Thickness-Dependent Molecular Chain and Lamellar Crystal

A remarkable influence of the orientations of the silicon backbone on the UV .... Shuang Yang , Er-Qiang Chen , Chun-Xia Sun , Alexander Stroeks , Tie...
1 downloads 0 Views 351KB Size
Download PDF
Langmuir 2004, 20, 3271-3277

3271

Thickness-Dependent Molecular Chain and Lamellar Crystal Orientation in Ultrathin Poly(di-n-hexylsilane) Films Zhijun Hu, Haiying Huang, Fajun Zhang, Binyang Du, and Tianbai He* State Key Laboratory of Polymer Physics and Chemistry, Changchun Institute of Applied Chemistry, Chinese Academy of Science, Changchun 130022, China Received October 29, 2003. In Final Form: January 18, 2004 The molecular chain and lamellar crystal orientation in ultrathin films (thickness < 100 nm) of poly(di-n-hexylsilane) (PDHS) on silicon wafer substrates have been investigated by using transmission electronic microscopy, wide-angle X-ray diffraction, atomic force microscopy, and UV absorption spectroscopy. PDHS showed a film thickness-dependent molecular chain and lamellar crystal orientation. Lamellar crystals grew preferentially in flat-on orientation in the monolayer ultrathin films of PDHS, i.e., the silicon backbones were oriented along the surface-normal direction. By contrast, the orientation of lamellar crystals was preferentially edge-on in ultrathin films thicker than ca. 13 nm, i.e., the silicon backbones were oriented parallel to the substrate surface. We interpret the different orientations of molecular chain and lamellar crystal as due to the reduction of the entropy of the polymer chain near the substrate surface and the particularity of the crystallographic (001) plane of flat-on lamellae, respectively. A remarkable influence of the orientations of the silicon backbone on the UV absorption of these PDHS ultrathin films was observed due to the one-dimensional nature of σ-electrons delocalized along the silicon backbone. With the silicon backbones perpendicular or parallel to the surface of the substrate, the UV absorbance increased or decreased with an increase of the angle between the incident UV beam direction and direction normal to the thin film, respectively.

Introduction Polymer thin and ultrathin films have received considerable attention both scientifically and technologically in recent years.1-4 Understanding and controlling the polymer chain organization and orientation in thin and ultrathin films are of central importance in relation to adhesion, surface wetting, liquid crystal alignment, and electronic and optical properties.5 Extensive studies on amorphous polymers have shown that thin films of submicrometer thickness exhibit different properties as compared with bulk material.6-22 Computer simulations6 showed an enrichment of the polymer/substrate and * To whom correspondence should be addressed. Phone: +86431-5262123. Fax: 86-431-5262126. E-mail: [email protected]. (1) de Gennes, P.-G. Rev. Mod. Phys. 1985, 57, 827. (2) Frank, C. W.; Bao, V.; Despotopoulou, M. M.; Pease, R. F. W.; Hinsberg, W. D.; Miller, R. D.; Rabolt, J. F. Science 1996, 273, 912. (3) Mellbring, O.; Kihlman Øiseth, S.; Krozer, A.; Lausmaa, J.; Hjertberg, T. Macromolecules 2001, 34, 7496. (4) Perahia, D.; Traiphol, R.; Bunz, U. H. F. Macromolecules 2001, 34, 151. (5) Seki, T.; Fukuda, K.; Ichimura, K. Langmuir 1999, 15, 5098. (6) Physics of Polymer Surfaces and Interfaces; Sanchez, I. C., Ed.; Butterworth-Heinemann: Boston, MA, 1992. (7) Jones, R. L.; Kumar, S. K.; Ho, D. L.; Briber, R. M.; Russell, T. P. Nature 1999, 400, 146. (8) Prest W. M.; Luca, D. J. J. Appl. Phys. 1980, 51, 5170. (9) Keddie, J. L.; Jones, R. A. L.; Cory, R. A. Europhys. Lett. 1994, 27, 59. (10) Forrest, J. A.; Mattson, J. Phys. Rev. E 2000, 61, R53. (11) Forrest, J. A.; Dalnoki-Veress, K.; Stevens, J. R.; Dutcher, J. R. Phys. Rev. Lett. 1996, 77, 2002. (12) Xie, L.; DeMaggio, G. B.; Frieze, W. E.; DeVries, J.; Gidley, D. W.; Hristov, H. A.; Yee, A. F. Phys. Rev. Lett. 1996, 74, 4947. (13) Kim, J. H.; Jang, J.; Zin, W. C. Langmuir 2000, 16, 4064. (14) Kim, J. H.; Jang, J.; Zin, W. C. Langmuir 2001, 17, 2703. (15) DeMaggio, G. B.; Frieze, W. E.; Gidley, D. W.; Zhu, M.; Hristov, H. A.; Yee, A. F. Phys. Rev. Lett. 1997, 78, 1524. (16) Fryer, D. S.; Nealey, P. F.; de Pablo, J. J. Macromolecules 2000, 33, 6439. (17) Torres, J. A.; Nealey, P. E.; de Pablo, J. J. Phys. Rev. Lett. 2000, 85, 3221. (18) Tsui, O. K. C.; Russell, T. P.; Hawker, C. J. Macromolecules 2001, 34, 5535.

