Active control of plasmon coupling via simple electrochemical surface oxidation/reduction of Au nanoparticle agglomerates

Abstract

Reversible tuning of plasmon coupling of Au nanoparticle (AuNP) agglomerates containing dimers as the main component was achieved via electrochemical surface oxidation/reduction of the AuNP surface. The system required no reactant except for water and was almost finished within a unit second, which leads to novel active plasmonic devices.

---

Metal nanoparticles and nanostructures absorb and scatter light due to localized surface plasmon resonance (LSPR). Electric field localized at the resonance sites, the intensity of which is much higher than the incident light, enables enhancement of optical signals such as fluorescence¹ and Raman scattering.² Au and Ag nanoparticles (AuNPs and AgNPs) have been widely utilized for these applications because of their chemical stability. The LSPR wavelength depends on various factors, such as the size and shape of the nanoparticles,^3,4 free electron density of the metal,⁵ refractive index of the surroundings^6,7 and interparticle distance.^8,9 Control of these parameters is therefore crucial to obtain nanoparticles with the desired LSPR wavelength.

On the other hand, studies on active control of the LSPR properties through reversible control of the above factors have attracted great attention.¹⁰ There are a variety of methods for reversible tuning of LSPR: electrochemical deposition and dissolution of AgNPs,¹¹ capacitive charging and discharging^5,12–14 and the use of nanoparticles covered with stimuli-responsive materials (e.g. thermoresponsive,¹⁵ redox-active^16,17 or photochromic polymers¹⁸). Reversible control of LSPR is expected to apply not only to chromic materials but also to the on-demand enhancement of optical signals.¹⁰

When a couple of nanoparticles are close to each other, drastic electric field enhancement occurs at the particle gap due to dipole–dipole plasmon coupling.^8,9 Thus, reversible control of plasmon coupling should increase the usefulness of the active plasmonic systems. Although some researchers reported plasmonic nanoparticles with tunable interparticle distances,^19,20 the particles were dispersed in a solution and extra molecules such as thermo- or pH-responsive polymers were required for LSPR tuning. Byers and coworkers demonstrated switching between conductive and capacitive coupling based on the electrochemical reaction of the Ag/AgCl shell around the AuNP core.²¹ The study is based on the electrochemical reaction of the nanoparticle itself on an electrode while extra ions (i.e. Cl⁻) are necessary for the switching. One of the simplest ways to control LSPR is capacitive charging discharging^5,12–14 without any chemical reaction. However, the shift of the LSPR wavelength is quite small because the charges can be stored only on the nanoparticle surface. If the fast and large spectral shift of the plasmon coupling peak is realized by the simplest possible system, that would become a new candidate for active plasmonics.

In this study, plasmon coupling of AuNP agglomerates deposited on a transparent electrode is controlled reversibly via surface oxidation/reduction of the nanoparticle surface. A larger LSPR wavelength shift than individual AuNPs is induced in an electrolyte only containing supporting electrolytes. Fast spectral changes are achieved because the system does not require any reactant except for AuNPs and water and is based on slight changes in the immediate vicinity of the Au surface.

Preparation of AuNPs with an average diameter of 13 nm and deposition of the AuNPs onto an indium tin oxide (ITO) electrode were carried out according to methods described in a previous paper¹⁴ (see ESI† for experimental conditions). The concentration of the AuNP solution was estimated to be ca. 1.3 nM from a relationship between AuNP diameter and extinction coefficient²² or from the amount of Au precursor used, the average diameter and Au density. After adjusting the pH of the solution using dilute sulfuric acid, a cleaned ITO electrode was immersed in the solution for 5 or 20 min to deposit AuNPs. AuNPs are adsorbed onto the ITO surface when the pH of the solution is around 2.8.¹⁴Fig. 1a and b show an extinction (= absorption + scattering) spectrum of the AuNPs on ITO and a scanning electron microscope (SEM) image of the electrode surface (immersion time: 20 min), respectively. The extinction peak at ∼530 nm and the SEM image indicate that most of the AuNPs were deposited individually on the electrode. Particle density on the electrode was controlled by changing the immersion time.

