Micelle-directed chiral seeded growth on anisotropic gold nanocrystals

See allHide authors and affiliations

Science  26 Jun 2020:
Vol. 368, Issue 6498, pp. 1472-1477
DOI: 10.1126/science.aba0980

Groovy chiral gold particles

Although plasmonic optical activity can arise from chiral assemblies of gold and silver nanoparticles, there are few examples of gold nanoparticles with intrinsic chirality and high optical activity. González-Rubio et al. show that morphological chirality can be induced during the seeded growth of gold nanoparticles, particularly for highly anisotropic nanorods. Chiral additives as cosurfactants formed helical micelles that directed the seeded growth to create grooves that maintained a chiral morphology. The resulting particles displayed high-intensity circular dichroism with anisotropy factors near 0.2 at near-infrared wavelengths.

Science, this issue p. 1472


Surfactant-assisted seeded growth of metal nanoparticles (NPs) can be engineered to produce anisotropic gold nanocrystals with high chiroptical activity through the templating effect of chiral micelles formed in the presence of dissymmetric cosurfactants. Mixed micelles adsorb on gold nanorods, forming quasihelical patterns that direct seeded growth into NPs with pronounced morphological and optical handedness. Sharp chiral wrinkles lead to chiral plasmon modes with high dissymmetry factors (~0.20). Through variation of the dimensions of chiral wrinkles, the chiroptical properties can be tuned within the visible and near-infrared electromagnetic spectrum. The micelle-directed mechanism allows extension to other systems, such as the seeded growth of chiral platinum shells on gold nanorods. This approach provides a reproducible, simple, and scalable method toward the fabrication of NPs with high chiral optical activity.

Chirality is a key signature of nature that can be found across length scales, from subatomic particles, through molecules and biological systems, to galaxies (13). Imparting handedness to selected materials may provide important advantages in terms of their interaction with living organisms, allowing the development of enantioselective catalysts or devices with spin selectivity in electron transport (46). For example, the demonstration of chiral plasmon modes in noble metal nanocrystals (NCs) has drawn interest in the field of metamaterials and the design of enantioselective sensing probes (79). The fabrication of plasmonic nanostructures has thus become an active field of research (1016). However, growing crystalline noble metals with dissymmetric morphology is challenging (1012). The formation of chiral plasmonic nanomaterials is usually achieved by assembling achiral plasmonic gold and silver nanoparticles (NPs) with molecular templates such as DNA, proteins, or polymeric fibers (1315). Although lithographic approaches have been successfully applied to fabricate chiral gold NPs, their scalability remains limited (9, 16).

Advances in the colloidal synthesis of noble metal NPs achieved during the past decades suggest realistic prospects for the production of plasmonic NCs with diverse morphologies, including chiral ones (1719). The presence of chiral amino acids during growth can guide the formation of NCs with distinct handedness. The most prominent example of this is the synthesis of helicoidal nanostructures assisted by cysteine and glutathione. The enantioselective interaction of amino acids with chiral geometrical elements naturally appearing at particular NC facets, has been claimed to induce the observed shape evolution into twisted geometries with high dissymmetry factors (11, 20).

In addition to chiral additives, the synthesis of colloidal chiral NPs requires surface ligands that prevent undesired aggregation (11, 21). However, most of these capping agents also play a role in the growth process and the resulting NC morphology (17, 22). Among the wide variety of ligands used for the synthesis of noble metal NCs, quaternary alkylammonium halide (CTAX, where X = Cl or Br) surfactants are some of the most extensively investigated. Through the adsorption of CTAX micellar aggregates on certain crystallographic facets, a wide variety of NP morphologies have been obtained, including nanorods, nanotriangles, and Platonic geometries (1719, 2224).

Recently, the addition of cosurfactants such as aromatic molecules, fatty acids, or long-chain alkyl alcohols has been shown to further improve the quality of colloidal gold NPs, and of nanorods in particular (2527). The ability of cosurfactants to intercalate within CTAX surfactant aggregates has been proposed to increase the rigidity of the micelles, induce phase transitions, or both. The impact of cosurfactants on the aggregation behavior of surfactant molecules can also be extended to the formation of chiral micelles. For example, 1,1′-bi(2-naphthol) [BINOL], a cosurfactant with axial chirality (i.e., atropisomerism), can induce the formation of chiroptically active, giant, worm-like micelles (28).

