Effect of Nanoconfinement on the Collapse Transition of Responsive

(29) However, more confined nanobrushes, for which the fanning-out affects all ...... Mario Tagliazucchi , Martin G. Blaber , George C. Schatz , Emily...
0 downloads 0 Views 821KB Size
Download PDF
NANO LETTERS

Effect of Nanoconfinement on the Collapse Transition of Responsive Polymer Brushes

2008 Vol. 8, No. 11 3819-3824

Alain M. Jonas,*,†,§ Zhijun Hu,† Karine Glinel,‡,⊥ and Wilhelm T. S. Huck*,‡ Research Center in Micro- and Nanoscopic Materials and DeVices (CeRMiN), UniVersite´ catholique de LouVain, Place Croix du Sud, 1, B1348 LouVain-la-NeuVe (Belgium), and MelVille Laboratory for Polymer Synthesis, Department of Chemistry, UniVersity of Cambridge, Lensfield Road, Cambridge CB2 1EW U.K. Received July 18, 2008; Revised Manuscript Received September 15, 2008

ABSTRACT Nanopatterned brushes of a thermo-responsive polymer, poly(2-(2-methoxyethoxy)ethyl methacrylate) (PMEO2MA), displaying a collapse temperature in the physiological range were synthesized for grafting diameters from a few micrometers down to 35 nm. The reversible collapse transition of the nanobrushes was studied in water as a function of their lateral confinement, down to ensembles of brushes containing only ∼300 chains. The confinement results in a considerable broadening of the collapse transition and in an increase of the degree of vertical swelling, which can be explained by the internal structure of the nanodroplets derived from a theoretical model of dry nanobrushes. These results enable the rational design of responsive surfaces having a tunable topography engineered at the nanometer scale, which is of direct interest for the development of soft nanoactuators and new substrates for cell adhesion studies.

Densely grafted polymer chains form submicrometer-thick brushes in which the chains are in a stretched conformation.1-4 Brushes of stimuli-responsive polymers have attracted considerable attention,5-7 because they offer a way to tune the behavior of a surface immersed in a solvent, depending on the application of external stimuli like temperature,8-10 pH,11 or electric field.12 Furthermore, the global change in chain conformation resulting from the application of this stimulus may be used to exert a force onto the surroundings of the brush, which is a crucial step in the fabrication of macromolecular actuators.13,14 For instance, responsive brushes grafted on one side of a microcantilever can be used to reversibly bend the cantilever after application of the proper stimulus.11,12,15,16 Alternatively, temperaturegated opening and closing of nanopores by grafted responsive brushes,17 and the use of brushes to move nanoparticles adsorbed on their surface (albeit in a so far nondirectional way),18 have been demonstrated as potential applications of responsive polymer brushes in soft nanotechnology. Nanopatterned polymer brushes have been fabricated,19-22 and the nanopatterning of stimuli-responsive brushes has also been reported recently.20,23-27 The latter holds promises for * To whom correspondence should be addressed. E-mail: (A.M.J.) [email protected]; (W.T.S.H.) [email protected]. † Universite ´ catholique de Louvain. ‡ University of Cambridge. § On sabbatical leave at the University of Cambridge. ⊥ Permanent address: PBM-UMR 6522, CNRS-Universite ´ de Rouen, France. 10.1021/nl802152q CCC: $40.75 Published on Web 10/04/2008

 2008 American Chemical Society

the development of soft nanoactuators, capable of converting chemical, thermal, or electrical energy into mechanical energy at a very local scale. However, for such applications to become effective, the responsiveness of a small ensemble of chains grafted in a nanopattern must be understood. Nanopatterned brushes are supposed to remain responsive when immersed in the proper solvent and exposed to the appropriate stimulus. However, the nanopatterning process results in a considerable perturbation of the average chain conformation, because the chains located at the periphery of the nanodroplet relax laterally, thereby increasing entropy and decreasing their average degree of stretching.28,29 This should have important consequences on the responsive properties of the nanopatterned brushes, although no experimental information on this issue has been published, most probably due to the difficulty to measure the response of spatially confined brushes under the application of a stimulus in a solvent. Here, we report on the nanopatterning of a brush of a thermo-responsive polymer which presents a reversible collapse temperature in water, poly(2-(2-methoxyethoxy)ethyl methacrylate)30 (PMEO2MA, Scheme 1). The responsive properties of the nanopatterned polymer brushes are studied in situ using AFM in water at variable temperature. Below 32.3 °C, the PMEO2MA chains in a brush immersed in water adopt a highly swollen configuration; above 32.3 °C, the chains collapse abruptly to a much less water-swollen

