Positionable Vertical Microfluidic Cell Based on Electromigration in a

Jul 30, 2014 - Minkyung Kang , Dmitry Momotenko , Ashley Page , David Perry , and ... Joshua C. Byers , Binoy Paulose Nadappuram , David Perry , Kim ...
0 downloads 0 Views 382KB Size
Download PDF
Article pubs.acs.org/Langmuir

Positionable Vertical Microfluidic Cell Based on Electromigration in a Theta Pipet Michael A. O’Connell,†,∥ Michael E. Snowden,† Kim McKelvey,†,‡ Florence Gayet,† Ian Shirley,§ David M. Haddleton,† and Patrick R. Unwin*,† †

Department of Chemistry, and ‡MOAC Doctoral Training Centre, University of Warwick, Coventry CV4 7AL, United Kingdom § Syngenta, Jealott’s Hill, International Research Centre, Bracknell, Berkshire, RG42 6EY, United Kingdom ABSTRACT: A microscale vertical fluidic cell system has been implemented, based on a simple theta pipet pulled to a sharp point (ca. 10−20 μm diameter for the studies herein) and positioned with a high degree of control on a surface. The dual channel arrangement allows an electric field to be generated between an electrode in each compartment of the pipet that can be used to control the electromigration of charged species between the two compartments, across a thin liquid meniscus in contact with the substrate of interest. By visualizing the interfacial region using laser scanning confocal microscopy, the adsorption of fluorescently-labeled materials on surfaces is monitored quantitatively in real time, exemplified through studies of the adsorption of anionic microparticles (1.1 μm diameter) on positively and negatively charged substrate surfaces of poly-L-lysine (PLL) and poly-Lglutamic acid (PGA), respectively, on glass. These studies highlight significant electrostatic effects on adsorption rates and also that the adsorption of these particles is dominated by the three phase meniscus/solid/air boundary. The technique is easily modified to the case of a submerged substrate, resulting in a much larger deposition area. Finite element method modeling is used to calculate local electric field strengths that are used to understand surface deposition patterns. To demonstrate the applicability of the technique to live biological substrates, the delivery of fluorescent particles directly to the surface of a single root hair cell of Zea mays is demonstrated. The mobile pipet allows deposition to be directed to specific regions of the cell, allowing discrete sites to be labeled with particles. Finally, the technique is used to study the uptake of fluorescent polymer molecules to single root hair cells, with quantitative analysis of the adsorption rates of vinyl-sulfonic acid copolymers, with varying rhodamine B content.

■

patterning with multiple species4,5 and for general surface and electrode analysis.13−24 This type of device represents the forefront of a general class of pipet-based approaches that have been used to analyze, inject, deposit, and fabricate materials on surfaces.25−35 Pipet-based cells are attractive because they are simple, quick and inexpensive to fabricate, and the aperture size can be tailored over a wide range from the nanoscale to the microscale. Moreover, as contact with the substrate of interest is via a meniscus, this technique is applicable to a broad range of samples. In this paper we demonstrate how a dual channel theta pipet can be used as a versatile, mobile microfluidic device for quantitative adsorption studies and for the localized delivery of material to surfaces. Experiments are supported by finite element method (FEM) modeling to describe the electric field in the theta pipet, from which fluxes and mass transport can be elucidated, enabling deeper insight into a range of interfacial

INTRODUCTION Vertically-mounted microfluidic cells are gaining increasing attention as a means of probing surface processes and delivering material to surfaces without needing to immerse the entire sample in solution.1−6 These devices expand the horizons of closed microfluidic systems, by allowing the ready coupling of a surface or interface of interest to a microfluidic system in an easy fashion, as well as opening up horizons in imaging and patterning. Hitherto, the probes used in many of these devices have tended to be fabricated by multistep processes and may incorporate flow reservoirs and pumping systems.1−3 The footprint of this type of vertical microfluidic device is in the range of 10−10 000 μm2, although the recently introduced Fluid-AFM probe takes the spatial resolution to the 1 μm2 level or smaller.7−12 An alternative approach for the delivery of charged species (simple ions, charged (macro)molecules and particles) is to employ electromigration. This approach is perhaps best illustrated by theta pipet devices containing 2 quasi-reference counter electrodes (QRCEs) and (typically) aqueous electrolyte, which have been employed in air and under oil to carry out © XXXX American Chemical Society

