Interrogating Charge Storage on Redox Active Colloids via Combined

Jun 16, 2017 - Noah B. SchorrAnnie G. JiangJoaquín Rodríguez-López ... Mihail R. KrumovBurton H. SimpsonMichael J. CounihanJoaquín Rodríguez-López...
5 downloads 0 Views 1MB Size
Subscriber access provided by University of Rochester | River Campus & Miner Libraries

Article

Interrogating Charge Storage on Redox Active Colloids via Combined Raman Spectroscopy and Scanning Electrochemical Microscopy Zachary T Gossage, Noah Benjamin Schorr, Kenneth Hernandez-Burgos, Jingshu Hui, Burton H. Simpson, Elena C. Montoto, and Joaquin Rodriguez-Lopez Langmuir, Just Accepted Manuscript • DOI: 10.1021/acs.langmuir.7b01121 • Publication Date (Web): 16 Jun 2017 Downloaded from http://pubs.acs.org on June 19, 2017

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 free 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 accessible to all readers and 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.

Langmuir 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 11

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

Langmuir

Interrogating Charge Storage on Redox Active Colloids via Combined Raman Spectroscopy and Scanning Electrochemical Microscopy Zachary T. Gossage,†,‡,ø Noah B. Schorr,†,‡,ø Kenneth Hernández-Burgos,†,‡,« Jingshu Hui,‡ ,φ Burton H. Simpson,‡ Elena C. Montoto,†,‡ Joaquín Rodríguez-López†,‡,«,* †

Joint Center for Energy Storage Research



Department of Chemistry, University of Illinois at Urbana-Champaign, 600 South Mathews Avenue, Urbana, IL 61801 φ

Department of Materials Science and Engineering, University of Illinois at Urbana-Champaign, 1304 W Green St, Urbana, IL 61801 «

Beckman Institute for Advanced Science and Technology, University of Illinois at Urbana-Champaign

* Corresponding author ø

Authors equally contributed

ABSTRACT: Redox active colloids (RACs) are dispersible, cross-linked polymeric materials that incorporate a high concentration of redox-active motifs, making them attractive for next-generation size-exclusion redox flow batteries. In order to tap into their full potential for energy storage, it is essential to understand their internal charge mobility, capacity, and cyclability. Here we focus on using a combined suite of Raman spectroscopy and scanning electrochemical microscopy (SECM) tools for evaluating three important parameters that govern charge storage in viologen-RACs: their intra-particle redox active concentration, their reduction/oxidation mechanism, and their charge transfer rate. We addressed RACs using SECM imaging and single-particle experiments, from which the intra-particle diffusion and concentration parameters were elucidated. By using Raman Spectroscopy coupled to Surface Interrogation SECM, we further evaluated their reversible redox properties within monolayer films of 80 and 135 nm sized RACs. Most notably we have confirmed that the concentration and redox mechanisms are essentially unchanged when varying the RAC size. As expected, we see that larger particles inherently require longer times for electrolysis independent of the methodology used for their study. Our simulations further verify the internal concentration of RACs and suggest their porosity enables solution redox active mediators to penetrate and titrate charge in their interior. The combined methodology presented here sets an important analytical precedent in decoupling the charge storage properties of new bulk materials for polymer batteries starting from probing lowdimensional assemblies and single particles using nano- and spectroelectrochemical approaches.

INTRODUCTION Polymeric redox active colloids (RACs) are emerging energy storage materials that combine size tunability, a defined spherical geometry, and high charge capacity, making them attractive for size exclusion battery approaches.1 Balancing charge accessibility in RACs while improving their size-exclusion by means of increasing their volume, requires a careful evaluation of rate, cyclability, and internal redox-active component concentration. Because RACs rely on long-distance intra-particle charge transfer, reliable access to their maximum charge requires efficient charge mobility with minimization of undesirable charge trapping.1,2 Charge diffusion in RACs is dominated by electron

hopping between redox units on the polymer chain with associated counter-ion and solvent transport.3-5 Diffusional processes can be evaluated in deposited redox active polymer films3, 6 or in solution.1, 5, 7-10 However, many methodologies for investigating extended redox networks convolute diffusion parameters with concentration, making it difficult to precisely acquire them. Viologen-containing RACs are particularly interesting not only because of their potential applications in all-organic polymer batteries, but also because they can be conveniently probed using simultaneous redox and spectroscopic approaches.1 The ethyl viologen pendants of RACs can exist in three oxidation states, undergoing reversible

