Mechanically Induced Switching of Molecular Layers - Nano Letters

Letter Cite This: Nano Lett. XXXX, XXX, XXX−XXX

pubs.acs.org/NanoLett

Mechanically Induced Switching of Molecular Layers Jonathan Berson, Markus Moosmann,† Stefan Walheim,* and Thomas Schimmel* Institute of Nanotechnology and Institute of Applied Physics, Karlsruhe Institute of Technology (KIT), 76131 Karlsruhe, Germany

Nano Lett. Downloaded from pubs.acs.org by EASTERN KENTUCKY UNIV on 01/29/19. For personal use only.

S Supporting Information *

ABSTRACT: Within the field of switchable surfaces, azobenzenes are an extensively studied group of molecules, known for reversibly changing conformation upon illumination with light of different wavelengths. Relying on the ability of the molecules to change properties and structure as a response to external stimuli, they have been incorporated in various devices, such as molecular switches and motors. In contrast to the well-documented switching by light irradiation, we report the discovery of mechanically triggered switching of self-assembled azobenzene monolayers, resulting in changes of surface wettability, adhesion, and friction. This mechanically induced cis−trans isomerization is triggered either locally and selectively by AFM or macroscopically by particle impact. The process is optically reversible, enabling consecutive switching cycles. Collective switching behavior was also observed, propagating from the original point of impact in a domino-like manner. Finally, local force application facilitated nondestructive and erasable nanopatterning, the cis−trans nanolithography. KEYWORDS: Molecular switch, mechanical switching, azobenzene, self-assembled monolayers, nanolithography

M

the energy needed for the release of the metastable cis conformation and triggers the switching process into the lower energy trans conformation. The switching behavior was tested on two different azobenzene monolayers and was corroborated and characterized by changes in UV−vis spectra and monolayer height as measured by AFM. Further measurements revealed that the monolayer-coated surface also changes its friction, adhesion, and wettability as a result of isomerization. Furthermore, our capability to locally affect the monolayer enabled the investigation of the collective switching behavior propagating laterally throughout the monolayer, as well as facilitated monolayer cis−trans nanolithography (CTNL). For a deep understanding of a switching monolayer system, the surface has to be characterized not only by averaging measurement methods like spectroscopy or electrical measurements but also by imaging techniques which map the differences between the states in terms of surface properties such as height or friction. However, the common scanning probe-based surface characterization techniques only determine relative differences and in lack of a reference yield no absolute values of the surface properties at each individual state. To include a reference, imaging efforts so far have relied on single molecules embedded in a background matrix,6 monolayers containing both switching states at the same time5,21 or incomplete and rough monolayers.22 In this study, we have been able to image the switching of densely packed and highly ordered monolayer regions. This was achieved by

olecular switches bear great potential for fabrication of nanoscale electromechanical and optoelectronic devices.1 They are based on molecules which reversibly change their characteristics, such as shape or structure, electronic, mechanical, or optical properties, as a response to external stimuli. A prominent class of such molecules is azobenzenes, which are known to reversibly photoswitch between the cis and trans conformations as a response to UV and blue light stimuli, respectively. As a test case for molecular switching, the behavior of azobenzene-functionalized systems, including polymers,2 supramolecular systems,3 liquid crystals,4 monolayers5 as well as single molecules,6 have been extensively studied. They were utilized, among other uses, for electrical,7,8 optical,9,10 biological,11,12 and mechanical13,14 applications. Some investigations have been dedicated to the mechanical work that is induced by the switching event between the cis (folded, higher energetic) and trans (extended, lower energetic) conformations of the azo bond in monolayer15 and polymeric systems.16,17 Calculations18 and experiments19 have also shown that two-side anchored azobenzene molecules are elongated to the trans conformation but only when both sides of the molecule are pulled apart, the same way other folded molecules elongate when stretched.20 In this report, we introduce immediate cis-to-trans isomerization of an azobenzene-functionalized silane monolayer induced by mechanical force, both locally with an AFM tip and macroscopically with a stream of CO2 snow particles (snowjet). Unlike previous work to stretch the azo bond, both the AFM and the snowjet particles introduce mechanical force into the system in a vector that actually works to compress the molecule rather than elongate it. However, this force provides © XXXX American Chemical Society

