Tailoring Molecular Specificity Toward a Crystal Facet: a Lesson From

Jan 15, 2013 - Li Huey Tan , Yuan Yue , Nitya Sai Reddy Satyavolu , Arzeena Sultana ... Chemistry of Materials 2014 26 (19), 5725-5734 .... Wiley Inte...
2 downloads 0 Views 4MB Size
Letter pubs.acs.org/NanoLett

Tailoring Molecular Specificity Toward a Crystal Facet: a Lesson From Biorecognition Toward Pt{111} Lingyan Ruan,† Hadi Ramezani-Dakhel,‡ Chin-Yi Chiu,† Enbo Zhu,† Yujing Li,†,∥ Hendrik Heinz,*,‡ and Yu Huang*,†,§ †

Department of Materials Science and Engineering, University of California, Los Angeles, California 90095, United States Department of Polymer Engineering, University of Akron, Akron, Ohio 44325, United States § California NanoSystems Institute, University of California, Los Angeles, California 90095, United States ‡

S Supporting Information *

ABSTRACT: Surfactants with preferential adsorption to certain crystal facets have been widely employed to manipulate morphologies of colloidal nanocrystals, while mechanisms regarding the origin of facet selectivity remain an enigma. Similar questions exist in biomimetic syntheses concerning biomolecular recognition to materials and crystal surfaces. Here we present mechanistic studies on the molecular origin of the recognition toward platinum {111} facet. By manipulating the conformations and chemical compositions of a platinum {111} facet specific peptide, phenylalanine is identified as the dominant motif to differentiate {111} from other facets. The discovered recognition motif is extended to convert nonspecific peptides into {111} specific peptides. Further extension of this mechanism allows the rational design of small organic molecules that demonstrate preferential adsorption to the {111} facets of both platinum and rhodium nanocrystals. This work represents an advance in understanding the organic−inorganic interfacial interactions in colloidal systems and paves the way to rational and predictable nanostructure modulations for many applications. KEYWORDS: Inorganic binding peptides, surfactants, metal nanocrystal, facet specificity, molecular dynamic simulation, organic−inorganic interface

T

Recently, more attention has been directed to the inspection of biomolecular behavior at various material surfaces and interfaces including specific surface binding and self-assembly processes.20,21 Despite extensive studies and various evidence, one of the key questions regarding how biomolecules recognize specific materials and crystal facets still exists in biomimetics.17−19,22 The development of the state-of-the-art interface detection techniques greatly aid the studies on organic− inorganic interfaces,23 however, great challenges remain including the complexity and the inaccessibility of the interface, for example, the multifunctionality of the molecules and the distinct atomic coordination on different material surface or crystal facets. Clarifications are often possible using simulation techniques in which interfaces can be well-defined, although reliable insight also strongly depends on the availability of robust binding couples as well as suitable interaction potentials for organic−inorganic interfaces.24 Here, we present detailed experimental and computational studies on understanding the biomolecular recognition

he organic−inorganic interface in various systems has been a subject of intense research interest and can impact a wide range of applications from syntheses, catalysis, sensing to energy devices. In colloidal nanocrystal (NC) syntheses, one important key to achieve desired shape control is the use of organic surfactants that can judiciously stabilize NC surfaces.1−3 Under this guidance, a number of surfactants have been explored to manipulate NC shapes and further to construct sophisticated architectures.4−8 However, mechanisms regarding the selective adsorption of surfactants remain unclear, which limit the choice of suitable surfactants.1−3 During the past decade, molecular evolution approaches borrowed from biochemistry render useful tools to identify artificial biomolecules with exquisite recognition properties to a variety of material surfaces therefore enriching the pool of surfactants for material fabrication and functionilization.9−16 Nevertheless, most efforts have been focused on identifications and utilizations of biomolecules, while the fundamental investigation of interactions between biomolecules and materials represents a formidable task that has been less touched upon due to the complexity of the system.17−19 © 2013 American Chemical Society

