Electrodissolution Inhibition of Gold Nanorods with Oxoanions - The

Subscriber access provided by UNIV OF LOUISIANA

Article

Electrodissolution Inhibition of Gold Nanorods with Oxoanions Charlotte Flatebo, Sean S. E. Collins, Benjamin S. Hoener, Yi-Yu Cai, Stephan Link, and Christy F. Landes J. Phys. Chem. C, Just Accepted Manuscript • DOI: 10.1021/acs.jpcc.9b01575 • Publication Date (Web): 06 May 2019 Downloaded from http://pubs.acs.org on May 6, 2019

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 32 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

Electrodissolution Inhibition of Gold Nanorods with Oxoanions Charlotte Flatebo,1,2 Sean S. E. Collins,2,3 Benjamin S. Hoener,2 Yi-yu Cai,2 Stephan Link,*2,3,4,5 Christy F. Landes*2,3,4,5 1

Applied Physics Program, 2Department of Chemistry, 3Smalley-Curl Institute, 4Department of Electrical and Computer Engineering, and 5Laboratory for Nanophotonics, Rice University, Houston, Texas 77005, United States Abstract Metal nanoparticles experience varied chemical environments that can cause corrosion and dissolution in electronics, electrocatalysis, and sensing applications. Understanding oxidative dissolution is critical for plasmonic nanoparticles because their optical properties strongly depend on size and shape. We demonstrate that the addition of low relative concentrations of oxoanions to aqueous halide electrolyte solutions improves the morphological stability of plasmonic gold nanorods at anodic electrochemical potentials that otherwise induce complete oxidative electrodissolution. Single particle hyperspectral dark-field imaging and correlated scanning electron microscopy show that oxoanions alter the electrodissolution onset potential, electrodissolution pathway, and nanoparticle reaction heterogeneity, as compared to chloride-only electrolyte solutions. We identify five mechanistic contributors to the corrosion inhibition capabilities of oxoanions in the presence of chloride ions, with the aim of expanding the range of electrochemical sensing and catalysis applications for plasmonic metal nanoparticles. Of the contributors investigated, the pH, adsorption potential and ionicity of the oxoanion are found to be the most influential factors, supporting the superior corrosion inhibition observed with bicarbonate and phosphate. *Corresponding authors: [email protected] (S.L), [email protected] (C.F.L.)

1 ACS Paragon Plus Environment

The Journal of Physical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 2 of 32

Introduction Controlling the dissolution of nanomaterials is essential for the stability and viability of functional nano-structured surfaces for electrochemical applications, which include catalysts,1-6 biomedical devices,7-8 energy storage devices,9 and biosensors.10-11 The chemical environment of each application varies.12-13 Furthermore, nanomaterials often exhibit different reactivity from their bulk counterparts,14 thus understanding dissolution and the nanotoxicological implications is paramount.15-17 Achieving a fundamental understanding of the underlying electrodissolution mechanisms is key for nanomaterial-based electrodes to find use in advanced energy applications.18-21 Gold nanorods (AuNRs) are a distinctive class of metal nanoparticles that offer appealing properties for electrochemical applications,22-25 with electronic tunability and electrochemical stability comparable to other nanomaterials.8,

11, 26-27

AuNRs have a relatively high oxidation

potential, which offers stability in electrolytic environments.2, 28-36 One of the most appealing properties of AuNRs is their extremely high absorption-cross section as a result of their localized surface plasmon resonance (LSPR).37-38 LSPR decay produces a high concentration of hot charge carriers that can be extracted to improve the catalytic efficiency of redox reactions.39-43 The LSPR of AuNRs is extremely responsive to the composition, size, shape, and environment of the metal nanoparticle resulting in precise tunability.37, 44 The drawback of optical tunability arises when dissolution occurs because changes in the shape and size alter the LSPR.22, 45 Metallic gold, in the presence of halide ions,46-49 can be electro-oxidatively dissolved under anodic potentials. Halide ions are common in both environmental and physiological conditions, but additional ions are often present.50 Investigations into the oxidation mechanisms of bulk gold have observed oxoanion inhibition of chloridation and electrodissolution.46, 51-52 Research groups 2 ACS Paragon Plus Environment