Received: October 4, 2018 Revised: December 5, 2018

A

DOI: 10.1021/acs.nanolett.8b03987 Nano Lett. XXXX, XXX, XXX−XXX

Letter

Nano Letters

Figure 1. Synthetic route to DABSYL@FDTS islands. AFM topography images and averaged cross sections of the marked areas of a silicon substrate structured by PBL. The structures are composed of FDTS monolayer background with (a) holes of bare SiOX, (b) holes filled with APTES monolayer, and (c) DABSYL-functionalized islands, obtained by reacting the APTES monolayer with DABSYL-Cl.

Figure 2. UV- and mechanically induced switching of monolayer islands. (a) AFM topography (left) and friction (right, linear color scale) images and cross sections (middle) of a DABSYL@FDTS island sample: The scan started at the top, was paused in the middle at the marked line and resumed after in situ UV illumination for 20 s. The height of the islands decreased by 0.2 nm and the friction increased after the illumination. (b) AFM topography (left and center) and friction (right) images of mechanically switched DABSYL islands in the inner area by scanning with a force of 30 nN.

holes. The 1.2 nm deep Si holes are functionalized with an azobenzene monolayer in a two-step process, first by the selfassembly of a (3-aminopropyl)triethoxysilane (APTES) monolayer,24 rendering the height difference to 0.6 nm below the FDTS background (Figure 1b), and then by reacting the ATPES monolayer with either 4-dimethylaminoazobenzene-4′-sulfonyl chloride25 (DABSYL-Cl, an azobenzene molecule of the pseudostilbenes class) or 4-phenylazobenzenesulfonyl chloride (PhABSYL-Cl). Subsequent to the reaction between the sulfonyl-chloride function of the azobenzene molecules and the amine groups, the holes gained 1.2 nm in height to now create “islands” 0.6 nm above the background

creating a surface architecture of well-defined islands of azobenzene-functionalized monolayer embedded in a matrix of a reference monolayer, which is chemically inert and unaffected by the switching stimuli. For this purpose, we have utilized polymer blend lithography (PBL)23 to create monolayer patterns on a smooth silicon substrate by polymer phase separation, leading to the creation of a structure of a 1H,1H,2H,2H-perfluorodecyltrichlorosilane (FDTS) monolayer with empty bare Si “holes” (Figure 1a) . The robust and inert FDTS monolayer matrix has a known height of 1.2− 1.3 nm23 and is applied from this stage of the experiment as a reference system to monitor changes that occur inside the B

DOI: 10.1021/acs.nanolett.8b03987 Nano Lett. XXXX, XXX, XXX−XXX

Letter

Nano Letters

Figure 3. Mechanical−optical switching cycles. (a) Schematic representation of the cis-to-trans and trans-to-cis switching mechanisms, optical (left) and mechanical (right). (b) UV−vis spectra of a quartz slide functionalized on both sides with a PhABSYL monolayer. The peak at 320 nm, associated with the trans conformation, loses intensity after illumination with UV light increases again after snowjet treatment. In contrast, the cisassociated peak at 445 nm appears as a result of UV illumination and disappears after snowjet. (c) Cycling between cis and trans DABSYL monolayer conformations measured by static water contact angle, showing average values of 74.4° ± 1.3° and 68.5° ± 0.8° after snowjet and UV illumination, respectively.

