Transcript Quantum, Classical & Multi

Computer-Aided Molecular Modeling of Materials
Instructor: Yun Hee Jang ([email protected], MSE 302, 2323)
TA: Eunhwan Jung ([email protected], MSE 301, 2364)
Web: http://mse.gist.ac.kr/~modeling/lecture.html
Reference:
- D. Frenkel & B. Smit, Understanding molecular simulations, 2nd ed. (2002)
- M. P. Allen & D. J. Tildesley, Computer simulation of liquids (1986)
- A. R. Leach, Molecular modeling: principles and applications, 2nd ed. (2001)
- and more
Grading:
- Homework: reading + 0.5-page summary
- Exam or Term report: Mid-term & Final
- Hands-on computer labs (report & presentation)
- Presence & Participation (questions, answers, comments, etc.)
Difficulty (cost, time, manpower, inaccuracy)
Why do we need a molecular modeling (i.e. computer
simulation at a molecular level) in materials science?
Emerging (future)
Materials science
N~102, L~10 nm
Simulation will lead.
Traditional (Past)
Materials science
N~1023, L~10 cm
Experiment didn’t need simulation.
Molecular simulation
in virtual space
too hard
hard
easy
easy
Experiment
in real space
N (number of atoms) or L (size) of a system of interest)
Nobel Prize History of Molecular Modeling
• 1918 – Physics – Max Planck – Quantum theory of blackbody radiation
• 1922 – Physics – Niels Bohr – Quantum theory of hydrogen spectra
• 1929 – Physics – Louis de Broglie – Matter waves
• 1932 – Physics – Werner Heisenberg – Uncertainty principle
• 1933 – Physics – Erwin Schrodinger & Paul Dirac – Wave equation
• 1945 – Physics – Wolfgang Pauli – Exclusion principle
• 1954 – Physics – Max Born – Interpretation of wave function
• 1998 – Chemisty – Walter Kohn & John Pople
Quantum Mechanics
• 1921 – Physics – Albert Einstein– Quantum theory of photoelectric effect
Quantum Chemistry
• 2013 – Chemisty – Martin Karplus, Michael Levitt, Arieh Warshel
Classical
Molecular
Simulation
Review of Nobel Information 2013 Chemistry
- Simulation or Modeling of molecules (in materials) on computers
- Classical (Newtonian) physics vs. Quantum (Schrodinger) physics
-
Quantum description of atoms and molecules: electrons & nuclei
Strength: applicable to describe electronic (photo)excitation
Strength: interatomic interactions described “naturally”
Strength: chemical reactions (bond formation/breaking)
Weakness: slow, expensive, small-scale (N < 102), @ 0 K
Classical description of atoms and molecules: balls & springs
Strength: fast, applicable to large-scale (large N) systems
Strength: close to our conventional picture of molecules
Strength: easy to code, free codes available, finite T
Weakness: interatomic interactions from us (force field)
Weakness: no chemical reactions, no electronic excitation
Application: structural, mechanical, dynamic properties
Difficulty (cost, time, manpower, inaccuracy)
Quantum vs. Classical description of materials
With reasonable amount of resources,
larger-scale (larger-N) systems can be described
with classical simulations than with quantum simulations.
Quantum simulation
in virtual space
hard
Classical simulation
in virtual space
easy
Experiment
in real space
N (number of atoms) or L (size) of a system of interest)
Example of multi-scale
molecular modeling:
CO2 capture project
MC
Monte Carlo Process Simulation
- Grand Canonical (GCMC) or Kinetic (KMC)
- Flue gas diffusion & Selective CO2 capture
MD
Large-scale Molecular Dynamics
- Validation: Interatomic potential (Force Field)
- Viscosity, diffusivity distribution: bulk solvent
QM
First-principles Quantum Mechanics
- Validation: DFT + continuum solvation
- Reaction: solvent molecule + CO2 complex
Step 1:
Quantum:
Reaction
PzH3+-CO2
N
H
10.6 kcal/mol
(MEA)
C
Piperazine
O
PzH2+-CO2-
+CO2
PzH+-2CO2-
7.8
+PzH2
+CO2
PzH-CO2H
+PzH2
+CO2
PzH3+
Pz(CO2)22-
PzH3+
PzHCO2solvent (PzH2)
PzH2 (regener)
PzH2+CO2-
PzH3+
HCO37
Quantum simulation Example No. 2:
Pd 촉매 반응, UV/vis spectrum 재현, 유기태양전지 효율 저하 설명
-1.96
EX2
3.26
PCE
3.1%
-3.26
EX1
1.96
-5.22
-2.12
PCE
0.4%
EX1
2.99
gone!
