Visualizing and Calculating Tip−Substrate Distance in NanoscaleScanning Electrochemical Microscopy Using 3‑Dimensional Super-Resolution Optical ImagingVignesh Sundaresan,† Kyle Marchuk,†,‡ Yun Yu,† Eric J. Titus,†,§ Andrew J. Wilson,†,∥

Chadd M. Armstrong,⊥ Bo Zhang,⊥ and Katherine A. Willets*,†

†Department of Chemistry, Temple University, Philadelphia, Pennsylvania 19122, United States⊥Department of Chemistry, University of Washington, Seattle, Washington 98195-1700, United States

*S Supporting Information

ABSTRACT: We report a strategy for the optical determination oftip−substrate distance in nanoscale scanning electrochemicalmicroscopy (SECM) using three-dimensional super-resolutionfluorescence imaging. A phase mask is placed in the emission pathof our dual SECM/optical microscope, generating a double helixpoint spread function at the image plane, which allows us to measurethe height of emitting objects relative to the focus of the microscope.By exciting both a fluorogenic reaction at the nanoscale electrode tipas well as fluorescent nanoparticles at the substrate, we are able tocalculate the tip−substrate distance as the tip approaches the surfacewith precision better than 25 nm. Attachment of a fluorescentparticle to the insulating sheath of the SECM tip extends thistechnique to nonfluorogenic electrochemical reactions. Correlatedelectrochemical and optical determination of tip−substrate distanceyielded excellent agreement between the two techniques. Not only does super-resolution imaging offer a secondary feedbackmechanism for measuring the tip−sample gap during SECM experiments, it also enables facile tip alignment and a strategy foraccounting for electrode tilt relative to the substrate.

Nanoscale scanning electrochemical microscopy (SECM) isevolving as a powerful tool to obtain electrochemical

images and investigate electron-transfer processes with nanoscaleresolution.1−3 Using nanometer-sized electrodes as scanningprobe tips, it is possible to image the topography ofelectrochemically active nano-objects,4−7 probe electrocatalyticreactions at single nanoparticles,6,7 detect short-lived reactionintermediates,8,9 and interrogate biological cells.10,11 In afeedback-mode SECM experiment, a voltage is applied to anelectroactive tip immersed in a solution of a redox-activemediator in order to generate a current, which is then measuredas the tip approaches either an electrochemically inert or activesubstrate. The current at the tip either decreases or increases asthe tip approaches the substrate due to the hindered diffusion ofthe redox mediator by the inert substrate (negative feedback) orthe regeneration of the redox mediator at the active substrate(positive feedback), respectively.1 By fitting the current versusdistance approach curve, it is possible to use measured currentsto extract information about the tip−substrate gap in subsequentexperiments. A feedback mode experiment utilizes very fast masstransport and can provide high axial spatial resolution, which hasbeen extensively used for topographic imaging4−7 anddetermining rapid electron-transfer kinetics.12−15

While the steady-state tip current strongly depends on thetip−substrate distance, obtaining accurate spatial informationalong this axial dimension is essential to interpret feedback modedata. Typically, the minimum electrochemically measurable tip−sample gap is limited by the size of the electrode and thethickness of the protective sheath (usually glass) that surroundsthe electrode, making it difficult to measure nanoscale tip−sample gaps. Moreover, to extract the separation distancebetween the tip and the substrate, approach curves for adiffusion-controlled process are fit to a theoretical model, whichdepends on the geometries of the tip electrode and substrate.1

Difficulties in characterizing the nanoelectrode geometry,compensating for thermal drift,16 and achieving perfect align-ment of the nanotip to the substrate17 often cause considerableuncertainties in the distance determination using an electro-chemical approach curve. Further, in an actual experiment, themeasured current will depend on both the tip−sample gap as wellas any electrochemical activity at the sample itself. Thus, if thesample drifts over time, it is extremely difficult to know whetherthe change in the measured current is due to drift in the tip−

