to filter out short-lived structural fluctuations. The trained neural networks perform extremely well, with accuracy up to 99.99% on a balanced validation dataset constructed from millions of labeled bulk phase structures. A typical analysis using the neural networks, including neighbor list generation, order parameter calculation, and class inference, is computationally inexpensive compared to MD simulations. As a demonstration of their accuracy and utility, the neural network classifiers are employed to follow the nucleation and crystal growth of two model alkali halide systems, crystallizing into distinct structures from the melt. We further demonstrate the classifiers by implementing them in automated MD melting point calculations. Melting points for model alkali halides using the most commonly employed rigid-ion interaction potentials are reported and discussed.

1.
M.
Matsumoto
,
S.
Saito
, and
I.
Ohmine
,
Nature
416
,
409
(
2002
).
2.
D.
Zahn
,
Phys. Rev. Lett.
92
,
040801
(
2004
).
3.
J.
Anwar
and
D.
Zahn
,
Angew. Chem., Int. Ed.
50
,
1996
(
2011
).
4.
D.
Chakraborty
and
G. N.
Patey
,
Chem. Phys. Lett.
587
,
25
(
2013
).
5.
D.
Chakraborty
and
G. N.
Patey
,
J. Phys. Chem. Lett.
4
,
573
(
2013
).
6.
F.
Jiménez-Ángeles
and
A.
Firoozabadi
,
J. Phys. Chem. C
118
,
11310
(
2014
).
7.
N. E. R.
Zimmermann
,
B.
Vorselaars
,
D.
Quigley
, and
B.
Peters
,
J. Am. Chem. Soc.
137
,
13352
(
2015
).
8.
M.
Salvalaglio
,
C.
Perego
,
F.
Giberti
,
M.
Mazzotti
, and
M.
Parrinello
,
Proc. Natl. Acad. Sci. U. S. A.
112
,
E6
(
2015
).
9.
G.
Lanaro
and
G. N.
Patey
,
J. Phys. Chem. B
120
,
9076
(
2016
).
10.
G. C.
Sosso
,
J.
Chen
,
S. J.
Cox
,
M.
Fitzner
,
P.
Pedevilla
,
A.
Zen
, and
A.
Michaelides
,
Chem. Rev.
116
,
7078
(
2016
).
11.
M.
Fitzner
,
G. C.
Sosso
,
F.
Pietrucci
,
S.
Pipolo
, and
A.
Michaelides
,
Nat. Commun.
8
,
2257
(
2017
).
12.
H.
Jiang
,
P. G.
Debenedetti
, and
A. Z.
Panagiotopoulos
,
J. Chem. Phys.
150
,
124502
(
2019
).
13.
P. J.
Steinhardt
,
D. R.
Nelson
, and
M.
Ronchetti
,
Phys. Rev. B
28
,
784
(
1983
).
14.
H.
Tanaka
,
H.
Tong
,
R.
Shi
, and
J.
Russo
,
Nat. Rev. Phys.
1
,
333
(
2019
).
15.
P.-L.
Chau
and
A. J.
Hardwick
,
Mol. Phys.
93
,
511
(
1998
).
16.
P. M.
Larsen
,
S.
Schmidt
, and
J.
Schiøtz
,
Modell. Simul. Mater. Sci. Eng.
24
,
055007
(
2016
).
17.
P. M.
Piaggi
and
M.
Parrinello
,
J. Chem. Phys.
147
,
114112
(
2017
).
18.
J.
Behler
and
M.
Parrinello
,
Phys. Rev. Lett.
98
,
146401
(
2007
).
19.
P.
Geiger
and
C.
Dellago
,
J. Chem. Phys.
139
,
164105
(
2013
).
20.
C. L.
Phillips
and
G. A.
Voth
,
Soft Matter
9
,
8552
(
2013
).
21.
E.
Boattini
,
M.
Ram
,
F.
Smallenburg
, and
L.
Filion
,
Mol. Phys.
116
,
3066
(
2018
).
22.
F.
Leoni
and
J.
Russo
,
Phys. Rev. X
11
,
031006
(
2021
).
23.
C.
Dietz
,
T.
Kretz
, and
M.
Thoma
,
Phys. Rev. E
96
,
011301
(
2017
).
24.
Q.
Wei
,
R. G.
Melko
, and
J. Z.
Chen
,
Phys. Rev. E
95
,
032504
(
2017
).
25.
M.
Fulford
,
M.
Salvalaglio
, and
C.
Molteni
,
J. Chem. Inf. Model.
59
,
2141
(
2019
).
26.
V.
Bapst
,
T.
Keck
,
A.
Grabska-Barwińska
,
C.
Donner
,
E. D.
Cubuk
,
S. S.
Schoenholz
,
A.
Obika
,
A. W. R.
Nelson
,
T.
Back
,
D.
Hassabis
, and
P.
Kohli
,
Nat. Phys.
16
,
448
(
2020
).
27.
G. M.
Coli
and
M.
Dijkstra
,
ACS Nano
15
,
4335
(
2021
).
28.
R. S.
DeFever
,
C.
Targonski
,
S. W.
Hall
,
M. C.
Smith
, and
S.
Sarupria
,
Chem. Sci.
10
,
7503
(
2019
).
29.
V. F.
Hernandes
,
M. S.
Marques
, and
J. R.
Bordin
,
J. Phys.: Condens. Matter
34
,
024002
(
2021
).
30.
E. D.
Cubuk
,
S. S.
Schoenholz
,
J. M.
Rieser
,
B. D.
Malone
,
J.
Rottler
,
D. J.
Durian
,
E.
Kaxiras
, and
A. J.
Liu
,
Phys. Rev. Lett.
114
,
108001
(
2015
).
31.
S. S.
Schoenholz
,
E. D.
Cubuk
,
D. M.
Sussman
,
E.
Kaxiras
, and
A. J.
Liu
,
Nat. Phys.
12
,
469
(
2016