Received: May 27, 2014 Revised: July 27, 2014

A

dx.doi.org/10.1021/la5020412 | Langmuir XXXX, XXX, XXX−XXX

Langmuir

Article

determined accurately by field emission-scanning electron microscopy (FE-SEM) (Supra 55, Zeiss) and optical microscopy (BH-2, Olympus). In order to form a stable meniscus, the outer wall of each pipet used was made hydrophobic by immersing the pipet in dimethyldichlorosilane (99% Acros), while passing Argon through the pipet to ensure that the inside was not coated. Both barrels were then filled with the analyte solution (∼50 μL per barrel), and chloridized silver wires (Ag/AgCl) were inserted into each barrel. Surface Modification. Positive and negatively charged surfaces were prepared on cleaned borosilicate glass coverslips (Mezsel-Glaser) by drop casting of poly-L-lysine (PLL) (MW 40 000−70 000) or PLL followed by poly-L-glutamic acid (PGA) (MW 50 000−100 000).38 Solutions of 1 mg mL−1 concentration of each were allowed to adsorb for 20 min, followed by rinsing with 18.2 MΩ cm Milli-Q water and drying under a nitrogen stream. Root Hair Cell Preparation. Zea mays seeds were germinated between two layers of damp paper at 25 °C for 3 days. This provided plants with roots of approximately 10−20 mm in length with a dense layer of root hair cells. Plants were used for measurements at this stage to ensure consistent size, with all nutrients for growth still provided by the seed. Cell viability was tested using fluorescein diacetate (FDA) dissolved in acetone (5 mg mL−1), which was further diluted with water at a ratio of 1 part FDA/acetone solution to 99 parts water.39 Imaging was performed using an I3 filter set with excitation between 450 and 490 nm and collection above 515 nm. Adsorption Experiments. SECCM probes were approached to the functionalized glass surfaces, so that the meniscus just made contact, with a 3-axis piezoelectronic positioning system (Nanocube P611.3S, Physik Instrumente) in a humidified cell.16,17 The pipet was oscillated, typically at a frequency of 70 Hz with a 100 nm peak-topeak amplitude, giving rise to an alternating current (ac) migration current upon surface meniscus contact.6 Control of, and data acquisition from, the piezoelectric/electrochemical apparatus was performed using an FPGA card (PCIe-7852R, National Instruments) with a Labview interface written in house.13,16−23 This apparatus was mounted on an inverted LSCM (Leica TCS SP5 X, Leica Microsystems) to allow particle adsorption to be quantified directly. Fluorescence excitation was achieved using a tunable white light laser (WLL) at a wavelength of 515 nm for particles (emission was collected between 525 and 600 nm) and 540 nm for Rhodamine B polymers (emission was collected between 550 and 600 nm). For experiments without meniscus confinement, a 20 μL droplet of 1 mM KCl was added around the pipet tip after contact of the meniscus with the surface. The zeta potential and electrophoretic mobility of the particles in this solution were measured by Escubed, Ltd. (Leeds, UK) using a ZetasizerNano ZS (Malvern, UK). Simulations and Modeling. Evaluation of the potential field involved solving the Nernst−Planck equations in a similar fashion as reported by us recently in detail.17 Briefly, 3D (spatial) simulations were performed using the commercial FEM modeling package Comsol Multiphysics v4.3 (Comsol AB, Sweden). A schematic of the generalized front face of the probe-surface geometry is shown in Figure 1a, where the simulated domain represents half of a theta pipet (Figure 1b), employing a plane of symmetry perpendicular to the theta pipet dividing wall to enhance computational efficiency. In this report, the liquid domain outside the pipet was either confined within a limited meniscus between the pipet and the surface in a similar manner to our previously reported work (e.g., Figure 1d)17 or extended into bulk solution (Figure 1b,c). The simulated potential fields are shown in Figure 1c for the bulk solution model and Figure 1d for the meniscus-confined model. For the cases depicted, which closely mimicked those employed experimentally, the Nernst−Planck equations were solved for a pipet of outer radius, r0 = 11 μm (also defining the meniscus radius for the confined meniscus case), inner radius, rp = 10 μm, a pipet semi angle, θ = 8°, central segment thickness, tw = 1 μm and tip-sample separation, mh = 2 μm, filled with a solution of 1 mM KCl, where the diffusion coefficient (Dj) and molar conductance (λj) were 20.3 × 10−6 cm2 s−1 and 73.2 S cm2 mol−1, respectively for Cl−; and 19.6 × 10−6 cm2 s−1 and 73.5 S cm2 mol−1, respectively, for K+.40 The potentials across the

physicochemical processes. To demonstrate the approach, we have studied the adsorption of fluorescently-labeled negativelycharged carboxylate-modified microparticles to both electrostatically favorable and unfavorable surfaces as a function of the transport rate, which was controlled by varying the potential applied between the QRCEs in the barrels of the pipet. Realtime particle counting by laser scanning confocal microscopy (LSCM) was employed to quantify microparticle adsorption rates. As well as employing the pipet in air with a confined meniscus, adsorption has also been studied at a substrate submerged in solution, illustrating the diverse range of environments open to study. The flexibility of the method was further demonstrated through the use of a root hair cell as the substrate, to which charged species (particles and fluorescently-labeled polymers) were delivered to several discrete areas in a controlled fashion.

