Transition-Metal Nanocluster Catalysts: Scaled-up Synthesis, Characterization, Storage Conditions, Stability, and Catalytic Activity before and after Storage of Polyoxoanion- and Tetrabutylammonium-Stabilized Ir(0) Nanoclusters

Previously, we reported the milligram-scale synthesis of tetrabutylammonium- and polyoxoanion (P 2 W 15 Nb 3 O 62 9 - )-stabilized Ir(0) nanoclusters. To increase the isolable yield of this nanocluster catalyst material to near 1 g, experiments were carried out optimizing the key synthetic variables including the solvent, the presence or absence of olefin during the nanocluster formation reaction, the isolation procedure, and the storage conditions. The ability to quantitatively monitor each nanocluster synthesis by the catalytic activity of the resultant material has led to an optimized synthesis yielding 880 ± 10 mg of 3.8 ± 0.6 nm Ir(0) ∼ 2000 nanoclusters. The scaled-up nanoclusters are characterized by TEM, elemental analysis, GC, TGA/MS, IR, 1 H, and 31 P NMR, resulting in compositionally well-characterized nanoclusters with an average stoichiometry of [( n -C 4 H 9 ) 4 N] ∼ 11000 Na ∼ 5000 Ir ∼ 2000 (P 4 W 30 Nb 6 O 123 ) ∼ 1000 (C 4 H 6 O 3 ) ∼ 50 00 (where C 4 H 6 O 3 = propylene carbonate). The isolated nanoclusters are active cyclohexene hydrogenation catalysts, possessing ∼65% of their as-formed catalytic activity even after isolation; we have also applied our recently developed CS 2 -poisoning method to determine the true number of catalytically active sites and thus the true turnover frequency (TOF) in the scaled-up nanoclusters, a rare but valuable number for any nanocluster catalyst. The isolated nanoclusters maintain their activity to within ±15% for 6 weeks when stored as a solid in a ≤5 ppm, nitrogen-filled drybox. However, a longer term storage and stability study, the first of its kind, shows that the activity of the scaled-up nanoclusters does decrease by 90% after almost three-fourths of a year (253 days) for an average loss of 2.5% of activity/week, despite the presence of the current “Gold Standard” nanocluster-stabilizing anion, P 2 W 15 Nb 3 O 62 9 -, and even though the nanoclusters are in the solid-state, are double-bottled, and are in a ≤5 ppm O 2 drybox. TEM studies of the resultant material show that agglomeration is not the cause of the deactivation process; instead, a surface deactivation process is implicated, possibly the nanoclusters titrating the ≤5 ppm O 2 out of the drybox atmosphere. The nanoclusters are, however, more stable when stored under 40 psig H 2, another valuable finding. The results point toward the needed studies of nanocluster surface structures under H 2, N 2, O 2, olefins, and other such reagents and ligands. Finally, three lines of evidence are provided indicating that the previously elucidated slow, continuous nucleation and then fast autocatalytic surface-growth nanocluster mechanism of formation also operates in the present synthesis. This in turn means that the multiple insights from that mechanism should also be applicable to the present, scaled-up Ir(0) ∼ 2000 nanocluster synthesis.