polymer/air interfaces with polymer chain ends, and the chains exhibited a strong tendency to orient with their longest dimension parallel to the surface. In accord with these ideas, Jones et al.7 found that the Gaussian conformation of amorphous polymer was retained parallel to the surface in ultrathin films by using small-angle neutron scattering. Similarly, Prest et al.8 found that the solventcasting process preferentially aligned amorphous polymer chains in the plane of thin film (1-5 µm). The confinement of amorphous polymers in ultrathin films is known to significantly alter numerous physical properties, such as the glass-transition temperature9-18 and molecular mobility.19-22 Because the critical film thickness, wherein the mobility begins to decrease, is comparable to the fully extended chain length, Calvert proposed20 that amorphous polymer chains at the polymer/substrate interface were not coils but stretched out straight into the films such as reeds on the pond bed. Although much progress has been achieved on amorphous polymers, the crystallization of polymers in thin and ultrathin films is still lacking comprehensive insight.23 Several conflicting results have been reported in the literature. For instance, Frank et al.2,24,25found that the crystallization of poly(di-n-hexylsilane) (PDHS) was substantially hindered in ultrathin films, wherein a critical thickness of 15 nm was needed for crystalline morphology to exist and in which the rate of crystallization was initially (19) Frank, B.; Gast, A. P.; Russell, T. P.; Brown, H. R.; Hawker, C. Macromolecules 1996, 29, 6531. (20) Calvert, P. Science 1996, 384, 311. (21) Zheng, X.; Rafailovich, M. H.; Sokolov, J.; Strzhemechny, Y.; Schwarz, S. A.; Sauer, B. B.; Rubinstein, M. Phys. Rev. Lett. 1997, 79, 241. (22) Lin, E. K.; Kolb, R.; Satija, S. K.; Wu, W.-L. Macromolecules 1999, 32, 3753. (23) Reiter, G.; Sommer, J. U. Phys. Rev. Lett. 1998, 80, 3771. (24) Despotopoulou, M. M.; Frank, C. W.; Miller, R. D.; Rabolt, J. F. Macromolecules 1995, 28, 6687. (25) Despotopoulou, M. M.; Frank, C. W.; Miller, R. D.; Rabolt, J. F. Macromolecules 1996, 29, 5797.

10.1021/la036033k CCC: $27.50 © 2004 American Chemical Society Published on Web 03/02/2004

3272

Langmuir, Vol. 20, No. 8, 2004

Hu et al.

crystals can be physically understood as due to the reduction of the entropy of the polymer chain near the substrate surface and the particularity of the crystallographic (001) plane of flat-on lamellae. A remarkable influence of the orientation of silicon backbone on the UV absorption was observed due to the one-dimensional nature of σ-electrons delocalized along the silicon backbone. Experimental Section

Figure 1. Schematic illustration for the edge-on lamellae (a) and flat-on lamellae (b) formed on substrates.

