Collision and Oxidation of Single LiCoO2 Nanoparticles Studied by

May 8, 2017 - State Key Laboratory of Analytical Chemistry for Life Science, School of ... microscopy (SPRM) and conventional electrochemical recordin...
0 downloads 0 Views 2MB Size
Download PDF
Article pubs.acs.org/ac

Collision and Oxidation of Single LiCoO2 Nanoparticles Studied by Correlated Optical Imaging and Electrochemical Recording Linlin Sun, Dan Jiang, Meng Li, Tao Liu, Liang Yuan, Wei Wang,* and Hong-Yuan Chen* State Key Laboratory of Analytical Chemistry for Life Science, School of Chemistry and Chemical Engineering, Nanjing University, Nanjing 210093, China S Supporting Information *

ABSTRACT: Electrochemists have witnessed the rapid development of single nanoparticle collision (SNC) in the past decade as a powerful technology to study the electrochemical activity of single nanoparticles (NPs). One of the ultimate goals of SNC is to build the structure−activity relationship in a bottom-up way, which requires the correlation between the electrochemical activity and the morphology/structure of the same individual. In the present work, we demonstrated the capability of combining surface plasmon resonance microscopy (SPRM) and conventional electrochemical recordings to correlate the size and activity of single LiCoO2 NPs. Electrochemical recordings provided the rate and amount of electron transfer associated with a single LiCoO2 NP during its collision onto the electrode. SPRM measured the size and the location of the collision of the very same individual, which further allowed for the morphology characterization with scanning electron microscopy. The present work was the first attempt to correlate the electrochemical activity and the morphology of single Li-ion storage NPs during collision events, which not only expanded the SNC technology to study versatile electro-active cathode and anode nanomaterials in different kinds of batteries, but also paved a way toward the structure−activity relationship of Li-ion storage nanomaterials with significant implications for Li-ion batteries.

C

to the dissolution of Ag NPs into Ag ions, allowing for inferring the size of NPs by Faraday’s law. Although powerful, the present SNC technology faces two major challenges. First, while conventional electrochemical recording of a collision event revealed the activity of a single NP, it often lacked the structure information associated with this particular NP such as its size, morphology, and the location of a collision. Such information was critical to understand the electron transfer at nanoscaled interfaces and to build the structure−activity relationship. The combination of SNC with optical microscopy16,17 and electron microscopy18 has proven to be an effective way to assist the identification of NPs. Second, existing SNC studies mainly focused on the methodology and reaction mechanism underneath several model electrochemical reactions. The utilization of SNC to other kinds of electro-active nanomaterials, particularly those with significant practical implications, is still desired for further fundamental and application developments of the SNC technology. LiCoO2 is one of the most successful positive electrode materials in commercial Li-ion batteries (LIBs) for their broad applications in portable electronic devices and hybrid electric

onventional electrochemistry often measures the electrode current at a bulk interface containing large populations of nanoparticles (NPs), which not only washes out the intrinsic heterogeneity among individual NPs, but also hampers the understanding of electron transfer occurring at nanoscaled electro-active regions. In order to investigate the electrochemical activity of single NPs, single nanoparticle collision (SNC) has emerged rapidly and has become an important research field in the past decade.1−5 SNC strategy has been mainly utilized in two categories of nanomaterialsinvolved electrochemical processes, namely, electrocatalysis and direct electrochemistry of nanomaterials. In the former case, the collision of a single NP formed a local and transient electrocatalytic site, resulting in an amplified electrochemical current contributed by solution-phase electrochemical reactions.6−10 Catalytic current went back to the baseline at the moment of the inactivation or the departure of the NP. The time-resolved current spikes allowed for the study on the electrocatalytic property of single nanocatalyst. In the latter case, the NP itself underwent oxidation or reduction reactions when it stroked at the electrode surface.11−15 So far, silver NPs and several other metal NPs have been most commonly adopted as model nanomaterials to study the direct electrochemistry of single NPs. Electron transfer stopped when the accessible electro-active atoms in the NPs were exhausted. The observed current spikes during collision events were attributed © XXXX American Chemical Society

