Insights into Nanoscale Electrophoresis of Single Dye Molecules in

Article pubs.acs.org/JPCC

Insights into Nanoscale Electrophoresis of Single Dye Molecules in Highly Oriented Mesoporous Silica Channels Melari Davies,† Bastian Rühle,† Chen Li,‡ Klaus Müllen,‡ Thomas Bein,*,† and Christoph Braü chle*,† †

Department of Chemistry and Center for NanoScience (CeNS), University of Munich (LMU), Butenandtstrasse 5-13 (E), D-81377 Munich, Germany ‡ Max Planck Institute for Polymer Research, Ackermannweg 10, D-55128 Mainz, Germany S Supporting Information *

ABSTRACT: Electrophoresis is a very important tool in chemistry, biochemistry, and biology and is commonly used for the separation and analysis of molecules. Typically, electrophoresis is performed at the macroscopic level. With a view on integrating molecular separations into micro- and nanofluidic and massively parallel systems, it would be highly desirable to obtain detailed knowledge about electrophoretic separation at the nanoscale. Here, we describe an in situ single molecule fluorescence imaging approach to provide insights into the controlled movement and separation of charged molecules in highly oriented silica nanochannels under the influence of an electric field. The velocity of the charged single molecules increases with increasing electric field strength and can therefore be directly controlled by an external stimulus. On the basis of the single molecule trajectories, we propose a model for the internal structure of the host and for the diffusion of charged and uncharged species in the nanoscale channels. In order to prove that the molecules move along the aligned pores, we applied an electric field diagonally to the pores. The molecules that were located in the mesopores showed a directed movement along the mesoporous silica channels, whereas molecules next to, but outside of, the structure followed the electric field lines. Leaky defects in the side walls of the pores were indicated by molecules switching into and out of the pores. In addition, we show the application of the mesoporous host material for separation of molecules according to their mobility and charge.

■

INTRODUCTION Because of their large structural and functional versatility, the scope of periodic mesoporous silica host systems1,2 for separation applications is enormous.3−5 Mesoporous silica thin films that feature an ordered array of well-defined mesopores in an extended porous network are especially intriguing for studying diffusion, electrophoretic transport, and separation and could provide a platform for next-generation separation devices. So far, besides theoretical studies and simulations,16−18 many experimental investigations on nanoscale electrophoresis employing silica hosts mainly focused on disordered, macroporous silica monoliths6−9 for electrochromatography. However, macroporous monoliths typically do not feature an ordered array of uniform, well-defined pores, and it is difficult to detect and analyze the exact locations and pathways of single molecules inside these structures in the presence of an electric field, ultimately leading to their separation. On the other hand, approaches featuring more well-defined nanoscale environments, including elaborate “top-down” systems for nanoscale electrophoresis19−24 and single pores in thin membranes10−13 for single molecule sensing, are typically limited to only a small number of nanochannels per device. Using mesoporous silica thin films that can be produced by an inexpensive, simple © 2014 American Chemical Society

bottom-up approach to create virtually millions of uniform nanochannels at once could be very beneficial for applications requiring a well-defined, massively parallel separation environment. Moreover, unlike in conventional gel electrophoresis,14,15 the template-filled nanoscale channels of mesoporous silica thin films can provide a mechanically stable, tunable, and highly ordered liquid-crystalline nanoenvironment for guest molecules, while an applied electric field can provide both the possibility for separation of molecules with different mobilities and charges and a means for controlling the movement and velocity of the guest species inside the porous host material on demand via an external stimulus. In this work, we describe a simple and inexpensive way to obtain well-defined, ordered mesoporous silica thin films that can serve as model systems for investigating the pathways of single guest molecules during nanoscale electrophoresis. We demonstrate that a direct observation of the movement of charged and uncharged single molecules under the influence of an electric field inside the mesoporous silica hosts is possible and that single molecule trajectories can be used for a detailed Received: April 17, 2014 Revised: September 17, 2014 Published: September 19, 2014 24013

dx.doi.org/10.1021/jp503780n | J. Phys. Chem. C 2014, 118, 24013−24024

The Journal of Physical Chemistry C

Article

Figure 1. Scanning electron microscopy (SEM) images of a patterned silica thin film. (a, b) Top-views and (c) an oblique view show the patterning (MIMIC process) resulting in bars of porous silica. (d) High-resolution cross-sectional SEM image of one bar showing the nanometer-sized mesoporous channels.

cross section of one of these silica bars. The high-resolution image of the cross section clearly shows the alignment of the mesopores perpendicular to the cross section of the silica bars. Please note that our investigations in this work focus on the movement of guest molecules inside the nanometer-sized mesoporous channels running through the silica structure (as can be seen in Figure 1d) and not on the micrometer-sized voids between the silica bars that result from the PDMS stamp in the alignment process (see Figure 1a−c). We also carried out further characterization by twodimensional grazing incidence small-angle X-ray scattering (2D-GISAXS) and krypton sorption analysis. The 2D-GISAXS data show strongly anisotropic patterns depending on the orientation of the substrate with respect to the incident beam (see Figure S1), which is in line with a 2D hexagonal arrangement of the mesopores that are highly oriented in the substrate plane and run parallel to the silica bars, as also suggested by single molecule fluorescence microscopy experiments. The Kr sorption data (see Figure S2) further indicate a highly porous system with a BET surface area of approximately 0.44 cm2/cm2 (value normalized to 1 nm film thickness), which is in line with results from Bartels et al., who obtained values of 0.38−0.42 cm2/cm2 for similar porous silica thin films.26 For further details on the characterization methods see the

structural and dynamical analysis of the host−guest system. Using single molecule fluorescence microscopy, even single pores inside the mesoporous host can be resolved and visualized, including leaky defects in the side walls of the pores. Moreover, a separation of molecules with different mobilities and charges inside the host structure can be directly observed on a single molecule basis.

