Organocatalysis affected by proline is an extremely active research area, and computational chemists have made considerable contributions (see Chapter 5.3 of my book – and this is expanded on in the 2nd edition which should be out in just a few months). Most importantly, the Houk-List model1 for the catalysis was largely developed on the basis of computations.
Recent experiments have indicated cocatalysts that can hydrogen bond to proline may increase the catalytic effect, including the enantioselectivity. Xue and co-workers have examined a series of potential cocatalysts for their ability to enhance the acidity of proline.2 This is important in that a proton transfer is a component to the key step of the Houk-List model.
The cocatalysts examined included such compounds as 1–6. The deprotonation energy of proline with the associated cocatalysts was compared with that of proline itself. The energies were computed at M06-2x/6-311++G(2df,2p)//B3LYP/6-31+G(d) with the SMD treatment of five solvents. The structure of 5 with proline is shown in Figure 1.
 5 with proline |
 5 with proline conjugate base |
Figure 1. M06-2x/6-311++G(2df,2p)//B3LYP/6-31+G(d) optimized structure of 5 with proline and its conjugate base.
The effect of the cocatalysts is striking. In the gas phase, these additives decrease the pKa of proline by 15 – 70 pKa units, with 2 showing the largest effect. In solvent, the effect of the cocatalyst is attenuated, especially in more polar solvents, but still a considerable decrease in the pKa is seen (as much as a 12 pKa unit increase in acidity). Further exploration of potential cocatalysts seems fully warranted.
References
(1) Allemann, C.; Gordillo, R.; Clemente, F. R.; Cheong, P. H.-Y.; Houk, K. N. "Theory of Asymmetric Organocatalysis of Aldol and Related Reactions: Rationalizations and Predictions," Acc. Chem. Res. 2004, 37, 558-569, DOI: 10.1021/ar0300524.
(2) Xue, X.-S.; Yang, C.; Li, X.; Cheng, J.-P. "Computational Study on the pKa Shifts in Proline Induced by Hydrogen-Bond-Donating Cocatalysts," J. Org. Chem. 2014, 79, 1166–1173, DOI: 10.1021/jo402605n.
InChIs
1: InChI=1S/CH4N2O/c2-1(3)4/h(H4,2,3,4)
InChIKey=XSQUKJJJFZCRTK-UHFFFAOYSA-N
2: InChI=1S/CH5N3/c2-1(3)4/h(H5,2,3,4)/p+1
InCiKey=ZRALSGWEFCBTJO-UHFFFAOYSA-O
3: InChI=1S/C4H4N2O2/c5-1-2(6)4(8)3(1)7/h5-6H2
InChIKey=WUACDRFRFTWMHE-UHFFFAOYSA-N
4: InChI=1S/C13H12N2S/c16-13(14-11-7-3-1-4-8-11)15-12-9-5-2-6-10-12/h1-10H,(H2,14,15,16)
InChIKey=FCSHMCFRCYZTRQ-UHFFFAOYSA-N
5: InChI=1S/C20H14O2/c21-17-11-9-13-5-1-3-7-15(13)19(17)20-16-8-4-2-6-14(16)10-12-18(20)22/h1-12,21-22H
InChIKey=PPTXVXKCQZKFBN-UHFFFAOYSA-N
6: InChI=1S/C7H13N3/c1-3-8-7-9-4-2-6-10(7)5-1/h1-6H2,(H,8,9)/p+1
InChIKey=FVKFHMNJTHKMRX-UHFFFAOYSA-O
Proline: InChI=1S/C5H9NO2/c7-5(8)4-2-1-3-6-4/h4,6H,1-3H2,(H,7,8)/t4-/m0/s1
InChIKey= ONIBWKKTOPOVIA-BYPYZUCNSA-N
Prolinator responded on 05 Mar 2014 at 4:49 am #
I think that explicit solvent molecules need to be included to expand the study. It has been shown previously that the polar solvents used in proline catalysis (DMSO for example) are also capable of interacting with the catalyst (and catalytic intermediates) in a stabilising manner. It is also relevent to consider the solvent-additive interaction (which would be lost upon catalyst-additive interaction).
The results with apolar solvents are the most interesting to me because solvent interactions should be minimal and proline alone would have extermely low solubility – so if nothing else the additive could assist solvation.
Henry Rzepa responded on 06 Mar 2014 at 2:39 am #
We recently reported insertion of a water-relay into a proline catalysis pathway (doi: 10.1039/C3SC53416B), but this did not lower the overall free energy barrier. It is certainly true that more basic/acidic molecules such as shown above do have a quite different effect from water! Thus the effect of such molecules on the kinetic vs thermodynamic enolisation of 1-methyl cyclohexanone can be observed and here for oxime formation from hydroxylamine and propanone.
Peter R. Schreiner responded on 26 Mar 2014 at 3:19 am #
Enhancing the acidity of an acid through cooperative interaction with an organocatalyst (formally also a Brønsted acid) was first shown both experimentally and computationally in Org. Lett. 2008, 10, 1513–1516, DOI: 10.1055/s-2008-1072763 and highlighted in Synfacts 2008 (06), 644. This concept of “cooperative organocatalysis” was then picked up again in Science 2010, 327, 986; see also ibid. page 965).