■

EXPERIMENTAL SECTION

Chemicals. 1.1 μm diameter carboxylate−modified fluorescent polystyrene microparticle suspensions (2% mass/volume, stabilized by NaN3) with absorption maximum at 505−515 nm were obtained from Invitrogen. The suspensions of microparticles were diluted to 0.001% particles (for studies of planar substrates) or 0.0005% (for root cell studies), both in 1 μM NaN3 using 18.2 MΩ cm (25 oC) Milli-Q water (Millipore Corporation). The ionic concentration of the solution was kept constant by addition of KCl to a concentration of 1 mM. All other chemicals were purchased from Sigma-Aldrich unless otherwise specified. Rhodamine B Butyl Acrylate Monomer. The monomer was prepared under an inert (nitrogen) atmosphere using a slightly modified literature procedure previously reported.36,37 To a solution of rhodamine B (6 g, 12.5 mmol) in dry dichloromethane (DCM) (80 mL) cooled to 0 °C, (dimethylamino)pyridine (DMAP) (0.23g, 1.8 mmol) and dicyclohexylcarbodiimide (99%) DCC (3.1 g, 15 mmol) were added, and the reaction was left to stir for 1 h. A solution of 1hydroxy butyl acrylate (2.52g, 18.5 mmol) in 10 mL DCM was added dropwise at 0 °C and left stirring for a further 1 h. The reaction vessel was then allowed to warm to ambient temperature and stirred under nitrogen overnight. The volatiles were then removed under reduced pressure, and the residues redissolved in acetonitrile and filtered to remove the insoluble dicyclohexylurea (DCU) residues. The product was purified by column chromatography with acetonitrile eluent to give the rhodamine B butyl acrylate as a purple glassy solid in 83% yield, which gave a single thin layer chromatography (TLC) spot. High-resolution electrospray mass spectrometry (MS-ES) calc. for C35H41ClN2O5(M+): 604.27; found: 579.30 (M−Cl). Fluorescence spectrometry in methanol solution gave absorption and emission spectra with maxima of 563 and 578 nm, respectively. Rhodamine B/Vinyl Sulfonic Acid Copolymerization: General Procedure. Free radical aqueous polymerization was achieved by adaptation of a literature procedure,36,37 involving the addition of rhodamine B butyl acrylate in isopropanol and vinyl sulfonic acid in H2O (not deaerated) to a Schlenk tube. The resulting mixture was deaerated for 40 min by bubbling with dry N2. 4,4-α-Azobis(4cyanovaleric acid) (>75%, Sigma-Aldrich) (1 wt %) was then added, and the reaction mixture was heated to 80 oC. The reaction was sampled at regular intervals, and conversion was determined by aqueous gel permeation chromatography (buffer pH 8.2) measured against a poly(methacrylic acid) calibration (Mn = 1560 g mol−1 for the 2% rhodamine copolymer; Mn = 1860 g mol−1 for the 4% rhodamine copolymer).36,37 Both polymers were dissolved in 1 mM aqueous KCl solution for adsorption experiments. Scannning Electrochemical Cell Microscopy (SECCM) Probe Fabrication and Manipulation. Dual barrel borosilicate glass theta capillaries (Harvard Apparatus, UK) were pulled using a laser pipet puller (Model P-2000, Sutter Instruments, UK) to have a radius of ∼5 μm for experiments employing root hair cells and ∼10 μm for experiments with model planar surfaces. The pipet geometry was B

dx.doi.org/10.1021/la5020412 | Langmuir XXXX, XXX, XXX−XXX

Langmuir

Article

were employed, which were characterized by two closely identical barrels acting as inlet and outlet. Figure 2 shows the adsorption versus time behavior of the particles on the two surfaces at three different potential

Figure 1. Theta pipet microfluidic cell set up showing: (a) a schematic of the cross-section of the simulated domain; (b) a transparent view of the simulated domain (with the plane of reflection symmetry along the face); (c) the relative potential field for an immersed pipet; and (d) the relative potential field for a pipet with a confined liquid meniscus. Parts b, c, and d are for a pipet of rp = 10 μm, r0= 11 μm, tw = 1 μm, θ = 8°, mh = 2 μm (see text for definitions). Scale bar is 20 μm. Figure 2. Adsorption versus time plots for 1.1 μm diameter carboxylate-modified particles from a 10 μm radius theta pipet on (a) PLL- and (b) PGA-modified surfaces. Solution used contained 0.001% particles and 1 mM KCl as background electrolyte. The bias between the QRCEs in the barrels of the theta pipets was 100 mV (black), 200 mV (red), and 500 mV (blue). Values shown of number of adsorbed particles are the average of a minimum of three separate runs.

two faces, E1 and E2 were uniform and the effective potential difference between the faces, for the case in Figure 1, was 0.1 V. The potential drops most sharply in the meniscus between the two channels of the theta pipet, indicating that this is the region of highest resistance. In order to analyze experimental data, the potential over the zoomed region of the simulations was adjusted until the simulated ion conductance current matched that measured experimentally, as described previously.17 No flux boundary conditions were applied normal to the pipet walls, the substrate, and meniscus boundary, whereas the upper faces of the pipet and bulk domain were defined by a bulk concentration boundary condition, ensuring that the simulation domain in the z-direction was sufficiently large that the bulk solution conditions would be recovered and have no influence on the simulation results: cj = c*j

differences between the QRCEs in the barrel of the theta pipet. A few important points are evident. First, it is clear that the electrostatically-favorable surface (PLL) results in a much higher adsorption rate than the electrostatically-unfavorable (PGA) surface. On PLL (Figure 2a) the extent of adsorption (number of adsorbed particles) is essentially linear with respect to time, as would be expected for favorable adsorption in a low coverage regime.41 Second, there is an increase in adsorption rate with increasing potential between the barrels of the theta pipet, corresponding to a higher flux of particles to the surface. This is particularly evident for the PLL surface (Figure 2a). On PLL the adsorption rates for each bias were calculated to be 1.2 ± 0.1 particles per minute (100 mV bias); 2.8 ± 0.2 particles per minute (200 mV) and 4.8 ± 0.2 particles per minute (500 mV). Reported data are the mean of a minimum of 3 measurement runs, which were consistent to ±2 particles of the mean value at all data points. In fact, for the range of potentials studied, the relationship between adsorption rate and bias between the QRCEs is close to linear. These data illustrate the control of mass transport rates by electro-migration simply by varying the bias between the barrels of the pipet, as shown further below.

