GIUSAXS and AFM Studies on Surface Reconstruction of Latex Thin

Oct 17, 2008 - HASYLAB am DESY, Notkestrasse 85, D-22607 Hamburg, Germany. ReceiVed May 19, 2008. ReVised Manuscript ReceiVed July 15, 2008...
0 downloads 0 Views 794KB Size
4230

Langmuir 2009, 25, 4230-4234

GIUSAXS and AFM Studies on Surface Reconstruction of Latex Thin Films during Thermal Treatment† Shanshan Hu,‡ Jens Rieger,§ Stephan V. Roth,| Rainer Gehrke,| Reinhold J. Leyrer,§ and Yongfeng Men*,‡ State Key Laboratory of Polymer Physics and Chemistry, Changchun Institute of Applied Chemistry, Chinese Academy of Sciences, Graduate School of the Chinese Academy of Sciences, Renmin Street 5625, 130022 Changchun, PR China, BASF SE, Polymer Research, 67056 Ludwigshafen, Germany, and HASYLAB am DESY, Notkestrasse 85, D-22607 Hamburg, Germany ReceiVed May 19, 2008. ReVised Manuscript ReceiVed July 15, 2008 The structural evolution of a single-layer latex film during annealing was studied via grazing incidence ultrasmallangle X-ray scattering (GIUSAXS) and atomic force microscopy (AFM). The latex particles were composed of a low-Tg (-54 °C) core (n-butylacrylate, 30 wt %) and a high-Tg (41 °C) shell (t-butylacrylate, 70 wt %) and had an overall diameter of about 500 nm. GIUSAXS data indicate that the qy scan at qz ) 0.27 nm-1 (out-of-plane scan) contains information about both the structure factor and the form factor. The GIUSAXS data on latex films annealed at various temperatures ranging from room temperature to 140 °C indicate that the structure of the latex thin film beneath the surface changed significantly. The evolution of the out-of-plane scan plot reveals the surface reconstruction of the film. Furthermore, we also followed the time-dependent behavior of structural evolution when the latex film was annealed at a relatively low temperature (60 °C) where restructuring within the film can be followed that cannot be detected by AFM, which detects only surface morphology. Moreover, compared to AFM studies GIUSAXS provides averaged information covering larger areas.

Introduction Polymeric latex dispersions are widely used as paints, paper coatings, water based-adhesive, and so on. A great number of studies on the film formation of latex systems have been carried out using different techniques. (For reviews, see refs 1-3.) An understanding of the structural evolution of latex films upon exposure to elevated temperatures is of special interest because of its relevance to application-related properties of the films (e.g., durability, permeability, and optical appearance). It is generally accepted that there are three steps during the filmformation process: evaporative drying and ordering, particle deformation, and polymer interdiffusion. The third step normally is promoted by applying elevated temperatures and has been investigated by many researchers using small-angle neutronscattering,4-7 nonradiative energy transfer,8-10 and atomic force microscopy (AFM).11-14 Although it was often claimed that capillary pressure is the dominant driving force in film formation, † Part of the Neutron Reflectivity special issue. * Corresponding author. E-mail: [email protected]. ‡ Chinese Academy of Sciences. § BASF SE. | HASYLAB am DESY.

(1) Keddie, J. L. Mater. Sci. Eng. Res. 1997, 21, 101–170. (2) Winnik, M. A. Curr. Opin. Colloid Interface Sci. 1997, 2, 192–199. (3) Steward, P. A.; Hearn, J.; Wilkinson, M. C. AdV. Colloid Interface Sci. 2000, 86, 195–267. (4) Joanicot, M.; Wong, K.; Cabane, B. Macromolecules 1996, 29, 4976– 4984. (5) Joanicot, M.; Wong, K.; Richard, J.; Maquet, J.; Cabane, B. Macromolecules 1993, 26, 3168–3175. (6) Rieger, J.; Hadicke, E.; Ley, G.; Lindner, P. Phys. ReV. Lett. 1992, 68, 2782–2785. (7) Hahn, K.; Ley, G.; Schuller, H.; Oberthur, R. Colloid Polym. Sci. 1986, 264, 1092–1096. (8) Zhao, C. L.; Wang, Y.; Hruska, Z.; Winnik, M. A. Macromolecules 1990, 23, 4082–4087. (9) Pekcan, O.; Winnik, M. A.; Croucher, M. D. Macromolecules 1990, 23, 2673–2678. (10) Oh, J. K.; Tomba, P.; Ye, X.; Eley, R.; Rademacher, J.; Farwaha, R.; Winnik, M. A. Macromolecules 2003, 36, 5804–5814. (11) Lin, F.; Meier, D. J. Langmuir 1995, 11, 2726–2733.