is a well-documented phenomenon, we were able to switch the azobenzene-functionalized monolayer in the reverse direction in a previously unprecedented way. The topographic AFM image in Figure 2b (left) shows a DABSYL@FDTS system that has been illuminated with UV light (0.5 W/cm2) for 2 min prior to scanning, reducing the height of the DABSYL islands to 0.2 nm above the FDTS level. After two AFM scans with a force of 30 nN in the inner 5 μm × 5 μm area, marked with the white dashed square, the tip force load is reduced to 5 nN and the scan size was increased to 10 μm × 10 μm to include the surrounding area. The image reveals an increase in height of the affected islands from 0.2 to 0.6 nm, as well as a decrease in friction (right). This increase in island topography seems a counterintuitive response to compression by the AFM tip, but it implies that the mechanical force provides the energy needed to induce monolayer relaxation from the energetically higher cis conformation to the lower trans, comparable to the spring in a ballpoint pen being released by pressing. The observed 0.4 nm increase in height is accompanied by a reduction of 55% in surface adhesion (Supporting Information Section 3). The middle image of Figure 2b zooms in on two islands at the top part of the scanned area. A noticeable phenomenon is that dabsylated monolayer islands that were on the border of the inner area and hence only partially scanned nevertheless switched in their entirety. This indicates a cooperative switching mechanism in which monolayer molecules induce conformational changes in neighboring ones in a “domino-like” effect, presumably through π−π interactions.5,26,32,33 Thus, the

FDTS level (i.e., total monolayer height of 1.8 nm, Figure 1c) in the DABSYL case and 0.2 nm (total monolayer height of 1.4 nm) in the PhABSYL case (Figure S9b). Figure 2a displays the trans-to-cis switching behavior of the DABSYL islands inside the FDTS matrix (DABSYL@FDTS) as a response to UV illumination (the additional dimethylamino head functional group in the DABSYL monolayer makes for more distinct height differences between the cis and trans states than in the case of PhABSYL monolayers, similar results of PhABSYL are shown in Figure S9). An AFM scan (force load of 5 nN) was paused halfway and the sample was then illuminated in situ with UV light through an optical fiber for 20 s (6 mW/cm2, in situ illumination length and intensity were limited due to heating of the AFM). When the scan was resumed, the partial switching of the dabsylated monolayer islands from trans to cis was evident by a decrease in island height from 0.6 to 0.4 nm above the FDTS level as well as a significant increase in friction. The fact that the monolayer can switch to the cis state is noteworthy, as it is only enabled by a certain monolayer density window. Excessive monolayer density has been reported to inhibit switching;26,27 on the other hand, although the monolayer density is not high enough to sterically hinder the switching, it appears to be dense enough to facilitate π−π interactions between neighboring monolayer molecules, which stabilize the otherwise unstable cis conformation of the pseudostilbene class,28−30 known to thermally relax to the trans conformation within milliseconds.31 Although UV light-induced trans-to-cis switching C

DOI: 10.1021/acs.nanolett.8b03987 Nano Lett. XXXX, XXX, XXX−XXX

Letter

Nano Letters

Figure 4. Site-selective mechanical switching. (a) Illustration of the conformational switching propagating from a single point of impact. (b) AFM topography (top) and friction (bottom) images of such DABSYL@FDTS islands. As schematically shown in (a), the middle islands were individually switched by pressing the AFM tip at a single central point, which induced a domino-effect-like spreading throughout the entire island.

intensity peak around 445 nm, characteristic of the n-π* transition of the cis conformation (blue curve). The reverse cis-to-trans switching event by a short snowjet treatment resulted in restoration of the initial spectrum, where the peak in the blue range is hardly visible and the UV peak retains its original intensity. Repeatable snowjet−UV illumination switching cycles were also demonstrated by contact angle measurements (Figure 3c). DABSYL monolayers were chosen for these measurements as the hydrophilicity differences between the trans state, in which the dimethylamino headgroup is exposed, and the folded molecule at the cis state are larger than in the case of the PhABSYL monolayer, where the differences between the phenyl ring headgroup in the trans state and the folded molecule in the cis case are negligible. The results display four switching cycles, in which the static water contact angle changed from 74.4° ± 1.3° for the trans state of the dabsylated monolayer (induced by snowjet treatment) to 68.5° ± 0.8° for the cis state after UV illumination. This difference of about 6° matches values measured for other dimethylaminoazobenzene monolayer systems.34 This switching behavior showed no signs of fatigue for over 30 switching cycles. Noticeably, AFM, UV− vis spectroscopy, and contact angle goniometry did not reveal any signs of monolayer cis-to-trans or trans-to cis isomerization, as a result of heating (90 °C for 3 h), cooling (4 °C for 18 h), or relaxation over time (two months). The intermolecular π−π interactions and the dense monolayer structure are a plausible explanation for the monolayer staying “locked” in a conformation in accordance with the last switching action that has taken place and showing no signs of thermal relaxation. This observation was not only true for entire cis or trans surfaces but also for those with mixed monolayer conformations such as the one displayed in Figure 2b or trans patterns created against a cis background by CTNL (vide infra). The lateral spreading upon mechanical impact, schematically presented in Figure 4a, was observed to be highly dependent on monolayer density (Supporting Information Section 2). Figure 4b shows topography (top) and friction (bottom) AFM

