6214
Langmuir 2006, 22, 6214-6221
Trapping Silicon Surface-Based Radicals Dong Wang and Jillian M. Buriak* Department of Chemistry and the National Institute for Nanotechnology, UniVersity of Alberta, Edmonton, AB T6G 2G2, Canada ReceiVed March 9, 2006. In Final Form: April 27, 2006 The spontaneous one-electron reduction of diazonium salts on hydride-terminated porous silicon (pSi) and flat silicon produces surface radicals that can be trapped chemically. These silicon radicals react with reagents such as alkyl/arylselenoethers, alkenes, alkynes, and alkylbromide groups to generate covalently bound functionalities in a manner analogous to the chemistry of molecular-based silicon radical species, prepared via different methods. When pSi is exposed to an acetonitrile solution of any of the three diazonium salts examined in this study, aryl groups from the diazonium precursor become covalently bound and significant oxidation is noted; if, however, a reactive trapping agent is added, such as an alkyl/arylselenoether or a carbon-carbon unsaturated bond, no aryl group attachment is observed and oxidation is circumvented due to the efficiency of the trapping chemistry. The reactions proceed rapidly, in less than 3 h to maximum coverage, at room temperature. The diazonium salt-initiated radical reaction with R,ω-alkenes and alkynes tolerates various functional groups including aryl, diene, diyne, carboxylic acid, and hydroxyl, reacting exclusively via the carbon-carbon unsaturated bond; R,ω-bromoalkenes are not, however, compatible with this chemistry. A silicon-based molecule, tris(trimethylsilyl)silane, in the presence of a diazonium salt initiator and a primary alkyne does not lead to the hydrosilylation product but to tris(trimethylsilyl)silylbromide and the hydrogenated arene, derived from the diazonium. The difference in reactivity between the molecule and the surface is due to the fact that the silicon surface is a source of electrons to reduce the diazonium salts to aryl radicals, whereas a heterolytic pathway is followed in the molecular silane case.
Introduction The functionalization of silicon surfaces with Si-C bonds is an area of intense and active investigation because of the potential for a myriad of practical applications.1 Interfacing organic molecules directly with silicon is predicted to be a desirable scaffold for molecular electronics, sensing applications, siliconbased drug delivery chips, and many others.2 From the fundamental side, the actual mechanisms for Si-C bond-forming chemistry are simultaneously perplexing and fascinating.3 Correlations are made between the vast knowledge of organosilicon molecules and the more limited arena of observed reactivity on silicon surfaces,4-6 but many inconsistencies exist. Excitonmediated surface reactions on nanocrystalline and flat silicon, * Corresponding author: tel +1 (780) 492-1821; fax +1 (780) 492-8231; e-mail