Received: February 21, 2017 Accepted: May 8, 2017 Published: May 8, 2017 A

DOI: 10.1021/acs.analchem.7b00649 Anal. Chem. XXXX, XXX, XXX−XXX

Article

Analytical Chemistry

Figure 1. (a) Schematic illustration of studying single LiCoO2 NP collision events with correlated SPRM imaging and electrochemical recording. (b) Transient current spike associated with the extraction of Li-ion during the oxidation of the LiCoO2 NP on an Au microelectrode (black curve). (c) Transient plasmonic image intensity correlated with the collision (initial intensity increase) and oxidation (subsequent intensity decrease) of the LiCoO2 NP (blue curve). Three snapshots show the representative SPRM images before the collision (t0), right after the collision (t1), and eventually after the oxidation (t2). The dashed line in (b) and (c) demonstrates that maximal current occurs at the moment of largest optical intensity drop, as the current represents the first-order derivative of optical intensity curve.

vehicles.19 Interfacial electron transfer rates and Li-ion diffusion kinetics of LiCoO2 nanomaterials significantly affected the charge rate and life cycle of LIBs. Because of the intrinsic heterogeneity of Li-ion storage NPs, studying the dynamics of Li-ion diffusion at single NP level was of significance to design electrode materials toward better performance.20,21 In comparison to metal NPs that were often used in previous SNC studies, LiCoO2 NPs exhibited two superior advantages. First, the charge−discharge process of LiCoO2 NPs is not only a fundamentally important electrochemical reaction, but also is of importance for promoting its practical applications. Second, different from metal NPs that are often either dissolved or deformed during electrochemical oxidation,17,22−24 LiCoO2 NPs exhibit subtle morphological change during electrochemical cycling.25 This feature is beneficial for the adoption of the other in situ characterization techniques such as scanning electron microscopy (SEM) or atomic force microscopy. However, to the best of our knowledge, a capability of SNC to study collision events of important Li-ion storage nanomaterials such as LiCoO2 has not been demonstrated yet. Here, we report the investigation on the collision and oxidation kinetics of single LiCoO2 NPs using the SNC technology coupled with surface plasmon resonance microscopy (SPRM). SPRM is a recently developed optical imaging technique which is sensitive to the refractive index (RI) of single NPs, and thus, it is suitable for imaging not only plasmonic NPs22 but also nonplasmonic NPs such as metal oxide NPs26 and organic hydrogel NPs.27 We have found that RI of a single Li1−xCoO2 NP linearly decreases with the amount of Li-ion (1 − x) during electrochemical delithiation, resulting in decreased SPRM intensity. Based on such dependence, a conversion model was proposed to calculate the electrochemical current of a single LiCoO2 NP from its SPRM signal during electrochemical cycling.26 In our previous study, preimmobilized LiCoO2 NPs were examined with cyclic voltammetry and potential step voltammetry techniques. It was difficult to experimentally measure the current due to the large background charging current associated with these potential sweeping techniques. In contrast, SNC strategy adopted in the present work would significantly reduce the charging current by applying a constant potential. It thus

provided a good opportunity to experimentally measure the electrochemical current and compare it with the result calculated from the conversion model. A single LiCoO2 NP stochastically collided onto the electrode surface under a positive potential and stayed there. The potential was high enough to extract Li-ion out of the NP, leading to a transient current spike and accompanying with a change in its RI. In addition to the recording of electrochemical current, the entire collision-n-oxidation process was continuously monitored with SPRM at the same time, leading to a correlated structure (optical microscopy) and activity (electrochemical current) relationship of the same single NP. High temporal resolution of SPRM made it possible to resolve fast electrochemical reactions of LiCoO2 NPs at a temporal resolution of 10 ms. Its spatial resolution identified the locations of collision events, which was required to further characterize the morphology of NPs with SEM. Furthermore, an optical-to-electrochemical conversion model was proposed in our previous work to calculate the cyclic voltammogram of a single LiCoO2 NP solely from its optical response under sweeping potentials.26 The present work provided direct evidence to support such optical-to-electrochemical conversion model during the SNC under a constant potential.

