Study of Alkyl Organic Monolayers with Different Molecular Chain

9962

Langmuir 2006, 22, 9962-9966

Study of Alkyl Organic Monolayers with Different Molecular Chain Lengths Directly Attached to Silicon Takahiro Ishizaki,*,† Nagahiro Saito,‡ Lee SunHyung,§ Kaoru Ishida,§ and Osamu Takai† EcoTopia Science Institute, Nagoya UniVersity, Furo-cho, Chikusa, Nagoya 464-8603, Japan, Department of Molecular Design and Engineering, Graduate School of Engineering, Nagoya UniVersity, Furo-cho, Chikusa, Nagoya 464-8603, Japan, and Department of Materials, Physics and Energy Engineering, Graduate School of Engineering, Nagoya UniVersity, Furo-cho, Chikusa, Nagoya 464-8603, Japan ReceiVed August 27, 2005. In Final Form: August 1, 2006 Alkyl organic monolayers with different alkyl molecular chain lengths directly attached to silicon were prepared at 160 °C from 1-decene (C10), 1-dodecene (C12), 1-tetradecene (C14), 1-hexadecene (C16), and 1-octadecene (C18). These monolayers were characterized on the basis of water contact angle measurement, ellipsometry, X-ray reflectivity (XR), X-ray photoelectron spectroscopy (XPS), and grazing incidence X-ray diffraction (GIXD) to elucidate the effect of the molecular chain length on the molecular arrangement and packing density of the monolayers. Water contact angle and XPS measurements showed that C12, C14, and C16 monolayers have a comparably higher quality, while the quality of C10 and C18 monolayers is worse. GIXD revealed that the alkyl monolayers directly attached to the Si were all amorphously structured regardless of their alkyl chain length. The amorphous structure of the alkyl monolayers could be attributed to the rigid Si-C bonding, low quality of hydrogen-terminated silicon substrate, and/or low mobility of physisorbed molecules.

Introduction There has been growing interest in organic monolayers directly attached to silicon through Si-C bonds due to their application to semiconductor technology.1,2 Such organic monolayers offer the potential for effective electron transfer at the silicon/organic layer interface, since there is no insulating layer, that is, no silicon oxide layer.3-5 It has been reported that organic monolayers can be employed for silicon surface passivation6,7 and for the incorporation of biochemical functionality at interfaces.8,9 In particular, among the several types of organic monolayers, alkyl monolayers directly attached to silicon are promising as resist films for nanofabrication.10,11 Moreover, since alkyl monolayers function as insulator layers,12,13 they are expected to be applied to organic field-effect transistors (OFET). It is vital for the practical application of OFETs to control the molecular arrangement and packing density of the organic monolayer directly attached to the silicon substrate.14 Thus, it is necessary to understand in detail the structure of such * Corresponding author. E-mail: [email protected]. † EcoTopia Science Institute. ‡ Department of Molecular Design and Engineering. § Department of Materials, Physics and Energy Engineering. (1) Sze, S. M. Physics of Semiconductor DeVices, 2nd ed.; Wiley-Interscience: New York, 1981. (2) Buczkowski, A.; Radzimski, Z. J.; Rozgonyi, G. A.; Shimura, F. J. Appl. Phys. 1991, 69, 6495. (3) Sieval, A. B.; Linke, R.; Zuilhof, H.; Sudho¨ter, E. J. R. AdV. Mater. 2000, 12, 1457. (4) Bansal, A.; Lewis, N. S. J. Phys. Chem. B 1998, 102, 4058. (5) Bansal, A.; Li, X.; Zi, S. I.; Weinberg, W. H.; Lewis, N. S. J. Phys. Chem. B 2001, 105, 10266. (6) Bansal, A.; Lewis, N. S. J. Phys. Chem. B 1998, 102, 1067. (7) Royea, W. J.; Juang, A.; Lewis, N. S. Appl. Phys. Lett. 2000, 77, 1988. (8) Strother, T.; Cai, W.; Zhao, X.; Hamers, R. J.; Smith, L. M. J. Am. Chem. Soc. 2000, 122, 1205. (9) Strother, T.; Hamers, R. J.; Smith, L. M. Nucleic Acids Res. 2000, 28, 3535. (10) Ara, M.; Graaf, H.; Tada, H. Appl. Phys. Lett. 2002, 80, 2565. (11) Sondaghuethorst, J.; Vanhelleputte, H.; Fokkink, L. Appl. Phys. Lett. 1994, 64, 285. (12) Kobayashi, S.; Nishikawa, T.; Takenobu, T.; Mori, S.; Shimoda, T.; Mitani, T.; Shimotani, H.; Yoshimoto, N.; Ogawa, S.; Iwasa, Y. Nat. Mater. 2004, 3, 317. (13) Ishii, H.; Sugiyama, K.; Ito, E.; Seki, K. AdV. Mater. 1999, 11, 605.