slow but increased rapidly as the ultrathin film approached ca. 50 nm in thickness. In contrast to these results, Reiter et al.23,26,27 found that poly(ethylene oxide) was able to crystallize by attaching molecules which diffused toward the edge of the crystal, even if the polymer was confined in ultrathin films. In addition, the chain axis of poly(ethylene oxide) was perpendicular to the substrate. Recently, Frank et al.28,29 found that the polymer chain orientation of poly(ethylene oxide) depended on the thickness of thin and ultrathin films. The helices of poly(ethylene oxide) were oriented along the surface-normal direction in films thinner than 300 nm. On the contrary, the molecules were perpendicular to the surface-normal direction in films thicker than ca. 1 µm. On the basis of these conflicting results, it is clear that even the most fundamental questions regarding the behavior of polymer chains near the surface are poorly understood.7 In a previous study,30 we reported the dependence of the silicon backbone and lamellar crystal orientation on the polymer-substrate interaction in ultrathin poly(din-butylsilane) (PDBS) and PDHS films. In the case of a strong polymer-substrate interaction, both PDBS and PDHS ultrathin films grew into edge-on lamellae on the surface of highly oriented pyrolytic graphite, i.e., the silicon backbones were parallel to the substrate surface (schematically illustrated in Figure 1a). In contrast, in the case of a weak van der Waals interaction between the polymer and carbon-coated substrate, lamellar crystals of PDHS in ultrathin films grew preferentially in a flat-on orientation with silicon backbones parallel to the surface-normal direction (Figure 1b). In this paper, we report the thickness-dependent molecular chain and lamellar crystal orientation in PDHS ultrathin films. PDHS was selected as the model polymer because polysilanes have been attracting considerable attention due to their unique electronic and optical properties.31 Polysilanes exhibit semiconducting characteristics such as photoconductivity,32 charge carriers (holes) transport property,33 and nonlinear optical effect,34 similar to those of π-conjugated polymers. The occurrence of different orientations of molecular chain and lamellae (26) Reiter, G.; Sommer, J. U. J. Chem. Phys. 2000, 112, 4376. (27) Sommer, J. U.; Reiter, G. J. Chem. Phys. 2000, 112, 4384. (28) Scho¨nherr, H.; Frank, C. W. Macromolecules 2003, 36, 1188. (29) Scho¨nherr, H.; Frank, C. W. Macromolecules 2003, 36, 1199. (30) Hu, Z.; Zhang, F.; Du, B.; Huang, H.; He, T. Langmuir 2003, 19, 9013. (31) Miller, R. D.; Michl, J. Chem. Rev. 1989, 89, 1359.

Sample Preparation. PDHS was prepared from the corresponding dichlorosilanes by dehalogenation coupling with sodium under standard conditions for polysilanes synthesis by the Wurtz coupling reaction.35 After purification from toluene with isopropyl alcohol (twice) and tetrahydrofuran with methanol, a flocculent, pure white, oligomer-free sample of PDHS was obtained. The molecular weight determined by gel permeation chromatography (GPC) in tetrahydrofuran solution was Mn ) 1.2 × 104 and Mw/ Mn ) 1.4. Solutions of the polymer were prepared in hexane at concentrations ranging from 0.05% to 1.0% (w/v). The substrates used for the experiments were silicon wafers. They were cleaned in a mixture of H2SO4, H2O2, and H2O (70/21/9 vol %) for 30 min at 120 °C. After cooling to room temperature, the substrates were thoroughly rinsed with deionized water. All substrates were dried in a stream of nitrogen and used immediately for film deposition. Film Preparation. The PDHS ultrathin films with different thicknesses were obtained by spin-coating solutions at various concentrations at 3000 rpm for 30 s at room temperature. During the spinning procedure, the color of the samples changed, indicating the thinning of the film. The residual solvent was removed under vacuum for 2 h at room temperature. Instruments. The film thickness of newly prepared films was measured with an Auto RLR-_ automatic ellipsometer with a He-Ne laser source. Wide-angle X-ray diffraction (WAXD) experiments were conducted with a Rigaku 18 kW rotating-anode generator (Cu KR) with a diffractometer with I ) 300 mA and V ) 50 kV. The experiments were carried out in a reflection mode to allow the study of the film on top of the silicon wafers. The X-ray beam was monochromatized using a graphite crystal. The diffraction peak positions and widths were calibrated through silicon crystals with known crystal sizes. The 2θ angle region ranged between 2° and 40° with a scanning rate of 0.3°/min. Transmission electronic microscopy (TEM) experiments were performed with a JEOL 2010 TEM with an accelerating voltage of 200 kV. Before the samples were examined, the thin films were coated with carbon, peeled off using HF, and transferred to a copper grid. Platinum and gold were evaporated on the thin film for morphology observations and calibration of the electron diffraction (ED) spacing, respectively. Atomic force microscopy (AFM) studies were performed with a SPA-300HV AFM with a SPI 3800N controller (Seiko Instruments Industry Co., Ltd.). All experiments were carried out using dynamic force mode. Etched Si probes with a resonant frequency of 250-300 kHz and spring constants of 42 N/m were used. The UV absorption experiments were conducted on a Cary 500 UV-Vis-NIR spectrometer (Varian Analytical Instruments). The experiments were carried out in reflection mode at different incident angles (the angle between the incident UV beam direction and normal direction of the thin film), which ranged from 20° to 70°.