(1)

where cj is the concentration of species j, and the superscript * indicates bulk concentration.

■

RESULTS AND DISCUSSION We first examined the theta pipet cell on model surfaces, where the substrate was wetted by a meniscus formed between the surface and a pipet containing a suspension of carboxylatemodified (negatively-charged) particles. The surfaces chosen were electrostatically favorable (positively-charged, PLL) and unfavorable (negatively-charged, PGA).The rate of particle transport through the meniscus above the surface was controlled by varying the bias between the pipet barrels. The size of the theta pipet aperture was tailored to provide a reasonable size for the microparticles. For the 1.1 μm diameter particles, pipets of approximately r0 = 11 μm and rp = 10 μm C

dx.doi.org/10.1021/la5020412 | Langmuir XXXX, XXX, XXX−XXX

Langmuir

Article

For the PGA surface (Figure 2b) the adsorption rates were estimated to be 0.2 ± 0.2 particles per minute (100 mV bias), 0.3 ± 0.2 particles per minute (200 mV), and 0.1 ± 0.1 particles per minute (500 mV). Reported data are again the mean of a minimum of three measurements, which were in agreement to ±1 particle of the mean value at each data point shown. These adsorption rates are at least an order of magnitude lower than for PLL and are essentially negligible, which clearly illustrates the impact of electrostatic interactions between the substrate and the particles on adsorption rates. Interestingly, for this surface where adsorption appears to be unfavorable, the higher applied bias (0.5 V) appears to discourage adsorption (promote desorption). As shown below, there is a strong (biasdependent) electric field in the meniscus between the inlet and outlet channels, and this serves not only to control the mass transport rate of the particles, but also their residence time in the meniscus and affinity for the surface. With the model described above, we were able to estimate the “sticking coefficient” for the particles. The average flux of particles was calculated from the modeled potential field at the end of the tip. The tip of one barrel (negative electrode) served as the inlet, and the potential gradient on this plane was used to calculate the average particle velocity, and hence the flux toward the surface. The electrophoretic mobility, μe, of these particles was 3.0 × 10−8 m2 s−1 V−1, and the average flux of particles was calculated to be 0.9 ± 0.4 particles per minute (Eb = 100 mV, mean conductance current, ib = 2.1 nA), 1.9 ± 0.9 particles per minute (Eb = 200 mV, mean ib = 4.1 nA) and 4.6 ± 2.2 particles per minute (Eb = 500 mV, mean ib = 10.2 nA), with the errors based on the range of experimental conductance currents. These values are very similar to those measured on PLL, indicating that the sticking coefficient is close to 1. In contrast, the sticking coefficient on the PGA surface is close to zero (below 0.1). Visualization with LSCM (Figure 3a,b) allowed the location of adsorbed particles to be determined both in the meniscus (in situ) and following withdrawal of the pipet probe. We carried out the analysis for the PLL surface, and it is apparent from Figure 3a that the particles adsorb preferentially at the threephase boundary. When the probe is withdrawn, it is clear the particles have some degree of mobility as the particles tend to aggregate together and the lower number on the surface indicates that some are withdrawn with the meniscus (Figure 3b). The high energy three-phase boundary is generally expected to be a key site for particle deposition.42 Furthermore, simulations of the potential field (Figure 3c), for a pipet of rp = 10 μm, tw = 1 μm, θ = 10°, and mh = 2 μm, show the areas with the lowest potential field gradient (hence, the slowest particle velocity) correlate to the areas of preferential particle deposition. We attribute the observed deposition pattern mainly to these two factors. The ability to deliver particles to submerged substrates would diversify the range of environments open to study, and is particularly important for in vivo studies. Therefore, we examined the impact on particle adsorption of submerging the substrate. Figure 4a shows a typical image for particle deposition from a pipet of r0 = 11 μm, rp = 10 μm, and θ = 10°, into bulk solution (1 mM KCl for 10 min) on a PLL substrate. Two observations are noteworthy in Figure 4. First, the particles spread out considerably from the tip. Second, particle deposition is, to some extent, directional with respect to the electric field induced by the theta pipet. It should be noted that the total deposition field extends over a much greater range

Figure 3. LSCM images of 1.1 μm diameter carboxylate-modified particles on a PLL surface after deposition at 200 mV for 10 min with (a) the meniscus in contact with the substrate and (b) following withdrawal of the probe. Scale bar denotes 10 μm, and +/− indicates the polarity of the electrodes in the theta pipet. (c) Simulated potential field gradient and particle velocity at the surface/meniscus boundary shown for a 200 mV bias (corresponding to an average current of 4.1 nA) for a probe of rp = 10 μm, tw = 1 μm, θ = 10°, and mh = 2 μm. Half of the meniscus is shown with the symmetry plane perpendicular to the dividing wall of the theta pipet. The arrows indicate the direction of the particle movement. Scale bar denotes 4 μm.

