Anal. Chem. 2002, 74, 317-323
Articles
Surface Modification and Patterning Using Low-Energy Ion Beams: Si-O Bond Formation at the Vacuum/Adsorbate Interface Chris Evans,† Nathan Wade,† Federico Pepi,‡ Greg Strossman,§ Tom Schuerlein,§ and R. Graham Cooks*,†
Department of Chemistry, Purdue University, West Lafayette, Indiana 47907, Dipartimento di Studi di Chimica e Technologia delle Sostanze Biologicamente Attive, University of Rome “La Sapienza”, 00185 Rome, Italy, and Charles Evans and Associates, Sunnyvale, California 94086
Modification of hydroxyl-terminated self-assembled monolayer (HO-SAM) surfaces by collision of low-energy (15 eV) hyperthermal Si(CH3)3+ ions is shown to lead to Si-O bond formation and terminal trimethylsilyl ether formation. Modification was verified by in situ mass spectrometry using chemical sputtering with CF3+ ions (70 eV), ex situ secondary ion mass spectrometric analysis (12 kV Ga+ primary ion beam), and through X-ray photoelectron spectroscopy by monitoring Si (2s). The nature of the surface modification was further established by analysis of synthetic SAM surfaces made up of mixtures of the trimethylsilyl-11-mercapto-1-undecane ether and various proportions of the hydroxyl-terminated mercaptan (11mercapto-1-undecanol). These mixed surfaces, as well as the spectroscopic data, indicate that ca. 30% of the hydroxyl chains are covalently modified at saturation coverage. Analogous surface transformations are achieved using Si(CH3)2F+ and Si(CH3)2C6H5+. Creation of nanopatterns within self-assembled monolayer (SAM) structures can be achieved using scanning probe lithography, a method that uses atomic force microscopy (AFM) to “nanograft” various thiols on the nanometer scale to write patterns of functionalized SAM chains onto an existing SAM surface.1-3 “Dip-pen” nanolithography delivers SAM chains to a bare gold substrate via capillary forces in a pattern established by the †
Purdue University. University of Rome “La Sapienza”. § Charles Evans and Associates. (1) Muller, W. T.; Klein, D. L.; Lee, T.; Clark, J.; Mceuen, P. L.; Schultz, P. G. Science 1995, 268, 272. (2) Xu, S.; Miller, S.; Laibinis, P. E.; Liu, G. Y. Langmuir 1999, 15, 7244. (3) Liu, G. Y.; Xu, S.; Qian, Y. Acc. Chem. Res. 2000, 33, 457-466. ‡
10.1021/ac010928p CCC: $22.00 Published on Web 12/11/2001
© 2002 American Chemical Society
motion of the tip.4-6 In work on a larger dimensional scale, micropatterened surfaces can be created through such procedures as photolithography7-9 and replicated by microcontact printing. These capabilities depend on simple chemistry (poly(dimethylsiloxane) (PDMS) polymerization and alkanethiolate self-assembly) and can be carried out on materials with special surface structural features, such as steps and edges. Typical of the chemistry involved is the creation of reactive carboxylic anhydride SAMs patterns through microcontact printing using a PDMS stamp to generate a mixed SAM consisting of anhydrides and amides.10,11 Although the microcontact printing method to create mixed SAM surfaces is a well-established technique for patterning SAMs on the meso scale (micrometers and larger) as well as on curved surfaces, the many steps required makes it an arduous task. Direct transformation of the chemical nature of a nanostructure would complement these capabilities,12 especially if surface functional groups could be altered after fabrication of the microdevice. The potential usefulness of low-energy ions to modify surfaces has become evident in the past few years.13-16 There are also (4) Piner, R. D.; Zhu, J.; Xu, F.; Hong, S. H.; Mirkin, C. A. Science 1999, 283, 661. (5) Hong, S. H.; Zhu, J.; Mirkin, C. A. Science 1999, 286, 523. (6) Amro, N. A.; Xu, S.; Liu, G. Y. Langmuir 2000, 16, 3006-3009. (7) Jeon, N. L.; Finnie, K.; Branshaw, K.; Nuzzo, R. G. Langmuir 1997, 13, 3382. (8) Jeon, N. L.; Choi, I. S.; Whitesides, G. M.; Kim, N. Y.; Laibinis, P. E.; Harada, Y.; Finnie, K. R.; Girolami, G. S.; Nuzzo, R. G. Appl. Phys. Lett. 1999, 75, 4201. (9) Ghosh, P.; Lackowski, W. M.; Crooks, R. M. Macromolecules 2001, 34, 1230-1236. (10) Yan, L.; Zhao, X. M.; Whitesides, G. M. J. Am. Chem. Soc. 1998, 120, 6179. (11) Yan, L.; Huck, W. T. S.; Zhao, X. M.; Whitesides, G. M. Langmuir 1999, 15, 1208. (12) Schreiber, F. Prog. Surf. Sci. 2000, 65, 151-256.