Received: October 17, 2016Accepted: December 6, 2016Published: December 6, 2016

Article

pubs.acs.org/ac

© 2016 American Chemical Society 922 DOI: 10.1021/acs.analchem.6b04073Anal. Chem. 2017, 89, 922−928

sample gap, changes in the electrochemical activity of the sample,or some combination of the two. To overcome these limitations,hybrid techniques using tip position modulation,18 shearforce,19,20 atomic force,21,22 and intermittent contact SECM23

have been developed to facilitate tip positioning and distancecontrol by providing a secondary distance-dependent feedbacksignal. In these hybrid techniques, either the tip has to bemodified in its shape, requiring extensive fabrication efforts,21,22

or is oscillated in solution, which perturbs the native environ-ment of the system.18 Moreover, several of these techniques havelimited dynamic range in determining the tip−sample gap due tothe tip−substrate interaction that is necessary to achieve thissecondary feedback.1

To address the aforementioned challenges, we have integratedSECM with three-dimensional (3-D) super-resolution opticalmicroscopy in order to use optical feedback to measure tip−sample gaps with a dynamic range of ∼2 μm and, in many cases,resolution better than 25 nm. In 3-D super-resolution imaging,an optic such as a cylindrical lens or spatial light modulator isinserted into the emission path of an optical microscope toinduce a distortion on the point spread function (PSF) of anemitter. This distortion depends strongly on the axial position ofthe probe relative to the focus of the microscope.24−29 Here, weuse the 3-D super-resolution approach described by Moernerand co-workers, in which they generated a double helix PSF(DH-PSF) to measure the axial position of a fluorescent emitterwith precision better than 25 nm.30−33 To use this technique formeasuring tip−substrate gaps in SECM, we introduce twoemission sources, one at the substrate surface and the other at theSECM tip, which produce different emission patterns based ontheir relative axial positions. By fitting the DH-PSF of the twoemitters, we are able to calculate the distance between the tip andthe substrate, independent of the electrochemical activity of thesystem and with no dependence on the geometry of thenanoelectrode. For our initial experiments, we used a fluorogenicresorufin redox reaction occurring at the nanoelectrode tomonitor the axial position of the tip, while later experimentsextended this technique to nonfluorescent redox reactions byattaching a fluorescent particle to the protective sheath of theSECM tip. Using this 3-D super-resolution approach, wedemonstrate excellent agreement between the optical andelectrochemical approach curve data, indicating that opticalfeedback provides a robust strategy for measuring tip−samplegaps with large dynamic range and excellent spatial resolution.

■ EXPERIMENTAL SECTIONChemicals and Materials. Ferrocenemethanol (FcMeOH,

97%), resorufin (dye content 95%, λemission = 585 nm),fluorescent nanodiamonds (>800 NV/particle, 100 nm averageparticle size, λemission = 690 nm), potassium chloride (BioXtra,≥99.0%), sodium hydroxide (BioXtra, ≥98%), D-(+)-glucose(≥99.5%, GC), and (3-aminopropyl)triethoxysilane (APTES,≥98%) were purchased from Sigma-Aldrich. Fluorescent skyblue nanospheres (220 nm, λemission = 610 and 660 nm) werepurchased from Spherotech Inc. Ethyl alcohol (200 proof) waspurchased from Pharmco-Aaper. Glass coverslips (25 × 25 mm,0.17 mm thickness) and glass slides were purchased from FisherScientific. Quartz capillaries (1 mm outer diameter, 0.3 mm innerdiameter and 7.5 cm in length) were purchased from Sutterinstruments. A 25 μm diameter Pt wire was purchased fromGoodfellow. Polishing materials were purchased from Ted PellaInc. All solutions were prepared with nanopure water (18 MΩ·cm, arium pro, Sartorius).

