In the previous chapters, various applications of microfluidic devices for a broad range of research areas including neuroscience and pharmacology have been presented. The combination of microfluidics with scanning probe microscopy (SPM) has been introduced as well. For instance, laterally translated micropipettes were used for cell patterning by the controlled and localized deposition of cell suspensions as discussed in Chapter 12. The localized mass spectrometry (MS) analysis of lipids extracted from live cells was demonstrated in Chapter 13. In Chapters 14 and 15, the so-called FluidFM was discussed, which is based on the implementation of microfluidics into tips for atomic force microscopy (AFM) with the aim to investigate and manipulate single live cells.
SPMs, such as AFM, scanning ion conductance microscopy (SICM), or scanning electrochemical microscopy (SECM), allow the precise positioning and scanning of a micro- or nanometric tip over a sample area of interest. The morphology as well as local reactivity of the sample can be detected (reading mode). Alternatively, patterns can be created on the sample, for instance, through the delivery or generation of local reactants or by a mechanically induced surface manipulation (writing mode). In SECM, amperometric microelectrodes (MEs) are typically used in an electrolyte solution to measure local surface reactivity through the detection of redox-active species generated or converted at active surface sites [1–3].
Recently, we have combined microfluidics and SECM by developing so-called soft probes that contain amperometric MEs and/or microfluidic channels. In this chapter, we will first introduce briefly the fundamentals, applications, and limitations of conventional SECM, followed by a detailed discussion about the soft SECM probes and the possibility to work on large extended samples. A major advantage of these soft probes is that they can be brushed in a gentle contact mode over rough, tilted, and delicate surfaces without inducing detectable damages while keeping a constant working distance. The implementation of the microfluidic channels for the controlled delivery or aspiration of solutions to work on initially dry surfaces for creating in-solution local pH changes and for analyzing sample extracts by complementary techniques (e.g., MS) will be discussed in Chapter 17.
SECM has been used for a wide range of applications including the investigation of live cells [4–7], screening of photo/electrocatalyst libraries , studying corrosion and corrosion inhibitors [9, 10], and analyzing the molecular transport across membranes . Herein, only a brief overview about the principles and applications of SECM can be given, and the interested reader is directed to several reviews on SECM [3, 12–14] and the recent edition of the seminal book Scanning Electrochemical Microscopy .
Generally, an ME (often also denoted as tip “T”) is scanned in an electrolyte solution vertically or horizontally in close proximity to a sample surface. The ME is typically composed of a microdisc made up of Pt, Au, or carbon fiber embedded in an insulating material such as glass. The electrolyte solution contains redox-active species that can interact with the sample or that are generated at active surface sites. A supporting electrolyte is added to decrease the solution resistance and to assure the measured current is based only on diffusion. The ME is biased at a potential where the present redox-active species is electrochemically converted. Due to the size of the microdisc (the radius rT is usually smaller than 12.5 µm), a hemispherical diffusion profile is formed around the ME, resulting in a diffusion-limited steady-state current iT,∞. Equation (16.1) is valid for a disc-shaped ME with infinite insulating sheath:
where n is the number of transferred electrons per molecule of the redox-active species, F the Faraday constant, D the diffusion coefficient of the redox-active species, and c* the bulk concentration of the redox-active species.
The basic SECM setup consists of a precise x,y,z-positioning system in order to control the relative position of the ME to the sample, which is placed at the bottom of an electrochemical cell. A (bi)potentiostat is used to apply a working potential at the ME acting as working electrode and optionally the identical or a different potential value at the substrate versus a reference and a counter electrode. The currents at both the ME and the sample are recorded as a function of the ME position.
Different SECM operation modes were developed from which the feedback and the generation/collection (G/C) modes are the most frequently used ones . They are introduced in the following two sections.
For feedback mode operation, a redox mediator, such as ferrocenemethanol (FcMeOH), is added to the electrolyte solution. The electrochemical oxidation according to the reaction R → O + ne− occurs at the ME under diffusion control when the right electrode potential ET is applied. R represents FcMeOH and O the oxidized from, that is, FcMeOH+. When the ME is positioned in bulk solution, that is, at a large distance between the tip and the substrate (Figure 16.1a), iT,∞ is recorded. If the ME is approached toward a conductive or chemically active substrate within a distance of a few tip radii, the species O can immediately be reduced back to R at the sample (Figure 16.1b). This recycling process, called “positive feedback,” causes an additional flux of species R to the ME. The result is a significant increase of the recorded current at the ME (iT) compared with the current measured in bulk solution (iT > iT,∞) . On the contrary, when the probe is approached to an electrically insulating or non-active substrate, the diffusion of species R toward the ME is physically hindered by the substrate lowering the flux of species R to the ME and consequently iT decreases (iT < iT,∞). This process is often called “negative feedback” (Figure 16.1c) . The recorded SECM signals depend on the ratio of the insulating glass sheath rglass and rT, called the RG value. A large RG value blocks more efficiently the diffusion of species R from the bulk toward the ME but also perturbs more the diffusion layers.