the microscopic details of film formation are still a topic of considerable debate. Ordered arrays of latex particles are attracting significant attention for their potential applications in optical devices for data processing. Particularly, this attention has encouraged researches to study particles ordered as colloidal crystals in thin layers.15,16 When the latex particles have a core-shell structure and if there is a sufficient difference between optical refractive indices of the core and shell, colored films can be produced without any dye or pigment.17 The surfaces of nanostructured polymer films are routinely characterized by AFM, which is able to give a real-space visualization of the surface topography. With the advent of grazing incidence small-angle X-ray scattering (GISAXS), a powerful new technique for investigating the structure of samples of thin layers yields additional insight into these systems. Whereas with AFM only surface structures are accessible, with GISAXS the buried information is also detected. Real-time GISAXS measurements of the structural evolution of metallic nanoparticles have been performed successfully by relying on the strong scattering intensity that can be obtained.18-20 Although the contrast in nanostructured polymer films is not as strong as in hybrid materials, synchrotron GISAXS has also (12) Goudy, A.; Gee, M. L.; Biggs, S.; Underwood, S. Langmuir 1995, 11, 4454–4459. (13) Perez, E.; Lang, J. Macromolecules 1999, 32, 1626–1636. (14) Aramendia, E.; Mallegol, J.; Jeynes, C.; Barandiaran, M. J.; Keddie, J. L.; Asua, J. M. Langmuir 2003, 19, 3212–3221. (15) Denkov, N. D.; Velev, O. D.; Kralchevsky, P. A.; Ivanov, I. B.; Yoshimura, H.; Nagayama, K. Nature 1993, 361, 26–26. (16) Adachi, E.; Dimitrov, A. S.; Nagayama, K. Langmuir 1995, 11, 1057– 1060. (17) Wohlleben, W.; Bartels, F. W.; Altmann, S.; Leyrer, R. J. Langmuir 2007, 23, 2961–2969. (18) Renaud, G.; Lazzari, R.; Revenant, C.; Barbier,; Noblet, A. M.; Ulrich, O.; Leroy, F.; Jupille, J.; Borensztein, Y.; Henry, C. R.; Deville, J. P.; Scheurer, F.; Mane-Mane, J.; Fruchart, O. Science 2003, 300, 1416–1419. (19) Narayanan, S.; Lee, D. R.; Guico, R. S.; Sinha, S. K.; Wang, J. Phys. ReV. Lett. 2005, 94, 145504. (20) Winans, R. E.; Vajda, S.; Lee, B.; Riley, S. J.; Seifert, S.; Tikhonov, G. Y.; Tomeczyk, N. A. J. Phys. Chem. B 2004, 108, 18105–18107.

10.1021/la801527y CCC: $40.75  2009 American Chemical Society Published on Web 10/17/2008

Surface Reconstruction of Latex Thin Films

been successfully applied to polymeric thin films, indicating that the contrast between two different polymeric materials provides sufficient scattering intensity in synchrotron GISAXS measurements.21-24 Furthermore, in situ GISAXS studies on nanoporous materials25 and on systems exhibiting swelling-induced surface reconstruction26 were performed successfully. When structures on a larger length scale (e.g., up to a few micrometers) are considered (found, for example, in polymer blend films), grazing incidence ultrasmall angle X-ray scattering (GIUSAXS) must be performed to achieve adequate resolution.27-29 In this article, we present a GIUSAXS study of the structural evolution of a latex film composed of a single layer of polymeric particles under thermal treatment. AFM was used to investigate the surface texture of the film. One of the largest differences between these two methods is that AFM detects only localized areas with typical dimensions on the micrometer scale whereas GISAXS can measure films with areas greater than 400 × 40 000 µm2 in our experiments (i.e., the GISAXS experiment covers a larger area with the corresponding averaging). In bulk samples, transmission geometry is usually applied. The beam stop in SAXS or USAXS geometry limits the minimum q value, whereas in the GISAXS setup the beam stop is just at the position of the specular peak. As a result, the structure factor obtained on both sides of the Yoneda peak (see below) even at very small q values is not affected by the beam stop. Therefore, GIUSAXS experiments also provide structural information for larger particles and their structural ordering.