Fig. 1 (a) Extinction spectra of AuNPs (red curve) and AuNP agglomerates including 50% of dimers (blue curve) deposited on ITO. (b) and (c) SEM images of (b) the AuNPs and (c) the AuNP agglomerates on ITO.

AuNP agglomerates were assembled on the AuNP-modified electrode through a method modified from that of previous literature.²² 3,6-Dioxa-1,8-octanedithiol (DODT) was diluted with water to 10⁵ times its volume and the electrode with AuNPs was immersed in the DODT solution for 1 h. After washing and drying, the electrode was immersed in a 10-times diluted AuNP solution for 2 h. An extinction peak assignable to plasmon coupling was observed at 610 nm (Fig. 1a) and AuNP agglomerates were formed on the electrode (Fig. 1c). The proportion of the dimers to all agglomerates and monomers was calculated to be ca. 50% (Fig. S1, ESI†) by counting randomly chosen 100 samples. The plasmon coupling peak was not observed clearly without DODT treatment (Fig. S2, ESI†), indicating that the agglomerates shown in Fig. 1c were formed through the dithiol linkers. Note that the experimental conditions were optimized by changing the immersion time and concentration of the AuNP solution (Fig. S1, ESI†).

Before investigating the electrochemical response of plasmon coupling, photoelectrochemical measurements were performed using the sample shown in Fig. 1b. The electrode with individually deposited AuNPs was immersed in an aqueous solution containing 0.01 M KNO₃ as the working electrode. A bare ITO and an Ag|AgCl (sat. KCl) electrode were used as the counter and reference electrodes, respectively. The extinction spectrum was measured after applying a potential for 50 s (see ESI† for experimental setup). The potential was changed from the rest potential in 0.20 or 0.30 V steps in the range of −0.55 to +1.15 V. The set of measurements including those from the rest potential to −0.55 V, −0.55 V to +1.15 V and +1.15 V to the rest potential was repeated three times.

Fig. 2a shows a wavelength–potential–extinction (λ–E–ext) plot of the individually deposited AuNPs. Extinction spectra at representative potentials are shown in Fig. 2b. Reversible changes in the extinction spectrum, blue and redshifts of the LSPR peak at more negative and positive potentials, respectively, were observed. Fig. 2c represents the relationship between applied potential and LSPR peak wavelength for the second and third cycles. Two linear slopes were observed below and above ca. +0.55 V. The slope below +0.55 V corresponds to the potential sensitivity ascribed to capacitive charging and discharging.¹⁴ The average slope of the second and third cycles was 6.5 nm V⁻¹ and relatively in accordance with the theoretical value (8.9 nm V⁻¹).¹⁴ It should be noted that the sensitivity of the first cycle was 2.7 nm V⁻¹, suggesting that citrate around the AuNPs prevented capacitive charging and discharging and was removed during the first cycle. The sensitivity was increased to 26.7 nm V⁻¹ at a more positive potential than +0.55 V, which is due to a surface oxidation of AuNPs.²³ AuNPs are covered with an Au oxide shell with a high refractive index (the real part is ∼2.8 in the visible region²⁴), so that the LSPR peak redshifts with increasing applied potential.

Fig. 2 (a) λ–E–ext plot of the individually deposited AuNPs in 0.01 M KNO₃. (b) Extinction spectra of the AuNPs at −0.55 V, +0.35 V and +1.15 V. (c) Relationship between applied potential and LSPR peak wavelength. (d) Relationship between Au₂O₃ shell thickness and LSPR peak wavelength obtained by FDTD calculation using a model shown in the inset.