■

EXPERIMENTAL SECTION Materials and Characterizations. Crude LiCoO2 powders were purchased from Sigma-Aldrich and milled to ∼200 nm (diameter) with a ball grinder. As-prepared LiCoO2 NPs were dispersed in the deionized water (18.2 MΩ·cm, Smart2Pure3 UF, Thermo Fisher) by sonication just before use. A 1 M lithium nitrate (LiNO3, Sigma-Aldrich) aqueous solution was used as the electrolyte solution throughout the research. SEM (Shimadzu, S-4800) was used to determine the morphology of single LiCoO2 NPs on the surface of the Au microelectrode. Apparatus. SPRM imaging was performed on a home-built SPRM microscope, whose detailed descriptions can be found in our previous work.28,29 Briefly, it was built on an inverted optical microscopy (TIRFM, Nikon Ti-E) equipped with a high numerical aperture oil immersion objective 60× (N.A. 1.49). A red super luminescent light-emitting diode with wavelength of B

DOI: 10.1021/acs.analchem.7b00649 Anal. Chem. XXXX, XXX, XXX−XXX

Article

Analytical Chemistry 680 nm (Q-photonics LLC, QSDM-680−2, operating power 0.2 mW) was used as the light source. A polarizer (Edmund Optics, Catalog No. 43−786) was inserted in the optical path to generate p-polarized light, which excited surface plasmon polaritons propagating at the interface between electrolyte solution and gold-coated coverslip (a quartz coverslip coated with a 47 nm thick gold film). A CCD camera (Pike F-032B, Allied Vision Technologies) was used to capture the SPRM images with an acquisition rate of 100 frames per second (fps) at pixel resolution of 320 × 240. The electrochemical measurement was carried out using an Axon Multiclamp 700B amplifier (Molecular Devices) interfaced to a PC through a Digidata 1550A digitizer (Molecular Devices). Current traces were recorded with a 100 Hz sampling rate and the low-pass filter of the amplifier was set to 20 Hz. A data acquisition card USB-6251 (National Instruments) was used to synchronize the camera TTL signal and electrochemical recordings. The temporal resolution (10 ms) of electrochemical measurement was the same as that of optical imaging (100 fps) for the convenience of synchronization. In order to perform electrochemical experiments, the gold-coated coverslip that was placed on the microscope sample stage acted as the working electrode. The gold film was specially patterned with an effective size of 50 × 50 μm2 (a microelectrode, see below for details). An Ag/AgCl wire was used as the counter and quasireference electrode to form a two-electrode recording required by the current amplifier. The time constant of as-prepared microelectrode in 1 M LiNO3 solution was determined to be 25 μs (Figure 3 and 4). We also adopted a regular sized 5 mm diameter gold electrode to measure the bulk material (Figure 2), whose time constant was 1 ms. In a typical SNC measurement, data acquisition card started to record camera TTL signal and electrode current before the camera and the current amplifier was initiated. After the potential was applied and the camera was running, 1 μL of

Figure 3. (a) Snapshots of plasmonic images of two sequential LiCoO2 NPs during the collision and delithiation processes on the Au microelectrode under a constant potential of +500 mV. (b) Sequential transient electrochemical current (top panel) corresponding to transient SPRM intensity curve (bottom panel) of the two individual LiCoO2 NPs.