Experimental Section The latex particles used in the present study are composed of a low-Tg (-54 °C) core (n-butylacrylate, 30 wt %) and a high-Tg (41 °C) shell (t-butylacrylate, 70 wt %) with a pronounced hairy layer on the surface of the particles. The core-shell particles were synthesized using two-step radical emulsion polymerization. The reactions were performed in a glass reactor equipped with a reflux condenser, temperature controller, nitrogen gas inlet, monomer and initiator inlets, and mechanical stirrer. All reactions were carried out at 358 K and at a stirring rate of 150 rpm (with a horseshoe mixer). Prior to polymerization, the reaction mixture was purged with nitrogen, and a slightly positive nitrogen pressure was maintained during the reaction. Particle size distributions were characterized in dispersion with analytical ultracentrifugation (AUC), which results in the triad D10/D50/D90 at 481/498/532 nm, respectively. Silicon substrates (1 × 4 cm2) were pretreated in 3:7 H2O2/H2SO4 overnight, making them superhydrophilic. Then the substrates were dipped into an open 20 mL glass with a 1 wt % diluted dispersion. The particles were deposited onto the vertical substrate and self-assembled during the evaporation of water at room temperature (22 ( 1 °C). The relative humidity was 30 ( 5%. The substrate was kept vertical during the whole drying process. The films on the silicon substrates were annealed by placing them directly on a copper plate in a preheated oven. The annealed samples were quenched to room temperature immediately after thermal treatment. (21) Mueller-Buschbaum, P. Anal. Bioanal. Chem. 2003, 376, 3–10. (22) Mueller-Buschbaum, P.; Wolkenhauer, M.; Wunnicke, O.; Stamm, M.; Cubitt, R.; Petry, W. Langmuir 2001, 17, 5567–5575. (23) Busch, P.; Posselt, D.; Smilgies, D. M.; Rauscher, M.; Papadakis, C. M. Macromolecules 2007, 40, 630–640. (24) Lee, B.; Park, I.; Yoon, J.; Park, S.; Kim, Jehan.; Kim, K.; Chang, T.; Ree, M. Macromolecules 2005, 38, 4311–4323. (25) Lee, R.; Yoon, J.; Oh, W.; Hwang, Y.; Heo, K.; Jin, K. S.; Kim, J.; Kim, K. W.; Ree, M. Macromolecules 2005, 38, 3395–3405. (26) Xu, T.; Goldbach, J. T.; Misner, M. J.; Kim, S.; Gibaud, A.; Gang, O.; Ocko; Ben.; Guarini, K. W.; Black, C. T.; Hawker, C. J.; Russell, T. P. Macromolecules 2004, 37, 2972–2977. (27) Mueller-Buschbaum, P. Prog. Colloid Polym. Sci. 2006, 132, 23–32. (28) Mueller-Buschbaum, P.; Bauer, E.; Maurer, E.; Roth, S. V.; Gehrke, R.; Burghammer, M.; Riekel, C. J. Appl. Crystallogr. 2007, 40, s341-s345. (29) Mueller-Buschbaum, P.; Bauer, E.; Maurer, E.; Schloegl, K.; Roth, S. V.; Gehrke, R. Appl. Phys. Lett. 2006, 88, 083144.

