Official Resources
- Homepage: https://github.com/lanl/NEXMD
- Documentation: https://nexmd.github.io/
- Source Repository: https://github.com/lanl/NEXMD
- License: BSD 3-Clause License
Overview
NEXMD is a software package developed at Los Alamos National Laboratory for simulating photoinduced adiabatic and non-adiabatic excited-state molecular dynamics. It uses semiempirical quantum chemistry methods (CEO package) with Tully's fewest-switches surface hopping algorithm, making it efficient for studying large conjugated systems like chromophores and polymers. Written in Fortran 90 with Python scripts for parallel execution.
Scientific domain: Photochemistry in large molecules, conjugated systems, chromophore dynamics, exciton dynamics
Target user community: Researchers studying organic chromophores, conjugated polymers, and large molecular photophysics
Theoretical Methods
- Semiempirical methods (AM1, PM3, PM6, PM7)
- Collective Electronic Oscillator (CEO) approach
- Configuration Interaction Singles (CIS)
- Tully's Fewest-Switches Surface Hopping (FSSH)
- Non-adiabatic coupling calculations
- Trivial crossing detection
- Decoherence corrections (EDC, AFSSH)
- Velocity rescaling schemes
Capabilities (CRITICAL)
- Ground and excited-state dynamics
- Non-adiabatic transitions between states
- Large molecular systems (100s of atoms)
- Exciton dynamics in conjugated systems
- Hot carrier relaxation
- Energy transfer simulations
- Absorption spectra simulation
- Time-resolved properties
- Parallel trajectory execution
- Ensemble averaging
Sources: Official GitHub documentation, LANL publications
Key Strengths
Semiempirical Efficiency:
- Fast electronic structure
- Large systems (>100 atoms)
- Many excited states feasible
- Long timescale dynamics
CEO Methodology:
- Collective modes description
- Efficient gradient computation
- Multi-state treatment
- Excited-state forces
LANL Development:
- National lab support
- Active maintenance
- Regular updates
- Scientific validation
Conjugated Systems:
- Polymer dynamics
- Organic chromophores
- Exciton migration
- Energy transfer
Inputs & Outputs
-
Input formats:
- NEXMD input files
- Geometry files (XYZ)
- Parameter files
- Python driver scripts
-
Output data types:
- Trajectory files
- Population dynamics
- Energy files
- Coupling data
- Statistical output
Interfaces & Ecosystem
- Electronic structure: Internal CEO (semiempirical)
- Scripting: Python for job management
- Parallelization: Multiple trajectory parallelism
- Analysis: Built-in analysis tools
Advanced Features
Trivial Crossing Detection:
- Automatic state relabeling
- Diabatic following
- Coupling analysis
- State tracking
Decoherence Methods:
- Energy-based decoherence (EDC)
- Augmented FSSH (AFSSH)
- Wavefunction collapse schemes
- Physical decoherence treatment
Large-Scale Dynamics:
- Efficient for 100+ atoms
- Many trajectories feasible
- Long timescales (ps)
- Statistical averaging
Performance Characteristics
- Speed: Fast (semiempirical)
- Accuracy: Good for trends
- System size: 100s of atoms
- Parallelization: Trajectory-level
Computational Cost
- Per step: Milliseconds (semiempirical)
- Full trajectory: Minutes to hours
- Ensemble: Highly parallel
- Typical: 100-500 trajectories
Limitations & Known Constraints
- Accuracy: Semiempirical limitations
- Parametrization: Requires validated parameters
- Heavy atoms: Limited treatment
- Spin-orbit: Not included
- Quantum nuclei: Classical only
Comparison with Other Codes
- vs SHARC/Newton-X: NEXMD faster but less accurate
- vs DFTBaby: Similar semiempirical approach
- vs Ab initio codes: NEXMD faster, less accurate
- Unique strength: Large conjugated systems, efficiency, LANL support
Application Areas
Conjugated Polymers:
- Polythiophenes
- PPV derivatives
- Donor-acceptor polymers
- Exciton dynamics
Organic Chromophores:
- Dye molecules
- Photosensitizers
- OLED materials
- Photovoltaic materials
Energy Transfer:
- FRET dynamics
- Exciton migration
- Hot carrier relaxation
- Charge separation
Biological Chromophores:
- Chlorophylls
- Carotenoids
- Flavins
- Photoactive proteins
Best Practices
Parameter Validation:
- Benchmark against ab initio
- Check excited-state ordering
- Validate geometries
- Compare spectroscopy
Trajectory Management:
- Sufficient ensemble size
- Convergence checking
- Statistical analysis
- Error estimation
System Setup:
- Proper initial conditions
- Adequate equilibration
- Careful state selection
- Documentation
Community and Support
- Open-source BSD license
- LANL development team
- GitHub repository
- Documentation available
- Active maintenance
Verification & Sources
Primary sources:
- GitHub repository: https://github.com/lanl/NEXMD
- Documentation: https://nexmd.github.io/
- S. Tretiak et al., publications on CEO methodology
- A. F. Fidler et al., J. Phys. Chem. Lett. 2013
Secondary sources:
- LANL software registry
- Published applications
- Conference presentations
Confidence: VERIFIED - LANL open-source
Verification status: ✅ VERIFIED
- Official homepage: ACCESSIBLE (GitHub)
- Documentation: ACCESSIBLE
- Source code: OPEN (BSD 3-Clause)
- Community support: LANL maintained
- Active development: Yes
- Specialized strength**: Large conjugated systems, semiempirical efficiency, exciton dynamics