Figure 1. AFM topographic images of dry nanobrushes of PMEO2MA grown from a single prepatterned wafer (the thickness of the laterally infinite brush grown in identical conditions is 49 nm). An oligo(ethylene oxide) silane was grafted as background before performing the synthesis of the brush. R0 is the radius of the dots of ATRP initiator from which the brushes were grown. All images have identical grayscales and lateral sizes, except the image shown as an inset in the top left image for which the grayscale and lateral size are five times smaller.

state, resulting in a significant variation of brush thickness and properties.31 This specific polymer is especially interesting for technological applications because its collapse transition is close to the physiological temperature of humans and can be easily adjusted by copolymerization with a small amount of oligoethyleneglycol methacrylate.32,33 In addition, PMEO2MA is noncytotoxic,34 which might not be the case for the comparable but more frequently studied thermoresponsive poly(N-isopropyl acrylamide) (PNIPAM). 35 The resulting biocompatible and nontoxic nanopatterned brushes should therefore also be of interest for the development of “smart” surfaces for cell adhesion studies. In this letter, we investigate experimentally the collapse transition of nanobrushes in water as a function of their lateral confinement and show that the confinement results in an amplification of the brush response, together with a considerable broadening of the collapse transition. These results are interpreted within the framework of the internal structure of nanobrushes derived from a general model of nanopatterned dry brushes,29 and provide for the first time a direct view of the intriguing effect of confinement on the collective conformational transition of a small ensemble of chains. The growth of dense brushes of PMEO2MA on silicon by atom-transfer radical polymerization (ATRP) was reported previously,31 as was the nanopatterning of silicon wafers by combining electron-beam lithography with gas phase silanation.36 Here, we combined these methods to grow PMEO2MA brushes from silicon wafers prepatterned with 3820

an ATRP initiator (full experimental details are given in the Supporting Information). Briefly, arrays of circular holes of varying diameter (from 35 nm to a few micrometers) were realized by electron beam or nanoimprint lithography in 100 nm thick PMMA films spin-coated on silicon wafers (Scheme 1). After a short exposure to an oxygen plasma to clean the bottom of the holes, the wafers were reacted with an ATRP silane initiator in a gas phase silanation reactor, and the PMMA mask was quantitatively removed by an acetone Soxhlet. When required, an oligo(ethylene oxide) silane was deposited from solution to form a continuous background surrounding the ATRP initiator.37 The PMEO2MA brush was then grown from the ATRP initiator-covered regions using standard aqueous ATRP conditions (Supporting Information). A 1 µm wide line of ATRP initiator was also drawn on each wafer in order to grow simultaneously for reference a quasilaterally infinite brush whose dry thickness h0 was determined by AFM. It should be noted that earlier work on (nano)patterned brushes found deviations from “infinite” brushes even for pattern sizes in the micron range.21,26 We did not see substantial differences for pattern sizes over 1 µm, probably because we are working with much thinner brushes. The wafers were then rinsed in methanol and water, and stored in the dark in nitrogen before further use. The nanopatterned dry brushes were first observed by tapping mode atomic force microscopy (AFM) in air. Figure 1 presents a series of AFM images (topography) of PMEO2MA nanobrushes of varying lateral sizes grown under Nano Lett., Vol. 8, No. 11, 2008