Figure 4. (a) SPRM snapshots of two sequential LiCoO2 NPs collided onto a Au microelectrode at a potential of +200 mV in 1 M LiNO3. (b) Sequential electrochemical current (top panel) and transient SPRM intensity curve (bottom panel) of the LiCoO2 NPs.

diluted LiCoO2 suspension was manually injected into and mixed with 500 μL of LiNO3 solutions. The subsequent collision and oxidation of single LiCoO2 NPs resulted in current spikes in the electrochemical recordings and optical features in the SPRM images. The locations of collision events were recorded with SPRM under a coordinate determined by the square Au microelectrode, which can be easily identified in SEM characterizations. After the electrochemical and optical measurements, the gold-coated coverslip was rinsed and dried in a vacuum chamber, and characterized by SEM. The very same NP can be identified by searching the location-of-collision previously revealed by SPRM. Fabrication of the Microelectrode. A square Au microelectrode (50 × 50 μm2) was fabricated by lithography on a glass coverslip (No. 1 BK7 glass from Fisher). In addition to the active area, there was a 30 μm width of gold band coated

Figure 2. (a) Typical bulk-electrode cyclic voltammogram of ensemble LiCoO2 NPs immobilized on a 5 mm diameter gold electrode in 1 M LiNO3. (b) SPRM intensity curve (blue) of a single LiCoO2 NP immobilized on a gold chip when a potential step was applied from 0.3 to 0.5 V (black curve). Time constant of this electrochemical cell is 1 ms. Inset: the SPRM snapshots of the LiCoO2 NP before and after the oxidation. C

DOI: 10.1021/acs.analchem.7b00649 Anal. Chem. XXXX, XXX, XXX−XXX

Article

Analytical Chemistry

intensity was determined by selecting a 5 pixel × 15 pixel rectangle region of interest (ROI) in the middle part (bright part) of the parabolic tail (bottom panel in Figure 2b, inset, red ROI). Note that the charging background of a gold film has been subtracted by using a reference-correction method we introduced previously.26 The decreased RI of LiCoO2 after oxidation was also identified from the reduced SPRM image contrast as shown in Figure 2b, inset. Simultaneous Optical and Electrochemical Recordings. The up-left image in Figure 3a shows the electro-active area of the microelectrode (50 × 50 μm2). This area was smaller than the view of field of SPRM (80 × 60 μm2), so that all collision events that contributed electrochemical signals would be monitored in SPRM images. Other images in Figure 3a display several representative snapshots of the sequential collision events of two individual LiCoO2 NPs onto the microelectrode in 1 M LiNO3 at a potential of +500 mV. A movie of the entire process is provided in Movie S1 in the Supporting Information. At 1.90 s, NP1 hit at the lower right corner of the Au microelectrode. SPRM intensity decreased by 38% in 230 ms and kept stable thereafter, demonstrating the reduced RI due to the delithiation (Figure 3b, blue curve in the bottom panel). At 20.46 s, NP2 hit at the lower left corner of the microelectrode. Its SPRM intensity dropped by 34% in 130 ms to complete the delithiation process (Figure 3b, green curve in the bottom panel). To distinguish the two SPRM intensity curves, the curve of NP1 was raised by 200 intensity unit (IU). Two oxidative peaks were simultaneously observed in the electrochemical current curve shown in Figure 3b (black curve in the top panel), corresponding to the two collision events. The collision frequency was optimized to be sufficiently low so as to avoid simultaneous collisions of multiple individuals and facilitate the identification of each individual in the subsequent SEM characterization. To confirm that the SPRM intensity decrease was indeed due to the electrochemical oxidation of LiCoO2 NPs, we performed a control experiment by applying a potential (+200 mV) that was too low to trigger the oxidation of LiCoO2 NPs. Figure 4a shows several snapshots of the sequential collisions of two individual LiCoO2 NPs onto the microelectrode. Figure 4b shows only a step but no decrease in the SPRM intensity curve, and no change in the electrical recording. This result demonstrated that the spikes in the electrochemical recording, as well as the subsequent drop in the SPRM intensity curve, were caused by the oxidation of LiCoO2 at +500 mV. Validation of the Optical-to-Electrochemical Conversion Model. The theoretical oxidation current of a single LiCoO2 NP was calculated from the optical images with the conversion model we previously proposed under sweeping potentials.26 It was found that the optical signal was proportional to the amount of Li-ion (1 − x) in the Li1−xCoO2 NPs (0 ≤ x ≤ 0.25). Therefore, the first-order derivative of time-dependent optical signal gave the Li-ion transfer rate, that is, the delithiation current, of a single LiCoO2 NP. The firstorder derivative relationship was also valid in the present work where constant potential was applied. The theoretical current of a single LiCoO2 NP was described by the following equation:

