In Situ Infrared Ellipsometric Study of Stimuli-Responsive Mixed

Sep 19, 2007 - So far, other in situ investigations with infrared spectroscopy are not known for ...... Synytska, Alla; Rogach, Andrey L.; Stamm, Manf...
0 downloads 0 Views 471KB Size
Download PDF
Anal. Chem. 2007, 79, 7676-7682

In Situ Infrared Ellipsometric Study of Stimuli-Responsive Mixed Polyelectrolyte Brushes Yulia Mikhaylova,† Leonid Ionov,‡ Jo 1 rg Rappich,§ Michael Gensch,† Norbert Esser,† Sergiy Minko,| ⊥ Klaus-Jochen Eichhorn, Manfred Stamm,⊥ and Karsten Hinrichs*,†

ISASsInstitute for Analytical Sciences, Department Berlin, Albert-Einstein-Strasse 9, 12489 Berlin, Germany, Max-Planck-Institute of Molecular Cell Biology and Genetics, Pfotenhauerstrasse 108, 01307 Dresden, Germany, Hahn-Meitner-Institut Berlin GmbH, Abt. Silizium-Photovoltaik, Kekule´ strasse 5, 12489 Berlin, Germany, Department of Chemistry and Biomolecular Science, Clarkson University, Potsdam, New York 13699, and Leibniz Institute of Polymer Research Dresden, Hohe Strasse 6, 01069 Dresden, Germany

The switching behavior of stimuli-responsive mixed polymer brushes (d ) 11 nm) was monitored for the first time in liquid phase in situ by infrared spectroscopic ellipsometry (IRSE). IRSE is presented as a new, sensitive in situ tool for online analysis of chemical changes in a thin complex film at the solid/liquid interface. Responsive behavior (protonation and deprotonation reaction) of the poly(acrylic acid)/poly(2-vinylpyridine) (PAA-mix-P2VP) brush was probed in aqueous solutions with pH ranging from pH 2 to pH 10. Structural and chemical changes in a thin polymer brush layer were identified from the analysis of infrared ellipsometric tanΨ spectra during the variation of pH. Systematic change in pH confirmed the reversible switching behavior of the PAA-mix-P2VP brush between three different states: swollen P2VP and compact PAA chains at pH 2, a compact “P2VP...PAA” complex at pH 6.5, and swollen PAA and compact P2VP chains at pH 10. Research and development of responsive materials such as functional polymer films and surfaces1-8 are strongly motivated by their high technological potential for adhesives,9 coatings,10 * To whom correspondence should be addressed. E-mail: [email protected]. † ISASsInstitute for Analytical Sciences. ‡ Max-Planck-Institute of Molecular Cell Biology and Genetics. § Hahn-Meitner-Institut Berlin GmbH. | Clarkson University. ⊥ Leibniz Institute of Polymer Research Dresden. (1) Luzinov, I.; Minko, S.; Tsukruk, V. V. Prog. Polym. Sci. 2004, 29, 635-698. (2) Nath, N.; Chilkoti, A. Adv. Mater. 2002, 14, 1243-1246. (3) Zhang, H.; Ito, Y. In Handbook of Polyelectrolytes and Their Applications; Tripathy, S. K., Kumar, J., Nalwa, S., Eds.; American Scientific Publishers: Stevenson Ranch, CA, 2002; p 183. (4) Senaratne, W.; Andruzzi, L.; Ober, C. K. Biomacromolecules 2005, 6, 24272448. (5) Gil, E. S.; Hudson, S. A. Prog. Polym. Sci. 2004, 29, 1173-1222. (6) Brittain, W. J.; Boyes, S. G.; Granville, A. M.; Baum, M.; Mirous, B. K.; Akgun, B.; Zhao, B.; Blickle, C.; Foster, M. D. Adv. Polym. Sci. 2006, 198, 125-147. (7) Russell, T. P. Science 2002, 297, 964-967. (8) Minko, S., Ed. Responsive polymer materials: design and applications; Blackwell Publishing: Ames, IA, 2006. (9) Koberstein, J. T.; Duch, D. E.; Hu, W.; Lenk, T. J.; Bhatia, R.; Brown, H. R.; Lingelser, J. P.; Gallot, Y. J. Adhes. 1998, 66, 229-249. (10) Uhlmann, P.; Ionov, L.; Houbenov, N.; Nitschke, M.; Grundke, K.; Motornov, M.; Minko, S.; Stamm, M. Prog. Org. Coat. 2006, 55, 168-174.