Considering the role of surfactants on the synthesis of gold NCs and the possibility of producing chiral micelles in the presence of BINOL-like cosurfactants, we envisioned the surfactant-assisted seeded growth of chiral gold NCs. We hypothesized that elongated rod-like particles would have the ideal shape to guide the adsorption of twisted, worm-like micelles on the gold surface into helix-like chiral structures. To investigate both hypotheses, we first performed a computational analysis on an ideal system comprising cetyltrimethyl ammonium chloride (CTAC) and BINOL. Molecular dynamics (MD) simulations revealed that BINOL molecules induced the assembly of surfactants into chiral, worm-like aggregates. Such elongated micelles tend to coil around gold nanorods, which can thus be considered as templates for the seeded growth of anisotropic NPs with chiral features.

Although experimental evidence of the proposed idea was obtained using BINOL as a cosurfactant, the results of chiral growth were largely improved by using its derivative 1,1′-binaphthyl-2,2′-diamine (BINAMINE). The growth of plasmonic gold nanorods patterned with a complex chiral surface was indeed confirmed by high-angle annular dark-field scanning transmission electron microscopy (HAADF-STEM) tomography. By tuning the dimensions of the final nanostructures, high anisotropy factors (g-factor ~0.20) were achieved within a wide wavelength range (from 500 to beyond 1350 nm). Theoretical modeling of the optical properties highlights the importance of growing well-defined chiral wrinkles to obtain intense circular dichroism (CD) responses.

MD simulations confirmed that in the presence of BINOL as a cosurfactant, CTAC can form giant cylindrical micelles that span across the simulation boxes (i.e., up to 60 nm in length; Fig. 1A and fig. S1). Regarding their chiral nature, we observed that unit vectors between the local micellar center and the principal micellar axis were distributed helically, with the handedness being dictated by the choice of cosurfactant stereoisomer. This effect can be shown by measuring the average angle formed between pairs of vectors (fig. S2). An achiral helix does not have any tendency to coil into a particular direction, so the angle formed between such vectors averages out to zero.

Fig. 1 MD simulations of BINOL-CTAC micelles and their assembly on gold nanorods.

(A) MD Simulations for (R)-BINOL-CTAC in solution. The aggregation of CTA+ (blue) in the presence of (R)-BINOL (red) results in chiral, worm-like aggregates spanning the whole simulation box 60 nm in length. (B) The shape of the micellar coil is described by a set of vectors along the principal micellar axis. The average angle between vectors is plotted as a function of their vertical distance for 60-nm-long rod-like micelles. The presence of (R)- or (S)-cosurfactant enantiomers in the micelles leads to coiling in opposite directions, as revealed by the sign of the average angle. (C and D) MD simulations for (R)-BINOL-CTAC adsorbed onto a gold nanorod in axial and lateral views, respectively.

In our simulations, the average angles not only were finite but also appeared sinusoidally correlated and displayed a well-defined handedness (Fig. 1B). Because of the marked atropisomerism of BINOL, we can assume that it could induce helical chirality onto the surfactant aggregates, likely by a “sergeants and soldiers” mechanism in which a chiral molecule directs many achiral molecules or atoms (29, 30). In this case, the adsorbed chiral micelles would assist chirality transfer during further deposition of metal atoms onto the nanorod surface.

We tested this effect by simulating a gold nanorod immersed in an aqueous mixed solution of CTAC and BINOL. Visual inspection of the simulation results revealed that CTAC and BINOL adsorbed into worm-like micelles coiled around the gold nanorod surface (Fig. 1, C and D). Although no preferred orientation of the assembly was observed by the use of different BINOL enantiomers in this case, in the absence of cosurfactant, the coil-like morphology was largely lost. Instead, small aggregates and a high fraction of individual surfactant molecules adsorbed on the gold surface (fig. S3). The simulations indicated that chiral worm-like micelles likely formed first and were subsequently adsorbed onto the gold nanorod. Thus, a complex cooperative chiral transfer mechanism would convey the atropisomerism from BINOL into the giant micelles and ultimately to gold nanorods.