Electrochemical Measurements. All electrochemicalmeasurements were performed using a CH750E potentiostat(CH Instruments). A Pt wire and an Ag/AgCl (1 M KCl; CHInstruments) were used as a counter and reference electroderespectively throughout the experiment. A Pt disk electrode withthe diameter of 230 nm (ratio of the insulator sheath radius to thePt disk electrode radius, RG: 40) and 510 nm (RG: 20) werefabricated using a laser puller and polished (see SupportingInformation for detailed procedure). The fabricated electrodeswere characterized electrochemically by performing cyclicvoltammetry with 1 mM FcMeOH in 0.1 M KCl (Figure S-1)and the radius of the electrode was calculated from the diffusion-limited steady state equation (Supporting Information).34

Negative feedback electrochemical approach curves wereobtained using 1 mM FcMeOH in 0.1 M KCl in a homemadespectroelectrochemical cell with a glass coverslip as the substrateand performed on top of an inverted optical microscope (seebelow). The tip electrode was positioned ∼20 μm above thesubstrate using a stepper motor (Microdrive, Mad City LabsInc.). The process is continuously monitored with the opticalmicroscope in a bright-field geometry. A piezo controller(Thorlabs) is employed for further approach of the tip towardsthe substrate with a step size of 40 nm. The current from thebipotentiostat is acquired using a data acquisition card (PCI-6221, National Instruments) and synchronized with the piezocontroller using home written LabView code (National Instru-ments).

3-D Super-Resolution Imaging. Optical measurementswere performed on an inverted microscope (Olympus IX-73)with an Olympus 60× oil immersion objective (OlympusPlanoApo N) in a widefield epi-illumination geometry (Figure1). The objective was attached to a piezo stage (Piezosystemjena), allowing its vertical position (and thus the focal plane ofthe imaging system) to be adjusted in 50 nm steps. A 532 nmlaser excitation (Spectra-Physics, 532−50-CDRH) was intro-duced through the objective to excite fluorescence from thesample, which was then collected by the same objective and

Figure 1. Schematic of experimental setup. The SECM cell containsthree electrodes connected to a bipotentiostat and mounted on aninverted optical microscope with a 60× 1.45 NA oil immersionobjective. A 532 nm laser excites fluorescence and the resulting emissionis imaged on an EM-CCD after passing through a double helix phasemask (DHPM), producing a double lobed emission pattern as shown(bottom left). (Far right) Schematic showing the reactions occurring inthe electrochemical cell: dihydroresorufin is oxidized to fluorescentresorufin at the Pt tip, while resorufin is reduced back todihydroresorufin by glucose in NaOH solution. Fluorescent surfacemarkers (red spheres) are also included to mark the substrate surface.

Analytical Chemistry Article

DOI: 10.1021/acs.analchem.6b04073Anal. Chem. 2017, 89, 922−928

923

filtered with a dichroic/long-pass/notch filter combination with acutoff near 540 nm (Semrock). The emission is then passedthrough a 4f imaging system with a double-helix phase-mask,DHPM (DoubleHelix LLC) in the detection path.33 The phasemask induces a double helix emission pattern, which is imagedonto an electron-multiplied CCD detector (ProEM, PrincetonInstruments). Due to the phase mask, two emission lobes areobserved for a single emitter, and the angle between the twolobes changes as a function of the axial position of the emitterrelative to the focal plane of the objective. The phase mask wascommercially purchased for a system with total magnification of60× and an emission wavelength centered at 665 nm.To create calibration curves relating the angle between the two

lobes to the position of the focal plane of the objective, either theobjective or the SECM tip is moved in steps of 50 nm using theirrespective piezo controllers, and synchronized images areacquired on the CCD camera. For each position, we acquiredfive frames with an acquisition time of 50 ms. The procedure forfitting each image to extract quantitative position information isdescribed below and in more detail in the SupportingInformation.To image the fluorogenic resorufin reaction at the SECM tip, a

solution of 50 μM resorufin and 67 mM D-(+)-glucose in 0.5 MNaOH was prepared fresh before each experiment.35 A 230 nmdiameter platinum nanoelectrode was used as the tip electrode.The resorufin is chemically reduced to nonfluorescentdihydroresorufin by glucose in NaOH, and the Pt nanoelectrodeis biased at an oxidation potential (+0.4 V) to generatefluorescent resorufin at the tip only (Figures 1 and S-2). Toreport on the substrate position, skyblue fluorescent beads ornanodiamonds were attached to the substrate (see SupportingInformation); we refer to these as surface markers in theremainder of this report. Additional experimental details can befound in the Supporting Information.

