Supporting Information ucleophile-catalyzed Additions to Activated Triple Bonds. Protection of Lactams, Imides, and ucleosides with MocVinyl and Related Groups Laura Mola, Joan Font, Lluís Bosch, Joaquim Caner, Anna M. Costa,* Gorka Etxebarria-Jardí, riol Pineda, David de Vicente, and Jaume Vilarrasa* 1. Relevant 1 H and 13 C MR spectra.... S2 2. Calculated energies of the main possible conformers of of 1a, 2a, 5a, 6a, the 1-Me analog of 7a, and the 9-Me analogs of 8a 11a (E stereoisomers), as well as of the lowest energy conformer of the corresponding Z isomers... S31 3. Calculated equilibrium energies for the exchanges of MocVinyl group... S32 4. Comparison of intermediates of E and Z configuration... S33 S1
1. Relevant 1 H and 13 C MR spectra S2
S3
S4
S5
S6
S7
CMe (E)-5a Ph S8
S9
S10
CMe Ac Ac Ac (E)-8a S11
CMe Ac (E)-9a S12
CMe TBS (E)-10a S13
S14
S15
S16
S17
S18
C t Bu TBS (E)-10b S19
S20
S21
CMe Ac Ac Ac (E)-8c S22
CMe Ac (E)-9c S23
S24
S25
S26
Ac (E)-7d 3 S27
S28
Me Ac (E)-7e 3 S29
Me Ac Ac Ac (E)-8e S30
2. Calculated energies of the main possible conformers of 1a, 2a, 5a, 6a, the 1-Me analog of 7a, and the 9-Me analog of 8a 11a (stereoisomers E), as well as of the lowest energy conformer of the corresponding Z isomers The main possible conformers of representative MocVinyl-protected heterocycles have been calculated with the Gaussian 03 package, 6 in CHCl 3 (with optimization, PCM), 7 at three different levels. Total energies are given in au or hartrees (1 au = 627.5 kcal/mol). The stationary points were characterized by frequency calculations to verify that minima have no imaginary frequency. Within parentheses, in red, energy differences in kcal/mol, referred to the lowest energy species of each series. As shown in the next figure, s-cis and s-trans conformers (by rotation through Cβ and through Cα CMe) were calculated, while only the s-cis conformers of the ester group (Me C=) were systematically examined, since, as expected, representative cases of derivatives with the CMe in s-trans arrangements turned out to be much higher in energy. (Row 1) (Row 2) (Row 3) B3LYP/6-31+G(d) (CHCl 3, optim.) MP2/6-31+G(d)//B3LYP/6-31+G(d) (CHCl 3, optim.) MP2/6-311+G(d,p)//B3LYP/6-31+G(d) (CHCl 3, optim.) S31
MocVinyl groups of configuration E are found to be practically coplanar with the heterocyclic ring (most CCβCα dihedral angles, θ, are 0±4º or 180±4º), unless otherwise indicated; only values of φ > 5 are pointed out. n the other hand, MocVinyl groups of configuration Z are outside the plane of the heterocyclic ring (largely or slightly, depending on the presence or absence of repulsive interactions between the CMe group and the atoms or substituents at one or both sides of the atom). It is observed in the preceding Table that most conformers of stereoisomers E are predicted to be more stable than the Z stereoisomer of lowest energy (which, according to the results at the highest level examined here, lies 1.9 4.9 kcal/mol above the corresponding absolute minimum). In principle, this suggests that, independently of the kinetic E/Z ratio, it could be feasible to isomerize Z isomers or nearly equimolar mixtures of E/Z isomers to almost stereopure E, if required; this could be confirmed experimentally in some cases. For the sake of simplification (and saving of computer time) G values were not calculated. We assume that the calculated relative energies among isomers can be correlated with their relative free energies. We have compared (previous figure, within a box) the relative energy of the -MocVinyl isomer of 6a with regard to (E)-6a (the - MocVinyl species of lowest energy at the B3LYP/6-31+G(d) level). Due to the aromaticity of the pyridine ring, it was the case in which an -MocVinyl species had more chances to be formed and detected. As explained in the main text, this -MocVinyl species was isolable. It could be fully converted into (E)-6a only by heating in CH 3 C in the presence of DABC. 3. Calculated equilibrium energies for the exchange of the MocVinyl group The energies calculated for the exchanges between succinimide or its anion and a model of 8 11 (inosine derivatives) are shown below. It was commented in the main text that pyrrolidine was unsuitable to deprotect MocVinyl derivatives 8a 11a due to the fact that mixtures of products of deprotection and attack at C2 were obtained, so we developed an alternative addition elimination process based on the use of succinimide anion. Representative reactions corresponding to the exchanges of MocVinyl groups of configuration E among the neutral molecules shown above (provided that a basic or nucleophilic catalyst could be involved to make the exchange kinetically feasible) and among their anions are indicated below. Calculations predict that the possible equilibria between succinimide (2) and inosine model (8 11) 8 for "trapping" a MocVinyl group are shifted to the right (we assume again that thermal and entropic terms are compensated). When equilibria in strong basic media are compared, the figures change but the reactions are predicted to be even more feasible. S32