7676 Analytical Chemistry, Vol. 79, No. 20, October 15, 2007

separation,11 microfluidic devices,12 surface transport,13,14 sensors,15-19 responsive colloids,20,21 and biomedical applications.22,23 Among all variety of smart materials, surface-immobilized polymer layerss mixed polymer brushess deserve a particular interest due to their unique switching properties.24 The mixed binary polymer brushes consist of two kinds of immiscible polymers that are grafted with sufficiently high density to the same substrate. Such mixed polymer brushes demonstrate a strong response to changes of environmental conditions such as pH, solvent, and temperature due to an interplay between a lateral and a vertical reorganization of polymer chains inside the polymer layer.25 Thereby, in the solvent selective to one of the polymers, the topmost brush layer is enriched with this polymer while the second polymer is segregated to the grafting surface and forms spherical domains. In nonselective solvents, both polymers are present equally in the topmost brush layer being laterally segregated. Due to such a switching behavior, mixed polymer brushes find a broad application for control of wetting, adhesion, and adsorption. A very interesting case of mixed polymer brushes is represented by mixed polyelectrolyte brushes consisting of two oppositely charged (11) Kanazawa, H. Anal. Bioanal. Chem. 2004, 378, 46-48. (12) Ionov, L.; Houbenov, N.; Sidorenko, A.; Stamm, M.; Minko, S. Adv. Funct. Mater. 2006, 16, 1153-1160. (13) Ichimura, K.; Oh, S. K.; Nakagawa, M. Science 2000, 288, 1624-1626. (14) Ionov, L.; Stamm, M.; Diez, S. Nano Lett. 2006, 6, 1982-1987. (15) Ionov, L.; Sapra, S.; Synytska, A.; Rogach, A. L.; Stamm, M.; Diez, S. Adv. Mater. 2006, 18, 1453-1457. (16) Tokareva, I.; Minko, S.; Fendler, J. H.; Hutter, E. J. Am. Chem. Soc. 2004, 126, 15950-15951. (17) Westenhoff, S.; Kotov, N. A. J. Am. Chem. Soc. 2002, 124, 2448-2449. (18) Lin, Y. H.; McConney, M. E.; LeMieux, M. C.; Peleshanko, S.; Jiang, C. Y.; Singamaneni, S.; Tsukruk, V. V. Adv. Mater. 2006, 18, 1157-1161. (19) Tokareva, I.; Tokarev, I.; Minko, S.; Hutter, E.; Fendler, J. H. Chem. Commun. 2006, 3343-3345. (20) Motornov, M.; Sheparovych, R.; Lupitskyy, R.; MacWilliams, E.; Hoy, O.; Luzinov, I.; Minko, S. Adv. Funct. Mater. In press. (21) Fujii, S.; Armes, S. P.; Binks, B. P.; Murakami, R. Langmuir 2006, 22, 68186825. (22) Hoffman, A. S.; Stayton, P. S.; Press, O.; Murthy, N.; Lackey, C. A.; Cheung, C.; Black, F.; Campbell, J.; Fausto, N.; Kyriakides, T. R.; Bornstein, P. Polym. Adv. Technol. 2002, 13, 992-999. (23) Lata, S.; Piehler J. Anal. Chem. 2005, 77, 1096-1105. (24) Minko, S. Polym. Rev. 2006, 46, 397-420. (25) Minko, S.; Muller, M.; Usov, D.; Scholl, A.; Froeck, C.; Stamm, M. Phys. Rev. Lett. 2002, 88, Art. No. 035502. 10.1021/ac070853a CCC: $37.00