Received: January 3, 2013 Published: January 15, 2013 840

dx.doi.org/10.1021/nl400022g | Nano Lett. 2013, 13, 840−846

Nano Letters

Letter

tetrahedral yield (tens of percent) are considered to gauge molecular selectivity. It has been suggested that interactions between peptidic molecules and material surfaces involve both peptide conformations and chemical functional groups on residues.10,11,17,25 To deconvolute contributions from these factors to the platinum {111} recognition, we designed a set of S7 variants that target one specific factor at a time (Figure 1b, Supporting Information Table S1) (see Supporting Information for peptide design). As shown in Figure 1b, S7-1 (randomly shuffling the sequence of S7) and S7-2 (replacing the two structural residues P in S7 with small flexible G, G: Glycine) are to test the effect of peptide conformation. Fragment SSF is used to examine the contribution from hydroxyl groups in the two S residues without amide groups in the Q and N residues, and fragment FPQPN is to examine the contribution from amide groups in the Q and N residues without hydroxyl groups in the two S residues. Both amide and hydroxyl groups have been suggested to contribute significantly to the binding on metal surfaces, and especially on {111} surface.10,11,17 Finally, S7-G (replacing the F in S7 with G) and fragment PQPN are designed to investigate the effect of the phenyl group on the F residue. Additionally, S7-Y and SSY (replacing the F with Y in S7 and SSF, Y: Tyrosine) are to compare the phenyl ring in F with the phenol ring in Y, which will be discussed later. These S7 variants were applied to platinum NC synthesis, and same reaction conditions (see Supporting Information for Pt NC synthesis) were applied for all variants to exclude possible influences from different reaction kinetics. The structures of the NCs were characterized by TEM (Supporting Information Figure S1) and yields of tetrahedra were counted based on HRTEM structural analysis of 100 NCs for each reaction (see Supporting Information Figures S2−S4 for two examples for SSF and PQPN). The majority of NCs (over 80%) are single crystals, identified as either “tetrahedra” (or tetrahedra with corners slightly truncated), or “cuboctahedra”, which is consistent with the monomer attachment growth model. Only a small population (less than 20%) shows branched or rodlike shapes with twin defects, counted as “others”. Since twin defects usually result from altered crystal growth mechanisms,26−30 we studied the facet recognition mechanism based on the single crystal shapes (tetrahedra or slightly truncated tetrahedra and cuboctahedra), and others were excluded from the comparison. We found the set of S7 variants falls into two subgroups that give significantly different tetrahedra yields: (1) peptides with the amino acid F (S7, S7-1, S7-2, SSF and FPQPN), and (2) peptides without amino acid F (S7-G, PQPN, S7-Y and SSY) (Figure 1c). All peptides with F give considerably high tetrahedra yields as the original S7, indicating their retained Pt{111} specificity, while peptides without F produce mainly cuboctahedra, suggesting the loss of Pt{111} specificity. Notably, the short fragment SSF results in particularly high tetrahedra yield, further suggesting that the specific residue (F) overrides the overall structural conformation in contribution to Pt{111} specificity. The difference observed in the tetrahedra yield indeed originates from the binding specificity the peptides possess, as confirmed by sets of control experiments (Supporting Information Figures S1, S5−S7) (see Supporting Information for Pt NC synthesis). First, the reactions without the presence of any peptides produced large NCs without preferential shapes, suggesting the clean synthetic backgrounds in the shape

mechanism toward platinum {111} through manipulating the conformations and compositions of a previously discovered facet-specific peptide S7 (sequence, SSFPQPN; S, Serine; F, Phenylalanine; P, Proline; Q, Glutamine; N, Asparagine).14 We employed an indirect strategy using nanocrystal shape as an indicator to infer the specific recognition properties of peptides to the chosen facet. In the classical model of NC growth through monomer attachment, it is generally established that the selective adhesion of surfactants to a chosen facets lowers its surface energy and hence its growth rate, leading to a resultant shape enclosed by this specific facet.1−3 The S7− Pt{111} facet represents a definitive binding pair we have demonstrated in our previous study that S7 could shape Pt NC exclusively into tetrahedra through its selective adsorption to Pt{111} (Figure 1a).14 Without selective stabilization,

Figure 1. Nanocrystal (NC) shape as an indicator of surfactant binding specificity. (a) Schematic illustration of using platinum (Pt) tetrahedral-shaped NC yield as a measure for the binding specificity of peptides to Pt{111} surfaces. (b) Sequences of S7 and its variants. The variants are designed to target one specific factor at a time that might potentially contribute to the binding specificity of peptidic molecules, including both peptide conformation and chemical functional groups on residues (indicated in each boxes). (c) Yields of Pt tetrahedral NCs by S7 and its variants under the same reaction conditions. TH represents tetrahedra and TTH represents truncated tetrahedra. The yield is averaged over three repeated syntheses for each variant and the error bar is the standard deviation.

cuboctahedra are the thermodynamically favorable shape for face-centered cubic (fcc) metals.1−3 Under this guideline, we can judge the facet specificity of a peptide surfactant by the yield of the shaped NCs it produced. Starting from the original S7 sequence, we designed a set of S7 variant peptides (Figure 1b) and compared with the original S7 their effectiveness as surfactants in directing platinum tetrahedra formation (Figure 1c). When the variant peptide surfactant produced similar yield of tetrahedral shapes as S7, we infer the molecular specificity toward platinum {111} of the modified sequence is retained, otherwise we declare the loss of the specificity as a result of the sequence modification. We note that only significant changes in 841

dx.doi.org/10.1021/nl400022g | Nano Lett. 2013, 13, 840−846

Nano Letters

Letter

Figure 2. The origin of Pt{111} specificity. (a,b) Correlation of tetrahedra yields with computed peptide adsorption energies to Pt{111} and {100} surfaces in solution (Ea{111} and Ea{100}), respectively. TH represents tetrahedra and uncertainties are