5924
Langmuir 2001, 17, 5924-5931
Hyperbranched Polyesters on Solid Surfaces A. Sidorenko,† X. W. Zhai,† S. Peleshanko,† A. Greco,† V. V. Shevchenko,‡ and V. V. Tsukruk*,† Materials Science & Engineering Department, Iowa State University, Ames, Iowa 50011, and Institute of Macromolecular Chemistry, Kiev, 252160, Ukraine Received March 15, 2001. In Final Form: May 17, 2001 The interfacial behavior of third and fourth generations of hyperbranched polyesters (HBP3 and HBP4) with 32 and 64 hydroxyl-terminal groups was studied with scanning probe microscopy. The molecular adsorption on a bare silicon surface of both hyperbranched polymers was described in the terms of the Langmuir isotherm. A higher adsorption amount under an identical adsorption condition was found for lower generation HPB3. The shape of HBP3 molecules within an adsorbed layer evolved from a pancake with a thickness less than 1 nm for very low surface coverage to densely packed wormlike bilayer structures with a thickness of about 3 nm for the highest surface coverage. The molecules of the fourth generation, HBP4, hold a stable, close-to-spherical shape with a diameter of 2.5 nm throughout the entire range of surface coverage including both dense monolayers and isolated molecules. High intramolecular flexibility of HBP3 molecules as compared with constrained mobility of bulkier branches of HBP4 is considered to be responsible for different surface behavior.
Introduction Hyperbranched polymers are macromolecular compounds built from multifunctional monomers ABn, where the function A can couple with the function B as proposed by Flory and demonstrated by Kim and Webster.1,2 Contrary to highly regular dendrimers obtained in complex multistep processes, the hyperbranched molecules do not have every B fragment coupled. The branching is not well defined for these molecules because of the random character of a one-step reaction. Instead of the regular structure of dendrimers, locally irregular, somewhat defective structure occurs in hyperbranched molecules. However, the general chemical microstructure of hyperbranched molecules still reminds us of tree-like branches, and their properties are very different from coiled polymers.3 The dimension of the hyperbranched molecules (or a generation number according to dendrimer terminology) can be controlled by initial chemical monomer composition. The great advantage of hyperbranched polymers is that they are synthesized in one step and still possess all major elements, which are characteristic of highly compact nanoparticle-like structures typical for regular dendrimers of high generations.2,4,5 A variety of applications were considered for hyperbranched materials. Hyperbranched polymers were used as multifunctional initiators and for rheology control.2,6 They are considered to be promising * To whom correspondence should be addressed. E-mail:
[email protected]. † Iowa State University. ‡ Institute of Macromolecular Chemistry. (1) Flory, P. J. J. Am. Chem. Soc. 1952, 74, 2718. (2) Kim, Y. H.; Webster, O. W. J. Am. Chem. Soc. 1990, 112, 4592. (3) Frechet, J. M. J.; Hawker, G. J.; Gitsov, I.; Leon, J. W. J. Macromol. Sci., Pure Appl. Chem. 1996, A33(10), 1399. Jansen, J. F.; de Brabandervan den Berg, E. M.; Meijer, E. W. Science 1994, 266, 1226. Tomalia, D. A. Adv. Materials 1994, 6, 529. Percec, V.; Kawasumi, M. Macromolecules 1994, 27, 4441. Tomalia, D. A.; Naylor, A. M.; Goddard, W. A., III Angew. Chem. 1990, 29, 138. (4) Hawker, C. J.; Lee, R.; Frechet, J. M. J. J. Am. Chem. Soc. 1991, 113, 3(12), 4583. (5) Voit, B. J. Polym. Sci. Part A 2000, 38, 2505. Maier, G.; Zech, C.; Voit, B.; Komber, H. Macromol. Symp. 2001, 163, 75. Huber, T.; Boehme, F.; Komber, H.; Kronek, J.; Luston, J.; Voigt, D.; Voit, B. Macromol. Chem. Phys. 1999, 200, 126.