1

ACS Paragon Plus Environment

Langmuir

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 11

Figure 1. Combined SECM and Raman spectroscopy approaches for evaluating RAC electrochemical properties. (a) SECM approach and contact to single particle (b) Simultaneous spectroscopic and electrochemical monitoring of RAC monolayer films with an inverted Raman microscope and SI-SECM. (c) Monolayer reduction and oxidation measurements using SISECM. and sequential electron transfers, first from the native state (V2+) to the radical cation (V+·) and then to the neutral form (V) (Figure S1).11, 12 Here, we apply three different interrogation techniques based on scanning electrochemical microscopy (SECM) (Figure 1) and its combination to Raman spectroscopy (Figure 1b) to contrast different modes of charge accessibility. This strategy effectively decouples the contributions from charge diffusion and redox probe concentration. We used SECM to make contact between a nanoelectrode tip and single RACs for direct reduction/oxidation with cyclic voltammetry (CV) or chronoamperometry (CA) (Figure 1a). This greatly improves the quality and reliability of the obtained data in comparison to bulk and ensemble electrochemical measurements which often display resistive effects and uncertainties regarding the extent of electrolysis. SECM has been previously used for studying single nano- and micro-sized entities,1, 13-15 making it an ideal platform to investigate individual RACs to understand intra-particle processes. We further employed in situ spectroelectrochemical measurements to decouple diffusion and concentration and evaluate charge movement within RACs. Coupling surface interrogation SECM (SI-SECM) to Raman spectroscopy enables us to comprehensively study the charge transfer mechanism in RAC monolayer films (Figure 1b). Viologen derivatives have multiple Raman-active peaks that allow structural changes to be observed during the reduction to the radical cation and neutral form.11, 12, 16-18 A

strong resonance Raman signal from V+· arising from an absorption region (540 nm) near the excitation wavelength (532 nm), allows monitoring of the signal on non-noble electrode surfaces, circumventing the need for surface-enhancement.17 Raman spectroscopy has previously been coupled to shear force SECM,19 but this is the first time that SECM measurements have been used in-situ with Raman spectroscopy with the intention of monitoring transient processes. SI-SECM is a powerful method for studying concentration within RAC films because it bypasses the need for electrode contact to the particles. Instead, it relies on the mediation by a redox species to fully access the interior of the particle’s redox loading. This technique has been applied to the titration and quantification of surface-confined redox active species and polymer films.2, 20, 21 SI-SECM experiments following either direct reduction with an electrode (Figure 1b) or indirectly using a reducing mediator (Figure 1c) illustrate the differences in the efficiency of charge accessibility. Comparison of the contact SECM mode, spectroelectrochemical, and SI-SECM measurements provides in depth detail to the nature of charge transfer in RACs. These localized measurements do not suffer the same assumptions necessary for bulk electrochemical methods, thus providing reliable metrics to evaluate the extent and rate of charge and discharge processes.

2

ACS Paragon Plus Environment

Page 3 of 11

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

Langmuir

EXPERIMENTAL SECTION Chemicals and Reagents Ferrocene (Fc, Aldrich, 98%) and ferrocenemethanol (FcMeOH, Sigma-Aldrich, 97%) were purchased from commercial sources and used as received as redox mediators for the SECM experiments. Lithium tetrafluoroborate (LiBF4, (Aldrich, 99.99%)) and tetrabutyl ammonium hexafluorophosphate (TBAPF6, (TCI, >98%)) were used as electrolytes in organic solutions. Potassium chloride (KCl, (BDH, 99%)) and HPLC grade water (Macron) were used for all aqueous experiments. [1,1’-diheptyl-4,4’-bipyridinium]2+ 2PF6- (heptyl viologen, HV) was prepared via an ion exchange reaction using ammonium hexafluorophosphate (Aldrich, 99.99%) and [1,1’-diheptyl-4,4’-bipyridinium]2+ 2Br- (Aldrich, 97%) in a solution of acetonitrile and minimal added water. The uptake of PF6- was confirmed with IR spectroscopy (Figure S2). RAC particles containing PF6- counter-ions (80 ± 11 nm, 135 ± 12 nm) were prepared as described previously.1 RAC particles containing Cl- counter-ions (1.2 µm) were prepared with the same methodology excluding the final ion exchange step. Anhydrous dimethyl formamide (DMF) was used as received from Sigma-Aldrich. SECM probes, including Wollaston electrodes (Goodfellow, purity 99.9%, 300 nm radius) and platinum ultramicroelectrodes (Goodfellow, purity 99.9%, 12.5 μm radius) were prepared as described in previous reports.22, 23 Substrate Preparation Glass coverslips (Ted Pella, 0.13-0.16 mm thick) were used as insulator substrates. Indium tin oxide (ITO) coverslips (SPI 15-30 Ω) were used as conductive substrates. For single particle measurements, a 5 µL droplet of RACs in acetonitrile (0.1 mg/mL, previously dispersed through sonication) was casted onto the center of a glass substrate and allowed to dry under ambient conditions. The substrate was checked under an optical microscope (Zeiss). For monolayer measurements on glass or ITO, substrates were prepared using an air-liquid interface method described in a previous publication.1 Briefly, a solution of RACs was prepared in acetonitrile (3 mg/mL) and sonicated. A small volume (40 μL) of the RAC suspension was dispensed on top of water in a small glass trough. After 24 hours, a substrate was pressed through the monolayer floating on the water and allowed to dry for 12 hours (Figure S3). SECM Contact to Single Particles

