Continuous and Reversible Tuning of ... - ACS Publications

Subscriber access provided by YORK UNIV

Article

Continuous and Reversible Tuning of Electrochemical Reaction Kinetics on Back-Gated 2D Semiconductor Electrodes: Steady-State Analysis using a Hydrodynamic Method Chang-Hyun Kim, Yan Wang, and C. Daniel Frisbie Anal. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.analchem.8b05216 • Publication Date (Web): 20 Dec 2018 Downloaded from http://pubs.acs.org on December 22, 2018

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 27 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

Analytical Chemistry

Continuous and Reversible Tuning of Electrochemical Reaction Kinetics on Back-Gated 2D Semiconductor Electrodes: Steady-State Analysis using a Hydrodynamic Method Chang-Hyun Kim,† Yan Wang,†,‡ and C. Daniel Frisbie†,* †Department of Chemical Engineering and Materials Science, University of Minnesota, 421 Washington Avenue SE, Minneapolis, Minnesota 55455, United States ‡Department of Chemistry, University of Minnesota, 207 Pleasant Street SE, Minneapolis, Minnesota 55455, United States

*Corresponding Author. Phone: 612-625-0779 / E-mail: [email protected] Author Contributions. The manuscript was written through contributions of all authors. All authors have given approval to the final version of the manuscript.

ACS Paragon Plus Environment

1

Analytical 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 27

ABSTRACT. Here we report the steady state kinetic analysis of field-effect-controlled outer-sphere electrochemistry on ultrathin back-gated ZnO working electrodes (i.e., 5 nm ZnO electrodes prepared on SiO2/degenerate Si back gates). To achieve steady state conditions in the electrolyte phase, gate-tunable electrochemical flow cells were prepared by integrating a silicone microfluidic channel on the back-gated ZnO electrode. In these flow cells, continuous supply of fresh electrolyte generates time-invariant diffusion layers near the ZnO surface, allowing steadystate kinetic analysis as in other hydrodynamic methods. From the steady-state analysis, it was found that the electron density on the ZnO surface increases with the voltage bias, VBG, applied to the back gate, while the rate constant for electron transfer decreases with VBG. The observed trend can be explained as a result of the field-effect-induced band alignment shift at the ZnO/electrolyte interface which is predicted by our conceptual model; a positive back gate bias shifts the conduction band edge down at a given working electrode potential, leading to an increased surface electron density on ZnO, but simultaneously less overlap of the band edge with the electron acceptor states in solution, which means a lower electron transfer rate constant. Overall, the results quantitatively demonstrate that back gates and the ensuing field effect can be used to control kinetics of interfacial electron transfer at two-dimensional (2D) semiconductor electrodes.


2

Page 3 of 27 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


Table of Contents (TOC) Graphic


3


Page 4 of 27

Introduction Recently, we have demonstrated continuous and reversible electrostatic tuning of outersphere redox reaction kinetics on ultrathin, back-gated semiconductor electrodes (note: the term back gate is adopted from the field-effect transistor (FET) literature).1,2 In the initial reports, the redox reaction on a two-dimensional (2D) electrode was found to be gate-tunable, in that the reaction rate on the front side of the electrode was freely modulated - at a given electrochemical potential - by a voltage bias applied to the back gate, Figure 1. The basic idea of using such gatetunable electrodes is to electrostatically induce extra charge carriers into the 2D electrode to modify the electrode/electrolyte interface. Because of the extreme thinness of the 2D electrode, the gate-induced charge carriers are accessible to the opposite (front) face of the electrode and thus alter the electric double layer (EDL) structure as well as electronic occupation at the electrode surface. In the previous studies using cyclic voltammetry (CV), we could successfully show that the onset potential of redox reactions can be modulated on gate-tunable electrodes. However, we could not fully clarify how the kinetic parameters of the redox reactions are influenced by the applied back gate bias because the current-to-voltage responses obtained from CV were not simple enough to interpret with well-known kinetic analysis techniques3–7, such as Nicholson’s method.3


4

Page 5 of 27 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


Figure 1. Schematic representation of the outer-sphere electron transfer reaction between redox species (TBBQ) and a gate-tunable ultrathin ZnO electrode used in this study. Note that a voltage bias VBG is applied to the back gate to control reaction kinetics on the ZnO surface. The electrolyte is 0.1 M THAP in acetonitrile. The symbols indicated in the scheme are as follows: working electrode potential (Vw), sourceto-drain bias (VSD), and back gate bias (VBG).

In the current investigation, we have introduced a more advanced electrochemical cell in which a microfluidic channel is integrated over the gate-tunable electrode. With this electrochemical flow cell, we can perform steady-state analysis of reaction kinetics on the gatetunable electrode because continuous supply of fresh electrolyte under forced convection leads to time-invariant diffusion layers formed near the electrode surface.8–10 Like other electrochemical analysis techniques based on hydrodynamic methods,11–18 such as the rotating disk electrode,7,11 our flow cell provides several important advantages compared to CV including: (1) simpler and more accurate analysis of charge transfer kinetics because the mass transport contribution to the overall electrode polarization is minimized under forced convection; (2) elimination of doublelayer charging current under steady-state conditions so that the measured current is due exclusively to redox reaction. Herein, with the gate-tunable electrochemical flow cells and a rate law based on the Gerischer model,19,20 we investigated how the charge transfer kinetics between an outer-sphere redox species, 2,3,5,6-tetrabromo-1,4-benzoquinone (TBBQ), and an ultrathin ZnO film are


5


Page 6 of 27

modulated by the back gate bias. Our results indicate that a positive back gate bias leads to both an increased electron density on the ZnO surface and a reduced rate constant for electron transfer at a given electrochemical potential. In terms of the net reaction current these two effects compete against each other, but overall, the overpotential to attain a given reaction current decreases at positive back gate bias because the electron density is more susceptible to the back gate bias than the rate constant. Consistent with the prediction of our conceptual model, these experimental results confirm that the continuous and reversible change of charge transfer kinetics on ZnO surfaces is attributed to the gate-induced band alignment shift at the ZnO/electrolyte interface. An appealing point of the gate-tunable electrodes is that gating is basically an electrostatic charging process which, in principle, requires no extra electrical work to maintain the charge induced in the electrode.21 Therefore, field effect modulation can be a potential means to improve energy efficiencies of electrochemical processes on 2D semiconductor electrode surfaces. Furthermore, electrochemical processes on a gate-tunable electrode can be continuously and reversibly controlled without changing the chemical composition of the electrode. This, in turn, allows one to investigate systematically how thermodynamics and/or kinetics of the electrode processes are determined by the electronic structure at the electrode/electrolyte interface while excluding chemical effects. In this regard, we believe that the gated electrode design described in this report will serve as a very useful platform to investigate and optimize electrochemical interfaces in various electrochemical systems including liquid junction solar cells,22–27 and (photo)electrocatalysts.28–35