Figure 4. (a) Deposition of 1.1 μm diameter carboxylate modified particles on a PLL surface at 200 mV for 10 min, from a theta pipet microfluidic cell, with the cell and substrate both submerged in 1 mM KCl. Scale bar denotes 20 μm. (b) Simulated potential field gradient and particle velocity on the substrate surface for a submerged theta pipet and a 200 mV bias (corresponding to an average current of 4.1 nA) for a pipet of rp = 10 μm, tw = 1 μm, θ = 10°, and mh = 2 μm. The arrows indicate the direction of the particle movement. Scale bar denotes 4 μm.

than shown in Figure 4a; however, particles were difficult to resolve on a larger field of view. The deposition pattern can be rationalized by modeling the potential field within the probe and solution (Figure 4b). It is apparent that the particle velocity is greatest beneath the septum and (to a lesser extent) from the negatively polarized barrel into bulk solution. The potential gradient into the bulk solution serves to propel the particles from the pipet tip, resulting in the observed wide dispersion of particles over the substrate. We next consider the application of the technique to individual cells. The Zea mays root hair cells were approximately 10−15 μm in width, and so 5 μm radius theta pipets were used. It was possible to mount the entire plant D

dx.doi.org/10.1021/la5020412 | Langmuir XXXX, XXX, XXX−XXX

Langmuir

Article

These images highlight that localized particle delivery and adhesion to the root hair cell can be achieved with the average deposition rate found to be 0.5 ± 0.2 particles per minute. Importantly, this methodology confined the particles to the area probed by the pipet, with no spreading of particles along the length of the root observed. Therefore, the dual barreled probe delivery method can be used to selectively label individual areas of cells, where there is strong interaction between the surface and delivered material as in this case, even when operated with the sample submerged. Finally, to demonstrate the versatility of the technique we investigated the delivery of charged macromolecules to living root hair cells. Two negatively-charged polymers consisting of copolymerized 2% rhodamine functionality/98% sulfonate functionality (Figure 7, I) and copolymerized 4% rhodamine functionality/96% sulfonate functionality (Figure 7, II) were chosen to test the viability of the delivery system with in situ fluorescence measurements on the LSCM platform used to quantify the accumulation of the polymers in the cell. The extent of adsorption represented by normalized fluorescence, is shown in Figure 7a,b respectively. The fluorescence intensity is normalized to the intensity of the droplet of solution at the apex of the probe, providing an internal calibration and thus an indication of the relative adsorption of the two polymers. Evidently, the 4% rhodamine polymer adsorbs quicker and more extensively than the 2% rhodamine polymer. It is important to note that this large difference cannot be attributed to a significant difference in mass transport rates of the two polymers. The measured currents were similar for each polymer solution. Furthermore, as the polymers are of fairly similar Mn and net charge, it is reasonable to assume that the mobility of the polymer molecules would be similar. Therefore, we attribute the increased surface adhesion of II to a greater binding affinity due to the higher concentration of rhodamine moieties in the polymer, which serves as an effective binder to the root cell.

using an adhesive tab, providing sufficient stability for adsorption experiments and access to individual root hair cells. Vitality of the cells was confirmed by “vital staining”, using fluorescein diacetate.39 The acetate-functionalized fluorescein molecule is fluorescently inactive but membrane mobile. On diffusion into the cell, the acetate group is cleaved via cytoplasmic enzymes, leaving the free form of fluorescein which is fluorescent. Figure 5 shows a mounted root system

Figure 5. Surface bound root hair cells after treatment with the “vital stain” fluorescein diacetate.

imaged via fluorescent microscopy after treatment with fluorescein diacetate. The surface bound root hair cells can be clearly seen on this image, showing that they are viable for study. Figure 6a,b shows fluorescence and optical microscopy images of the root hair cell following delivery of particles in three areas using a 0.0005% particle solution in 1 mM KCl with a deposition time of 10 min at 200 mV barrel bias in each area.

■

CONCLUSIONS The controlled delivery of charged species to substrate surfaces by a dual barrel microfluidic theta pipet cell in air and under liquid has been successfully demonstrated. Substrates modified by the polymers PLL and PGA show significant differences for the adhesion of negatively-charged particles, with the positivelycharged PLL substrate providing a surface which promotes microparticle adhesion (sticking coefficient close to unity), compared to the PGA surface (sticking coefficient close to zero). FEM modeling allowed the calculation of the electric field in the pipet and meniscus (or solution), from which the magnitude and directionality of the local migration of particles within the probe and meniscus could be evaluated. For the confined meniscus system, in particular, areas near the substrate that experience a small potential gradient, and hence a small particle velocity, correlated with sites of preferential particle adhesion. There is growing interest in methods for controlling and analyzing the delivery of microparticles and nanoparticles to surfaces.43−49 The methodology herein is particularly attractive in terms of its versatility and the range of substrate surfaces open to study. Thus, we have successfully demonstrated the controlled deposition of microparticles onto living cells. LSCM measurements showed the particles to be deposited within an area defined by the size of the pipet, even with the substrate under solution, providing a technique for the localized labeling