Figure 16.1d shows the normalized current–distance curves where the normalized tip current IT = iT/iT,∞ is plotted versus the normalized working distance L = d/rT where d is the measured working distance. The two limiting cases of positive and negative feedback are indicated in Figure 16.1d. However, the redox mediator regeneration at the sample can be kinetically limited, and the reactions at the sample take place with a finite rate. The corresponding current–distance curves of finite kinetics lie between the two limiting cases of negative and positive feedback. These curves are characterized by a dimensionless heterogeneous reaction rate constant κ = keff·rT/D where keff is the heterogeneous reaction rate for a first-order reaction with respect to O.
Analytical approximations developed by using numerical simulations are available in the literature [18–21]. They can be used to fit experimental approach curves to the theory in order to determine κ, which can then be used to calculate keff. For an inkjet-printed and thermally cured gold electrode, keff becomes 0.042 cm/s (DFcMeOH = 7.8·10−6 cm2/s , rT = 12.5 µm, RG = 6.8, κ = 6.73) using the theory from Cornut and Lefrou . Pure positive feedback (κ → ∞) has not been observed over the printed gold electrode most likely due to the influence of some particle stabilizers that remained after curing. Hindered diffusion (κ = 0) is obtained over glass. For the feedback mode imaging, a constant working distance is essential to measure a current contrast between different surface reactivities as characterized by κ (see Figure 16.1d).
One of the two SECM G/C modes is the substrate generation/tip collection (SG/TC) mode where an analyte, initially not present in the solution, is generated at the substrate and diffuses to the ME where it is electrochemically collected. Consequently, iT reflects the concentration profile of the redox-active species above a reactive sample spot. The SG/TC mode can be used for monitoring enzymatic reactions  or detecting the metabolic response of living cells, which released chemical species to the extracellular matrix (ECM) .
The second G/C mode is the tip generation/substrate collection (TG/SC) mode where the redox-active species is generated at the ME and collected at the sample. It is often used for screening electrocatalysts for the oxygen reduction reaction (ORR) where O2 is generated at the ME and reduced at the catalyst spot .
The G/C modes show a larger sensitivity compared with feedback mode experiments due to a background-free signal, but the lateral resolution is lower. If the sample spots are too large, the diffusion layers around the spots grow continuously, and thus the SECM signal becomes time dependent. Typically, the working distance in G/C experiments is relatively large in order to avoid redox mediator regeneration and hence feedback effects.
Conventional SECM experiments are carried out in constant height mode, which poses limitations with respect to the requirement of a constant working distance. As it can be seen in Figure 16.2a, scanning over tilted or topographic samples leads to a change of d, which in consequence varies IT, even for the same κ. Therefore, reactivity and topography information need to be decoupled. In addition, the probe and the sample can irreparably be damaged during a mechanical probe-sample contact. Flat but tilted substrates can be aligned with a tilt table by extracting relative height differences in the x–y-plane from approach curves. However, for large, corrugated, and rough samples, this procedure becomes pointless.
A constant d can be achieved mainly by three strategies: (i) combining SECM with other techniques such as AFM , electrochemical scanning tunneling microscopy (EC-STM) , SICM , or shear-force detection , including a tip modification; (ii) measuring signals with the ME that can be correlated to topography, such as AC-SECM ; and (iii) employing soft probes in contact mode . For a comprehensive overview, the reader is directed to some recent reviews [32, 33]. The soft SECM probes made of thin flexible materials were introduced in 2009 by the groups of Girault and Wittstock . Upon mechanical contact, the probes bend and can be brushed in contact mode over irregularly curved and tilted substrates (Figure 16.2a). The setup itself operates in standard constant height mode configuration and does not require additional constant distance accessories.