6

Page 7 of 27 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


Experimental Section The fabrication procedure of gate-tunable ZnO electrodes (before integrating microfluidic channels on them) is thoroughly described elsewhere.1 Briefly, a 5 nm ZnO film was grown on a SiO2/p-Si substrate (∼300 nm-thick oxide thermally grown on 525 μm thick, highly boron-doped silicon wafer) by atomic layer deposition (ALD),36 and a series of annealing, photolithography, and wet etching steps were performed to pattern the ZnO film into an array of microband electrodes with desired dimensions. Then a series of photolithography and electron beam evaporation steps were followed to prepare Au/Ti metal contacts (30 nm thick Au top layer on 3 nm thick Ti adhesion layer) and a SiO2 passivation layer that prevents electrochemical reaction from occurring on the metal contacts. Similarly, Au microband electrodes were prepared on a SiO2/p-Si wafer by a series of photolithography and electron beam evaporation steps (3 nm Ti and 30 nm Au, respectively). Figures 2a and 2b show Au band (50 μm wide) and ZnO (60 μm x 60 μm) electrodes prepared for this study. For the hydrodynamic steady-state kinetic analysis, a microfluidic channel was prepared in poly(dimethylsiloxane) (PDMS) and integrated on an Au or a ZnO electrode as shown in Figure 2c. The width and the height (d and 2h in Figure 2d) of the microfluidic channel were 200 μm and 50 μm, respectively. The detailed procedures to prepare and integrate the PDMS channel onto the Au or ZnO electrodes are described in the Supporting Information (SI).


7


Page 8 of 27

Figure 2. Optical images of (a) Au and (b) gate-tunable ZnO electrodes underneath a PDMS microfluidic channel. (c) Structure of the gate-tunable electrochemical flow cell with Au or ZnO electrode. (d) Geometry of the microfluidic channel (width d and height 2h) and active electrode area (length xe and width w). (e) Electrical configuration and electrolyte solution streams connected to the flow cell. The symbols indicated in the scheme are as follows: WE potential (Vw), source (Is) and drain (Id) current, source-to-drain bias (VSD), and back gate bias (VBG).

To perform electrochemical measurements while applying variable back gate biases, we configured the instrumental setup as schematically shown in Figure 2e. In this study, TBBQ and tetrahexylammonium hexafluorophosphate (THAP) were used for the outer-sphere redox species and the supporting electrolyte, respectively. 0.5-2.0 mM TBBQ and 0.1 M THAP in acetonitrile (MeCN) solution were supplied to the microfluidic channel using a syringe pump (Model 00323VE, Cole-Parmer). An Ag/Ag+ reference electrode (BASi Co., filled with 10 mM AgNO3 and 0.1 M THAP in MeCN) was plugged to a Luggin capillary filled with 0.1 M THAP in MeCN, and the tip of the capillary was placed at the outlet of the microfluidic channel. A platinum wire used for the counter electrode was placed ~3−5 mm away from the outlet of the microfluidic channel. Note that, if not specified, 𝑉𝑆𝐷 in Figure 2e was set to 0 V. The reaction current 𝐼𝑤 at the grounded working electrode (WE) was obtained from the sum of the currents on source and drain (i.e., 𝐼𝑆 and 𝐼𝐷 in Figure 2e, respectively). Throughout this study, 𝑉𝑤 was corrected by the “IR compensation” technique,37 which subtracts the potential drop due to ohmic resistance between


8

Page 9 of 27 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 WE and the reference electrode (RE) (see SI for details). For ZnO electrodes, 𝑉BG was applied to the back gate (i.e., p-Si) to induce the field effect. The complete details of the electrical connections to multiple source meters for electrochemical measurements can be found in the previous report.1 All electrochemical measurements were performed at room temperature in ambient conditions and the data were collected with an in-house LabVIEW program. The hydrodynamic and mass transport behaviors in our electrochemical cells are described by the mathematical model reported by Compton et al.,9,38 which was developed to analyze quasireversible redox reactions with the form: Ox + 𝑒 ⇄Red

(1)

in a “high-speed channel flow cell”, where both oxidized (Ox) and reduced (Red) species are kinetically stable on the timescale of the experiment, and the bulk solution contains only Ox before it enters the flow cell (see SI for details). All experiments in this work were conducted under laminar flow conditions with a negligible horizontal diffusion effect; i.e., dimensionless 𝑉𝑓

3𝑉𝑓𝑥𝑒2

parameters Re = ℎ𝜈 and Pe = 2𝑑𝔇

2 𝑜𝑥ℎ

, where 𝑉𝑓 is the volumetric flow rate of the feed solution, ν is

the kinematic viscosity of the feed solution, 𝔇ox is the diffusion coefficient of Ox, and the other terms represent the geometrical parameters given in Figure 2d, were kept in 0.93−14.9 and 1.72 × 103−2.76 × 104 ranges, respectively, where the adopted mathematical model is valid.

