Interplay between the Directing Group and Multifunctional Acetate Ligand in Pd-Catalyzed anti-Acetoxylation of Unsymmetrical Dialkyl-Substituted Alkynes

The cooperative action of the acetate ligand, the 2-pyridyl sulfonyl (SO2Py) directing group on the alkyne substrate, and the palladium catalyst has been shown to be crucial for controlling reactivity, regioselectivity, and stereoselectivity in the acetoxylation of unsymmetrical internal alkynes under mild reaction conditions. The corresponding alkenyl acetates were obtained in good yields with complete levels of β-regioselectivity and anti-acetoxypalladation stereocontrol. Experimental and computational analyses provide insight into the reasons behind this delicate interplay between the ligand, directing group, and the metal in the reaction mechanism. In fact, these studies unveil the multiple important roles of the acetate ligand in the coordination sphere at the Pd center: (i) it brings the acetic acid reagent into close proximity to the metal to allow the simultaneous activation of the alkyne and the acetic acid, (ii) it serves as an inner-sphere base while enhancing the nucleophilicity of the acid, and (iii) it acts as an intramolecular acid to facilitate protodemetalation and regeneration of the catalyst. Further insight into the origin of the observed regiocontrol is provided by the mapping of potential energy profiles and distortion–interaction analysis.


General methods.
All reagents and solvents were purchased from commercial sources and used as received. Starting materials were prepared according to previous method reported in the literature. 1,2 All reactions were carried out in anhydrous solvents and air atmosphere, unless otherwise noted. Column liquid chromatographies were performed on silica gel (230-400 mesh ASTM). TLC analysis was performed on 0.2 mm aluminium-based plates (60 230-400 mesh). 1 H, 13 C, and 19 F NMR spectra were recorded in CDCl3 or AcOD-d4 solutions at 25 °C (indicated on each case) on AV-300, AVII-300 y AVIII-HD-300 (300, 75, and 282 MHz, respectively) spectrometers (δ, ppm; J, Hz). 1 H and 13 C NMR spectra were referenced using the solvent signal as internal standard. HRMS electrospray ionization (ESI+) were recorded using an API-QToF ESI with a mass range from 20 to 3000 m/z and mass resolution 15000 (FWHM). Melting points were determined in open-end capillary tubes.

Pd-catalyzed acetoxylation of unsymmetrical internal alkynes
General procedure for the acetoxylation of propargyl 2-pyridyl sulfones: In a scintillation vial charged with a magnetic stir bar, the corresponding propargyl 2-pyridyl sulfone 1 (0.2 mmol) and Pd(OAc)2 (2.24 mg, 0.01 mmol, 5 mol%) were dissolved in 1.0 mL of acetic acid. The mixture was allowed to stir at room temperature for 10 minutes. Then, the reaction vial was placed into a pre-heated oil bath and stirred at 80-100 °C (indicated in each case). The reaction was followed by TLC analysis after full completion was observed. After that, the reaction was cooled to room temperature, and the residue was filtered off through a pad of silica gel employing AcOEt as eluent. The solvent was removed in vacuo and the resulting residue was further purified by flash column chromatography.

KIE experiments
The determination of the KIE was performed by comparison of the reaction rate of propargyl sulfone 1a in either acetic acid or acetic acid-d4 taking aliquots of the reaction at the indicated time ( Figure S1). The reaction kinetics was performed on a 0.4 mmol scale of 1a taking aliquots of 0.1 mL at the indicated time. The resulting sample was further passed through a pad of silica gel and 0.1 mL of a 0.067 M stock solution of 1,3,5-trimethoxybenzene (TMB) in CH2Cl2 was added. After that, the solvent was evaporated in vacuo and the mixture was analyzed by 1 H NMR spectroscopy employing CDCl3 as the solvent. The conversion of starting material and yield of acetoxylation product 2a were determined by comparing the integrals taking the singlet at 6 ppm from TMB as the reference (see Table S1 for the data).  Figure S1. Reaction conversions for acetoxylation of propargyl sulfone 1a in acetic acid (red) and acetic acid-d4.
Plotting ln(100-conv) vs time for each kinetic data ( Figure S2), the corresponding k-values for the Hand the D-experiments can be obtained: Figure S2. Determination of KIE for acetoxylation in AcOH (red) and AcOD-d4 (black).