[email protected]. (1) (a) Buriak, J. M. Chem. ReV. 2002, 102, 1272. (b) Solares, S. D.; Yu, H.; Webb, L. J.; Lewis, N. S.; Heath, J. R.; Goddard, W. A., III. J. Am. Chem. Soc. 2006, 128, 3850. (c) Yu, H.; Webb, L. J.; Ries, R. S.; Solares, S. D.; Goddard, W. A., III; Heath, J. R.; Lewis, N. S. J. Phys. Chem. B. 2005, 109, 671. (d) Webb, L. J.; Nemanick, E. J.; Biteen, J. S.; Knapp, D. W.; Michalak, D. J.; Traub, M. C.; Chan, A. S. Y.; Brunschwig, B. S.; Lewis, N. S. J. Phys. Chem. B 2005, 109, 3930. (e) Sailor, M. J.; Lee, E. J. AdV. Mater. 1997, 9, 783. (f) Wayner, D. D. M.; Wolkow, R. A. J. Chem. Soc., Perkin Trans. 2, 2002, 23. (g) Boukherroub, R.; Morin, S.; Bensebaa, F.; Wayner, D. D. M. Langmuir 1999, 15, 3831. (h) Wojtyk, J. T. C.; Tomietto, M.; Boukherroub, R.; Wayner, D. D. M. J. Am. Chem. Soc. 2001, 123, 1535. (i) Asanuma, H.; Lopinski, G. P.; Yu, H.-Z. Langmuir 2005, 21, 5013. (j) Lasseter, T. L.; Clare, B. H.; Abbott, N. L.; Hamers, R. S. J. Am. Chem. Soc. 2004, 126, 10220. (k) Cicero, R. L.; Linford, M. R.; Chidsey, C. E. D. Langmuir 2000, 16, 5688. (l) Linford, M. R.; Chidsey, C. E. D. Langmuir 2002, 18, 6217. (m) Niederhauser, T. L.; Lua, Y.-Y.; Jiang, G.; Davis, S. D.; Matheson, R.; Hess, D. A.; Mowat, I. A.; Linford, M. R. Angew. Chem., Int. Ed. 2002, 41, 13, 2353. (n) Schmeltzer, J. M.; Porter, L. A., Jr.; Stewart, M. P.; Buriak, J. M. Langmuir 2002, 18, 2971. (o) Liu, Y.-J.; Navasero, N. M.; Yu, H.-Z. Langmuir 2004, 20, 4039. (p) Cattaruzza, F.; Cricenti, A.; Flamini, A.; Girasole, M.; Longo, G.; Mezzi, A.; Prosperi, T. J. Mater. Chem. 2004, 14, 1461. (n) Zhang, L.; Wesley, K.; Jiang, S. Langmuir 2001, 17, 6275. (q) Wallart, X.; de Villeneuve, C. H.; Allongue, P. J. Am. Chem. Soc. 2005, 127, 7871. (r) Allongue, P.; de Villeneuve, C. H.; Cherouvrier, G.; Cortes, R.; Bernard, M. C. J. Electroanal. Chem. 2003, 550, 161. (s) Boukherroub, R.; Petit, A.; Loupy, A.; Chazalviel, J.-N.; Ozanam, F. J. Phys. Chem. B 2003, 107, 13459. (t) Fellah, S.; Boukherroub, R.; Ozanam, F.; Chazalviel, J.-N. Langmuir 2004, 20, 6359. (u) Fellah, S.; Teyssot, A.; Ozanam, F.; Chazalviel, J.-N.; Vigneron, J.; Etcheberry, A. Langmuir 2002, 18, 5851.
for instance, have no obvious parallels with molecules, as well as some electrochemically driven reactions.7-9 In 1997, it was demonstrated that aryldiazonium reagents can serve as a source of aryl groups to be grafted directly to hydride(2) (a) Stewart, M. P.; Buriak, J. M. AdV. Mater. 2000, 12, 859. (b) Link, J. R.; Sailor, M. J. Proc. Natl. Acad. Sci. U.S.A. 2003, 100, 10607. (c) Dorvee, J. R.; Derfus, A. M.; Bhatia, S. N.; Sailor, M. J. Nat. Mater. 2004, 3, 896. (d) Canham, L. T.; Reeves, C. L.; Newey, J. P.; Houlton, M. R.; Cox, T. I.; Buriak, J. M.; Stewart, M. P. AdV. Mater. 