Results & Discussion Steady-State Voltammetry of TBBQ on Gold Electrodes: To confirm the validity of the experimental setup and the mathematical model used for this study and to measure the limiting


9


Page 10 of 27

currents that are required to correct the mass transport effect on ZnO electrodes in the following section, we first performed a hydrodynamic analysis of TBBQ reaction on gold microband electrodes. It is assumed the redox reaction follows Butler-Volmer kinetics with the forward and backward rate constants (𝑘𝑓 and 𝑘𝑏, respectively) given by:

(

)

(2)

((1 ―𝑅𝑇𝛼)𝐹(𝑉𝑤 ― 𝑈0 )) = 𝑘0exp((1 ― 𝛼)𝜃)

(3)

𝛼𝐹

𝑘𝑓 = 𝑘0exp ― 𝑅𝑇(𝑉𝑤 ― 𝑈0 ) = 𝑘0exp( ―𝛼𝜃) 𝑘𝑏 = 𝑘0exp

′

′

where 𝛼 is the transfer coefficient, 𝑈0′ is the formal potential of the redox couple ( ―0.286 V vs. 𝐹

Ag/Ag+), 𝜃 is a dimensionless potential (𝜃 = 𝑅𝑇(𝑉𝑤 ― 𝑈0′)), 𝑘0 is the standard rate constant, R is the ideal gas constant, and T is the temperature. Figures 3a and 3b show that the steady-state voltammograms of TBBQ reduction, which were obtained at different flow rates (𝑉𝑓) and bulk concentrations of Ox (𝑐ox ∗ ), match well with the curves simulated with Equations S3-S5, 2, and 3 assuming 𝑘0 = 0.3 cm/s, 𝛼 = 0.5, and 𝔇ox = 𝔇red ≈ 1.4 × 10 ―5 cm2/s (note: Ox and Red correspond to TBBQ and TBBQ●−, respectively). The parameters used to fit the experimental data are consistent with the previously reported values (𝑘0 = 0.3 cm/s, 𝛼 = 0.5, and 𝔇ox = 𝔇red = 1.25 × 10 ―5 cm2/s).10 The insets of Figures 3a and 3b show that the limiting current 𝐼𝑤,𝑙𝑖𝑚 depends linearly on 𝑉𝑓1/3 and 𝑐ox ∗ , as predicted in Equation S6. From these results, we confirmed (1) a stable, time-invariant diffusion layer is formed near the electrode in our devices in the given 𝑉𝑓 range, (2) TBBQ●− radical formed by reduction of TBBQ is kinetically stable over the timescale of the experiment and is not further reduced to TBBQ2 ― as long as 𝑉𝑤> ―0.7 V. We observed that sweeping 𝑉𝑤 to more negative values leads to an insoluble organic film formed on the gold electrode and large hysteresis in the voltammograms. These


10

Page 11 of 27 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


unfavorable behaviors become more severe when tetrabutylammonium hexafluorophosphate (TBAP), which has smaller cation than THAP, was used for the supporting electrolyte. It is believed that TBBQ2 ― reacts with cations in the supporting electrolyte to form an insoluble film on the electrode, which inhibits further outer-sphere electron transfer, like other quinone-type molecules.39

Figure 3. (a) Steady state voltammograms of 2 mM TBBQ solution on an Au band electrode at different flow rates 𝑉𝑓. The inset of (a) shows the limiting current 𝐼𝑤,𝑙𝑖𝑚 vs 𝑉𝑓1/3. (b) Steady state voltammograms of TBBQ solution on an Au band electrode with different concentrations 𝑐ox ∗ and the flow rate fixed at 10 μL/min. The inset of (b) shows the limiting current 𝐼𝑤,𝑙𝑖𝑚 vs 𝑐ox ∗ . The symbols and the lines in (a) and (b) indicate the experimental data and the simulated curves, respectively. For both panels the supporting electrolyte is 0.1 M THAP in acetonitrile.

Steady-State Voltammetry of TBBQ on Gate-Tunable ZnO Electrodes: In this study, we adopted the Gerischer model19,20 (see SI for more detail) to analyze the charge transfer process between the outer-sphere redox species (TBBQ) and the gate-tunable ZnO semiconductor electrodes. The energy diagram in Figure 4 schematically represents the ZnO/electrolyte interface, where equimolar Ox and Red species are dissolved in the electrolyte and the applied 𝑉𝑤 is slightly more negative than the formal potential 𝑈0′ (i.e., 𝑉𝑤 =

𝐸𝐹 ― 𝐸ref ―𝑒

< 𝑈0 = ′

𝐸0′ ― 𝐸ref ―𝑒

where 𝐸𝐹 and 𝐸ref


11


Page 12 of 27

represent the Fermi levels of the WE and RE, respectively). While the left side of the diagram represents the electronic states of an ideal ZnO surface which consists of the conduction band (CB) and the valence band (VB) separated by the bandgap energy (Eg=~3.2 eV for ZnO), the right side of the diagram represents the donor (TBBQ●−) and the acceptor (TBBQ) states (corresponding to Red and Ox, respectively) at the electrode/solution interface (namely, outer Helmholtz plane, OHP). The Gerischer model explains that the energy difference between the donor and the acceptor states is due to the solvent reorganization energy λ (generally in the range of 0.5-2 eV), and that their Gaussian-type energy distribution is a consequence of the thermal fluctuation of the solvation shell.40 Since charge transfer rate at a given E is essentially determined by the overlap of occupied and empty states between the two phases, the cathodic and anodic current densities (𝑗𝑤― and 𝑗𝑤+ , respectively) on the ZnO surface can be described as follows (see SI for details):

{

[𝐸𝑐 ― (𝐸0 + 𝜆)]2

{

[𝐸𝑐 ― (𝐸0 ― 𝜆)]2

𝑗𝑤― ≈ ―𝑒𝑘et,max exp ―

𝑗𝑤+

≈ 𝑒𝑘et,max exp ―

}

′

4𝑘𝐵𝑇𝜆 ′

4𝑘𝐵𝑇𝜆

}

𝑛 𝑐ox = ―𝑒𝑘et― 𝑛s 𝑐ox

(4)

s

𝑝 𝑐red = 𝑒𝑘et+ 𝑝s 𝑐red

(5)

s

where 𝑘et’s are the rate constants (cm3 s-1) for cathodic and anodic charge transfer, and 𝑛s and 𝑝s are the densities (cm-2) of electrons and holes on the 2D ZnO electrode, respectively. Note that 𝐸𝐹 is assumed to be more than 2𝑘𝐵𝑇 below CB edge (𝐸𝐶) when deriving these equations. Considering 𝑛s~𝑓(𝐸𝑐)𝐷(𝐸𝑐) and 𝑝s~(1 ― 𝑓(𝐸𝑐))𝐷(𝐸𝑐), where 𝑓(𝐸𝑐) is the value of the Fermi function at 𝐸𝑐 and 𝐷(𝐸𝑐) is the density of states at 𝐸𝑐, we obtain:


12

Page 13 of 27 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


| | [ 𝑗𝑤― 𝑗𝑤+

𝐸𝐹 ― 𝐸0 𝑐ox ≈ exp 𝑘𝐵𝑇 𝑐red ′

]

(6)

Note that, because 𝑐ox/𝑐red ≫ 1 in our system, |𝑗𝑤― /𝑗𝑤+ | will be much greater than 1 as long as 𝐸𝐹 ′ ′ ― 𝐸0 ≫ 𝑘𝐵𝑇 (i.e., 𝑉𝑤 ― 𝑈0 ≪ ―26 mV). Therefore, the anodic current density 𝑗𝑤+ is negligible

compared to the total current density (𝑗𝑤 = 𝑗𝑤― + 𝑗𝑤+ ) and 𝑗𝑤 ≈ 𝑗𝑤― . Finally, note also that when 𝐸𝐹 lies within the bandgap, 𝑛s in Equation 4 can be approximated as:

[

𝑛s = 𝑁𝑐exp ―

]

(𝐸𝑐 ― 𝐸𝐹) 𝑘B𝑇

(7)

where 𝑁𝑐 is the effective DOS at the lower edge of the CB.