■

RESULTS AND DISCUSSION We have recently demonstrated that defined macroscopic alignment of large-pore mesochannels (diameter about 9 nm) in mesoporous silica can be achieved by confinement in poly(dimethylsiloxane) microgrooves. The detailed synthesis and analysis of these systems have been published elsewhere.25 In short, we filled microgrooves that were formed by a PDMS stamp and a glass substrate with a silica precursor solution and used the evaporation-induced self-assembly (EISA) process to obtain mesoporous silica thin films inside the microgrooves. By tuning the synthesis parameters, the mesopores inside the obtained silica bars can be aligned parallel to the microgroove walls. Figure 1 shows electron microscopy images that display a top view of the silica bars (Figure 1a,b), an oblique view (Figure 1c), and a high-resolution SEM image (Figure 1d) of a 24014



Article

Figure 2. Overview of the dye molecules incorporated into the mesoporous silica channels. Atto633 and Atto532 are commonly available dyes with one charge (Atto633 is positively charged and Atto532 is negatively charged). PhPy-PDI and Py-PDI both bear four positive charges on a perylenediimide (PDI) chromophore. The neutral dyes Dip-TDI, As-TDI, and Sw-TDI and also the four times negatively charged Ws-TDI molecules have a terrylenediimide (TDI) core.

was reached during the recording time. The final overlay consists only of the brightest pixels of the movie. We call this procedure a “maximum projection” (see also Rühle et al.25). This method gives a global overview of the pore alignment. In addition, single molecule tracking allowed for a more detailed structural and dynamic analysis of the guest molecule’s movement. In order to verify the mesopore alignment of the porous host material, we used uncharged guest molecules (Dip-TDI, AsTDI, and Sw-TDI, chemical structure see Figures 2 and 3) in a preliminary experiment. The movies recorded with our widefield microscope (see Supporting Information Movies S1−S3) clearly display a structured movement of these TDI molecules parallel to the pore direction of the silica channels. The maximum projections of the individual frames of the movies directly depict the pathways of the molecules and thus the macroscopic alignment of the mesoporous channels (Figure 3a−c). Single molecule trajectories reveal the structure of the

Materials and Methods section and the Supporting Information. To investigate the feasibility of electrophoresis experiments in such highly oriented silica channels on a single molecule basis, differently charged fluorescent dye molecules were incorporated into mesoporous silica channels at single molecule concentrations (10−10−10−11 mol/L) and directly manipulated by an externally applied electric field. We also carried out investigations with uncharged dyes to ensure the integrity of the mesoporous structure under the experimental conditions, i.e., the presence of strong electric fields. The dyes used in the experiments are presented in Figure 2. By determining the position of the single molecules in each frame of a movie recorded with a wide-field fluorescence microscope and overlaying the individual frames of the movie, the pathway of the single molecules can be depicted. Specifically, each pixel in the overlay assumes the maximum value of the corresponding pixels in the individual frames that 24015



Article

Figure 3. Maximum projection of several consecutive wide-field images demonstrating the alignment of the mesoporous silica channels. (a) Dip-TDI (N,N′-bis(2,6-diisopropylphenyl)terrylene-3,4,11,12-tetracarboxdiimide), (b) As-TDI (N-(2,6-diisopropylphenyl)-N′-octylterrylene-3,4,11,12-tetracarboxydiimide), and (c) Sw-TDI (N,N′-(pentadecan-8-yl)terrylene-3,4,11,12-tetracarboxydiimide) molecules were incorporated as fluorescent guest dyes into the mesopores during the synthesis in order to gain information about pore directionality. Typical single molecule trajectories of (d) DIPTDI, (e) As-TDI (inset shows the enlarged trajectory of guest molecule 3 switching from one channel to another), and (f) Sw-TDI reveal the structure and alignment of the mesopores.

By incorporating charged dye molecules (listed in Figure 2) into the same samples, we observed that uncharged molecules present structured movement along the well-aligned mesochannels, while charged molecules took totally different, more random diffusional pathways. We mixed uncharged As-TDI molecules with charged PhPy-PDI molecules (Figure 4) and

channels in even more detail, exhibiting defects such as pore interconnections and dead ends present in the porous system (Figure 3d−f). By analyzing the trajectories of the molecules in space and time, one can even observe the switching of guest molecules from one channel to another channel through defects in the silica walls (see inset in Figure 3e as an example). 24016



Article

Figure 4. Different diffusional behavior of charged and uncharged dyes in the porous host system. Overlay of the individual frames of (a) Movie S4 and (b) Movie S5, revealing the pathways of (a) PhPy-PDI and (b) As-TDI molecules. (c, d) Typical trajectories of (c) PhPy-PDI and (d) As-TDI with an inset for each showing a magnified trajectory. As-TDI molecules show structured, aligned pathways while PhPy-PDI molecules feature an unstructured, random movement.

