Coarse-Grained Force Fields Built
on Atomistic Force Fields
Huai Sun, Liang Wu, Zhao Jin, Fenglei Cao, Gong Zheng, and Hao Huang
1 Introduction
For predictions of phase separation and self-assembly, the atomistic simulations
using all-atom or united-atom force fields (AAFFs or UAFFs) [1–11] are hindered
by the requirement of large spatiotemporal scales. Moreover, it is often unnecessary
to follow the atomistic details in order to predict the phase behaviors, which take
place at much larger spatiotemporal scales than the atomic vibrations. As significant
savings in computational resources can be achieved by coarse graining several atoms
into one interacting bead, coarse-grained force fields (CGFFs) have been developed
for polymers, [12] bio-macromolecules [11, 13], and surfactant related materials [14].
In addition, various applications including predictions of polymer phase behaviors
using specifically developed CGFFs [15–21] have been published. In this article, we
report our recent works on CGFFs [22–26].
We build CGFFs based on validated AAFFs. This approach is advantageous not
only because it bridges a gap in multiscale simulations as shown in Fig. 1, but also
due to the concept can be applied to build force fields at different levels consistently.
The approach is a scale-up as it builds an upper level force field using data generated
at lower level. Since the AAFF is made from quantum mechanics data (QMD), [5–7]
the CGFF derived from the AAFF is essentially built on first principle. The first
principle force fields are more accurate and transferrable than empirical force fields.
With a few experimental data as guidance, the force field can be optimized to quantitatively predict thermodynamic properties [7, 26–31]. Furthermore, the concept of
scale-up can be extended to high-level of coarse-graining. On the technical side, the
H. Sun (B) · L. Wu · Z. Jin · F. Cao · G. Zheng · H. Huang
School of Chemistry and Chemical Engineering, Materials Genome Initiative Center, and Key
Laboratory of Scientific and Engineering Computing of Ministry of Education, Shanghai Jiao
Tong University, Shanghai 200240, China
e-mail: huaisun@sjtu.edu.cn
© The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2021
E. J. Maginn and J. Errington (eds.), Foundations of Molecular Modeling
and Simulation, Molecular Modeling and Simulation,
https://doi.org/10.1007/978-981-33-6639-8_7
143
on Atomistic Force Fields
Huai Sun, Liang Wu, Zhao Jin, Fenglei Cao, Gong Zheng, and Hao Huang
1 Introduction
For predictions of phase separation and self-assembly, the atomistic simulations
using all-atom or united-atom force fields (AAFFs or UAFFs) [1–11] are hindered
by the requirement of large spatiotemporal scales. Moreover, it is often unnecessary
to follow the atomistic details in order to predict the phase behaviors, which take
place at much larger spatiotemporal scales than the atomic vibrations. As significant
savings in computational resources can be achieved by coarse graining several atoms
into one interacting bead, coarse-grained force fields (CGFFs) have been developed
for polymers, [12] bio-macromolecules [11, 13], and surfactant related materials [14].
In addition, various applications including predictions of polymer phase behaviors
using specifically developed CGFFs [15–21] have been published. In this article, we
report our recent works on CGFFs [22–26].
We build CGFFs based on validated AAFFs. This approach is advantageous not
only because it bridges a gap in multiscale simulations as shown in Fig. 1, but also
due to the concept can be applied to build force fields at different levels consistently.
The approach is a scale-up as it builds an upper level force field using data generated
at lower level. Since the AAFF is made from quantum mechanics data (QMD), [5–7]
the CGFF derived from the AAFF is essentially built on first principle. The first
principle force fields are more accurate and transferrable than empirical force fields.
With a few experimental data as guidance, the force field can be optimized to quantitatively predict thermodynamic properties [7, 26–31]. Furthermore, the concept of
scale-up can be extended to high-level of coarse-graining. On the technical side, the
H. Sun (B) · L. Wu · Z. Jin · F. Cao · G. Zheng · H. Huang
School of Chemistry and Chemical Engineering, Materials Genome Initiative Center, and Key
Laboratory of Scientific and Engineering Computing of Ministry of Education, Shanghai Jiao
Tong University, Shanghai 200240, China
e-mail: huaisun@sjtu.edu.cn
© The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2021
E. J. Maginn and J. Errington (eds.), Foundations of Molecular Modeling
and Simulation, Molecular Modeling and Simulation,
https://doi.org/10.1007/978-981-33-6639-8_7
143