© 2007 American Chemical Society Published on Web 09/19/2007

polyelectrolytes.26,27 Due to particular sensitivity to pH and ion concentration in aqueous environment, mixed polyelectrolyte brushes are especially promising to control interactions in aqueous environment for many biotechnological applications. Analytical methods for in situ studies (in the changing environment of the brush) of the structural changes in responsive thin polymer films are of key importance for effective design of smart surfaces and development of their technological applications. It was shown during the past decade that spectroscopic ellipsometry in the mid-infrared range is well suited to study the thin polymer films providing information about thickness, molecular structure, and chemical compositions.28 Recently, the high potential of a combined visible monochromatic ellipsometry and infrared spectroscopic ellipsometry (IRSE) study was shown for characterization of polymer brush films in the dry state.29 Often, optical spectroscopic methods such as Raman spectroscopy and infrared spectroscopy have the advantage of being noninvasive and noncontact methods that offer manifold spectral analysis. In Raman spectroscopy and infrared spectroscopy (e.g., IRSE), the interaction mechanisms of radiation with the sample are completely different. A quantitative evaluation of Raman spectra is typically more complex than that of IRSE. Sensitive in situ measurements of Raman spectra for ultrathin films are approached either by resonant Raman scattering or surfaceenhanced Raman scattering.30 However, in both methods, the vibrational signatures can give direct access to the microscopic and molecular structure of a thin film. For in situ spectroscopy in the mid-IR spectral range (wavelengths from 5 to 25 µm), it is important to consider that the penetration depth of radiation in water is in the range of several tens of micrometers. This behavior leads to strong physical restrictions for the applications of the in situ measurements through an aqueous solution. Therefore, for in situ measurements at the solid/liquid interface in the infrared spectral range, often a strategy with transparent substrates is applied. Most of these investigations use attenuated total reflection (ATR) geometries31-33 with multiple internal reflections in order to increase the signal-to-noise ratio. Several studies were carried out to characterize the adsorption of different molecules at air/ liquid, liquid/liquid, and solid/liquid interfaces using IR techniques; e.g., adsorption of proteins at solid/aqueous solution interface.32,34 Complementary results from other experimental methods like vacuum ultraviolet-visible ellipsometry,29,35 X-ray (26) Houbenov, N.; Minko, S.; Stamm, M. Macromolecules 2003, 36, 58975901. (27) Ionov, L.; Houbenov, N.; Sidorenko, A.; Stamm, M.; Luzinov, I.; Minko, S. Langmuir 2004, 20, 9916-9919. (28) Hinrichs, K.; Gensch, M.; Esser, N. Appl. Spectrosc. 2005, 59, 272a-282a. (29) Ionov, L.; Sidorenko, A.; Eichhorn, K.-J.; Stamm, M.; Minko, S.; Hinrichs, K. Langmuir 2005, 21, 8711-8716. (30) Weaver, M. J. J. Raman Spectrosc. 2002, 33, 309-317. (31) McClellan, S. J.; Franses, E. I. Colloid Surf., A 2005, 260, 265-275. (32) Muller, M.; Werner, C.; Grundke, K.; Eichhorn, K.-J.; Jacobasch, H. J. Mikrochim. Acta 1997, 14, 671-674. (33) Muller, M. In Handbook on polyelectrolytes and their application; Tripathy, S. K., Kumar, J., Nalwa, H. S., Eds.; American Scientific Publisher: Stevenson Ranch, CA, 2002; Vol. 1, p 293. (34) Kallay, N.; Hlady, V.; Jednacak-Biscan, J.; Milonjic, S. In Investigations of Surfaces and Interfaces; Rossiter, B. W., Baetzold, R. C., Eds.; John Wiley & Sons: New York, 1993; p 73. (35) Hinrichs, K.; Silaghi, S. D.; Cobet, C.; Esser, N.; Zahn, D. R. T. Phys. Status Solidi B 2005, 242, 2681-2687.