feedback while rastering the electrode across the surface for SECM imaging. After discrete particles were identified via imaging, the electrode was positioned directly over the particle and verified via line scans (Figure S4). Contact to the particle was then made through a manual approach (100-300 nm steps toward the particle) with CVs (20 mV/s) taken at each step. Once the particle was contacted, as evaluated by a faradaic contribution from viologen reduction (Figure S5), CV was conducted at various scan rates. The steady state current and total charge passed/collected were then extracted after subtracting the background current obtained in the absence of particle contact. CA measurements were also recorded using a potential step to the first reduction process. CA data from single particles were compared with simulations using COMSOL (described below) or with the method proposed by Li et al.25 In Situ Raman Spectroscopy Coupled to SI-SECM The Raman microscope used in this study is an in-lab constructed inverted configuration setup (Figure 5a). Incorporation of a Raman microscope to an SECM was accomplished as follows. The laser line from a 532 nm diode laser (Melles Griot) was first directed through 532 nm bandpass filter (Semrock). The filtered laser line was then passed through a short pass dichroic beamsplitter (Semrock). The beam line was then focused to the sample by a 20x long working distance objective 0.42 NA (Mitutoyo). For Raman SI-SECM measurements the UME was approached to the surface, and the laser focused to the same region using micro positioners. Scattered light was collected by the objective and reflected off the dichroic filter and focused through two adjustable mechanical slits opened to 5 x 5 µm (Thorlabs) onto a 10:90 partially reflective mirror (Semrock). Reflected light was focused onto a CCD camera (Thorlabs) for sample visualization. Transmitted light was directed through a 532 nm notch filter (Semrock) and then focused to a fiber optic leading to a spectrometer (Ocean Optics QE Pro). Raman-SECM measurements were made on 80 and 135 nm RAC monolayer films on ITO. A Pt wire as the counter electrode and a Ag/AgCl reference electrode (CH Instruments, Inc.) were used for all Raman-SECM measurements. 0.5 mM FcMeOH was used as a redox mediator and titrant with 100 mM KCl as the supporting electrolyte in water. For chronoamperometric reduction and oxidation of RAC films, a potential of -0.6 V vs Ag/AgCl was applied to the ITO substrate for 99 s to reduce the RAC film to V+·,

All electrochemical measurements were performed using a CHI900D Scanning Electrochemical Microscope (CH Instruments, Inc.). Measurements were performed inside an oxygen and moisture free glove box. For SECM procedures, a 300 nm Pt UME was used as a working electrode with a Pt wire as the counter electrode and Ag wire as a quasi-reference. The substrate was first leveled using a redox mediator and negative feedback. Then an SECM probe was approached to the substrate and positioned using the theory of Cornut and Lefrou.24 We further used negative

3

ACS Paragon Plus Environment

Langmuir

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 4 of 11

then 0 V vs Ag/AgCl was applied for 99 s to oxidize the RAC film back to V2+. For SI-SECM measurements a Pt UME was approached to the film surface using FcMeOH and negative feedback (Figure S6). Once approached to the surface, CA was performed with the microelectrode to determine the steady-state current from the redox mediator under negative feedback conditions. A potential of -0.6 V vs Ag/AgCl was then applied to the ITO substrate for 30 s, CA reduction/oxidation experiments showed that 30 s was sufficient to fully reduce the monolayer film to the V+· state. The film was then titrated by FcMeOH+ formed at the microelectrode tip while it recorded the current (Figure S15). Raman measurements were taken continuously for all electrochemical measurements with an acquisition time of 3 s for each spectrum unless otherwise specified. The measurements were taken in solutions covered with Parafilm that were purged with argon and left under an argon positive pressure blanket to avoid oxygen interference. Localized SI-SECM on RAC Films SI-SECM was used to study small regions of a monolayer film on an insulator in DMF with LiBF4 as the electrolyte. The films were both reduced and oxidized with tip generated mediators. A 1.2 mM solution of HV2+ was used to first reduce a portion of the film below the tip. While the electrode was held near the film, a chronoamperometric step was applied to reduce HV2+ for 300 s. An electron transfer occurs between HV+ and the film species because of the

Figure 3. Background subtracted CA reduction of single particle on glass substrate in DMF after making contact with an SECM electrode. One fitting is based on a COMSOL model (orange curve) that allows particle size, mediator concentration and DEX to be defined. The other fitting, black curve, is based on an exponential model previously described by Li et al.25 more negative redox potential (~150 mV) for the mediator.11 Thereafter, Fc (2.3 mM) acted as a titrant to oxidize the film. In this case the Fc species was also oxidized for 300 s. These experiments were performed inside a water and oxygen free glove box. Numerical Simulations Simulations were completed using the Transport of Diluted Species module within COMSOL Multiphysics 4.4. For our simulations, we utilized a 2-D axisymmetric geometry. SI-SECM simulations used a bimolecular reaction between a solution species formed at the electrode probe and the film, while CA simulations explored charge transport within a single particle (Figure S7). A more thorough description is provided in the supplemental (Supplemental, pg 5-6).

RESULTS AND DISCUSSION Direct Electrical Contact to Single RACs

Figure 2. (a) Optical microscopy of RACs on glass after drop-casting. (b) SECM imaging of a single RAC on glass using negative feedback.