■ RESULTS AND DISCUSSION

The first step of these experiments is to measure calibrationcurves, in which we relate the emission pattern induced by thedouble helix phase mask to the distance of the emitter from thefocus of the microscope. Figure 2A shows representativeemission patterns from a single nanodiamond surface markeras the focus of themicroscope is changed relative to the substrate.When the surface marker is in focus, it exhibits two lobeshorizontal to each other; we define this angle as 0°. When thesurface marker is below or above the focus of the microscope, theangle between the two lobes rotates, as shown. By fitting eachlobe to a two-dimensional Gaussian function, we are able toextract the angle between the two lobes (Figure S-4). To createthe calibration curve shown in Figure 2B, we move the focus ofthe microscope by 50 nm steps roughly 1 μm above and belowthe substrate surface and calculate the angle between the twolobes. These data then allow us to calculate the distance betweenthe surface marker and focal plane (ds) based on the anglebetween the lobes.Similarly, a calibration curve was obtained for the electrode

using resorufin fluorescence at the electrode tip as shown inFigure 2C,D. The focal plane of the microscope was set roughly3−5 μm above the substrate, and the electrode was held close tothis focal plane while an oxidation potential of +0.4 V was appliedto generate fluorescent resorufin (Figure S-2). When theelectrode is at the focal plane of the microscope, the two lobesassociated with resorufin emission appear roughly horizontal,and they rotate as the electrode is moved in steps of 50 nmroughly 1 μm above and below the focal plane (Figure 2C). Wecreate the calibration curve shown in Figure 2D by stepping theposition of the electrode relative to the fixed focal plane of themicroscope. The calibration curve allows us to calculate thedistance between the electrode and the focal plane (de) based onthe angle between the two lobes. We point out that the two lobesare not as clearly resolved in the images in Figure 2C in

Figure 2. (A) DH-PSF images of a nanodiamond surface marker when it is below, in, and above the focal plane. (B) Optical calibration curve relating theangle between the two lobes in the DH-PSF images and the distance between the surface marker and the focus of the microscope (ds). (C) DH-PSFimages of the resorufin fluorescence at the SECM tip when it is below, in and above the focal plane. The insulting sheath of the SECM tip can also beobserved in these images. Insets show zoomed-in images of the lobes at the electrode. (D)Optical calibration curve of the resorufin fluorescence at the Ptnanoelectrode relating the angle between the two lobes in the DH-PSF images and the distance between the electrode and the focus of the microscope(de).

Analytical Chemistry Article

DOI: 10.1021/acs.analchem.6b04073Anal. Chem. 2017, 89, 922−928

924

comparison to the surface marker images from Figure 2A. This ismost likely due to diffusion of the resorufin probe from theelectrode surface, which causes broadening in the resultingdiffraction-limited spot. To reduce this diffusion-based effect, wework in a solution of glucose in NaOH, which chemically reducesthe resorufin produced at the electrode to nonfluorescentdihydroresorufin, thereby mitigating the impact of diffusion onour images. Further, we also use electrodes for these measure-ments that have diameters at or below the diffraction limit oflight, such that we observe clean double helix point spreadfunction images associated with the resorufin emission. Thiselectrode size restriction is a limitation of monitoring fluorogenicreactions at the electrode surface, which will be addressed inmore detail below.Using the two calibration curves, we can now determine the

tip−substrate distance as the height of the electrode is changedrelative to the substrate. Figure 3 shows a series of images where

both the emission from the resorufin at the tip (yellow circle) andthe emission of a surface marker at the substrate (white circle) areobserved. Note that the two emitters are spatially separated suchthat the two images do not overlap; otherwise, we would beunable to determine their individual angles and therefore axialposition. The focus of the microscope is set between thesubstrate and the initial position of the electrode, as indicated bythe opposite polarity of the lobes in Figure 3A (Figure S-5). Next,the electrode is stepped closer to the surface, as shown in Figure3B−H, causing the angle between the lobes associated with theresorufin emission to rotate (note that the angle between thelobes of the surface markers is constant unless the electrodecomes into contact with the surface, Figure S-6). In panels A−D,the angles between the lobes of the resorufin emission have apositive rotation, indicating that the electrode is above the focal