On the basis of the simulation results, we implemented experimentally the growth of chiral nanorods. (R)-BINOL was solubilized by surfactant micelles in aqueous solution at an optimum molar ratio of 40 and used to template the growth of preformed gold nanorods (130 ± 10 nm long, 29 ± 3 nm thick; for details, see Fig. 2A and the supplementary materials; note that (R)-BINOL is insoluble in water in the absence of surfactant). Preliminary characterization by CD spectroscopy revealed the appearance of moderate, chiral-localized surface plasmon resonance bands (g-factor ~0.002; figs. S4 and S5). HAADF-STEM analysis of the NPs grown in the (R)-BINOL and surfactant mixture revealed a complex morphology characterized by the presence of wrinkles on the nanorod surface (Fig. 2B and fig. S4). Previous theoretical and experimental studies demonstrated that intense CD responses are exhibited by NPs with distinct chiral distortions, stellated-like morphologies, or both (9, 11, 31).

Fig. 2 Growth of chiral gold nanorods in micellar systems and effects of size and shape.

(A) TEM image of gold nanorods 130 nm in length and 29 in width used as seeds. (B) High-magnification HAADF-STEM image of a chiral gold nanorod grown in (R)-BINOL displaying a complex surface containing wrinkles. (C) HAADF-STEM image at low magnification of gold nanorods obtained in the presence of (R)-BINAMINE displaying a complex surface containing sharp wrinkles. Scale bars, 200 nm [(A) and (C)] and 100 nm (B). (D to I) Gold nanorods of 165 × 73 nm [(D) and (E)], 210 × 112 nm [(F) and (G)], and 270 × 175 nm [(H) and (I)] grown in (R)-BINAMINE-surfactant mixtures were analyzed by HAADF-STEM. Scale bars, 20 nm [(D) and (E)], 50 nm [(F) to (I)], and 10 nm [(J) and (K)]. Tomography reconstructions [(D), (F), and (H)] reveal their surface topography, and selected orthoslices show the growth of wrinkles from the gold nanorod seeds and the internal structure of the wrinkle network [(E), (G) and (I)]. (J and K) Tomography reconstruction and selected orthoslice of an NP obtained by overgrowth of a 30-nm gold nanosphere in an (R)-BINAMINE-surfactant mixture. Scale bar, 100 nm. Corresponding animated reconstructions are provided in movies S1 (D), S3 (E), S7 (F), S9 (G), S18 (H), S20 (I), S43 (J), and S45 (K).

Encouraged by the appearance of plasmonic bands in the CD spectra, our efforts focused on improving chiral seeded growth. We hypothesized that the use of stronger stabilizing agents would help to direct the growth of more structured chiral wrinkles (32). We investigated (R)-BINAMINE, an analog of (R)-BINOL in which the hydroxyl functionalities are replaced by amine moieties. Although the interaction of amines with gold surfaces is moderately strong (Au–N bond energy ~6 kcal/mol), the presence of two amines in a chelating configuration should ensure a greater affinity (33, 34). MD simulations of the BINAMINE-CTAC system confirmed the formation of chiral, worm-like aggregates, similar to those for BINOL-CTAC, with a helical distribution of unit vectors (fig. S6). HAADF-STEM analysis of the particles obtained by seeded growth in (R)-BINAMINE-CTAC revealed the presence of gold nanorods displaying a highly complex surface morphology resembling an intricate network of wrinkles (Fig. 2C and figs. S7 to S10). Scanning electron microscopy (SEM) imaging of the same NPs also revealed a complex surface structure (fig. S11).

Notwithstanding, proper understanding of such complex structures demands a more detailed characterization than that provided by either HAADF-STEM or SEM, which only retrieve two-dimensional (2D) and surface information, respectively. We used HAADF-STEM tomography, which has been shown to be a powerful technique to investigate the 3D morphology of complex NCs. We prepared similar particles, but with increasing dimensions, using 130- × 29-nm gold nanorods as seeds and characterized them with HAADF-STEM (see the supplementary materials for synthetic details). Electron tomography reconstructions of nanorods with the smallest degree of growth, i.e., 165 nm long and 73 nm thick, revealed the presence of sharp wrinkles oriented in a seemingly radial direction and displaying tilt angles between 0° and 45° with respect to the short axis (Fig. 2D). Detailed analysis of the internal structure suggested that the wrinkles grew from the surface of the gold nanorod seed in a quasiradial direction up to ~15 to 20 nm in height and with a constant width of 3 to 4 nm. The average intergroove separation was measured to be 2 to 3 nm (Fig. 2E), which is near the 3.75-nm width of worm-like micelles obtained from MD simulations. This finding would suggest a micelle-directed chiral growth of the gold nanorods, as we initially hypothesized.