different channels. Channel 1 depicts the fluorescence signal of PhPy-PDI molecules, and channel 2 displays the signal of AsTDI molecules. While the charged molecules (PhPy-PDI) moved more randomly (see Movie S4 and Figure 4a,c), the uncharged molecules (As-TDI) indicate structured movement

incorporated them into the same sample. PhPy-PDI was excited with a 532 nm diode laser in order to be able to spectrally separate its fluorescence signal from the signal of As-TDI molecules which were excited with a He−Ne laser (633 nm), and we detected both fluorescence signals simultaneously in 24017



Article

Figure 5. Different diffusional behavior of charged and uncharged dyes in the porous host system. (a) Maximum projection of the individual frames of Movie S7, indicating the pathways of two different species of single molecules (Atto633 and As-TDI) in the mesoporous structure. (b) Overlay after photobleaching revealing preferentially one species (As-TDI), which shows structured movement. (c) Direct comparison of typical trajectories of As-TDI (red) and Atto633 (green). As-TDI molecules show structured, aligned pathways while Atto633 molecules feature an unstructured, random movement.

is known that Atto dyes are not as photostable as TDI dyes,27 we assigned these signals to Atto633 molecules. The molecules that moved in a structured way had a good photostability and could therefore be preferably identified as As-TDI molecules (see Movie S7 in the Supporting Information). Figure 5a depicts the maximum projection of all frames of Movie S7 (Supporting Information) recorded with our fluorescence microscope, revealing the pathways of both species of single molecules in the mesoporous structure. A maximum projection of the last 100 frames of the same Movie S7 preferably exposes the pathway of one species (As-TDI) moving in a structured way (Figure 5b), while most of the unstructured Atto633 molecules have already been photobleached. Even though there is the possibility of statistical outliers in this experiment, the qualitative trends corroborate the observations from our other experiments; i.e., uncharged dyes show structured diffusional movement, while the movement of charged dyes is much more random and unstructured. In order to compare the patterns of movement of both dyes, typical trajectories of each dye are plotted in Figure 5c. As-TDI molecules (red trajectories in Figure 5c) move up and down in the y-direction (which is the direction of the mesoporous silica

up and down (see Movie S5 and Figure 4b,d). By switching on the electric field (1200 V/cm), we could observe a directed yet still unstructured movement of the charged PhPy-PDI molecules before photobleaching (Movie S6). As expected, the diffusional movement of the uncharged As-TDI molecules was not affected by the electric field. They continued moving in a structured way exactly as they did without the electric field. We consider this a strong indication that the highly ordered sample structure is maintained under the influence of a (strong) electric field and that the rather unstructured movement of the charged dyes is due to other effects than the destruction of the ordered mesoporous structure (see also discussion below). In order to investigate the unstructured movement of charged dyes in more detail, we used Atto633a much smaller and less charged moleculein a second experiment, instead of the large, highly charged PhPy-PDI in the former experiment. As before, we introduced Atto633 molecules together with uncharged As-TDI molecules into the same sample. In this case we could distinguish the two dyes due to their photophysical stability. Both dyes were excited with a 633 nm He−Ne laser and detected in the same channel. The species that moved randomly was bleaching faster, and since it 24018



Article

this effect to the existence of holes in the pore walls, disruptions in the channel structure, and wormlike interstitial porosity introduced by the block copolymer during synthesis.37 In unextracted (i.e., template-filled) mesoporous silica films, we assume that these outer parts of the micellar template structure are preferred by hydrophilic, charged molecules whereas hydrophobic, uncharged molecules prefer the hydrophobic core of the micelles from where they cannot access the defects in the pore wall. This could be a possible explanation why uncharged molecules move in a structured way inside the channels, while charged molecules moving in close proximity to the pore wall can access defects and switch to other channels. As a further proof of this model and in order to ensure that the charged molecules are moving inside the pores of the mesoporous silica structure despite their rather random movement patterns, we prepared a sample containing only Atto633 in a single dye molecule concentration and applied an electric field diagonally to the aligned pores (Movie S9 in the Supporting Information). We could observe that the molecules that were located in the mesopores showed a directed movement along the mesoporous silica channels whereas molecules next to, but outside of, the structure followed the electric field lines. The maximum projection of the individual frames of Movie S9 is depicted in Figure 7a. We also observed molecules next to the silica structure changing their direction abruptly when slipping out of it (see trajectory 1 in Figure 7b) and also slipping into the mesoporous structure (see trajectory 3 in Figure 7b). This is possible because of numerous defects in the silica walls as explained above. Moreover, the experiment demonstrates that the molecules are guided by the mesoporous silica channels (see trajectory 2 in Figure 7b) aligned in the ydirection. The trajectories of the charged dye molecules are again not as straight as in the case of uncharged dye molecules, as explained above in the guest−host model (Figure 6). We then investigated the influence of the electric field strength on the average velocity of single Atto633 dye molecules. For this purpose the electric field was applied along the channel direction. At the beginning, the electric field was adjusted to 1800 V/cm and then reduced stepwise to 100 V/cm. The velocity of the molecules decreases with decreasing electric field strength and could therefore be directly influenced by the electric field acting as an external stimulus. Figure 8 presents the maximum projection of the individual frames of Movie S10 (see Supporting Information) revealing the pathway

bars) in a structured way, while Atto633 molecules (green trajectories in Figure 5c) move rather randomly, having a greater propagation in the x-direction and lacking the preferential movement in the y-direction. Similar unstructured trajectories have also been obtained with and without an electric field for all other charged dyes in Figure 2, i.e., WS-TDI, Py-PDI, and Atto532 (see Movie S8 where an electric field strength of 1400 V/cm was applied). The directed motion of Atto532 molecules seems to occur with “jumps” from pore to pore as they traverses upward. Therefore, it is not the size of the molecule or the type and number of charges of the molecules which cause unstructured movementit is simply the fact that a molecule bears a charge at all. Taking into account that all the charged dyes that were investigated showed the same behavior in the material, we suggest that this phenomenon can be explained by the following model of the host−guest system (see Figure 6).