Figure 6. Localized delivery of 1.1 μm diameter carboxylate-modified particles to three separate areas of a root hair cell. (a) LSCM image and (b) brightfield microscopy image. Deposition conditions of 0.0005% particle solution in 1 mM KCl for 10 min at 200 mV barrel bias. The approximate position of the capillary is denoted by the white circles. Scale bars denote 10 μm. E

dx.doi.org/10.1021/la5020412 | Langmuir XXXX, XXX, XXX−XXX

Langmuir

Article

Figure 7. Schematic of the polymers used for deposition in 1 mM KCl: (I) 2% rhodamine functionality/98% sulfonate functionality and (II) 4% rhodamine functionality/96% sulfonate functionality. Adsorption rates (points) measured by normalized fluorescent intensity (see text) for polymers containing (a) 2% rhodamine functionality/98% sulfonate functionality and (b) 4% rhodamine functionality/96% sulfonate functionality, respectively, at 500 mV barrel bias for 10 min at 1 mg mL−1 concentration (giving bulk concentrations of 0.64 mM and 0.53 mM for 2% rhodamine and 4% rhodamine copolymers, respectively).

Materials project with support from Advantage West Midlands and the European Regional Development Fund.

of live cells with particles. In the future, it would be possible to have different compositions in the pipet and bathing solution, expanding the range of cell systems and environments open to study. Experiments with polymer species with different absorption properties demonstrated that the technique is sensitive to variations in fluorophore to substrate adhesion. Combined with the lateral resolution of the general SECCM platform,16,17 it is evident that the methodology described herein should be widely applicable as a means of probing localized surface (adsorption) properties.

■

■

REFERENCES

(1) Kaigala, G. V.; Lovchik, R. D.; Drechsler, U.; Delamarche, E. A Vertical Microfluidic Probe. Langmuir 2011, 27 (9), 5686−5693. (2) Cortes-Salazar, F.; Lesch, A.; Momotenko, D.; Busnel, J. M.; Wittstock, G.; Girault, H. H. Fountain Pen for Scanning Electrochemical Microscopy. Anal. Methods 2010, 2 (7), 817−823. (3) Momotenko, D.; Cortes-Salazar, F.; Lesch, A.; Wittstock, G.; Girault, H. H. Microfluidic Push−Pull Probe for Scanning Electrochemical Microscopy. Anal. Chem. 2011, 83 (13), 5275−5282. (4) Rodolfa, K. T.; Bruckbauer, A.; Zhou, D. J.; Schevchuk, A. I.; Korchev, Y. E.; Klenerman, D. Nanoscale Pipetting for Controlled Chemistry in Small Arrayed Water Droplets Using a Double-Barrel Pipet. Nano Lett. 2006, 6 (2), 252−257. (5) Rodolfa, K. T.; Bruckbauer, A.; Zhou, D. J.; Korchev, Y. E.; Klenerman, D. Two-Component Graded Deposition of Biomolecules with a Double-Barreled Nanopipette. Angew. Chem., Int. Ed. 2005, 44 (42), 6854−6859. (6) Ebejer, N.; Güell, A. G.; Lai, S. C. S.; McKelvey, K.; Snowden, M. E.; Unwin, P. R. Scanning Electrochemical Cell Microscopy: A Versatile Technique for Nanoscale Electrochemistry and Functional Imaging. Annu. Rev. Anal. Chem. 2013, 6, 329−351. (7) Kim, K.-H.; Moldovan, N.; Espinosa, H. D. A Nanofountain Probe with Sub-100 nm Molecular Writing Resolution. Small 2005, 1 (6), 632−635. (8) Basabe-Desmonts, L.; Wu, C.-C.; van der Werf, K. O.; Peter, M.; Bennink, M.; Otto, C.; Velders, A. H.; Reinhoudt, D. N.; Subramaniam, V.; Crego-Calama, M. Fabrication and visualization of metal-ion patterns on glass by dip-pen nanolithography. ChemPhysChem 2008, 9 (12), 1680−1687. (9) Grueter, R. R.; Voeroes, J.; Zambelli, T. FluidFM as a Lithography Tool in Liquid: Spatially Controlled Deposition of Fluorescent Nanoparticles. Nanoscale 2013, 5 (3), 1097−1104.

AUTHOR INFORMATION

Corresponding Author

*E-mail [email protected]. Present Address ∥

National Physical Laboratory, Teddington, Middlesex, TW11 0LW, United Kingdom. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS M.A.O’C. and M.E.S. contributed equally to this work. The European Research Council has provided financial support under the European Community’s (EC’s) Seventh Framework Programme (FP7/2007-2013)/ERC-2009-AdG2471143QUANTIF. Additional funding was provided by the EPSRC CTA scheme (Syngenta) for M.A.O’C., and the EPSRC (MOAC/DTC studentship scheme) for K.M. D.M.H. thanks the Royal Society for a Wolfson Merit Award. We thank Dr. Paul Wilson for discussion. Some of the equipment used in this work was obtained through the Science City Advanced F

