Preparation of Submicron-Structured Alkylsiloxane Monolayers Using

Apr 2, 2004 - ... used to create oxide patterns on H-terminated Si(100) samples under ambient conditions. Depending on the laser power and the writing...
1 downloads 11 Views 72KB Size
Langmuir 2004, 20, 3525-3527

3525

Preparation of Submicron-Structured Alkylsiloxane Monolayers Using Prepatterned Silicon Substrates by Laser Direct Writing Thorsten Balgar, Steffen Franzka, Nils Hartmann,* and Eckart Hasselbrink Fachbereich Chemie, Universita¨ t Essen, Universita¨ tsstrasse 5, 45141 Essen, Germany Received January 12, 2004. In Final Form: February 26, 2004 A new constructive method for the preparation of laterally structured alkylsiloxane monolayers is demonstrated. Laser direct writing has been used to create oxide patterns on H-terminated Si(100) samples under ambient conditions. Depending on the laser power and the writing speed, oxide structures with a lateral resolution below 500 nm are prepared routinely. The patterned samples are suitable as temporary templates for the preparation of laterally structured octadecylsiloxane monolayers. Prior to immersion in an octadecyltrichlorosilane solution, however, hydration of the samples in water is essential to facilitate a selective coating of the oxidized areas. After coating, atomic force microscopy reveals the formation of octadecylsiloxane islands exclusively on top of the oxide lines.

Laterally structured self-assembled monolayers (SAMs) of organic molecules provide a versatile means for the design of complex surface architectures on the nanometer and micrometer length scale.1-5 Examples range from the spatial arrangement of biomolecules1,2 and the controlled growth of cells2,3 to the selective adsorption of nanoparticles4 and the fabrication of organic circuits.5 For these purposes, appropriate patterning techniques are needed. Previous approaches include studies using photolithography,2,6 microcontact printing (µCP),7 and various scanning beam8,9 and scanning probe techniques.10-12 Most techniques are based on a local degradation of the SAM on an initially homogeneously covered surface.2,6,8-10 The initiated processes are, however, in most cases rather complex and can result in an incomplete removal of the monolayer in the affected areas. In this respect, constructive methods generally appear to provide a more controlled and better defined procedure for the preparation of laterally structured SAMs.7,11-12 µCP, for example, allows for a direct deposition of self-assembled monolayers in * Corresponding author. E-mail: [email protected]. (1) Wadu-Mesthrige, K.; Xu, S.; Amro, N. A.; Liu, G.-y. Langmuir 1999, 15, 8580. (2) Singhvi, R.; Kumar, A.; Lopez, G. P.; Stephanopoulos, G. N.; Wang, D. I. C.; Whitesides, G. M.; Ingber, D. E. Science 1994, 264, 696. (3) Dulcey, C. S.; Georger, J. H.; Krauthamer, V.; Stenger, D. A.; Fare, T. L.; Calvert, J. M. Science 1991, 252, 551. (4) Qin, D.; Xia, Y.; Xu, B.; Yang, H.; Zhu, C.; Whitesides, G. M. Adv. Mater. 1999, 11, 1433. Liu, X.; Fu, L.; Hong, S.; Dravid, V. P.; Mirkin, C. A. Adv. Mater. 2002, 14, 231. (5) Gorman, C. B.; Biebuyck, H.; Whitesides, G. M. Chem. Mater. 1995, 7, 526. (6) Huang, J.; Dahlgren, D. A.; Hemminger, J. Langmuir 1994, 10, 626. Yang, X. M.; Peters, R. D.; Kim, T. K.; Nealey, P. F. J. Vac. Sci. Technol., B 1999, 17, 3203. (7) Kumar, A.; Whitesides, G. M. Appl. Phys. Lett. 1993, 63, 2002. Xia, Y.; Whitesides, G. M. Angew. Chem. Int. Ed. 1998, 37, 550. (8) Gillen, G.; Wight, S.; Bennett, J.; Tarlov, M. J. Appl. Phys. Lett. 1994, 65, 534. Ada, E. T.; Hanley, L.; Etchin, S.; Melngailis, J.; Dressik, W. J.; Chen, M.-S.; Calvert, J. M. J. Vac. Sci. Technol., B 1995, 13, 2189. (9) Tiberio, R. C.; Craighead, H. G.; Lercel, M.; Lau, T.; Sheen, C. W.; Allara, D. L. Appl. Phys. Lett. 1993, 62, 476. Lercel, M.; Craighead, H. G.; Parikh, A. N.; Seshadri, K.; Allara, D. L. Appl. Phys. Lett. 1996, 68, 1504. Go¨lzha¨user, A.; Geyer, W.; Stadler, V.; Eck, W.; Grunze, M.; Edinger, K.; Weimann, Th.; Hinze, P. J. Vac. Sci. Technol., B 2000, 18, 3414. (10) Lercel, M. J.; Redinbo, G. F.; Craighead, H. G.; Sheen, C. W.; Allara, D. L. Appl. Phys. Lett. 1994, 65, 974. Xu, S.; Miller, S.; Laibinis, P. E.; Liu, G.-y. Langmuir 1999, 15, 7244. (11) Maoz, R.; Cohen, S. R.; Sagiv, J. Adv. Mater. 1999, 11, 55. (12) Piner, R. D.; Zhu, J.; Xu, F.; Hong, S.; Mirkin, C. A. Science 1999, 283, 661.