Soft probes comprised up to eight microchannels and are produced by photoablation with an excimer laser (193 nm) into polyethylene terephthalate (PET) sheets (100 µm thick or less). A sickle-like microchannel profile of approximately 15 µm depth and 30 µm width is achieved. The channels are filled with a carbon paste (Figure 16.2b). Alternatively, aerosol jet printing of Au or inkjet printing of various conductive materials can be used directly on a plastic support without the microchannel approach . After thermal curing, the conductive electrode traces are sealed with a thin polyethylene (PE)/PET or parylene C film [34, 36]. The active ME areas are exposed either by razor blade or by laser-assisted cutting (Figure 16.2b). Before each experiment a new cut can be made to obtain a fresh electrode surface, avoiding polishing and cleaning procedures as required for conventional MEs. Single soft MEs are called “soft stylus probes” (Figure 16.2c), while paralleled probes are denoted “soft linear microelectrode arrays” (in the following simply “soft arrays”; Figure 16.2d).
The soft probes are assembled in a specific probe holder to adjust an inclination angle γ with respect to the surface normal (Figure 16.3a). The PE/PET or parylene C-coated site faces the substrate surface. Feedback mode approach curves of a soft stylus probe in 2 mM FcMeOH solution over an insulating (glass) and a conductive (gold) sample are shown in Figure 16.3a. As expected, an increase of iT was obtained over the gold film due to the diffusion-controlled regeneration of FcMeOH (positive feedback). Negative feedback was observed when the soft probe approached the insulating glass due to hindered diffusion. At a certain point, the probe contacts the substrate. While approaching further, the probe starts to slide with assistance of γ, which at 20° enables stable scanning with a good current contrast. The vertical position of the soft probe is described by the geometric parameter hp (Figure 16.3a) that equals hA − lT. The attachment point hA is defined as the height between the sample surface and the position of the probe holder (varies with z), whereas lT describes the vertical length of the probe. Similarly to d, hp decreases when the soft probe approaches the sample. By definition, hp becomes zero when the probe contacts the substrate and becomes negative when approached further. The working distance dc in contact mode is defined by the angle α and the thickness of the sealing layer tL (see schemes and equations in Figure 16.3a).
SECM imaging is performed by translating the probe in the horizontal x–y-plane while recording a current at defined grid points. The scanning mode using the soft probes is different from typical SPMs where the probe is moved forwardly and reversely in an identical manner. Due to the predefined bending direction, the soft probe needs to be retracted using a lift-off procedure after a contact mode line scan and while repositioning the probe with a small displacement perpendicularly to the scanning direction (Figure 16.3b). In this way, overbending and mechanical stress to the probe as well as to the substrate do not occur. In order to avoid double scanning of surface areas with the adjacent MEs of a soft array (the typical step size is 5–10 µm, ME separation 500 µm), a larger displacement is performed to image a new frame next to the previously measured region.
Although many approaches have been reported to apply SECM and related techniques with nanometric tips to investigate electroanalytical and electrocatalytical phenomena on the nanoscale [33, 37–40], there is still a deep interest in studying samples as large as square centimeters such as combinatorial material libraries or tissues. In the following, three main applications of the soft SECM probes are introduced.
Conventional SECM is operated under diffusion control, which limits the probe translation rates to the lower micrometers per second scale . Faster probe movements induce convective disturbances, lowering the image resolution. Therefore, the imaging time for square centimeter-sized areas can exceed 24 h, causing experimental difficulties arising from electrode fouling, electrolyte evaporation, or sample aging. One solution would be imaging subregions of a large sample with intermittent renewal of the electrolyte solution and electrode surfaces. But this concept increases further the experimental time, and a renewal of the sample surface is difficult. A larger step size could principally be applied but will lower significantly the lateral resolution. Arrays of individually addressable MEs as SECM probes overcome these experimental drawbacks by enlarging the scanned image area per time unit by the number of array electrodes while compromising the image quality [31, 42, 43].
Figure 16.4a shows the original feedback mode SECM image in FcMeOH solution of a gold microstructure on glass by using a soft array with eight MEs (250 µm electrode separation) . A sample area of 16 mm2 was imaged. The working potential ET of all MEs was 0.3 V versus an Ag quasi-reference electrode, and the current values at each ME were recorded in parallel. The gold microbands were identified with high resolution, but the current scales of each ME do not correspond well to each other. This is due to differences in electrode size and shape upon mechanical cutting of the probe cross section. To compensate for this, a current offset iT,offset,i is subtracted from the measured currents of each ME i, and a dimensionless scale factor si is employed to obtain a calibrated current i′T = (iT − iT,offset,i) × si . Slight positional offsets of the individual MEs are corrected with the help of small topographic or reactive sample textures, such as a cross, located next to the sample area of interest.