dx.doi.org/10.1021/la5020412 | Langmuir XXXX, XXX, XXX−XXX

Langmuir

Article

(10) Li, Y.; Sun, H.; Chu, H. Controlled Preparation of Inorganic Nanostructures on Substrates by Dip-Pen Nanolithography. Chem. Asian J. 2010, 5 (5), 980−990. (11) Lee, K. B.; Park, S. J.; Mirkin, C. A.; Smith, J. C.; Mrksich, M. Protein Nanoarrays Generated by Dip-Pen Nanolithography. Science 2002, 295 (5560), 1702−1705. (12) Piner, R. D.; Zhu, J.; Xu, F.; Hong, S. H.; Mirkin, C. A. “Dippen” Nanolithography. Science 1999, 283 (5402), 661−663. (13) Kinnear, S. L.; McKelvey, K.; Snowden, M. E.; Peruffo, M.; Colburn, A. W.; Unwin, P. R. Dual-Barrel Conductance Micropipet as a New Approach to the Study of Ionic Crystal Dissolution Kinetics. Langmuir 2013, 29 (50), 15565−15572. (14) Eng, L.; Wirth, E.; Suter, T.; Bohni, H. Non-contact Feedback for Scanning Capillary Microscopy. Electrochim. Acta 1998, 43 (19-20), 3029−3033. (15) Bohni, H.; Suter, T.; Schreyer, A. Microtechniques and Nanotechniques to Study Localized Corrosion. Electrochim. Acta 1995, 40 (10), 1361−1368. (16) Ebejer, N.; Schnippering, M.; Colburn, A. W.; Edwards, M. A.; Unwin, P. R. Localized High Resolution Electrochemistry and Multifunctional Imaging: Scanning Electrochemical Cell Microscopy. Anal. Chem. 2010, 82 (22), 9141−9145. (17) Snowden, M. E.; Güell, A. G.; Lai, S. C. S.; McKelvey, K.; Ebejer, N.; O’Connell, M. A.; Colburn, A. W.; Unwin, P. R. Scanning Electrochemical Cell Microscopy: Theory and Experiment for Quantitative High Resolution Spatially-Resolved Voltammetry and Simultaneous Ion-Conductance Measurements. Anal. Chem. 2012, 84 (5), 2483−2491. (18) Patten, H. V.; Meadows, K. E.; Hutton, L. A.; Iacobini, J. G.; Battistel, D.; McKelvey, K.; Colburn, A. W.; Newton, M. E.; Macpherson, J. V.; Unwin, P. R. Electrochemical Mapping Reveals Direct Correlation between Heterogeneous Electron-Transfer Kinetics and Local Density of States in Diamond Electrodes. Angew. Chem., Int. Ed. 2012, 51 (28), 7002−7006. (19) Miller, T. S.; Ebejer, N.; Güell, A. G.; Macpherson, J. V.; Unwin, P. R. Electrochemistry at Carbon Nanotube Forests: Sidewalls and Closed Ends Allow Fast Electron Transfer. Chem. Commun. 2012, 48 (60), 7435−7437. (20) Güell, A. G.; Ebejer, N.; Snowden, M. E.; Macpherson, J. V.; Unwin, P. R. Structural Correlations in Heterogeneous Electron Transfer at Monolayer and Multilayer Graphene Electrodes. J. Am. Chem. Soc. 2012, 134 (17), 7258−7261. (21) Lai, S. C. S.; Dudin, P. V.; Macpherson, J. V.; Unwin, P. R. Visualizing Zeptomole (Electro)Catalysis at Single Nanoparticles within an Ensemble. J. Am. Chem. Soc. 2011, 133 (28), 10744−10747. (22) Güell, A. G.; Ebejer, N.; Snowden, M. E.; McKelvey, K.; Macpherson, J. V.; Unwin, P. R. Quantitative Nanoscale Visualization of Heterogeneous Electron Transfer Rates in 2D Carbon Nanotube Networks. Proc. Natl. Acad. Sci. U. S. A. 2012, 109 (29), 11487−11492. (23) Patel, A. N.; McKelvey, K.; Unwin, P. R. Nanoscale Electrochemical Patterning Reveals the Active Sites for Catechol Oxidation at Graphite Surfaces. J. Am. Chem. Soc. 2012, 134 (50), 20246−9. (24) Suter, T.; Bohni, H. A New Microelectrochemical Method to Study Pit Initiation on Stainless Steels. Electrochim. Acta 1997, 42 (2022), 3275−3280. (25) Bruckbauer, A.; Ying, L.; Rothery, A. M.; Zhou, D.; Shevchuk, A. I.; Abell, C.; Korchev, Y. E.; Klenerman, D. Writing with DNA and Protein Using a Nanopipet for Controlled Delivery. J. Am. Chem. Soc. 2002, 124 (30), 8810−8811. (26) Bruckbauer, A.; Zhou, D. J.; Kang, D. J.; Korchev, Y. E.; Abell, C.; Klenerman, D. An Addressable Antibody Nanoarray Produced on a Nanostructured Surface. J. Am. Chem. Soc. 2004, 126 (21), 6508− 6509. (27) Bruckbauer, A.; Zhou, D. J.; Ying, L. M.; Korchev, Y. E.; Abell, C.; Klenerman, D. Multicomponent Submicron Features of Biomolecules Created by Voltage Controlled Deposition from a Nanopipet. J. Am. Chem. Soc. 2003, 125 (32), 9834−9839.