plane, while in panel E the lobes are roughly horizontal,indicating the electrode is near/in the focal plane. Panels F and Gshow negative angles between the lobes of the resorufin emission,indicating the electrode has passed below the focal plane. Theangle between the lobes of the resorufin and surface markeremission are fit using a 2-D Gaussian model function (Figure S-4) and de and ds are then calculated from the individualcalibration curves. The sum of de and ds provides the tip−substrate distance (Figure S-5), as indicated in each of the panelsof Figure 3.One challenge with using the optical feedback approach we

have described thus far is that it requires a fluorogenic redoxprobe to define the electrode surface, which limits theapplicability of the technique. Moreover, because we use glucoseto quickly oxidize the resorufin produced at the SECM tip backto dihydroresorufin, the currents that wemeasure as we approachthe electrode to the surface do not follow the expected negativefeedback profile, making it difficult to compare our optical data toelectrochemical data. To overcome these limitations andimprove the generality of this technique, we attached afluorescent nanodiamond to the nonconducting glass sheath ofour SECM tip, which provides an optical readout of the tipposition while allowing us to probe any electrochemical reactionof interest at the electrode (see Supporting Information fordetails about nanodiamond attachment). We will refer to thenanodiamonds attached to the sheath of the SECM tip aselectrode markers, for clarity.For correlated 3-D super-resolution optical and electro-

chemical experiments, we immersed the nanodiamond-function-alized electrode (here, a 510 nm diameter Pt electrode) in 1 mMFcMeOH in 0.1 MKCl and applied +0.4 V versus Ag/AgCl (1MKCl). The focal plane of the microscope was set severalnanometers above the substrate (as defined by the surfacemarkers) and the nanodiamond-functionalized tip was allowed toapproach the substrate in steps of 40 nm using the piezocontroller. The current was continuously monitored and whenthe electrode reached ∼1 μm above the focal plane, we started toobserve emission corresponding to the electrode markers (seevideo in the Supporting Information). Figure 4A shows fourrepresentative optical images, indicating the surface marker(white circle) and the brightest electrode marker (yellow circle).The lobes associated with the electrode marker rotate as theelectrode approaches the surface (Figure 4A, panels 1−3) whilethe lobes associated with surface marker remain at a consistentangle. Using calibration curves for this system, we calculate thedistance of both the electrode and the surface marker relative tothe focus (Figure 4B). As expected, ds is constant as the electrodeapproaches, while de changes smoothly as the electrode movesthrough the focal plane. The summation of de and ds results in thetip−substrate distance (1.24, 0.8, and 0.27 μm, respectively, forFigure 4A, panels 1−3). Once the tip hits the substrate, the lobesin the PSF of both the electrode and the surface markers rotate(Figure 4A, panel 4), resulting in changes in both ds and de alongwith significant error in their determination (Figure 4B).Next, we compared the results of the optical experiment to the

electrochemical data that was obtained simultaneously. Thenormalized current versus distance approach curve is shown inFigure 5A. As expected, as the tip approaches the insulatingsubstrate, the tip current gradually decreases, and then abruptlylevels off, indicating that the glass sheath of the tip touched thesubstrate surface. The electrochemical data was fit to negativefeedback theory,36 using the tip radius and RG ratio determinedfrom the steady-state voltammogram and the optical microscope

Figure 3. 3-D super-resolution optical images to determine the tip−substrate distance for various positions of the tip relative to the substrate.The emission in the white circle is from a surface marker at the substrateand the emission in the yellow circle is due to resorufin fluorescence atthe electrode. The calculated tip−substrate distance at various positionare (A) 1.92, (B) 1.51, (C) 0.92, (D), 0.74, (E) 0.50, (F) 0.36, (G) 0.27,and (H) 0.15 μm. Scale bars: 3 μm.