Figure 4. Schematic representation of the DOS and electronic occupation at the surface of the ZnO WE (left) and in the electrolyte solution containing equimolar TBBQ and TBBQ●− (right). The symbols used in this diagram are as follows: CB edge (Ec) and VB edge (Ev) of ZnO; Fermi levels of ZnO WE (EF) and RE (Eref); CB edge offset (δ) from EF in ZnO; electrochemical potential (vs reference) of the ZnO WE (Vw); energy level corresponding to the formal potential of TBBQ/TBBQ●− (E0’); and the solvent reorganization energy (λ). The red arrows indicate the electron transfer across the interface. Note that the electron transfer from CB to TBBQ is fast (expressed as a solid line) because the overlap between the filled (blue shadow in CB) and the empty states (TBBQ) is significant, while the electron transfer from TBBQ●− to CB is negligible (expressed as a dotted line) because there are not acceptor states available in ZnO that can receive electrons from TBBQ●−.


13


Page 14 of 27

The mass transport effect in electrolyte can be corrected using the well-known KouteckýLevich equation (see SI for details) as follows7,41:

(

)

1 1 𝑗𝑤,k = ― 𝑗𝑤 𝑗𝑤,𝑙𝑖𝑚

―1

= ―𝑒𝑘et― 𝑛s 𝑐ox ∗

(8)

Here 𝑗𝑤 is the measured reaction current, 𝑗𝑤,𝑙𝑖𝑚 is the mass transfer-limited reaction current, and 𝑗𝑤,k is pure reaction current absent mass-transfer limitations. Note also that 𝑗𝑤,k can be viewed as the reaction current density when the mass transport rate of Ox is extremely fast such that 𝑐ox equals 𝑐ox ∗ . Importantly, as shown in Equation S6, the limiting current density 𝑗𝑤,𝑙𝑖𝑚( =

𝐼𝑤,𝑙𝑖𝑚

𝑥 𝑒w )

is

determined not by electrode materials but by a given set of 𝑐ox ∗ , 𝔇ox, 𝑉𝑓, and the geometry of the flow cell. Therefore, in this study, 𝑗𝑤,𝑙𝑖𝑚 measured on the Au electrodes is used to correct the mass transport effect on the ZnO electrodes. The maximum error that can be potentially introduced by this correction method is less than 2% (see SI for details). To identify the relationship between 𝑗w,k and 𝑐ox ∗ , and thus to extract 𝑘et― via Equation 8, we first obtained the steady-state 𝑗w vs. 𝑉w curves on ZnO electrodes at four different 𝑐ox ∗ of TBBQ while keeping constant 𝑉BG = 0 V and 𝑉𝑓 = 10 𝜇L/min. Figure 5a shows apparently twostep reduction of TBBQ with onset potential 𝑉w,on at ―0.2 V and ―0.45 V. Considering 𝑉w,on for an Au electrode is ―0.2 V (Figure 3), the apparent first reduction at ―0.2 V is believed to come from the Au/Ti metal contacts to the ZnO electrode (most likely from their edges which are not fully passivated with the SiO2 film). The reduction current on the exposed Au should reach a constant value when 𝑉w< ―0.4 V because mass transport limiting condition is met in such a negative 𝑉w range (Figure 3). Therefore, the reaction current 𝑗w on ZnO electrode should be


14

Page 15 of 27 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


obtained by subtracting the constant background current (indicated as the dashed base line in Figure 5a for 𝑐ox ∗ = 0.5 mM case) from the total current for each 𝑐ox ∗ . It is noteworthy that the background current may not be merely proportional to 𝑐ox ∗ , as is in Figure 5a, because the area of exposed Au in each device can be very different from one another. Importantly, the Figure 5a data show that 𝑗w is reaction limited; the 𝑗w vs. 𝑉w response for the ZnO electrode does not plateau as it does for gold band electrodes in Figure 3b, but rather continues to increase as 𝑉w becomes more negative. The inset shows the exponential dependence of the mass-transfer corrected reaction current 𝑗w,k on 𝑉w, typical of Butler-Volmer kinetics. Figure 5b shows the good linear relationship between 𝑗w,k and 𝑐ox ∗ in the 𝑉w range from ―0.5 V to ―0.7 V, which is consistent with the rate law proposed in Equation 8.

Figure 5. (a) Steady state voltammograms (𝑗w vs 𝑉𝑤) of TBBQ solution on gate-tunable ZnO electrode (where back gate bias 𝑉BG=0 V) with different concentrations 𝑐ox ∗ and flow rate fixed at 10 μL/min. The inset of (a) shows the mass transport corrected reaction current density 𝑗w,k (in a log scale) at different 𝑐ox ∗ and 𝑉𝑤. (b) 𝑗w,k (in a linear scale) at different 𝑐ox ∗ and 𝑉𝑤. The supporting electrolyte is 0.1 M THAP in acetonitrile.


15


Page 16 of 27

To investigate how the charge transfer kinetics between TBBQ and a ZnO electrode is affected by the transverse field effect (i.e., the back gate), 𝑗w vs. 𝑉w curves were obtained at different 𝑉BG while maintaining constant 𝑐ox ∗ and 𝑉𝑓 at 0.5 mM and 10 𝜇L/min, respectively, (Figure 6a). We observed that the reaction current for TBBQ reduction (i.e., 𝑗w) at a given 𝑉w increases at a positive 𝑉BG and decreases at a negative 𝑉BG. In other words, the electrode polarization 𝜂 = 𝑉w ― 𝑈0′ to generate a given 𝑗w systematically decreases with 𝑉BG. This is a central result of this study.