Determination of the stereochemistry
The configuration of the alkene Z-3a was determined by performing quantitative nOe experiments. The corresponding correlation between the olefinic proton and allylic protons is given: Figure S3. Correlation between HB and HC by nOe experiment. The HB proton was irradiated showing a correlation of 1.23% with HC. Figure S4. Correlation between HB and HA by nOe experiment. The HA proton was irradiated showing a correlation of 0.41% with HB.

Computational Details
All calculations have been carried out within the framework of density functional theory, using Gaussian 09 package. 3 In particular, gas-phase geometry optimizations and frequency analysis, the meta-GGA M06-L exchange-correlation functional 4 was employed, in combination with the cc-pVDZ basis set 5 for C, H, N, O and S atoms, and the def2-SVP basis set and ECP 6 for Pd. More accurate electronic energies were obtained through single-point energy calculations at the M06-L/cc-pVTZ(C,H,N,O,S),def2-TZVP(Pd) level of theory, including solvent effects (AcOH) using the SMD model. 7 The condensed Fukui functions 8 can be used to estimate the relative electrophilicity or nucleophilicity of a given atomic position in a molecule. In contrast to the "global" Fukui functions, which are obtained from the difference between the unperturbed electron density (no additional electrons) and the electron density upon addition/removal of one electron, condensed Fukui functions come from the variation in the atomic charges (populations) under the same circumstances. Therefore, to evaluate the electrophilicity of the Cα and Cβ positions of propargyl sulfone 1a, NBO analysis allowed us to calculate their population variation upon addition of an extra electron, f - . Larger values of fi are related to larger electrophilicities.

Electrophilicity of Cα and Cβ positions
We have calculated the condensed Fukui functions at the Cα and Cβ positions (fα and fβ) to qualitatively assign the intrinsic reactivity of the triple bond in different scenarios (Table S1). We considered the free propargyl sulfone I1, the I1-Pd(OAc)2 complex (I2, with and without Pd-alkyne bond), and I2 with one or two additional AcOH molecules (linked by hydrogen bonds to the acetate ligands). We also evaluated the final adducts I7 and I8 (see Figure 2 of the main text), to assess if a second acetoxylation is feasible given that there exists a Pd-alkene bond in those structures. These results (see below) show that formation of a Pd-alkyne bond is critical to increase the electrophilicity of the triple bond, and that subsequent coordination of AcOH molecules slightly rises it. On the other hand, the lower electrophilicity of the double bond prevents diacetoxylation.

Evaluation of alternative reaction pathways
Apart from intermolecular AcOH α-and β-addition, we also calculated three additional pathways. One of them is the intramolecular syn insertion of the alkyne into the Pd-OAc bond, which would lead to the β-adduct but with opposite stereochemistry, and unable to give an intramolecular protodepalladation ( Figure S5). The other pathways are also intermolecular AcOH additions but considering a second AcOH molecule linked to the complex (see below, AcOH marked with dashed circles). None of these pathways are more favorable than the intermolecular β-addition with just one AcOH, either due to larger activation barriers or endergonic coordination of the second AcOH molecule. Figure S5. Alternative pathways.

IRC analysis with the Distortion/Interaction model
In the main text, we have presented the distortion-interaction analysis for the IRC of the acetoxylation considering one AcOH molecule. We also calculated the reaction pathways including an additional AcOH molecule, linked through hydrogen bonding to the non-participating AcO ligand. Thus, this second AcOH molecule should account for explicit solute-solvent interaction. In the previous section, we showed that this additional AcOH does not have any substantial effect on the activation barriers, and that the energy profile is just up-shifted due to entropy loss. Furthermore, identical conclusions can be extracted if the distortion-interaction analysis is carried out for these reaction pathways considering 2 AcOH molecules, as shown below ( Figure S6). In this case, [Pd] stands for 1a-Pd(OAc)2-AcOH (the AcOH that does not participate in the reaction). Figure S6. Distortion-interaction model considering 2 molecules of AcOH.