Results and Discussion Lamellar Crystal Morphology and Silicon Backbone Orientation in Monolayer Ultrathin Film. Figure 2a shows a typical AFM height image of PDHS (32) Kepler, R. G.; Zeigler, J. M.; Havrah, L. A.; Kurtz, S. R. Phys. Rev. B 1987, 35, 2818. (33) van der Laan, G. P.; de Haas, P.; Hummel, A.; Sheiko, S.; Mo¨ller, M. Macromolecules 1994, 27, 1897. (34) Hasegawa, T.; Iwasa, Y.; Kishida, H.; Koda, T.; Tokura, Y.; Tachibana, H.; Kawabata, Y. Phys. Rev. B 1992, 45, 6317. (35) Trefonas, P.; West, R. Inorg. Synth. 1988, 25, 58.

Ultrathin Poly(di-n-hexylsilane) Films

Figure 2. (a) AFM height image of a PDHS ultrathin film with thickness of 12.7 nm prepared by spin coating from solution with a concentration of 0.05% w/v onto the surface of a silicon wafer. The contrast covers the surface corrugations in the 0-17.9 nm range. (b) Profile line of the cross section between A and B shown in a.

ultrathin film with a thickness of 12.7 nm spin-coated from solution (0.05% (w/v)) onto the surface of a silicon wafer. Ramified lamellar pattern covers the whole surface of the silicon wafer. A clear characteristic of these crystals is their lack of regular crystallographic faceting. This may very likely be a consequence of the relatively rapid crystallizing rate and pseudohexagonal unit cell of PDHS crystal.36 The fact that the substrate used in this case is amorphous eliminates any possible interfacial influence of a crystallographic or epitaxial nature. Another characteristic of these crystals is that the height of the lamellae varies in the range of 10.0-14.1 nm, which can be clearly seen from the profile line over the lamellae between A and B as shown in Figure 2b. Since the all-trans conformation is taken for the silicon backbone in PDHS crystals,36 the average contour length of the silicon backbone of the sample used here with Mn ) 1.2 × 104 is calculated to be 12.1 nm according to the standard bond lengths and angles for silicon.37 The consistency between the average thickness and average contour length of the silicon backbone likely indicates that the lamellar structures represent a monolayer with the extended silicon backbone perpendicular to the substrate surface, similar to that formed on carbon-coated substrate.30 This assertion is confirmed by WAXD in reflection mode and ED experiments in transmission mode. The WAXD taken in reflection mode and ED taken in transmission mode can reflect the crystalline faces parallel and per(36) Rabolt, J. F.; Hofer, D.; Miller, R. D.; Fickes, G. N. Macromolecules 1986, 19, 611. (37) Patnaik, S. S.; Farmer, B. L. Polymer 1992, 33, 4443.

Langmuir, Vol. 20, No. 8, 2004 3273

Figure 3. WAXD pattern (a) and ED pattern (b) of a PDHS ultrathin film (12.7 nm). WAXD pattern of bulk PDHS sample at room temperature is also shown in a.