Figure 6. Structural model of guest molecules inside the templatefilled pores of mesoporous silica.

For the synthesis of the mesoporous silica channels we used the triblock copolymer Pluronic F127 in an evaporation-induced self-assembly (EISA) approach. This template creates mesopores by forming cylindrical micelles in the polar solvent mixture with a diameter of about 9 nm, with the micelles featuring a hydrophobic interior and a hydrophilic exterior. The hydrophilic ends of the template protrude into the silica walls and can cause intrawall porosity.28−30 The existence of such intrawall pores has already been described in the literature for Pluronic-templated mesoporous silica and has been studied by sorption measurements and electron microscopy of nanomolded platinum replicas.31−36 Moreover, in good agreement with our results, Ulrich et al. also observed diffusion perpendicular to the channel direction in template-filled SBA15 particles in a pulsed field gradient NMR study and attributed

Figure 7. Movement of charged dyes in a diagonal electric field. (a) Maximum projection of the individual frames of Movie S9 revealing the pathway of Atto633 molecules influenced by a diagonal electric field. Molecules inside the mesoporous structure are guided by the aligned mesopore channels, whereas molecules next to the microgroove follow the electric field lines. The yellow arrow indicates the direction of movement of the molecules inside the mesopores. (b) Typical trajectories 1, 2, and 3 of Atto633. 24019



Article

Figure 8. Maximum projection and average velocity. (a) Maximum projection of the individual frames of Movie S10 revealing the pathway of Atto633 molecules pulled along the silica channels by an electric field. The white arrow indicates the direction of movement of the molecules inside the mesopores. (b) Plot of the average velocity of Atto633 molecules versus electric field strength increasing from 100 to 1800 V/cm. A linear fit (y = a + bx) with a = 6 × 10−6 cm/s and b = 5 × 10−8cm2/(V s) is shown in red. (c) Absolute values of the average velocities of PhPy-PDI, WS-TDI, PyPDI, and Atto633 at 1200 V/cm.

(Py-PDI) with four positive charges, and this is in turn much faster than a very large molecule (PhPy-PDI) with also four positive charges. However, the absolute values of the mobilities of the 4-fold negatively charged WS-TDI and the 4-fold positively charged PhPy-PDI, which are comparable in size, are very similar. Additionally, other effects such as interactions with the surface of the silica nanochannels, the template inside the mesopores, and the zeta-potential can influence the mobility and movement of the dye molecules as well, but they seem to be small. A particularly instructive electrophoretic separation is shown in Figure 9, where we incorporated WS-TDI (negatively charged, molecular structure, see Figure 9) and Py-PDI (positively charged, molecular structure, see Figure 9) into the same sample. Again, we detect both fluorescence signals simultaneously in different channels. Channel 1 shows the fluorescence signal of Py-PDI molecules (see maximum projection in Figure 9a and Movie S11 in the Supporting Information) moving in the opposite direction as WS-TDI molecules in channel 2 (see maximum projection in Figure 9b

of Atto633 single molecules pulled along the silica channels. By averaging the positions (pixels converted to distance) versus time of at least five individual molecules for each electric field strength, the dependence of the average velocity on the electric field strength could be calculated (Figure 8b). The error bars depict the standard deviations of the averages. A linear trend can be seen for the increase of the average velocities with the electric field strength. From the slope we can calculate the mobility (b = 5 × 10−8 cm2/(V s)) of the Atto633 molecules in the silica host. In addition, we plotted the absolute values of the average velocities of PhPy-PDI, WS-TDI, Py-PDI, and Atto633 for 1200 V/cm into a diagram (Figure 8c). From this a clear trend of the absolute value of the mobility of the investigated single molecule species can be given as Atto633 ≥ Py−PDI ≫ PhPy−PDI ≈ WS−TDI

This trend shows that the size of the molecules plays a more important role than the number of charges per molecule. For the positively charged molecules the smallest dye (Atto633) with one charge is in a similar range than another dye molecule 24020



Article

Figure 9. Separation of differently charged dyes. The maximum projections of the individual frames of Movie S11 reveal the pathways and separation of (a) Py-PDI and (b) Ws-TDI molecules. The white arrows indicate the direction of movement of the molecules inside the mesopores under the influence of an electric field (1000 V/cm).

While this effect makes the system more complex, it can also be beneficial. On the one hand, it can provide further degrees of freedom when optimizing the system for a specific separation application. On the other hand, it also helps charged guest molecules to circumvent defects and dead ends in the mesoporous structure and hence to travel further within the mesoporous silica thin films and to cover longer distances, beyond the microscopic length scale. Besides the structural analysis, the influence of the electric field strength on charged molecules was also investigated. The velocity of the charged molecules increased monotonically with increasing electric field strength, where the slope of the plot is the mobility of the guest molecule inside the host. By applying a diagonal electric field, we could even observe molecules switching into the mesoporous structure, guided by the wellaligned pores and also switching out of the mesoporous material through defects in the silica walls. Finally, we illustrate the controlled separation of differently charged molecules in the template-filled mesoporous silica thin films that can be directly observed with our single molecule fluorescence wide-field microscope. According to the measured mobilities of different single molecule species, the separation of equally charged molecules should also be feasible with the investigated system.

and Movie S11 in the Supporting Information). By switching the electric field on and off, the directed movement of the molecules could be controlled. Thus, we clearly demonstrated the feasibility of separating differently charged individual molecules in a porous silica host. In addition, the separation of two molecules with the same charge, e.g. Ph-PDI and PhPyPDI, also appears to be possible due to their different mobilities (as indicated in Figure 8c).