Scheme 1. Schematic Description of the Fabrication of Nanopatterned PMEO2MA Brushes

strictly identical conditions from a single prepatterned wafer for which the dry thickness of the corresponding laterally infinite brush, h0, is 48.6 ( 0.5 nm as determined by AFM. As reported previously,22,26,29 the maximum height of the brush nanodots decreases with decreasing diameter of the grafting region, 2R0 (i.e., the diameter of the region onto which the ATRP initiator is grafted). For instance, whereas the maximum height of the brush (hmax,dry) is ∼48-50 nm for grafting diameters ranging from 210 nm to 5 µm, it falls to 29, 23, and 10 ( 0.5 nm for diameters of 110, 60, and 35 nm, respectively. The reason for this behavior can be obtained by using a recent theoretical model of the shape of dry nanobrushes.29 This model takes into account both the entropy penalty associated with chain stretching and the energetic advantage to spread chains on a wetted substrate, which are the two factors controlling the shape of nanobrushes, hence their maximal height. For the present system, the chains at the periphery of the droplet try to spread on the surface because the di(ethylene oxide) segments of the polymer have an affinity for the oligo(ethylene oxide) silanized background or for the H-bonding hydroxylated native silicon oxide of bare wafers. The relaxation of the chains at the rim of the features creates extra room for neighboring chains farther inside the nanobrush, which relax by tilting away from the normal to the surface and by decreasing the distance between their ends, which increases entropy. The result is a decrease of the maximum height of the nanobrush, which can be observed for grafting diameters Nano Lett., Vol. 8, No. 11, 2008

Figure 2. Swelling in water and thermo-responsiveness of nanobrushes of PMEO2MA. (a-c) Composite AFM topographic images of nanobrushes in the dry state and in water at two temperatures. Sample details (2R0, h0, substrate background): (a)115 nm, 39 nm, bare silicon; (b) 60 nm, 48.6 nm, oligo(ethylene oxide) silane); (c) 35 nm, 48.6 nm, oligo(ethylene oxide) silane. (d) AFM-determined variation with temperature of the vertical swelling of the nanobrushes in water. Closed circles, laterally infinite brush; open circles, sample displayed in panel a; closed squares, sample displayed in panel b; open squares, sample displayed in panel c. The lines are guides for the eye. The inset is a composite three-dimensional representation of the experimental average nanobrush shape for 2R0 ) 35 nm.

below ∼10h0,29 and which is more pronounced for smaller diameters. The PMEO2MA nanobrushes were then imaged in water in the region of the collapse transition temperature of the laterally infinite brush, which is 32.3 °C.31 Because of their softness, water-swollen nanobrushes are difficult to measure by AFM;38 for this reason, we selected very soft cantilevers usually used for contact mode AFM, and used them for tapping-mode imaging with an amplitude setpoint almost equal to the free amplitude value, corresponding to very soft tapping conditions. All other tested conditions did not provide meaningful data. Typical images obtained for brushes of ∼40-50 nm dry reference thickness, grafted on initiator dots of 35, 60, and 115 nm diameter, are shown in the dry state and for two temperatures in the wet state in Figure 2a-c. The average three-dimensional shape of the nanobrushes is also displayed for different conditions in the inset of Figure 2d (grafting diameter of 35 nm). These AFM studies confirm that the brushes swell considerably in water compared to the dry state; in addition, a large variation of height can be detected over a small temperature range in the wet state. The degree of vertical swelling of the nanobrushes in the wet state, which is defined as the ratio between the maximal heights of the nanobrushes in the wet and dry states (hmax,wet/ hmax,dry), is plotted versus temperature in Figure 2d and compared to the vertical swelling of a laterally infinite brush.31 3821

Figure 3. Correlation between the characteristics of chains in dry nanobrushes and their swelling and thermo-responsiveness in water. (a) Chain end-to-end vectors in the dry nanobrushes of Figure 2. The free chain ends are indicated by dots. (b) Dependence of the chain end-to-end distance l on the location of the grafting point on the surface. The end-to-end distances are normalized by h0, and rg is the distance of the grafting point to the nanobrush center. (c) Histograms of the normalized end-to-end distance of the chains in the nanobrushes. (d) Experimental vertical swelling of the nanobrushes determined by AFM at two temperatures versus the normalized average chain endto-end distance in the dry state obtained from the model. The lines are guides for the eye.