1999, 11, 1505. (e) Faber, E. J.; de Smet, L. C. P. M.; Olthuis, W.; Zuilhof, H.; Sudho¨lter, E. J. R.; Bergveld, P.; van den Berg, A. ChemPhysChem 2005, 6, 2153. (f) Lopinski, G. P.; Wayner, D. D. M.; Wolkow, R. A. Nature 2000, 406, 48. (g) Piva, P. G.; DiLabio, G. A.; Pitters, J. L.; Zikovsky, J.; Rezeq, M.; Dogel, S.; Hofer, W. A.; Wolkow, R. A. Nature 2005, 435, 658. (h) Fabre, B.; Lopinski, G. P.; Wayner, D. D. M. Chem. Commun. 2002, 2904. (i) Zhu, X. Y.; Jun, Y.; Staarup, D. R.; Major, R. C.; Danielson, S.; Boiadjiev, V.; Gladfelter, W. L.; Bunker, B. C.; Guo, A. Langmuir 2001, 17, 7798. (j) Lin, Z.; Strother, T.; Cai, W.; Cao, X.; Smith, L. M.; Hamers, R. J. Langmuir 2002, 18, 788. (k) Jun, Y.; Cha, T.; Guo, A.; Zhu, X. Y. Biomaterials 2004, 25, 3503. (l) Pike, A. R.; Lie, L. H.; Eagling, R. A.; Ryder, L. C.; Patole, S. N.; Connolly, B. A.; Horrocks, B. R.; Houlton, A. Angew. Chem., Int. Ed. 2002, 41, 615. (3) (a) Brook, M. A. Silicon in Organic, Organometallic, and Polymer Chemistry; John Wiley and Sons: New York, 2000. (b) Rappoport, Z.; Apeloig, Y. The Chemistry of Organic Silicon Compounds, Vol. 3; John Wiley and Sons: New York, 2001. (4) (a) Linford, M. R.; Fenter, P.; Eisenberger, P. M.; Chidsey, C. E. D. J. Am. Chem. Soc. 1995, 117, 3145. (b) Bateman, J. E.; Eagling, R. D.; Worrall, D. R.; Horrocks, B. R.; Houlton, A. Angew. Chem., Int. Ed. 1998, 37, 2683. (5) (a) Kim, N. Y.; Laibinis, P. E. J. Am. Chem. Soc. 1999, 121, 7162. (b) Kim, N. Y.; Laibinis, P. E. J. Am. Chem. Soc. 1998, 120, 4516. (c) Song, J. H.; Sailor, M. J. J. Am. Chem. Soc. 1998, 120, 2376. (d) Song, J. H.; Sailor, M. J. Inorg. Chem. 1999, 38, 1503. (6) (a) Buriak, J. M.; Allen, M. J. J. Am. Chem. Soc. 1998, 120, 1339. (b) Buriak, J. M.; Stewart, M. P.; Geders, T. W.; Allen, M. J.; Choi, H. C.; Smith, J.; Raftery, D.; Canham, L. T. J. Am. Chem. Soc. 1999, 121, 11491. (7) (a) Stewart, M. P.; Buriak, J. M. J. Am. Chem. Soc. 2001, 123, 7821. (b) Stewart, M. P.; Buriak, J. M. Angew. Chem., Int. Ed. 1998, 37, 3257. (c) Sun, Q.-Y.; de Smet, L. C. P. M.; van Lagen, B.; Giesbers, M.; Thune, P. C.; van Engelenburg, J.; de Wolf, F. A.; Zuilhof, H.; Su¨dholter, E. J. R. J. Am. Chem. Soc. 2005, 127, 2514. (d) Eves, B. J.; Sun, Q. Y.; Lopinski, G. P.; Zuilhof, H. J. Am. Chem. Soc. 2004, 126, 14318. (8) (a) Robins, E. G.; Stewart, M. P.; Buriak, J. M. Chem. Commun. 1999, 2479. (b) Hurley, P. T.; Ribbe, A. E.; Buriak, J. M. J. Am. Chem. Soc. 2003, 125, 11334. (c) Gurtner, C.; Wun, A. W.; Sailor, M. J. Angew. Chem., Int. Ed. 1999, 38, 1966. (9) (a) Allongue, P.; deVilleneuve, C. H.; Pinson, J.; Ozanam, F.; Chazalviel, J. N.; Wallart, X. Electrochim. Acta 1998, 43, 2791. (b) deVilleneuve, C. H.; Pinson, J.; Bernard, M. C.; Allongue, P. J. Phys. Chem. B 1997, 101, 2415.