Drop-casting from a dilute solution (0.1 mg/mL) was sufficient for isolating RACs on top of glass as shown in Figure 2a. The optical image shows individual particles with a dry diameter of 1.2 ± 0.2 µm (standard deviation, n = 34). SECM allowed us to position a 300 nm (radius) disk-shaped electrode above the particles at the surface. By using Fc as a redox mediator, we observed negative feedback, i.e., a decrease in current when approaching a glass substrate (Figure S8).26 24, 27 Negative feedback imaging allowed us to identify individual RAC particles and evaluate their size by using an edge-to-edge measurement from the SECM image. We found that the swollen diameter of the particles

4

ACS Paragon Plus Environment

Page 5 of 11

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

Langmuir

Figure 4. (a) CV at 20 mV/s on two single RAC particles (2 µm diameter) after background subtraction. (b) Scan rate dependence of CV when in contact with a single particle. in DMF was 2.0 ± 0.5 µm (n=34, Figure 2b). After contact to a particle with an SECM probe, the viologen within the RAC can be electrolyzed to explore intra-particle charge transfer. In Figure 3, results of reducing a single particle on top of glass are shown after background subtraction (SECM particle image, Figure S9). For this measurement we biased the SECM tip 180 mV more negative than the E1/2 to ensure the quantitative reduction of V2+ to V+·. The chronoamperogram shows a current decay for a 300 s time interval. A best fit to this curve was obtained using two different models. The first model consisted of a modified bulk electrolysis procedure assuming a single contact point between a nanoelectrode and a particle which does not explicitly require assumptions about the intra-particle concentration. This model is based on the equation:25 𝑖(𝑡) = 𝑖𝑝 𝑒 −(

4𝑟𝑒 𝐷𝐸𝑋 )𝑡 𝑉

(1)

where ip is the maximum current after the potential step, re is the effective contact radius, DEX is the diffusion coefficient, V is the volume of the particle, and t is time. Using this method, we obtained a charge diffusion coefficient of 2.4 x 10-10 cm2/s with a least squares fit to a single exponen-

tial. Following this initial analysis, finite element simulations using COMSOL Multiphysics were performed to gain a more precise understanding of charge transport within single RAC particles. In the COMSOL simulations, diffusion of the redox couple was simulated in a 2D axisymmetric domain representing a radial cross section of a spherical particle in contact with a nanoelectrode (Figure S7a). At the boundary between the particle and the electrode, a zero concentration for V2+ was simulated and the current was extracted using a boundary probe. All other boundaries were insulating. Best fits were selected by eye for evaluating the effect of different diffusion coefficients on the simulated transients and in comparison to the experiment (Figure S10). A suitable profile for the current decay obtained via COMSOL using the diffusion coefficient from the first approach was obtained when using an intra-particle concentration of 1.25 M of V2+. The concentration and diffusion coefficients we found are in good agreement with trends estimated previously1 through CV at a RAC monolayer electrode. The larger diffusion coefficient of the 1.2 µm colloid follows the same tendency found for the smaller RAC particles in organic and aqueous media (Figure S17; table S1), where the diffusion coefficient increases for particles with a larger diameter.1 The combination of methods reported here, and applied on a single particle, provides us with a strong degree of confidence on the accuracy of the charge properties determined for these RACs. We confirmed the results from CA by performing CV in contact with a RAC (Figure 4), where intra-particle charge transfer diffusion is approximated via the steady state equation:27 𝑖𝑠𝑠 = 4𝑛𝐹𝐷𝑎𝐶

(2)

where iss is the measured steady state current, n is the number of electrons for the process, F is Faraday’s constant, D is the diffusion coefficient, a is the radius of the electrode, and C is the concentration of the system. This is only an approximation, because diffusion in our experiment is restricted to the particle, while this equation applies for a semi-infinite boundary condition.26 By using various scan rates (Figure 4b), we found deviation from steady-state behavior at scan rates below 20 mV/s, which exhibited a decreasing limited current upon reduction beyond E0. This is likely related to the larger degree of electrolysis attained over the longer time required for this voltammogram to proceed, given the restricted geometry for the particle. By using the background-subtracted current for a relatively fast scan rate of 20 mV/s (Figure 4a) and the concentration derived from CA (1.25 M) a measurement of the electron diffusion through different RACs using the steady state equation (2) was determined to be 3 x 10-10 ± 2 x 10-10 cm2/s (standard deviation for 9 particles). If we use the diffusion coefficient from our bulk electrolysis simulations and the average steady state current for a single particle we find an intra-particle concentration of 1.1 M. Within a reasonable accuracy, these results are consistent and also on the same order of magnitude with previous bulk measurements.1

5

ACS Paragon Plus Environment

Langmuir Spectroelectrochemical Measurements

method that does not require physical contact with an electrode. We found SI-SECM to be ideally suited for this determination, in addition its coupling with Raman spectroscopy enables a direct monitoring of compositional

In order to rule out contributions from charge trapping, and to increase our confidence on the parameters determined using single-particle analysis, we now turn to a

Microelectrode

a)

SECM

CCD camera

Fiber optic to spectrometer

Objective lens Slits Dichroic filter laser line

c)

6 4 2 0 -2 -4 0.0

-0.1

-0.2

-0.3

-0.4

-0.5

e)

70 60 50 40 30 20 10 0 -10 -20 -30 0

10

20

30

40

50

60

70

0.8

0.6

0.4

0.2

0.0 -0.15 -0.30 -0.45 -0.60

Potential (V vs Ag/AgCl)

d)