compounds for surface modification,7 medical applications,8 nanofillers for polymer nanocomposites,9 and nonlinear optics.10 The surface properties of hyperbranched polymers are controlled by the nature of terminal groups located on the periphery of the branches.11 Highly branched polymers demonstrated unusual thermal, mechanical, and rheological properties as compared with linear flexible polymers because of the absence of intermolecular entanglements.9,12 Hyperbranched polyesters with hydroxyl-terminal groups were studied for their transport properties13 and were used as tougheners in thermoplastics and thermosets.9,14 It seems that bulk material mechanical and transport properties were weakly dependent on generation number for low generations.13,14 However, to date, only a few articles have been dedicated to the investigation of the microstructure of hyperbranched polymers within adsorbed layers, and their interfacial behavior remains unexplored. Fujiki et al.15 developed a new approach of grafting of hyperbranched polyamidoamines on surface of fine SiO2 particles. Crooks and colleagues 16 synthesized surface-grafted hyperbranched poly(acrylic acid)s and other dendrimers and characterized their multilayered organization. Viville et al.17 and Huck (6) Kim, Y. H.; Webster, O. W. Macromolecules 1992, 25, 5561. (7) Zhao, M.; Zhou, Y.; Bruening, M. L.; Bergbreiter, D. E.; Crooks, R. M. Langmuir 1997, 13, 1388. Vicari, R.; Fruckey, O. S.; Juneau, K. N.; Thames, S. F.; Rawlins, J. W. (Herberts GmbH, Germany) U. S. 2000, 22pp. USXXAM US 6114489 A 20000905. Tsukruk, V. V. Prog. Polym. Sci. 1997, 22, 247. Newkome, G. R., Moorefield, C. N., Vogtle, F., Eds. Dendritic Molecules; VCH Publishers: Weinheim, 1996. (8) Ulrich, K. Trends Polym. Sci. 1997, 5(12), 388. (9) Xu, J.; Wu, H.; Mills, O. P.; Heiden, P. A. J. Appl. Polym. Sci. 1999, 72, 1065. Gopala, A.; Wu, H.; Xu, J.; Heiden, P. J. Appl. Polym. Sci. 1999, 71, 1809. (10) Zhang, Y.; Wang, L.; Wada, T.; Sasabe, H. J. Polym. Sci. 1996, 34, 1359. (11) Hult, A.; Johansson, M.; Malstro¨m, E. Adv. Polym. Sci. 1999, 143, 1. Kricheldorf, H. R.; Bolender, O.; Wollheim, T. Macromolecules 1999, 32, 3878. (12) Kim, Y. J. Polym. Sci. Part A: Polym. Chem. 1998, 36, 1685. (13) Hedenqvist, M. S.; Yousefi, H.; Malmstro¨m, E.; Johansson, M.; Hult, A.; Gedde, U. W.; Trolsås, M.; Hedrick, J. L. Polymer 2000, 41, 1827. (14) Xu, J.; Wu, H.; Liu, Y.; Heiden, P. A. J. Appl. Polym. Sci. 1999, 72, 151. (15) Fujiki, K.; Sakamoto, M.; Sato, T.; Tsubokawa, N. J. Macromol. Sci., Pure Appl. Chem. 2000, A37(4), 357.
10.1021/la0103970 CCC: $20.00 © 2001 American Chemical Society Published on Web 08/24/2001
Hyperbranched Polyesters on Solid Surfaces
Langmuir, Vol. 17, No. 19, 2001 5925
Figure 1. Idealized chemical structure of HBP4 molecule.