■

CONCLUSION In this study, we have demonstrated the controlled movement of single molecules in highly oriented, template-filled mesoporous silica thin films under the influence of an electric field. Single molecule trajectories directly reveal the mesoporous structure of the silica thin films with their defects and dead ends. By incorporating various uncharged dyes into the mesochannels, we observed structured movement along the well-defined large-pore mesochannels in the y-direction, while charged molecules took totally different, more random diffusional pathways and showed a greater propagation in the x-direction. On the basis of these observations, we present a host−guest model that explains the different pathways of the molecules in such mesoporous materials. These structural features of the material have important implications when considering implementation of mesoporous silica thin films in nanoscale electrophoresis devices. For charged molecules (which are in general considered for electrophoretic separation), partitioning through intrawall pores has to be considered besides the separation resulting from interaction of the guests with the main mesoporous structure inside the host material.

■

MATERIALS AND METHODS Single molecule fluorescence images were recorded with a wide-field setup, using an Eclipse TE200 (Nikon) epiflourescence microscope with a high numerical aperture (NA) oil-immersion objective (Nikon Plan Apo 100×, NA = 1.40). The fluorescent uncharged dyes N-(2,6-diisopropylphenyl)-N′24021



Article

2D grazing incidence small-angle X-ray scattering (2DGISAXS) patterns were obtained with a SAXSess system from Anton Paar, equipped with a 2-D CCD detector. The wavelength of the incident beam was 0.154 nm (copper Kα radiation), and the sample−detector distance was 308 mm. Samples were measured with a tilt angle of 1° or 2° (as indicated) with respect to the primary beam. When the orientation of the silica bars is parallel to the plane of incidence of the X-rays, two strong out-of-plane reflections can be seen that vanish almost completely when rotating the sample by 90°. This is in line with a 2D hexagonal arrangement of the mesopores that are highly oriented in the substrate plane and run parallel to the silica bars. Scanning electron microscopy (SEM) images were recorded with a Jeol JSM6500F scanning electron microscope equipped with a field-emission gun, typically operated at an acceleration voltage of 5 kV and a working distance of 5−6 mm. Kr sorption experiments were carried out with an Autosorb iQ from Quantachrome Instruments at 77.3 K (liquid nitrogen temperature), assuming a saturation vapor pressure of p0 = 217 Pa for Kr. Sample outgassing was performed in vacuo for 12 h at room temperature. BET surface area analysis was carried out in the partial pressure range of p/p0 = 0.21−0.38 (p = 46−83 Pa). For the BET surface area calculation, a saturation vapor pressure of the supercooled liquid (321 Pa at 77.3 K) and a molecular cross-sectional area of 0.205 nm2 were used.26 The Kr sorption data indicate a highly porous system with a BET surface area of approximately 760 cm2/cm2 (see Figure S2 in the Supporting Information). Considering that the actually patterned area corresponds only to roughly 43% of the total substrate area, the BET surface area with respect to the porous material can be estimated to approximately 1800 cm2/cm2 for the whole film thickness or 0.44 cm2/cm2 when normalized to a film thickness of 1 nm. These results are in line with results from Bartels et al., who performed Kr sorption experiments on similar porous silica thin films and obtained values of 0.38−0.42 cm2/cm2 (also normalized to a film thickness of 1 nm).26 In some cases, thin silver layers acting as electrodes were deposited on the samples using an Oerlikon Leybold Vacuum UNIVEX 350 sputter coater system operated at a base pressure of 1 × 10−6 mbar, an argon pressure of 1 × 10−2 mbar, a power of 25 W, and a sputtering time of 30 min. The mesoporous silica thin films were prepared as described in our previous work.25 Briefly, 700 μL of Milli-Q water and 100 μL of aqueous HCl (1 mol/L) were diluted in 4.40 g of absolute ethanol. Then, 1.00 g of tetraethyl orthosilicate (TEOS) was added, and the solution was stirred (∼400 rpm) at 65 °C for 1 h. Simultaneously, a mixture of 475 mg of Pluronic F127 in 4.40 g of ethanol was stirred at room temperature for 1 h. Then, this solution was poured slowly into the ethanolic TEOS solution, and the resulting clear solution was stirred at room temperature for 2 h. Afterward, the dye molecules of interest were added to the solution (yielding a dye concentration of 10−10−10−11 mol/L); the resulting solution was used to infiltrate the voids of a PDMS stamp that was brought into close contact with a precleaned, plasma-treated standard glass coverslip and aged at room temperature for 21 h in a saturated ethanolic atmosphere and for another 24−48 h under ambient conditions. Prior to XRD and sorption measurements, the PDMS stamps were removed, and the films were calcined in air by first heating at 60 °C for 90 min (ramp: 1 °C/min), then at 160 °C for 90 min (ramp: 1 °C/ min), and finally at 400 °C for 4 h (ramp: 1 °C/min).