10.1021/la060653e CCC: $33.50 © 2006 American Chemical Society Published on Web 06/03/2006
Trapping Silicon Surface-Based Radicals Scheme 1. Summary of Known Reactivity of Diazonium Salts with (a) Silicon Surfaces and (b) Molecular Silanes, (c) Structure of the Tris(trimethylsilyl)silane Molecule, and (d) Chemical Structures of the Three Diazonium Salts Used in This Study
Langmuir, Vol. 22, No. 14, 2006 6215 Table 1. Absolute Rate Constants for the Reaction of Silyl Radicals with Several Typical Quenchers, from Previous Molecular Silane Literature15 entry
substrate
absolute rate constanta (M-1 s-1)
1 2 3 4 5 6 7 8
C6H5Br C5H11Br C2H3Ph C2HPh C2H3Bu C2Ht-Bu C10H21SePh C10H21SPh
4.6 × 106 b 2.0 × 107 5.9 × 107 1 × 108 c 3.9 × 106 2.3 × 106 c 9.6 × 107 1 µm), whereas in the pSi unfunctionalized control, substantially lower levels of C and Se are observed; the carbon seen in the pSi control is most likely due to adventitious carbon incorporation, a commonly observed phenomenon with this high surface area, hydrophobic material. The levels of oxidation of the ≡Si-Se-Ph-functionalized surface and the pSi controls are superimposable, substantiating the FTIR observation that this reaction does not induce oxidation of the interface. X-ray
photoelectron spectroscopy (XPS, shown in the Supporting Information) was applied to the analysis of this chemistry on a flat Si(100)-Hx surface to further verify the presence of selenium. The Se 3d feature appears at ∼55 eV, as expected.24,25 Some oxidation of the silicon interface is noted, based on the small feature in the Si 2p spectrum around 104 eV, suggesting that reaction of trace oxygen with the flat silicon surface is competitive with the selenide trapping agent. Surface Si-Se-Ar bonds have been previously prepared via a direct, high-temperature thermal (400 °C) reaction of a hydride-terminated Si(100)-Hx surface with an acetylselenoarene (Ar-Se-OAc) for short times, but mechanistic elucidation was not pursued, and thus it remains unknown as to whether parallels exist between these two reaction schemes.25 According to Table 1, the rates of reaction of alkene/alkyne addition (hydrosilylation products) to silicon-based radicals are comparable to that of the alkyl/arylselenoether (∼107-108 M-1 s-1). Addition of a range of different aryl and alkyl alkenes and alkynes, in the presence of a diazonium salt, leads to clean and efficient hydrosilylation reactions on the pSi surface, in less than 3 h. The simplest alkyne, 1-dodecyne, shown in Figure 4, reveals the expected profile including the ν(CdC) at 1595 cm-1 resulting from reduction of the alkyne to a ≡Si-(CdCR) group. The trans-olefinic out-of-plane deformation, γ(dCH), appears at ∼980 cm-1, indicating that some fraction of the silicon-bound vinyl groups are trans;7a this mode is not observed in reactions that exclusively produce a cis vinyl.6 Surface coverage does not increase above 3 h of contact time, and about 90% of complete incorporation is reached within 30-60 min. Comparison of the integrated ν(Si-Hx) region around 2100 cm-1 and the ν(C-Hx)
(20) Lebarillier, L.; Outurquin, F.; Paulmier, C. Tetrahedron 2000, 56, 7483. (21) Han, S. W.; Lee, S. J.; Kim, K. Langmuir 2001, 17, 6981. (22) Li, Y.-H.; Buriak, J. M. Inorg. Chem. 2006, 45, 1096. (23) Canham, L. T. AdV. Mater. 1995, 7, 1033.
(24) Shaporenko, A.; Cyganik, P.; Buck, M.; Terfort, A.; Zharnikov, M. J. Phys. Chem. B. 2005, 109, 13630. (25) Yasseri, A. A.; Syomin, D.; Loewe, R. S.; Lindsey, J. S.; Zaera, F.; Bocian, D. F. J. Am. Chem. Soc. 2004, 126, 15603.
Figure 4. Transmission FTIR of dodecenyl-modified porous Si surface initiated by different diazonium salts: (a) NBD, (b) FBD, and (c) BBD.
