Multidimensional Conducting Polymer Nanotubes for Ultrasensitive

Apr 30, 2012 - Fabrication of Various Conducting Polymers Using Graphene Oxide as a Chemical Oxidant. Minkyu Kim , Choonghyeon Lee , Young Deok Seo ...
0 downloads 0 Views 4MB Size
Download PDF
Letter pubs.acs.org/NanoLett

Multidimensional Conducting Polymer Nanotubes for Ultrasensitive Chemical Nerve Agent Sensing Oh Seok Kwon,†,§ Seon Joo Park,†,§ Jun Seop Lee,† Eunyu Park,† Taejoon Kim,‡ Hyun-Woo Park,‡ Sun Ah You,† Hyeonseok Yoon,*,‡ and Jyongsik Jang*,† †

World Class University program of Chemical Convergence for Energy & Environment, School of Chemical and Biological Engineering, Seoul National University, Seoul 151-742, Korea ‡ Alan G. MacDiarmid Energy Research Institute, Department of Polymer and Fiber System Engineering, Chonnam National University, Gwangju 500-757, South Korea S Supporting Information *

ABSTRACT: Tailoring the morphology of materials in the nanometer regime is vital to realizing enhanced device performance. Here, we demonstrate flexible nerve agent sensors, based on hydroxylated poly(3,4-ethylenedioxythiophene) (PEDOT) nanotubes (HPNTs) with surface substructures such as nanonodules (NNs) and nanorods (NRs). The surface substructures can be grown on a nanofiber surface by controlling critical synthetic conditions during vapor deposition polymerization (VDP) on the polymer nanotemplate, leading to the formation of multidimensional conducting polymer nanostructures. Hydroxyl groups are found to interact with the nerve agents. Representatively, the sensing response of dimethyl methylphosphonate (DMMP) as a simulant for sarin is highly sensitive and reversible from the aligned nanotubes. The minimum detection limit is as low as 10 ppt. Additionally, the sensor had excellent mechanical bendability and durability. KEYWORDS: Multidimensional nanostructures, conducting polymers, flexible sensors, chemical nerve agents, DMMP

O

agents are organophosphate compounds that can cause rapid and severe inhibition of serine proteases in the nervous. Typically, the inhibition of acetylcholinesterase induces acute effects in humans by reducing the breakdown of acetylcholine. Recently, diverse technical methodologies have been devised to detect the nerve gas simulant dimethyl methylphosphonate (DMMP), which is also one of organophosphate compounds, such as colorimetric assay,36 fluorometric analysis,32 surface acoustic wave,37 enzymatic assay,38 gas chromatography−mass spectrometry,39 and atomic emission detection.40 Although all of these methods have individual strengths, significant limitations, including low sensitivity, slow response times, operational complexity, high cost, and limited portability, remain.36−41 Herein, we report the fabrication of flexible sensors based on multidimensionally structured CP nanotubes that can detect the simulant of nerve agents. Specifically, hydroxylated poly(3,4-ethylenedioxythiophene) nanotubes (HPNTs) are fabricated, whose surface is decorated with substructures such as nanonodules (NNs) and nanorods (NRs). These unique multidimensional nanostructures have enhanced surface-to-volume ratio. In addition, the introduced hydroxyl groups provide differentiated responses to various analytes. Lastly, the nanotubes are aligned parallel to each other

ne-dimensional (1D) electronic nanomaterials with welldefined structures have attracted tremendous attention in various promising application areas including transistors, lasers, light-emitting diodes, solar cells, and sensors.1−9 Particularly, the 1D geometry at the nanometer scale facilitates efficient charge transport along the long-axis direction, which can lead to high sensitivity in sensor applications.10−18 To realize highperformance sensors with better sensitivity and selectivity, the following issues need to be further considered in the materials design: enlarged surface area for enhanced interaction, chemical functionality for selective responses, and structural ordering for efficient signal transfer.19 Various kinds of nanomaterials have been employed for sensor application so far, and much effort has been devoted to the control of the morphology of the nanomaterials.20−23 Conducting polymer (CP) nanomaterials have been also demonstrated as the excellent signal transducer, and especially they have strong potential for fabricating true flexible or wearable sensors.24−30 Unfortunately, however, it is still highly difficult to control the morphology of polymer materials in the nanometer regime because polymers consist of directional and thermosensitive covalent bonds. Chemical weapons were designed to kill large numbers of people in warfare. Unfortunately, they have been often used to terrorize people even in peace, not only in war. Especially, nerve gas agents such as sarin, VX, soman, and tabun are of great concern due to their odorless and colorless characteristics. Thus, intensive research has been carried out to develop sensitive, selective, and portable gas sensors.31−35 Such nerve © XXXX American Chemical Society