octylterrylene-3,4:11,12-tetracarboxdiimide (called As-TDI in this work and ref 38), N,N′-bis(2,6-diisopropylphenyl)terrylene-3,4:11,12-tetracarboxdiimide (called Dip-TDI in this work), and N,N′-bis(pentadecan-8-yl)terrylene-3,4:11,12-tetracarboxdiimide (called Sw-TDI in this work; for their chemical structures see Figure 2) with a high photostability and excellent quantum yield and the charged fluorescent dyes N,N′-bis(2,6diisopropylphenyl)-1,6,7,12-tetra(1-methyl-3-(4-phenoxy)pyridinium)perylene-3,4:9,10-tetracarboxdiimide tetraiodide (called PhPy-PDI in this work; chemical structure in Figure 2 and synthesized as described previously39), N,N′-bis(2,6diisopropylphenyl)-1,6,7,12-tetra(1-methylpyridinium-3-oxy)perylene-3,4:9,10-tetracarboxdiimide tetramethanesulfonate (called Py-PDI in this work and synthesized as described previously40), and N,N′-bis(2,6-diisopropylphenyl)-1,6,9,14tetra(4-sulfonylphenoxy)terrylene-3,4:11,12-tetracarboxdiimide (called Ws-TDI in this work and synthesized as described previously;41 chemical structures in Figure 2) were incorporated into the material in single molecule concentrations (10−10−10−11 mol/L). The molecules were excited either at 633 nm with a He−Ne gas laser (Coherent, 75 mW maximum at 632.8 nm) with an intensity of 0.3 kW cm−2 or at 532 nm with a diode pumped solid-state laser (Cobolt Samba TM laser) with an intensity of 0.05 kW cm−2. Their fluorescence was detected with a backilluminated electron multiplying charge-coupled device (EMCCD) camera in frame transfer mode (AndoriXon DV897, 512 × 512 pixels). Incident laser light was blocked by a dichroic mirror (dual line beamsplitter 532/633, AHF Analysentechnik) and by bandpass filters (675/250 and 560/40 for channel 1 and 730/140 for channel 2, AHF Analysentechnik). One pixel on the camera chip corresponds to 154 nm on the sample. The movement of the single molecules was detected in dependence of an electric field. In this study, electric field strengths up to 1800 V/cm were used. Single molecule tracking was done by fitting the fluorescence spot by a 2D-Gaussian function 2

2

f1 (x , y , A , w) = A e−((x − x0)/ w) e−((y − y0 )/ w)

where A and w are the amplitude and the width of the Gaussian curve, to the point-spread function of the signals recorded with a back-illuminated EM-CCD camera. The method was described in detail in previous studies.42−44 The resulting positions can be combined frame by frame to form molecular trajectories. If not stated otherwise, all chemicals were used as received. Absolute ethanol, tetraethyl orthosilicate (TEOS), and Pluronic F127 were purchased from Sigma-Aldrich Co. HCl (1 M) was purchased from AppliChem GmbH. Hellmanex cleaning solution was purchased from Hellma GmbH & Co. KG. PDMS and the PDMS curing agent were purchased from Dow Corning Co. Standard glass coverslips (#1, 22 × 22 mm2) were purchased from Gerhard Menzel, Glasbearbeitungswerk GmbH & Co. KG. Substrates were precleaned prior to usage by submerging them in a water/Hellmanex solution (1:100 v/v) at 60 °C for 1 h, followed by ultrasonic agitation for 3 min and rinsing with DI water. For further surface modification, the cleaned substrates were plasma-treated by exposing them to oxygen plasma for 15 min. Exposure to oxygen plasma was carried out with a Femto Plasma System from Diener Electronic typically operated at a power of 50 W and an oxygen flow of 4−5 sccm. 24022



Article

Electrodes for applying the electric field were obtained by depositing a thin silver film (typically 100 nm) with a gap of the desired width (typically 1−5 mm) on the glass coverslips by a sputter-coating process using a shadow mask, or they were made by cutting the mesoporous silica structure to the desired width after the aging process (typically 1−5 mm) with a razor blade and applying a conductive silver paste to both sides of the remaining structure, forming two pads for connection to the high-voltage generator.