photoelectron spectroscopy,36 scanning microscopy (e.g., atomic force microscopy23 or scanning near-field infrared microscopy37) can help to reveal the microscopic structure and the density of the film and interface. In this paper, we introduce IR spectroscopic ellipsometry as a highly sensitive technique for in situ investigations of thin polymer films through an infrared-transparent substrate. In contrast to IR microscopy, the high sensitivity of IRSE allows the collection of important spectral information in single reflection geometry even for monolayers36,38 or interfaces.39 We report that in situ IRSE in single reflection geometry can be successfully used for online studies of the structural changes of a mixed polyelectrolyte brush upon change of environmental conditions. So far, other in situ investigations with infrared spectroscopy are not known for this kind of ultrathin binary polymer brushes. EXPERIMENTAL SECTION Film Preparation. A thin homogeneous film (d ) 11 ( 0.3 nm in a dry state) of poly(acrylic acid)/poly(2-vinylpyridine) (PAA-mix-P2VP) mixed brush with 50:50 component ratio was prepared on a silicon wedge according to the published protocol (see ref 26) using an anchoring poly(glycidyl methacrylate) layer (1.5 nm thick) for grafting the polymers: carboxylic acid endterminated poly(tert-butyl acrylate) (PtBA) Mn ) 42 000 g/mol and carboxylic acid end-terminated poly(2-vinylpyridine) (P2VP) Mn ) 39 200 g/mol (both polymers synthesized by anionic polymerization were purchased from Polymer Source, Dorval, Canada). PtBA afterward was hydrolyzed in organic solvent to produce PAA chains.26 The sample was rinsed in organic solvents methylethylketone (MEK and 1,3-dioxane) and dried. The thickness (d ) 11 ( 0.3 nm) of the polymer brush was determined at λ ) 632.8 nm by a single-wavelength ellipsometer (SE402, Sentech Instruments GmbH, Berlin, Germany). IR Transmission Measurements. A Bruker IFS 66v/S spectrometer with DTGS detector was used to measure the dry bulk polymers. All transmission spectra were recorded in the spectral range of 4000-400 cm-1 with a resolution of 2 cm-1. For comparison, P2VP and PAA were prepared as several micrometerthick films by casting from ethanol and aqueous solutions on KBr and KRS5 disks, respectively. A preparation of the P2VP/PAA mixture as casted film was not possible since these two polymers form an insoluble complex in water. Therefore, the P2VP/PAA mixture (1:1) was prepared as a KBr pellet with a ratio of 2 mg of the polymer mixture per 400 mg of KBr. IR In Situ Ellipsometry. General details about the infrared ellipsometric setups can be found in refs 28 and 40. The used setup was purged with dry air. The in situ IR spectroscopic ellipsometry measurements on the brush layers were performed through an IR transparent trapezoidal Si wedge (1.5° with a (111) surface in contact with the aqueous solution). A schematic (36) Gensch, M.; Roodenko, K.; Hinrichs, K.; Hunger, R.; Guell, A. G.; Merson, A.; Schade, U.; Shapira, Y.; Dittrich, T.; Rappich, J.; Esser, N. J. Vac. Sci. Technol. B 2005, 23, 1838-1842. (37) Raschke, M. B.; Molina, L.; Elsaesser, T.; Kim, D. H.; Knoll, W.; Hinrichs, K. Chemphyschem 2005, 6, 2197-2203. (38) Meuse, C. W. Langmuir 2000, 16, 9483-9487. (39) Hinrichs, K.; Gensch, M.; Esser, N.; Ro¨seler, A. J. Phys. Condens. Matter 2004, 16, S4335-S4343. (40) Ro ¨seler, A.; Korte, E. H. Infrared Spectroscopic Ellipsometry. In Handbook of Vibrational Spectroscopy; Chalmers, J. M., Griffiths, P. R., Eds.; Wiley: Chichester, 2001; pp 1065-1090.

Analytical Chemistry, Vol. 79, No. 20, October 15, 2007

7677

Figure 2. Calculated (formula 1) normalized in situ tan Ψ spectra for a 10-nm P2VP film with different molecular orientations in contact with water. For normalization, the spectrum of 10-nm P2VP was divided by the spectrum of the clean surface. The optical constants of P2VP were taken from a dry spin-coated film, and the spectra were calculated for the same optical conditions as used for the experiments (see Figure 1).

Figure 1. (A) Principle of IR ellipsometry. (B) The in situ cell was placed in the ellipsometer, which was adjusted at an incidence angle of 50°. The cell was adjusted on the reflex from the interface of the mixed brush (in contact with solution) grafted to the wedged silicon (front window of the cell). In this configuration (adjustment on the reflex of the back side, which is marked as off-back) the incidence angle was 59.2° and the angle for the reflected beam was 40.8° with respect to the normal of the front surface of the silicon wedge. The opening angle of the incident beam is (3.5 °.

illustration of the setup is shown in Figure 1. A wedge was used in order (i) to suppress interferences, which would arise from multiple reflections in plane silicon substrate, and (ii) to enable separate measurements of the reflected radiation from the front (silicon) and back side (solid/liquid interface), respectively. The angle of incidence is shown in Figure 1b. The horizontal wedge (52 × 20 mm) was adjusted to the maximum signal for the reflection from the back side. The relevant angles for the incident and for the reflected beams from the solid/liquid interface are given with respect to the surface normal in Figure 1. The reflex from front and back sides is well separated since the difference between these two angles is larger than double the opening angle of (3.5°, which was used in our optical setup. The spot size on the surface of the silicon wedge is ∼50 mm2 at an incidence angle of 60° (Jacquinot aperture of 2.4 mm). The reflected radiation is described by the complex ratio of the two mutually orthogonal polarized components, rs and rp: F ) rp/rs ) tan Ψ ei∆.41 Here tan Ψ ) |rp|/|rs| is the absolute amplitude ratio and ∆ is the phase shift between the p- and s-polarized components of reflected waves. The polarized reflectances Rp and Rs are given by Rp ) |rp2| and Rs ) |rs2|. The transport of radiation in the silicon wedge is described by formulas 1 and 2. The incidence angles on the (41) Graf, R. T.; Koenig, J. L. Anal. Chem. 1986, 58, 64-68.