A first observation is that the smaller the diameter of the nanobrushes, the larger their vertical swelling compared to the laterally infinite brush. For instance, the degree of swelling at 22 °C is 1.6 for the laterally infinite brush, but increases to 1.7, 2.1, and 2.6 for nanobrushes of 115, 60, and 35 nm grafting diameter, respectively. In addition, whereas a neat collapse transition is observed for the laterally infinite brush and the brush of limited confinement (2R0/h0 ∼ 2.9), the transition is considerably broadened for more confined brushes, resulting in the loss of a well-defined collapse temperature. This does not mean that the collapse transition is lost; on the contrary, the relative collapse is larger for the more confined brushes, but it now occurs over a wider range of temperatures. It should also be mentioned that the collapse transition is reversible, as could be checked by measuring the samples again after cooling to room temperature. The broadening and amplification of the collapse transition with confinement can be understood with reference to the structure of the nanobrushes in the dry state (Figure 3). Figure 3a shows the shape of the dry nanobrushes of Figure 2 and the approximate end-to-end vectors of their chains, as obtained from a theoretical model fully described elsewhere.29 The stretching of the peripheral chains and the relaxation of the central chains for the more confined brushes 3822

are readily apparent on this picture. This is also seen in Figure 3b, where the chain end-to-end distances (l) normalized to their end-to-end distances in the laterally infinite brush (h0), are plotted versus the distance of their grafting point from the center of the brush, rg. The end-to-end distances of chains lying at the edges of the brushes are longer due to the spreading of chains on the substrate because of partial wetting, whereas the chains deeper in the brush are less extended. The distributions of normalized end-to-end distances in the dry state are shown in Figure 3c, from which it can be seen that the average end-to-end distance decreases for nanobrushes in smaller patterns, while the width of the distribution of end-to-end distances increases. Please note that for simplicity we assume monodisperse polymer chains for this model; small deviations are expected experimentally, where some degree of polydispersity is expected. The experimentally determined vertical swelling of the nanobrushes in water for two temperatures is plotted in Figure 3d together with the predicted normalized average end-toend distance in the dry state. The degree of swelling correlates well with the normalized average end-to-end distance in the dry state, as chains which are initially less extended are able to swell more in water, since their initial average entropy is higher. According to Flory’s classical theory of the swelling of polymer coils,39 the chains would Nano Lett., Vol. 8, No. 11, 2008

swell until their entropy decreases to the point where it compensates the decrease of energy associated with the uptake of solvent. Hence, because the chains in more confined nanobrushes start from a higher entropic state characterized by smaller chain stretching and shorter endto-end distances, they are able to swell to a much larger extent than chains in the laterally infinite brush, as observed experimentally. Likewise, the increased width of the distribution of end-to-end distances also correlates with the observed broadening of the collapse transition, because the collapse transition temperature of thermo-responsive polymer brushes depends on the degree of chain stretching in the brush, as was reported recently for PNIPAM.40 Another way to explain these results is by reference to the theoretical work of Zhulina and co-workers on polymer brushes.5 They reported on the collapse transition of brushes grafted on surfaces of varying geometry from planar to spherical with the radius of the spheres being small compared to the brush thickness. For a planar brush, the height of the brush h0 scales as L both below and above the collapse transition, where L is the extended length of the chains. By contrast, for a brush grafted on a sphere of small radius the scaling laws are h0 ∼ L1/3 in the collapsed state and ∼ L3/5 in the swollen state, which shows that the degree of swelling changes much more at the collapse transition for a brush grafted on a nanosphere than on a planar surface. For our chains grafted in nanodots of very small diameter, all chains of the nanostructure fan out (Figure 2a) and adopt a conformation which is reminiscent of the one of chains grafted on a nanosphere. In contrast, when the grafting diameter increases the central chains feel an environment almost identical to the one of a laterally infinite planar brush. The observed increased degree of swelling of the more confined nanobrushes is thus in good agreement with these scaling arguments. In conclusion, nanobrushes of a responsive biocompatible polymer with pattern diameters down to 35 nm have been prepared and the thermal response in water of the brush nanodroplets was assessed. The reduction of size down to droplets containing only a very small number of chains does not suppress the collective transition of the chains (a brush grafted on a 35 nm diameter initiator dot contains about 300 chains only, considering a chain footprint of 3 nm2).29 However, more confined nanobrushes, for which the fanningout affects all chains in the nanostructure, display a much larger swelling in water and a considerably broadened collapse transition to the point that no precise collapse transition temperature can be defined anymore. The results of the modeling of the structure of the brushes in the dry state underpins our understanding of the effect of confinement on the thermo-responsiveness of the nanobrushes, since the average end-to-end distance of the chains in the dry state correlates with the degree of swelling of the nanobrushes in water, and since the distribution of dry end-to-end distances correlates with the width of the collapse transition. These results indicate that the response of polymer brushes can be modulated by controlling their lateral size in the nanometer range, which introduces a completely new parameter in the Nano Lett., Vol. 8, No. 11, 2008