).
32.
S. S.
Schoenholz
,
E. D.
Cubuk
,
E.
Kaxiras
, and
A. J.
Liu
,
Proc. Natl. Acad. Sci. U. S. A.
114
,
263
(
2017
).
33.
A.
Ziletti
,
D.
Kumar
,
M.
Scheffler
, and
L. M.
Ghiringhelli
,
Nat. Commun.
9
,
2775
(
2018
).
34.
H.
Doi
,
K. Z.
Takahashi
,
K.
Tagashira
,
J.-i.
Fukuda
, and
T.
Aoyagi
,
Sci. Rep.
9
,
16370
(
2019
).
35.
T.
Terao
,
Soft Mater.
18
,
215
(
2020
).
36.
K.
Swanson
,
S.
Trivedi
,
J.
Lequieu
,
K.
Swanson
, and
R.
Kondor
,
Soft Matter
16
,
435
(
2020
).
37.
W. F.
Reinhart
,
A. W.
Long
,
M. P.
Howard
,
A. L.
Ferguson
, and
A. Z.
Panagiotopoulos
,
Soft Matter
13
,
4733
(
2017
).
38.
M.
Spellings
and
S. C.
Glotzer
,
AIChE J.
64
,
2198
(
2018
).
39.
E.
Boattini
,
M.
Dijkstra
, and
L.
Filion
,
J. Chem. Phys.
151
,
154901
(
2019
).
40.
C. S.
Adorf
,
T. C.
Moore
,
Y. J. U.
Melle
, and
S. C.
Glotzer
,
J. Phys. Chem. B
124
,
69
(
2019
).
41.
E.
Boattini
,
S.
Marín-Aguilar
,
S.
Mitra
,
G.
Foffi
,
F.
Smallenburg
, and
L.
Filion
,
Nat. Commun.
11
,
5479
(
2020
).
42.
J.
Paret
,
R. L.
Jack
, and
D.
Coslovich
,
J. Chem. Phys.
152
,
144502
(
2020
).
43.
S.
Becker
,
E.
Devijver
,
R.
Molinier
, and
N.
Jakse
,
Phys. Rev. E
105
,
045304
(
2022
).
44.
R.
Tamura
,
M.
Matsuda
,
J.
Lin
,
Y.
Futamura
,
T.
Sakurai
, and
T.
Miyazaki
,
Phys. Rev. B
105
,
075107
(
2022
).
45.
Y.
Wang
,
W.
Deng
,
Z.
Huang
, and
S.
Li
,
J. Chem. Phys.
156
,
154504
(
2022
).
46.
E. P. L.
Van Nieuwenburg
,
Y.-H.
Liu
, and
S. D.
Huber
,
Nat. Phys.
13
,
435
(
2017
).
47.
J. F.
Rodriguez-Nieva
and
M. S.
Scheurer
,
Nat. Phys.
15
,
790
(
2019
).
48.
J.
Carrasquilla
and
R. G.
Melko
,
Nat. Phys.
13
,
431
(
2017
).
49.
M.
Walters
,
Q.
Wei
, and
J. Z.
Chen
,
Phys. Rev. E
99
,
062701
(
2019
).
50.
F.
Musil
,
S.
De
,
J.
Yang
,
J. E.
Campbell
,
G. M.
Day
, and
M.
Ceriotti
,
Chem. Sci.
9
,
1289
(
2018
).
51.
J. C.
Schön
and
M.
Jansen
,
Comput. Mater. Sci
4
,
43
(
1995
).
52.
Ž. P.
Čančarević
,
J. C.
Schoen
, and
M.
Jansen
,
Chem. - Asian J.
3
,
561
(
2008
).
53.
K.
Doll
,
J. C.
Schön
, and
M.
Jansen
,
Phys. Chem. Chem. Phys.
9
,
6128
(
2007
).
54.
H. O.
Scheiber
and
G. N.
Patey
,
J. Chem. Phys.
154
,
184507
(
2021
).
55.
A.
Bach
,
D.
Fischer
, and
M.
Jansen
,
Z. Anorg. Allg. Chem.
635
,
2406
(
2009
).
56.
Y.
Liebold-Ribeiro
,
D.
Fischer
, and
M.
Jansen
,
Angew. Chem., Int. Ed.
47
,
4428
(
2008
).
57.
Ž.
Čančarević
,
J.
Schön
,
D.
Fischer
, and
M.
Jansen
,
Current Research in Advanced Materials and Processes
, Materials Science Forum Vol. 494 (
Trans Tech Publications Ltd.
,
2005
), pp.
61
66
.
58.
O. I.
Abiodun
,
A.
Jantan
,
A. E.
Omolara
,
K. V.
Dada
,
N. A.
Mohamed
, and
H.
Arshad
,
Heliyon
4
,
e00938
(
2018
).
59.
A.
Dhillon
and
G. K.
Verma
,
Prog. Artif. Intell.
9
,
85
(
2020
).
60.
I. S.
Joung
and
T. E.
Cheatham
,
J. Phys. Chem. B
112
,
9020
(
2008
).
61.
D. E.
Smith
and
L. X.
Dang
,
J. Chem. Phys.
100
,
3757
(
1994
).
62.
E.
Lindahl
,
B.
Hess
,
M.
Abraham
,
D.
van der Spoel
, and
GROMACS Development Team
, GROMACS Reference Manual Version 2019,
2019
.
63.
H. J. C.
Berendsen
,
J. R.
Grigera
, and
T. P.
Straatsma
,
J. Phys. Chem.
91
,
6269
(
1987
).
64.
W. L.
Jorgensen
,
J.
Chandrasekhar
,
J. D.
Madura
,
R. W.
Impey
, and
M. L.
Klein
,
J. Chem. Phys.
79
,
926
(
1983
).
65.
H. W.
Horn
,
W. C.
Swope
,
J. W.
Pitera
,
J. D.
Madura
,
T. J.
Dick
,
G. L.
Hura
, and
T.
Head-Gordon
,
J. Chem. Phys.
120
,
9665
(
2004
).
66.
M. P.
Tosi
and
F. G.
Fumi
,
J. Phys. Chem. Solids
25
,
45
(
1964
).
67.
F. G.
Fumi
and
M. P.
Tosi
,
J. Phys. Chem. Solids
25
,
31
(
1964
).
68.
M. J.
Abraham
,
T.
Murtola
,
R.
Schulz
,
S.
Páll
,
J. C.
Smith
,
B.
Hess
, and
E.
Lindahl
,
SoftwareX
1-2
,
19
(
2015
).
69.
E.
Lindahl
,
M. J.
Abraham
,
B.
Hess
, and
D.
van der Spoel
(
2019
). “
GROMACS 2019.6 Source Code
,” Zenodo.
70.