by a 50 μm width of SiO2 (as insulating layer), which connected the Au microelectrode to an electrochemical device. These microsized areas were defined from AZ5214 photoresist lithography. Note that this design was to ensure the entire microelectrode surface was in the view-field of SPRM.

■

RESULTS AND DISCUSSION Principle of the Present Work. Figure 1a shows the schematic illustration of the experimental setup for simultaneous optical and electrochemical recordings. It consists of a SPRM microscope for optical recording, a current amplifier for electrochemical control and recording, and a data acquisition card for signal synchronization. The detailed description of SPRM can be found in previous work.28,29 Briefly, a gold film that served as a substrate was placed on an inverted microscope. A red beam (wavelength = 680 nm) was collimated to illuminate the gold film with a certain incident angle and excited the surface plasmon polaritons on the gold film− electrolyte interface.30 The reflected light was captured with a CCD camera to produce a SPRM image. When NPs are present on the surface of gold film, they interact with the surface plasmon wave and produce a parabolic pattern,31 in which the parabolic tail represents the propagating direction of surface plasmon wave. The SPRM intensity relies on the RI of the NPs, and thus, higher RI would exhibit larger optical contrast in the SPRM image. The principle of the present work is to measure the RI change of single LiCoO2 NPs during electrochemical reactions with plasmonic imaging, as shown in Figure 1a. Single LiCoO2 NPs stochastically collided and stayed on the gold film microelectrode. When the electrode (gold film) was kept at a constant potential (0.5 V vs Ag/AgCl) that was sufficient to induce the oxidation of LiCoO2 NPs (LiCoO2 − xe− = Li1−xCoO2 + xLi+), Li-ion was extracted from the LiCoO2 NP, resulting in an oxidative current (Figure 1b). At the same time, the change in the Li amount in the NP decreased its RI as the chemical composition and electronic structure was modified.26 Consequently, SPRM intensity first increased due to the collision of a LiCoO2 NP onto the coverslip (SPRM is a nearfield phenomenon and it is mostly sensitive to the object closed to the coverslip), which subsequently decreased because of the reduced RI after delithiation (Figure 1c). It is anticipated that the maximal electron transfer rate appeared at the moment when SPRM intensity exhibited largest decrease (dash line in Figure 1b,c). Note that this principle is different from our previous report that measured the volume evolution of a single Ag NP during electrochemical oxidation.22,24 Reduced RI During LiCoO2 Oxidation. We first demonstrated the electrochemical activity of as-prepared LiCoO2 NPs. Cyclic voltammogram of ensemble LiCoO2 NPs immobilized on a regular gold electrode exhibited a pair of current peaks at 0.44/0.41 V (vs Ag/AgCl) in the presence of 1 M LiNO3 aqueous solution (Figure 2a). This result was well consistent with previous studies on the aqueous electrochemistry of LiCoO2.32,33 This pair of peaks corresponded to the first-order Mott transition from metallic to insulator phase. Our previous work has shown that delithiated Li1−xCoO2 (oxidized form) exhibited smaller RI compared with the lithiated state LiCoO2 (reduced form),26 and the RI was nearly proportional to the lithiation state (1 − x). For this reason, when a potential step from 0.3 to 0.5 V was applied to a single LiCoO2 NP immobilized on a gold chip, a decreased SPRM intensity was observed (blue curve in Figure 2b). The SPRM

i=α

dISPR dt

(1)

where i, α, and ISPR are the theoretical current, conversion factor, the SPRM intensity, respectively. The conversion factor, α, describes how much the amount of Li-ion will be transferred D