Analytical Chemistry Article

DOI: 10.1021/acs.analchem.6b04073Anal. Chem. 2017, 89, 922−928

925

image, respectively (equation shown in the SupportingInformation). The piezo displacement of zero separationdistance (z0) was used as an adjustable parameter. While thecurvature of the distance-current curve essentially depends onthe tip radius (a) and RG, the z0 value could be confidentlydetermined from the fitting. The tip−substrate separationdistance of each data point was evaluated from the tip currentand compared with optical results (see Supporting Information).Figure 5B shows the tip−substrate separation distancescalculated using both the electrochemical and optical data, andthey show excellent overall agreement, indicating that the opticalapproach is able to reproduce electrochemical data with highfidelity.One additional advantage of our 3-D super-resolution optical

feedback approach is that we are able to calculate the tilt of theelectrode. In traditional SECM experiments, electrode tilt cannotbe measured directly and is instead inferred when an approach

curve deviates significantly from ideal feedback behavior. Usingthe optical approach, if there are multiple markers on the SECMtip, we can measure the vertical distance between them: if theelectrode is tilted, the emission from electrode markers will havedifferent angles between the lobes in the DH-PSF images,indicating that the markers are in different axial planes (Figure S-7), whereas if the electrode is flat, the electrode markers willappear in the same axial plane (Figure S-8). We havedemonstrated the ability to measure tilt angles as large as 22°and as small as 1.8° using this technique, allowing us to haveexcellent control over tilt. Moreover, because we can placemultiple markers on both the substrate and the tip, we are able tomonitor both tilts simultaneously, providing us with the ability toensure that the two electrochemical components are well-alignedrelative to each other.

■ CONCLUSIONIn summary, we have developed a nano SECM combined with a3-D super-resolution optical microscope to determine tip−substrate distance optically with axial resolution better than 25nm. The optical approach curve data agrees with simultaneouslyobtained electrochemical approach data, providing a secondaryfeedback mechanism for SECM tip−sample gap measurements

Figure 4. (A) Optical images in panels 1−4 represent four differentpositions of the electrode, corresponding to the labeled points in panel(B). The white and yellow circles correspond to emission from thesurface and electrode markers, respectively. The optically calculatedtip−substrate distances are indicated in panels 1−3. The angle betweenthe lobes of the surface marker changes when the electrode hits thesurface (panel 4). Scale bars: 5 μm. (B) Calculated distances (angles) ofa surface marker (red triangles) and an electrode marker (blue circles)from the focal plane as a function of piezo positions. The green dottedline indicates the position of the focal plane and the red dashed linerepresents the position at which the tip hits the substrate. The distancesbetween the focal plane and surface marker as well as the focal plane andelectrode marker are shown as ds and de, respectively. The calculateddistances (angles) of the surface and tip electrode markers fluctuate afterthe electrode hits the surface. Note that the piezo displacement increasesrelative to its starting position as the tip gets closer to the substratesurface.

Figure 5. (A) Experimental electrochemical approach curve (symbols)fitted to the theory for pure negative feedback (blue solid line) with RG= 20. The distance is normalized by tip radius, a = 255 nm and the tipcurrent is normalized by i∞ = 75 pA. (B) Tip−substrate separationdistances vs piezo displacement determined from the optical measure-ment (blue squares), SECM approach curve fit (black solid line), and tipcurrents (red triangle). Note that the piezo displacement increasesrelative to its starting position as the tip gets closer to the substratesurface.

Analytical Chemistry Article

DOI: 10.1021/acs.analchem.6b04073Anal. Chem. 2017, 89, 922−928

926

in an experiment. Moreover, our approach allows us to measurethe tilt of both an electrode and a surface without the risk ofelectrode damage.Despite the success of these initial experiments, there are some

improvements that can be made to yield even better axialresolution. For example, our DH-PSF phase mask is optimizedfor a 665 nm emitter, which is red of the emission of both theresorufin and the nanodiamonds. Thus, using a wavelength-optimized phase mask (or a spatial light modulator which allowsdynamic phase mask control) would improve the quality of theoptical images and yield even better resolution. Second, we havetested two different fitting algorithms to extract the anglebetween the two lobes in the DH-PSF: fitting to two separatetwo-dimensional Gaussians (Figure S-4) and using radialsymmetry fitting.37 While the former yielded better precisionin the fits, the latter is significantly faster and would be easier toimplement in a real-time SECM experiment. Thus, finding analgorithm that can determine the angle between the two lobes ofthe different PSFs andmake drift corrections in real time remainsa future improvement to this experiment. Finally, minimizingvibration of the tip is a significant experimental challenge thatfurther limits our resolution. For some of our experiments, weobtain precision better than∼25 nm (e.g., Figure 2), but in othercases, we observe small vibrations in the tip that limit theresolution to ∼50 nm (Figures 4 and 5). Further improvementsto the tip holder and introducing improved vibration isolationwill help address these issues and allow even better axialprecision. However, even with these experimental challenges, webelieve this initial demonstration of correlated 3-D super-resolution and SECM offers a promising avenue towardsimproved performance as we push towards even morenanoresolved electrochemical imaging.