et al.18 studied the surface microstructure of hyperbranched polymers and observed well-defined nanoparticle shapes for high molecular weight specimens. Sheiko et al. considered surface behavior of well-defined hyperbranched macromolecules on various surfaces and demonstrated several examples of the individual behavior of isolated macromolecules controlled by fine intermolecular and interfacial interactions.19 In this study, we addressed the question of the interfacial molecular behavior of hyperbranched molecules in regard to their ability to hold a nanoparticle-like shape under a wide range of conditions. We studied the adsorbed conditions ranging from sparse, isolated molecules to dense, packed solid films. We report the results of the investigation of the surface adsorption and microstructure of hyperbranched polyesters (HBPs) with hydroxylterminal groups (Figure 1) adsorbed on a silicon surface. Two generations of these polyesters, HBP3 and HBP4, with the number of terminal groups of 32 and 64, respectively, were studied and discussed here. Experimental Section Hydroxyl-terminated polyesters of third and fourth generations were used as received from Aldrich. Gel permeation chromatography (GPC) measurements of the polymers were performed to compare actual molecular weight and theoretical molecular weight for the idealized AB2 structure (Figure 1). These measurements were conducted for HBPs in tetrahydrofuran (16) Bruening, M. L.; Zhou, Y.; Aguilar, G.; Agee, R.; Bergbreiter, D. E.; Crooks, R. M. Langmuir 1997, 13, 770. Franchina, J. G.; Dermody, D. L.; Peez, R.; Jones, S. J.; Bergbreiter, D. E.; Bruening, M. L.; Crooks, R. M. Polym. Mater. Sci. Eng. 1998, 78, 1. Hierlemann, A.; Campbell, J. K.; Baker, L. A.; Crooks, R. M.; Ricco, A. J. J. Am. Chem. Soc. 1998, 120, 5323. Wells, M.; Crooks, R. M. J. Am. Chem. Soc. 1996, 118, 3988. (17) Viville, P.; Deffieux, A.; Schappacher, M.; Leclere, Ph.; Bredas, J. L.; Lazzaroni, R. Polym. Prepr. (Am. Chem. Soc., Div. Polym. Chem.) 2000, 41(2), 1501. (18) Huck, W. T. S.; Snellink-Ruel, B. H. M.; Lichtehbelt, J. W. Th.; van Veggel, F. C. J. M.; Reinhoudt, D. N. Chem. Commun. 1997, 1, 9. (19) Sheiko, S.; Moller, M. Top. Curr. Chem. 2001, 212, 137. Prokhorova, S. A.; Sheiko, S. S.; Mourran, A.; Azumi, R.; Beginn, U.; Zipp, G.; Ahn, C.-H.; Holerca, M. N.; Percec, V.; Moeller, M. Langmuir 2000, 16(17), 6862. Sheiko, S. S. Adv. Polym. Sci. 2000, 151, 61. Sheiko, S. S.; Gauthier, M.; Moller, M. Macromolecules 1997, 30, 2343.
solutions by using a Waters-GPC equipped with Mini Dawn (Wyatt Technology) light-scattering detector. The substrates for adsorption were atomically smooth silicon wafers of the {100} orientation with one side polished (Semiconductor Processing Co.). Silicon wafers were treated in an ultrasonic bath for 10 min followed by a “piranha” solution (30% concentrated hydrogen peroxide, 70% concentrated sulfuric acid) bath for 1 h. After the “piranha” bath, the samples were rinsed several times with “Nanopure” water (resistivity, 18 MΩ cm) and dried under the stream of dry nitrogen. The HBPs adsorption was performed from acetone solution (Fisher, reagent grade) with different concentrations. After adsorption, the samples were moved to a pure acetone bath by six-step fractional solution replacement and, finally, dried under a stream of dry nitrogen. The amount and thickness of adsorbed HBPs on the silicon surfaces were measured with a Compel Ellipsometer (InOmTech, Inc.). The thickness of a native SiO2 layer of every silicon wafer was measured in three to five different locations before polymer deposition. It was about 1.2 nm for the wafer batch used in this study and the data scattering was within the typical ellipsometry accuracy of 0.1 nm. The averaged thickness of the SiO2 layer was taken into account when the ellipsometry measurements of the adsorbed amount were performed, and polymer layer thickness was calculated from double-layer model. Again, for the adsorbed layers, the measurements of the several locations were carried out and averaged with typical data scattering within range of 0.2 nm. For the calculation of the HBP layer thickness, the refractive indices of polymers should be known. To ensure reliable ellipsometric measurements, we found the refractive indices of HBPs from reference measurements. There is a well-known restriction in the measurements of both thickness and refractive index of thin layer.20 To overcome this limitation, thicker (about 40 nm) spin-coated polymer films were deposited from 10 g/L acetone solutions. An incident angle was found for reference spin-coated films of polystyrene with known refractive index of 1.59. The incident angle was measured with an accuracy of 0.03° and assumed to be the same for both reference sample and HPB films. By fixing this angle, both thickness and refractive index of HBP spin-coated films were calculated with an accuracy of 0.1 nm and 0.02, respectively. This way, the refractive indices of HBP3 and HBP4 were found to be 1.48 ( 0.01. It is worth to note (20) Azzam, R. M. A.; Bashara, N. M. Ellipsometry and Polarized Light; Amsterdam: North-Holland Publishing Co., 1977; 529 p.