Figure 16.4b shows the final corrected surface reactivity image. Recently, the calibration of current values was simplified by leveling the negative feedback response to zero using iT,offset,i and converting the positive feedback current to unity with si. This procedure was applied for scanning a large curved metallic pin with insulated and graved letters forming the text “Carl von Ossietzky University of Oldenburg” (Figure 16.4c–e). Although the size of the sample was as large as 1.5 × 0.8 cm2 and showed significant topographic features, the imaging time was 7 h with a point density of 1603 mm−2 and 2 h for 401 points·mm−2, using a soft array with eight MEs in feedback mode (500 µm electrode separation). These imaging times cannot be achieved using single electrodes.
Besides imaging, SECM can be applied for surface modification in many different ways. A comprehensive review can be found elsewhere . Generally, local surface modification can be initiated by electrogeneration of reactive compounds, such as bromine, at the ME . A quantitative control of the modification result is enabled by adjusting d and the electrolysis current as well as electrolysis time.
Soft arrays reduce considerably the process time by the number of individually addressable MEs. Another important advantage of the soft probes is that they exert a relatively low pressure onto the sample surface, allowing the operation on samples as delicate as self-assembled monolayers (SAMs) . An estimated value of 100 N/m is some orders of magnitude lower compared with conventional AFM tips and corresponding AFM modes . This is the result of the relatively large contact area between the soft probe and the substrate together with the flexibility of the thin polymeric films of which the probes are made of.
The groups of Wittstock and Girault used soft arrays with eight carbon MEs to “write” and “read” on an oligo(ethylene glycol) (OEG)-terminated SAM on Au by the electrogeneration of Br2/HOBr . The Br2/HOBr degrades the OEG units and leaves terminal alkyl chains on the Au surface [49, 50]. The alkyl units degrade with a much slower rate compared with the OEG units. In the pristine state, the OEG SAM shows cytophobic (i.e., cell-repellent) properties, while after modification ECM proteins for cell patterning can be adhered. Typically, the size of the patterns corresponds to the lateral dimensions of various types of living cells. Switching surfaces from cell repellent to cell adhesive by generating Br2/HOBr at MEs had been shown first by Nishizawa and coworkers [51, 52] and found many further applications.
One 0.1 M phosphate-buffered solution (pH 7) containing both 50 mM KBr as source for Br2/HOBr and 1 mM [Ru(NH3)6]Cl3 as redox mediator for subsequent SECM feedback mode imaging was used. By biasing the MEs at an “on” potential (1.8 V vs an Ag quasi-reference electrode), the chemical degradation of the OEG units by electrogenerated Br2/HOBr set in. Three “writing” procedures were developed: (i) creating arrays of micrometric spots (ET,on is applied for a few seconds with the probe resting), (ii) drawing lines (ET,on is applied while the probe is moved), and (iii) creating complex patterns (ET,on is applied at specific positions as defined in a detailed patterning file). The latter procedure is schematically shown in Figure 16.5a to write alphabetic characters . The array probe is resting and the modification pulse is only applied to certain electrodes while the remaining electrodes are biased at an “off” potential ET,off (0.1 V). During the probe translation, no reaction is initiated at any ME. Drawing complex structures was enabled by using a small step size in between each modification step so that an overlap of adjacent modified spots was achieved. By performing a line-by-line process similarly to SECM imaging, an optical photograph, which was previously converted into a two-color code black–white image, could be drawn onto the sample. The number of pixels in the photograph was reduced to 69 × 80 (Figure 16.5b) to match the dimensions of the modified spot size. Consequently, each of the MEs in the array with an electrode separation of 500 µm used 69 resting positions for the modification during each of 10 modification line scans (Figure 16.5c). The pulse length was 5 s, resulting in a process time of 1.5 h for the patterning.
Subsequently, the result was visualized by SECM feedback mode imaging where all MEs were biased at –0.3 V for the diffusion-controlled reduction of [Ru(NH3)6]3+ to [Ru(NH3)6]2+. It is known that [Ru(NH3)6]3+ is regenerated at the Au surface under the modified OEG surface causing a larger current while the pristine OEG SAM is impermeable for the redox mediator. A high resolution image was obtained within 6.5 h using three MEs in a soft array (Figure 16.5d). Furthermore, the successful transformation from the hydrophilic OEG SAM into a hydrophobic alkyl-terminated monolayer was confirmed by the site-selective adsorption of fluorescently labeled proteins . The parallel modification and imaging using soft arrays demonstrate an elegant possibility of high-throughput patterning and readout by SECM.