pendicular to the substrate surface, respectively. The WAXD and ED patterns of the PDHS ultrathin film with the same thickness as 12.7 nm are shown in Figure 3a and b, respectively. Obviously, there is no diffraction peak observed in this WAXD pattern. However, many diffraction peaks are observed for bulk samples (also shown in Figure 3a), which allowed an assignment of the solidstate structure of PDHS.38 An orthorhombic unit cell with a ) 1.376, b ) 2.386, and c ) 0.399 nm monoclinic unit cell was proposed as the structure of crystallized PDHS.37 Compared with the abundant diffraction peaks of the bulk sample, the disappearance of diffraction peaks in the ultrathin film sample indicates that the (001) plane in this ultrathin film sample is parallel to the surface of substrate, since nearly all the sharp diffraction peaks of bulk sample correspond to the (hk0) reflections. The ED pattern shown in Figure 3b confirms this. Many orders of diffractions are observed in both directions, which indicate the high crystallographic order in these crystals. According to the structure of PDHS,37 all diffraction dots can be recognized as the (hk0) projection of the reciprocal lattice. Therefore, the lamellar structures observed in Figure 2a are chain-extended flat-on lamellar crystals with the silicon backbone perpendicular to the substrate surface. Frank et al.2,24,25 showed that the crystallization of PDHS is substantially hindered in ultrathin films, in which a critical thickness of 15 nm is needed for crystalline morphology to exist. Obviously, their conclusions are (38) Lovinger, A. J.; Schilling, F. C.; Bovey, F. A.; Zeigler, J. M. Macromolecules 1986, 19, 2657.

3274

Langmuir, Vol. 20, No. 8, 2004

Hu et al.

inconsistent with what we described above. To interpret the difference between their conclusions and our results, the following three aspects should be taken into account. First is the effect of physical structure and chemical composition of the substrate. The substrates used in their experiments were quartz, and in our case it is a silicon wafer. We note that the surfaces of both substrates are chemically similar, i.e., each substrate has many hydroxyl groups on the surface.24 Therefore, the effects of physical structure and chemical composition of the substrates can be excluded. Second is the molecular weight effect. Though the molecular weight of PDHS in their experiments (Mw ) 2.6 × 106, Mw/Mn ) 2.4) is much higher than that in our case (Mn ) 1.2 × 104, Mw/Mn ) 1.4), they also found39 that the result of low molecular weight material (Mw ) 4.3 × 104) was similar to that obtained with the high molecular weight material. Third is the experiment method. The conclusions Frank et al. drew2,24,25 are based on the fact that PDHS thick films show two strong UV absorption maxima, one around 364 nm and the other around 316 nm. The low-energy peak (364 nm) corresponds to an extended all-trans conformation of the silicon backbone, and the high-energy absorption (316 nm) arises from a conformationally disordered backbone in a presumably helical conformation. The low-energy absorption is correlated with the presence of a crystalline phase, while the low-energy one corresponds to a disordered phase.38 Therefore, they estimated the sample crystallinity from the relative absorption of the all-trans and disordered segments in the sample. However, they only carried out the UV absorption experiments with light propagating along the film normal. Since UV absorptions of polysilanes are generally attributed to the σ-electron delocalization along the silicon backbone, no absorption should occur with the normal incident beam if the silicon backbones align perpendicularly to the substrate surface (see below). As a result, the sample crystallinity cannot be determined from the relative UV absorbance in the thin film sample with different molecular chain orientations. Our results discussed above not only prove that PDHS thin film with an average thickness of 12.7 nm crystallizes at room temperature, but also indicate that the silicon backbones are perpendicular to the substrate in the thin film. We note that nearly all the crystallized polymer chains close to the polymer/substrate interface are perpendicular to the surface of the substrate.23,26-30 Even if the polymer chains cannot crystallize, Calvert also proposed20 that they are inclined to be stretched out straight into the films. We interpret these phenomena as due to the following reasons. It has been reported40,41 that the presence of a substrate in contact with dilute polymer solutions reduces the entropy of polymer chains near the surface of substrate. Populations of polymer chains near the surface of substrate are thus depleted due to thermodynamics. As a consequence, the polymer chains will, in general, tend to partition into the bulk solution unless there is a strong attractive force between the surface of the substrate and polymer.40,41 As a result, the concentration of chain ends will be increased at the interface between the substrate and the polymer solution. Therefore, if the polymer chains in a very dilute solution can rapidly crystallize due to rapid solvent evaporation, it is possible to orient them perpendicularly on the surface of the substrate. Therefore,

