We have investigated the thermochemical stability of the carbon skeleton in polycyclic aromatic (halo) hydrocarbons using a systematic collection of molecules (the PAHH343 set). With high-level quantum chemistry methods such as W1X-2, we have obtained chemically accurate (i. e.,±~5 kJmol^-1 ) “normalized carbon skeleton” bond energies. They are calculated by removing the C H and C X (X=F, Cl) bond energies from the total atomization energy, and then normalizing on a per-carbon basis. For species with isomeric halogen-substitution pattern, the energetic variation is generally small, though larger difference can also be seen due to structural distortion from steric repulsion. The skeleton energy becomes smaller with an increasing number of halogen atoms due to the withdrawal of electron density from the bonding orbitals, mainly through the σ-bonds. We have further assessed the performance of some low-cost quantum chemistry methods for the PAHH343 set. The deviations from reference values are largely systematic, and can thus be compensated for, yielding errors that are on average below 10 kJmol^-1. This provides the prospect for the study of an even wider range of PAHH and related systems.

Composite ab initio methods based on local coupled-cluster approaches calculate the CCSD(T)/CBS energy at a significantly reduced computational cost than the corresponding canonical methods. While showing promising performance for general thermochemistry, local composite ab initio methods have not been tested for fullerenes. Here we examine the performance of several local G4(MP2)-based methods for calculating the relative stability of a diverse set of C₄₀ fullerenes. We use canonical G4(MP2) isomerization energies as reference data. Fullerenes provide a challenging problem for DLPNO-G4(MP2)-based methods. The DLPNO-based methods result in overall root-mean-square deviations (RMSDs) of 28.6 (NormalPNO with CCSD(T₀)), 23.0 (NormalPNO with CCSD(T₁)), and 16.1 (TightPNO with CCSD(T₀)) kJ mol⁻¹. The local natural orbital LNO-G4(MP2) method provides the best overall performance with an RMSD of 4.9 kJ mol⁻¹. The DLPNO-G4(MP2) and LNO-G4(MP2) methods systematically overestimate the canonical G4(MP2) isomerization energies. Therefore, they provide valuable upper limits of the fullerene isomerization energies.

In the present study, we have reformulated the G4(MP2) and G4(MP2)-6X procedures for use with a restricted-open-shell (RO) formalism. We find that the resulting ROG4(MP2) and ROG4(MP2)-6X procedures generally perform comparably to the original unrestricted (U) variants, including their performance on radicals. Our analysis suggests that this is due mainly to the inclusion of empirical parameters that overcome the slightly less good performance of the U variants. However, a major practical advantage of ROG4(MP2) and ROG4(MP2)-6X is that they can be used in a wider range of computational chemistry software packages than the U analogs. We have demonstrated the importance of this aspect with a large-scale ROG4(MP2)-6X computation for the dissociation of the dodecahedryl dimer (C₂₀H₁₉)₂.

In the present study, we have investigated the performance of RIJCOSX DLPNOCCSD(T)-F12 methods for a wide range of systems. Calculations with a high-accuracy option ["DefGrid3 RIJCOSX DLPNO-CCSD(T₁)-F12"] extrapolated to the complete-basis-set limit using the maug-cc-pV[D+d,T+d]Z basis sets provides fairly good agreements with the canonical CCSD(T)/CBS reference for a diverse set of thermochemical and kinetic properties [with mean absolute deviations (MADs) of ~1– 2 kJ mol^-1 except for atomization energies]. On the other hand, the low-cost "RIJCOSX DLPNO-CCSD(T)-F12D" option leads to substantial deviations for certain properties, notably atomization energies (MADs of up to tens of kJ mol^-1). With the high-accuracy CBS approach, we have formulated the L-W1X method, which further includes a low-cost core–valence plus scalar-relativistic term. It shows generally good accuracy. For improved accuracies in specific cases, we advise replacing maug-cc-pV (n+d)Z with jun-cc-pV(n+d)Z for the calculation of electron affinities, and using well-constructed isodesmic-type reactions to obtain atomization energies. For medium-sized systems, DefGrid3 RIJCOSX DLPNO-CCSD(T₁)-F12 calculations are several times faster than the corresponding canonical computation" the use of the local approximations (RIJCOSX and DLPNO) leads to a better scaling than that for the canonical calculation (from ~6–7 down to ~2–4 for our test systems). Thus, the DefGrid3 RIJCOSX DLPNO-CCSD(T₁)-F12 method, and the L-W1X protocol that based on it, represent a useful means for obtaining accurate thermochemical quantities for larger systems.

We have used computational quantum chemistry to investigate the thermochemistry of α-hydrogen abstraction from the full set of amino acids normally found in proteins, as well as their peptide forms, by •OH and •SH radicals. These reactions, with their reasonable complexity in the electronic structure (at the α-carbon), are chosen as a consistent set of models for conducting a fairly robust assessment of theoretical procedures. Our benchmarking investigation shows that, in general, the performance for the various classes of theoretical methods improves in the order nonhybrid DFT → hybrid DFT → double-hybrid DFT → composite procedures. More specifically, we find that the DSD-PBE-P86 double-hybrid DFT procedure yields the best agreement with our high-level W1X-2 vibrationless barriers and reaction energies for this particular set of systems. A significant observation is that, when one considers relative instead of absolute values for the vibrationless barriers and reaction energies, even nonhybrid DFT procedures perform fairly well. To exploit this feature in a cost-effective manner, we have examined a number of multilayer schemes for the calculation of reaction energies and barriers for the abstraction reactions. We find that accurate values can be obtained when a "core" of glycine plus the abstracting radical is treated by DSD-PBE-P86, and the substituent effects are evaluated with M06-2X. Inspection of the set of calculated thermochemical data shows that the correlation between the free energy barriers and reaction free energies is strongest when the reactions are either endergonic or nearly thermoneutral.