monolayer structure was dense enough to enable long-ranged switching based on short-ranged interactions, with entire islands being switched even when as little as 4% of their area were scanned. In fact, it was even possible to switch entire islands from a single point of AFM tip impact (vide infra). To learn the maximal distance of the switching spreading, similar AFM tip-induced switching was conducted on entire uniform surfaces with the dabsylated self-assembled monolayer, displaying an increase in height of 0.6 nm (Figure S7) and a decrease in friction (Figure S5). Noticeable spreading could be observed at up to 1.5 μm away from the edge of the inner scanned area. Analogous control experiments on both FDTS and APTES monolayers did not result in any differences between a prescanned area and its periphery. Further support to our conclusion of the mechanical and optical conformational switching mechanisms schematically shown in Figure 3a was obtained from the investigation of macroscopic surface properties of uniform azobenzenefunctionalized monolayer samples. To complement the nanoscale mechanical cis−trans switching of the monolayer by AFM, a suitable tool for macroscopic switching was found in the snowjet system. This system, which employs a concentrated jet of CO2 snow and is conventionally used to remove contamination from surfaces by mechanical impact, was utilized here for mechanical cis−trans switching (Figure S7). Spectroscopic investigation of UV and snowjet switching of the monolayer was performed on quartz slides modified with uniform PhABSYL monolayers (Figure 3b). PhABSYL was preferred over DABSYL as the choice monolayer for spectroscopy as the cis and trans peaks of pseudostilbene overlap and the differences between the states are harder to distinguish (spectroscopic results of the dabsylated monolayer are shown in Figure S8). The initial spectrum (black) displays a large trans peak around 320 nm, characteristic of the π−π* transition of the trans state. Thirty seconds of UV illumination induced a transto-cis conformational switching as indicated by the peak decreasing in intensity and the appearance of a wide, lowD

DOI: 10.1021/acs.nanolett.8b03987 Nano Lett. XXXX, XXX, XXX−XXX

Letter

Nano Letters

Figure 5. CTNL process. (a) Schematic illustration of CTNL, here with the tip of an AFM scanning from left to right and inducing local conformational change of the monolayer molecules. (b) AFM topography image of the logo of the Karlsruhe Institute of Technology (KIT), created by CTNL on a uniform dabsylated monolayer. (c) AFM topography (left) and friction (right) images of a cis−trans grid pattern with linewidths of about 50 nm, as displayed in the cross section (middle) along the marked line.

images of DABSYL@FDTS islands, where the dabsylation reaction was performed at room temperature. The switching of the selected islands, creating a smiley pattern of trans islands with a 0.5 nm increase in height and a notably lower friction, was in this case not conducted by scanning the sample but rather by pressing the AFM tip at a single impact point at the middle of each island (see schematic animation of the process in Supporting Information Movie S1). However, in order to create higher resolution cis−trans patterns on a uniform dabsylated monolayer, the creation of monolayers with reduced spreading was required. This was achieved by elevating the temperature of the derivatization of the APTES monolayer with DABSYL-Cl to 70 °C as suggested in the original procedure25 or by illuminating the reaction vessel with UV light during dabsylation, which yielded monolayers of lower density (with DABSYL functionalization density estimated at about 75% of that of room-temperature dabsylated monolayers, see Supporting Information Sections 1 and 2). As a result, the short-ranged interaction between neighboring molecules decreased, which limited spreading. The utilization of the lower-density uniform dabsylated monolayer surface synthesized at 70 °C for cis−trans nanolithography (CTNL) by raster patterns of the AFM tips is schematically described in Figure 5a. An example of CTNL is shown with an AFM topography image of an inscribed logo of the Karlsruhe Institute of Technology (Figure 5b reproduced with permission of the Karlsruhe Institute of Technology). Using this process, we were able to create structures with features as narrow as few tens of nanometers (Figure 5c). As an expansion to the field of switching surfaces, we show here the switching of monolayers by application of mechanical force. While azobenzene-based systems have already been proposed for numerous applications and devices,1 the introduction of an additional activation method for the conformational change broadens the scope of possible applications not only in specific cases where molecules are addressed and anchored at both edges but generally in systems where an external mechanical energy stimulus can be introduced.