The width of the wrinkles was compatible with the diffusion of metal ion–loaded CTAC micelles from solution (35), leading to reduction at available gold sites between (R)-BINAMINE-surfactant micelles on the nanorod surface. By varying the concentration of nanorod seeds in the growth mixture, chiral nanorods were obtained with increased dimensions of 210 nm in length and 112 nm in thickness, and their surface was more entangled (Fig. 2F). Nonetheless, the presence of tilted wrinkles was still visible and angles between 0° and 45° were again observed [between 0° and –45° in the case of (S)-BINAMINE; fig. S9]. Analysis of the internal structure revealed wrinkles of ~35 to 45 nm in height and a similar width to those in the smaller rods (Fig. 2G). Finally, even larger chiral nanorods (270 nm long and 175 nm thick) were investigated. Although the surface appeared more undefined in this case (Fig. 2H), analysis of the internal structure showed the characteristic features described for the previous samples (Fig. 2I). For completeness, 3D animations of all discussed systems are provided as supplementary materials (movies S1 to S48).

Diffraction of ideal helical structures results in X-shaped patterns (36, 37). Thus, we applied fast Fourier transformations (FFTs) to our 3D reconstructions. We exemplify this analysis with chiral nanorods of 210 × 112 nm (Fig. 3, A and B), from which the 3D FFT indeed shows an X-shaped pattern (Fig. 3C); further details, as well as comparison with an idealized model (fig. S12), with a smooth nanorod (fig. S13), and with a (R)-BINOL-grown nanorod (fig. S13), are provided in the supplementary materials. Next, the spots in reciprocal space can be linked to the corresponding features in real space by manually segmenting the 3D FFT to minimize noise. By using the segmented 3D FFT as a mask (see details in the supplementary materials), an inverse FFT (38) was computed and overlayed (pink fringes) with the original reconstruction, as exemplified in Fig. 3D, so that the helical features are visually highlighted. Although the use of a 3D FFT does not provide a quantitative value such as, e.g., the Hausdorff chirality measure (39), it provides a qualitative and visual way to estimate the presence of chiral features. The inverse FFT enables direct location of specific periodicities at a local level.

Fig. 3 Qualitative analysis of chiral features.

(A) Tomography reconstruction of a gold nanorod grown in an (R)-BINAMINE-surfactant mixture (see also movie S7). (B) Low-resolution HAADF-STEM image of chiral gold nanorods obtained in the presence of (R)-BINAMINE. Scale bar, 100 nm. (C) 3D FFT from which the inverse FFT was computed after segmentation (see also movie S10). (D) Inverse FFT (see also movie S12) showing the areas of the particle with chiral features overlapped with the projection of the tomography, indicating a right-handed angle. (E) Tomography reconstruction in HAADF-STEM (left) and EDX (right, Pt displayed in green) modes of an Au@Pt nanorod grown in an (R)-BINOL-surfactant mixture (see also movies S24 and S25). (F) HAADF-STEM image at low magnification of Au@Pt nanorods. Scale bar, 100 nm. (G) 3D FFT from which the inverse FFT was computed after segmentation (see also movies S28 and S29). (H) Inverse FFT (see also movie S30) showing the areas of the particle with chiral features overlapped with the projection of the tomography, indicating a right-handed lower angle.

Fringes with a right-handed angle can be seen in Fig. 3D, and the wrinkles growing on the sides of the gold nanorod seeds seemed to display a better-defined chiral arrangement than those located at the hemispherical tips. The observation of a curvature-dependent growth was confirmed by a control experiment, in which seeded growth was performed on 30-nm gold nanospheres (figs. S14 and S15). Although some chiral features could still be observed, the spheres yielded a more random wrinkle organization, as revealed by the corresponding HAADF-STEM tomography reconstruction (Fig. 2, J and K), and further supported by their 3D FFT and corresponding inverse FFT (fig. S14). Accordingly, more intense plasmonic CD bands were recorded when gold nanorods were used as seeds (g-factor ~0.2; Fig. 4A) compared with spheres (g-factor ~0.003; fig. S15).

Fig. 4 Effect of nanorod dimensions on chiral plasmonic activity.

(A) Spectral evolution of the anisotropy factor for chiral Au nanorods with increasing particle size: 165 × 73 nm (red), 210 × 112 nm (blue), and 270 × 175 nm (magenta). (B) Models of chiral gold nanorods used to simulate the chiral plasmonic properties (from left to right: 165 × 73, 210 × 112, and 270 × 175 nm). (C) Calculated anisotropy factor spectra [from left to right in (C): red, blue, and magenta].