More recent versions of GROMACS are not compatible with the CBHM interaction potential as tabulated potentials are no longer supported.

71.
M.
Parrinello
and
A.
Rahman
,
J. Appl. Phys.
52
,
7182
(
1981
).
72.
B.
Xu
,
Q.
Wang
, and
Y.
Tian
,
Sci. Rep.
3
,
3068
(
2013
).
73.
G.
Bussi
,
D.
Donadio
, and
M.
Parrinello
,
J. Chem. Phys.
126
,
014101
(
2007
).
74.
U.
Essmann
,
L.
Perera
,
M. L.
Berkowitz
,
T.
Darden
,
H.
Lee
, and
L. G.
Pedersen
,
J. Chem. Phys.
103
,
8577
(
1995
).
75.
S. W.
de Leeuw
,
J. W.
Perram
, and
E. R.
Smith
,
Proc. R. Soc. London, Ser. A
373
,
27
(
1980
).
76.

GROMACS employs a long-range dispersion correction that assumes dispersion is given by a single interaction term proportional to r−6.

77.
W.
Mickel
,
S. C.
Kapfer
,
G. E.
Schröder-Turk
, and
K.
Mecke
,
J. Chem. Phys.
138
,
044501
(
2013
).
78.
T.
Stöttner
, “
Why data should be normalized before training a neural network
,” towardsdatascience.com/why-data-should-be-normalized-before-training-a-neural-network-c626b7f66c7d (
2019
).
79.
J. A.
van Meel
,
L.
Filion
,
C.
Valeriani
, and
D.
Frenkel
,
J. Chem. Phys.
136
,
234107
(
2012
).
80.