monolayers through characterization by various experimental techniques. Nonetheless, there is still little information available on the properties of such films. In the case of alkyl monolayers on hydrogen-terminated silicon surfaces, the possible structure of the assembled molecules has been inferred from theoretical studies15,16 and experimental results17-20 obtained by infrared (IR) spectroscopy, ellipsometry, and X-ray reflectivity. Sieval et al. reported that the maximum coverage of alkyl molecules on Si(111) is about 0.5-0.55 on the basis of molecular modeling simulation.15,16 Bansal et al. demonstrated by XPS and ellipsometric measurements that the thickness of an alkyl monolayer varied monotonically with the length of the alkyl group in the reactant.6 However, no attempt has yet been made to directly investigate the effect of the molecular chain length on the molecular arrangement of alkyl monolayers directly attached to silicon by grazing incidence X-ray diffraction (GIXD). In this study, alkyl monolayers with different molecular chain lengths were prepared on hydrogen-terminated Si substrates by the radical-initiated reaction of Si-H bonds with olefins. Their molecular arrangements were investigated on the basis of grazing incidence X-ray diffraction (GIXD), X-ray reflectivity (XR), X-ray photoelectron spectroscopy (XPS), ellipsometry, and water contact angle measurement. Experimental Section As raw materials for alkyl monolayers with different molecular chain lengths, 1-decene (C10; Aldrich, 94%), 1-dodecene (C12; (14) Park, J. S.; Vo, A. N.; Barriet, D.; Shon, Y-.S.; Lee, T. R. Langmuir 2005, 21, 2902. (15) Sieval, A. B.; Hout, B.; Zuilhof, H.; Sudho¨ter, E. J. R. Langmuir 2000, 16, 2987. (16) Sieval, A. B.; Hout, B.; Zuilhof, H.; Sudho¨ter, E. J. R. Langmuir 2001, 17, 2172. (17) Linford, M. R.; Chidsey, C. E. D. J. Am. Chem. Soc. 1993, 115, 12631. (18) Linford, M. R.; Fenter, P.; Eisenberger, P. M.; Chidsey, C. E. D. J. Am. Chem. Soc. 1995, 117, 3145. (19) Sieval, A. B.; Demirel, A. L.; Nissink, J. W. M.; Lindord, M. R.; van der Maas, J. H.; de Jeu, W. H.; Zuilhof, H.; Sudho¨ter, E. J. R. Langmuir 1998, 14, 1759. (20) Effenberger, F.; Gotz, G.; Bidlingmaier, B.; Wezstein, M. Angew. Chem., Int. Ed. Engl. 1998, 37, 2462.

10.1021/la052342u CCC: $33.50 © 2006 American Chemical Society Published on Web 10/21/2006