Additional evidence behind the general validity of this chiral growth method and the proposed mechanism was provided by experiments in which we varied the nature of the metal deposited during seeded growth. We thus implemented the seeded growth of Pt on (R)-BINOL-CTAC–covered gold nanorods using identical dimensions and similar growth conditions. Representative results in Fig. 3, E to H, show that seeded growth resulted in very regular wrinkled platinum coating. The smaller dimensions of the wrinkles, as well as the different electron configuration between platinum and gold (see a 3D EDX reconstruction in Fig. 3E, right), allow a better distinction of the obtained pattern. Although the tilting angle is obviously lower than that measured for gold, a helical character is demonstrated by the 3D-FFT analysis (Fig. 3, G and H). Unfortunately, the lossy character of platinum does not allow recording of meaningful plasmonic optical activity for these samples.

We next demonstrate that the optical activity (CD bands) can be modulated by varying the dimensions of chiral nanorods (by varying the amount of seeds in the growth solution). In all cases, an intense negative Cotton effect was recorded for the (R)-enantiomer (29). As the size of the nanorods was increased from 165 × 73 up to 270 × 175 nm, the positive band redshifted from 700 up to 1300 nm, and the maximum of the negative band shifted from 1100 up to beyond 1350 nm (fig. S16). Analysis of the anisotropy factor showed values ranging from 0.1 to 0.2 (Fig. 4A), which are among the highest values reported for colloidal plasmonic NPs in the visible and near-infrared (NIR) ranges (10, 11). By extending the analysis further into the NIR, up to the limit of our CD spectrometer (1600 nm), we recorded even higher dissymmetry factors of 0.25 to 0.28 (fig. S17). However, the noise in the 1400- to 1500-nm range precluded us from obtaining more compelling evidence to confirm the precise values.

When gold nanorods of different dimensions and degrees of anisotropy were used to seed the chiral growth (figs. S18 and S19), similar trends in optical activity were observed. However, the anisotropy factor was influenced by the nanorod dimensions. In general, slightly lower g-factor values were obtained when shorter and thinner gold nanorods were used as seeds (100 × 12 nm; i.e., higher aspect ratio), whereas a decrease was observed for seeds with similar length to the optimal one but thicker in diameter (135 × 52 nm; i.e., lower aspect ratio).

The relation between the chiral features observed by electron microscopy and the chiroptical activity measured by CD spectroscopy was further investigated by electromagnetic modeling. We used an accurate full-wave solver based on Maxwell’s surface integral equations and the method of moments to model the plasmonic properties of chiral gold nanorods (see the supplementary materials for details) (4042). Three-dimensional computer-aided design (CAD) models were constructed on the basis of the information provided by the experimental electron tomography reconstructions (Fig. 4B). Although suitable reproduction of the observed morphologies appeared challenging, given the intricate network of wrinkles covering the chiral gold nanorod surface, we created models for the three different dimensions characterized by electron microscopy: 165 × 73, 210 × 112, and 270 × 175 nm. To mimic the regions with wrinkles displaying different angles, each rod was built with 16 helices having four leveled and four inclined steps per pitch (pitch of 90 nm). The wrinkle width was fixed at 3.5 nm and the separation distance between wrinkles was 2.5 nm.

The simulations revealed chiral plasmonic bands that shifted toward longer wavelengths with increasing nanorod dimensions, in agreement with the experimental results (Fig. 4C and figs. S20 to S22). Differences in the anisotropy factors can be attributed to the polydispersity in size and shape of the real sample, as well as the difficulty in modeling the exact morphology of the NPs. Overall, our theoretical model confirmed the origin of the plasmonic chiroptical activity and the importance of sharp wrinkles for the emergence of strong CD responses. Replacement of BINAMINE by L-cysteine did not induce any clear modification on the CD signal obtained for chiral nanorods synthesized with either (R)-BINAMINE or (S)-BINAMINE. A substantial modification would be expected if the chiroptical properties stemmed from the coupling of the chiral molecules with the NP plasmon rather than from the chiral structural features (fig. S23) (43).