As mentioned earlier, SECM has been applied on a broad range to probe the biochemical activity of living cells [5, 6, 53–57]. However, it has not been widely employed for tissue imaging mainly due to several SECM limitations discussed previously as well. The soft SECM probes are ideal for the contact mode scanning of tissues. Topographic artifacts are irrelevant due to the gentle brushing in contact mode, and, most importantly, the tissue material is not affected by the mechanical impact (Figure 16.6a). This has been demonstrated for skin cancer tissues from biopsies , which represents a promising approach for a diagnostic SECM application. Indeed, cutaneous melanoma is the most aggressive type of skin cancer that strikes more than hundred thousand people around the world per year . The survival rate is strongly linked to the stage of the cancer when it is diagnosed [60, 61]. Based on this diagnosis, medical treatments may vary. In consequence, it is inevitable to distinguish between the early non-metastatic stages (stages 0, I, and II) and later metastatic stages (stages III and IV) [60, 61]. Skin biopsy is usually performed to prepare tissue sections for a pathological analysis using optical microscope techniques. In order to reveal the presence of specific cancer biomarkers, staining procedures are carried out using colorimetry of samples treated with immunohistochemistry (IHC) [62, 63] or employing fluorescence in situ hybridization (FISH) .
However, optical techniques suffer from one major drawback, namely, optical interferences. For melanoma, this can be a result of the presence of melanin, which is a color pigment in the skin. Fluorescence detection can suffer from autofluorescence and photobleaching . In contrast, SECM in the SG/TC mode detects electrochemically specific analytes that are exclusively generated by cancer biomarkers present inside or linked to the sample. The group of Girault has developed a strategy to map with the soft probes the local distribution of the prognostic indicator tyrosinase (TyR), which is overexpressed in melanoma . For the SECM detection, an immunoassay for TyR was applied to tissue sections of different melanoma stages and normal skin from nine patients, similarly to optical detection methods. Primary antibodies (Abs) were used to recognize TyR, while secondary Abs labeled with horseradish peroxidase (HRP) binds the primary Abs. The sample solution contained hydrogen peroxide (H2O2) and tetramethylbenzidine (TMB). The oxidation of TMB in presence of H2O2 is catalyzed by HRP, and the reaction product TMBox can electrochemically be reduced at the SECM probe, causing the recorded SECM signal (Figure 16.6b) . A significant current is only recorded over positions where the tissue section contains a certain amount of TyR.
Figure 16.7a shows an SECM image of stage II and stage III melanoma and normal skin tissues recorded with a soft stylus probe in contact mode. Generally, the highest expression level of TyR was observed in stage II tissue samples as indicated by the higher SECM currents over the according regions. A lower level of TyR was found in normal skin, where TyR is produced at moderate levels by normal melanocytes and is mainly located at the basal layer of the epidermis (close to the dermis). Furthermore, the TyR distribution in stage II was clearly homogeneous, whereas a heterogeneous distribution was observed in stage III. This observation is supported by reports in literature . During the progression of tumor metastasis, the tumor marker concentration might decrease in order to facilitate metastasis and proliferation [69, 70]. This explains that a heterogeneous distribution and a slightly decreased TyR concentration are detected in stage III melanoma. In the normal skin tissue, the TyR content increased slightly toward the basal epidermis where normal melanocytes are located. The SECM results were compared with the optical images obtained from an IHC (Figure 16.7b–d) where TyR was labeled with a specific pink-colored chromogen. The detection was strongly interfered by melanin, particularly in stage III melanoma (Figure 16.7c). The average current values from nine representative locations on each tissue section reflect the estimated global TyR concentration, which would electrochemically be detected with laterally less resolved techniques. Indeed, stage II melanoma shows the highest average current (left bar in Figure 16.7e), followed by normal skin (right bar) and then stage III melanoma. Importantly, the lower averaged current for stage III melanoma could be misinterpreted as normal skin. Therefore, the SECM image together with its 2D plot (Figure 16.7f) shows unequivocally the TyR distribution in the different melanoma stages. It is important to point out that conventional SECM imaging in constant height mode with MEs made in glass is limited by the height differences of tissue samples and the fragility of the biomaterial in the case of a mechanical contact between probe and tissue. In the future, the soft probe-based screening strategy could be implemented directly or complementarily as a diagnostic tool for skin cancer screening. Soft arrays are supposed to accelerate the analysis to screen cancer tissue libraries, drug distribution in cancers, or disorders of redox metabolism. The weak forces exerted by the brushing probe indicate that it might be possible to image live cells and to study the impact of dynamic stimuli on cells.