To elucidate a relationship between applied potential and Au oxide shell thickness, spectral simulation of Au oxide-coated nanoparticles was carried out on the basis of a finite-difference time-domain (FDTD) method. The simulation model is shown in Fig. 2d (inset). Au oxide was assumed to be Au₂O₃. Complex refractive indices of Au and Au₂O₃ were referenced from previous papers.^24,25 The surrounding refractive index was set to 1.333 (water). The Au core radius r_Au and the outer radius of the core–shell structure r_Au2O3 were determined according to the following equation:²⁶

where r₀, V_Au2O3 and V_Au are the initial radius of the non-oxidized AuNP (6.5 nm), the molar volume of Au₂O₃ and that of Au, respectively. V_Au2O3 and V_Au were calculated to be 38.97 and 10.21 cm³ mol⁻¹, respectively, from their densities and molar masses.²⁷Fig. 2d shows the calculated LSPR peak wavelength depending on Au₂O₃ shell thickness (r_Au2O3 − r_Au). The slope was calculated to be 36.5 nm per unit Au₂O₃ thickness (nm). Assuming that capacitive charging and discharging affect the LSPR wavelength even above +0.55 V and the oxidation reaction was initiated from +0.55 V, the Au oxide thickness at +1.15 V, for instance, can be calculated as (26.7 − 6.5) × (1.15 − 0.55)/36.5 ≈ 0.3 nm. When the thickness is 0.3 nm, r_Au2O3 and r_Au are calculated to be ca. 6.6 and 6.3 nm, respectively.

Similar experiments to those in Fig. 2 were performed using the electrode modified with AuNP agglomerates shown in Fig. 1c. Fig. 3a shows a λ–E–ext plot of the electrode. The plasmon coupling peak was located at around 620 nm at the beginning of the first cycle whereas the peak appeared at ∼680 nm and changed reversibly after the first application of +1.10 V. Extinction spectra in the first and second cycles at +0.30 V are shown in Fig. 3b. The peak-to-peak separations (Δλ = λ₁ − λ_s), the difference between the LSPR wavelength for plasmon coupling (longitudinal mode) λ₁ and that for the mode perpendicular to the interparticle axis (transverse mode) λ_s, were 75 and 160 nm, respectively. A relationship between Δλ and interparticle distance for spherical AuNP agglomerates, s, can be calculated approximately by the following equation:²⁸

where d is the particle diameter. It is suggested from eqn (2) that s was decreased from ∼0.5 nm to ∼0.3 nm in the first cycle. The decrease is probably due to the oxidative desorption²⁹ of DODT around the AuNPs. The desorption is supported by the following results: an anodic current was observed at a more positive potential than ∼1.0 V in a cyclic voltammogram of an AuNP-modified electrode after the DODT treatment (Fig. S3, ESI†) and the increase in Δλ was not observed in cycles ranging from −0.50 to +0.30 V.

Fig. 3 (a) λ–E–ext plot of AuNP agglomerates including dimers in 0.01 M KNO₃. (b) Extinction spectra of the agglomerates at +0.30 V in the first and second cycles. (c) Extinction spectra of the agglomerates at −0.50 V, +0.50 V and +1.10 V in the second cycle. (d) Relationship between applied potential and wavelength of the plasmon coupling peak in the third cycle.

It is noteworthy in the second and third cycles that the plasmon coupling peak was changed significantly (Fig. 3c and d) compared to that of the individual AuNPs (Fig. 2b and c) or the transverse mode at more positive potential than +0.50 V. The peak was shifted bathochromically by 42 nm when the potential was changed from +0.50 V to +1.10 V. The potential sensitivity in the third cycle was determined to be 69.8 nm V⁻¹, reflecting that plasmonic nanoparticle dimer structures exhibit higher refractive index sensitivity than single nanoparticles.^30,31 In order to confirm the mechanism, FDTD simulation was carried out using a dimer model in water (Fig. 4a). The Au oxide shell thickness r_Au2O3 − r_Au was set to 0.3 nm, which was calculated from Fig. 2c and d and eqn (1). Fig. 4b shows calculated extinction spectra before (red curve) and after (blue curve) the surface oxidation. The experimental result was reproduced well by the simulation: the plasmon coupling peak was redshifted by 41 nm and the extinction intensity decreased after the surface oxidation. Note that the peak shift reached 244 nm when assuming that the Au₂O₃ shell was formed without changing the core diameter (Fig. 4b, black broken curve). Thus, the reproduced result can be interpreted as the balance between the high refractive index sensitivity of the plasmon coupling peak and a decrease in the effective interparticle distance because a part of the metallic Au surface region is transformed to dielectric Au₂O₃.