Analytical Chemistry, Vol. 74, No. 2, January 15, 2002 317
limitations in some of these experiments, including the fact that the modification has not always been confined to the outermost surface region, and sometimes there has been a lack of chemical specificity or an inadequate chemical characterization. In still other cases, the efficiency of reaction has been low.17-19 Specific chemical transformations of surface functional groups by selected ions are known; for example, carboxylate-terminated surfaces can be converted to phenyl-terminated surfaces, but only in low yield.20 In the present work, chemically specific covalent surface modifications are achieved in relatively high yield, and the resulting covalently modified surfaces are characterized by a variety of spectroscopic methods. The functional group transformation we report occurs on a microscale through the agency of low-energy, reactive polyatomic ions. Specifically, we show that gas-phase trimethylsilyl cations, Si(CH3)3+, can be covalently bound to a hydroxyl-terminated selfassembled monolayer (11-mercapto-1-undecanol) surface. Proof is provided that covalent bond formation is involved, and estimates are made regarding the quantitative aspects of this highly surfacesensitive reaction. Remarkable features of this experiment are (i) this covalent modification is extremely efficient and occurs in situ without fragmentation of the molecular projectile to form a trimethylsilyl ether; (ii) when the treated surface is probed using a more energetic ion beam, the trimethylsilyl cation is released, and it constitutes the major ion observed in the secondary ion spectrum; (iii) spatial control of surface functionalization can be achieved by masking; and finally, (iv) verification of modification was achieved using two independent mass spectrometric analysis techniques, as well as a through X-ray photoelectron spectroscopy (XPS).
EXPERIMENTAL SECTION The ion/surface scattering experiments were conducted in a four-analyzer tandem BEEQ mass spectrometer described previously.21 Primary ions were generated by 70 eV electron impact and subjected to mass and energy analysis. This mass- and kineticenergy-analyzed ion beam was decelerated to the required collision energy prior to collision with a self-assembled monolayer surface. Mass distributions of the scattered product ions were determined using a quadrupole analyzer prior to detection by an electron multiplier. During ion/surface scattering experiments, the scattering angle was set to 90° (incident angle 45° to normal), and while performing modification experiments, the incident angle was set to 0°. (13) Ast, T.; Mabud, M. A.; Cooks, R. G. Int. J. Mass Spectrom. Ion Proc. 1988, 82, 131. (14) Cooks, R. G.; Ast, T.; Mabud, M. A. Int. J. Mass Spectrom. Ion Proc. 1990, 100, 209. (15) Rabalais, J. W. Low Energy Ion-Surface Interactions; J. Wiley and Sons: New York, 1994. (16) Shen, J. W.; Grill, V.; Evans, C.; Cooks, R. G. J. Mass Spectrom. 1999, 34, 354. (17) Rabalais, J. W.; Marton, D. Nucl. Instrum. Methods B 1992, 67, 287. (18) Miller, S. A.; Luo, H.; Pachuta, S.; Cooks, R. G. Science 1997, 275, 1447. (19) Wijesundara, M. B. J.; Hanley, L.; Ni, B. R.; Sinnott, S. B. Proc. Nat. Acad. Sci. 2000, 97, 23. (20) Shen, J. W.; Evans, C.; Wade, N.; Cooks, R. G. J. Am. Chem. Soc. 1999, 121, 9762. (21) Winger, B. E.; Laue, H. J.; Horning, S. R.; Julian, R. K.; Lammert, S. A.; Riederer, D. E.; Cooks, R. G. Rev. Sci. Instrum. 1992, 63, 5613.
