Mixed Monolayers on Nanoparticles

Mixed Ligand Monolayer Self-Assembly on Nanoparticle Surfaces

Figure 1: Comparison of Ligand Fragment Distributions from MALDI-MS and Atomistic Simulations. The two measures show an excellent agreement and clear evidence of phase separation for a mixture of dodecanethiol (DDT) and mercaptoethanol (ME) ligands on the surface of ultrasmall silver NPs (Merz, ACS Nano, 2018).
 

Nanoparticles (NPs) protected with self-assembled ligand monolayers (SAMs) have applications ranging from photonics and catalysis to drug delivery and biosensing. SAMs with a mixture of ligands may self-organize into random, striped, patchy, and Janus-like surface morphologies. These patterns impact their functional properties but are difficult to predict. Even the characterization of SAM morphology on NPs <10 nm has proven difficult. In an NSF-funded collaboration with Prof. David Green at UVA, who employs matrix-assisted laser desorption/ionization mass spectrometry (MALDI-MS) to indirectly characterize NP-SAMs, we adapted an advanced Monte Carlo (MC) technique developed by Siepmann and Frankel to equilibrate atomistic models of these sluggish mixed ligand monolayers – monolayers which take days to equilibrate in the lab.
Key Findings: The Green group synthesized silver NPs with SAMs consisting of dodecanethiol (DDT) and mercaptoethanol (ME) at varying ratios and analyzed them using MALDI-MS. The results suggested that DDT and ME phase separated on the NP surface. We then employed atomistic MC simulations to generate equilibrated SAM morphologies. To quantitatively compare the computational and experimental results, we developed a method to calculate expected MALDI-MS spectra from the resulting atomistic models. The simulations agreed quantitatively with the experiments, both indicating that the DDT/ME ligands undergo phase separation in the SAMs, resulting in large, patchy Janus-like domains (Merz, ACS Nano, 2018). We then applied the same MC approach to successfully predict MALDI-MS results for three new ligand combinations, which were chosen to test the ability to separately tune SAM morphology and chemical functionality. The resulting morphologies ranged from randomly-mixed to Janus-like, demonstrating that chain-length modifications are an effective way to tune SAM morphology without needing to alter chemical functionality. Our results also provided atomistic detail of the SAM structures, such as patch sizes and co-crystallization patterns (Merz, Soft Matter, 2019).

Figure 2: Simulated Structures and MALDI-MS Distributions from Simulation and Experiment. Images of the final simulated configurations are shown here for each of the three ligand combinations: DDT/MHA (left), HT/MUDA (center), and DDT[D25]/MUDA (right). A comparison of the computationally-predicted and experimentally-observed MALDI-MS fragment distributions are shown below the simulated structures. Error bars (in orange) show the standard deviations. For the simulations, the results from six randomly initialized simulations for each starting configuration (Janus, Striped, Random) were included (18 simulations per data point). For the experiments, all samples that were within 0.50 ± 0.05 χalkane were included (Merz, Soft Matter, 2019).