Figure 6. (a) Steady state voltammograms (𝑗w vs 𝑉𝑤) of 0.5 mM TBBQ solution on gate-tunable ZnO electrodes at different back gate biases 𝑉BG with flow rate 𝑉𝑓 fixed at 10 μL/min. The inset in (a) shows how overall polarization (filled symbols) and in-plane polarization (open symbols) of the ZnO electrode at given current densities 𝑗w change with 𝑉BG. (b) Sheet conductance of ZnO electrode (inset in a log scale) simultaneously obtained with the data in (a).

As we demonstrated in our previous report1 that gate-induced charge in ultrathin ZnO electrodes is fully accessible at the front face of the electrode, we attribute the reduction of 𝜂 to improved charge transfer kinetics at the ZnO surface. However, there is another possibility for the reduced 𝜂 and that is the reduction of ohmic resistance in the ZnO film as VBG increases. The gatetunable ZnO working electrodes are prepared on insulating SiO2 rather than on a metallic current


16

Page 17 of 27 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


collector, and thus in principle the effect of in-plane polarization (due to ohmic resistance) to overall reaction rate can be significant and should also be taken into account. To separate in-plane polarization due to ohmic resistance from out-of-plane polarization due to charge-transfer kinetics, we performed in-situ monitoring of the sheet conductance of the ZnO film by applying a small and constant 𝑉SD during electrochemical measurements (Figure 6b, see SI for experimental details). The inset of Figure 6a reveals that the 𝑉BG-induced change in the electrode polarization is mostly attributed to the out-of-plane polarization (e.g., up to 0.2 V change for 𝑗w = ―10 𝜇A/cm2 when 𝑉BG changes from −60 to 90 V) rather than in-plane polarization (e.g., up to 0.04 V change for the same 𝑗w and 𝑉BG range). Therefore, the systematic changes of 𝑗w vs. 𝑉w curves at different 𝑉BG in Figure 6a are not attributable to the 𝑉BG-induced change in ohmic resistance of the electrode but indeed reflect the 𝑉BG-induced change in charge transfer kinetics at the electrode/electrolyte interface. With the Figure 6 data we are now in position to assess the quantitative impact of 𝑉BG on the kinetic parameters 𝑛s and 𝑘et― . Figure 7a shows the mass-transfer corrected reaction current 𝑗w,k as a function of 𝑉w, calculated from 𝑗w in Figure 6a using Equation 8. Note that 𝑗w,k could not be obtained for 𝑉w> ―0.4 V range because the background current from the non-passivated Au is not constant and 𝑗w cannot be corrected in this regime. The electron density 𝑛s on the ZnO electrode shown in Figure 7b was estimated from the sheet conductance in Figure 6b (see SI for details). The heterogeneous charge transfer rate constant 𝑘et― in Figure 7c was obtained from the quotient of 𝑗w,k and ―𝑒𝑛s𝑐ox ∗ at each data point in Figure 7a and 7b (see Equation 8). The results reveal that 𝑛s and 𝑘et― at a fixed 𝑉BG generally increase with ―𝑉w. The apparent exponential dependence of 𝑛s on ―𝑉w is an expected behavior for an n-type semiconductor electrode, consistent with


17


Page 18 of 27

Equation 7. Importantly, at fixed 𝑉w, 𝑛s increases by ~2 orders of magnitude with 𝑉BG changing from -60 V to 90 V, while 𝑘et― generally decreases (down to 5% of initial value) at 𝑉w = ―0.68 V over the same gate voltage range. 𝑉BG has a major impact on both 𝑛s and 𝑘et― . In terms of the reaction current, the trends in 𝑛s and 𝑘et― oppose one another. However, since 𝑛s is more susceptible to 𝑉BG than 𝑘et― , the reaction current density 𝑗w,k at a fixed 𝑉w increases with 𝑉BG (Figure 7a).

Figure 7. (a) Mass transport corrected reaction current density 𝑗w,k plotted vs. 𝑉𝑤 for different VBG. (b) Surface electron density ns plotted vs 𝑉𝑤 for different VBG. (c) Rate constant 𝑘et― for TBBQ reduction on ZnO plotted vs 𝑉𝑤 for different VBG. VBG values in (a-c) are adjusted from −60 to 90 V with 30 V intervals. Note that all the data in this figure are calculated from the data shown in Figure 6.

The observed trends of 𝑛s and 𝑘et― at different 𝑉w and 𝑉BG can be better understood with the schematic energy diagrams shown in Figure 8, which show a specific case where 𝑉w is fixed. In creating these diagrams, we assumed that the 5 nm thick ZnO is essentially a 2D semiconductor in which the energy band bending within the electrode can be ignored and the EDL is described as a Helmholtz capacitor. Figure 8a depicts the initial state with 𝑉𝐵𝐺=0 V, where δ0 represents the initial offset of the ZnO CB edge from 𝐸𝐹. Figure 8b corresponds to the application of 𝑉𝐵𝐺 > 0 V while 𝑉w is fixed. In this case, the field effect shifts the ZnO CB edge closer to 𝐸𝐹; the offset is


18

Page 19 of 27 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


now δ< δ0 and electron accumulation in the thin ZnO layer has occurred. Note that the 𝑉BG induces charge not only in the ZnO electrode but also in the EDL. The vacuum level shift in the EDL ∆ 𝜙EDL due to 𝑉BG-induced charge leads to the band alignment shift at the electrode/electrolyte interface. In our previous report,1 the CB edge offset with respect to 𝑉BG at a fixed 𝑉w was derived from charge and energy balances as follows:

( ) ∂𝛿 ∂𝑉𝐵𝐺

𝑉𝑤

―𝑒

= 1+

𝐶EDL 𝐶BG

+

𝐶𝑤(𝛿)

(9)

𝐶BG

where 𝐶BG and 𝐶EDL are the back-gate and double-layer capacitances, respectively, and 𝐶𝑊(𝛿) is the quantum capacitance of the electrode given by:

𝐶𝑊(𝛿) = 𝑒

𝑑𝑄𝑊(𝛿) 𝑑𝛿

(10)