Alkyl Organic Monolayers Directly Attached to Si

Langmuir, Vol. 22, No. 24, 2006 9963

Figure 1. Schematic diagram of the reaction between 1-alkenes and a hydrogen-terminated Si surface. Aldrich, 95%), 1-tetradecene (C14; Sigma, 99%), 1-hexadecene (C16; Tokyo Kasei, 99%), and 1-octadecene (C18; Fulka, 97%) were used as received. The alkyl monolayers were prepared from each raw material on p-type Si(111) wafers with electrical resistivity of 1020 Ω cm. First, the silicon substrates were cleaned in acetone, ethanol, and ultrapure water, in that order. Next, the silicon substrates were cleaned in a piranha solution (H2SO4:H2O2 ) 3:1) at 100 °C for 10 min,17 then rinsed thoroughly with ultrapure water (18.2 MΩ). After rinsing, the oxidized silicon substrates were dried with N2 gas and immersed in a 40% aqueous ammonium fluoride solution (NH4F) at room temperature for 15 min in order to remove the native silicon oxide layer, resulting in hydrogen-termination of the surface.21 Each hydrogen-terminated Si substrate was placed in a solution containing one of the raw materials for 5 h at the temperature of 160 °C. Alkyl monolayers with different molecular chain lengths were formed on the silicon substrates through Si-C bonds.22-26 Figure 1 shows a schematic reaction diagram of this synthesis. Alkyl monolayer formation was confirmed by X-ray photoelectron spectroscopy (Shimadzu-Kratos, AXIS), water contact angle measurement (KRUSS, DSA10-MK2), and ellipsometry (Philips, PZ 2000). XPS was performed with Al KR (1486.3 eV) radiation operating at 10 mA and 12 kV with a takeoff angle of 30°. All binding energies were referenced to the Si 2p peak of bulk silicon (99.3 eV). Water contact angles were collected at room temperature under ambient humidity. The values at five different points on each sample were averaged. Film thickness was estimated by ellipsometry and X-ray reflectivity (XR) (Rigaku, ATX-G). The ellipsometer was equipped with a HeNe laser (λ ) 632.8 nm) and a 45° polarizer. The incident angle of the laser was 70° from the surface normal. The refraction index and the absorption coefficient for the silicon substrate were 3.865 and -0.020, respectively. The respective constants for silicon oxide were 1.465 and 0, which were applied to those of the alkyl monolayers.27 In XR, the angular dependence of the specular reflectivity was measured by a series of RRi-2RRi scans, where RRi indicates the incident angle of the monochromatic X-ray beam.28,29 The reflected beam was detected by a scintillation counter through a receiving slit. GIXD (Rigaku, ATX-G) was conducted to evaluate the molecular arrangement of the different alkyl monolayers. Each sample was placed on the sample stage of a three-axis goniometer for surface X-ray diffraction. The sample surface was then irradiated by a monochromatic X-ray beam at an incident angle of 0.2°.30,31 The wave vector is defined as qxy () 4π(sinθ)/λ), where θ and λ are the angle and the wavelength of the X-ray (Cu KR, λ ) 0.154 nm), respectively. The surfaces of the hydrogen-terminated Si(111) substrate and the alkyl monolayers directly attached to the Si(111) substrate were (21) Saito, N.; Youda, S.; Hayashi, K.; Sugimura, H.; Takai, O. Surf. Sci. 2003, 532, 970. (22) Hurley, P. T.; Ribbe, A. E.; Buriak, J. M. J. Am. Chem. Soc. 2003, 125, 11334. (23) Buriak, J. M.; Allen, M. J. J. Am. Chem. Soc. 1998, 120, 1339. (24) Boukherroub, R.; Morin, S.; Bensebaa, F.; Wayner, D. D. M. Langmuir 1999, 15, 3831. (25) Boukherroub, R.; Morin, S.; Sharpe, P.; Wayner, D. Langmuir 2000, 16, 7429. (26) Munford, M. L.; Maroun, F.; Cortes, R.; Allongue, P.; Pasa, A. A. Surf. Sci. 2003, 537, 95. (27) In Annual Book of ASTM Standards; 1990; p F 576. (28) Tidswell, I. M.; Ocko, B. M.; Pershan, P. S.; Wasserman, S. R.; Whitesides, G. M. Phys. ReV. B 1990, 41, 1111. (29) Kojio, K.; Takahara, A.; Omote, K.; Kajiyama, T. Langmuir 2000, 16, 3932. (30) Kojio, K.; Takahara, A.; Kajiyama, T. Bull. Chem. Soc. Jpn. 2001, 74, 1397. (31) Takahara, A.; Kojio, K.; Kajiyama, T. Ultramicroscopy 2002, 91, 203.

Figure 2. XPS Si 2p spectra of Si sample surfaces obtained (a) before and (b) after NH4F treatment. (c) Topographic image of the hydrogen-terminated Si surface. observed in air with an atomic force microscope (AFM; Seiko Instruments, SPA-300HV+SPI-3800N). AFM images were acquired at the scan rate of 1 Hz in contact mode using a sharpened silicon nitride tip with a force constant of 0.08 N/m and a resonance frequency of 34 kHz (Seiko Instruments Inc., Micro Cantilever, type SI-AF01).