design of responsive surfaces of complex tunable topography. They also show that the collective motion of chains can be harnessed at the level of a few hundred chains only, which should help the fabrication of nanoactuators at an unprecedentedly small scale. Acknowledgment. Financial support was provided by the French Community of Belgium (ARC 06/11-339), by the Belgian Federal Science Policy (IAP/V/27), and by the Leverhulme Trust (U.K.). A.M..J and K.G. thank the University of Louvain and CNRS, respectively, for having supported this work through a sabbatical leave. Supporting Information Available: Materials, description of the patterning and synthesis of brushes of PMEO2MA, AFM measurements, and image analysis. This material is available free of charge via the Internet at http://pubs.acs.org. References (1) Alexander, S. J. Phys. (Paris) 1977, 38, 983. (2) de Gennes, P.-G. Macromolecules 1980, 13, 1069. (3) Jones, R. A. L.; Richards, R. W. Polymers at Surfaces and Interfaces; Cambridge University Press: Cambridge, 1999. (4) Polymer Brushes; Advincula, R. C., Brittain, W. J., Caster, K. C., Ru¨he, J., Eds.; Wiley-VCH: Weinheim, Germany, 2004. (5) Zhulina, E. B.; Borisov, O. V.; Pryamitsyn, V. A.; Birshtein, T. M. Macromolecules 1991, 24, 140. (6) Luzinov, I.; Minko, S.; Tsukruk, V. V. Prog. Polym. Sci. 2004, 29, 635. (7) Minko, S. Polym. ReV. 2006, 46, 397. (8) Ista, L. K.; Mendez, S.; Perez-Luna, V. H.; Lopez, G. P. Langmuir 2001, 17, 2552. (9) Jones, D. M.; Smith, J. R.; Huck, W. T. S.; Alexander, C. AdV. Mater. 2002, 14, 1130. (10) Plunkett, K. N.; Zhu, X.; Moore, J. S.; Leckband, D. E. Langmuir 2006, 39, 3420. (11) Zhou, F.; Shu, W.; Welland, M. E.; Huck, W. T. S. J. Am. Chem. Soc. 2006, 128, 5326. (12) Zhou, F.; Biesheuvel, P. M.; Choi, E.-Y.; Shu, W.; Petes, R.; Steiner, U.; Huck, W. T. S. Nano Lett. 2008, 8, 725. (13) Prokhorova, S. A.; Kopyshev, A.; Ramakrishnan, A.; Zhang, H.; Ru¨he, J. Nanotechnology 2003, 14, 1098. (14) Zhou, F.; Huck, W. T. S. Phys. Chem. Chem. Phys. 2006, 8, 3815. (15) Bumbu, G. G.; Kircher, G.; Wolkenhauer, M.; Berger, R.; Gutmann, J. S. Macromol. Chem. Phys. 2004, 205, 1713. (16) Abu-Lail, N. I.; Kaholek, M.; LaMattina, B.; Clark, R. L.; Zauscher, S. Sens. Actuators, B 2006, 114, 371. (17) Alem, H.; Duwez, A.