Computational Chemistry

seeking to migrate to an end carbon. To summarize: the HF calculations led to a
hilltop and a transition state, the MP2 calculations to a transition state and a relative
minimum, and the B3LYP calculations to two transition states. We see below that
many more stationary points resembling 1,3-propanediyl can be found by appropri-
ate methods.
1,4-Butanediyl, tetramethylene.Three different starting geometries were used
(Fig.8.8), with symmetry C 2 ,C2h, and C 1 , and each was submitted to a geometry
optimization/frequency calculation by the HF, the MP2, and the B3LYP method. In
all cases the U-shaped C 2 input geometry closed to a cyclobutane molecule and the
zigzag C2hand C 1 geometries dissociated to two ethene molecules. We see below
that many stationary points resembling 1,4-butanediyl can be found by appropriate
methods.

8.2.3 (1) Singlet Diradicals: Beyond Model Chemistries.

(2) Complete Active Space Calculations (CAS)

8.2.3.1 (1) Singlet diradicals: Beyond model chemistries

We now look at results of calculations on the 1,3- and 1,4-diradicals by methods
more appropriate than the model chemistries just employed.

Table 8.3 Results of attempted geometry optimization of the.CH 2 CH 2 CH 2.

singlet diradical by different model chemistries; the 6-31G* basis was used in all

cases. See Fig.8.7for the input structures and text for clarification

Symmetry of

input structure

HF MP2 B3LYP

C 1 Cyclopropane Cyclopropane Cyclopropane

C 2 Cyclopropane Cyclopropane Cyclopropane

Cs Cyclopropane Cyclopropane Cyclopropane

C2v p-cyclopropane? p-cyclopropane? p-cyclopropane?

Fig. 8.8 The input structures for attempted model chemistry optimizations on 1,4-butanediyl
(.CH 2 CH 2 CH 2 CH 2 .). All bond lengths and angles in these structures were standard, e.g. C–C ca.
1.5 A ̊, C–H ca. 1.1 A ̊, bond angles ca. 110–120

8.2 Singlet Diradicals 537