Results and Discussion Figure 2, parts a and b, shows XPS Si 2p spectra of silicon substrate surfaces before and after immersion in a 40% ammonium fluoride aqueous solution, respectively. In the spectrum of the sample surface before immersion, a peak attributed to silicon dioxide was observed at the binding energy of 104 eV. On the other hand, no or little peak attributed to silicon dioxide was observed in the spectrum after immersion. This indicates that the silicon dioxide layer on the Si substrate had been almost removed. Figure 2c shows an AFM topographic image of the Si substrate surface confirming terrace and step structures of silicon. The in-plane intervals of the structures were 180 nm, and the difference in height between adjacent terraces was 3.2 ( 0.3 Å. Densely packed methyl-terminated monolayers generally show water contact angles greater than 110°. The water contact angle becomes smaller with decreasing packing density.17,18,32,33 Thus, the water contact angle strongly depends on the molecular structure of the monolayer. Figure 3 shows the water contact angles on the alkyl monolayers as a function of the number of carbon atoms in their alkyl chains, n. In the present case, the alkyl chain length is proportional to the number of carbon atoms. Henceforth, the respective monolayers are referred to by their (32) Porter, M. D.; Bright, T. B.; Allara, D. L.; Chidsey, C. E. D. J. Am. Chem. Soc. 1987, 109, 3559. (33) Laibinis, P. E.; Whitesides, G. M.; Allara, D. L.; Tao, Y. T.; Parikh, A. N.; Nuzzo, R. G. J. Am. Chem. Soc. 1991, 113, 7152.

9964 Langmuir, Vol. 22, No. 24, 2006

Ishizaki et al.

Figure 3. Static contact angles of water on the alkyl monolayers as a function of the linear alkyl molecular chain length, n. n indicates the number of carbon atoms in the alkyl monolayer (CnH2n+1).

number of carbon atoms, that is, as C10, C12, C14, C16, and C18 monolayers. The water contact angles of the C12, C14, and C16 monolayers were greater than 110°, while those of the C10 and C18 monolayers were less than 110°. This indicates that the packing density of the C10 and C18 monolayers was lower than that of the C12, C14, and C16 monolayers. In studies on alkylthiol monolayers on gold and alkyl monolayers prepared from radical reaction of 1-alkenes on (111)-oriented silicon, it has been reported5,17,18 that the alkyl monolayers with a close packing structure had water contact angles in the range of 110-115°. The water contact angles of the C12, C14, and C16 monolayers agreed well with this report. This indicates that the surface of the C12, C14, and C16 monolayers was densely covered with methyl groups. XPS scans of the monolayers with different alkyl chain lengths showed signals originating from Si, C, and O. The C 1s/Si 2p ratio monotonically increased with increasing alkyl chain length. XPS spectra were acquired in narrow ranges to obtain more detailed information on the alkylated silicon surfaces. Figures 4 and 5 show XPS C 1s and Si 2p spectra, respectively. The XPS C 1s spectra showed a peak located at 285.0 eV, which is assigned to C-H34 bonding. There was an inflection point at 283.8 eV in the spectra, indicating a peak at this binding energy. This peak may correspond to a chemical bonding state originating from Si-C bonds.35 The Si 2p spectra of the C10-C16 monolayers were nearly identical to that of the hydrogen-terminated Si(111) surface. Although silicon oxide was hardly detected in the narrow spectra of these monolayers, little silicon oxide might be formed on the Si(111) surface. However, the C 1s spectra revealed that the hydrogen-terminated Si(111) surface was chemically passivated by the C10-C16 alkyl monolayers. These results provide strong evidence that the alkylation of the hydrogen-terminated silicon substrate was accomplished. In contrast, two peaks were clearly observed in the Si 2p spectrum of the C18 monolayer. One was a strong peak at 99.3 eV attributed to metallic silicon, while the other was a weak peak at 103.4 eV attributed to silicon oxide. This indicates that silicon oxide formed on the silicon substrate. The presence of this silicon oxide would inhibit the formation of the alkyl monolayer and may have led to the lower packing density of the C18 monolayer. From another respect, the reaction temperature of 160 °C was not high enough to prepare a highly packed C18 monolayer. The water contact angle and XPS results indicate that C12, C14, and C16 monolayers have a comparably higher quality, while the quality of C10 and C18 monolayers is worse. (34) Saito, N.; Youda, S.; Hayashi, K.; Sugimura, H.; Takai, O. Chem. Lett. 2002, 31, 1194. (35) Muehlhoff, L.; Choyke, W. J.; Bozack, M. J.; Yates, J. T., Jr. J. Appl. Phys. 1986, 60, 2842.