𝑄𝑊(𝛿) is the charge induced in the working electrode as a function of the Fermi level offset 𝛿 = 𝐸𝑐 ― 𝐸𝐹. Note that 𝐶𝑊(𝛿) essentially represents the DOS of the WE in capacitance units, which is a function of . When 𝐸𝐹 lies within the bandgap, 𝐶𝑊(𝛿) essentially becomes zero if there are no in-gap surface trap states near 𝐸𝐹. Considering the estimated 𝐶BG (~20 nF/cm2) is much smaller than the estimated 𝐶EDL (~10 μF/cm2), Equation 9 reduces to:

( ) ∂𝛿 ∂𝑉BG

≈ ―𝑒 𝑉w

𝐶BG 𝐶EDL

≈ ―0.002 𝑒

(11)

Therefore, the band alignment shift that can be achieved with 𝑉BG varying between −60 V to 90 V is estimated to be Δ𝛿 = 𝛿 ― 𝛿0 ≈ −0.3 eV. Importantly, Δ𝛿 can be increased by preparing a gate dielectric layer with a higher dielectric constant or a reduced thickness. For example, if we use 50


19


Page 20 of 27

nm HfO2, which has 4-6 times higher dielectric constant than SiO2, we can achieve Δ𝛿 in excess of 1 eV.

Figure 8. Energy diagrams of a back-gated 2D ZnO electrode in electrolyte (a) before and (b) after a positive VBG is applied to the back gate while VW is fixed. The red arrows indicate electron transfer processes. The additional symbols used in the diagram, which are not shown in Figure 4, are as follows: local vacuum level (Evac); Fermi level of back gate (EBG); work functions of back gate (ΦBG) and RE (Φref); electron affinity of ZnO (χ); vacuum level shifts in SiO2 (∆𝜙BG) and EDL (∆𝜙EDL); energy distribution function of the oxidized species (Dox). Figure 8 also schematically shows how this VBG-induced band alignment shift influences the charge transfer kinetics at the ZnO/electrolyte interface. In these energy diagrams, it is assumed that the maximum of the energy distribution function of TBBQ (𝐷ox(𝐸)), which occurs at 𝐸0′ +𝜆 (Equation S9), is above the CB edge in most cases, based on the fact that 𝜆 = ~0.76 eV for TBBQ.10 Note that only 𝐷ox(𝐸) is illustrated in the diagrams because, practically, anodic reaction does not occur in our system. The DOS of the electrode 𝐷(𝐸) can be shifted freely by 𝑉BG, but 𝐷ox (𝐸) is not subject to the field effect because the electric field from the back gate is completely screened outside the Helmholtz plane by the abundant mobile ions in the electrolyte. As a result, 𝑉BG can lead to a significant change in 𝑛s and 𝑘et― at a given 𝑉w. In Figure 8b, the CB edge shift to


20

Page 21 of 27 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


a lower energy level gives rise to an increased 𝑛s (by Equation 7) and a reduced energy state overlap (i.e., smaller 𝑘et― ) between the 𝐷(𝐸) and 𝐷ox(𝐸) near the CB edge. This is consistent with the general trends observed in Figures 7b and 7c. Lastly, it is noteworthy that the proposed model assuming an ideal n-type semiconductor electrode catches the essence of our observations but does not fully explain several behaviors including the response of 𝑘et― to 𝑉w. For an ideal n-type semiconductor the CB edge does not move with 𝑉w as the effect of 𝑉w is simply to change 𝛿. That is, the CB edge of an ideal n-type semiconductor at a fixed 𝑉BG is essentially pinned with respect to the reference energy 𝐸ref. This means that to a first approximation 𝑘et― in Equation 4 should not be a function of 𝑉w, which is contrary to our observation in Figure 7c. This deviation is most likely attributed to in-gap surface states of the ZnO electrodes prepared in this study. According to the conceptual model proposed in our previous report,1 the variation in the CB edge position with respect to 𝑉w at a fixed 𝑉BG is given by:

( ) ∂𝐸𝑐

∂𝐸𝐹

𝑉BG

=

𝐶BG + 𝐶𝑤(𝛿) 𝐶EDL + 𝐶BG + 𝐶𝑤(𝛿)

(12)

If 𝐸𝐹 lies within the bandgap and there are no in-gap surface states in the energy range, Equation 12 becomes zero because 𝐶𝑊(𝛿) is zero and 𝐶BG is much smaller than 𝐶EDL. In other words, in an ideal n-type semiconductor having no surface states in the bandgap, 𝐸𝑐 is essentially pinned with respect to the reference potential 𝐸ref when 𝑉𝑊 (and thus 𝐸𝐹) is swept at a fixed 𝑉BG. In reality, the ZnO film grown by ALD is generally amorphous or polycrystalline, and has intrinsic surface states due to dangling bonds at the surface or grain boundaries.42 If the ZnO electrode has a significant number of surface states so 𝐶𝑤(𝛿) is comparable to 𝐶EDL, Equation 12 has a positive value and


21


Page 22 of 27

thus a positive sweep of 𝐸𝐹 (i.e., a negative sweep of 𝑉w) pushes up the CB edge to a higher energy level (i.e., 𝐸𝐶 moves proportional to 𝐸𝐹). This explains the trend observed in Figure 7c; 𝑘et― increases with ― 𝑉w as long as 𝐸𝑐 is below the energy maximum of 𝐷ox(𝐸) located at 𝐸0′ +𝜆 = 1.05 eV vs. 𝐸ref. The presence of the surface states is also supported by the trend in 𝑛s with 𝑉w. According to Equation 7, ― 𝑉w (=𝐸𝐹/𝑒) should show ~60 mV increase per decade of 𝑛s when 𝐸𝑐 is pinned (see Section S6 for details). Therefore, the ~150 mV increase per decade observed in Figure 7b suggests that a significant density of surface states on ZnO surface comes into play at the ZnO/electrolyte interface, which causes 𝐸𝑐 to shift up during a positive sweep of 𝐸𝐹.