Received: December 30, 2011 Revised: April 18, 2012

A

dx.doi.org/10.1021/nl204587t | Nano Lett. XXXX, XXX, XXX−XXX

Nano Letters

Letter

NN (48 m2 g−1) < NR (62 m2 g−1). The nanotubes were inherently functionalized by introducing hydroxylated EDOT (HEDOT) as a comonomer during the polymerization ([EDOT]/[HEDOT] = 3.6/1) (Figure S2). The role of the hydroxyl group is to act as a reactive unit toward organophosphates that are the basis of the nerve agents.31−41 The multidimensional HPNT was integrated into a custom-made sensor substrate (Figure 2a, inset), where the HPNT functions as both a sensing element and an electrode (first and second transducers). Figure 2a shows the current−voltage (I−V) curves of the HPNTs on the sensor substrate. All HPNTs displayed ohmic behavior in their I−V characteristics, informing that the HPNTs made reliable electrical contact on the sensor substrate. The dI/dV values increased slightly, in the order SL < NN < NR, due primarily to internanotube contact resistance. Using the above sensor substrate, the ability of the random accumulated HPNTs (the thickness of HPNTs was fixed with ca. 25 μm) to perceive organophosphorus nerve agents, 5 ppb DMMP as a simulant for the nerve agent sarin, was confirmed through preliminary experiments (Figure 2b). However, several crucial factors were further considered to optimize sensor performance, such as conductive pathway and effective surface area. First, the nanofiber template was axially aligned via the electrospinning process under a magnetic field to produce highly oriented nanotubes (Figure 2c and Figure 3S). This enhanced orientation of CP nanotubes reduced conformational defects in the molecular structures and thus allowed efficient electronic delocalization.42,44 Figure 2d shows the responses of aligned and nonaligned SL-HPNTs to 5 ppb DMMP. The aligned HPNTs had better charge-transport properties (Figure 2c), and their responses were also about 30% larger than those of nonaligned HPNTs over a wide range of concentrations (Figure 2d,e and Figure S4). Figure 2f shows the effect of the nanotube surface substructure on the electrical response. The response intensity increased with increasing population of the substructures, reconfirming that precise control of the morphology of the transducer allows modulation of sensor performance. The response intensity was determined as the ΔR/R0 (%) measured when the saturated value was reached after exposure to DMMP. On the basis of these optimized conditions, the sensing capability of the HPNTs was systematically investigated. The change in resistance was recorded in real time upon cyclic exposure to DMMP and N2 stream (Figure 3a). Exposing HPNTs to DMMP elicited a precipitous rise in resistance. After the DMMP vapor was replaced by a N2 flow, the resistance recovered to the original level. The NN-HPNTs and NRHPNTs showed better responses to DMMP than did SLHPNTs (area, 10 × 10 mm; thickness, ca. 25 μm). Notably, NR-HPNTs had the lowest detection limit of 10 ppt (signal-tonoise ratio: 3.2), which is 2−3 orders of magnitude more sensitive than previously reported DMMP sensors.36−41 In previous work, we demonstrated that the oxidation level of PEDOT was extrinsically affected by hydrogen bonding rather than dielectric properties.51 Similarly, the phosphoryl group provides great strength in hydrogen-bond basicity, and thus DMMP acts as a strong hydrogen-bond base that accepts protons from the HPEDOT. Because PEDOT is a p-type semiconductor, this phenomenon is accompanied by a decrease in PEDOT conductivity. As a control, pristine PEDOT nanotubes with no functionalized hydroxyl groups did not show remarkable responses on exposure to DMMP vapors (Figure 3b). The HPNTs had rapid response times (less than 1

for efficient charge carrier transport. Consequently, the sensing response of dimethyl methylphosphonate as a simulant for sarin was highly sensitive, selective, and reversible from the HPNTs. To the best of our knowledge, this is the first example of a highperformance flexible chemical nerve agent sensor based on multidimensional CP nanostructures. Figure 1 shows the scanning electron microscopy (SEM) and transmission electron microscopy (TEM) images of hydroxy-