1031 cm-1 1363 cm-1 1522 cm-1 1531 cm-1 1660 cm-1

1.0

-0.6

80

90

100

Normalized 1530 cm-1 Intensity

Current (A)

8

Normalized Peak Intensity

Partially reflective mirror

10

b)

Current (A)

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 6 of 11

-0.45 -0.30 -0.15

Potential (V vs Ag/AgCl) 1.0

0.8

0.6

0.4

0.2

0.0 0

Time (s)

50

100

150

200

Time (s)

Figure 5. (a) Schematic of instrumentation for in situ spectroelectrochemical measurements. (b) 5 mV/s CV of 135 nm RAC monolayer film on ITO. (c) Normalized peak intensities of Raman spectra of 135 nm RAC monolayer film on ITO taken during 5 mV/s CV shown in (b) with 10 s acquisition time. (d) CA 99 s reduction, orange, and oxidation, blue, of 135 nm RAC monolayer film on ITO. (e) Normalized intensity of 1530 cm -1 peak during reduction and subsequent oxidation of monolayer film shown in (d). Error bars represent standard deviation for n=4 measurements. Normalization was performed with respect to the maximum intensity detected for V+ in a given experiment.

6

ACS Paragon Plus Environment

Page 7 of 11

Scheme 1: Electrochemical reactions involved in Raman coupled SI-SECM of RAC films. Reaction (a) is the reduction 0f a RAC film deposited on an ITO surface. Reactions (b) and (c) are the production of titrant species FcMeOH+ at the microelectrode and subsequent reaction with the reduced viologen film.

during CV, and starting at a potential of -0.35 V vs. Ag/AgCl, the 135 and 80 nm RAC films displayed rapid appearance and growth of Raman active vibrational modes which are associated with the radical cation viologen moiety (Figure 5c).16, 17, 28 The high baseline intensity of the 1660 cm-1 vibrational mode originates from the weak Raman scattering from water. Other than changes in intensity, the Raman spectra showed no variation in peak position or full width half maximum intensity (Figure S11). The similar trend between vibrational modes in peak growth and decay allowed us to select only the strongest peak (1530 cm-1) to monitor the behavior of the viologen pendant. The return to baseline of the Raman signal upon the reverse scan of the CV indicates that all of the generated radical cation is reduced to the native state. This result supports the idea that there is little to no charge trapping present in RACs.

We first analyzed the redox behavior using a conductive ITO substrate. Before any potential was applied to the film, there were no visible Raman peaks. From spectra taken

The oxidation of the RAC films was evaluated by titration using SI-SECM (Scheme 1). By aligning the SECM tip to the area interrogated by the focus of the objective, we were

a)

b) 1.0 0.8 0.6 0.4 0.2 0.0 0

20

40

60

80

100

Normalized 1530 cm-1 Intensity

changes upon reversible redox processes in RACs. We performed in situ Raman spectroscopy simultaneously with chronoamperometry and SI-SECM on 80 and 135 nm RAC monolayer films, which are expected to have a similar density and substitutional yield of redox motifs as the larger RAC particles.

Following the Raman signal during the chronoamperometric reduction of the RAC films the intensity reaches a maximum and plateaus as seen in Figure 5e and Figure S12. The leveling of intensity suggests that all viologen pendants are in the radical cation state. Upon application of an oxidative potential the signal returns to baseline, again supporting there is no charge trapping in this system. The radical cation state under an inert atmosphere was found to be stable for the time scale of all experiments, so return to baseline intensity is likely not caused by interference with O2.

Normalized 1530 cm-1 Intensity

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

Langmuir

1.0 0.8 0.6 0.4 0.2 0.0 0

20

40

60

80

Time (s)

Time (s)

Figure 6. (a) 1530 cm-1 peak intensity during titration of 135 nm RAC monolayer film. Slope = -0.0083 s-1. (b) 1530 cm-1 peak intensity during titration of 80 nm RAC monolayer film. Slope = -0.01409 s-1. All error bars and baseline are representative of standard deviation for n=4 measurements taken at different locations on monolayer film. Baselines are average of signal before reductive potential is applied. Normalization was performed with respect to the maximum intensity detected for V+ in a given experiment.

7

ACS Paragon Plus Environment

Langmuir

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 8 of 11

able to follow the intensity of the 1530 cm-1 peak over time during oxidation with tip-generated FcMeOH+ (Figure 6a,b). This film oxidation can be induced by using a redox species with a sufficiently positive redox potential.2 The oxidized form of redox species, FcMeOH+, will have a large rate constant when interacting with reduced viologen due to their large ΔE0 (~700 mV). The titration process elicits a linear decay of the normalized intensity of the 1530 cm-1 peak over time for the 80 and 135 nm RAC films. A linear decay is consistent with the rate of oxidation of the film controlled by a constant flux of titrant generated at the tip. When comparing the slopes for the 80 nm and 135 nm RAC films, their ratio was 1.691 ± 0.0086. Differences in the slope of the normalized Raman intensities versus time are expected since larger particles contain a greater number of redox sites, therefore taking longer to titrate. Multiplying these slopes by the diameter of the particles yields -1.121 and -1.127 for the 135 and 80 nm RACs respectively. This agreement implies that the intraparticle concentration of redox active units is equivalent for both particle sizes. With the aid of Raman spectroscopy, we conclude that both methods of accessing charge, i.e. direct electrode contact and chemical titration, were able to confidently monitor redox processes in RACs. Localized Reduction/Oxidation Monitored by SISECM SI-SECM experiments allow us to probe redox processes within RACs, while eliminating uncertainties due to charge transfer and transport, and providing an additional route

Figure 7. (a) Reduction of a 135 nm film using HV+ with a fit to a COMSOL model. (b) Titration of the reduced film using Fc+ (blue) fit to a COMSOL model (red). CA curve with Fc before film reduction (orange) and fitting with a COMSOL model with no interaction between the film and Fc+ solution species (black).