selected areas on an initially bare substrate.7 Due to its simplicity and convenience, the technique gained remarkable popularity in the fabrication of laterally structured alkanethiolate monolayers on metallic substrates, such as gold and silver. In contrast to alkanethiolate monolayers, alkylsiloxane monolayers form on various oxidic substrates.13 The distinct set of addressable substrates and the unique robustness of these SAMs make them advantageous over alkanethiolate monolayers in applications requiring nonconducting substrates or a high chemical, thermal, and mechanical stability. Apart from µCP,7 though, only few inherently constructive methods can be used to prepare patterned alkylsiloxane monolayers.11 Moreover, the intricate growth mechanism and its sensitive dependence on various parameters can complicate the direct preparation of laterally structured alkylsiloxane monolayers using µCP.14 In this letter, we present a new constructive approach toward patterned alkylsiloxane monolayers. The method is based on a laser direct writing technique originally introduced by Mu¨llenborn and co-workers.15,16 Starting with an H-terminated Si(100) surface, this technique allows for the preparation of arbitrary oxide patterns on wafer-scale areas under ambient conditions. Depending on the wavelength, the incident laser power, and the writing speed, structures can be prepared with a lateral resolution ranging from several microns down to the 100 nm regime. The patterned substrates represent temporary templates, which provide an opportunity for the preparation of laterally structured alkylsiloxane monolayers. The laser direct writing setup consists of a 10 W continuous-wave argon ion laser (Innova 310, Coherent), an acousto-optic tunable filter (AOTF), a 4× Galilean beam expander, and a standard microscope objective. The AOTF (13) Ulman, A. Chem. Rev. 1996, 96, 1533. Schreiber, F. Prog. Surf. Sci. 2000, 65, 151. (14) Jeon, N. L.; Finnie, K.; Branshaw, K.; Nuzzo, R. G. Langmuir 1997, 13, 3382. Walheim, S.; Mu¨ller, R.; Sprenger, M.; Loser, E.; Mlynek, J.; Steiner, U.; Adv. Mater. 1999, 11, 1431. Pompe, T.; Fery, A.; Herminghaus, S.; Kriele, A.; Lorenz, H.; Kotthaus, J. P. Langmuir 1999, 15, 2398. (15) Mu¨llenborn, M.; Birkelund, K.; Grey, F.; Madsen, S. Appl. Phys. Lett. 1996, 69, 3013. (16) Laser direct writing is commonly used in materials processing. For an introduction and an overview of recent advances, see: Ba¨uerle, D. Laser Processing and Chemistry, 3rd ed.; Springer: Berlin, 2000. LIA Handbook of Laser Materials Processing; Ready, J. F., Farson, D. F., Eds.; Laser Institute of America: Orlando, 2001.

10.1021/la040006s CCC: $27.50 © 2004 American Chemical Society Published on Web 04/02/2004