Figure 1. SEM images of HPNTs with (a) NN and (b) NR surface substructures (insets: TEM images). The TEM images reveal the hollow interior of the nanotubes.

lated poly(3,4-ethylenedixoythiophene) (HPEDOT) nanotubes with NNs (Figure 1a) and NRs (Figure 1b) surfaces, fabricated by vapor deposition polymerization (VDP) techniques on the electrospun nanofiber template.42−50 Conventional HPEDOT nanotubes with smooth layer (SL) surface were also prepared and employed as a control (Supporting Information, Figure 1S). Precise control over the morphology of nanostructures is of great importance for realizing a variety of future technologies.1,4 From this point of view, the above CP nanostructures can have a high surface area and excellent charge-transport properties by virtue of their unique morphology and anisotropic geometry.24−30 These advantages are expected to induce a synergetic effect that enhances device performance, particularly in sensors. Thus, the multidimensional nanostructures were adopted to fabricate high-performance flexible gas sensors for detecting chemical nerve agents. The Brunauer−Emmett−Teller (BET) surface areas of the nanotubes increased in the following order: SL (31 m2 g−1) < B

dx.doi.org/10.1021/nl204587t | Nano Lett. XXXX, XXX, XXX−XXX

Nano Letters

Letter

Figure 2. Construction of chemical nerve agent sensors based on HPNTs. (a) Schematic illustration of the custom-made sensor substrate used (left top inset) and I−V characteristics of HPNTs integrated in the sensor substrate (scan rate, 1 mV s−1). (b−e) Effect of nanotube orientation on the sensor response: (b) the response intensities of random SL, NN and NR at 5 ppb DMMP (the insets indicate SEM images); (c) I−V curves of nonaligned and aligned SL-HPNTs (the insets indicate SEM images of (upper left) aligned and (under right) nonaligned SL-HPNTs, scan rate was 1 mV s−1); (d) comparison of response intensity from nonaligned and aligned SL-HPNTs at 5 ppb DMMP; (e) the difference in response intensity between nonaligned and aligned HPNTs was plotted for 5−100 ppb DMMP (ΔR/R0Align and ΔR/R0Random note the response intensities from aligned and random HPNTs, respectively). (f) Effect of NN (blue triangle) and NR (red circle) populations on response intensity.

Figure 3. Sensing performance of chemical nerve agent sensors based on HPNTs. (a) Real-time responses of HPNTs upon cyclic exposure to DMMP (10 ppt to 50 ppb) and N2 streams. (b) Real-time response of pristine PEDOT nanotubes without hydroxyl groups to DMMP gas: the signal-to-noise ratio at 100 ppb was much low as 0.13. (c) Real-time responses of HPNTs on periodic exposure to 5 ppb DMMP. (d) Changes in response intensity of HPNTs as a function of DMMP vapor concentration: the response intensity was determined as the ΔR/R0 (%) measured when the saturated value was reached after exposure to DMMP. (e) Real-time responses of HPNTs upon consecutive exposure to DMMP (10 ppt to 50 ppb), in which the response intensity was determined as the difference from the ΔR/R0 (%) measured at the previous concentration. (f) Response differences between continuous addition and cycle DMMP/N2. The response difference was calculated by the difference between the response intensities obtained from cyclic DMMP/N2 exposure and consecutive DMMP exposure.