Conclusion In summary, we investigated continuous and reversible modulation of outer-sphere charge transfer kinetics on gate-tunable ultrathin ZnO electrodes, in which the charge densities on the ZnO surface and the EDL are controlled by a voltage bias applied to the back gate, as well as by its electrochemical potential set independently. A microfluidic channel integrated on this gatetunable 2D electrode allows electrochemical measurements at a steady-state condition, in which mass transport effects that introduce difficulties in analysis of reaction kinetics are minimized. We observed that, at a given working electrode potential, the surface electron density 𝑛s increases with the back gate bias (up to ~2 orders of magnitude) while the kinetic constant 𝑘et― generally decreases by a factor of 20 at the same time. Overall, the charge transfer at a given electrode potential gets faster at a positive back gate bias. The observed trends of the kinetic parameters at different gate biases are consistent with our predictions; the transverse field effect induces band alignment shifts at the electrode/electrolyte interface, which lead to the modulation of (1) charge carrier density at


22

Page 23 of 27 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 electrode surface and (2) the energy overlap of empty and occupied states in the two phases at a given electrochemical potential. Although only simple outer-sphere electrochemistry has been explored in the current investigation, the same concept should be transferrable to continuous and reversible modulation of other interfacial phenomena, including inner-sphere charge transfer and surface binding of reaction species to the electrode. We believe that the concept and the system introduced here opens new opportunities to investigate and optimize solid/liquid interfaces in various electrochemical systems including liquid junction solar cells, and (photo)electrocatalysts.

ASSOCIATED CONTENT Supporting Information. Materials, fabrication of electrochemical flow cells, cyclic voltammetry of TBBQ on Au electrodes, correction of 𝑉w with IR compensation technique, mathematical model for outer-sphere electrochemistry in electrochemical flow cells, Gerischer model for outer-sphere charge transfer on semiconductor electrodes, correction of 𝑗𝑤 based on the linear diffusion layer model, in-situ measurements of sheet conductance on ZnO electrodes, electron density on the ZnO electrode. This material is available free of charge via the Internet at http://pubs.acs.org.

AUTHOR INFORMATION Corresponding Author *[email protected]


23


Page 24 of 27

Notes The authors declare no competing financial interests.

ACKNOWLEDGMENT The authors thank Dr. Scott White for valuable discussions. C.D.F. acknowledges support by NSF ECCS-1407473. Parts of this works were carried out in the College of Science and Engineering Characterization Facility, University of Minnesota, which has received capital equipment funding from the NSF through the UMN MRSEC program under Award Number DMR-1420013, and in the Minnesota Nano Center which receives partial support from NSF through the NNIN program.

REFERENCES (1)

Kim, C.-H.; Frisbie, C. D. Field Effect Modulation of Outer-Sphere Electrochemistry at Back-Gated, Ultrathin ZnO Electrodes. J. Am. Chem. Soc. 2016, 138 (23), 7220–7223.

(2)

Wang, Y.; Kim, C.-H.; Yoo, Y.; Johns, J. E.; Frisbie, C. D. Field Effect Modulation of Heterogeneous Charge Transfer Kinetics at Back-Gated Two-Dimensional MoS2 Electrodes. Nano Lett. 2017, 17 (12), 7586–7592.

(3)

Nicholson, R. S. Theory and Application of Cyclic Voltammetry for Measurement of Electrode Reaction Kinetics. Anal. Chem. 1965, 37 (11), 1351–1355.

(4)

Klingler, R. J.; Kochi, J. K. Electron-Transfer Kinetics from Cyclic Voltammetry. Quantitative Description of Electrochemical Reversibility. J. Phys. Chem. 1981, 85 (12), 1731–1741.

(5)

Nicholson, R. S.; Shain, I. Theory of Stationary Electrode Polarography. Single Scan and Cyclic Methods Applied to Reversible, Irreversible, and Kinetic Systems. Anal. Chem. 1964, 36 (4), 706–723.

(6)

Irma, L.; Riccarda, A.; Franco, M. An Extended Method for the Practical Evaluation of the Standard Rate Constant from Cyclic Voltammetric Data. Electroanalysis 2004, 16 (6), 505–


24

Page 25 of 27 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


506. (7)

Bard, A. J.; Faulkner, L. R. Electrochemical Methods: Fundamentals and Applications, 2nd ed.; John Wiley & Sons: New York, 1980.

(8)

Rees, N. V; Dryfe, R. A. W.; Cooper, J. A.; Coles, B. A.; Compton, R. G.; Davies, S. G.; McCarthy, T. D. Voltammetry under High Mass Transport Conditions. A High Speed Channel Electrode for the Study of Ultrafast Kinetics. J. Phys. Chem. 1995, 99 (18), 7096– 7101.

(9)

Rees, N. V.; Klymenko, O. V.; Coles, B. A.; Compton, R. G. The High Speed Channel Electrode Applied to Heterogeneous Kinetics: The Oxidation of 1,4-Phenylenediamines and Related Species in Acetonitrile. J. Electroanal. Chem. 2002, 534 (2), 151–161.

(10)

Rees, N. V.; Clegg, A. D.; Klymenko, O. V.; Coles, B. A.; Compton, R. G. Marcus Theory for Outer-Sphere Heterogeneous Electron Transfer: Predicting Electron-Transfer Rates for Quinones. J. Phys. Chem. B 2004, 108 (34), 13047–13051.

(11)

Levich, V. G. Physicochemical Hydrodynamics.; Prentice-Hall: New Jersey, 1962.

(12)

Frumkin, A. N.; Nekrasov, L. N. Ring Disk Electrodes. Dokl. Akad. Nauk SSSR 1959, 126, 115.

(13)

Ivanov, Y. B.; Levich, V. G. Investigation of Unstable Intermediate Products of Electrode Reactions by Means of a Rotating Disc Electrode. Dokl. Akad. Nauk SSSR 1959, 126, 1029.

(14)

Skoog, D. A.; Holler, F. J.; Crouch, S. R. Principles of Instrumental Analysis, 7th ed.; Brooks Cole: Pacific Grove, CA, 2017.

(15)

Macpherson, J. V. Recent Advances in Hydrodynamic Modulation Voltammetry. Electroanalysis 2000, 12 (13), 1001–1011.

(16)

Schuette, S. A.; McCreery, R. L. Efficient Hydrodynamic Modulation Voltammetry with a Microcylinder Electrode. Anal. Chem. 1986, 58 (8), 1778–1782.

(17)

Macpherson, J. V; Unwin, P. R. Hydrodynamic Modulation Voltammetry with a VariableHeight Radial Flow Microring Electrode. Anal. Chem. 1999, 71 (14), 2939–2944.

(18)

Dewald, H. D.; Peterson, B. A. Ultrasonic Hydrodynamic Modulation Voltammetry. Anal. Chem. 1990, 62 (7), 779–782.