Scheme 2: Electrochemical reactions involved in SISECM measurements on glass. Reactions (a) is the reduction of a solution species, HV2+, at the electrode probe. The reduced HV+ then reduces the RAC film as shown in reaction (b). Reactions (c) and (d) are the production of titrant species, Fc+, and subsequent reaction with the reduced viologen film

to determine the intra-particle redox concentration. For these experiments, we used a 135 nm RAC monolayer deposited on glass in order to solely evaluate SECM feedback on the particles and avoid possible lateral charge transfer from ITO. In this case, we were able to reduce the film using reduced heptyl viologen (HV+) which has an E0 that is ~ 150 mV more negative than the ethyl viologen RAC species.11 HV+ can be generated at an SECM tip from a solution HV2+ species and further react with a RAC film as shown in Scheme 2 (reactions a and b). This reaction between HV+ and V2+ in the film regenerates the original HV 2+ species that can be collected as a transient positive feedback (Figure 7a, blue curve). The reduction of a 135 nm film could be fit with our simulations for a 1.1 M film as shown in Figure 7a. Due to the large difference in E0 between HV+ and V2+ we assumed the bimolecular reaction was mass-transfer limited.

8

ACS Paragon Plus Environment

Page 9 of 11

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

Langmuir

For further evaluation of the concentration in the RAC film, we used oxidized ferrocene (Fc+) to titrate the localized region of V+ in the RAC film. After first reducing the film with HV+, we immediately biased the SECM tip for oxidation of ferrocene (Fc) to Fc+. We observed transient feedback (Figure 7b, blue curve) from the Fc+ interaction with V+ in the film (reactions c and d in Scheme 2). Without first reducing the film, we observed a typical chronoamperometric background for oxidation of Fc (Figure 7b, orange curve). Simulations for the titration process suggested that the short-lived plateau observed throughout the first 2 s in the SI-SECM transient in Figure 7b, corresponds to the titrating species being able to penetrate the film (Figure S13 and S14). This accessibility of some redox species into polymer films has been shown previously.29, 30 Interestingly, the same feature is missing in the film reduction with HV+, suggesting that this species does not permeate the film to the same extent as Fc+ (Figure S14). Through simulations of the SI-SECM of the RAC monolayer films we confirmed that the intra-particle redox concentration is near 1.0 M (Figure S15 and S16). This is in line with our single particle measurements and verifies that differences in reduction/oxidation are related to the colloid size, not concentration. Conclusion Three approaches to interrogate RACs were used to evaluate the diffusion of charge and the concentration of redox active motifs on low-dimensional assemblies of particles and on single entities. The combination of different chronocoulometric and chronoamperometric methods on single particles allowed us to preliminarily decouple concentration and diffusion contributions to the electrode response. Surface interrogation experiments using a redox mediator for titrating charge on RAC monolayers were then used as resourceful alternatives that remove uncertainties due to electrode-particle contact and help confirm the values obtained for viologen-motif concentration within RACs. Direct in situ spectroelectrochemical measurements on monolayer films indicated that 80 and 135 nm RAC particles have the same concertation of intra-particle redox active units, strongly supporting the scalability of the modular synthetic methods reported. While the calculated diffusion coefficients of these smaller RACs are slower than that observed on 1.2 µm RACs, the findings agree with previous studies of RACs.1 This encourages us to evaluate the impact of cross-linking strategies and potential differences in RAC structure triggered by the synthesis procedure for small and large RACs. Improving the diffusion coefficient in smaller RAC particles would overall have a profound impact in the rate of charge and discharge, while still retaining the desirable size-exclusion properties observed for large polymers on commercial porous separators.1,7 The use of powerful nano- and spectroelectrochemical approaches used here provides us with a strong confirmation of previous observations1 which would otherwise remain obscured

by experimental artifacts arising from solution resistance and experimental uncertainties. Insights into the reduction/oxidation of RACs were provided by Raman spectroscopy, which confirmed negligible charge trapping in RACs by conveniently using the viologen motif simultaneously as a redox and spectroscopic probe. Its combination with localized reduction/oxidation of monolayer films monitored by SI-SECM was also useful to determine that 80 nm and 135 nm RACs displayed a similar concentration of redox motifs. Finite element simulations allowed us to explore the use of SI-SECM as a useful technique for probing the full titration of RACs, thus verifying their viologen loading, but also setting an important precedent for the use of this technique in cases where imperfect contact with the electrode introduces experimental uncertainties. Our combined nano-, spectroscopic, and interrogation toolset adds new capabilities in the study of colloids and nanomaterials for charge storage, enabling insightful comparisons to their bulk behavior and bringing new opportunities to inform the formulation of design hypothesis to improve charge storage performance.