6218 Langmuir, Vol. 22, No. 14, 2006
in the 2900 cm-1 region shows that incorporation is similar to room-temperature Lewis acid-mediated hydrosilylation.6 Radicals certainly have been invoked for silicon surface hydrosilylation, but they require either high temperatures (>100 °C), UV irradiation, or longer times (24 h at room temperature, for instance, with TEMPO as the radical initiator).1h,4,13d,26 Choice of diazonium reagent has some importance, with little or no oxidation observed for 1-dodecyne hydrosilylation initiated by BBD and FBD; the nitro-substituted aryldiazonium, on the other hand, leads to small amounts of oxidation, as shown by the ν(Si-O) observed by FTIR (Figure 4). In addition, residual aromatic modes can be seen with NBD and FBD at 1515, 1345, and 1525 cm-1, respectively, whereas the BBD spectrum is clean. BBD appears to be the ideal diazonium initiator in terms of oxidation and residual aryl incorporation. Functional group tolerance of the BBD reagent as a hydrosilylation initiator was then investigated. Figure 5 and the Supporting Information show the FTIR spectra of different surfaces obtained with a range of alkenes and alkynes containing the following ω-substituents: hydroxyl, alkyne, aryl, carboxylic acid, alkene, and bromo groups. In terms of compatibility with an ω-hydroxyl, Figure 5a shows the reaction with 10-undecyn1-ol; the ν(CdC) of the surface-bound vinyl group is obvious at 1595 cm-1, as well as the broad ν(O-H) mode at ∼3300 cm-1. The feature at ∼1000 cm-1 can be assigned to the C-O stretch of the alcohol but could also obscure silicon oxidation in this region, although the lack of oxygen back-bonded Si-H features at 2200 cm-1 suggests, in fact, that oxidation is low. Figure 5b indicates good incorporation levels for styrene, and in the case of 1,7-octadiyne (Figure 5c), the ω-alkyne is clearly intact since the strong ν(≡C-H) peak is observed at 3300 cm-1. A simple alkene, 1-dodecene, shown in Figure 5d, is straightforward to interpret, with only ν(CHx) just above 2900 cm-1, and the δ(CH2) at 1465 cm-1 attributable to the aliphatic chain. The GATR25,27 (germanium attenuated total reflection, or GATR) IR spectrum of 1-dodecyne is shown in Figure 5e. The νas(CH2) stretch of the aliphatic chain can be seen at 2922 cm-1, indicating that these monolayers are not particularly ordered but certainly are not liquidlike. XPS analysis confirms that the reaction is relatively free of oxidation, based upon examination of the Si 2p profile (Supporting Information). Both N and Br analysis showed no incorporation of residual nitrogen or the aryl group from diazonium salts, to the detection limits of the instrument. Other alkenes and alkynes such as 10-undecylenic acid, cyclohexene, 1,13-tetradecene, and phenylacetylene react smoothly, yielding the expected FTIR spectra (Supporting Information). The R,ω-terminated alkene 8-bromo-1-octene, on the other hand, leads to hydrocarbon incorporation as shown by the modes above 2900 cm-1, but the presence of substantial oxidation as well as the 2960 cm-1 methyl mode suggests reactivity between the silicon surface radical and the bromo moiety (Supporting Information and vide infra). A summary of the terminations prepared via this method is shown in Figure 6. To try to confirm the involvement of radicals, electron spin resonance (ESR) experiments were carried out on free-standing pSi on frozen samples at 10 K, in the presence and absence of the diazonium salts, with the goal of observing an increase in the integrated silyl radical intensity with an initiator present. Silyl radicals are observed in freshly etched pSi samples, as has (26) (a) Perring, M.; Dutta, S.; Arafat, S. N.; Mitchell, M.; Kenis, P. J. A.; Bowden, N. B. Langmuir 2005, 21, 10537. (b) Dutta, S.; Perring, M.; Barrett, S.; Mitchell, M.; Kenis, P. J. A.; Bowden, N. B. Langmuir 2006, 22, 2146. (27) Lummerstorfer, T.; Hoffmann, H. Langmuir 2004, 20, 6542.
Wang and Buriak
Figure 5. FTIR spectra of hydrosilylated hydrogen-terminated silicon surfaces with the diazonium salt, BBD, as the initiator. The alkenes/alkynes used in each spectrum are as follows: (a) 10undecyn-1-ol on pSi, (b) styrene on pSi, (c) 1,7-octadiyne on pSi, (d) 1-dodecene on pSi, and (e) 1-dodecyne on Si(111)-H.