7678

Analytical Chemistry, Vol. 79, No. 20, October 15, 2007

different interfaces as used in the presented experiments are given in Figure 1b. Feff ) F′air/silicon-59.2°Fsilicon/siliconoxide/water-13.8°F′silicon/air -12.3° (1)

with F)

|rp| i∆r e |rs|

and

F′ )

|tp| i∆t e |ts|

(2)

where rp, rs and tp, ts are the p- and s-polarized reflection and transmission coefficients, respectively. Principally formula 1 allows for a quantitative interpretation of the measured in situ IR ellipsometric spectra. For example, Figure 2 shows calculated in situ tan Ψ spectra, which represent a 10-nm P2VP film adsorbed on the silicon wedge in water. The P2VP film spectra (normalized to the spectrum of the clean surface) are calculated under the same optical conditions as used for the experiments. The presented calculations in Figure 2 rely on the optical constants of the dry P2VP film. Therefore, they do not take into account chemical changes and band broadening due to the contact of the brush with aqueous solution. However, the band amplitudes are in the same order of magnitude as in the measured spectra. Some qualitative criteria for interpretation of band shapes in measured ellipsometric spectra can be drawn from these model calculations. Under the assumption of an isotropic distribution of transition dipole moments for a certain molecular vibration, an upward pointing band can be interpreted as an increase in the number of correlated molecular groups. In contrast, e.g., for an anisotropic film structure with preferential in-plane transition dipole moments, the peaks would point downward for an increasing amount of correlated molecular vibrations. An interesting feature of the referenced spectra can be seen around 1650 cm-1, where a broad band due to the water deformation

Figure 3. (a) IR spectra of thick-film samples: PAA (I), P2VP (IV), and the PAA:P2VP ) 1:1 mixture (II) in comparison to a tan Ψ spectrum of an 11-nm-thick dry PAA-mix-P2VP (50:50) brush on the silicon wafer (III). (b) Scheme illustrating hydrogen bonds in the system: if P2VP is added to PAA, a new type of hydrogen bond between pyridine rings and carboxylic groups appears.

vibration arises due to the fact that the dielectric function of the solid/liquid interface is different after adsorption of a thin polymer film. In general, when the optical properties of the solid/liquid interface are changed, a contribution in this spectral range will be observed in in situ measurements with aqueous solutions. Thus, a band can arise even without the adsorption or incorporation of water at the interface. In particular, such a contribution due to the change of the optical constants at the solid/liquid interface can overlap with the absorption of complex water42 in the polymer film. Therefore, a reliable interpretation of this band in the later discussed in situ spectra is difficult. Tan Ψ spectra were recorded in the spectral range of 2000400 cm-1 with a spectral resolution of 4 cm-1. In situ ∆ spectra showed higher noise levels than the corresponding tan Ψ spectra. The signal-to-noise ratio for ∆ spectra could be improved by the use of a rotating retarder, which enables optimization of the photon fluxes in the single spectra within the ellipsometric measurement scheme by the choice of an optimal phase shift. Since some of the small peaks in the original spectra (dotted) are slightly above the 2σ level, the signal-to-noise level of the shown tan Ψ spectra was increased by a factor up to 10 by adjacent averaging over 8 points (corresponding to 16-cm-1 resolution. Before the PAA-mix-P2VP brush was placed into the cell, the measurements in a dry state were performed. Afterward, the PAAmix-P2VP brush was equilibrated with the reagent-grade water (pH 6.5) produced by the Milli-Q filtration system. Then, KOH(pH 10) and HCl (pH 2)-containing aqueous solutions were pumped through the cell to induce switching of the PAA-mix-P2VP brush. Measurements with shorter steps showed for the pH range from 5 to 7 that a variation of pH by 0.5 produces spectral changes that could not be distinguished from the noise of the presented spectra (2σ criterion). Thereby, the pH value was measured during cycling of a corresponding solution with a pH meter from Hannah instruments with an accuracy of 0.1. RESULTS AND DISCUSSION IR Transmission Spectroscopic and IR Spectroscopy Ellipsometry in the Dry State. Figure 3a shows the IR transmission spectra of (I) PAA, (IV) P2VP, and (II) PAA/P2VP (50/50) (42) Mikhaylova, Y.; Adam, G.; Haussler, L.; Eichhorn, K.-J.; Voit, B. J. Mol. Struct. 2006, 788, 80-88.