Relative free energy (kcal/mol)
-5.11
10
0
-10
8.9
H
0
N
N
N
N
N
Pd
+ Pd
-20
H
-18.4
-30
-32.4
-40
-34.7
-26.5
-29.1
N
N
-51.3
N
+ Pd
-50
Pd+
22BI
TS1
I1
TS2
I2
TS3
I3
H
N
Pd
N
N Pd
H
Pd+
+2BI
N Pd
N
N
H
+
Pd
N
What quantum/classical molecular modeling can bring to you: Examples.
Reduction-oxidation potential, acibity/basicity (pKa), UV-vis spectrum, density profile, etc.
J. Phys. Chem. B (2006)
Oscillator strength (f)
J. Phys. Chem. A (2009, 2001), J. Phys. Chem. B (2003),
Chem. Res. Toxicol. (2003, 2002, 2000), Chem. Lett. (2007)
2.5
2.0
1.5
1.0
0.5
0.0
200
400
600
800
cm-1
1000
J. Phys. Chem. B (2011), J. Am. Chem. Soc. (2005, 2005, 2005)
Step 2:
Classical:
2-species (AMP and PZ) distribution in water
Which one (among AMP and PZ) is less soluble in water?
Which one is preferentially positioned at the gas-liquid interface?
Which one will meet gaseous CO2 first?
Hopefully PZ to capture CO2 faster, but is it really like that?
Let’s see with the MD simulation on a model of their mixture solution!
First-principles multi-scale molecular modeling
제일원리 다단계 분자모델링
► 물질구조 분자수준 이해 ► 선험적 특성 예측
► 신물질 설계 ► 물질특성 향상
2. 고전역학 분자동력학 모사 (컴퓨터 구축 102~107 개 원자계의 뉴턴방정식 풀기)
- 전자 무시, ball (원자) & spring (결합) 모델로 분자/물질 표현 (힘장)
- Cheap ► 대규모 시스템에 적용, 시간/온도에 따른 구조/형상 변화 모사
1. 양자역학 전자구조 계산 (컴퓨터 구축 101~103 개 원자계 슈레딩거방정식 풀기)
- 정확, 경험적 패러미터 불필요, 제일원리계산, but expensive ► 소규모 시스템
MD
atomistic
molecular
FF
snapshot
CG-FF
MULTI
SCALE
MODELING
CGMD
coarsegrained
nanoscale
morphology
KMC
chargetransport
QM
electronic
structure
transport
parameter
understanding
new design
prediction
test
validation
EXPERIMENT
synthesis
fabrication
characterization
Lecture series I-IV: Molecular Modeling of Materials
I. 2013 Spring: Elements of Quantum Mechanics (QM)
- Birth of quantum mechanics, its postulates & simple examples
 Particle in a box (translation)
 Harmonic oscillator (vibration)
 Particle on a ring or a sphere (rotation)
II. 2013 Fall: Quantum Chemistry
- Quantum-mechanical description of chemical systems
 One-electron & many-electron atoms
 Di-atomic & poly-atomic molecules
III. 2014 Spring: Classical Molecular Simulations of materials
- Large-scale simulation of chemical systems (or any collection of particles)
 Monte Carlo (MC) & Molecular Dynamics (MD)
IV. 2014 Fall: Molecular Modeling of Materials (Project-oriented class)
- Application of a combination of the above methods to understand
structures, electronic structures, properties, and functions of various materials
A typical experiment in a real (not virtual) space
1. Some material is put in a container at fixed T & P.
2. The material is in a thermal fluctuation, producing lots
of different configurations (a set of microscopic states)
for a given amount of time. It is the Mother Nature
who generates all the microstates.
3. An apparatus is plugged to measure an observable
(a macroscopic quantity) as an average over all
the microstates produced from thermal fluctuation.
P
T
How do we mimic the Mother Nature in a virtual space
to realize lots of microstates, all of which correspond to
a given macroscopic state?
How do we mimic the apparatus in a virtual space
to obtain a macroscopic quantity (or property or observable)
as an average over all the microstates?
How do we mimic the Mother Nature in a virtual space
to realize lots of microstates, all of which correspond to
a given macroscopic state? By MC & MD methods!
microscopic states (microstates)
or microscopic configurations
under external constraints
(N or , V or P, T or E, etc.)
 Ensemble (micro-canonical,
canonical, grand canonical, etc.)
Average over
a collection of
microstates
In a real-space experiment
they’re generated naturally
from thermal fluctuation
In a virtual-space simulation
it is us who needs to generate
them by QM/MC/MD methods.
Macroscopic quantities (properties, observables)
• thermodynamic –  or N, E or T, P or V, Cv, Cp, H, S, G, etc.
• structural – pair correlation function g(r), etc.
• dynamical – diffusion, etc.
These are what are
measured in true
experiments.