been seen previously by many beforehand.28 In the presence of the BBD diazonium reagent in acetonitrile, only negligible increases in the integrated intensity of the silyl radical defects were observed, most likely due to their low concentration relative to the intrinsic defect population of a pSi matrix. Finally, to contrast with molecular silicon-containing molecules, we investigated the reaction between TTMSS, a diazonium reagent, and an alkene, in CD3CN solution. As shown in the (28) (a) von Bardeleben, H. J.; Ortega, C.; Grosman, A.; Morazzani, V.; Siejka J.; Stievenard D. J. Luminesc. 1993, 57, 301. (b) Nakamura, T.; Sasaki, K.; Hayashi, K.; Mimura, H. Kobayashi, J. Appl. Surf. Sci. 1996, 92, 291. (c) Rong, F. C.; Harvey, J. F.; Poindexter, E. H.; Gerardi, G. J. Appl. Phys. Lett. 1993, 63, 920. (d) Chiesa, M.; Amato, G.; Boarino, L.; Garrone, E.; Geobaldo, F.; Giamello, E. Angew. Chem., Int. Ed. 2003, 42, 5031.
Trapping Silicon Surface-Based Radicals
Langmuir, Vol. 22, No. 14, 2006 6219
Figure 6. Summary of surface terminations prepared on silicon via quenching of radicals produced through diazonium salt initiation.
Supporting Information (Figure S5a), 1H NMR clearly indicates that a 1:1 mixture of TTMSS and the FBD diazonium salt, ArN2+BF4- (Ar ) 3,4,5-trifluorophenyl), produces the hydrogenated arene (and presumably TTMSS fluoride) in 24 hsa dramatically different set of products compared to the reaction of pSi with a diazonium salt.11 If 1 equiv of 1-dodecyne is added to the 1:1 TTMSS/FBD acetonitrile solution for the same length of time, the alkyne remains untouched, and the TTMSS + ArN2+BF4- reaction proceeds, as observed in the control experiment in absence of 1-dodecyne (Supporting Information, Figure S5b). Figure S5c (Supporting Information) shows the 1H NMR of the FBD diazonium salt in CD3CN, exposed to a freshly etched shard of pSi. All the diazonium salt is consumed, with the major product being the hydrogenated arene, although other unidentified arene products are observed. The pSi has a high surface area, allowing for reduction of sufficient quantities of FBD to permit detection of the organic products by solutionphase NMR.
Discussion A summary of the reactivity observed in this work is shown in Figure 7. The critical step that commences all chemistry is the reduction of the diazonium salt (Figure 7a). The key to the case of silicon, and any reaction on a metal or semiconductor, is the 1 e- reduction of the diazonium cation, leading to N2 loss, formation of the aryl radical, and a hole, or positive charge, in the silicon. It has previously been surmised that, under OCP conditions on Si(111)-H, the positive charge is eliminated as a proton in the form HX, with the counteranion of the diazonium salt, X-, as the conjugate base.10 While entirely plausible, it is also possible that the hole migrates into the silicon lattice, drawn toward the edges and other roughened areas of the silicon wafer (Figure 7b). Hole migration in this fashion has been postulated to be important in NH4F (aq) etching of Si(111) as the edges and roughened spots can act as sacrificial anodes.29 In either case, the aryl radical derived from the diazonium salt abstracts a hydrogen from the hydride-terminated surface, leading to a silicon radical. The 1H NMR studies of FBD in a CD3CN solution in the presence of pSi clearly reveal formation of the expected hydrogenated arene, trifluorobenzene, although other unidentified (29) Allongue, P.; de Villeneuve, C. H.; Morin, S.; Boukherroub R.; Wayner, D. D. M. Electrochim. Acta 2000, 45, 4591.
Figure 7. Summary of silicon surface reactivity: (a) reduction of the diazonium salt with one electron derived from the bulk silicon; (b) production of silicon surface radicals and elimination of the positively charged hole, through either HX production or migration of the hole into the bulk; (c) quenching of the silicon surface radicals with a selenoether, an alkene, and an alkylbromide.
arene products are observed. In the case of an Si-H-containing silane molecule (i.e., TTMSS), there is no available reducing electron sink and so the reaction proceeds in a heterolytic fashion;30 the Si-H group acts as a hydride, attacking the +N2Ar diazonium group, which then collapses to release N2 and the hydrogenated arene.11 Therefore, no hydrosilylation of 1-dodecyne by TTMSS is observed under these conditions. In terms of harnessing the silicon radicals produced via contact of hydride-terminated silicon, Figure 7c outlines the chemistry observed in this work. Quenching of the silicon surface radicals with secondary reagents, such as a selenoether or alkyne/alkene, is favored over combination with the diazonium-derived aryl radical due to a large excess of secondary reagents. The (30) Galli, C. Chem. ReV. 1988, 88, 765.