3526

Langmuir, Vol. 20, No. 9, 2004

allows one to select the desired wavelength and switch the laser beam on and off. The objective is mounted on a high-resolution stepper motor which allows one to focus the laser beam onto the sample. Two equivalent stepper motors are used to translate the sample in the focal plane of the objective. Internal encoders ensure accurate positioning and scanning with a bidirectional repeatability of 100 nm over a travel range of 25 mm at scanning speeds up to 25 mm/s. The incident laser power on the samples has been measured using a thermal sensor from Coherent. The 1/e2 focal beam diameter has been determined using a knife-edge system in conjunction with a semiconductor sensor from Coherent. For patterning, H-terminated Si(100) samples are prepared by dipping the silicon substrates in 5% hydrofluoric acid at room temperature for about 2 min followed by drying in a stream of high-purity argon (Messer, 4.6). All samples shown here have been patterned in air. Prior to coating, the patterned samples are first hydrated in deionized water at 100 °C and dried in a stream of argon. Dissolved oxygen in the water was removed beforehand by purging with argon. After hydration, the substrates are immersed in a 2.5 mM octadecyltrichlorosilane (OTS) solution in toluene at room temperature, rinsed with toluene and ethanol, and dried in a stream of high-purity argon. The residual water content of the coating solution was adjusted to about 12 mM. Throughout the patterning and coating process, atomic force microscopy (AFM) has been used for inspection of the samples. AFM images were recorded in contact mode using either an Autoprobe CP Research with a 100 µm scanner or a Multimode/ Nanoscope IIIa with a 10 µm scanner (Veeco). To prepare laterally structured alkylsiloxane monolayers, H-terminated Si(100) substrates have been locally oxidized using 514 nm light and a microscope objective with a numerical aperture (NA) of 0.25. Figure 1a shows typical AFM images displaying the topographic contrast and the friction contrast after patterning. Depending on the experimental parameters, such as the incident laser power and the writing speed, the local oxidation also results in a topographic contrast between the oxidized and the remaining H-terminated areas. However, as indicated in Figure 1a the topographic contrast remains negligible at the conditions used here, that is, at comparatively low incident laser powers and high writing speeds. The oxide lines, though, can be clearly identified in the friction image of the same area. In the next step, the patterned substrates are used as temporary templates for the preparation of laterally structured octadecylsiloxane monolayers upon immersion into a millimolar OTS solution. A selective coating of the oxidized areas can be expected for two reasons: First, various investigations indicate that the formation of alkylsiloxane monolayers requires the presence of a thin water layer on the substrate surface.17 Considering the hydrophobic nature of the H-terminated areas on the patterned substrate, it appears likely, however, that such a water layer is absent or largely incomplete in these domains. Second, surface hydroxyl groups are necessary for the permanent grafting of the monolayer to the substrate.18,19 Clearly, these chemical functionalities are (17) Sagiv, J. J. Am. Chem. Soc. 1980, 102, 92. Parikh, A. N.; Allara, D. L.; Azouz, I. B.; Rondelez, F. J. Phys. Chem. 1994, 98, 7577. Brzoska, J. B.; Azouz, I. B.; Rondelez, F. Langmuir 1994, 10, 4367. Allara, D. L.; Parikh, A. N.; Rondelez, F. Langmuir 1995, 11, 2357. (18) Le Grange, J. D.; Markham, J. L.; Kurkjian, C. R. Langmuir 1993, 9, 1749. (19) Rye, R. R.; Nelson, G. C.; Dugger, M. T. Langmuir 1997, 13, 2965.

Letters

Figure 1. (a) Topographic image (left frame) and friction image (right frame) of a H-terminated sample after patterning by laser direct writing. The incident laser power and the writing speed were 447 mW and 5 mm/s, respectively. (b) Topographic images of a patterned sample after hydration in boiling water for 30 s and subsequent coating in 2.5 mM OTS solution for 20 s. (c) Height profile at the position marked by the white double arrow line in Figure 1b.

missing on the H-terminated areas. Nevertheless, in the first attempts to coat the substrates, both the H-terminated areas and the oxide lines remained uncoated. Obviously, at the conditions used here, the laser-induced oxidation results in the formation of a rather unreactive oxide. It has been reported, for example, that the growth of alkylsiloxane monolayers strongly depends on the degree of hydration of the silicon oxide; that is, a densely packed monolayer only forms on hydrophilic silicon oxide substrates.18,19 Very thin silicon oxide layers, however, are hydrophobic.20 Also, whereas freshly cleaned silicon oxide substrates after immersion in oxidizing acids bear a maximum number of surface hydroxyl groups and hence are hydrophilic,21 annealing of these samples results in the sucessive removal of the hydroxyl groups and the silicon oxide finally becomes hydrophobic.18 The hydrophilicity of annealed samples, though, can be restored upon rehydration in water.18 Thus, in ensuing experiments the patterned substrates have been first immersed into boiling water in order to hydrate the oxidized areas. With an increasing immersion time, this procedure also results in an oxidation of the H-terminated areas. An immersion time below 1 min, however, appears to suffice for hydration of the oxide lines, whereas the H-terminated areas remain unaffected. Subsequent analysis of the substrate surface revealed results similar to those displayed in Figure 1a, that is, AFM images with a negligible topographic but a significant friction contrast. After hydration, the patterned silicon substrates are coated by dipping into an OTS solution. As shown in the (20) Williams, R.; Goodman, A. M. Appl. Phys. Lett. 1974, 25, 531. (21) Zhuravlev, L. T. Langmuir 1987, 3, 316.