-S.; Lussis, P.; Lipnik, P.; Jonas, A. M.; Demoustier-Champagne, S. J. Membr. Sci. 2008, 308, 75. (18) Santer, S.; Kopyshev, A.; Donges, J.; Yang, H.-K.; Ru¨he, J. AdV. Mater. 2006, 18, 2359. (19) Tsujii, Y.; Ejaz, M.; Yamamoto, S.; Fukuda, T.; Shigeto, K.; Mibu, K.; Shinjo, T. Polym. Commun. 2002, 43, 3837. (20) Ahn, S. J.; Kaholek, M.; Lee, W.-K.; LaMattina, B.; LaBean, T. H.; Zauscher, S. AdV. Mater. 2004, 16, 2141. (21) Schmelmer, U.; Paul, A.; Kueller, A.; Steenackers, M.; Ulman, A.; Grunze, M.; Goelzhaeuser, A.; Jordan, R. Small 2007, 3, 459. (22) Steenackers, M.; Kueller, A.; Ballav, N.; Zharnikov, M.; Grunze, M.; Jordan, R. Small 2007, 3, 1764. (23) Kaholek, M.; Lee, W.-K.; LaMattina, B.; Caster, K. C.; Zauscher, S. Nano Lett. 2004, 4, 373. (24) Kaholek, M.; Lee, W.-K.; Ahn, S.-J.; Ma, H.; Caster, K. C.; LaMattina, B.; Zauscher, S. Chem. Mater. 2004, 16, 3688. (25) Kaholek, M.; Lee, W.-K.; Feng, J.; LaMattina, B.; Dyer, D. J.; Zauscher, S. Chem. Mater. 2006, 18, 3660. (26) Lee, W.-K.; Patra, M.; Linse, P.; Zauscher, S. Small 2007, 3, 63. (27) He, Q.; Kueller, A.; Grunze, M.; Li, J. Langmuir 2007, 23, 3981. (28) Patra, M.; Linse, P. Nano Lett. 2006, 6, 133. (29) Jonas, A. M.; Hu, Z.; Glinel, K.; Huck, W. T. S. Macromolecules, in press (DOI: 10.1021/ma801584k). (30) Han, S.; Hagiwara, M.; Ishizone, T. Macromolecules 2003, 36, 8312. (31) Jonas, A. M.; Glinel, K.; Oren, R.; Nysten, B.; Huck, W. T. S. Macromolecules 2007, 40, 4403. 3823

(32) Lutz, J.-F.; Hoth, A. Macromolecules 2006, 39, 893. (33) Lutz, J.-F.; Akdemir, O.; Hoth, A. J. Am. Chem. Soc. 2006, 128, 13046. (34) Lutz, J.-F.; Andrieu, J.; Uzgu¨n, S.; Rudolph, C.; Agarwal, S. Macromolecules 2007, 40, 8540. (35) Vihola, H.; Laukkanen, A.; Valtola, L.; Tehnhu, H.; Hirnoven, J. Biomaterials 2005, 26, 3055. (36) Pallandre, A.; Glinel, K.; Jonas, A. M.; Nysten, B. Nano Lett. 2004, 4, 365.

3824

(37) Buron, C. C.; Callegari, V.; Nysten, B.; Jonas, A. M. Langmuir 2007, 23, 9667. (38) Patra, M.; Linse, P. Macromolecules 2006, 39, 4540. (39) Teraoka, I. Polymer Solutions; Wiley-Interscience: New York, 2002. (40) Yim, H.; Kent, M. S.; Mendez, S.; Lopez, G. P.; Satija, S.; Seo, Y. Macromolecules 2006, 39, 3420.

NL802152Q

Nano Lett., Vol. 8, No. 11, 2008