Langmuir, Vol. 25, No. 7, 2009 4231 Synchrotron GIUSAXS measurements were performed at the BW4 beamline at HASYLAB, DESY, Hamburg, Germany.30 The energy of the X-ray radiation was 8.979 keV, resulting in a wavelength of 0.13808 nm. The size of the primary X-ray beam at the sample position was 0.4 × 0.4 mm2. The sample-detector distance was 13 037 mm, resulting in GIUSAXS geometry. In this geometry, at a very large sample-detector distance a high resolution of 2.75 × 10-4 nm-1 and the maximum accessible lateral length scale (21 µm) were achieved.28,29 The scattering vector q was decomposed as follows: b q )b qy + b qz, where b qy and b qz are the horizontal and vertical components, respectively, and |q b| ) q ) (4π/λ) sin θ (where 2θ is the scattering angle and λ is the wavelength). The GIUSAXS experiments were performed in a vacuum chamber equipped with a sample holder mounted on a goniometer. The scattering intensity was recorded at a fixed angle of incidence (Ri ) 0.3°) of the X-ray beam with respect to the sample surface. The specular peaks were always kept at the same position on the detector. GIUSAXS data were recorded with an exposure time of 15 min and with a 2D detector array (2048 pixels × 2048 pixels, pixel size 79.1 µm). The AFM measurements were carried out with a scanning probe microscope (SPA-300HV, Seiko Instruments Inc., Japan).

Theory When reflection/refraction effects at interfaces have to be accounted for, the scattering factor has to be calculated according to the distorted wave Born approximation (DWBA). In the framework of the DWBA, the differential cross-section is given by31

dσ Cπ2 b ) ∝ F(q b) ) 4 (1 - n2)2|Ti|2|Tf|2F(q dΩ λ

(1)

where C is the illuminated surface area, n is the refractive index, Ti and Tf are the Fresnel transmission functions, and F(q) is the diffuse scattering factor. If the incident or exit angle is equal to the critical angle of the material, then the transmission functions have a maximum, which is called the Yoneda peak.32 For N identical and centrosymmetric objects with random orientation, the diffuse scattering factor can be approximated by

F(q b ) ∝ NP(q b ) S(q b)

(2)

where P(q) is the form factor of the individual objects and S(q) is the structure factor. In our experiments, the particles can be considered to be spheres. The core-shell structure becomes apparent only at larger q values, which are not of interest here. The theoretical scattering intensity distribution of a sphere (i.e., the form factor P(q)) is given by33

P(q) ∝

(sin qR - qR cos qR)2 (qR)6

(3)

where R is the radius of the particle. The finite particle size distribution is neglected in a first-order approximation.

Results and Discussion GIUSAXS Map Analysis. Figure 1a shows the 2D GIUSAXS pattern of the latex film before annealing where distinct outof-plane diffractions and off-specular reflections are detected. Figure 1b shows the corresponding AFM image. From the AFM (30) Roth, S. V.; Doehrmann, R.; Dommach, M.; Kuhlmann, M.; Kroeger, I.; Gehrke, R.; Walter, H.; Schroer, C.; Lengeler, B.; Mueller-Buschbaum, P. ReV. Sci. Instrum. 2006, 77, 085106. (31) Salditt, T.; Metzger, T. H.; Peisl, J.; Reinecker, B.; Moske, M.; Samer, K. Europhys. Lett. 1995, 32, 331–336. (32) Yoneda, Y. Phys. ReV. 1963, 131, 2010–2013. (33) Roe, R.-J. Methods of X-ray and Neutron Scattering in Polymer Science; Oxford University Press ; New York , 2000; pp 155-162.

4232 Langmuir, Vol. 25, No. 7, 2009

Hu et al.

Figure 1. Grazing incidence ultra-small-angle X-ray scattering (GIUSAXS) patterns of a latex film on a silicon substrate dried at room temperature (a), annealed at 60 °C for 2 h (c), and annealed at 140 °C for 2 h (e). Tapping-mode atomic force microscopy phase images (5 × 5 µm2) of the film after annealing at the respective temperatures (b, d, and f).

images, it is derived that the film consists of a single layer of particles. It is seen that the latex particles locally self-assemble into hexagonal close packing during water evaporation. However, the particles form a multiple-domain structure rather than a continuous film because of the low concentration of the latex dispersion. Because the reflection-refraction effects at interfaces have to be accounted for, the DWBA must be applied to analyze the data.34,35 The contrast between the particle and the air contributes to the scattering intensity. The scattering intensity distribution along a horizontal slice parallel to the wave vector component qy is called out-of-plane scan; an example is shown for the present system in Figure 2a. (34) Vineyard, G. H. Phys. ReV. B 1982, 26, 4142–4159. (35) Rauscher, M.; Salditt, T.; Spohn, H. Phys. ReV. B 1995, 52, 16855– 16863.