318
Analytical Chemistry, Vol. 74, No. 2, January 15, 2002
The functionalized target surface, a hydroxyl-terminated selfassembled monolayer (HO-SAM), HO-(CH2)11-S-Au, was prepared by exposing a clean gold substrate to a 1 mM ethanolic solution of the corresponding thiol for a minimum of 2 days. The commercial substrate (International Wafer Service) consisted of a Si wafer covered with 5 nm of chromium as an adhesion layer and 200 nm of polycrystalline vapor-deposited gold. Before exposure to the thiol, the surface was cleaned with piranha solution, H2SO4:H2O2 in 3:1 volume ratio, and thoroughly washed in deionized water and ethanol. Because piranha solution is highly oxidizing, care should be taken while handling it. Other functionalized SAM surfaces were synthesized as described and were allowed to assemble under identical conditions. The authentic silyl ether SAM was prepared by adding 1:1:1 molar ratios of (CH3)3SiCl and (CH3CH2)3N to the HO-SAM starting material, 11mercapto-1-undecanol. The mixture was stirred at 25° C for 8 h in acetonitrile. The final product was analyzed using a Finnigan TSQ 700 triple quadrupole mass spectrometer (Thermo Finnigan, San Jose, CA) to reveal the creation of the species HS-(CH2)11O-Si(CH3)3, a trimethylsilyl-11-mercapto-1-undecane ether. This purified product was dissolved in CH3CN to create an ∼1 mM thiol solution into which Au surfaces were placed for self-assembly. Ex situ analysis of modified SAM surfaces was performed by time-of-flight secondary ion mass spectrometry (TOF-SIMS) and X-ray photoelectron spectroscopy (XPS). Samples were removed from the vacuum system and delivered overnight to the analytical services laboratory of Charles Evans & Associates for surface analysis. The TOF-SIMS instrument utilized was a Charles Evans & Associates PHI TRIFT I series instrument. A liquid metal ion gun created a pulsed 12 kV primary ion beam consisting of 69Ga+ (600 pA primary ion current), which was rastered across the SAM sample surface in order to ionize and desorb surface species. An entire mass spectrum was acquired from each primary ion beam raster position, and both chemically specific ion images and exact mass measurements were made. While obtaining ion images, the instrument was operated with the beam buncher off, providing a primary ion beam with a spatial resolution of 0.5 µm. Exact mass measurements were made with the beam buncher on, providing a 1 µm spatially resolved beam and a mass resolution (at halfheight) of ca. 5000. The XPS instrument utilized was a PHI Quantum 2000 series instrument. The 100 W Al KR (1486.7 eV) X-ray beam was rastered over an area, 500 µm by 1400 µm, of the SAM surface, ejecting photoelectrons with binding energies characteristic of the various elements on the surface. Collection times of 2 h were used. The binding energies of the ejected electrons were normalized under standard conditions, correcting the C (ls) signal to a binding energy of 284.8 eV. Using appropriate sensitivity factors for the elements that were detected, integrated peak areas provided elemental percent composition of the SAM surfaces analyzed. RESULTS AND DISCUSSION Evidence of Covalent Modification by Chemical Sputtering with CF3+. Figure 1A shows a mass spectrum recorded upon collision of 70 eV CF3+ projectiles, m/z 69, with a virgin hydroxylterminated self-assembled monolayer (HO-SAM) surface. Most of the secondary ions generated during this hyperthermal ion/ surface collision arise by the low-energy charge exchange process known as chemical sputtering.22 This charge-transfer event des-
Figure 1. Scattered ion mass spectra recorded upon collision of 70 eV CF3+ ions with an 11-mercapto-1-undecanol self-assembled monolayer surface (A) before and (B) after surface modification using 15 eV Si(CH3)3+ ions.
