It’s Okay to say “I don’t know!” There is no shame in that! The only shame is to pretend that we know everything.
-- Richard Feynman
- curiosity-driven and problem-solving oriented (it's really fun to figure out how things work!)
- consider failure as a way of making progress (it's more fun when it does not work inititally, but figure out why and make it work again)
- use whatever suits the best to solve problems (always happy to learn new things)
Expertise
- Computational molecular mechanics on the nanoscale:
- polymer hydration (PEO/PPO); block copolymer (PEO-PPO diblock and PEO-PPO-PEO triblock) assembly; polymer behavior in aqueous solutions and under nanoconfinement;
- interface phenomena between a polymer layer and hard (hydrophobic) materials;
- polymer brush formed by grafted on hard surface (PEO-grafted gold nanopores) and to (another) polymer (PVA-g-PEO bottlebrush polymer);
- materials design based on architectural parameters, such as side chain length, grafting density and backbone length/topology in the bottlebrush polymer;
- cosolvent-induced nanopore gated control; temperature-responsiveness controlled nanopore gated control;
- nanofluidics based on the polymer-grafted nanopores;
- coarse grained MD simulations of bottlebrush polymer assembly in solutions and melts;
- nonequilibrium molecular dynamics simulations/free energy calculation/enhanced sampling (e.g., umbrella sampling);
- Computational solid mechanics at the continuum level:
- finite element method (FEM) for general problems and meshfree methods (RKPM/SPH) for extremely large deformation problems;
- multiphysics modeling of finite-strain elasticity/plasticity, elastoplasticity/viscoplasticity, and viscoelasticity;
- multiscale modeling of deformation behaviors of elastomers/thermoplastics with deformation mechanism informed from the nanoscale modeling;
- Artificial intelligence/Machine learning:
- cheminformatics of drug-like molecules, polymers, and computational constitutive modeling;
- supervised machine learning: deep neural networks (DNN), convolutional neural networks (CNN), and recurrent neural networks (RNN);
- unsupervised machine learning: clustering (k-means/hierarchical clustering), dimensionality reduction (PCA, t-SNE), feature learning (VAE);
- semi-supervised learning: Reinforcement learning; Transfer learning;
- deep generative model: GAN/VAE/RNN;
- Computer programing, design and modeling:
- massive parallel computation using OpenMP, MPI, and OpenACC/CUDA;
- large scale simulations using super computer (Expanse/Bridges2/Stampede2/Frontera);
- computer cluster maintenance (up to 16 nodes) and user management;
- programming language: C++, Python, Fortran, Matlab, bash/batch, Makefile, tcl/tk;
- deep learning framework: PyTorch, Tensorflow, Jax;
- cheminformatics: RDKit, PubChemPy, Openbabel;
- version control: git/GitHub/Gitlab;
- molecular design: Avogadro/Maestro/Pymol/Atomsk;
- molecular modeling: Gromacs/LAMMPS/NAMD;
- molecular visualization: VMD/Ovito/Blender;