s) as well as recovery times (3−25 s). Additionally, they demonstrated excellent reproducibility and reversibility in responses (Figure 3a,c). The change in resistance at which the signal responses were saturated is plotted over a concentration range of 10 ppt to 50 ppb in Figure 3d. The signal intensity increased in the order SL < NN < NR, which is consistent with the tendency in surface area. The response of the HPNTs was further examined upon consecutive exposure

to different concentrations of DMMP vapor (Figure 3e and Figure S5). Compared with the case of the cyclic DMMP/N2 exposure, the response intensity was similar at low concentrations. However, the response intensity appeared to decrease at higher concentrations of more than 5 ppb, as seen in Figure 3f. A portion of the reaction sites of the HPNTs would remain bound with DMMP under the consecutive exposure, leading to the reduction in response intensity. Thus, the HPNTs need to C

dx.doi.org/10.1021/nl204587t | Nano Lett. XXXX, XXX, XXX−XXX

Nano Letters

Letter

Figure 4. Sensing performance of chemical nerve agent sensor based on HPNTs. (a) Histogram showing the response of HPNTs toward similar organophosphorus compounds at 1 ppb (TCP, MDCP, DMMP, TMP). (b) 3D graphics showing the formation of hydrogen bonds between nerve agent stimulant molecules and HEDOT. (c) Principal components analysis plot using response intensity inputs from four CP nanomaterials (NNHPNT, NR-HPNT, pristine PEDOT nanotubes, and PPy nanotubes) to the 16 analytes (including DMMP): each analyte concentration was fixed at around 4 ppm.

Figure 5. Flexible HPNT sensors. (a) Variation in resistance of NR-HPNT deposited on a 80 μm thick PET substrate for different bend radii, which were adjusted by changing distances between holding stages (see the inset): the resistances perpendicular (Rx) and parallel (Ry) to the bending direction, x and y, were measured. (b) Variation in the intensity of the response of NR-HPNT deposited on the PET substrate for different bend radii (at 100 ppt DMMP): the insets show the consequent changes in shape of the HPNT at each bend radius. (c) Photograph (the top) and schematic illustration (the bottom) of flexible HPNT sensor in wearable system. (d) Representative photographs showing finger motions during fatigue test. (e) Sensing behaviors of the flexible HPNT sensor when measured in a flat state (red) and in a curved state (blue). The fatigue test (green) was carried out by bending and relaxing the sensor for 100 times and measuring it on a flat state. The concentration of DMMP was 100 ppt. The real-time responses were measured by the data based on the parallel resistances (Ry direction).

be recovered to their initial state for the quantitative analysis of target DMMP vapor. The responses of HPNTs toward similar organophosphorus compounds were also explored to evaluate their selectivity. Figure 4a summarizes the responses of HPNTs on exposure to trimethyl phosphate (TMP), methyl dichlorophosphate (MDCP), and trichlorophosphate (TCP), used as nerve gas simulants. The ability of a phosphoryl group to form a hydrogen bond depends on the chemical environment of the compound to which it is attached. Those compounds also

contain methoxy groups that can act as weak hydrogen-bond acceptors. The hydrogen-bond strength increased in the order TCP < MDCP (QH = 1.64) < sarin (QH = 2.26) ≈ DMMP (QH = 2.28) < TMP (QH = 2.79) (the absolute value (QH) value is the difference between the negatively charged atoms and the positively charged atoms calculated with Hückel charges), which leads to different response intensities with the same tendency (Figure 4b and Table S1). The HPNTs was further tested against 15 volatile organic compounds, selected as possibly interfering with the response, and its response was D

dx.doi.org/10.1021/nl204587t | Nano Lett. XXXX, XXX, XXX−XXX

Nano Letters

Letter

gain highly aligned nanofiber arrays, magnetic field-assisted electrospinning technique was employed, where two magnets (50 000−80 000 G) were placed in parallel on the collector. The distance between two magnets was ca. 10 cm, and they were connected to the ground terminal of the power supply. The resulting PMMA nanofibers were immersed into ferric chloride (FeCl3, Aldrich)/methanol solution, as an initiator (40 mL). After drying under vacuum, the nanofibers were placed in the pressure-controllable reactor. Subsequently, the liquid monomer (EDOT or EDOT/hydroxymethyl EDOT ([EDOT]/[HEDOT] = 3.6/1) mixture, Aldrich) was injected at a controlled temperature and pressure, which resulted in the multidimensional formation of HPEDOT-coated PMMA nanofibers with nanonodules (NNs, 90 °C and 760 Torr) and nanorods (NRs, 60 °C and 760 Torr) surfaces after 4 h. The nanofiber with smooth layer (SL, 90 °C and 1 Torr) surface was also prepared. Finally, the multidimensional HPNF structures with NNs and NRs surfaces were obtained by dissolving the PMMA core with DMF solution (the diameter was ca. 100 nm). Furthermore, the population of the NN and NR substructures increased with increasing initiator concentration (2−10 wt % of FeCl3/methanol solution).