Soft ME probes for gentle contact mode scanning of fragile and extended samples broaden the field of SECM applications. The contact point of the probe with the sample can easily be identified without the risk to damage the probe or the sample. Scanning large surface areas such as tissue sections can be performed rapidly without the need to level the substrate surface. Furthermore, no additional equipment for constant distance mode operation is required. The implementation of soft arrays enables high-throughput imaging and surface modification while keeping the high resolution of SECM under diffusion-controlled conditions. In the newest generation of soft arrays, the MEs are shaped as individual “fingers” acting independently from each other. Such probes will be valuable for the investigation of tumor blocks and possibly for performing experiments directly on human or animal skin.
The integration of microfluidic channels into the soft probes is discussed in detail in the next chapter. Such probes are intended to deliver various reagents or drugs for medical treatment or for manipulating microenvironments of cells and tissues close to the MEs while performing electrochemical and chemical analyses in parallel [71–73].
- 1 Bard, A.J., Fan, F.R.F., Kwak, J., and Lev, O. (1989) Anal. Chem., 61, 132–138.
- 2 Bard, A.J., Fan, F., Pierce, D.T., Unwin, P.R., Wipf, D.O., and Zhou, F. (1991) Science, 254, 68–74.
- 3 Wittstock, G., Burchardt, M., Pust, S.E., Shen, Y., and Zhao, C. (2007) Angew. Chem. Int. Ed., 46, 1584–1617.
- 4 Yasukawa, T., Kondo, Y., Uchida, I., and Matsue, T. (1998) Chem. Lett., 27, 767–768.
- 5 Liu, B., Rotenberg, S.A., and Mirkin, M.V. (2000) Proc. Natl. Acad. Sci. U.S.A., 97, 9855–9860.
- 6 Bergner, S., Vatsyayan, P., and Matysik, F.M. (2013) Anal. Chim. Acta, 775, 1–13.
- 7 Holzinger, A., Steinbach, C., and Kranz, C. (2016) in Electrochemical Strategies in Detection Science (ed. D.W.M. Arrigan), Royal Society of Chemistry, pp. 125–169.
- 8 Rodríguez-López, J., Zoski, C.G., and Bard, A.J. (2012) in Scanning Electrochemical Microscopy, 2nd edn (eds A.J. Bard and M.V. Mirkin), CRC Press, pp. 525–568.
- 9 Simões, A.M., Bastos, A.C., Ferreira, M.G., González-García, Y., González, S., and Souto, R.M. (2007) Corros. Sci., 49, 726–739.
- 10 Izquierdo, J., Santana, J.J., González, S., and Souto, R.M. (2010) Electrochim. Acta, 55, 8791–8800.
- 11 White, H.S. and Kanoufi, F. (2012) in Scanning Electrochemical Microscopy, 2nd edn (eds A.J. Bard and M.V. Mirkin), CRC Press, pp. 233–274.
- 12 Mirkin, M.V. and Horrocks, B.R. (2000) Anal. Chim. Acta, 406, 119–146.
- 13 Amemiya, S., Bard, A.J., Fan, F.R.F., Mirkin, M.V., and Unwin, P.R. (2008) Annu. Rev. Anal. Chem., 1, 95–131.
- 14 Mirkin, M.V., Nogala, W., Velmurugan, J., and Wang, Y. (2011) Phys. Chem. Chem. Phys., 13, 21196–21212.
- 15 Bard, A.J. and Mirkin, M.V. (eds) (2012) Scanning Electrochemical Microscopy, 2nd edn, CRC Press.
- 16 Cornut, R. and Lefrou, C. (2008) J. Electroanal. Chem., 621, 178–184.
- 17 Wipf, D.O. and Bard, A.J. (1991) J. Electrochem. Soc., 138, 469–474.
- 18 Mirkin, M.V., Fan, F.R.F., and Bard, A.J. (1992) J. Electroanal. Chem., 328, 47–62.
- 19 Amphlett, J.L. and Denuault, G. (1998) J. Phys. Chem. B, 102, 9946–9951.