Letters

AFM images displayed in Figure 1b, a partial coating on top of the oxide lines can be clearly identified. Longer coating times result in the formation of a complete monolayer on the oxide lines. However, the incomplete coating chosen here nicely demonstrates that the growth of the octadecylsiloxane monolayer proceeds in the same way as on unstructured silicon oxide substrates, that is, via the formation of branched islands.22 Figure 1c displays a height profile across a coated oxide line. The height difference between the line and the surrounding areas is close to 2 nm, which is consistent with the height values reported in previous AFM investigations.22 In comparison with the generally accepted height of the final octadecylsiloxane monolayer of 2.5 nm,23 though, this value appears to be rather low. Clearly, this could simply result because of the general difficulty of height measurements on substrates with distinctly terminated regions. Also, reoxidation of the originally H-terminated areas in air takes place after some hours, which might change the relative height level of these areas and hence hamper absolute measurements. An additional indication for a selective coating of the oxide lines is provided by a pronounced friction contrast between the coated lines and the surrounding areas. To probe the achievable lateral resolution of the patterning technique, that is, the width of the oxide lines, H-terminated Si(100) substrates have been structured at distinct incident laser powers and writing speeds. Figure 2 displays the dependence of the line width on the writing speed for three different laser powers at a wavelength of 514 nm using an objective with a NA of 0.25. At a given laser power, the line width drops exponentially approaching a threshold, which appears to remain constant up to high writing speeds. Clearly, there will be a limit for the patterning process at a certain writing speed above the value of 25 mm/s tested here. Using lower laser powers generally results in decreased line widths. At a laser power of 400 mW, oxide lines with a width of 500 nm can routinely be prepared. Patterning using less than 400 mW, however, was limited to comparatively slow writing speeds. A minimum line width close to 500 nm has also been reported by Mu¨llenborn et al.15 Considering the focal beam diameter of 2.8 µm used in our experiments, a line width of 500 nm indicates a highly nonlinear patterning process.24 Contrary to standard photolithography, this generally offers the opportunity for patterning with a (22) Bierbaum, K.; Grunze, M.; Baski, A. A.; Chi, L. F.; Schrepp, W.; Fuchs, H. Langmuir 1995, 11, 2143. Carraro, C.; Yauw, O. W.; Sung, M. M.; Maboudian, R. J. Phys. Chem. B 1998, 102, 4441. Vallant, T.; Brunner, H.; Mayer, U.; Hoffmann, H.; Leitner, T.; Resch, R.; Friedbacher, G. J. Phys. Chem. B 1998, 102, 7190. Balgar, Th. Bautista, R.; Hartmann, N.; Hasselbrink, E. Surf. Sci. 2003, 532-535, 963. (23) Wasserman, S. R.; Tao, Y.-T.; Whitesides, G. M. Langmuir 1989, 5, 1074.

Langmuir, Vol. 20, No. 9, 2004 3527

Figure 2. Dependence of the oxide line width on the incident laser power and the writing speed. The lines are to guide the eyes only. For line width measurements, the patterned samples were etched in a solution of potassium hydroxide in order to conserve the pattern and increase the topographic contrast between the previously irradiated and nonirradiated areas.

lateral resolution well below the respective diffraction limit. Experiments in our group using an objective with a NA of 0.65, in fact, allow for routine patterning with a lateral resolution of 280 nm. Using an objective with a NA of 0.85, the narrowest oxide lines with sharp line edges exhibited a width of 180 nm. Implying diffraction-limited performance, a lateral resolution below 100 nm appears feasible. Note, however, that in conjunction with the focal beam diameter also the depth of the focus decreases, which complicates the work with high-aperture objectives. In conclusion, a simple constructive procedure for the preparation of submicron-structured alkylsiloxane monolayers is proposed. Laser direct writing allows for routine preparation of oxide patterns on H-terminated Si(100) samples with a lateral resolution well below the diffractionlimited laser spot size. After hydration of these templates, a selective coating of the oxide pattern takes place upon immersion into alkylsilane solutions. Current work focuses on the preparation of nanostructured templates using a scanning probe microscope as well as the secondary functionalization of the remaining uncoated areas. In conjunction with standard wet etching procedures, this offers the opportunity for the preparation of multifunctional three-dimensionally structured organic monolayers over several length scales. Acknowledgment. Financial support by the Deutsche Forschungsgemeinschaft (HA 1424/5-3, Graduiertenkolleg 689) is gratefully acknowledged. T.B. thanks the Studienstiftung des Deutschen Volkes for a stipend. LA040006S (24) Ehrlich, D. J.; Tsao, J. Y. Appl. Phys. Lett. 1984, 44, 267.