Peptide-Programmable Nanoparticle Superstructures with Tailored Electrocatalytic Activity

Abstract

Biomaterials derived via programmable supramolecular protein assembly provide a viable means of constructing precisely defined structures. Here, we present programmed superstructures of AuPt nanoparticles (NPs) on carbon nanotubes (CNTs) that exhibit distinct electrocatalytic activities with respect to the nanoparticle positions via rationally modulated peptide-mediated assembly. De novo designed peptides assemble into six-helix bundles along the CNT axis to form a suprahelical structure. Surface cysteine residues of the peptides create AuPt-specific nucleation site, which allow for precise positioning of NPs onto helical geometries, as confirmed by 3-D reconstruction using electron tomography. The electrocatalytic model system, i.e., AuPt for oxygen reduction, yields electrochemical response signals that reflect the controlled arrangement of NPs in the intended assemblies. Our design approach can be expanded to versatile fields to build sophisticated functional assemblies.

Keywords: artificialy designed peptide; electrocatalytic oxygen reduction; electron tomography; nanoparticle superstructure; peptide-based catalyst; peptide-based superstructure 3-D reconstruction; supramolecular protein self-assembly.

Figure 1.

Peptide-mediated synthesis of NP superstructures on a SWNT. (a) Synthetic scheme. (b) Genetic modification of outward-facing sites in peptide sequences (the red-to-green coloring indicates the N-to-C termini) with Cys residues for specific nucleating NPs. One of the modified peptides, with E8 and Q26 changed to Cys, is shown (bottom box: peptide sequence). (c) Computational model depicting the assembly geometry of (8,26) NPs. The midpoint between two adjacent Cys residues indicates the NP position (inset figures).

Figure 2.

Two types of helical NP superstructures defining interparticle distances formed by antiparallel six-helix bundles. (a) Antiparallel dimeric unit cells in which two distinct Cys–Cys distances (d_intra, on a monomer; d_inter, between monomers) are defined to determine the arrangement of NPs in helical geometries, and top-down and side views of computational models of 3 × 2-fold and three-fold helical arrays. (b) Calculated distances between a centrally situated NP and its three first-nearest neighbors (interparticle distance (ID) type i, ID type ii, and ID type iii, indicated by the arrows in part a) for all the submodels. Striped and solid bars indicate 3 × 2- and three-fold helical array models, respectively. (c) Four different computational models predicting the geometry of an NP assembly aligned on the peptide/SWNT superhelix, highlighting the interparticle distances.

Figure 3.

Characterization of AuPt/HC/SWNT superstructures: (a) TEM and HRTEM images and (b) STEM-EDS elemental mapping images of the (8,26)-modeled sample. (c) 3-D reconstructed model of rendered tomographic volume through electron tomography. (d) Comparison of helical-bundle parametric model based on the theoretical geometries with 3-D reconstruction model. The correlation is shown between the AuPt NPs (yellow dots) and theoretical helical model (green line) both in plane and (e) vertical views. (f) Statistical analysis of the particle size distribution in TEM images of the (8,26)-modeled AuPt/HC/SWNT samples with average particle sizes of 1.6, 2.1, 2.4, and 2.9 nm (for particle counts >200). (g) Plots of calculated and observed interparticle distances as a function of particle size. The calculated distance data were obtained from simulated models. The observed distances were obtained by subtracting the average particle size from the average center-to-center distance between two particles in TEM images. The center-to-center distances longer than theoretically possible were excluded from the calculation, reducing errors.

Figure 4.

(a) Specific activities of (8,26), (8,22), (8,19), and (12,19) for these samples at 0.9 V (with respect to a reversible hydrogen electrode, RHE), averaged for four different electrodes. The particle size and interparticle distance were obtained from TEM imaging and model-simulated results, respectively. The areas marked in pale yellow indicate the expected merging of NPs. (b) Loss of specific activities and calculated portion of overlapping surface between AuPt NPs for (8,19), (8,22), and (8,26). The losses were determined using the relationship [(j_k,f – j_k,m)/j_k,m] × 100, where j_k,m and j_k,f are the specific activities of each model before and after particle merging, respectively. The portion of the overlapped surface area was calculated using the formula for the surface area of a sphere intersected by a plane (see Supporting Information Figure S8). (c) ORR polarization curves of the (8, 26)-catalysts with different particle sizes and a commercial Pt/C catalyst (Premetek, 20% Pt) on a rotating disk electrode in an O₂-saturated 0.1 M KOH solution at a sweep rate of 10 mV s^–1 and a rotation rate of 1600 rpm.