Dense packing of azobenzene molecules has been reported to be a good solution to stabilize an otherwise unstable cis conformation.35,36 On the other hand, in some systems the cis conformation is so stable that it slows down back-switching.37 We maintain the stabilizing effect provided by a densely packed monolayer but suggest the mechanical isomerization as a solution for back-switching, as it has proven to be a potent and fast trigger. Moreover, although applications are often limited by long switching times, mechanically induced isomerization is practically immediate. Additionally, we have shown in this work that CO2 snowjet, which is known as a simple and widely used tool for cleaning surfaces, can also be utilized as a tool to induce changes on the molecular level. Further studies may discover that snowjet may be applicable to other bistable systems. The results demonstrate that adjustment of the synthesis conditions enables control of the interplay between short ranged interactions in the monolayer and localization of the switching. This in turn facilitates the creation of erasable and rewritable cis−trans surface patterns for possible applications in the design of devices. Materials and Methods. All materials were used as received from the manufacturers Fabrication of the FDTS-DABSYL and FDTS-PhABSYLFunctionalized Monolayer Islands Samples (DABSYL@FDTS and PhABSYL@FDTS). On the basis of the procedure reported in ref 23, polymer blend lithography (PBL) was applied for the creation of a monolayer island pattern. Poly(methyl methacrylate) (PMMA, MW = 9.59 kg/mol, PDI = 1.05) and polystyrene (PS, MW = 96 kg/mol, PDI = 1.04) were purchased from Polymer Standards Service and dissolved in methyl ethyl ketone (MEK, Aldrich) at a PS/PMMA ratio of 3:7 and a solution concentration of 15 mg polymer/ml. Thirtyfive microliters of the polymer solution was spin-cast on a ∼1.5 cm2 Si substrate (single side polished ⟨100⟩ Si, thickness 675 ± 50 μm, Siegert Consulting e.K., Germany) freshly cleaned by the snowjet procedure, in which a stream of CO2 particles hits the surface (Snow Jet model K4-05, Tectra, Germany). The spin coating process was performed at a rate of 1500 rpm and at ambience with controlled humidity of 40% RH. The subsequent polymer pattern of PS islands against a PMMA E

DOI: 10.1021/acs.nanolett.8b03987 Nano Lett. XXXX, XXX, XXX−XXX

Letter

Nano Letters

Inc., U.S.A.) were recorded with a Cary-500 instrument (Varian, U.S.A.) against a reference of a clean quartz substrate. Ellipsometry. Ellipsometry of azobenzene monolayers, grown in situ on APTES-modified Si substrates, was performed using an EL X-02C ellipsometer (DRE, Germany).