5926
Langmuir, Vol. 17, No. 19, 2001
Sidorenko et al.
Table 1. Molecular Weight/Density Characteristics of the Hyperbranched Polyesters
HBP3 HBP4 a
theoretical molecular weight, g/mol
actual peak molecular weight, GPC, g/mol
Mw, GPC, g/mol
polydispersity index
F, g/cm3
height of complete layer,a nm
theoretical height, nm
3604 7316
3408 6443
2522 4243
1.88 1.32
1.19 0.92
3.0 2.5
1.7 2.5
As measured by scanning probe microscopy.
Figure 2. Kinetics of adsorption from the 1 g/L solution normalized to the saturation level of HBP3 (filled circles) and HBP4 (hollow circles). that the bulk refractive index of polymer with similar chemical structure, polylactic acid, is 1.45.21 The adsorbed amount, Γ, mg/m2, was calculated as:
Γ)F*d
(1)
where F is the HBP density and d is the thickness found from ellipsometry measurements. For this evaluation, we measured independently the bulk polymer density and assumed this density in our evaluation for adsorbed monolayers. This assumption was supported by very similar refractive indices for thin polymer layers and bulk polymers. Possible variations of the density within reasonable limits could not significantly affect the values discussed here (less than 10% deviations). Contact angles for adsorbed layers were recorded on a custom-built instrument combining a microscope and a digital camera. Three to five successive measurements were prepared for each sample. Imaging of the HBP layers was performed with an atomic force microscope (AFM) Dimension-3000 (Digital Instrument, Inc.) in the tapping mode according to an experimental procedure described in detail elsewhere.22 Ultrasharp silicon tips with the spring constant of 3.0 N/m were used. The regime of very “light” tapping (driving amplitude/free amplitude ratio was within 0.950.99) was set to avoid the compliant polymer layer damaging. The tip radius was estimated from scanning of the reference sample of tethered gold nanoparticles with the radius of 5 and 28 nm according to the known procedure.23,24 The tip radii were found to be 12.4 ( 0.9 and 28.4 ( 1.9 nm for two tips used for monolayer measurements of HBP3 and HBP4, respectively. To visualize molecular dimensions and shapes, molecular models were built with the HyperChem 6.0. The geometry optimization and molecule volume/dimensions calculations were performed with Cerius2 3.9 with the Universal force field on a SGI workstation. The globular conformations were obtained at (21) Aranishi, Y.; Takahashi, H. Jpn. Kokai Tokkyo Koho 2000, 6pp. JKXXAF JP 2000226727 A2 20000815. (22) Tsukruk, V. V. Rubber Chem. Technol. 1997, 70(3), 430. Tsukruk, V. V.; Reneker, D. H. Polymer 1995, 36, 1791. Scanning Probe Microscopy of Polymers; Ratner, B., Tsukruk, V. V., Eds.; ACS Symposium Series 694; American Chemical Society: Washington, DC, 1998. (23) PELCO Atomic Force Microscopy Gold Calibration Kit, Copyright TED PELLA, Inc., June 1999. (24) Vesenka, J.; Manne, S.; Giberson, R.; Marsh, T.; Henderson, E. Biophys. J. 1993, 65(9), 992.
Figure 3. (a) The isotherm of adsorption of HBP3 (filled circles) and contact angles (hollow circles) The solid line shows fitting of experimental adsorption data to the Langmuir isotherm equation. (b) Ellipsometry thickness (squares) and the average height of the HBP3-adsorbed layer (triangles) obtained by AFM vs solution concentration. temperature 500 K with dynamic simulations of the molecules in an initially extended conformational state for at least 10|000 steps and repeated several times. The “flattened” conformations were calculated by applying unidirectional “external stress” in the minimization mode with equivalent hydrostatic pressure of 66 GPa. After this procedure, the molecules were allowed to relax using dynamic simulations at temperature 300 K followed by the energy-minimization cycle. We should mention that, because of the flexible nature of the hyperbranched macromolecules, conformations presented here reflect only one of endless possibilities. However, we believe that such a representation is important to demonstrate that overall macromolecular dimensions discussed in the article are not forbidden by the chemical microstructure. On the other hand, such a “visualization” is important in the interpretation of the experimental data and understanding possible molecular shapes and their correlation with direct experimental results.