a monolayer with vertically oriented silicon backbones is formed after a dilute solution of PDHS was spin-coated onto the silicon wafer surface. This is also in accord with the enrichment of the polymer/substrate and polymer/air interfaces with polymer chain ends in amorphous polymer ultrathin films,6,42 which leads to reduction of the glasstransition temperature9-18 and molecular mobility19-22 with respect to bulk sample. Lamellar Crystal Morphology and Chain Orientation in Ultrathin Films Thicker than Monolayer. When polymer chains crystallize on lamellae with vertically oriented polymer chains, for instance, in the case of melted crystallized extended chain poly(tetrafluoroethylene),43 chain ends or fold surfaces (crystallographic (001) plane) of the lamellae provide a rather special surface for further growth. The (001) surface provides support for a nucleus only with little or no crystallographic orientation to the parent lamellae.43 The chainlike geometry of macromolecules compels polymer chains to preferentially orient on the (001) plane. If nucleation sites are of sufficient concentration, the resulting morphology is thus a stack of lamellae growing close to perpendicular to the parent flat-on lamellae. The AFM height image shown in Figure 4a indicates that sheaf-like bundle structures are produced in the ultrathin film with a thickness of 45.0 nm produced by spin coating from solution with a concentration of 0.5% (w/v) onto the surface of silicon wafer. A higher resolution AFM phase image shows that the sheaf-like bundle structures are made up of narrow lamellae (Figure 4b). The profile line between A and B (in Figure 4b) shown in Figure 4c indicates that the width of the lamellae is ca. 11.96 nm. This stacked lamellar morphology has been occasionally observed in extended chain polyethylene44 and frequently observed in extended chain poly(tetrafluoroethylene)43 due to the similar nucleation process as discussed above. To determine the structure of the thicker ultrathin films, WAXD measurements of the PDHS ultrathin films with various thicknesses are performed and shown in Figure 5. For the thinner ultrathin films, i.e., the film thickness is 12.7 and 13.2 nm, no diffraction peak is observed. On the contrary, for the thicker ultrathin films, i.e., those films thicker than 23.1 nm, two strong peaks centered at 7.5° and 15.0° as well as a weaker peak centered at 22.5° are observed in the WAXD pattern. According to the crystalline structure of bulk PDHS,37 these three peaks correspond to the first-, second-, and third-level diffraction, respectively, from the (110) and (020) planes. Compared with the diffraction pattern of bulk sample shown in Figure 3a, the disappearance of other diffraction peaks in the thicker ultrathin films indicates that the (110) or (020) plane in the film sample of these thicknesses is parallel to the surface of the substrate. That means the silicon polymer backbones in the thicker ultrathin films are parallel to the surface of substrate. Therefore, the narrow lamellae found in the thicker ultrathin film with a thickness of 45.0 nm, shown in Figure 4b, are edge-on lamellar crystals. The following problem is whether the silicon backbone is parallel to the narrow lamella or perpendicular to it. Since the average width of the lamellae is 11.96 nm and consistent with the average contour length of the silicon backbone, it is reasonable to conclude that the silicon backbone is perpendicular to the long axis direction of edge-on lamellae with extended chains.

(39) Despotopoulou, M. M.; Miller, R. D.; Rabolt, J. F.; Frank, C. W. J. Polym. Sci., Part B: Polym. Phys. 1996, 34, 2335. (40) Schuetjens, J. M. H. M.; Fleer, G. J. J. Chem. Phys. 1979, 71, 1619. (41) DiMarzio, E. A.; Rubin, R. J. J. Chem. Phys. 1971, 55, 4318.

(42) Tanaka, K.; Taura, A.; Ge, S.-R.; Takahara, A.; Kajiyama, T. Macromolecules 1996, 29, 2040. (43) Melillo, L.; Wunderlich, B. Kolloid-Z. u. Z. Polym. 1972, 250, 417. (44) Wunderlich, B.; Melillo, L. Makromol. Chem. 1968, 118, 250.

Ultrathin Poly(di-n-hexylsilane) Films

Langmuir, Vol. 20, No. 8, 2004 3275

Figure 5. WAXD pattern of the PDHS ultrathin films with various thicknesses.

Figure 4. AFM height image (a) and phase image (b) of a PDHS ultrathin film with a thickness of 45.0 nm prepared by spin coating from solution with a concentration of 0.5% (w/v) onto the surface of a silicon wafer. (c) Profile line of the cross section between A and B shown in b.