and the ellipsometric tan Ψ spectrum of the dry PAA-mix-P2VP mixed brush film on a silicon substrate. Characteristic bands are marked. A similar spectral signature is found in the thin film (III) and bulk mixture (II). Observed frequency shifts of the carboxyl vibration originate from the formation of hydrogen bonds between carboxylic groups of PAA and pyridine rings of P2VP as indicated in Figure 3b.43 In Situ Infrared Ellipsometry. Irreversible Change. Analysis of tan Ψ spectra collected upon multiple switching of the PAAmix-P2VP brush in aqueous solutions with different pH values has demonstrated that all observed changes are reversible changes, except for the first treatment of the mixed brush with pH 10 where some irreversible changes occurred. Here and in the following we label the switching experiments with a pH value and a digit, which corresponds to the sequence of the brush sample treatment with different aqueous solutions. For example, the label pH 6.5 (1) denotes the spectrum obtained for another (not previously used) sample of the same kind of brush, which was immersed in water pH 6.5 for the first time. The label pH 10 (2) denotes the second step where water was substituted by aqueous solution at pH 10 (adjusted with KOH), while the label pH 6.5 (3) means that in the third step the alkaline aqueous solution was substituted with water at pH 6.5, and finally, the label pH 2 (4) denotes the fourth step of treatment of the brush with acidic aqueous solution at pH 2 (adjusted with HCl). The following sequence of treatment is discussed in this work. Switching steps of sample 1: pH 6.5 (1) f pH 10 (2) f pH 6.5 (3) f pH 2 (4) f pH 6.5 (5) f pH 10 (6) f pH 2 (7) f pH 6.5 (8) f drying with He. For a second sample (sample 2), only spectra for the initial step are shown for comparison in Figure 4b and c. First, we discuss the irreversible changes observed in tan Ψ spectra of the PAA-mix-P2VP brush. The experiments began with the PAA-mix-P2VP brush equilibration in the cell with Milli-Q water at pH 6.5. For such a film, we expected because of the formation of a P2VP-PAA complex (according to the previous report26) that the thickness was approximately the same as for the dry film. We have used the spectrum of PAA-mix-P2VP brush (labeled pH 6.5 (1)) treated for the first time with water at pH 6.5 as the spectrum for normalization of tan Ψ spectra (single spectrum as shown in Figure 4a). The normalized spectra in Figure 4b have been obtained by dividing the tan Ψ spectra at different pH 10 (2) by the tan Ψ (pH 6.5 (1)) spectrum (Figure 4a). This procedure was applied because the original tanΨ spectra are dominated by the deformation band of H2O molecules that hides changes in spectra affected by the formation of solid/liquid interface. Figure 4b shows the normalized spectra pH 10 (2) (tan Ψ pH 10 (2)/ tan Ψ pH 6.5 (1)) for two different samples of the mixed brush obtained after injection of alkaline solution into the cell. As can be seen from these spectra, new absorption bands at about 1650, 1551, and 1405 cm-1 were detected. For sample 2, additionally, the corresponding difference ∆ spectrum (pH 10 (2) (∆pH 10 (2) - ∆ pH 6.5 (1))) is shown in Figure 4c. In this spectrum, the Kramers-Kronig consistent counterparts of the bands as observed in the tan Ψ spectrum are identified. For further (43) Zhou, X.; Goh, S. H.; Lee, S. Y.; Tan, K. L. Polymer 1998, 39, 3631-3640.

Analytical Chemistry, Vol. 79, No. 20, October 15, 2007

7679

Figure 5. Tan Ψ and ∆ spectra of the PAA-mix-P2VP brush: initial dry state, before contact with aqueous solution (1, black curves); after switching experiment in aqueous solution and subsequent drying with He (2, blue curves). Baselines were shifted for convenience.

Figure 4. Typical single spectrum at pH 6.5, which served for normalization of tan Ψ spectra shown below (a), normalized tan Ψ spectra of PAA-mix-P2VP brush at pH 10 for samples 1 and 2 (b), and difference ∆ spectrum of PAA-mix-P2VP brush at pH 10 for sample 2 (c). Characteristic bands are marked.

discussions, we will focus on tan Ψ spectra. The new absorption bands at 1551 and 1405 cm-1 are attributed to the asymmetric (νasCOO-) and symmetric (νssCOO-) stretching vibration of the negatively charged PAA carboxylate groups. For sample 1, a broad band around 1650 cm-1 remained unchanged upon the repeated injection of water pH 6.5 (3). However, for the measurement with a second sample (sample 2) of the same brush in Figure 4b for this band, a different amplitude was found. Taking into account that a band in this spectral range can also be seen due to the change of the optical constants of the solid/liquid interface as discussed with respect to the calculations shown in Figure 2, an unequivocal assignment is not possible. A band associated with water complex molecules42 might overlap with such a contribution, which just arises because the refractive index of the film is changing. For sample 1, the spectral change around 1650 cm-1 was irreversible as long as the brush was in contact with any aqueous solution. The final washing with water and drying of a sample resulted in vanishing of the broad band at 1650 cm-1 (Figure 5). Similar amplitudes of the material specific bands in the spectra of the dry brush before and after the switching experiment in Figure 5 confirm that no significant amount of material was lost in the pH-dependent measurements. Spectral changes in the range of CH2-stretching vibrations are not revoked after drying. The origin of these bands is not clear yet. They could be due to an irreversible structural change after the first treatment with pH 10 or organic contaminations. Reversible Switching. Here we discuss the reversible changes that occur with PAA-mix-P2VP brushes. In order to resolve the 7680 Analytical Chemistry, Vol. 79, No. 20, October 15, 2007