We have experimentally demonstrated that chiral gold nanostructures can be readily obtained by seeded growth of gold nanorods in BINOL-surfactant mixtures, whereas NP colloids with intense CD responses were obtained using BINAMINE-surfactant mixed micelles. The chiroptical properties stem from the formation and stabilization of steep chiral wrinkles. These findings point to a dual role of mixed micelles: (i) templating the growth of steep wrinkles and (ii) the subsequent stabilization of such morphological features.

We propose that the formation of wrinkles during seeded growth can be explained by the presence of elongated micelles coiled on the gold nanorod seeds, which act as patterns directing the diffusion of micellar aggregates containing gold ions, from the aqueous solution toward the core NC at intermicellar regions. Indeed, the size of micelles in solution would fit the interwrinkle distance on the nanorod surface. However, wrinkle formation is only induced when the rate of gold ion deposition is faster than their diffusion on the surface (44). We observed that upon lowering the concentration of reducing agent (ascorbic acid, from 160 down to 1.6 mM), the CD intensity decreased by an order of magnitude and less-defined surface roughness was observed in HAADF-STEM (fig. S24).

An additional source of surface stabilization is required to preserve the wrinkles after growth. In this case, BINAMINE plays such a role during the growth process, likely because of its two amine moieties in chelating conformation (33). When one of the amine groups in BINAMINE was replaced by a hydroxyl moiety, the passivation effect was suppressed and the growth of chiral features was hindered, which resulted in a decrease of CD intensity (fig. S25). Once the growth is completed, addition of a stronger stabilizing agent (e.g., cysteine) is necessary because the adsorption of surfactant micelles is not sufficiently strong and the NPs may slowly reshape, as reflected by the loss of chiroptical properties (fig. S26).

The approach described here for the synthesis of chiral plasmonic anisotropic NCs is simple, reproducible, and holds great potential for the large-scale production intrinsic to colloid chemistry methods. Compared with the growth of inorganic chiral nanostructures based on the direct chemisorption of small additives on the NC surface to induce chiral growth, the micelle-directed growth method relies on the ability of the chiral cosurfactant to direct the formation of helical micelles. Such supramolecular assemblies have sufficient interaction points with the NPs to effectively transfer their chirality to them during the growth step through a multivalency effect.

Supplementary Materials

Materials and Methods

Figs. S1 to S26

Tables S1 to S6

References (4560)

Movies S1 to S48

References and Notes

Acknowledgments: Funding: L.M.L.-M. acknowledges funding from the European Research Council (ERC AdG grant no. 787510). G.G.-R. and J.M. thank the Spanish MICIU for FPI (BES-2014-068972) and Juan de la Cierva (FJCI-2015-25080) fellowships. S.B., L.M.L.-M., V.K., and A.P.-T. acknowledge financial support from the European Commission under the Horizon 2020 Programme by grant no. 731019 (EUSMI) and ERC Consolidator grant no. 815128 (REALNANO). J.M.T. and F.O. acknowledge financial support from the Spanish MICIU (grants TEC2017-85376-C2-1-R and TEC2017-85376-C2-2-R), as well as from the ERDF and the Galician Regional Government as part of the agreement for funding the Atlantic Research Center for Information and Communication Technologies (AtlantTIC). A.G.-M. acknowledges financial support from the Spanish MICIU (grant no. RTI2018-095844-B-I00). E.G.N. and L.G.M. acknowledge funds from the Spanish MICIU (grant no. FIS2017-89361-C3-2-P), as well as the use of the Mare-Nostrum supercomputer and the technical support provided by Barcelona Supercomputing Center from the Spanish Network of Supercomputing (grant nos. QCM-2018-3-0039 and QCM-2019-1-0038). This work was performed under the Maria de Maeztu Units of Excellence Program from the Spanish State Research Agency (grant no. MDM-2017-0720). Author contributions: G.G.-R., J.M., and L.M.L.-M. conceived the project. G.G.-R., J.M., and V.K. performed NP synthesis and CD experiments. A.P.-T., I.L., and S.B. performed EM characterization and analysis. P.L., E.G.N., and L.G.M. executed and analyzed MD simulations. D.M.S., E.G.N. J.M.T., and F.O. performed electromagnetic modeling of plasmonic properties of chiral NPs. G.G.-R., J.M., A.G.-M., S.B., and L.M.L.-M. wrote the manuscript with comments from all authors. L.M.L.-M. supervised the project. Competing interests: The authors declare no competing interests. Data and materials availability: All data needed to evaluate the conclusions in the paper are presented in the main text or the supplementary materials.

Stay Connected to Science

Navigate This Article