The direct approach is not always the best: butadiene plus dichlorocarbene

The four-electron thermal cycloaddition (in reverse a cheletropic elimination) of dichlorocarbene to ethene is a classic example of a forbidden pericyclic process taking a roundabout route to avoid directly violating the Woodward-Hoffmann rules. However, a thermal six-electron process normally does take the direct route, as in for example the Diels-Alder cycloaddition as Houk and co have recently showed using molecular dynamics[1]. So can one contrive a six-electron cycloaddition involving dichlorocarbene?

Surely, it should now form the two new C-C bonds at the same time (synchronously)? Well, here comes a ωB97XD/6-311G(d,p)/SCRF=dichloromethane intrinsic reaction coordinate calculation:

1. The reaction starts at IRC -5, 
2. and proceeds with only a small barrier to the transition state (IRC =0.0) 
3. At IRC +4, the potential flattens out and the gradients drop, with formation of the first C-C bond completed. But the gradients do not quite go to zero, which would have implied the formation of a discrete intermediate such as:
4. The concerted reaction continues and by IRC ~ +11, the two chlorine atoms now exhibit quite different C-Cl lengths. The one that is orthogonal to the second forming C-C bond is normal (1.815Å), whereas the one antiperiplanar to the C-C bond is 1.92Å. There are some interesting stereoelectronic alignments involved.
5. Coincidentally perhaps, but these phenomena of an intermediate almost forming in a system containing a CCl₂ group with concomitant lengthening of one C-Cl bond compared to the other, was also observed in my IRC for the addition of thiolate to a dichlorobuteneone. For that system,  Dan Singleton’s work had shown that molecular dynamics is necessary to obtain a more complete picture, and that may well be also true for the example here!  Perhaps Ken Houk might give it a go!
6. The second C-C bond then completes at around IRC +16.

Well, this shows that a reaction only modestly removed from the classical six-electron Diels-Alder can change character dramatically from the synchrony expected of the latter. I am hunting for a simple explanation of this phenomenon, but perhaps participation of the C-Cl bonds makes this different from a simple cycloaddition. Or possibly, the explanation will only properly emerge when the molecular dynamics is studied?

References

1. K. Black, P. Liu, L. Xu, C. Doubleday, and K.N. Houk, "Dynamics, transition states, and timing of bond formation in Diels–Alder reactions", Proceedings of the National Academy of Sciences, vol. 109, pp. 12860-12865, 2012. http://dx.doi.org/10.1073/pnas.1209316109