DOI: 10.1021/acs.analchem.7b00649 Anal. Chem. XXXX, XXX, XXX−XXX

Article

Analytical Chemistry

thickness. This result also demonstrated that the Li-ion extraction first occurred at the bottom portion of the NP and then top portion. Different from the oxidation-induced dissolution of Ag NPs, the delithiation of a single LiCoO2 NP maintained its location and the morphology. It allowed us to characterize its morphology with SEM after the electrochemical and optical recording of SNC. Its location was provided by SPRM as a guide for SEM characterizations. The corresponding SEM images of NP1 and NP2 were displayed in the inset of Figure 5a,b. SEM images indicated an equivalent diameter of 218 and 330 nm for NP1 and NP2, respectively, resulting in a volume ratio of 1:3.5. This ratio was consistent with the SPRM intensity ratio (1:4.2) between NP1 (96 IU) and NP2 (403 IU). Such consistency was expected because the SPRM intensity was proportional to the volume. By integrating the current peak, the charge quantity was found to be 6 and 33 pC for NP1 and NP2, respectively, suggesting a volume ratio of 1:5.5. The volume ratio determined by three independent techniques was generally consistent with each other. The difference might be due to the inaccurate estimation from SEM image to the volume for an irregular NP.

from a single LiCoO2 NP in the case of the change of 1 IU. For the NP1 shown in Figure 3b (bottom panel), the extraction of 38% of total amount of Li-ion in this LiCoO2 NP led to a SPRM intensity decrease by 64 IU. We can calculate α using the equation below: α=

4πr 3ρNAex 3M ΔISPR

(2)

where ρ, NA, e, and M are density of LiCoO2, Avogadro’s constant, electron charge, and molecular weight of LiCoO2, respectively; r is the radius of the NP1, which is measured by SEM (see Figure 5); x is delithiation state (x = 0.38); and ΔISPR

■

CONCLUSION We demonstrated a study of correlating SPRM and SNC technology to measure the electrochemical activity of a single LiCoO2 NP and to correlate it with its size and morphology. We summarize several advantages of LiCoO2 NPs when acting as a new model nanomaterial in SNC studies. First, LiCoO2 is an important electrode material in commercial LIBs. The traditional SNC of studying model reactions can be expanded to the field of LIBs. Second, LiCoO2 NPs have great electrochemical activity and reversibility, which is required for performing effective SNC measurements. Third, LiCoO2 NPs maintain their morphology and locations after the oxidation. We can thus combine SPRM with SEM or other in situ characterization techniques, making it possible to establish the structure−activity relationship. Finally, LiCoO2 NPs avoid the plasmonic coupling effect with surface plasmon wave in comparison with previously adopted plasmonic materials such as Au and Ag NPs. Therefore, LiCoO2 is anticipated to a promising model nanomaterial for SNC studies in future. We also believe that the SNC technology coupled with optical microscopy can help the study and design of Li-ion storage nanomaterials with better performances.

Figure 5. Comparison of transient current spikes obtained from electrochemical recording (black curve) and the conversion of the SPRM intensity for NP1 (a) and NP2 (b). (c) Correlation between the electric quantity (i.e., integration of the electrochemical current) and the decrease of SPRM intensity for the two NPs. Inset: SEM images of the corresponding LiCoO2 NPs.

represents the change of the SPRM intensity (64 IU). Combining eqs 1 and 2, we can calculate theoretical current. As shown in Figure 5a,b, the theoretical current was consistent with the experimental current very well, further demonstrating that the previous reported model was general for SPRM. The spikes in the electrochemical current curve were systematically wider than those in the optical current because of the peak broadening effect associated with the low-pass electronic filter at 20 Hz (50 ms). It further underscored the value of optical recording as it was free of electronic filters. We observed the correlation between the electric quantity, that is, the integration of the electrochemical current, and the change of SPRM intensity (the decrease between maximal SPRM intensity and transient SPRM intensity) for the two NPs, as shown in Figure 5c. When transferring 1 pC charge from individual LiCoO2 NP, SPRM intensity dropped approximately 10 IU for both NP1 and NP2. However, only 3 IU per 1 pC charge was observed at the late stage of oxidation. This was because the surface plasmon wave exponentially decayed along the vertical distance. As a result, SPRM was less sensitive to the Li-ion extraction occurring at the top portion of NPs. This effect was mostly obvious for NP2, which exhibited much larger size and vertical