orbs secondary ions that are characteristic of both the HO-SAM surface and the adventitious hydrocarbon species on the surface. Peaks characteristic of the monolayer surface are observed at m/z 19, 31, and 45, corresponding to H3O+, CH2OH+, and C2H4OH+, respectively, namely, these are the products of C-O and C-C bond cleavage in the adsorbed hydroxyl alkane thiol. A small portion of the impinging CF3+ projectiles are elastically scattered from the surface and are observed at m/z 69. Adventitious hydrocarbons give rise to the hydrocarbon ion series m/z 27, 29, 39, 41, and 43. The characterized surface was then treated with the trimethylsilyl cation, Si(CH3)3+ at a laboratory collision energy of 15 eV for 1 h. At this collision energy, the surface current measured 2 × 10-10 A and is estimated to interact with an exposed HO-SAM area of 9 mm2. The treated surface was then interrogated with a 70 eV CF3+ ion beam, 2 × 10-10 A, to again induce chemical sputtering. The resulting mass spectrum is shown in Figure 1B. It is immediately evident that the base peak in the resulting spectrum now corresponds to Si(CH3)3+, m/z 73. In fact, the species observed at m/z 73, the initial projectile ion, is ca. 3 times more abundant than the characteristic cation, m/z 31, generated from the unmodified surface. An additional significant change is the sizable increase of the peak at m/z 45, ascribed to the known trimethylsilyl fragmentation product Si(CH3)H2+. Subsequent modification experiments performed at a slightly higher ion flux revealed that Si(CH3)3+ became the base peak in the mass spectrum after only 10 min of ion bombardment. The interaction of low-energy (10-70 eV) trimethylsilyl cations with surfaces has been studied previously, both theoretically23 and experimentally.24,25 These investigations included nonfunctionalized self-assembled alkanethiol monolayer surfaces, so the observed (22) Vincenti, M.; Cooks, R. G. Org. Mass Spectrom. 1988, 23, 317. (23) Schultz, D. G.; Wainhaus, S. B.; Hanley, L.; deSaint-Claire, P.; Hase, W. L. J. Chem. Phys. 1997, 106, 10337. (24) Schultz, D. G.; Hanley, L. J. Chem. Phys. 1998, 109, 10976. (25) Wainhaus, S. B.; Lim, H.; Schultz, D. G.; Hanley, L. J. Chem. Phys. 1997, 106, 10329.
processes were restricted to inelastic scattering and surfaceinduced dissociation. Surface science studies are commonly plagued by surface contamination, a common form of which is represented by adventitious poly(dimethylsiloxane) (PDMS). PDMS contamination results in the observation of silyl cations, including the ion Si(CH3)3+, in the chemical sputtering mass spectrum at m/z 73. Although the intense Si(CH3)3+ peak seen in Figure 1B is unlikely to arise from PDMS contamination, experiments were performed with Si(CD3)3+, m/z 82, to rule out this possibility. Shown in the inset of Figure 1B is the chemical sputtering mass spectrum of an HO-SAM surface treated with the deuterated silyl cation. New peaks in the mass spectrum, corresponding to Si(CD3)D2+ and Si(CD3)3+, are observed at m/z 50 and 82, respectively. It is interesting to note that the Si(CH3)H2+ peak observed using the nondeuterated projectile, shifts to m/z 50, confirming the assignment Si(CD3)D2+ and confirming its origin by ethene elimination from the impinging projectile ion. In additional blank experiments, an unmodified HO-SAM surface was treated with CF3+ at 15 eV at a similar ion dosage in order to determine if the trimethylsilyl signal could be produced by treatment with other low-energy ions; no trimethylsilyl signal was observed upon chemical sputtering of these surfaces. Chemical Evidence for Covalent Bonding. To investigate the chemical binding of Si(CH3)3+ at the HO-SAM surface, experiments were completed to determine the strength of this interaction. Experiments were conducted by testing various Si(CH3)3+-modified HO-SAM surfaces outside the ultrahigh vacuum scattering chamber using solutions of various nucleophiles. For example, after rinsing with anhydrous methanol, the surface was reexamined and revealed only a minor decrease (