4. Comparison of intermediates of E and Z configuration A full study, at a high level of theory, of the potential hypersurface of the addition reactions leading to one or another stereoisomer was outside the scope of the present work and of our computational facilities. evertheless, we have compared the first steps of the process the two alternative set of intermediates as an entry to the subject, in order to try and understand the origin of the observed stereoselectivity when excellent nucleophiles such as DABC are involved as the catalysts. The energies are in au or hartrees, unless otherwise indicated. 6. Gaussian 03: Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb, M. A.; Cheeseman, J. R.; Vreven, T.; Kudin, K..; Burant, J. C.; Millam, J. M.; Iyengar, S. S.; Tomasi, J.; Barone, V.; Mennucci, B.; Cossi, M.; Scalmani, G.; Rega,.; Petersson, G. A.; akatsuji, H.; Hada, M.; Ehara, M.; Toyota, K.; Fukuda, R.; Hasegawa, J.; Ishida, M.; akajima, T.; Honda, Y.; Kitao,.; akai, H.; Klene, M.; Li, X.; Knox, J. E.; Hratchian, H. P.; Cross, J. B.; Bakken, V.; Adamo, C.; Jaramillo, J.; Gomperts, R.; Stratmann, R. E.; Yazyev,.; Austin, A. J.; Cammi, R.; Pomelli, C.; chterski, J. W.; Ayala, P. Y.; Morokuma, K.; Voth, G. A.; Salvador, P.; Dannenberg, J. J.; Zakrzewski, V. G.; Dapprich, S.; Daniels, A. D.; Strain, M. C.; Farkas,.; Malick, D. K.; Rabuck, A. D.; Raghavachari, K.; Foresman, J. B.; rtiz, J. V.; Cui, Q.; Baboul, A. G.; Clifford, S.; Cioslowski, J.; Stefanov, B. B.; Liu, G.; Liashenko, A.; Piskorz, P.; Komaromi, I.; Martin, R. L.; Fox, D. J.; Keith, T.; Al-Laham, M. A.; Peng, C. Y.; anayakkara, A.; Challacombe, M.; Gill, P. M. W.; Johnson, B.; Chen, W.; Wong, M. W.; Gonzalez, C.; Pople, J. A. Gaussian 03, Revision E.01; Gaussian, Inc.: Wallingford CT, 2004. 7. For an entry to the Polarized Continuum Model (PCM, solvent effects), see: Tomasi, J.; Mennucci, B.; Cammi, R. Chem. Rev. 2005, 105, 2999 3094. 8. For some entries to DFT or MP2 calculations of 1, 2, and the nucleobase models of 7 11 studied in this SI, see: (a) Vogt,.; Khaikin, L. S.; Grikina,. E.; Karasev,. M.; Vogt, J.; Vilkov, L. V. J. Phys. Chem. A 2009, 113, 931 937. (b) Jang, Y. H.; Hwang, S.; Chung, D. S. Chem. Lett. 2007, 36, 1496 1497. (c) Kawahara, S.; Uchimaru, T.; Taira, K.; Sekine, M. J. Phys. Chem. A 2002, 106, 3207 3212. (d) Galetich, I.; Kosevich, M.; Shelkovsky, V.; Stepanian, S. G.; Blagoi, Yu. P.; Adamowicz, L. J. Mol. Struct. 1999, 478, 155 162. (e) Portalone, G.; Bencivenni, L.; Colapietro, M.; Pieretti, A.; Ramondo, F. Acta Chem. Scand. 1999, 53, 57 68. S33
The true TS structures only one imaginary frequency leading to the most stable allenolates/zwitterions could be located. The energy barriers are relatively low (quick first steps, in agreement with the experimental results). The kinetically favored zwitterion is the Z stereoisomer, whereas the E stereoisomer is thermodynamically favored. The process appears to be under thermodynamic control. The energy gap (without solvent) between the two zwitterionic allenolates (with non-linear structures of E type and Z type) was predicted to be 2 kcal/mol, at the MP2/6-31+G(d) level. In a polar solvent (in CH 3 C, without geometry optimization), the energy gap increased up to 3 kcal/mol at the same level. In summary, although we recognize that the in silico study described above is preliminary, the calculations predict that intermediates of type E are always few kcal/mol more stable than the corresponding intermediates of type Z, independently of their zwitterionic features and their cationic structure. The differences are similar to those calculated for the final products (isomers E and Z), which was not so obvious at first glance. Thus, along the nucleophile-catalyzed stepwise process the thermodynamic bias in favor of the E isomers seems to be maintained, in agreement with the experimental outcome. S34