compared to responses from other sensing materials. Principal components analysis was performed on the detection data that were collected from two HPEDOT (NN and NR) nanotubes, pristine PEDOT nanotubes, and polypyrrole nanotubes. The first three principal component scores, accounting for a total 99.9% of the total data variance, are plotted in Figure 4c. Unique signatures of the analytes were observed, denoted by their segregation into separate regions of the plot, allowing identification of individual analytes by their responses. In particular, DMMP had clearly differentiable components, which validated the selective recognition ability of HPNTs. CPs are excellent candidate materials for developing flexible, wearable, or even implantable sensors because they have chemical, thermal, and mechanical properties that are similar to common plastic substrates.24−30 The HPNT deposited on a cellulose substrate was transferred to a plastic substrate, poly(ethylene terephthalate) (PET) film, via the dry-transfer method (Figure S6). To assess the mechanical flexibility of the aligned HPNT sensor on the PET substrate, the dependence of the HPNT resistance on the bend radius was examined (Figure 5a). The resistance varied little up to a bend radius of 17 mm and was completely restored to the original level after unbending even for a bend radius as high as 3 mm. The change in resistance affected the response intensity of the sensor. As shown in Figure 5b, the response intensity decreased by a maximum of 5% at a bend radius range of 3 mm, indicating an almost folded state. However, there was only a small decrease (less than 2%) in response intensity under a moderate bend radius, in the range of 10−20 mm. The real-time responses were measured by the data based on the parallel resistances (Ry direction, Figure S7). To further demonstrate the flexibility of the HPNT sensors, a wearable measuring system was built by assembling a HPNT sensor substrate on gloves (Figure 5c). Conductive silver epoxy was used to achieve stable electrical contact between the HPNT and connecting wires, and the HPNT sensor substrate was secured on gloves with aid of a polydimethylsiloxane (PDMS) rubber glue. Figure 5d shows the HPNT sensor positioned on a subject’s hand. The bending of finger caused a deformative effect comparable to a bend radius of 10 mm. A fatigue test was conducted for the flexible sensor device. Figure 5e shows the typical change in response intensity after repeated bending/ relaxing. The response decreased by less than 5% after 100 bending cycles. Thus, the HPNT sensor has excellent mechanical bendability and durability, opening the possibility of fabricating reliable flexible wearable sensors after further optimization. In conclusion, we have demonstrated the ultrasensitive detection of nerve agent stimulant, DMMP, using aligned HPNT sensors. From materials point of view, HPNT was designed to achieve highly sensitive and selective sensor performance. Fortunately, integrating the HPNT on flexible PET sensor substrate allowed detection of nerve gas agents at concentrations as low as 10 ppt. The sensing response was reversible and also durable under mechanical deformation. Moreover, the excellent flexibility of HPNT sensor was demonstrated in the wearable system, where the sensor substrate was attached on gloves. Methods. Fabrication of Multidimensional HPNTs. PMMA (1 g, Mw = 350 000, Aldrich) was dissolved in dimethylformamide (DMF) at 70−80 °C, and PMMA nanofibers with an average diameter of 60 nm were electrospun from the PMMA/DMF solution (Nano NC 60 kV/2 mA). To

■

ASSOCIATED CONTENT

* Supporting Information S

Additional information about experimental details, sample preparation, characterization, physicochemical properties, sensor performances, and SEM observation. This material is available free of charge via the Internet at http://pubs.acs.org.

■

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected] (J.J.); [email protected] (H.Y.). Author Contributions §

These authors contributed equally to this work.

Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS This research was supported by the National Research Foundation of Korea (NRF) grant (No. 2011-0017125), World Class University (WCU) program through the NRF funded by the Ministry of Education, Science and Technology (MEST) (R31-10013), and International S&T Cooperation Program (K20903002024-11E0100-04010) through the NRF funded by the MEST.