Hyperbranched Polyesters on Solid Surfaces
Figure 4. (a) The isotherm of adsorption of HBP4 (filled circles) and contact angles (hollow circles). The solid line shows fitting of experimental adsorption data to the Langmuir isotherm equation. (b) Ellipsometry thickness (squares) and the average height of HBP4-adsorbed layer (triangles) obtained by AFM vs solution concentration.
Results and Discussion GPC data shown in Table 1 confirmed that for both generations peak molecular weight was fairly close to the theoretical value estimated from the idealized chemical structure such as presented in Figure 1 for HBP4 polymer. Molecular weight distribution was modest for both polymers. (See polydispersity indices in Table 1.) Adsorption Behavior. The kinetics of adsorption for both HBP generations is presented in Figure 2. The adsorbed amount of material reached the saturation level within 30 min. The saturation time varied slightly for different concentrations but never exceeded 1 h. HBP3 demonstrated a distinctly slower rate of adsorption after the adsorption amount reached one half of the saturation level. Unlike HBP3, HBP4 polyester showed a very fast saturation time of less than one minute. This behavior can be a sign of significant differences in the microstructure of the adsorbed layers. There were no considerable changes of the adsorbed amount after 1 h of adsorption and within 24 h of adsorption as was tested for several concentrations. Only a minor increase of adsorbed amount (≈10%) was observed after 24 h of
Langmuir, Vol. 17, No. 19, 2001 5927
Figure 5. AFM images of HBP4-adsorbed molecules (topography, left; phase, right) obtained from different solution concentrations: 0.5 g/L (A), 1.5 g/L (B), and 5.0 g/L (C). Image 5A and 5B, C are taken from two independent series. Scan size is 2 × 2 µm, height scale is 5 nm, phase scale is 15°.
adsorption time from the most concentrated (1 g/L) solution of HBP3. The isotherms of adsorption for both polymers at the adsorption time of 3 h are presented in Figures 3a and 4a. Isotherms for both hyperbranched polymers demonstrated the typical characteristics of classic adsorption, which can be described in terms of the Langmuir equation25
Γ ) Γ∞ ×
KC 1 + KC
(2)
where Γ∞ is the maximum adsorption amount, C is the equilibrium concentration, and K is the adsorption constant showing the adsorption ability. The value of the adsorption constant K was obtained from Langmuir fitting and was about 0.84 ( 0.02 L/g for both polymers (Figures 3, 4). This is evidence that both HBP3 and HBP4 have a similar affinity to the silicon substrate because of the same nature of terminal groups. However, the third generation HBP3, reached higher adsorption amount (Γ∞ ) 3.2 mg/m2) than HBP4 under similar conditions (Γ∞ ) 2.1 mg/m2). The contact angle for adsorbed polymer films increased monotonically with solution concentration (Figures 3a, 4a). The contact angle reached 42 ( 2° and 60 ( 2° for (25) Atkins, P. W. Physical Chemistry; Oxford University Press: New York, 1982.
5928
Langmuir, Vol. 17, No. 19, 2001
Figure 6. High-resolution image (1 × 1 µm) of HBP4 molecules adsorbed from the solution of 0.3 g/L concentration, height scale is 5 nm, and the cross-section shows height variation along the lines shown on the image.
HBP3 and HBP4, respectively. These values were fairly close to the values obtained for annealed spin-coated films (35 ( 3° and 58 ( 2° for HBP3 and HBP4 films). This indicates full surface coverage of silicon substrates by the adsorbed polymer layers at saturation limit. Surface Morphology and Microstructure. The images of HBP4 layers on the silicon substrate at different concentrations are presented in Figure 5. The adsorption of HBP4 from a very diluted solution (