Fig. 4 (a) Dimer model used for the spectral simulation based on an FDTD method. (b) Calculated extinction spectra of the dimer (b) before (red curve) and (c) after (blue curve) the surface oxidation. The black broken curve represents a calculated spectrum if the Au₂O₃ shell with a thickness of 0.3 nm is formed without changing the core diameter.

Finally, the response time of the spectral changes was evaluated using an ITO electrode with AuNP agglomerates fabricated separately. Fig. 5a shows changes in extinction at the plasmon coupling peak (664 nm) during alternative application of −0.50 V and +1.10 V in 0.01 M KNO₃. Enlarged images of Fig. 5a are shown in Fig. 5b and c. Extinction changes caused by the oxidation and reduction reactions almost finished within ∼1.0 s and ∼0.6 s, respectively. Even though the response times were measured as an ensemble average, they were much faster than or comparable to previously reported ones determined by single particle spectroscopy.²¹ The fast response is probably because the oxidation reaction of Au requires only water present in large amounts in the system and takes place only at the surface region of the AuNPs. The simple oxidation reaction, forming only a subnanometer oxide layer, had good chemistry with AuNP agglomerates having high refractive index sensitivity^30,31 especially in the nanoparticle gap.

Fig. 5 (a) Extinction changes at 664 nm during alternative application of –0.50 V and +1.10 V. (b) and (c) Enlarged images of panel a to determine the switching time for the (b) oxidation and (c) reduction processes.

In conclusion, we succeeded in controlling plasmon coupling by electrochemical surface oxidation/reduction of AuNP agglomerates including 50% of dimers. The redox reactions proceeded only at the surface region of AuNPs and required only water, resulting in fast active control of plasmon coupling properties. A higher sensitivity system such as Au nanocube dimers³¹ or nanoparticle or hole arrays^32,33 would lead to superior switching systems.

This work was supported in part by a Grant-in-Aid for Scientific Research (B) (JP23H01932) from the Japan Society for the Promotion of Science (JSPS). The author also acknowledges a research grant from the Proterial Materials Science Foundation. SEM measurements were performed at the Division of Instrumental Analysis at University of Toyama.

Data availability

The data that support the findings of this study are available from the corresponding author upon reasonable request.

Conflicts of interest

There are no conflicts to declare.