■

ASSOCIATED CONTENT

* Supporting Information S

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.analchem.7b00649. Movie S1: Top panel: Transient electrochemical current of two sequential LiCoO2 nanoparticles during the collision and delithiation processes on the Au microelectrode under a constant potential of +500 mV. Middle panel: Simultaneously recorded SPRM intensity curves. The SPRM intensity was determined by averaging the pixel intensity of a 5 pixel × 15 pixel rectangle ROI located in the center bright part of each parabolic pattern. Initial intensity increase indicates the collision of nanoparticle. Subsequent decrease indicates the oxidation of nanoparticle. Bottom panel: Differential SPRM images E

DOI: 10.1021/acs.analchem.7b00649 Anal. Chem. XXXX, XXX, XXX−XXX

Article

Analytical Chemistry

■

during two sequential “collision and oxidation” events of LiCoO2 nanoparticles (AVI).

(24) Sun, L. L.; Fang, Y. M.; Li, Z. M.; Wang, W.; Chen, H. Y. Nano Res. 2017, 10, 1740−1748. (25) Mukhopadhyay, A.; Sheldon, B. W. Prog. Mater. Sci. 2014, 63, 58−116. (26) Jiang, D.; Jiang, Y. Y.; Li, Z. M.; Liu, T.; Wo, X.; Fang, Y. M.; Tao, N. J.; Wang, W.; Chen, H. Y. J. Am. Chem. Soc. 2017, 139, 186− 192. (27) Cho, K.; Fasoli, J. B.; Yoshimatsu, K.; Shea, K. J.; Corn, R. M. Anal. Chem. 2015, 87, 4973−4979. (28) Shan, X. N.; Diez-Perez, I.; Wang, L. J.; Wiktor, P.; Gu, Y.; Zhang, L. H.; Wang, W.; Lu, J.; Wang, S. P.; Gong, Q. H.; Li, J. H.; Tao, N. J. Nat. Nanotechnol. 2012, 7, 668−672. (29) Wang, W.; Foley, K.; Shan, X.; Wang, S. P.; Eaton, S.; Nagaraj, V. J.; Wiktor, P.; Patel, U.; Tao, N. J. Nat. Chem. 2011, 3, 249−255. (30) Homola, J. Chem. Rev. 2008, 108, 462−493. (31) Demetriadou, A.; Kornyshev, A. A. New J. Phys. 2015, 17, 013041. (32) Heli, H.; Yadegari, H.; Jabbari, A. J. Appl. Electrochem. 2012, 42, 279−289. (33) Ruffo, R.; Wessells, C.; Huggins, R. A.; Cui, Y. Electrochem. Commun. 2009, 11, 247−249.

AUTHOR INFORMATION

Corresponding Authors

*E-mail: [email protected]. *E-mail: [email protected]. ORCID

Linlin Sun: 0000-0003-0900-5038 Dan Jiang: 0000-0001-5408-4949 Wei Wang: 0000-0002-4628-1755 Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS We thank financial support from the National Natural Science Foundation of China (NSFC, Grants 21327902, 21522503, 21527807) and the Natural Science Foundation of Jiangsu Province (BK20150013, BK20140592, BK20150570).