A distinct peak in the q range of 0.01-0.015 nm-1 is observed. The inset shows the two typical distances in the 2D hexagonal close-packed arrangement of particles. The first distance, d1, which is equal to the particle diameter d, is related to two adjacent particles with a spacing in reciprocal space of q1 ) 2π/d. The second one, d2, is responsible for the reflection of Bragg rods (10) with a rod spacing in reciprocal space of q2 ) 4π/d3. The detected broad intensity distribution in the peak is thus assumed to be due to the overlapping of the two Bragg peaks caused by the local hexagonally ordered structures. It is possible to simulate GIUSAXS patterns with software developed by Lazzari,36 as is evidenced in refs 18 and 37. Froemsdorf et al. successfully simulated the DWBA form factor (36) Rauscher, M.; Paniago, R.; Metzger, H.; Kovats, Z.; Domke, J.; Peisl, J.; Pfannes, H.-D.; Schulze, J.; Eisele, I. J. Appl. Phys. 1999, 86, 6763–6769.

Surface Reconstruction of Latex Thin Films

Langmuir, Vol. 25, No. 7, 2009 4233

Figure 3. GIUSAXS patterns along the qy direction at qz ) 0.27nm-1. The latex film was annealed at different temperatures (60, 100, and 140 °C) for 2 h.

Figure 2. Scan of the GIUSAXS pattern of the latex film before annealing (Figure 1a) along the qy direction at qz ) 0.27 nm-1 (a); the inset indicates the structure of the latex film. Scan along the qy direction at qz ) 0.49 nm-1 compared to the scattering of sphere particles (b).

of spheres in a 2D hexagonal lattice;38 their results are in line with our findings with respect to the scattering behavior around the specular peak in our experiments. As seen in Figure 1a, the high-resolution GIUSAXS experiment shows a very pronounced interference pattern. Figure 2b presents the qy scan at qz ) 0.49 nm-1 (off-specular scan) with the theoretical scattering curve, using eq 3, assuming spherical particles with a radius of 260 nm. Hence, the qy scan at qz ) 0.49 nm-1 reveals the form factor of the particles, whereas the scan at qz ) 0.27 nm-1 (out-of-plane scan) gives both the structure factor and the form factor, which will be discussed later. Structural Evolution. GIUSAXS patterns of the latex films after thermal treatment are shown in Figure 1c for films annealed at 60 °C for 2 h and (e) annealed at 140 °C for 2 h. In Figure 3, we present the evolution of the 1D scattering intensity distributions of the out-of-plane scans of the latex films annealed at various temperatures ranging from room temperature to 140 °C. These data indicate that the structure of the latex film changes distinctly during annealing. For example, the Bragg peaks at q ) 0.01-0.015 nm-1 disappeared when the films were annealed at the given temperatures for 2 h. Besides, a Bragg peak at q ) 0.1 nm-1 was detected when the films were annealed at higher temperatures (100 and 140 °C). The AFM image of the film annealed at 140 °C shows that the particles are deformed and coalesce into perfect hexagonal close packing. Because there was no water in the film, it is safe to assume that the deformation and coalescence are driven by the polymer/air interfacial tension. Furthermore, the large domains originating from particle coalescence contribute to the increase in the intensity at very small q values around the Yoneda peak. (37) Metwalli, E.; Couet, S.; Schlage, K.; Roehlsberger, R.; Korstgens, V.; Ruderer, M.; Wang, W.; Kaune, G.; Roth, S. V.; Mueller-Buschbaum, P. Langmuir 2008, 24, 4265–4272. (38) Froemsdorf, A.; Capek, R.; Roth, S. V. J. Phys. Chem. B 2006, 110, 15166–15171.