■

REFERENCES

(1) Xia, Y.; Yang, P.; Sun, Y.; Wu, Y.; Mayron, G.; Yin, Y.; Kim, F.; Yan, H. Adv. Mater. 2003, 15, 353. (2) Dong, H.; Jiang, S.; Jiang, L.; Liu, Y.; Li, H.; Hu, W.; Wang, E.; Yan, S.; Wei, Z.; Xu, W.; Gong, X. J. Am. Chem. Soc. 2009, 131, 17315. (3) Yang, P.; Fardy, M.; Yan, R. Nano Lett. 2010, 10, 1529. (4) D’Arcy, J. M.; Tran, H. D.; Tung, V. C.; Tucker-Schwartz, A. K.; Wong, R. P.; Yang, Y.; Kaner, R. B. Proc. Natl. Acad. Sci. U. S. A. 2010, 107, 19673. (5) Krogman, K. C.; Lowery, J. L.; Zach, N. S.; Rutledge, G. C.; Hammond, P. T. Nat. Mater. 2009, 8, 512. (6) Goldberger, J.; Fan, R.; Yang, P. Acc. Chem. Res. 2006, 39, 239. (7) Gao, A.; Lu, N.; Dai, P.; Li, T.; Pei, H.; Gao, X.; Gong, Y.; Wang, Y.; Fan, C. Nano Lett. 2011, 11, 3974.

E

dx.doi.org/10.1021/nl204587t | Nano Lett. XXXX, XXX, XXX−XXX

Nano Letters

Letter

(48) Nair, S.; Kim, S. H. Chem. Mater. 2009, 21, 115. (49) Laforgue, A.; Robitaille, L. Macromolecules 2010, 43, 4194. (50) Bai, X.; Hu, X.; Zhou, S.; Yan, J.; Sun, C.; Chena, P.; Li, L. J. Mater. Chem. 2011, 21, 7123. (51) Yoon, H.; Hong, J.-Y.; Jang, J. Small 2007, 3, 1774.