- 20 Wei, C., Bard, A.J., and Mirkin, M.V. (1995) J. Phys. Chem., 99, 16033–16042.
- 21 Kwak, J. and Bard, A.J. (1989) Anal. Chem., 61, 1221–1227.
- 22 Zoski, C.G. (2002) Electroanalysis, 14, 1041–1051.
- 23 Wittstock, G. (1997) Anal. Chem., 69, 5059–5066.
- 24 Amemiya, S., Guo, J., Xiong, H., and Gross, D.A. (2006) Anal. Bioanal. Chem., 386, 458–471.
- 25 Fernández, J.L. and Bard, A.J. (2003) Anal. Chem., 75, 2967–2974.
- 26 Kranz, C., Friedbacher, G., Mizaikoff, B., Lugstein, A., Smoliner, J., and Bertagnolli, E. (2001) Anal. Chem., 73, 2491–2500.
- 27 Treutler, T.H. and Wittstock, G. (2003) Electrochim. Acta, 48, 2923–2932.
- 28 Takahashi, Y., Shevchuk, A.I., Novak, P., Murakami, Y., Shiku, H., Korchev, Y.E., and Matsue, T. (2010) J. Am. Chem. Soc., 132, 10118–10126.
- 29 Ballesteros Katemann, B., Schulte, A., and Schuhmann, W. (2003) Chem. Eur. J., 9, 2025–2033.
- 30 Etienne, M., Schulte, A., and Schuhmann, W. (2004) Electrochem. Commun., 6, 288–293.
- 31 Cortés-Salazar, F., Momotenko, D., Girault, H.H., Lesch, A., and Wittstock, G. (2011) Anal. Chem., 83, 1493–1499.
- 32 O'Connell, M.A. and Wain, A.J. (2015) Anal. Methods, 7, 6983–6999.
- 33 Kranz, C. (2014) Analyst, 139, 336–352.
- 34 Cortés-Salazar, F., Träuble, M., Li, F., Busnel, J.-M., Gassner, A.-L., Hojeij, M., Wittstock, G., and Girault, H.H. (2009) Anal. Biochem., 81, 6889–6896.
- 35 Lesch, A., Momotenko, D., Cortés-Salazar, F., Wirth, I., Tefashe, U.M., Meiners, F., Vaske, B., Girault, H.H., and Wittstock, G. (2012) J. Electroanal. Chem., 666, 52–61.
- 36 Cortés-Salazar, F., Lesch, A., Momotenko, D., Busnel, J.M., Wittstock, G., and Girault, H.H. (2010) Anal. Methods, 2, 817–823.
- 37 Sun, T., Yu, Y., Zacher, B.J., and Mirkin, M.V. (2014) Angew. Chem. Int. Ed., 53, 14120–14123.
- 38 Kang, M., Perry, D., Kim, Y.R., Colburn, A.W., Lazenby, R.A., and Unwin, P.R. (2015) J. Am. Chem. Soc., 137, 10902–10905.
- 39 Zoski, C.G. (2016) J. Electrochem. Soc., 163, H3088–H3100.
- 40 Clausmeyer, J. and Schuhmann, W. (2016) TrAC Trends Anal. Chem., 79, 46–59.
- 41 Combellas, C., Fermigier, M., Fuchs, A., and Kanoufi, F. (2005) Anal. Chem., 77, 7966–7975.
- 42 Barker, A.L., Unwin, P.R., Gardner, J.W., and Rieley, H. (2004) Electrochem. Commun., 6, 91–97.
- 43 Kanno, Y., Ino, K., Inoue, K.Y., Şen, M., Suda, A., Kunikata, R., Matsudaira, M., Abe, H., Li, C.-Z., Shiku, H., and Matsue, T. (2015) J. Electroanal. Chem., 741, 109–113.
- 44 Cortés-Salazar, F., Momotenko, D., Lesch, A., Wittstock, G., and Girault, H.H. (2010) Anal. Chem., 82, 10037–10044.
- 45 Lesch, A., Momotenko, D., Cortés-Salazar, F., Roelfs, F., Girault, H.H., and Wittstock, G. (2013) Electrochim. Acta, 110, 30–41.
- 46 Mandler, D. (2012) in Scanning Electrochemical Microscopy, 2nd edn (eds A.J. Bard and M.V. Mirkin), CRC Press, pp. 489–524.
- 47 Mandler, D. and Bard, A.J. (1989) J. Electrochem. Soc., 136, 3143–3144.