■

(10) Venkatesan, B. M.; Bashir, R. Nanopore Sensors for Nucleic Acid Analysis. Nat. Nanotechnol. 2011, 6, 615−624. (11) Kumar, H.; Lansac, Y.; Glaser, M. A.; Maiti, P. K. Biopolymers in Nanopores: Challenges and Opportunities. Soft Matter 2011, 7, 5898−5907. (12) Dekker, C. Solid-State Nanopores. Nat. Nanotechnol. 2007, 2, 209−215. (13) Chen, P.; Gu, J.; Brandin, E.; Kim, Y.-R.; Wang, Q.; Branton, D. Probing Single DNA Molecule Transport Using Fabricated Nanopores. Nano Lett. 2004, 4, 2293−2298. (14) Serwer, P. Agarose Gels: Properties and Use for Electrophoresis. Electrophoresis 1983, 4, 375−382. (15) Zhu, Z.; Lu, J. J.; Liu, S. Protein Separation by Capillary Gel Electrophoresis: A Review. Anal. Chim. Acta 2012, 709, 21−31. (16) Daiguji, H.; Oka, Y.; Shirono, K. Nanofluidic Diode and Bipolar Transistor. Nano Lett. 2005, 5, 2274−2280. (17) Daiguji, H.; Yang, P. D.; Szeri, A. J.; Majumdar, A. Electrochemomechanical Energy Conversion in Nanofluidic Channels. Nano Lett. 2004, 4, 2315−2321. (18) Leebeeck, A. D.; Sinton, D. Ionic Dispersion in Nanofluidics. Electrophoresis 2006, 27, 4999−5008. (19) Garcia, A. L.; Ista, L. K.; Petsev, D. N.; O’Brien, M. J.; Bisong, P.; Mammoli, A. A.; Brueck, S. R. J.; Lopez, G. P. Electrokinetic Molecular Separation in Nanoscale Fluidic Channels. Lab. Chip 2005, 5, 1271−1276. (20) Karnik, R.; Castelino, K.; Majumdar, A. Field-Effect Control of Protein Transport in a Nanofluidic Transistor Circuit. Appl. Phys. Lett. 2006, 88, 123114−1−123114−3. (21) Karnik, R.; Fan, R.; Yue, M.; Li, D. Y.; Yang, P. D.; Majumdar, A. Electrostatic Control of Ions and Molecules in Nanofluidic Transistors. Nano Lett. 2005, 5, 943−948. (22) Oh, Y.-J.; Gamble, T. C.; Leonhardt, D.; Chung, C.-H.; Brueck, S. R. J.; Ivory, C. F.; Lopez, G. P.; Petsev, D. N.; Han, S. M. Monitoring FET Flow Control and Wall Adsorption of Charged Fluorescent Dye Molecules in Nanochannels Integrated Into a Multiple Internal Reflection Infrared Waveguide. Lab Chip 2008, 8, 251−258. (23) Oh, Y.-J.; Garcia, A. L.; Petsev, D. N.; Lopez, G. P.; Brueck, S. R. J.; Ivory, C. F.; Han, S. M. Effect of Wall-Molecule Interactions on Electrokinetic Transport of Charged Molecules in Nanofluidic Channels During FET Flow Control. Lab Chip 2009, 9, 1601−1608. (24) Zhang, Y.; Gamble, T. C.; Neumann, A.; Lopez, G. P.; Brueck, S. R. J.; Petsev, D. N. Electric Field Control and Analyte Transport in Si/SiO2 Fluidic Nanochannels. Lab Chip 2008, 8, 1671−1675. (25) Rühle, B.; Davies, M.; Lebold, T.; Bräuchle, C.; Bein, T. Highly Oriented Mesoporous Silica Channels Synthesized in Microgrooves and Visualized with Single-Molecule Diffusion. ACS Nano 2012, 6, 1948−1960. (26) Bartels, O.; Zukal, A. Krypton Adsorption Technique for Assessment of Structural Properties of Mesoporous Silica and Titania Thin Films. J. Mater. Sci. 2005, 40, 2603−2605. (27) Jung, C.; Ruthardt, N.; Lewis, R.; Michaelis, J.; Sodeik, B.; Nolde, F.; Peneva, K.; Müllen, K.; Bräuchle, C. Photophysics of New Water-Soluble Terrylenediimide Derivatives and Applications in Biology. ChemPhysChem 2009, 10, 180−190. (28) Alexandridis, P.; Hatton, T. A. Poly(ethylene Oxide)-Poly(propylene Oxide)-Poly(ethylene Oxide) Block Copolymer Surfactants in Aqueous Solutions and At Interfaces: Thermodynamics, Structure, Dynamics, and Modeling. Colloids Surf., A 1995, 96, 1−46. (29) Almgren, M.; Brown, W.; Hvidt, S. Self-Aggregation and Phase Behavior of Poly(ethylene Oxide)-Poly(propylene Oxide)-Poly(ethylene Oxide) Block Copolymers in Aqueous Solution. Colloid Polym. Sci. 1995, 273, 2−15. (30) Wanka, G.; Hoffmann, H.; Ulbricht, W. Phase Diagrams and Aggregation Behavior of Poly(oxyethylene)-Poly(oxypropylene)-Poly(oxyethylene) Triblock Copolymers in Aqueous Solutions. Macromolecules 1994, 27, 4145−4159. (31) Galarneau, A.; Cambon, H.; Renzo, F. D.; Ryoo, R.; Choi, M.; Fajula, F. Microporosity and Connections Between Pores in SBA-15

ASSOCIATED CONTENT

S Supporting Information *

Figures S1 and S2, Table S1, and Movies S1−S11. This material is available free of charge via the Internet at http://pubs.acs.org.

■

AUTHOR INFORMATION

Corresponding Authors

*E-mail [email protected] (C.B.). *E-mail [email protected] (T.B.). Author Contributions

M.D. and B.R. contributed equally to this work. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS Financial support from SFB 749, the cluster of excellence Nanosystems Initiative Munich (NIM), and the Center for NanoScience (CeNS) is gratefully acknowledged. B.R. thanks the Römer Foundation for support of his PhD thesis in 2013. M.D. thanks the Römer Foundation for support of her master thesis in 2010. All authors thank Dr. K. Römer for his continuous support of science at the Department of Chemistry at the LMU.

