(427h) MALDI-MS Quantification of Ligand Phase-Separation on Mixed Thiol-Functionalized Silver Nanoparticles

Silver nanoparticles with mixed-thiol ligand self-assembled monolayers were functionalized with dodecanethiol and one other ligand from a homologous series of thiols (dodecanethiol[D25], butanethiol, pentanethiol, hexanethiol, heptanethiol, octanethiol, nonanethiol, decanethiol, undecanethiol, or 2-mercaptoethanol). These ligands were chosen on the basis of our hypothesis that increasing chemical and physical mismatch (and thus stronger entropic and enthalpic drivers) between the two ligands should lead to increased ligand phase separation on the nanoparticle surface. The baseline case for minimum phase separation was hypothesized to be dodecanethiol/dodecanethiol[D25] due to the chemical and physical similarities between the two ligands (only isotopic differences). At the opposite extreme, dodecanethiol/2-mercaptoethanol was expected to exhibit extreme phase separation in the form of hemispherical ligand phase separation into a Janus particle morphology due to the large chemical and physical differences between the ligands. All other pairings of dodecanethiol with remaining alkanethiol ligands (4-11 carbons) exhibit varying degrees of physical (but not chemical) mismatch, leading us to expect phase separation intermediate between the well-mixed and Janus morphologies.

            Matrix-Assisted Laser Desorption Ionization Mass Spectroscopy (MALDI) was used to quantify ligand phase separation via the sum of squares error, r, calculated between the experimentally measured ion distributions and the ion distributions predicted by a binomial distribution, which models a well-mixed system. For dodecanethiol/dodecanethiol[D25], which is expected to be a well-mixed ligand monolayer, the experimental and binomial results match closely with sum of squares error of r=0.002. In the opposite case, the pairing of dodecanethiol/2-mercaptoethanol results in r=0.538, indicating bulk ligand phase separation and the formation of a Janus particle. In nanoparticles functionalized with ligand pairs in between these two extremes, r was observed to increase with increasing ligand length difference, as hypothesized. To infer further information about the onset of phase separation, t-testing was applied to compare the set of r-values for the well-mixed dodecanethiol/dodecanethiol[D25] case with r-values for other ligand pairs. Based on this approach, measurable phase separation does not occur until a ligand length difference of four carbons, indicating the transition from a well-mixed monolayer to ligand microphase separation. These approaches offer robust methods of characterization for mixed ligand functionalized metallic nanoparticles with phase separated ligand domains.