background was shortly immersed in acetic acid (VWR International) to selectively dissolve the PMMA, leaving the PS cylinders as a mask for monolayer deposition. This stage was immediately followed by 1H,1H,2H,2H-perfluorodecyltrichlorosilane (FDTS, Alfa Aesar, 96%) gas phase self-assembly on the bare Si background by placing the samples overnight in a desiccator containing a few drops of liquid FDTS, evacuated to a pressure of 50 mbar. After the FDTS monolayer had been formed, the PS islands were dissolved by tetrahydrofuran (THF, VWR International) and the surface was cleaned by snowjet treatment. Functionalization of the bare Si sites starts with selfassembly of (3-aminopropyl)-triethoxysilane (APTES, 99% Aldrich). The self-assembly reaction was performed by immersion for 20 min of the samples in a 0.04 M APTES solution in methanol (max. 0.003% H2O, Merck) with 5% v/v Millipore water (drawn from Mill-Q Biocel, Millipore) and 1 mM acetic acid.24 The samples were then rinsed in water and ethanol (VWR International) and were immediately reacted to attach the azobenzene moiety. In the case of dabsylated monolayers, 4-dimethylaminoazobenzene-4′-sulfonyl chloride (DABSYL-Cl, ≥97.5%, Sigma-Aldrich) was used. In the reaction, adapted from ref 25, freshly prepared 1 mM DABSYL-chloride solution in acetone (VWR International) and 0.2 M sodium bicarbonate buffer (pH 9.0) (Sigma) were mixed at 65:35 volume ratio. Depending on the desired outcome, the samples were immersed in the reaction mixture for 5 min at room temperature or 10 min at 70 °C (Supporting Information Section 2) and rinsed with acetone, ethanol, and water. PhABSYL monolayers were created with similar reaction mixtures of 65:35 1 mM 4-phenylazobenzenesulfonyl chloride (98%, ABCR, Germany) and sodium bicarbonate buffer (pH 9.0). The reaction kinetics proved to be slower and needed at least 4 h at room temperature to create well-ordered monolayers. AFM Imaging and Mechanical Monolayer Switching. AFM imaging and surface patterning was carried out in contact mode using a Bruker Dimension ICON system. The tips that were used were Mikromasch (HQ/CSC37/AlBS and HQ/ CSC37/Pt (typical force constant: 0.3 N/m). The force constant of each individual tip was calibrated prior to work in order to control the exact force exerted by the tip on the sample. Images were obtained with a force load of 5 nN whereas monolayer switching by scanning was feasible at force loads starting at 10 nN with standard patterning being performed at 30 nN (see discussion of the threshold forces required for switching at Supporting Information Section 4). Sample Illumination. Monolayer samples were illuminated by a LQ-HXP-120 CUR UV light source (Leistungselektronik JENA GmbH, Germany) through a 400 nm Techspec Shortpass filter (Edmund Optics, USA). For the in situ illumination during AFM operation, we used an optical fiber to guide the light to a distance of about 1 cm from the sample. The intensities using different filters and distances between source and sample were determined with a NanoCalc 2000 spectrometer (mikropack, now ocean optics) by comparison to a light source of known spectral intensities. Contact Angle Measurements. The contact angle of static water drops (Millipore) was measured using an OCA 20 goniometer (Data Physics, Germany). UV−vis Spectroscopy. UV−vis spectra of monolayer samples on quartz substrates (1 mm thickness, Ted Pella

■

ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.nanolett.8b03987. Estimation of the density ratio between monolayer synthesized at room temperature and at 70 °C by ellipsometry; discussion of the dependence of monolayer density and properties on synthesis conditions; changes in surface adhesion upon switching; discussion of threshold forces required to switch; AFM imaging of a switching cycle; UV−vis spectroscopy of DABSYL monolayer; AFM switching of PhABSYL monolayers (PDF) Schematic animation of monolayer switching spreading from a single impact point (AVI)

■

AUTHOR INFORMATION

Corresponding Authors

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

Jonathan Berson: 0000-0002-4979-6359 Present Address †

(M.M.) Nees Institute for Biodiversity of Plants, University of Bonn, 53115 Bonn, Germany Author Contributions

J.B. and M.M. contributed equally. J.B. and M.M. designed the experimental setup, prepared the samples, UV/vis spectroscopy, contact angle goniometry, AFM measurements, data analysis and summary, and wrote the manuscript. S.W. conceived and supervised the research. T.S. initiated, conceived, and supervised the research. Funding

This research was funded by the Baden-Wuerttemberg Stiftung gGmbH, Stuttgart. Notes

The authors declare no competing financial interest.