ASSOCIATED CONTENT Supporting Information. CV of ethyl viologen RAC film, IR spectrum of ion exchanged HV, line scan across single RAC particle, CV in contact with single RAC, approach curves to monolayer RAC films, diagrams of COMSOL simulations, approach curves to RACs on insulating substrate, SECM image of RAC used for CA, potential dependent spectra of RACs, 1530 cm-1 intensity during charge/discharge 80 nm RAC monolayer, SI-SECM simulations, SEM of RAC monolayer film.

AUTHOR INFORMATION Corresponding Author *E-mail: [email protected]

Author Contributions øThese

authors contributed equally.

Funding Sources This work was supported as part of the Joint Center for Energy Storage Research (JCESR), an Energy Innovation Hub funded by the U.S. Department of Energy, Office of Science, Basic Energy Sciences.

Notes The authors declare no competing financial interest.

ACKNOWLEDGMENT Sample characterization was carried out in part in the Frederick Seitz Materials Research Laboratory Central Research Facilities and the Beckman Institute for Advanced Science and Technology, University of Illinois. K.H.B. gratefully acknowledges the Arnold and Mabel Beckman Foundation for a Beckman Institute Postdoctoral Fellowship. E.C.M. acknowledges support by the Ford Foundation Fellowship Program. J.R.L. acknowledges additional support from the 2015 SACP Starter

9

ACS Paragon Plus Environment

Langmuir

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

Grant for the construction of the experimental Raman microscope-SECM setup used in this work and the Alfred P. Sloan Research Foundation. The authors thank Mr. Kevin Cheng and Prof. Jeffrey S. Moore for providing synthetic guidance and samples of RACs for the experiments reported here.

REFERENCES 1. Montoto, E. C.; Gavvalapalli, N.; Hui, J.; Burgess, M.; Sekerak, N. M.; Hernández-Burgos, K.; Wei, T.; Kneer, M.; Grolman, J.; Cheng, K. J. Redox Active Colloids as Discrete Energy Storage Carriers. J. Am. Chem. Soc. 2016, 138 (40), 13230-13237. 2. Burgess, M.; Hernández-Burgos, K.; Cheng, K. J.; Moore, J. S.; Rodríguez-López, J., Impact of electrolyte composition on the reactivity of a redox active polymer studied through surface interrogation and ion-sensitive scanning electrochemical microscopy. Analyst 2016, 141 (12), 3842-3850. 3. Dalton, E. F.; Murray, R. W., Viologen (2+/1+) and viologen (1+/0) electron-self-exchange reactions in a redox polymer. J. Phys. Chem. 1991, 95 (16), 6383-6389. 4. Pollak, K. W.; Leon, J. W.; Fréchet, J. M. J.; Maskus, M.; Abruña, H. D., Effects of dendrimer generation on site isolation of core moieties: electrochemical and fluorescence quenching studies with metalloporphyrin core dendrimers. Chem. Mater. 1998, 10 (1), 30-38. 5. Amatore, C.; Bouret, Y.; Maisonhaute, E.; Goldsmith, J. I.; Abruña, H. D., Precise Adjustment of Nanometric‐Scale Diffusion Layers within a Redox Dendrimer Molecule by Ultrafast Cyclic Voltammetry: An Electrochemical Nanometric Microtome. Chem.--Eur. J. 2001, 7 (10), 2206-2226. 6. Ghimire, G.; Yi, Y.; Derylo, M. A.; Baker, L. A.; Ito, T., Electron Propagation within Redox-Active Microdomains in Thin Films of Ferrocene-Containing Diblock Copolymers. Langmuir 2015, 31 (44), 12307-12314. 7. Nagarjuna, G.; Hui, J.; Cheng, K. J.; Lichtenstein, T.; Shen, M.; Moore, J. S.; Rodríguez-López, J., Impact of redox-active polymer molecular weight on the electrochemical properties and transport across porous separators in nonaqueous solvents. J. Am. Chem. Soc. 2014, 136 (46), 16309-16316. 8. Burgess, M.; Hernández-Burgos, K.; Simpson, B. H.; Lichtenstein, T.; Avetian, S.; Nagarjuna, G.; Cheng, K. J.; Moore, J. S.; Rodríguez-López, J., Scanning Electrochemical Microscopy and Hydrodynamic Voltammetry Investigation of Charge Transfer Mechanisms on Redox Active Polymers. J. Electrochem. Soc. 2016, 163 (4), H3006-H3013. 9. Flanagan, J. B.; Margel, S.; Bard, A. J.; Anson, F. C., Electron transfer to and from molecules containing multiple, noninteracting redox centers. Electrochemical oxidation of poly (vinylferrocene). J. Am. Chem. Soc. 1978, 100 (13), 4248-4253. 10. Wang, W.; Balasubramanian, R.; Murray, R. W., Electron transport and counterion relaxation dynamics in neat ferrocenated imidazolium ionic liquids. J. Phys. Chem. C 2008, 112 (46), 18207-18216. 11. Bird, C. L.; Kuhn, A. T., Electrochemistry of the viologens. Chem. Soc. Rev. 1981, 10 (1), 49-82. 12. Liu, B.; Blaszczyk, A.; Mayor, M.; Wandlowski, T., Redoxswitching in a viologen-type adlayer: an electrochemical shell-isolated nanoparticle enhanced Raman spectroscopy study on Au (111)-(1× 1) single crystal electrodes. ACS Nano 2011, 5 (7), 56625672.