(19)

Gerischer, H. Semiconductor Electrochemistry. In Physical Chemistry: An Advanced Treatise Vol. 9A; Eyring, H., Henderson, D., Yost, W., Eds.; Academic: New York, 1970; pp 463–542.

(20)

Nozik, A. J.; Memming, R. Physical Chemistry of Semiconductor−Liquid Interfaces. J. Phys. Chem. 1996, 100 (31), 13061–13078.


25


Page 26 of 27

(21)

Neamen, D. A. Semiconductor Physics and Devices: Basic Principles; McGraw-Hill: New York City, USA, 2003.

(22)

Miller, B.; Heller, A. Semiconductor Liquid Junction Solar Cells Based on Anodic Sulphide Films. Nature 1976, 262, 680.

(23)

Gronet, C. M.; Lewis, N. S.; Cogan, G.; Gibbons, J. N-Type Silicon Photoelectrochemistry in Methanol: Design of a 10.1% Efficient Semiconductor/Liquid Junction Solar Cell. Proc. Natl. Acad. Sci. 1983, 80 (4), 1152 LP-1156.

(24)

Gibbons, J. F.; Cogan, G. W.; Gronet, C. M.; Lewis, N. S. A 14% Efficient Nonaqueous Semiconductor/Liquid Junction Solar Cell. Appl. Phys. Lett. 1984, 45 (10), 1095–1097.

(25)

Hsu, H.-Y.; Ji, L.; Ahn, H. S.; Zhao, J.; Yu, E. T.; Bard, A. J. A Liquid Junction Photoelectrochemical Solar Cell Based on P-Type MeNH3PbI3 Perovskite with 1.05 V Open-Circuit Photovoltage. J. Am. Chem. Soc. 2015, 137 (46), 14758–14764.

(26)

Foley, J. M.; Price, M. J.; Feldblyum, J. I.; Maldonado, S. Analysis of the Operation of Thin Nanowire Photoelectrodes for Solar Energy Conversion. Energy Environ. Sci. 2012, 5 (1), 5203–5220.

(27)

Lewis, N. S. Progress in Understanding Electron-Transfer Reactions Semiconductor/Liquid Interfaces. J. Phys. Chem. B 1998, 102 (25), 4843–4855.

(28)

Maldonado, S.; Stevenson, K. J. Influence of Nitrogen Doping on Oxygen Reduction Electrocatalysis at Carbon Nanofiber Electrodes. J. Phys. Chem. B 2005, 109 (10), 4707– 4716.

(29)

Jaramillo, T. F.; Jørgensen, K. P.; Bonde, J.; Nielsen, J. H.; Horch, S.; Chorkendorff, I. Identification of Active Edge Sites for Electrochemical H2 Evolution from MoS2 Nanocatalysts. Science 2007, 317 (5834), 100–102.

(30)

Li, W.; Tan, C.; Lowe, M. A.; Abruña, H. D.; Ralph, D. C. Electrochemistry of Individual Monolayer Graphene Sheets. ACS Nano 2011, 5 (3), 2264–2270.

(31)

Hui, J.; Pakhira, S.; Bhargava, R.; Barton, Z. J.; Zhou, X.; Chinderle, A. J.; Mendoza-Cortes, J. L.; Rodríguez-López, J. Modulating Electrocatalysis on Graphene Heterostructures: Physically Impermeable Yet Electronically Transparent Electrodes. ACS Nano 2018, 12 (3), 2980–2990.

(32)

Bard, A. J. Inner-Sphere Heterogeneous Electrode Reactions. Electrocatalysis and Photocatalysis: The Challenge. J. Am. Chem. Soc. 2010, 132 (22), 7559–7567.

(33)

Plaza, M.; Huang, X.; Ko, J. Y. P.; Shen, M.; Simpson, B. H.; Rodríguez-López, J.; Ritzert, N. L.; Letchworth-Weaver, K.; Gunceler, D.; Schlom, D. G.; Arias, T. A.; Brock, J. D.; Abruña, H. D. Structure of the Photo-Catalytically Active Surface of SrTiO3. J. Am. Chem. Soc. 2016, 138 (25), 7816–7819.

at


26

Page 27 of 27 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


(34)

Qu, L.; Liu, Y.; Baek, J.-B.; Dai, L. Nitrogen-Doped Graphene as Efficient Metal-Free Electrocatalyst for Oxygen Reduction in Fuel Cells. ACS Nano 2010, 4 (3), 1321–1326.

(35)

Hinnemann, B.; Moses, P. G.; Bonde, J.; Jørgensen, K. P.; Nielsen, J. H.; Horch, S.; Chorkendorff, I.; Nørskov, J. K. Biomimetic Hydrogen Evolution: MoS2 Nanoparticles as Catalyst for Hydrogen Evolution. J. Am. Chem. Soc. 2005, 127 (15), 5308–5309.

(36)

George, S. M. Atomic Layer Deposition: An Overview. Chem. Rev. 2010, 110 (1), 111– 131.

(37)

Britz, D. IR Elimination in Electrochemical Cells. J. Electroanal. Chem. Interfacial Electrochem. 1978, 88 (3), 309–352.

(38)

Blaedel, W. J.; Klatt, L. N. Reversible Charge Transfer at the Tubular Platinum Electrode. Anal. Chem. 1966, 38 (7), 879–883.

(39)

Rüssel, C.; Janicke, W. Heterogeneous Electron Exchange of Quinones in Aprotic Solvents. Part III. The Second Reduction Step of p-Benzoquinone and Its Dependence on the Supporting Electrolyte. J. Electroanal. Chem. 1986, 199 (1), 139–151.

(40)

Calef, D. F.; Wolynes, P. G. Classical Solvent Dynamics and Electron Transfer. 1. Continuum Theory. J. Phys. Chem. 1983, 87 (18), 3387–3400.

(41)

Koutecký, J.; Levich, V. G. The Use of the Rotating Disk Electrode in the Studies of Electrochemical Kinetics and Electrolytic Processes. Zhurnal Fiz. Khimii 1958, 32, 1565.

(42)

Morrison, S. R. Electrochemistry at Semiconductor and Oxidized Metal Electrodes; Plenum Press: New York, 1980.


27

Continuous and Reversible Tuning of ... - ACS Publications

Recommend Documents