On the basis of the above discussion, the edge-on lamellae should grow on the basis of flat-on lamellae. However, the above WAXD results only indicate the existence of edge-on lamellae. To further determine the structure of the thicker ultrathin films, the PDHS film with a thickness of 45.0 nm is rubbed with nylon cloth in a fixed direction at room temperature and then observed with TEM. The bright-field image of the film (Figure 6a) clearly shows that the edge-on lamellae are oriented perpendicular to the rubbing direction. The electron diffraction pattern (Figure 6b) clearly shows the coexistence of diffraction arcs and dots. The diffraction arcs are similar to the X-ray diffraction pattern of uniaxially oriented PDHS fibers at room temperature.38 According to the crystalline structure of PDHS,37 the diffraction arcs along the rubbing direction correspond to the (hk1) diffraction. Therefore, it is reasonable to conclude that the silicon backbones are oriented along the rubbing

Figure 6. TEM bright-field image (a) and ED pattern (b) of a rubbed PDHS ultrathin film with thickness of 45.0 nm.

direction and perpendicular to the lamellae. These results also indicate that the lamellae are edge-on crystals with extended silicon backbone. In addition, the ED pattern in

3276

Langmuir, Vol. 20, No. 8, 2004

Hu et al.

Figure 8. (a) Schematic illustration of the relationship between the UV probe beam and structure of PDHS ultrathin film with silicon backbone perpendicular to the surface of substrate. (b) UV absorption spectra of PDHS ultrathin film with a thickness of 12.7 nm at various θ* angles between the incident UV beam direction and surface-normal direction of the ultrathin film.

Figure 7. AFM images of rubbed a PDHS ultrathin film (45.0 nm). (a) Height image. The contrast in the height image covers the surface corrugations in the 0-50.0 nm range. (b) Higher resolution phase image over the rectangle area marked in a. (c) Profile line of the cross section between a and b shown in a.

Figure 6b also shows the same (hk0) diffraction dots as observed for monolayer flat-on lamellar crystals with the silicon backbone perpendicular to the surface of substrate, indicating that a layer with the same structure as the monolayer in Figure 2 still exists. Figure 7 shows the AFM height images of the rubbed PDHS film. The basal flat-on lamellae can be clearly seen in the AFM height image of rubbed thin film as shown in Figure 7a. The region marked C is the naked substrate. On the basis of the substrate, flat-on lamellae are clearly observed (region B). The profile line of the cross section between a and b shown in Figure 7a shows that the height of flat-on lamellae is consistent with the contour length of the silicon backbone with nearly an all-trans conformation. The region marked A is the edge-on lamellae, which is on the basis of flat-on lamellae. A higher resolution phase image of the rectangular area in Figure 7a is shown in Figure 7b, which clearly indicates the edge-on lamellae are

oriented perpendicular to the rubbing direction. These results clearly show that edge-on lamellae are formed on the sub-flat-on lamellae in the thicker PDHS ultrathin films. The similar phenomena that flat-on lamellae are formed near the polymer/substrate interface and edge-on lamellae are grown on the flat-on lamellae were observed in poly(ethylene terephthalate)45 and recently in poly(ethylene oxides) ultrathin films.28,29 Influence of Chain Orientations on the UV Absorption Property. Since the existence of strong UV absorptions is a primary driver for the development of polysilanes as functional materials, the UV absorption spectra of the PDHS films described above which have silicon backbones perpendicular or parallel to the surface are measured. According to the principle of optical absorption,46 the absorbance is proportional to M2E2 cos2 θ, where M is the transition moment of conjugated molecule, E the electric filed of incident UV beam, and θ the angle between the electric field and transition moment. Since the transition moment of a PDHS molecule lies along the silicon backbone and the electric field of the UV probe beam is normal to the propagation direction, the absorbance of a PDHS thin film will vary with the angle of incidence θ*, i.e., the angle between the incident UV beam direction and surfacenormal of ultrathin film. In this work, reflection UV (45) Sakai, Y.; Imai, M.; Kaji, K.; Tsuji, M. Macromolecules 1996, 29, 8830. (46) Jaffe´, H. H.; Orchin, M. Theory and applications of ultraviolet sepectroscopy; John Wiley & Sons: New York, 1962.