(8) Sorgenfrei, S.; Chiu, C.-Y; Johnston, M.; Nuckolls, C.; Shepard, K. L. Nano Lett. 2011, 11, 3739. (9) Kauffman, D. R.; Star, A. Angew. Chem., Int. Ed. 2008, 47, 6550. (10) Yoon, H.; Jang, J. Angew. Chem., Int. Ed. 2009, 48, 2755. (11) Schierhorn, M.; Boettcher., S. W.; Kraemer, D.; Stuck, G. D.; Moskovits, M. Nano Lett. 2009, 9, 3262. (12) Zhang, S.-W.; Swager, T. M. J. Am. Chem. Soc. 2003, 125, 3420. (13) Ner, Y.; Asemota, C.; Olson, R. R.; Sotzing, G. ACS Appl. Mater. Interfaces 2009, 1, 2093. (14) Potyrailo, R. A. Angew. Chem., Int. Ed. 2006, 45, 702. (15) Lee, J. I.; Cho, S. H.; Park, S.-M.; Kim, J. K.; Kim, J. K.; Yu, J.W.; Kim, Y. C.; Russell, T. P. Nano Lett. 2008, 8, 2315. (16) Roberts, M. E.; LeMieux, M. C.; Bao, Z. ACS Nano 2009, 3, 3287. (17) Liao, Y.; Zhang, C.; Zhang, Y.; Strong, V.; Tang, J.; Li, X.-G.; Kalantar-zadeh, K.; Hoek, E. M. V.; Wang, K. L.; Kaner, R. B. Nano Lett. 2011, 11, 954. (18) Yan, H.; Choe, H. S.; Nam, S. W.; Hu, Y.; Das, S.; Klemic, J. F.; Ellenbogen, J. C.; Lieber, C. M. Nature 2011, 470, 240. (19) Yoon, H.; Jang, J. Adv. Funct. Mater. 2009, 19, 1567. (20) Dua, V.; Surwade, S. P.; Ammu, S.; Agnihotra, S.; Jain, S.; Roberts, K.; Park, S.-J.; Ruoff, R. S.; Manohar, S. K. Angew. Chem., Int. Ed. 2010, 49, 2154. (21) Willner, I.; Willner, B. Nano Lett. 2010, 10, 3805. (22) Mohseni, P. K.; Lawson, G.; Couteau, C.; Weihs, G.; Adronov, A.; LaPierre, R. R. Nano Lett. 2008, 8, 4075. (23) Wu, P.; Miao, L.-N.; Wang, E.-F.; Shao, X.-G.; Yan, X.-P. Angew. Chem., Int. Ed. 2011, 50, 8118. (24) Zilberman, Y.; Ionescu, R.; Feng, X.; Mullen, K.; Haick, H. ACS Nano 2011, 5, 6743. (25) Heeger, A. J. Angew. Chem., Int. Ed. 2001, 40, 2591. (26) Long, Y.-Z.; Li, M.-M.; Gu, C.; Wan, M.; Duvail, J.-L.; Liu, Z.; Fan, Z. Prog. Polym. Sci. 2011, 36, 1415. (27) Pedrosa, V. A.; Luo, X.; Burdick, J.; Wang, J. Small 2008, 4, 738. (28) Chelawat, H.; Vaddiraju, S.; Gleason, K. Chem. Mater. 2010, 22, 2864. (29) MacDiarmid, A. G. Angew. Chem., Int. Ed. 2001, 40, 2581. (30) Huo, L.; Hou, J.; Zhang, S.; Chen, H.-Y.; Yang, Y. Angew. Chem., Int. Ed. 2010, 49, 1500. (31) Bencic-Nagale, S.; Sternfeld, T.; Walt, D. R. J. Am. Chem. Soc. 2006, 28, 5041. (32) Dale, T. J.; Rebek, J. J. Am. Chem. Soc. 2006, 28, 4500. (33) Montoro, C.; Linares, F.; Procopio, E. Q.; Senkovska, I.; Kaskel, S.; Galli, S.; Masciocchi, N.; Barea, E.; Navarro, J. A. R. J. Am. Chem. Soc. 2011, 133, 11888. (34) Ma, F.-J.; Liu, S.-X.; Sun, C.-Y.; Liang, D.-D.; Ren, G.-J.; Wei, F.; Chen, Y.-G.; Su, Z.-M. J. Am. Chem. Soc. 2011, 133, 4178. (35) Huang, J.; Miragliotta, J.; Becknell, A.; Katz, H. E. J. Am. Chem. Soc. 2007, 129, 9366. (36) Pavlov, V.; Xia, Y.; Willner, I. Nano Lett. 2005, 5, 649. (37) Yang, Y.; Ji, H.-F.; Thundat, T. J. Am. Chem. Soc. 2003, 125, 1124. (38) Ji, X.; Zheng, J.; Xu, J.; Rastogi, V. K.; Cheng, T.-C.; DeFrank, J. J.; Leblanc, R. M. J. Phys. Chem. B 2005, 109, 3793. (39) Kientz, C. E. J. Chromatogr., A 1998, 814, 1. (40) Karnati, C.; Du, H.; Ji, H.-F.; Xua, X.; Lvov, Y.; Mulchandani, A.; Mulchandani, P.; Chen, W. Biosens. Bioelectron. 2007, 22, 2636. (41) Lee, J. S.; Kwon, O. S.; Jang, J. ACS Nano 2011, 5, 7992. (42) McCann, J. T.; Lim, B.; Ostermann, R.; Rycenga, M.; Marquez, M.; Xia, Y. Nano Lett. 2007, 7, 2470. (43) Rutledge, G. C.; Fridrikh, S. V. Adv. Drug Delivery Rev. 2007, 59, 1384. (44) Greiner, A.; Wendorff, J. H. Angew. Chem., Int. Ed. 2007, 46, 5670. (45) Wei, C.; Dai, L.; Roy, A.; Tolle, T. B. J. Am. Chem. Soc. 2006, 128, 1412. (46) Peng, Q.; Sun, X.-Y.; Spagnola, J. C.; Spontak, R. J.; Parsons, G. N. Nano Lett. 2007, 7, 719. (47) Laforgue, A.; Robitaille, L. Chem. Mater. 2010, 22, 2474. F

dx.doi.org/10.1021/nl204587t | Nano Lett. XXXX, XXX, XXX−XXX