A case study of proton shuttling in palladium catalysis

Thanks to mechanistic studies, the catalytic performance of SCS indenediide Pd pincer complexes has been spectacularly enhanced using catechol additives as proton shuttles.

II.b. Partial order determination: The partial order of each reaction's components (substrate and catalyst) was determined by the initial rate method. The data of the concentration of product versus time plot were fitted with Excel. The obtained slope of the linear fitting represents the initial rate. The partial order was then determined by plotting the initial rates versus the initial concentrations.

II.b.1. Partial order determination for [Pd]
To determine the partial order of the reaction on catalyst, the initial kinetic profiles at different initial concentrations of palladium center were recorded. The final data were obtained by averaging the results of three independent trials for each experiment. Experiment % mol [Pd] x mg of Cat. I Initial concentration [Pd] 1 3 1.5 0.0042 2 5 2.5 0.007 3 7.5 3.8 0.0105 4 10 5.1 0.014 Figure S1. Reaction conditions for the partial order determination of [Pd].
General procedure: 10.8 -hexynoic acid 1a (0.0098 mmol, 0.14 M) a specific quantity of Catalyst I according to the above table and 0.7 mL of CDCl3 were introduced in a pressure NMR tube. The S3 reaction mixture was heated at 90°C and 1 H NMR spectra were recorded every five minutes until a conversion of 10%.

II.b.2. Partial order determination for 5-hexynoic acid 1a.
To determine the partial order of the reaction in 5-hexynoic acid, the initial kinetic profiles at different initial concentrations of 5-hexynoic acid were recorded. The final data were obtained by averaging the results of three independent trials for each experiment.

II.b.3. Partial order determination for 5-hexynoic acid 1a in presence of 1 mol% of Tetrachlorocatechol 4u.
To determine the partial order of the reaction on 5-hexynoic acid in presence of 1 mol% of Tetrachlorocatechol 4u, the initial kinetic profiles at different initial concentrations of 5-hexynoic acid were recorded. The final data were obtained by averaging the results of two independent trials for each experiment.  [1a] (mol.L -1 ) V0(1) (mol.L -1 .min -1 ) V0(2) (mol.L -1 .min -1 ) Average of V0 (mol.L -1 .min -1 ) 0.14 0. To determine the partial order of the reaction on 5-hexynoic acid in presence of 5 mol% of Tetrachlorocatechol 4u, the initial kinetic profiles at different initial concentrations of 5-hexynoic acid were recorded. The final data were obtained by averaging the results of two independent trials for each experiment.  To determine the partial order of the reaction on 5-hexynoic acid in presence of 10 mol% of Tetrachlorocatechol 4u, the initial kinetic profiles at different initial concentrations of 5-hexynoic acid were recorded. The final data were obtained by averaging the results of two independent trials for each experiment.  II.b.6. Partial order determination for 5-hexynoic acid 1a in presence of 20 mol% of Tetrachlorocatechol 4u.
To determine the partial order of the reaction in 5-hexynoic acid in presence of 20 mol% of Tetrachlorocatechol 4u, the initial kinetic profiles at different initial concentrations of 5-hexynoic acid were recorded. The final data were obtained by averaging the results of two independent trials for each experiment.   Figure S22. Partial order of 5-hexynoic 1a versus the quantity of tetrachlorocatechol 4u.

III.
Study of the self association 5-hexynoic acid in CHCl3 by IR spectroscopy. The self-association of 5-hexynoic acid was evidenced by IR spectroscopy in CHCl3 at different acid concentrations (Vide Infra). IR spectra were recorded with a resolution of 4 cm -1 in 16 scans. Two well defined absorptions bands were observed, at 1711 and 1751 cm -1 corresponding to the dimeric and monomer forms, respectively. The observed bands are shown in Figure S23 at different concentrations and reported in Table S8 with the intensities ratio.

31
P NMR analysis of the Palladium indenediide dimer I at variable temperature. The dissociation-association behavior of the Palladium indenediide dimer I was evidenced by 31 P NMR spectroscopy at variable temperature using a 400 MHz NMR spectrometer. The association activation barrier was estimated from these experiments ( Figure S24), using the following formula: The association activation barrier was estimated to be at least of 15.8 kcal.mol -1 , considering the 31 P NMR spectroscopic data: the coalescence temperature Tc = 363 K, and the chemical-shift difference │AB │ = 1017.33 Hz.

VII.
General procedure for cycloisomerization of alkynoic acids in presence of additives: In a NMR pressure tube, alkynoic acid (0.098 mmol), dried additive (x mol%) and complex I (2.5 mg, 5 mol% [Pd]) in 0.7 mL of CDCl3 were heated at the corresponding temperature under argon atmosphere. The progress of the reaction was monitored by 1 H NMR.   Figure S26. H-bond additives library used in the cycloisomerization of 5-hexynoic acid 1a catalyzed by indenediide dimer I. Figure S27. Evaluation of the impact of weak H-donor compounds 4 (30 %mol) on the cyclization of 5-hexynoic acid 1a. Table S10. Evaluation of the additives in the cyclization of 5-hexynoic acid 1a and optimization of the reaction conditions.

IX.
Computational details Calculations were carried out with the Gaussian 09 program S3 on the real experimental palladiumpincer system at the B3PW91 level of theory. S4 Palladium atom was treated with the corresponding Stuttgart-Dresden RECP (relativistic effective core potential) in combination with its adapted basis set, S5 augmented by an extra set of a f polarization function. S6 Phosphorus atoms were represented by the ECP from Dolg et al. and its associated basis set, S7 augmented also by d polarization functions. S8 For the remaining atoms the 6-31G(d,p) basis set was used. S9 Geometry optimizations carried out without any symmetry restrictions, and were followed by analytical frequency calculations to confirm that a minimum or a transition state had been reached. The connection between the transition state and the corresponding minima were done by performing IRC calculations. S10 Finally, the CYLview program was used for the representation of 3D structures. S11