Page 10 of 11

13. Sun, T.; Yu, Y.; Zacher, B. J.; Mirkin, M. V., Scanning electrochemical microscopy of individual catalytic nanoparticles. Angew. Chem., Int. Ed. 2014, 53 (51), 14120-14123. 14. Boika, A.; Thorgaard, S. N.; Bard, A. J., Monitoring the electrophoretic migration and adsorption of single insulating nanoparticles at ultramicroelectrodes. J. Phys. Chem. B 2012, 117 (16), 43714380. 15. Yasukawa, T.; Kondo, Y.; Uchida, I.; Matsue, T., Imaging of cellular activity of single cultured cells by scanning electrochemical microscopy. Chem. Lett. 1998, 27 (8), 767-768. 16. Melendres, C. A.; Lee, P. C.; Meisel, D., Cyclic voltammetry and in situ laser Raman spectroscopy studies of electrogenerated methyl viologen radical cation. J. Electrochem. Soc. 1983, 130 (7), 15231527. 17. Datta, M.; Jansson, R. E.; Freeman, J. J., In situ resonance Raman spectroscopic characterization of electrogenerated methyl viologen radical cation on carbon electrode. Appl. Spectrosc. 1986, 40 (2), 251-258. 18. Misono, Y.; Shibasaki, K.; Yamasawa, N.; Mineo, Y.; Itoh, K., Time-resolved resonance Raman and surface-enhanced resonance Raman scattering study on monocation radical formation processes of heptylviologen at silver electrode surfaces. J. Phys. Chem. 1993, 97 (22), 6054-6059. 19. Etienne, M.; Dossot, M.; Grausem, J. r. m.; Herzog, G. g., Combined Raman microspectrometer and shearforce regulated SECM for corrosion and self-healing analysis. Anal. Chem. 2014, 86 (22), 11203-11210. 20. Rodríguez-López, J.; Alpuche-Avilés, M. A.; Bard, A. J., Interrogation of Surfaces for the Quantification of Adsorbed Species on Electrodes: Oxygen on Gold and Platinum in Neutral Media. J. Am. Chem. Soc. 2008, 130 (50), 16985-16995. 21. Simpson, B. H.; Rodríguez-López, J., Redox Titrations via Surface Interrogation Scanning Electrochemical Microscopy at an Extended Semiconducting Surface for the Quantification of Photogenerated Adsorbed Intermediates. Electrochim. Acta 2015, 179, 74-83. 22. Kwak, J.; Bard, A. J., Scanning electrochemical microscopy. Theory of the feedback mode. Anal. Chem. 1989, 61 (11), 1221-1227. 23. Hui, J.; Burgess, M.; Zhang, J.; Rodríguez-López, J., Layer Number Dependence of Li+ Intercalation on Few-Layer Graphene and Electrochemical Imaging of Its Solid–Electrolyte Interphase Evolution. ACS nano 2016, 10 (4), 4248-4257. 24. Cornut, R.; Lefrou, C., A unified new analytical approximation for negative feedback currents with a microdisk SECM tip. J. Electroanal. Chem. 2007, 608 (1), 59-66. 25. Li, Y.; Deng, H.; Dick, J. E.; Bard, A. J., Analyzing benzene and cyclohexane emulsion droplet collisions on ultramicroelectrodes. Anal. Chem. 2015, 87 (21), 11013-11021. 26. Bard, A. J.; Mirkin, M. V., Scanning electrochemical microscopy. CRC Press: 2012. 27. Lefrou, C., A unified new analytical approximation for positive feedback currents with a microdisk SECM tip. J. Electroanal. Chem. 2006, 592 (1), 103-112. 28. Wen, B.-Y.; Yi, J.; Wang, Y.-H.; Madasamy, K.; Zhang, H.; Kathiresan, M.; Li, J.-F.; Tian, Z.-Q., In-situ monitoring of redox processes of viologen at Au (hkl) single-crystal electrodes using electrochemical shell-isolated nanoparticle-enhanced Raman spectroscopy. Electrochem. Commun. 2016, 72, 131-134. 29. Kudo, K.; Nagaoka, M.; Kuniyoshi, S.; Tanaka, K., Optical and electrical evaluation of adsorption Langmuir-Blodgett films of C

10

ACS Paragon Plus Environment

Page 11 of 11

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

Langmuir

18 TCNQ/TMPD and C 18 TCNQ/PPD. Synthetic Met. 1995, 71 (1), 2059-2060.

30. Buttry, D. A.; Anson, F. C., Electron hopping vs. molecular diffusion as charge transfer mechanisms in redox polymer films. J. Electroanal. Chem. Interfacial Electrochem. 1981, 130, 333-338.

SYNOPSIS TOC

11

ACS Paragon Plus Environment