To study the resultant structure after annealing, we analyzed the out-of-plane scans of the GIUSAXS patterns when the latex film was annealed at 140 °C. The Bragg peak at q ) 0.1 nm-1 is assumed to be due to coherent scattering from a structure with a typical length scale of 60 nm. Because the corresponding Bragg peak appears after prolonged annealing at temperatures far above the glass-transition temperature, it is assumed that this hypothetical structure is caused by some phase-separated structures. At present, it is not known which component of the latex might be responsible for this phase separation. The AFM phase image of the latex film annealed at 140 °C for 2 h shows perfect hexagonal close packing of latex particles whereas no information about the above ordering can be observed from the GIUSAXS pattern. As mentioned before, AFM reveals only the surface texture of the film; GIUSAXS can detect the buried information. The analysis of GIUSAXS pattern reveals that these latex particles are coalesced together or at least there is no contrast between the particles after annealing. It should be mentioned that the restructuring process was finished within 5 min at 100 °C because further annealing for 2 h at this temperature yielded identical GIUSAXS patterns. Hence, particle reconstruction proceeds too fast at this temperature to be followed. The data for the latex film annealed at 60 °C for 2 h (Figure 3) show only a weak peak at large q values (around qy ) 0.1 nm-1), which we believe is due to an intermediate stage in the structural evolution. Therefore, a temperature of 60 °C was chosen to investigate further the time-dependent behavior of structural evolution. Figure 4a presents the time-dependent structural evolution, as reflected in the scattering behavior, when annealing the latex film at 60 °C. It can be clearly seen that at this temperature the structure developed rapidly during the first 30 min, followed by a slower evolution at longer annealing times. The intensity of the Bragg peak in the q range of 0.01-0.015 nm-1 decreased until it vanished after prolonged annealing at 60 °C. Interestingly, additional peaks appeared when the latex film was annealed at 60 °C for 5 min, accompanied by the decrease in the intensity of the Bragg peak in the q range of 0.01-0.015 nm-1. We analyzed the data obtained when the latex film was annealed at 60 °C for 5 min (Figure 4b). The scan in Figure 4b clearly shows the scattering of single spheres because the data can be fitted by the theoretical scattering of spherical particles with a radius of 260 nm. This phenomenon can be attributed to the gradual coalescence of adjacent particles, resulting in a loss of internal contrast within the hexagonally packed domains. The scattering of disordered individual particles, as described by their form factor, thus appears in this intermediate stage. In later stages of the annealing process, the scattering intensity due to this form factor vanishes because

4234 Langmuir, Vol. 25, No. 7, 2009

Figure 4. Scans of a series GIUSAXS patterns along the qy direction at qz ) 0.27 nm-1, where the latex film was annealed at 60 °C for various times (a). Latex film annealed at 60 °C for 5 min and theoretical scattering behavior of spherical particles with a radius of 260 nm (b).

the majority of the particles have been fused. However, the AFM images of the films annealed at 60 °C from 5 to 800 min show nearly identical surface textures. Thus, the internal film structure provides the scattering intensity distributions that cannot be observed by the AFM technique. Accordingly, valuable information about structural changes within the film during aging can be obtained by GIUSAXS.

Hu et al.

AFM. The AFM images showed the expected smoothening of the surface morphology of the latex film during annealing. From the analysis of the GIUSAXS maps, one finds that the qy scan at qz ) 0.27 nm-1 (out-of-plane scan) is determined both by the structure factor and the form factor of the polymeric particles. Interference patterns around the specular peak (qz ) 0.49 nm-1) clearly correspond to scattering dominated by single-particle scattering as described by the form factor. The GIUSAXS experiments on latex films annealed at various temperatures ranging from room temperature to 140 °C indicate that the structure of the latex thin film changed distinctly. The disappearance of the Bragg peaks representing large-scale ordering and the appearance of a new Bragg peak representing smallscale ordering indicate a thermally induced reconstruction of the latex film. Compared to GIUSAXS measurements, the AFM images of the films annealed at 60 °C from 5 to 800 min show nearly identical surface textures. Thus, the changes in the GIUSAXS patterns result from changes in the internal structure of the films. A complete structural model for the evolution of close-packed polymeric particles as a function of time and temperature is not yet available, but the present data indicate that there are more restructuring processes that are active than might be assumed by a superficial consideration. It turns out that GIUSAXS is a method that is ideally suited to the study of such phenomena in thin films. Acknowledgment. Y.M. thanks the Hundred-Talent Project of the Chinese Academy of Sciences, the National Science Foundation of China (the fund for Creative Research Groups 50621302), and HASYLAB project II-20052011. We thank Dr. A. Timman for assistance with GIUSAXS experiments at HASYLAB.

Conclusions The structural evolution of a latex film made of a monolayer of polymeric particles was studied by means of GIUSAXS and

LA801527Y