Figure 6. Normalized tan Ψ spectra of PAA-mix-P2VP brush demonstrating the consecutive treatment as follows (from bottom to top): pH 10 (2) f pH 6.5 (3) f pH 2 (4) f pH 6.5 (5) f pH 10 (6) f pH 2 (7) f pH 6.5 (8). As reference spectrum for normalization served the spectrum at pH 6.5 (3).

reversible switching of a brush upon its exposure to pH 10 and pH 2 aqueous solutions, the following steps have been investigated: pH 10 (2) f pH 6.5 (3) f pH 2 (4) f pH 6.5 (5) f pH 10 (6) f pH 2 (7) f pH 6.5 (8). Figure 6 shows the normalized tan Ψ spectra for each step of brush treatment. The observed spectral changes demonstrated in Figure 6 are in good agreement with the expected switching behavior of the brush as shown schematically in Figure 7. The positive bands revealed after a pH 10 cycle are attributed to the formation of dissociated PAA carboxylic

Figure 7. Schematic drawing showing the changes in the mixed PAA-mix-P2VP brush structure due to treatment by aqueous solution with different pH. At neutral conditions, both polymers are collapsed and build an intermolecular complex (Figure 3b). Treatment by acidic and alkaline aqueous solutions results in protonation of pyridine and in dissociation of carboxylic groups.

groups (1558 (νasCOO-) and 1414 cm-1 (νssCOO-)). The negative band at ∼1705-1730 cm-1 correlates with corresponding conversion of carboxyl groups. The band at ∼1705-1730 cm-1 at pH 2 originates from ν(CdO) of the carboxylic groups in acidic environment. The peak centered at 1630 cm-1 might be associated with the in-plane stretching vibration of the protonated pyridine ring.44 However, it cannot be excluded that a changing contribution from water complex bands42 might also overlap in this spectral range. None of the discussed bands is correlated with an error in the normalization procedure, because the direction of the band amplitude is changing selectively. Under the assumption that the film structure is in average isotropic, positive peaks can be interpreted as increase and negative peaks as a decrease of the number of corresponding molecular vibrations. Thus, we were able to distinguish spectroscopically three different states of the PAA-mix-P2VP brush in contact with aqueous solutions at pH 10, pH 6.5, and pH 2. Water (pH 6.5) was replaced by KOH (pH 10) solution in order to induce switching of the brush by the dissociation of PAA carboxylic groups. Following rinsing with water promoted protonation of the negative charged COO- groups and switching back to the “PAA...P2VP” complex. Injection of HCl (44) Mauser, T.; Dejugnat, C.; Sukhorukov, G. B. J. Phys. Chem. B 2006, 110, 20246-20253.

(pH 2) solution leads to the protonation of the pyridine nitrogen and increase of the density of positive charges on P2VP chains. The observed spectral changes upon treatment with KOH and HCl solutions are accompanied by changes in segregated phases in the brush where one of the polymers shrinks back to the grafting surface and the other one is dangling into solution.26 Beside the observed chemical changes during contact with the different aqueous solvents (bottom of Figure 7), a change of physical properties could also influence the infrared spectra. A considerable pH-dependent swelling of the brush (see also schematic in Figure 7) might induce changes in observed band amplitudes and shapes. Such an influence can principally be taken into account by optical calculations. This has not been done quantitatively so far, because the optical constants of the film material in the aqueous state are unknown and might change considerably in comparison to the dry state. Future work will be directed to resolve these dependencies and to study the adsorption of proteins and nanoparticles on the mixed polyelectrolyte brushes. Conclusive Remarks. With respect to quantitative interpretations of in situ infrared ellipsometric spectra of ultrathin polymer brush films, a straightforward evaluation is not possible at the present stage of research since the optical constants for the film Analytical Chemistry, Vol. 79, No. 20, October 15, 2007

