Domain 3 of 10
From Frequencies to Rate Constants
From frequencies to free energies, transition states to rate constants, and composite recipes toward chemical accuracy.
6 Checked11 Partial1 UnverifiedEngine 1.10.12 live
What You Can Compute
Free Energies
RRHO thermochemistry with quasi-RRHO, hindered rotors and standard state shifts.
Transition States
P-RFO with Bofill updates, IRC and climbing-image NEB, with constraints and relaxed scans.
Rate Constants
TST with tunnelling, variational TST with ZCT and SCT, RRKM, Troe fall-off and microkinetic models with degree of rate control.
Composite Accuracy
Complete basis set extrapolation, runnable G3, G4 step plans, conformer ensembles and bond dissociation energies.
Reaction Discovery
Scans along bond, angle and dihedral coordinates, and artificial force induced reaction searches with reaction networks and protomer enumeration.
Every Method, with Its Level of Support
Generated from the engine's own capability report (engine 1.10.12), so this list never runs ahead of the code.
Checked reproduces printed values Checked (exact) no printed value exists; checked against exact mathematics or brute force Partial code exists, unchecked parts badged in the app Unverified built, no printed check yet
| Method | Support | Since | Primary Sources |
|---|---|---|---|
| Global optimisation, crystal structure prediction, kinetic Monte CarloBasin and minima hopping, genetic search, crystal structure prediction in eight common space groups, and lattice kinetic Monte Carlo on square and triangular lattices. Crystal candidates are exported as CIF for ranking with another tool; ranking inside the search uses force fields only. Lattice kinetic Monte Carlo has no lateral interactions. | Unverified | 1.9.0 | |
| Automated reaction path discoveryArtificial force induced reaction searches on the quantum chemistry energy and gradient with climbing-image NEB refinement, reaction network exploration and protomer enumeration. Artificial force follows Maeda and Morokuma 2010 Eq. 1 and 2, with the collision energy gamma as input. GSM string growth is not built. No discovered network reproduced against a printed one yet. | Partial | 1.9.0 | Jensen 2017; Dugundji, Ugi 1973; Maeda, Morokuma 2010 |
| Geometry optimisation with constraintsBFGS with trust radius, bond, angle and dihedral constraints. | Checked | 0.1 | Baker 1986 |
| RRHO thermochemistryZPE, H, S, G and Cp from harmonic frequencies. | Checked | 0.7.40 | McQuarrie 1976; Fernandez-Ramos et al. 2007 |
| Quasi-RRHO and hindered rotorsLow-mode entropy corrections, hindered rotors and standard state shifts. | Partial | 0.8.0 | Grimme 2012; Ribeiro et al. 2011 |
| Transition-state search (P-RFO)Partitioned rational function optimisation with Bofill updates. | Checked | 0.7.40 | Baker 1986; Bofill 1994; Baker, Chan 1996 |
| IRCGonzalez-Schlegel reaction path in mass-weighted coordinates. | Checked | 0.7.40 | Gonzalez, Schlegel 1989 |
| Climbing-image NEBMinimum energy paths with the improved tangent. | Partial | 0.7.45 | Henkelman, Uberuaga, Jonsson 2000 |
| Rigid and relaxed scansRigid scans along a bond, and relaxed scans along a bond, angle or dihedral with a constrained optimisation at every point (1 to 200 points), written as an energy profile and a trajectory. Rigid scans move one bond only; every angle and dihedral scan is relaxed. MMFF94 torsion barriers reproduce Halgren and Nachbar 1996 Table V (30 barriers, largest difference 0.0054 kcal/mol). | Checked | 0.1 | Halgren, Nachbar 1996 |
| TST rate constants with tunnellingConventional TST with Wigner or Skodje-Truhlar tunnelling. | Checked | 0.7.40 | Eyring 1935; Skodje, Truhlar 1981 |
| Variational TST with ZCT and SCTCanonical variational TST along the IRC with zero- and small-curvature tunnelling. Checked by exact identities and limits of the source papers only (projection of the path direction, a variational rate never above the conventional one, the zero-temperature limit, small-curvature equal to zero-curvature tunnelling on a straight path, the parabolic-barrier closed form of Skodje, Truhlar, Garrett 1981). No printed variational rate or tunnelling factor is reproduced yet. Large-curvature tunnelling refused. | Partial | 0.9.5 | Truhlar, Garrett 1984; Liu et al. 1993 |