Accurate double-hybrid density functional theory and isodesmic-type reaction schemes are utilized to report accurate estimates of the heats of formation (Δ_fH) for all 24 isolated-pentagon-rule isomers of the third most abundant fullerene, C₈₄. Kinetic stabilities of these C₈₄ isomers are also considered via C-C bond cleavage rates (P_cleav) calculated using density functional theory. Our results show that the relative abundance of C₈₄ fullerene isomers observed in arc discharge synthesis is the result of both thermochemical and kinetic factors. This provides timely insight regarding the characterization of several C₈₄ isomers that have been obtained experimentally to date. For instance, the established assignments of C₈₄ isomers of (using the Fowler-Manolopoulos numbering scheme) 22 [D₂(IV)], 23 [D_2d(II)], 19 [D_3d], 24 [D_6h], 11 [C₂(IV)], and 4 [D_2d(I)] are consistent with the relative Δ_fH and P_cleav values for these structures. However, our thermochemical and kinetic stabilities of C_s isomers 14, 15, and 16 indicate that the two experimentally isolated C_s isomers are 15 and 16, contrary to some previous assignments. Of the remaining isolated isomers of symmetry C₂ and D₂, definitive assignment was not possible with consideration of only Δ_fH and P_cleav.

We present a systematic assessment of the density functional tight binding (DFTB) method for calculating heats of formation of fullerenes with isodesmic-type reaction schemes. We show that DFTB3-D/3ob can accurately predict Δ_fH values of the 1812 structural isomers of C₆₀, reproduce subtle trends in Δ_fH values for 24 isolated pentagon rule (IPR) isomers of C₈₄, and predict Δ_fH values of giant fullerenes that are in effectively exact agreement with benchmark DSD-PBEP86/def2-QZVPP calculations. For fullerenes up to C₃₂₀, DFTB Δ_fH values are within 1.0 kJ mol⁻¹ of DSD-PBEP86/def2-QZVPP values per carbon atom, and on a per carbon atom basis DFTB3-D/3ob yields exactly the same numerical trend of (Δ_fH [per carbon] = 722n^−0.72 + 5.2 kJ mol⁻¹). DFTB3-D/3ob is therefore an accurate replacement for high-level DHDFT and composite thermochemical methods in predicting of thermochemical stabilities of giant fullerenes and analogous nanocarbon architectures.

We have examined the use of systematic bond-separation reactions and purposely constructed chemistry-preserving isodesmic reactions for the thermochemical calculation of aromatic hydrocarbon species. The bond-separation approach yields somewhat disappointing accuracy even when the reaction energies are obtained with generally robust composite and double-hybrid (DH) density functional theory (DFT) methods. In contrast, for the purposely constructed reactions, we find a dramatic improvement in the accuracy for energies calculated with all methods examined. Notably, for medium-sized aromatic hydrocarbons, we find that an effective approach for formulating a well-balanced reaction is to split the target species into two halves with an aromatic overlapping region. Overall, the G4(MP2)-XK, MPW2PLYP, MN15, PBE, and DC-DFTB3 methods are reasonable within their respective classes of methods for the calculation of bond-separation as well as chemistry-preserving isodesmic reactions. We have further computed per-carbon atomization energy (AE) for a series of D_6h benzene-type molecules, and thus obtained a formula for extrapolation to the graphene limit [AE_n = 711.5 × (1 - 1/n^0.640) kJ mol^-1, where n = number of carbons]. It suggests that nano-graphene with a length larger than 10 nm would resemble properties of bulk graphene, and conversely, downsizing a nano-graphene beyond this point may lead to considerably altered properties from the bulk.

We introduce a database of 14 accurate bond dissociation energies (BDEs) of noble gas compounds. Reference CCSD(T)/CBS BDEs are obtained by means of W1 theory. We evaluate the performance of contemporary density functional theory (DFT), double-hybrid DFT (DHDFT), and composite ab initio procedures. A general improvement in performance is observed along the rungs of Jacob's Ladder; however, only a handful of functionals give good performance for predicting the bond dissociation energies in the NGC14 database. Thus, this database represents a challenging test for DFT methods. Most of the conventional DFT functionals (71%) result in root-mean-square deviations (RMSDs) between 10.0 and 82.1 kJ mol^-1. The rest of the DFT functionals attain RMSDs between 2.5 and 8.9 kJ mol^-1. The best performing functionals from each rung of Jacob's Ladder are (RMSD given in parenthesis): HCTH407 (30.9); M06-L (5.4); PBE0 (2.8); B1B95, M06, and PW6B95 (2.7-2.9); CAM-B3LYP-D3 (5.4); and B2T-PLYP (2.5 kJ mol^-1).

In the present study, we have devised the G4(MP2)-XK composite method that covers species with up to fifth-row main-group elements (i.e., up to Rn). This new protocol is based on the previously published G4(MP2)-6X method, which has a general accuracy of ∼5 kJ mol^-1 for a diverse range of first- and second-row systems. The main difference between G4(MP2)-6X and G4(MP2)-XK is that the Pople-type basis sets in the former are replaced by Karlsruhe-type basis sets, with adjustments to the standard Karlsruhe basis sets to mimic the ones that they replace. Generally, G4(MP2)-XK is comparable in accuracy to G4(MP2)-6X. It is somewhat computationally more efficient than G4(MP2)-6X for the larger species that we have examined (e.g., a pentaglycine peptide). Importantly, the accuracy of G4(MP2)-XK for heavier elements is similar to that for first- and second-row species, even though it contains parameters that are fitted only to systems of the first two rows. This is indicative of the transferability of G4(MP2)-XK, and it paves the way for further expansion of its scope in future studies.