Notes and references

1. D. Semeniak, D. F. Cruz, A. Chilkoti and M. H. Mikkelsen, Adv. Mater., 2023, 35, 2107986.
2. N. S. Mueller, E. Pfitzner, Y. Okamura, G. Gordeev, P. Kusch, H. Lange, J. Heberle, F. Schulz and S. Reich, ACS Nano, 2020, 14, 28.
3. P. K. Jain, K. S. Lee, I. H. El-Sayed and M. A. El-Sayed, J. Phys. Chem. B, 2006, 110, 7238.
4. H. Chen, X. Kou, Z. Yang, W. Ni and J. Wang, Langmuir, 2008, 24, 5233.
5. T. Ung, M. Giersig, D. Dunstan and P. Mulvaney, Langmuir, 1997, 13, 1773.
6. K. A. Willets and R. P. Van Duyne, Annu. Rev. Phys. Chem., 2007, 58, 267.
7. K. M. Mayer and J. H. Hafner, Chem. Rev., 2011, 111, 3828.
8. W. Rechberger, A. Hohenau, A. Leitner, J. R. Krenn, B. Lamprecht and F. R. Aussenegg, Opt. Commun., 2003, 220, 137.
9. P. K. Jain, W. Huang and M. A. El-Sayed, Nano Lett., 2007, 7, 2080.
10. N. Jiang, X. Zhuo and J. Wang, Chem. Rev., 2018, 118, 3054.
11. A. Tsuboi, K. Nakamura and N. Kobayashi, Adv. Mater., 2013, 25, 3197.
12. A. H. Ali, R. J. Luther and C. A. Foss, Jr., Nanostruct. Mater., 1997, 9, 559.
13. A. Toyota and T. Sagara, Electrochim. Acta, 2008, 53, 2553.
14. H. Nishi, S. Hiroya and T. Tatsuma, ACS Nano, 2015, 9, 6214.
15. M. Nguyen, X. N. Sun, E. Lacaze, P. M. Winkler, A. Hohenau, J. R. Krenn, C. Bourdillon, A. Lamouri, J. Grand, G. Levi, L. Boubekeur-Lecaque, C. Mangeney and N. Félidj, ACS Photonics, 2015, 2, 1199.
16. J.-W. Jeon, P. A. Ledin, J. A. Geldmeier, J. F. Ponder, Jr., M. A. Mahmoud, M. El-Sayed, J. R. Reynolds and V. V. Tsukruk, Chem. Mater., 2016, 28, 2868.
17. W. Z. Lu, N. N. Jiang and J. F. Wang, Adv. Mater., 2017, 29, 1604862.
18. H. Nishi, T. Asahi and S. Kobatake, J. Phys. Chem. C, 2009, 113, 17359.
19. Y. D. Liu, X. G. Han, L. He and Y. D. Yin, Angew. Chem., Int. Ed., 2012, 51, 6373.
20. Z. H. Sun, W. H. Ni, Z. Yang, X. S. Kou, L. Li and J. F. Wang, Small, 2008, 4, 1287.
21. C. P. Byers, H. Zhang, D. F. Swearer, M. Yorulmaz, B. S. Hoener, D. Huang, A. Hoggard, W.-S. Chang, P. Mulvaney, E. Ringe, N. J. Halas, P. Nordlander, S. Link and C. F. Landes, Sci. Adv., 2015, 1, e1500988.
22. H. Cha, J. H. Yoon and S. Yoon, ACS Nano, 2014, 8, 8554.
23. H. Nishi and T. Tatsuma, Angew. Chem., Int. Ed., 2016, 55, 10771.
24. Y.-L. Liu, C.-Y. Fang, C.-C. Yu, T.-C. Yang and H.-L. Chen, Chem. Mater., 2014, 26, 1799.
25. P. B. Johnson and R. W. Christy, Phys. Rev. B: Solid State, 1972, 6, 4370.
26. Y. Ma, A. L. Highsmith, C. M. Hill and S. Pan, J. Phys. Chem. C, 2018, 122, 18603.
27. K. M. Cook and G. S. Ferguson, J. Phys. Chem. C, 2011, 115, 22976.
28. G. A. Kelesidis, D. Gao, F. H. L. Starsich and S. E. Pratsinis, Anal. Chem., 2022, 94, 5310.
29. H. Hagenström, M. A. Schneeweiss and D. M. Kolb, Langmuir, 1999, 15, 2435.
30. A. Bonyár, Biosensors, 2021, 11, 527.
31. A. Bonyár, I. Csarnovics and G. Szántó, Photonics Nanostructures: Fundam. Appl., 2018, 31, 1.
32. P. Offermans, M. C. Schaafsma, S. R. K. Rodriguez, Y. Zhang, M. Crego-Calama, S. H. Brongersma and J. G. Rivas, ACS Nano, 2011, 5, 5151.
33. V. E. Bochenkov, M. Frederiksen and D. S. Sutherland, Opt. Express, 2013, 21, 14763.

---

Footnote

† Electronic supplementary information (ESI) available: Experimental details, SEM images of AuNP agglomerates prepared by various conditions, control experiment without using DODT and cyclic voltammogram to investigate oxidative desorption of DODT. See DOI: https://doi.org/10.1039/d4cc02024c

---