- 48 Lesch, A., Vaske, B., Meiners, F., Momotenko, D., Cortés-Salazar, F., Girault, H.H., and Wittstock, G. (2012) Angew. Chem. Int. Ed., 51, 10413–10416.
- 49 Zhao, C., Witte, I., and Wittstock, G. (2006) Angew. Chem. Int. Ed., 45, 5469–5471.
- 50 Zhao, C., Zawisza, I., Nullmeier, M., Burchardt, M., Träuble, M., Witte, I., and Wittstock, G. (2008) Langmuir, 24, 7605–7613.
- 51 Kaji, H., Tsukidate, K., Matsue, T., and Nishizawa, M. (2004) J. Am. Chem. Soc., 126, 15026–15027.
- 52 Kaji, H., Kanada, M., Oyamatsu, D., Matsue, T., and Nishizawa, M. (2004) Langmuir, 20, 16–19.
- 53 Barker, A.L., Gonsalves, M., Macpherson, J.V., Slevin, C.J., and Unwin, P.R. (1999) Anal. Chim. Acta, 385, 223–240.
- 54 Takahashi, Y., Shevchuk, A.I., Novak, P., Babakinejad, B., Macpherson, J., Unwin, P.R., Shiku, H., Gorelik, J., Klenerman, D., and Korchev, Y.E. (2012) Proc. Natl. Acad. Sci. U.S.A., 109, 11540–11545.
- 55 Mauzeroll, J. and Schougaard, S.B. (2012) in Scanning Electrochemical Microscopy, 2nd edn (eds A.J. Bard and M.V. Mirkin), CRC Press, pp. 489–524.
- 56 Nebel, M., Grützke, S., Diab, N., Schulte, A., and Schuhmann, W. (2013) Angew. Chem. Int. Ed., 52, 6335–6338.
- 57 Rapino, S., Marcu, R., Bigi, A., Soldà, A., Marcaccio, M., Paolucci, F., Pelicci, P.G., and Giorgio, M. (2015) Electrochim. Acta, 179, 65–73.
- 58 Lin, T.E., Bondarenko, A., Lesch, A., Pick, H., Cortés-Salazar, F., and Girault, H.H. (2016) Angew. Chem. Int. Ed., 55, 3813–3816.
- 59 Taylor, J.-S. (2015) Science, 347, 824.
- 60 Gray-Schopfer, V., Wellbrock, C., and Marais, R. (2007) Nature, 445, 851–857.
- 61 Rothberg, B.E.G., Bracken, M.B., and Rimm, D.L. (2009) J. Natl. Cancer Inst., 101, 452–474.
- 62 Taylor, C. and Levenson, R. (2006) Histopathology, 49, 411–424.
- 63 Shi, S.-R., Liu, C., and Taylor, C.R. (2007) J. Histochem. Cytochem., 55, 105–109.
- 64 Bernet, D., Schmidt, H., Meier, W., Burkhardt-Holm, P., and Wahli, T. (1999) J. Fish Dis., 22, 25–34.
- 65 Monici, M. (2005) Biotechnol. Annu. Rev., 11, 227–256.
- 66 Orchard, G.E. (2000) Histochem. J., 32, 475–481.
- 67 Lin, T.-E., Cortés-Salazar, F., Lesch, A., Qiao, L., Bondarenko, A., and Girault, H.H. (2015) Electrochim. Acta, 179, 57–64.
- 68 Hofbauer, G.F.L., Kamarashev, J., Geertsen, R., Böni, R., and Dummer, R. (1998) J. Cutan. Pathol., 25, 204–209.
- 69 Alexander, P. (1974) Cancer Res., 34, 2077–2082.
- 70 Igney, F.H. and Krammer, P.H. (2002) J. Leukocyte Biol., 71, 907–920.
- 71 Bondarenko, A., Cortés-Salazar, F., Gheorghiu, M., Gaspar, S., Momotenko, D., Stanica, L., Lesch, A., Gheorghiu, E., and Girault, H.H. (2015) Anal. Chem., 87, 4479–4486.
- 72 Momotenko, D., Cortés-Salazar, F., Lesch, A., Wittstock, G., and Girault, H.H. (2011) Anal. Chem., 83, 5275–5282.
- 73 Momotenko, D., Qiao, L., Cortés-Salazar, F., Lesch, A., Wittstock, G., and Girault, H.H. (2012) Anal. Chem., 84, 6630–6637.