Figure 4. XPS C 1s spectra of alkyl monolayers directly attached to Si obtained by XPS.

Table 1 shows ellipsometrically determined thicknesses of the alkyl monolayers. The film thicknesses monotonically increased with increasing alkyl chain length. The film thicknesses of the C10-C18 monolayers were 1.2, 1.3, 1.5, 1.6, and 1.9 nm, respectively. We also calculated the molecular length by semiempirical molecular orbital (MO) calculation using the AM1 Hamiltonian. Here, the calculation was made on (CH3)(CH2)n-1SiH3 for simplicity. The molecular chain length was defined as the distance between the Si atom and the C atom of the methyl group. The film thicknesses were obtained by considering both the molecular chain lengths and the chain tilt angles. These calculations showed that the film thicknesses of the C10, C12, C14, C16, and C18 monolayers were 1.0, 1.2, 1.4, 1.6, and 1.8 nm, respectively. The measured film thicknesses were greater than the calculated ones. However, this difference is very reasonable since it is likely due to contributions from adventitious carbon and the chain tilt angle. We applied a value of 1.465 as the refractive index to the alkyl monolayers. However, the refractive index value might have been considerably influenced by both the surface coverage and the chain length. Thus, it was necessary to confirm the measured ellipsometric thicknesses with other analytical data. We therefore measured the X-ray reflectivity (XR) of the monolayers to precisely determine their film thickness. XR measurements are intrinsically more precise than ellipsometry, since the X-ray probe has a wavelength comparable to the monolayer thickness. Figure 6 shows the XR profiles for a hydrogen-terminated silicon surface and for the alkyl monolayers with different chain lengths. The first minimum seen in the profiles would originate from destructive interference between X-ray reflections from the top and bottom regions of the alkyl

Alkyl Organic Monolayers Directly Attached to Si

Langmuir, Vol. 22, No. 24, 2006 9965

Figure 6. X-ray reflectivity profiles of (a) hydrogen-terminated Si and the alkyl monolayers directly attached to Si: (b) C10, (c) C12, (d) C14, (e) C16, and (f) C18.

Figure 5. XPS Si 2p spectra of alkyl monolayers directly attached to Si obtained by XPS. Table 1. Physical Properties of Alkyl-Terminated Si(111) Surfaces

-R

observed ellipsometric thickness (Å)

analytical thickness from XR measurement (Å)

calculated film thicknessa (Å)

-C10H21 -Χ12H25 -C14H29 -C16H33 -C18H37

12.4 ( 2 13.0 ( 1 15.2 ( 1 16.4 ( 1 19.2 ( 2

11.1 12.8 14.7 16.3 18.1

10.2 12.2 14.2 16.2 18.2

a

Film thickness was calculated on the basis of the assumption that the alkyl groups are robustly attached to the silicon surface and the Si-C bonds are normal to the surface.

monolayers. The value of qz for the first minimum in the XR curves was defined as qz,min. The film thickness, L, was calculated using the equation L ) π/qz,min.28 At grazing incidence, the reflectivity for all the samples was high, although it rapidly decreased as the angle between the surface and the incident X-ray beam increased. Unlike the smooth Fresnel-like decay in the XR profile of the hydrogen-terminated Si(111) surface, all the alkyl monolayers showed a local minimum in reflectivity at the qz of 1-3. Compared to the hydrogen-terminated Si substrate, drastic decays in reflectivity were observed with the alkyl monolayers. The film thicknesses of the C10-C18 monolayers were determined to be 1.11, 1.28, 1.47, 1.63, and 1.81 nm, respectively, as shown in Table 1. These values agree well with the ellipsometrically calculated ones. To roughly estimate the chain tilt angles of the alkyl monolayers, the thicknesses obtained through XR measurements

Figure 7. GIXD profiles for the alkyl monolayers directly attached to Si: (a) C10, (b) C12, (c) C14, (d) C16, and (e) C18.

