Cure
kinetics control of epoxy resins is critical for the realization
of many structures and processes and is often manipulated by catalyst
design. We here show an example of switchable Lewis base catalytic
activity through ligand-controlled metal coordination. Divalent first-row
transition-metal (Co, Ni, Cu, Zn) β-diketonates with methyl
or trifluoromethyl end groups have found distinguished thermal latent
curing behaviors in triphenylphosphine (TPP)-catalyzed epoxy resins,
namely, a deceleration pattern for metal acetylacetonates (acac2) and an inhibition pattern for metal hexafluoroacetylacetonates
(6Facac2). Comparative analysis exposed the major initiation
mechanism as phosphine attack on epoxide rings, where the phosphine
reactivity was regulated by metal coordination whose strength depends
on the original diketone ligands. TPP further stabilized the metal
chelates and suppressed their dissociation. Feed ratio studies of
Co(II) chelates revealed an equilibrium built upon TPP, metal chelate,
and the formed passivated complex through numerical analysis. Further,
temperature dependence of the equilibrium constants suggested a reversed
metal-base affinity evolution of the two chelates during heating,
which determines the equivalent TPP concentration. Chemical and thermal
characterizations on the formed complexation states identified structural
changes during high-temperature treatment and, along with density
functional theory (DFT) calculation, verified the Co–P binding
energy that marks the TPP “effectiveness” in each stage
to catalyze epoxy cure. It was found that the competition between
incoming phosphine and original diketone ligands, depending on the
basicity of the latter, dictates the initial relative affinity between
metal and phosphine, while beyond phosphine ligand stabilization,
the diketone ligand dynamics at elevated temperatures were accompanied
by the respective Co–P affinity change. Across different metals,
the deviation from the “natural order” in metal-phosphine
affinity can also be qualitatively understood from the ligand competition
concept, where the same ligand effects on the field stabilization
schemes are expected as the distinctions caused by ligand fluorination
were consistent throughout d7–d10 metal
cations. The knowledge gained from this work could benefit future
design of thermal latent catalysts and shed light on the capability
of Lewis base reactivity control through adjusting transition-metal
coordination spheres.