Benchmarking Isomerization Energies for C5$$ {\mathrm{C}}_5 $$–C7$$ {\mathrm{C}}_7 $$ Hydrocarbons: The ISOC7 Database

Highly accurate benchmark databases are critical for the development of robust and computationally efficient electronic structure methods. We introduce the ISOC7 database, a diverse collection of 1308 unique constitutional isomers of <mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML"> <mml:semantics> <mml:mrow> <mml:msub><mml:mrow><mml:mtext>C</mml:mtext></mml:mrow> <mml:mrow><mml:mn>5</mml:mn></mml:mrow> </mml:msub> </mml:mrow> <mml:annotation>$$ {\mathrm{C}}_5 $$</mml:annotation></mml:semantics> </mml:math> - <mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML"> <mml:semantics> <mml:mrow> <mml:msub><mml:mrow><mml:mtext>C</mml:mtext></mml:mrow> <mml:mrow><mml:mn>7</mml:mn></mml:mrow> </mml:msub> </mml:mrow> <mml:annotation>$$ {\mathrm{C}}_7 $$</mml:annotation></mml:semantics> </mml:math> saturated and unsaturated hydrocarbons with reference isomerization energies at the CCSD(T)/CBS level of theory, obtained via the W1-F12 thermochemical protocol. The isomerization energies in this dataset span over 146 <mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML"> <mml:semantics><mml:mrow><mml:mtext>kcal</mml:mtext> <mml:mspace></mml:mspace> <mml:msup><mml:mrow><mml:mtext>mol</mml:mtext></mml:mrow> <mml:mrow><mml:mo>-</mml:mo> <mml:mn>1</mml:mn></mml:mrow> </mml:msup> </mml:mrow> <mml:annotation>$$ \mathrm{kcal}\kern0.3em {\mathrm{mol}}^{-1} $$</mml:annotation></mml:semantics> </mml:math> . This database was used to conduct a rigorous benchmark assessment of a wide hierarchy of computational methods. The performance of 40 contemporary density functional theory (DFT) functionals reveals a general, albeit not monotonic, improvement along the rungs of Jacob's Ladder, with lower-rung GGA and MGGA functionals providing generally poor performance. The range-separated hybrid-meta-GGA functional <mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML"> <mml:semantics><mml:mrow><mml:mi>ω</mml:mi></mml:mrow> <mml:annotation>$$ \omega $$</mml:annotation></mml:semantics> </mml:math> B97M-D4 emerges as the top DFT performer with a root-mean-square deviation (RMSD) of 1.62 <mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML"> <mml:semantics><mml:mrow><mml:mtext>kcal</mml:mtext> <mml:mspace></mml:mspace> <mml:msup><mml:mrow><mml:mtext>mol</mml:mtext></mml:mrow> <mml:mrow><mml:mo>-</mml:mo> <mml:mn>1</mml:mn></mml:mrow> </mml:msup> </mml:mrow> <mml:annotation>$$ \mathrm{kcal}\kern0.3em {\mathrm{mol}}^{-1} $$</mml:annotation></mml:semantics> </mml:math> . We also evaluated computationally economical semiempirical and tight-binding methods. While traditional semiempirical approaches are inadequate, the modern g-xTB tight-binding method achieves a respectable RMSD of 4.14 <mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML"> <mml:semantics><mml:mrow><mml:mtext>kcal</mml:mtext> <mml:mspace></mml:mspace> <mml:msup><mml:mrow><mml:mtext>mol</mml:mtext></mml:mrow> <mml:mrow><mml:mo>-</mml:mo> <mml:mn>1</mml:mn></mml:mrow> </mml:msup> </mml:mrow> <mml:annotation>$$ \mathrm{kcal}\kern0.3em {\mathrm{mol}}^{-1} $$</mml:annotation></mml:semantics> </mml:math> . Remarkably, the machine-learned neural network potential AIMNet2 delivers exceptional accuracy, achieving an RMSD of 1.67 <mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML"> <mml:semantics><mml:mrow><mml:mtext>kcal</mml:mtext> <mml:mspace></mml:mspace> <mml:msup><mml:mrow><mml:mtext>mol</mml:mtext></mml:mrow> <mml:mrow><mml:mo>-</mml:mo> <mml:mn>1</mml:mn></mml:mrow> </mml:msup> </mml:mrow> <mml:annotation>$$ \mathrm{kcal}\kern0.3em {\mathrm{mol}}^{-1} $$</mml:annotation></mml:semantics> </mml:math> , rivaling the performance of the best DFT functional at a fraction of the computational cost. The ISOC7 database provides a challenging benchmark for advancing the development and validation of both quantum chemical and machine learning methods.