Page 3 of 32 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


studying gold nanoparticle electrochemistry have primarily examined anodic processes in electrolyte solutions containing single anion species.49, 53-60 Gold nanoparticle oxidation in the presence of halide-oxoanion electrolytes has been demonstrated, but the role of the oxoanions was not examined.61-63 In this study, we show that introducing low relative concentrations of oxoanions inhibits chloride-mediated electrodissolution of AuNRs. Although other single particle techniques exist such as inductively coupled plasma mass spectrometry27 or identical location transmission electron microscopy,64 optical screening of morphological changes is a high throughput, highly sensitive49 methodology that can be performed in situ.65 Using hyperspectral dark-field imaging, we monitored the change in scattering intensity and LSPR peak position of single AuNRs undergoing electrodissolution in chloride-oxoanion electrolyte solution mixtures. The changes in the LSPR were attributed to electrodissolution, which was verified by correlated scanning electron microscopy (SEM). Single particle spectroscopy was particularly useful here because of the heterogeneity of nanoparticle synthesis and electrode surfaces.65 Hyperspectral imaging allowed us to gain mechanistic insights about how competing oxoanions inhibit non-equilibrium, chloridemediated electrodissolution of metal nanocrystals. Methods Working Electrode Preparation ITO coated, 22 × 22 mm glass coverslips (Evaporated Coatings Incorporated, 55 ohm/sq.) were patterned with microscopic markings by milling with a focused ion beam (FIB) in a SEM (FEI Helios 660 NanoLab DualBeam, FIB set to 30 kV and 0.23 nA). FIB milled arrays of 5 µm lines, 25 nm deep, and 50 µm apart were used as position markers to directly correlate single AuNR



Page 4 of 32

SEM micrographs with spectroscopic and electrochemical data.66 After FIB milling, the ITO coverslips were plasma cleaned immediately before drop-casting 96 × 39 nm gold AuNRs (size characterized by transmission electron microscopy analysis, Figure S1), synthesized by the binary surfactant method.67 The stock solution of AuNRs was sonicated for an hour and then diluted with water to one-fiftieth of the as-synthesized stock solution concentration. The dilute solution was sonicated for 15 minutes and then 50 µL were immediately drop-casted on the ITO coated coverslips. The average particle density was about three AuNRs per 200 μm2. The wetted ITO coverslips were undisturbed for 45 seconds before drying with nitrogen gas. Electrolyte Solution Preparation Electrolyte solutions were prepared with Millipore filtered water (>2 MΩ.cm-1) and the respective sodium salt mixtures. Sodium chloride (≥ 99.0%) was purchased from JBaker; sodium bicarbonate (certified ACS grade) and hydrochloric acid (1 N) were purchased from Fisher Scientific; sodium acetate (≥ 99.0%), sodium phosphate monobasic (≥ 98%), and sodium fluoride (99.99%) were purchased from Sigma Aldrich. Electrochemical Cell Preparation To prepare a three-electrode electrochemical cell for each experiment, the following electrodes were used: ITO/AuNRs working electrodes, platinum wire (Sigma-Aldrich, 0.076 mm diameter) counter electrode (surface area: ~36 mm2), and side-insulated platinum wire (A-M Systems, 0.002 mm2 geometric area on the exposed end) quasi-reference electrode, fabricated with an electropolymerized polypyrrole coating (Pt/PPy) via the Bard method.68 To insulate the quasireference electrode and counter electrode from the working electrode, 0.12 mm thick adhesive spacers with 13 mm chamber diameters (Grace Bio-Laboratories) were placed between the


Page 5 of 32 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


electrodes leaving a working electrode geometric area of 133 mm2. To provide volume for the electrolyte solution, a 0.5 mm thick silicon spacer (Grace BioLaboratories) was modified to have a 17 mm chamber diameter and an electrolyte thickness of 740 μm, above the limit for a thin layer electrolyte.69 The electrolyte solution was filtered through a 0.22 µm polyethersulfone filter (Merck Millipore). A plasma-cleaned 0.17 mm thick glass coverslip (Fisher Scientific) was placed on top of the spacer to create a closed cell volume of 0.145 mL. The closed cell was pressure sealed with a custom-made aluminum sample holder. To determine adsorption

potentials

(Eads),

bulk,

polycrystalline gold electrodes (Deposition Research Laboratory, Inc. St. Charles, MO;

Figure 1. Hyperspectral measurement of single AuNR electrodissolution in 100 mM NaCl aqueous electrolyte solution. (A) Schematic of an optically transparent electrochemical cell with illumination and collection optics and the working (WE), reference (RE), and counter (CE) electrodes. (B) Schematic of experimental voltammetry switching pattern (gray) with colored dashed lines indicating hyperspectral image acquisition. (C) Example hyperspectral scattering images corresponding to acquisition conditions outlined in (B), (scale bars = 2 µm). A single AuNR undergoing electrodissolution is circled in green. The initial image, before E was applied, has a brown border. The E1 image (orange) was acquired at the open circuit potential after four CV scans with an Emax of 0.2 V. The Ed, Cl image (blue) was acquired after the CV scan Emax reached the electrodissolution onset potential. The Ef, Cl image (black) was acquired after a CV scan that yielded ~20% of initial scattering intensity (D) Single particle scattering spectra for the AuNR circled in (C). The insets are SEM micrographs of a representative starting AuNR (brown) and the circled AuNR after Ef, Cl (black). The scale bars represent 50 nm.

