Official Resources
- Homepage: https://github.com/FreeON/freeon
- Documentation: https://github.com/FreeON/freeon/wiki
- Source Repository: https://github.com/FreeON/freeon
- License: GNU General Public License v3.0
Overview
FreeON is an experimental, open-source suite of programs for linear-scaling quantum chemistry calculations. Formerly known as MondoSCF, it performs Hartree-Fock, pure DFT, and hybrid DFT calculations using a Cartesian-Gaussian LCAO basis. FreeON emphasizes O(N) and O(N log N) algorithms for non-metallic systems.
Scientific domain: Molecules, clusters, large organic systems, biomolecules
Target user community: Researchers needing linear-scaling molecular quantum chemistry for large non-metallic systems
Theoretical Methods
- Hartree-Fock (HF)
- Density Functional Theory (DFT)
- Hybrid HF/DFT (B3LYP and others)
- Cartesian-Gaussian LCAO basis
- Linear-scaling O(N) algorithms
- O(N log N) Coulomb evaluation
- DIIS convergence acceleration
- Norm-conserving pseudopotentials (optional)
Capabilities (CRITICAL)
- Ground-state electronic structure
- Hartree-Fock calculations
- Pure DFT (LDA, GGA)
- Hybrid DFT (B3LYP, etc.)
- Linear-scaling O(N) Fock matrix construction
- Geometry optimization
- Molecular dynamics
- Forces and gradients
- Frequency calculations (Hessian)
- Large molecular systems
Sources: Wikipedia, GitHub repository, academic publications
Key Strengths
Linear-Scaling Focus:
- O(N) Fock matrix construction
- O(N log N) Coulomb operator
- Targeted at non-metallic systems
- Scales to large molecules
- Efficient for insulators
Open Source GPL v3:
- Fully open source
- Community contributions welcome
- Transparent implementation
- Academic and commercial use
Hybrid Functionals:
- B3LYP support
- PBE0 support
- Exact exchange with O(N)
- Accurate thermochemistry
Robust Implementation:
- DIIS convergence
- Multiple SCF algorithms
- Fortran95 and C code
- Well-structured codebase
Inputs & Outputs
-
Input formats:
- Native input file format
- XYZ coordinates
- Basis set specifications
- Calculation parameters
-
Output data types:
- Total energies
- Orbital energies
- Forces and gradients
- Optimized geometries
- Molecular dynamics trajectories
- Frequency data
Interfaces & Ecosystem
-
Standalone operation:
- Self-contained package
- Built-in basis sets
- Integrated geometry optimizer
-
Build system:
- Autoconf/Automake
- GNU toolchain
- Linux/Unix platforms
- FreeBSD support
Advanced Features
Linear-Scaling Methods:
- Fast Multipole Method (FMM)
- Hierarchical matrix algebra
- Sparse matrix techniques
- Locality exploitation
Geometry Optimization:
- Quasi-Newton methods
- Conjugate gradient
- Constraints support
- Transition state search
Molecular Dynamics:
- Born-Oppenheimer MD
- Velocity Verlet integrator
- Temperature control
- Trajectory analysis
Frequency Calculations:
- Hessian evaluation
- Vibrational analysis
- IR intensities
- Thermodynamic properties
Performance Characteristics
- Speed: O(N) for large non-metallic systems
- Accuracy: Standard Gaussian basis precision
- System size: Hundreds to thousands of atoms
- Memory: Efficient sparse matrix storage
- Parallelization: Some parallel capabilities
Computational Cost
- Linear scaling: Achieved for insulators/semiconductors
- Hybrid DFT: Efficient O(N) exact exchange
- Break-even: ~100-500 atoms vs cubic codes
- Typical: Competitive for large molecules
Limitations & Known Constraints
- Metallic systems: Linear scaling breaks down
- Development status: Experimental, less active recently
- Community: Smaller than major codes
- Documentation: Academic-level
- Periodic systems: Not primary focus
- Platform: Linux/Unix only
- Basis sets: Limited built-in selection
Comparison with Other Codes
- vs Gaussian: FreeON O(N) focus vs Gaussian robustness
- vs NWChem: FreeON specialized linear-scaling
- vs ONETEP: FreeON molecular, ONETEP materials
- vs BigDFT: FreeON Gaussian, BigDFT wavelets
- Unique strength: Open-source O(N) molecular quantum chemistry
Application Areas
Large Organic Molecules:
- Natural products
- Pharmaceuticals
- Polymers
- Dendrimers
Biomolecular Systems:
- Protein fragments
- Nucleic acid segments
- Enzyme models
- Drug-receptor complexes
Molecular Clusters:
- Water clusters
- Hydrogen-bonded systems
- Molecular crystals (fragments)
- Host-guest complexes
Materials Fragments:
- Molecular solids
- Organic semiconductors
- Self-assembled monolayers
- Nanoparticle ligands
Best Practices
System Selection:
- Best for non-metallic systems
- Check band gap for linear scaling
- Use for insulators/semiconductors
- Avoid highly metallic systems
Basis Set Choice:
- Start with standard basis (6-31G*)
- Test convergence with larger basis
- Document basis for reproducibility
SCF Convergence:
- Use DIIS for difficult cases
- Monitor energy convergence
- Adjust mixing if needed
Linear-Scaling Settings:
- Tune cutoff parameters
- Balance accuracy and speed
- Verify against standard calculation
Community and Support
- Open-source GPL v3
- GitHub repository
- Academic publications
- Legacy from LANL development
- Contributor community
Verification & Sources
Primary sources:
- GitHub: https://github.com/FreeON/freeon
- Wikipedia: FreeON article
- M. Challacombe et al., J. Chem. Phys. publications
Secondary sources:
- Linear-scaling quantum chemistry literature
- Open-source chemistry software surveys
Confidence: VERIFIED - Open source, published methodology
Verification status: ✅ VERIFIED
- Source code: OPEN (GitHub, GPL v3)
- Academic use: Published applications
- Documentation: Wiki and papers
- Development status: Experimental/Mature
- Specialty: O(N) linear-scaling molecular quantum chemistry