(28) Morris, C. A.; Friedman, A. K.; Baker, L. A. Applications of Nanopipettes in the Analytical Sciences. Analyst 2010, 135 (9), 2190− 2202. (29) Suryavanshi, A. P.; Yu, M.-F. Electrochemical Fountain Pen Nanofabrication of Vertically Grown Platinum Nanowires. Nanotechnology 2007, 18 (10), 105305. (30) Suryavanshi, A. P.; Yu, M. F. Probe-Based Electrochemical Fabrication of Freestanding Cu Nanowire Array. Appl. Phys. Lett. 2006, 88 (8), 083103. (31) Yang, D.; Han, L.; Yang, Y.; Zhao, L.-B.; Zong, C.; Huang, Y.-F.; Zhan, D.; Tian, Z.-Q. Solid-State Redox Solutions: Microfabrication and Electrochemistry. Angew. Chem., Int. Ed. 2011, 50 (37), 8679− 8682. (32) Laslau, C.; Williams, D. E.; Travas-Sejdic, J. The Application of Nanopipettes to Conducting Polymer Fabrication, Imaging and Electrochemical Characterization. Prog. Polym. Sci. 2012, 37 (9), 1177−1191. (33) Shiku, H.; Yamakawa, T.; Nashimoto, Y.; Takahashi, Y.; Torisawa, Y.-s.; Yasukawa, T.; Ito-Sasaki, T.; Yokoo, M.; Abe, H.; Kambara, H.; Matsue, T. A Microfluidic Dual Capillary Probe to Collect Messenger RNA from Adherent Cells and Spheroids. Anal. Biochem. 2009, 385 (1), 138−142. (34) Chen, C. C.; Zhou, Y.; Baker, L. A. Scanning Ion Conductance Microscopy. Annu. Rev. Anal. Chem. 2012, 5, 207−228. (35) Kranz, C. Recent Advancements in Nanoelectrodes and Nanopipettes Used in Combined Scanning Electrochemical Microscopy Techniques. Analyst 2014, 139 (2), 336−352. (36) Nicolas, J.; San Miguel, V.; Mantovani, G.; Haddleton, D. M. Fluorescently Tagged Polymer Bioconjugates from Protein Derived Macroinitiators. Chem. Commun. 2006, 4697−4699. (37) Cramm, J. R.; Johnson, B. S.; Reed, P. E.; Ward, W. J. Derivatized Rhodamine Dye and Its Copolymers U.S. Patent 5,772,894 (US00577289A). (38) O’Connell, M. A.; de Cuendias, A.; Gayet, F.; Shirley, I. M.; Mackenzie, S. R.; Haddleton, D. M.; Unwin, P. R. Evanescent Wave Cavity Ring-Down Spectroscopy (EW-CRDS) as a Probe of Macromolecule Adsorption Kinetics at Functionalized Interfaces. Langmuir 2012, 28 (17), 6902−6910. (39) Widholm, J. M. Use of Fluorescein Diacetate and Phenosafranine for Determining Viability of Cultured Plant-Cells. Stain Technology 1972, 47 (4), 189−194. (40) Lide, D. R. CRC Handbook of Chemistry and Physics; CRC Press: Boca Raton, FL, 2001; Vol. 82. (41) Adamczyk, Z. Particle Adsorption and Deposition: Role of Electrostatic Interactions. Adv. Colloid Interface Sci. 2003, 100, 267− 347. (42) Deegan, R. D.; Bakajin, O.; Dupont, T. F.; Huber, G.; Nagel, S. R.; Witten, T. A. Capillary Flow As the Cause of Ring Stains from Dried Liquid Drops. Nature 1997, 389 (6653), 827−829. (43) Fosdick, S. E.; Anderson, M. J.; Nettleton, E. G.; Crooks, R. M. Correlated Electrochemical and Optical Tracking of Discrete Collision Events. J. Am. Chem. Soc. 2013, 135 (16), 5994−5997. (44) Bard, A. J.; Zhou, H.; Kwon, S. J. Electrochemistry of Single Nanoparticles via Electrocatalytic Amplification. Isr. J. Chem. 2010, 50 (3), 267−276. (45) Kwon, S. J.; Fan, F.-R. F.; Bard, A. J. Observing Iridium Oxide (IrOx) Single Nanoparticle Collisions at Ultramicroelectrodes. J. Am. Chem. Soc. 2010, 132 (38), 13165−13167. (46) Quinn, B. M.; van ’t Hof, P.; Lemay, S. G. Time-Resolved Electrochemical Detection of Discrete Adsorption Events. J. Am. Chem. Soc. 2004, 126 (27), 8360−8361. (47) van den Heuvel, M. G. L.; Butcher, C. T.; Lemay, S. G.; Diez, S.; Dekker, C. Electrical Docking of Microtubules for Kinesin-Driven Motility in Nanostructures. Nano Lett. 2005, 5 (2), 235−241. (48) Boika, A.; Thorgaard, S. N.; Bard, A. J. Monitoring the Electrophoretic Migration and Adsorption of Single Insulating Nanoparticles at Ultramicroelectrodes. J. Phys. Chem. B 2012, 117 (16), 4371−4380. G

dx.doi.org/10.1021/la5020412 | Langmuir XXXX, XXX, XXX−XXX

Langmuir

Article

(49) Welsher, K.; Yang, H. Multi-Resolution 3D Visualization of the Early Stages of Cellular Uptake of Peptide-Coated Nanoparticles. Nat. Nanotechnol. 2014, 9, 198−203.

H

dx.doi.org/10.1021/la5020412 | Langmuir XXXX, XXX, XXX−XXX