5 nm Ti adhesion layer, 10 nm Au films) were used as the electrode. Hyperspectral Dark-Field Imaging and Electrochemical Apparatus



Page 6 of 32

Hyperspectral imaging was used to measure scattering spectra of single AuNRs on the ITO working electrode. Figure 1A shows a simplified schematic of the transparent electrochemical cell. An inverted dark-field microscope (Zeiss AxioObserver m1) was used with a dark-field oilimmersion condenser (numerical aperture NA = 1.4). The condenser focused white light from a 100 W halogen lamp filtered with a 304-785 nm bandpass filter (Thorlabs) to the working electrode surface. An oil-immersion objective (Zeiss, PlanAchromat 63x, NA = 0.7-1.4, set to 0.7 for all experiments) collected the scattered light from the illuminated AuNRs. The sparsity of AuNRs on the ITO surface allowed for the acquisition of multiple spectra from individual AuNRs simultaneously. Hyperspectral images were collected with a spectrograph (Princeton Instruments, Acton SP2150i, 500 nm blaze wavelength 150 gr/mm diffraction grating) coupled to a chargecoupled device camera (Princeton Instruments, PIXIS 400) on an electronically-driven translation stage (Newport Linear Actuator, LTA-HL). The entrance to the spectrograph had a mechanical slit installed that was set to a width of 20 µm, projecting a strip of the image from the microscope onto the diffraction grating. The spectrometer was translated across the image plane to construct a hyperspectral data-cube using customized LabVIEW software (National Instruments). The electrochemical potential of the working electrode (E) was externally controlled with a threeelectrode potentiostat (CH Instruments, 630 D). To spectroelectrochemically monitor the dissolution during cyclic voltammetry (CV), a second mode of operation simultaneously measured the optical scattering spectrum and electrochemical potential of a single AuNR, shown in Figure S2.70 Electrodissolution Data Acquisition and Analysis A hyperspectral scattering image of a region of interest within a FIB milled array was acquired at the open circuit potential (Figure 1B, colored dashed lines) to establish initial scattering 6 ACS Paragon Plus Environment

Page 7 of 32 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


images and spectra of AuNRs (Figure 1C). The size of the area of interest ranged between 1,400 µm2 – 3,000 µm2, depending on the local AuNR density. An exposure time of 1 second was used for each pixel column, resulting in a total hyperspectral image acquisition time of two to four minutes. Following the acquisition of the hyperspectral image, CV scans were run while monitoring the scattering signal of a single AuNR to track the electrodissolution process in realtime. The initial CV scan had a starting and final potential of 0 V (vs Pt/PPy), a minimum potential of -0.6 V, and a maximum potential (Emax) of +0.2 V for four cycles at a rate of 40 mV/s as shown in Figure 1B. We used sets of repeat CV scans to increase the total time at Emax and, subsequently, increase the extent of electrodissolution to improve the detection accuracy of electrodissolution. After each set of four cycles was completed, a hyperspectral image was acquired. The electrodissolution of AuNRs was monitored by acquiring hyperspectral images at the open circuit potential after increasing Emax. CV scan sets, followed by hyperspectral image acquisition, were performed, increasing Emax by 100 mV for each set. The process was continued until the scattering intensity of the AuNR had decreased to 20% ± 10% of the initial intensity, with this Emax defined as the final potential (Ef). Figure S2 shows the decrease in maximum scattering intensity for each cycle of a single particle at Ef. Establishing Ef for each AuNR ensured that SEM micrographs of individual AuNRs could be acquired by preventing total dissolution. The ensemble threshold potential for electrodissolution (Ed) of the AuNRs was determined post-acquisition using a cumulative distribution function (CDF) of the spectral intensities of all AuNRs. Each single AuNR spectrum was fit with a Lorentzian curve using a nonlinear leastsquares fitting algorithm. The intensities and peak resonance wavelengths acquired were compared between hyperspectral images to quantify electrodissolution. The Ed threshold was set as the Emax resulting in ≥ 70% of particles decreasing in intensity by ≥ 20%. 7 ACS Paragon Plus Environment


Page 8 of 32

To compare results between different electrolyte solutions, all potentials were converted to the reversible hydrogen electrode (RHE) post-experiment and the CV ferrocyanide standards for the conversion are shown in Figure S3. The pH of each electrolyte solution can be found in Table S1. All potentials are reported relative to the RHE reference electrode unless stated otherwise. To acquire statistics on the final size and shape of the AuNRs, SEM, set to 5 kV, 100 pA with a 10 µs dwell time, was used. An in-house MATLAB script was employed to determine the lengths and widths of the final AuNRs. The volumes of the individual AuNRs were estimated as spherically capped cylinders.66 SEM was not performed prior to electrodissolution experiments to prevent electron beam surface modifications. Results and Discussion Hyperspectral imaging after CV scan sets reflects AuNR electrodissolution as a function of Emax for different electrolytes. Figure 1D shows the initial scattering spectrum for a representative AuNR in 100 mM chloride before CV. The scattering peak above 700 nm corresponds to the longitudinal LSPR.22 As has been observed previously,49 single particle scattering spectra collected with Emax

Electrodissolution Inhibition of Gold Nanorods with Oxoanions - The

Recommend Documents