■ ASSOCIATED CONTENT*S Supporting InformationThe Supporting Information is available free of charge on theACS Publications website at DOI: 10.1021/acs.anal-chem.6b04073.

Attachment of fluorescent particles as surface markers tothe substrate, nanoelectrode fabrication, fluorogenicreaction of resorufin occurring at Pt nanoelectrode,attaching fluorescent nanodiamond to the nonconductivesheath of the nanoelectrode, calculation of angle betweenthe double-helix lobes, optical determination of tip−substrate distance, deviation in the position of fluorescentsurface marker when electrode hits the substrate andcalculating the tilt of the nanoelectrode (PDF).Video shows the emission from surface and electrodemarkers, in which the electrode is approaching towards thesubstrate (AVI).

■ AUTHOR INFORMATIONCorresponding Author*Phone: 215-204-7990. E-mail: kwillets@temple.edu.ORCIDBo Zhang: 0000-0002-1737-1241Katherine A. Willets: 0000-0002-1417-4656Present Addresses‡Department of Pathology, Biological Imaging DevelopmentCenter, University of California, San Francisco, CA, UnitedStates (K.M.).§Geico Data Science, Washington, DC.United States (E.J.T.).

∥Department of Chemistry, University of Illinois, Urbana-Campaign, IL, United States (A.J.W.)NotesThe authors declare no competing financial interest.

■ ACKNOWLEDGMENTSWe acknowledge support of this research from the AFOSRMURI (FA9550-14-1-0003).