■ ■

ACKNOWLEDGMENTS The authors wish to thank Moritz Fischer for the animation and Christian Grupe for the schematic drawing. ABBREVIATIONS AFM, atomic force microscope; UV, ultraviolet; UV−vis, ultraviolet and visible light; CTNL, cis−trans nanolithography; PBL, polymer blend lithography; FDTS, 1H,1H,2H,2Hperfluorodecyltrichlorosilane; APTES, (3-aminopropyl)triethoxysilane; DABSYL-Cl, 4-dimethylaminoazobenzene-4′sulfonyl chloride; PhABSYL-Cl, 4-phenylazobenzenesulfonyl chloride; PMMA, poly(methyl methacrylate); PS, polystyrene

■

REFERENCES

(1) Pathem, B. K.; Claridge, S. A.; Zheng, Y. B.; Weiss, P. S. Annu. Rev. Phys. Chem. 2013, 64 (1), 605−630.

F

DOI: 10.1021/acs.nanolett.8b03987 Nano Lett. XXXX, XXX, XXX−XXX

Letter

Nano Letters (2) Hosono, N.; Kajitani, T.; Fukushima, T.; Ito, K.; Sasaki, S.; Takata, M.; Aida, T. Science 2010, 330 (6005), 808−811. (3) Nalluri, S. K. M.; Voskuhl, J.; Bultema, J. B.; Boekema, E. J.; Ravoo, B. J. Angew. Chem., Int. Ed. 2011, 50 (41), 9747−9751. (4) Ikeda, T.; Tsutsumi, O. Science 1995, 268 (5219), 1873−1875. (5) Pace, G.; Ferri, V.; Grave, C.; Elbing, M.; von Hanisch, C.; Zharnikov, M.; Mayor, M.; Rampi, M. A.; Samori, P. Proc. Natl. Acad. Sci. U. S. A. 2007, 104 (24), 9937−42. (6) Kumar, A. S.; Ye, T.; Takami, T.; Yu, B. C.; Flatt, A. K.; Tour, J. M.; Weiss, P. S. Nano Lett. 2008, 8 (6), 1644−8. (7) Mativetsky, J. M.; Pace, G.; Elbing, M.; Rampi, M. A.; Mayor, M.; Samorì, P. J. Am. Chem. Soc. 2008, 130 (29), 9192−9193. (8) Li, S.; Feng, Y.; Long, P.; Qin, C.; Feng, W. J. Mater. Chem. C 2017, 5 (21), 5068−5075. (9) Tong, X.; Wang, G.; Yavrian, A.; Galstian, T.; Zhao, Y. Adv. Mater. 2005, 17 (3), 370−374. (10) Kuzyk, A.; Yang, Y.; Duan, X.; Stoll, S.; Govorov, A. O.; Sugiyama, H.; Endo, M.; Liu, N. Nat. Commun. 2016, 7, 10591. (11) Gorostiza, P.; Isacoff, E. Y. Science 2008, 322 (5900), 395−399. (12) Dong, M.; Babalhavaeji, A.; Collins, C. V.; Jarrah, K.; Sadovski, O.; Dai, Q.; Woolley, G. A. J. Am. Chem. Soc. 2017, 139 (38), 13483− 13486. (13) Yamada, M.; Kondo, M.; Mamiya, J.-i.; Yu, Y.; Kinoshita, M.; Barrett, C. J.; Ikeda, T. Angew. Chem. 2008, 120 (27), 5064−5066. (14) Abendroth, J. M.; Bushuyev, O. S.; Weiss, P. S.; Barrett, C. J. ACS Nano 2015, 9 (8), 7746−7768. (15) Ferri, V.; Elbing, M.; Pace, G.; Dickey, M. D.; Zharnikov, M.; Samorì, P.; Mayor, M.; Rampi, M. A. Angew. Chem. 2008, 120 (18), 3455−3457. (16) Yu, Y.; Nakano, M.; Ikeda, T. Nature 2003, 425 (6954), 145− 1450000. (17) Hugel, T.; Holland, N. B.; Cattani, A.; Moroder, L.; Seitz, M.; Gaub, H. E.; et al. Science 2002, 296 (5570), 1103−6. (18) Turansky, R.; Konopka, M.; Doltsinis, N. L.; Stich, I.; Marx, D. Phys. Chem. Chem. Phys. 2010, 12 (42), 13922−32. (19) Surampudi, S. K.; Patel, H. R.; Nagarjuna, G.; Venkataraman, D. Chem. Commun. 