7681

material in the aqueous phase are unknown. However, as can be seen from discussion with respect to Figure 2, some qualitative interpretations can be drawn when the optical constants of the dry material are used. Possible approaches for determination of optical constants of the material in aqueous phase would be the characterization of thicker films in aqueous phase or to consider the water contribution in a swollen thin film by an adequate model, e.g., effective medium approaches. So far, a quantitative evaluation of in situ ellipsometric spectra was possible only when the swelling of the film could be neglected, as, for example, the determination of thicknesses during the etching of thin silicon oxide films. In this case, it was possible to determine the optical constants, as necessary for the calculation, from ex situ measurements of a dry film. From a methodical point of view, standard reflectance and transmission experiments can be performed in principle in the same setup as used for IRSE. However, for ellipsometry measurements, typically additional polarizer/analyzer settings and a retarder (introducing a definite phase shift) are used. The optical setup is often better defined in the ellipsometric setup; e.g., in our case, the opening angle of the probing radiation is below (3.5°. Based on our measurement protocol,40 which calculates the ellipsometric parameters from in short order measured polarization-dependent amplitude ratios, the influence of disturbing absorptions of water vapor in the surrounding atmosphere is reduced even when the setup is not purged. Interpretation of measured ellipsometric spectra by optical simulations can give information on the thickness and the optical constants or the molecular orientation. It is noteworthy that the same optical models can be applied for interpretation of spectra from transmission or reflection measurements. Ellipsometry provides up to three parameters of the analyzed sample (tan Ψ, ∆, and polarization degree), which are method independent. That is not possible with reflectance or transmission spectroscopy alone, where often a Kramers-Kronig transformation is used to calculate a second parameter. On the other hand, polarized reflectance spectroscopy45 can measure s- and p-polarized reflectances (Rs and Rp) separately, whereby the ellipsometric parameter tan Ψ and ∆ depend on a mixture of both. Tan Ψ is defined as the amplitude ratio and ∆ as the phase shift (∆) of the two orthogonally polarized complex reflection coefficients (rs and rp). Polarized reflectance spectroscopy can be included in the ellipsometric measurement scheme, since tan Ψ ) (Rp/Rs)1/2. The different information content of ellipsometry can be used advantageously for quantitative evaluations of complex spectra. Thus, ellipsometry might be seen as a (45) Porter, M. D.; Anal. Chem. 1988, 60, A1143. (46) Roseler, A; Korte E. H., J Mol. Struct. 1995, 349, 321-324.

7682

Analytical Chemistry, Vol. 79, No. 20, October 15, 2007

valuable extension to reflectance and transmission experiments. Due to its high sensitivity and new possibilities for quantitative evaluation in situ IR spectroscopic ellipsometry could give new chemical and structural insights into the film structure in contact with aqueous solutions. With respect to the comparison of IRSE with ATR techniques, it must be mentioned that this special condition for the incidence angle can be included into ellipsometric experiments (ATR-ellipsometry46), if needed. The setup presented in this work can be adapted to ATR geometry by the use of a silicon prism as substrate and thereby the sensitivity could be increased (incidence angle at the silicon/liquid interface above ∼18-27° must be realized). For ATR geometry, the probing depth is limited to the penetration depth of the evanescent wave, which however would be far enough for investigations of films in the nanometer range. CONCLUSION The reversible switching of the mixed polyelectrolyte brush upon treatment with pH 6.5, pH 10, and pH 2 aqueous solutions has been monitored in terms of reversible changes of characteristic bands in tan Ψ spectra. The reported experiments clearly show high potential of infrared spectroscopic ellipsometry for online in situ studies of structural and chemical properties of stimuli-responsive polymer brushes in contact with an aqueous solution. ACKNOWLEDGMENT We thank Gudrun Adam and Ilona Fischer for their help in FT-IR spectroscopy and IR-ellipsometry measurements. The financial support by the Senatsverwaltung fu¨r Wissenschaft, Forschung und Kultur des Landes Berlin and support of the Bundesministerium fu¨r Bildung und Forschung as well as DFG (Hi 793/3-1; Sta 324/28-1; Ei 317/5-1) in Materials World Network program is gratefully acknowledged. J.R. and K.H. acknowledge the grant of European Union through the EFRE program (ProFIT grant, contract 10131870/71), in which the in situ cell application of IR spectroscopy ellipsometry was developed. Y.M. acknowledges the funding of European Union through the EFRE program (ProFIT grant, contract 10125494). S.M. acknowledges financial support provided by NSF (grant DMR 0602528 in the frame of the program Materials World Network).

Received for review April 26, 2007. Accepted August 6, 2007. AC070853A