were corrected for the contribution of adventitious carbon, assuming from the XPS results that it was constant at approximately 0.3 nm for the series of alkyl monolayers.5 The chain tilt angles of the C10-C18 monolayers derived from the corrected film thicknesses were 43.1, 40.0, 37.3, 35.3, and 33.5°, respectively. The predicted chain tilt angles on the basis of MO calculation were 41.6, 40.8, 40.3, 39.9, and 38.2°, respectively. The respective differences were less than 5°. Assumptions made in the estimations would be taken into consideration; however, all the tilt angles for the alkyl monolayers derived from the XR data were very close to those reported derived from IR measurements.18 GIXD measurements were performed to evaluate the in-plane arrangement of the alkyl monolayers. Figure 7 shows GIXD patterns of the alkyl monolayers with different alkyl chain lengths. If the alkyl monolayer has a periodic molecular arrangement, a peak corresponding to that periodic arrangement would appear on its GIXD pattern at the wave vector qxy between 14 and 16 nm-1. However, no peak related to periodic arrangement was observed in the pattern of any of the alkyl monolayers regardless

9966 Langmuir, Vol. 22, No. 24, 2006

of the alkyl molecular chain length. This indicates that all the alkyl monolayers directly attached to the Si substrate had amorphous structure. This is most likely due to low in-plane orientational order, that is, the presence of defects in the alkyl monolayer directly attached to the Si substrate, since such defects inhibit the formation of a lattice with long-range periodicity. Matthew et al. reported that actual organic monolayers could have a significant number of such defects.18 In addition, the low orientational order would lead to low conformational order, resulting in the presence of gauche defects. These structural defects could arise from two factors as follows: (i) a low quality of the H-Si substrate, initially having a high inhomogeneity and a high density of defects, and (ii) a low mobility of the physisorbed molecules, which cannot assemble in a proper way after they get a grip onto the substrate. In addition, the van der Waals radius of a methylene unit is sufficiently large. Thus, packing an alkyl chain onto every Si atom would be highly unlikely due to the steric hindrance. On the other hand, self-assembled monolayers (SAMs), such as octadecyltrichlorosilane (OTS)-SAM, are known to be composed of a lattice with long-range periodicity. The Si-C bonds in alkyl monolayers are more rigid than the Si-O bonds in alkylsilane SAMs. Moreover the siloxane layer under the OTSSAM is amorphous with some degree of flexibility. Due to this flexibility, the formation of the OTS monolayer is influenced by van der Waals attraction between the alkyl chains. This is the origin of “self-assembling.” In the case of our alkyl monolayers, the structure at the interface between the organic monolayer and the silicon substrate was mainly determined by the Si-C-C bond angle. This bond angle remained approximately constant since it originated from the sp3 chemical bonding state. The results of our tilt angle evaluation agree with the molecular bond

Ishizaki et al.

angle. Thus, the agreement originates from the rigid Si-C-C bond. On the other hand, there are many gauche defects in the alkyl monolayers due to the rigid bonding, low quality of hydrogen-terminated silicon substrate, and/or low mobility of physisorbed molecules although the defects decrease with the increase of the number of carbon atoms. These factors, which lead to the structural defects, prevent the adsorbed molecules on the surface from self-assembling transforming the arrangements during the formation process unlike the organosilane SAMs.

Conclusion Alkyl monolayers with different molecular chain lengths from C10 to C18 were prepared at 160 °C on hydrogen-terminated silicon substrates through a radical-initiated reaction. The effect of the chain length on the alkyl monolayers was investigated by water contact angle measurement, ellipsometry, X-ray reflectivity, X-ray photoelectron spectroscopy, and grazing incidence X-ray diffraction. XPS and ellipsometry indicated that the thickness of the alkyl monolayers monotonically increased with increasing length of the alkyl chains. Water contact angle measurements indicated that the alkyl monolayers from C12 to C16 were very densely packed, whereas GIXD measurements showed no peak related to periodic arrangement for any of the alkyl monolayers. This disordered in-plane structure could be due to the rigid Si-C bonding, low quality of hydrogen-terminated silicon substrate, and/or low mobility of physisorbed molecules. Acknowledgment. This study was partially supported by the Ministry of Education, Science, Sports and Culture, Grant-inAid for Young Scientists (B) (No. 17760577), and the Aichi Science and Technology Foundation. LA052342U