■ REFERENCES(1) Bard, A.; Mirkin, M. Scanning Electrochemical Microscopy, 2nd ed.;CRC Press, 2012.(2) Amemiya, S.; Chen, R.; Nioradze, N.; Kim, J. Acc. Chem. Res. 2016,49, 2007−2014.(3) Mirkin, M. V.; Sun, T.; Yu, Y.; Zhou, M. Acc. Chem. Res. 2016, 49,2328−2335.(4) Laforge, F. O.; Velmurugan, J.; Wang, Y.; Mirkin, M. V. Anal. Chem.2009, 81, 3143−3150.(5) Kim, J.; Xiong, H.; Hofmann, M.; Kong, J.; Amemiya, S. Anal.Chem. 2010, 82, 1605−1607.(6) Sun, T.; Yu, Y.; Zacher, B. J.; Mirkin, M. V. Angew. Chem., Int. Ed.2014, 53, 14120−14123.(7) Kim, J.; Renault, C.; Nioradze, N.; Arroyo-Curras, N.; Leonard, K.C.; Bard, A. J. J. Am. Chem. Soc. 2016, 138, 8560−8568.(8) Cao, F.; Kim, J.; Bard, A. J. J. Am. Chem. Soc. 2014, 136, 18163−18169.(9) Chang, J.; Bard, A. J. J. Am. Chem. Soc. 2014, 136, 311−320.(10) Sun, P.; Laforge, F. O.; Abeyweera, T. P.; Rotenberg, S. A.;Carpino, J.; Mirkin, M. V. Proc. Natl. Acad. Sci. U. S. A. 2008, 105, 443−448.(11) Takahashi, Y.; Shevchuk, A. I.; Novak, P.; Babakinejad, B.;Macpherson, J.; Unwin, P. R.; Shiku, H.; Gorelik, J.; Klenerman, D.;Korchev, Y. E.; Matsue, T. Proc. Natl. Acad. Sci. U. S. A. 2012, 109,11540−11545.(12) Sun, P.; Mirkin, M. V. Anal. Chem. 2006, 78, 6526−6534.(13) Nioradze, N.; Kim, J.; Amemiya, S. Anal. Chem. 2011, 83, 828−835.(14) Chen, R.; Nioradze, N.; Santhosh, P.; Li, Z.; Surwade, S. P.;Shenoy, G. J.; Parobek, D. G.; Kim, M. A.; Liu, H.; Amemiya, S. Angew.Chem., Int. Ed. 2015, 54, 15134−15137.(15) Bae, J. H.; Yu, Y.; Mirkin, M. V. J. Phys. Chem. C 2016, 120,20651−20658.(16) Kim, J.; Shen, M.; Nioradze, N.; Amemiya, S. Anal. Chem. 2012,84, 3489−3492.(17) Han, L.; Yuan, Y.; Zhang, J.; Zhao, X.; Cao, Y.; Hu, Z.; Yan, Y.;Dong, S.; Tian, Z.-Q.; Tian, Z.-W.; Zhan, D. Anal. Chem. 2013, 85,1322−1326.(18) Edwards, M. A.; Whitworth, A. L.; Unwin, P. R. Anal. Chem. 2011,83, 1977−1984.(19) Hengstenberg, A.; Kranz, C.; Schuhmann,W.Chem. - Eur. J. 2000,6, 1547−1554.(20) Lee, Y.; Ding, Z.; Bard, A. J. Anal. Chem. 2002, 74, 3634−3643.(21) Macpherson, J. V.; Unwin, P. R.; Hillier, A. C.; Bard, A. J. J. Am.Chem. Soc. 1996, 118, 6445−6452.(22) Kueng, A.; Kranz, C.; Lugstein, A.; Bertagnolli, E.; Mizaikoff, B.Angew. Chem., Int. Ed. 2005, 44, 3419−3422.(23)McKelvey, K.; Edwards, M. A.; Unwin, P. R.Anal. Chem. 2010, 82,6334−6337.(24) Huang, B.; Wang, W.; Bates, M.; Zhuang, X. Science 2008, 319,810−813.(25) Jia, S.; Vaughan, J. C.; Zhuang, X. Nat. Photonics 2014, 8, 302−306.(26) Jones, S. A.; Shim, S.-H.; He, J.; Zhuang, X. Nat. Methods 2011, 8,499−505.(27) Shechtman, Y.; Weiss, L. E.; Backer, A. S.; Sahl, S. J.; Moerner, W.E. Nano Lett. 2015, 15, 4194−4199.(28) Gu, Y.; Di, X.; Sun, W.; Wang, G.; Fang, N. Anal. Chem. 2012, 84,4111−4117.

Analytical Chemistry Article

DOI: 10.1021/acs.analchem.6b04073Anal. Chem. 2017, 89, 922−928

927

(29) Grover, G.; Mohrman, W.; Piestun, R. Opt. Express 2015, 23,23887.(30) Backlund, M. P.; Joyner, R.; Weis, K.; Moerner, W. E. Mol. Biol.Cell 2014, 25, 3619−3629.(31) Lee, M. K.; Rai, P.; Williams, J.; Twieg, R. J.; Moerner, W. E. J. Am.Chem. Soc. 2014, 136, 14003−14006.(32) Thompson, M. A.; Lew, M. D.; Badieirostami, M.; Moerner, W. E.Nano Lett. 2010, 10, 211−218.(33) Pavani, S. R. P.; Thompson, M. A.; Biteen, J. S.; Lord, S. J.; Liu, N.;Twieg, R. J.; Piestun, R.; Moerner, W. E. Proc. Natl. Acad. Sci. U. S. A.2009, 106, 2995−2999.(34) Bard, A. J.; Faulker, L. R. Electrochemical Methods, 2nd ed.; Wiley:New York, 2001.(35) Oja, S. M.; Guerrette, J. P.; David, M. R.; Zhang, B. Anal. Chem.2014, 86, 6040−6048.(36) Cornut, R.; Lefrou, C. J. Electroanal. Chem. 2007, 604, 91−100.(37) Parthasarathy, R. Nat. Methods 2012, 9, 724−726.

Analytical Chemistry Article

DOI: 10.1021/acs.analchem.6b04073Anal. Chem. 2017, 89, 922−928

928