■

REFERENCES

(1) Kresge, C. T.; Leonowicz, M. E.; Roth, W. J.; Vartuli, J. C.; Beck, J. S. Ordered Mesoporous Molecular Sieves Synthesized by a LiquidCrystal Template Mechanism. Nature 1992, 359, 710−712. (2) Themed Collection on Mesoporous Materials. Chem. Soc. Rev. 2013, 9, 3649. (3) Sanchez, C.; Boissiere, C.; Grosso, D.; Laberty, C.; Nicole, L. Design, Synthesis, and Properties of Inorganic and Hybrid Thin Films Having Periodically Organized Nanoporosity. Chem. Mater. 2008, 20, 682−737. (4) Chang, C.; Wang, X.; Bai, Y.; Liu, H. Applications of Nanomaterials in Enantioseparation and Related Techniques. TrAC, Trends Anal. Chem. 2012, 39, 195−206. (5) Walcarius, A.; Collinson, M. M. Analytical Chemistry with Silica Sol-Gels: Traditional Routes to New Materials for Chemical Analysis. Annu. Rev. Anal. Chem. 2009, 2, 121−143. (6) Levy, M. H.; Plawsky, J.; Cramer, S. M. Photopolymerized SolGel Monoliths for Separations of Glycosylated Proteins and Peptides in Microfluidic Chips. J. Sep. Sci. 2013, 36, 2358−2365. (7) Hsieh, M.-L.; Li, G.-Y.; Chau, L.-K.; Hon, Y.-S. Single-Step Approach to Beta-Cyclodextrin-Bonded Silica as Monolithic Stationary Phases for CEC. J. Sep. Sci. 2008, 31, 1819−1827. (8) Zhong, H.; Rassi, Z. E. Capillary Electrochromatography with Monolithic Silica Columns. IV. Electrochromatographic Characterization of Polar Bonded Monolithic Stationary Phases Having SurfaceBound Cyano Functionalities. J. Sep. Sci. 2006, 29, 2031−2037. (9) Nischang, I.; Höltzel, A.; Seidel-Morgenstern, A.; Tallarek, U. Concentration Polarization and Nonequilibrium Electroosmotic Slip in Hierarchical Monolithic Structures. Electrophoresis 2008, 29, 1140− 1151. 24023



Article

Mesostructured Silicas as a Function of the Temperature of Synthesis. New J. Chem. 2003, 27, 73−79. (32) Gobin, O. C.; Wan, Y.; Zhao, D.; Kleitz, F.; Kaliaguine, S. Mesostructured Silica SBA-16 with Tailored Intrawall Porosity Part 1: Synthesis and Characterization. J. Phys. Chem. C 2007, 111, 3053− 3058. (33) Kocherbitov, V.; Alfredsson, V. Assessment of Porosities of SBA-15 and MCM-41 Using Water Sorption Calorimetry. Langmuir 2011, 27, 3889−3897. (34) Kruk, M.; Jaroniec, M.; Ko, C. H.; Ryoo, R. Characterization of the Porous Structure of SBA-15. Chem. Mater. 2000, 12, 1961−1968. (35) Ravikovitch, P. I.; Neimark, A. V. Characterization of Micro- and Mesoporosity in SBA-15 Materials From Adsorption Data by the NLDFT Method. J. Phys. Chem. B 2001, 105, 6817−6823. (36) Ryoo, R.; Ko, C. H.; Kruk, M.; Antochshuk, V.; Jaroniec, M. Block-Copolymer-Templated Ordered Mesoporous Silica: Array of Uniform Mesopores or Mesopore-Micropore Network? J. Phys. Chem. B 2000, 104, 11465−11471. (37) Ulrich, K.; Galvosas, P.; Kärger, J.; Grinberg, F.; Vernimmen, J.; Meynen, V.; Cool, P. Self-Assembly and Diffusion of Block Copolymer Templates in SBA-15 Nanochannels. J. Phys. Chem. B 2010, 114, 4223−4229. (38) Holtrup, F. O.; Müller, G. R. J.; Quante, H.; Feyter, S. D.; Schryver, F. C. D.; Müllen, K. Terrylenimides: New NIR Fluorescent Dyes. Chem.Eur. J. 1997, 3, 219−225. (39) Kohl, C.; Weil, T.; Qu, J.; Müllen, K. Towards Highly Fluorescent and Water-Soluble Perylene Dyes. Chem.Eur. J. 2004, 10, 5297−5310. (40) Qu, J.; Kohl, C.; Pottek, M.; Müllen, K. Ionic Perylenetetracarboxdiimides: Highly Fluorescent and Water-Soluble Dyes for Biolabeling. Angew. Chem., Int. Ed. 2004, 43, 1528−1531. (41) Nolde, F.; Qu, J.; Kohl, C.; Pschirer, N. G.; Reuther, E.; Müllen, K. Synthesis and Modification of Terrylenediimides as HighPerformance Fluorescent Dyes. Chem.Eur. J. 2005, 11, 3959−3967. (42) Kirstein, J.; Platschek, B.; Jung, C.; Brown, R.; Bein, T.; Bräuchle, C. Exploration of Nanostructured Channel Systems with Single-Molecule Probes. Nat. Mater. 2007, 6, 303−310. (43) Hellriegel, C.; Kirstein, J.; Bräuchle, C. Tracking of Single Molecules as a Powerful Method to Characterize Diffusivity of Organic Species in Mesoporous Materials. New J. Phys. 2005, 7, 23. (44) Hellriegel, C.; Kirstein, J.; Bräuchle, C.; Latour, V.; Pigot, T.; Olivier, R.; Lacombe, S.; Brown, R.; Guieu, V.; Payrastre, C.; et al. Diffusion of Single Streptocyanine Molecules in the Nanoporous Network of Sol-Gel Glasses. J. Phys. Chem. B 2004, 108, 14699− 14709.

24024


Insights into Nanoscale Electrophoresis of Single Dye Molecules in

Recommend Documents