Ultrathin Poly(di-n-hexylsilane) Films

Figure 9. (a) Schematic illustration of the relation between the UV probe beam and structure of PDHS ultrathin film with silicon backbone parallel to the surface of substrate. (b) UV absorption spectra of PDHS ultrathin film with a thickness of 45.0 nm at various θ* angles between the incident UV beam direction and surface-normal direction of the ultrathin film.

absorption measurements are carried out with θ* ranging from 20° to 70°. When the silicon backbone is perpendicular to the surface of the substrate, the angle θ between the electric field of the incident UV beam and transition moment of the PDHS molecules is equal to 90° - θ*, i.e., θ ) 90° θ*, as shown schematically in Figure 8a. Figure 8b shows the UV absorption spectra of the thinner PDHS ultrathin film with a thickness of 12.7 nm with silicon backbone perpendicular to the substrate. The absorption spectrum at θ* ) 20° is characterized by a strong absorption at 314 nm and a small one at 342 nm. It has been suggested47 that the short wavelength peak arises from a conformationally disordered silicon backbone in a presumably helical conformation and the long wavelength peak corresponds to an extended all-trans conformation with the side chains largely extended but not completely alltrans. However, multiple long wavelength absorption peaks have been reported for PDHS. Kyotani and coworkers showed that cast films of this polymer, dried under (47) Miller, R. D.; Hofer, D.; Rabolt, J. F.; Fickes, G. N. J. Am. Chem. Soc. 1985, 107, 2172.

Langmuir, Vol. 20, No. 8, 2004 3277

different conditions, produce UV maxima at 357, 370, or 375 nm.48 Matsumoto et al. attributed the different long wavelength absorption peaks to the effects of kinks in the silicon backbone.49 The backbone chains have many kinks, and the segments vary greatly in length. As the kink content decreases, the long wavelength absorption peak shifts to higher energy but the integrated intensity remains constant. According to Matsumoto’s calculation,49 the fraction of Si-Si bones in trans-link is 0.97. Nevertheless, the trends of the absorbance accompanied with θ* are obvious. The UV absorption spectra shown in Figure 9b indicate a stronger absorption at 342 nm at larger θ*. That not only proves that silicon backbone in this sample is perpendicular to the surface of the substrate, but also indicates that the UV absorption can be controlled by adjusting the angle between the incident UV beam direction and normal direction of the ultrathin film. In contrast, in the case of an ultrathin film with silicon backbones parallel to the surface of the substrate, the angle θ between the electric field and transition moment is equal to θ*, i.e., θ ) θ*, as shown schematically in Figure 9a. The UV absorption spectra of the ultrathin film of thickness of 45.0 nm shown in Figure 9b indicate stronger absorbance at 342 nm at smaller θ* angles. Although a monolayer with the silicon backbone perpendicular to the substrate in this ultrathin film exists, the thickness of the monolayer is only one-fourth that of the film. UV absorbance in Figure 9b arises mainly for the silicon backbones oriented parallel to the surface of the substrate. The results not only confirm the existence of silicon backbones parallel to substrate in the thicker film, but also again indicate UV absorption can be controlled by adjusting the angle between the incident UV beam direction and normal direction of the thin film. Conclusions The thickness dependence of molecular chain and lamellar crystal orientation in ultrathin PDHS film was investigated by a combination of WAXD, TEM, AFM, and UV absorption spectroscopy. Flat-on lamellar crystals with the silicon backbones perpendicular to the surface of substrate are produced in the monolayer ultrathin film of PDHS. When additive PDHS chains crystallize on the monolayer ultrathin film formed with the silicon backbone perpendicular to the surface of substrate, the chain ends provide a rather special surface for further growth of crystals. Edge-on lamellar crystals with the silicon backbones parallel to the surface of substrate are grown on the sub-flat-on lamellae in the thicker ultrathin films. The remarkable influence of orientation of silicon backbone on the UV absorption of PDHS films is observed due to the one-dimensional nature of σ-electron delocalized along the silicon backbone. Acknowledgment. This work was supported by the National Science Foundation of China. We would thank Prof. Chengji Shan at the Changchun Institute of Applied Chemistry for assistance during synthesizing the samples. LA036033K (48) Kyotani, H.; Shimomura, M.; Miyazaki, M.; Ueno, K. Polymer 1995, 36, 915. (49) Matsumoto, N. Jpn. J. Appl. Phys. 1998, 37, 5425.