2013, 49 (68), 7519−7521. (20) Huang, W.; Zhu, Z.; Wen, J.; Wang, X.; Qin, M.; Cao, Y.; Ma, H.; Wang, W. ACS Nano 2017, 11 (1), 194−203. (21) Wen, Y.; Yi, W.; Meng, L.; Feng, M.; Jiang, G.; Yuan, W.; Zhang, Y.; Gao, H.; Jiang, L.; Song, Y. J. Phys. Chem. B 2005, 109 (30), 14465−14468. (22) Ishikawa, D.; Ito, E.; Han, M.; Hara, M. Langmuir 2013, 29 (14), 4622−31. (23) Huang, C.; Moosmann, M.; Jin, J.; Heiler, T.; Walheim, S.; Schimmel, T. Beilstein J. Nanotechnol. 2012, 3, 620−8. (24) Hermanson, G. T. Silane Coupling Agents. In Bioconjugate Techniques, second ed.; Hermanson, G. T., Ed.; Academic Press: New York, 2008; Chapter 13, pp 562−581. (25) Stocchi, V.; Cucchiarini, L.; Piccoli, G.; Magnani, M. Journal of Chromatography A 1985, 349 (1), 77−82. (26) Moldt, T.; Brete, D.; Przyrembel, D.; Das, S.; Goldman, J. R.; Kundu, P. K.; Gahl, C.; Klajn, R.; Weinelt, M. Langmuir 2015, 31 (3), 1048−1057. (27) Samanta, D.; Gemen, J.; Chu, Z.; Diskin-Posner, Y.; Shimon, L. J. W.; Klajn, R. Proc. Natl. Acad. Sci. U. S. A. 2018, 115, 9379. (28) Barrett, C. J.; Mamiya, J.-i.; Yager, K. G.; Ikeda, T. Soft Matter 2007, 3 (10), 1249−1261. (29) Raimondo, C.; Kenens, B.; Reinders, F.; Mayor, M.; Uji-i, H.; Samorì, P. Nanoscale 2015, 7 (33), 13836−13839. (30) Qian, H.; Wang, Y.-Y.; Guo, D.-S.; Aprahamian, I. J. Am. Chem. Soc. 2017, 139 (3), 1037−1040. (31) Whitten, D. G.; Wildes, P. D.; Pacifici, J. G.; Irick, G. J. Am. Chem. Soc. 1971, 93 (8), 2004−2008. (32) Weber, C.; Liebig, T.; Gensler, M.; Zykov, A.; Pithan, L.; Rabe, J. P.; Hecht, S.; Bléger, D.; Kowarik, S. Sci. Rep. 2016, 6, 25605. (33) Pechenezhskiy, I. V.; Cho, J.; Nguyen, G. D.; Berbil-Bautista, L.; Giles, B. L.; Poulsen, D. A.; Fréchet, J. M. J.; Crommie, M. F. J. Phys. Chem. C 2012, 116 (1), 1052−1055.

(34) Zhang, X. Y.; Wen, Y. Q.; Li, Y. F.; Li, G.; Du, S. X.; Guo, H. M.; Yang, L. M.; Jiang, L.; Gao, H. J.; Song, Y. J. Phys. Chem. C 2008, 112 (22), 8288−8293. (35) Han, M.; Ishikawa, D.; Honda, T.; Ito, E.; Hara, M. Chem. Commun. 2010, 46 (20), 3598−3600. (36) Klajn, R. Pure Appl. Chem. 2010, 82 (12), 2247. (37) Elbing, M.; Błaszczyk, A.; von Hänisch, C.; Mayor, M.; Ferri, V.; Grave, C.; Rampi, M. A.; Pace, G.; Samorì, P.; Shaporenko, A.; Zharnikov, M. Adv. Funct. Mater. 2008, 18 (19), 2972−2983.

G

DOI: 10.1021/acs.nanolett.8b03987 Nano Lett. XXXX, XXX, XXX−XXX