We trained several 0-0 NNs on order parameters calculated with neighbor sizes up to Nb(i) = 18 and saw steady improvement in accuracy.

81.
W.
Lechner
and
C.
Dellago
,
J. Chem. Phys.
129
,
114707
(
2008
).
82.
V.
Ramasubramani
,
B. D.
Dice
,
E. S.
Harper
,
M. P.
Spellings
,
J. A.
Anderson
, and
S. C.
Glotzer
,
Comput. Phys. Commun.
254
,
107275
(
2020
).
83.

AntiNiAs is NiAs with the metal and halide positions reversed. The other crystal structures studied here are symmetric to this operation.

84.
H. J. C.
Berendsen
,
J. P. M.
Postma
,
W. F.
van Gunsteren
,
A.
DiNola
, and
J. R.
Haak
,
J. Chem. Phys.
81
,
3684
(
1984
).
85.
M. R.
Shirts
,
J. Chem. Theory Comput.
9
,
909
(
2013
).
86.
M.
Abadi
,
P.
Barham
,
J.
Chen
,
Z.
Chen
,
A.
Davis
,
J.
Dean
,
M.
Devin
,
S.
Ghemawat
,
G.
Irving
,
M.
Isard
 et al, in
12th USENIX Symposium on Operating Systems Design and Implementation (OSDI 16)
(
USENIX Association
,
2016
), pp.
265
283
.
87.
F.
Chollet
 et al, Keras, https://keras.io,
2015
.
88.
J.
Snoek
,
H.
Larochelle
, and
R. P.
Adams
, in
Advances in Neural Information Processing Systems 25: 26th Annual Conference on Neural Information Processing Systems 2012
,
2012
.
89.
V.
Nair
and
G. E.
Hinton
, in
ICML
,
2010
.
90.
S.
Sharma
,
S.
Sharma
, and
A.
Athaiya
, “
Activation functions in neural networks: Sigmoid, tanh, Softmax, ReLU, Leaky ReLU EXPLAINED!!!
,” https://towardsdatascience.com/activation-functions-neural-networks-1cbd9f8d91d6 (
2017
)
91.
D. P.
Kingma
and
J.
Ba
, arXiv:1412.6980 (
2014
).
92.
J. L.
Aragones
,
E.
Sanz
,
C.
Valeriani
, and
C.
Vega
,
J. Chem. Phys.
137
,
104507
(
2012
).
93.
M.-M.
Walz
and
D.
van der Spoel
,
Chem. Commun.
55
,
12044
(
2019
).
94.
R. S.
DeFever
,
H.
Wang
,
Y.
Zhang
, and
E. J.
Maginn
,
J. Chem. Phys.
153
,
011101
(
2020
).
95.
A.
Stukowski
,
Modell. Simul. Mater. Sci. Eng.
18
,
015012
(
2009
).
96.
S.
Martiniani
,
P. M.
Chaikin
, and
D.
Levine
,
Phys. Rev. X
9
,
011031
(
2019
).
97.
G.
Lanaro
and
G. N.
Patey
,
J. Chem. Phys.
146
,
154501
(
2017
).
98.
S.
Karthika
,
T. K.
Radhakrishnan
, and
P.
Kalaichelvi
,
Cryst. Growth Des.
16
,
6663
(
2016
).
99.
L.
Pauling
,
J. Am. Chem. Soc.
51
,
1010
(
1929
).
100.
D. J.
Wales
and
R. S.
Berry
,
Phys. Rev. Lett.
73
,
2875
(
1994
).
101.
R.
Stephen Berry
,
Strength from Weakness: Structural Consequences of Weak Interactions in Molecules, Supermolecules, and Crystals
(
Springer
,
2002
), pp.
143
168
.
102.
R. J.
Tibshirani
and
B.
Efron
,
Monogr. Stat. Appl. Probab.
57
,
1
(
1993
).
103.
CRC Handbook of Chemistry and Physics
, 102nd ed., edited by
J.
Rumble
(
CRC Press
,
London, England
,
2021
).
104.
J. W.
Menary
,
A. R. J. P.
Ubbelohde
, and
I.
Woodward
,
Proc. R. Soc. London, Ser. A
208
,
158
(
1951
).
105.
B. S.
Sæthre
,
A. C.
Hoffmann
, and
D.
van der Spoel
,
J. Chem. Theory Comput.
10
,
5606
(
2014
).
106.
J.
Shi
,
M.
Fulford
,
H.
Li
,
M.
Marzook
,
M.
Reisjalali
,
M.
Salvalaglio
, and
C.
Molteni
,
Phys. Chem. Chem. Phys.
24
,
12476
(
2022
).

Supplementary Material

You do not currently have access to this content.