■

REFERENCES

(1) Quinn, B. M.; Van ’t Hof, P. G.; Lemay, S. G. J. Am. Chem. Soc. 2004, 126, 8360−8361. (2) Xiao, X. Y.; Bard, A. J. J. Am. Chem. Soc. 2007, 129, 9610−9612. (3) Wang, W.; Tao, N. Anal. Chem. 2014, 86, 2−14. (4) Cheng, W.; Compton, R. G. TrAC, Trends Anal. Chem. 2014, 58, 79−89. (5) Anderson, T. J.; Zhang, B. Acc. Chem. Res. 2016, 49, 2625−2631. (6) Xiao, X. Y.; Fan, F. R. F.; Zhou, J. P.; Bard, A. J. J. Am. Chem. Soc. 2008, 130, 16669−16677. (7) Kwon, S. J.; Bard, A. J. J. Am. Chem. Soc. 2012, 134, 7102−7108. (8) Dasari, R.; Robinson, D. A.; Stevenson, K. J. J. Am. Chem. Soc. 2013, 135, 570−573. (9) Kang, M.; Perry, D.; Kim, Y. R.; Colburn, A. W.; Lazenby, R. A.; Unwin, P. R. J. Am. Chem. Soc. 2015, 137, 10902−10905. (10) Zhao, L. J.; Qian, R. C.; Ma, W.; Tian, H.; Long, Y. T. Anal. Chem. 2016, 88, 8375−8379. (11) Zhou, Y. G.; Rees, N. V.; Compton, R. G. Angew. Chem., Int. Ed. 2011, 50, 4219−4221. (12) Krause, K. J.; Yakushenko, A.; Wofrum, B. Anal. Chem. 2015, 87, 7321−7325. (13) Oja, S. M.; Robinson, D. A.; Vitti, N. J.; Edwards, M. A.; Liu, Y.; White, H. S.; Zhang, B. J. Am. Chem. Soc. 2017, 139, 708−718. (14) Ustarroz, J.; Kang, M.; Bullions, E.; Unwin, P. R. Chem. Sci. 2017, 8, 1841−1853. (15) Ma, W.; Ma, H.; Chen, J. F.; Peng, Y. Y.; Yang, Z. Y.; Wang, H. F.; Ying, Y. L.; Tian, H.; Long, Y. T. Chem. Sci. 2017, 8, 1854−1861. (16) Fosdick, S. E.; Anderson, M. J.; Nettleton, E. G.; Crooks, R. M. J. Am. Chem. Soc. 2013, 135, 5994−5997. (17) Brasiliense, V.; Patel, A. N.; Martinez-Marrades, A.; Shi, J.; Chen, Y.; Combellas, C.; Tessier, G.; Kanoufi, F. J. Am. Chem. Soc. 2016, 138, 3478−3483. (18) Kleijn, S. E. F.; Lai, S. C. S.; Miller, T. S.; Yanson, A. I.; Koper, M. T. M.; Unwin, P. R. J. Am. Chem. Soc. 2012, 134, 18558−18561. (19) Bruce, P. G.; Scrosati, B.; Tarascon, J. M. Angew. Chem., Int. Ed. 2008, 47, 2930−2946. (20) Huang, J. Y.; Zhong, L.; Wang, C. M.; Sullivan, J. P.; Xu, W.; Zhang, L. Q.; Mao, S. X.; Hudak, N. S.; Liu, X. H.; Subramanian, A.; Fan, H. Y.; Qi, L. A.; Kushima, A.; Li, J. Science 2010, 330, 1515−1520. (21) Zheng, G.; Lee, S. W.; Liang, Z.; Lee, H. W.; Yan, K.; Yao, H.; Wang, H.; Li, W.; Chu, S.; Cui, Y. Nat. Nanotechnol. 2014, 9, 618−623. (22) Fang, Y. M.; Wang, W.; Wo, X.; Luo, Y. S.; Yin, S. W.; Wang, Y. X.; Shan, X. N.; Tao, N. J. J. Am. Chem. Soc. 2014, 136, 12584−12587. (23) Nizamov, S.; Kasian, O.; Mirsky, V. M. Angew. Chem., Int. Ed. 2016, 55, 7247−7251. F

DOI: 10.1021/acs.analchem.7b00649 Anal. Chem. XXXX, XXX, XXX−XXX