| RRKM, Troe fall-off and master equationk(E), thermal fall-off curves and a master equation kernel. | Partial | 0.7.55 | Robinson, Holbrook 1972; Troe 1983; Miller, Klippenstein 2006 |
| Complete basis set extrapolationSCF and correlation energy extrapolation from cardinal series. | Partial | 0.7.60 | Helgaker et al. 1997; Halkier et al. 1998 |
| G3, G3X and G4 composite stepsStep plans, atomic energies and supplied-energy assembly for G3, G3X and G4. A full molecular G4 run is not claimed. | Partial | 1.2.0 | Curtiss et al. 2007; Curtiss et al. 1998; Curtiss et al. 2001 |
| G3 full runRunnable G3 steps 1 to 7 for H, C, N, O and F, assembled into the G3 energy (Unchecked). G4 E0 still refused. | Partial | 1.3.0 | Curtiss et al. 1998 |
| Conformer ensembles, BDE and atomisation energiesBoltzmann ensemble free energies, bond dissociation energies, total atomisation energies and two-point CBS extrapolation. W1, W2, CBS-QB3 and full G4 refused: papers not held. | Partial | 1.5.0 | Pracht, Bohle, Grimme 2020; Karton, Daon, Martin 2011 |
| Microkinetic modelling and energetic spanStiff microkinetic models, degree of rate control, reaction networks, Curtin-Hammett, Arrhenius and Eyring fits, TOF and energetic span. | Partial | 1.4.0 | Kozuch, Shaik 2006; Kozuch, Shaik 2011; Campbell 2001 |
| Dimer method saddle searchTransition states without a Hessian by the dimer method. | Partial | 1.4.0 | Henkelman, Jonsson 1999 |
Printed Value, Our Value
| Quantity | Source | Printed | Ours | Result |
|---|---|---|---|---|
| Skodje-Truhlar Tunnelling Table I | Skodje, Truhlar 1981 | 28 rows | 25 rows (3 inconsistent in print) | Pattern |
| Bofill TS Searches, HF/STO-3G | Baker, Chan 1996 | 10 reactions | 10 of 10 converge | Match |
| HF/6-31G(d,p) Optimised Energies of HF, HCN, H2O, HCCH, Eh (Geometry Optimisation) | Simon, Duran, Dannenberg 1996 Tables I, IV, VII | -100.011691, -92.877138, -76.023615, -76.821837 | all within 5e-7 Eh | Match |
| CH4 Entropy and Heat Capacity at 298.15, 500 and 1000 K, J/(K mol) (RRHO Thermochemistry) | NIST-JANAF tables, Chase 1998 | printed tables | largest difference 0.0009 J/(K mol) | Match |
| Reaction Symmetry Numbers for TST, H + CH4 and CF3 + CH4 | Fernandez-Ramos et al. 2007 Eqs. (6), (7) | 4, 12 | 4, 12 | Match |
| IRC Points on a Model Surface at Step 0.05, Largest Distance from the Exact Path | Gonzalez, Schlegel 1989 | second-order method | 1.7e-4, ends connect to both minima | Pattern |
| MMFF94 Torsion Barriers, 30 Molecules, Kcal/mol (E.g. Ethane, Butane Syn) | Halgren, Nachbar 1996 Table V, doi 10.1002/(SICI)1096-987X(199604)17:5/6<587::AID-JCC4>3.0.CO;2-Q | 3.22, 5.21 | 3.215, 5.210 (largest difference 0.0054) | Match |
Not Offered Today
- Large-curvature tunnelling
- A full molecular G4 run (step plans and atomic energies only)
- Multidimensional master equations beyond the single-well kernel
- Growing string method
In the Chemistry Lab
Start, then Reaction and rates. Give reactant, product or a guess structure; the guided flow proposes TS search, IRC check and rates, each step approved by you.
Thermochemistry and Kinetics Questions
How do I know a transition state is real?
Frequencies show exactly one imaginary mode, and the IRC connects it to the reactant and product you expect. The Chemistry Lab suggests both checks after a TS search.
Are the rate constants reliable?
They are as reliable as the barrier. Hartree-Fock barriers are qualitative; use a correlated or DFT barrier, and read the tunnelling model beside each number.
Can the program find reactions I did not specify?
AFIR searches, reaction networks and protomer enumeration are built and marked Partial: the artificial force follows Maeda and Morokuma 2010, and no discovered network has been reproduced against a printed one yet. Relaxed scans along a bond, angle or dihedral are available for guided exploration, and rigid scans for a single bond.
See It on a Molecule of Your Own
Pair your own server once, then let your team queue computations from the browser.