6220 Langmuir, Vol. 22, No. 14, 2006
involvement of silicon surface radicals is demonstrated chemically through the contact of an alkyl/arylselenoether with both pSi and flat silicon, in the presence of a diazonium initiator. By analogy to the molecular silane literature, the selenide reagent associates reversibly with the silicon radical to form a selenanyl radical.15c,31 This step is followed by an R-cleavage to give the phenyselenolate modified Si surface and a new alkyl radical, R•, that will then abstract H• from either the surface or solvent. Competitive kinetics experiments with molecular silyl radicals point to a stepwise mechanism,15c,31 but such conclusions cannot be made at this point for the analogous surface reaction. The fact that the thioether does not lead to ≡Si-S-Ph termination is similar to the molecular case; the R S-C bond cleavage is not favorable. Hydrosilylation of alkenes and alkynes involving radicals is a reaction that is well-established in both the molecular and surface areas. The reaction tolerates a broad range of functional groups, although alkyl bromide is clearly a problem. In the case of 8-bromo-1-octene, the silicon surface radical most likely abstracts the Br,15d forming a hydrolytically sensitive Si-Br bond and an alkyl radical that could go on to abstract H• from the surface or solvent. The product of these steps is 1-octene, which then undergoes hydrosilylation with the surface, leading to the observed 2960 cm-1 methyl feature due to the terminal ω-methyl group (Figure 7c, part iv). The Si-Br bond, upon exposure to laboratory ambient conditions, hydrolyzes, leading to substantial oxidation.
Conclusions Surface silicon radicals are formed on both hydride-terminated pSi and flat silicon surfaces in the presence of a diazonium salt reagent. The presence of these radicals is confirmed through screening of various reagents known to react with molecular silicon radicals. The diazonium salts can be viewed not only as a convenient source of aryl radicals for forming ≡Si-Ar bonds but also as room-temperature radical initiators. As would be predicted on the basis of the molecular reactivity of radicals derived from tris(trimethylsilyl)silane and related molecules, the quenching of these silicon radicals is rapid and efficient, with incorporation levels reaching 90% of their maximum in 30-60 min and completion achieved in 3 h. The key difference between the direct reaction of a diazonium salt with either a silane or a hydride-terminated silicon surface is that, under open cell potential conditions, the bulk silicon acts as a source of electrons to reduce the +N2Ar species, leading to the aryl radical intiator. In the case of silane molecules, on the other hand, the mechanism is heterolytic and radical quenching in the form of hydrosilylation products is not observed. Experimental Section Generalities. CH2Cl2 was purified by use of the Innovative Technology solvent purification system. Anhydrous CH3CN packed in a SureSeal bottle was purchased from Aldrich and left with dry alumina for 24 h before use. Ethyl 2-phenylselanylpropanoate20 and 3,4,5-trifluorobenzenediazonium tetrafluoroborate (FBD)10 were synthesized following literature procedures. Alkenes and alkynes (Aldrich) were sparged with argon, stored in the -40 °C refrigerator in a nitrogen-filled Vacuum Atmospheres glovebox, and passed through dry, neutral alumina before each use. The silicon wafers utilized were obtained from Silicon Quest International Inc. and were 150 mm prime grade Si(100), n-type, with a resistivity of 0.5-1.2 Ω‚cm, and Si(111), n-type, with a resistivity of 10-30 Ω‚cm, unless stated otherwise. (31) Chatgilialoglu, C. In The Chemistry of Sulphenic Acids and Their DeriVatiVes; Patai, S., Ed.; Wiley: Chichester, U.K., 1990; pp 549-569.
Wang and Buriak Techniques. FTIR (Fourier transform infrared) spectra were collected on a Nicolet Nexus 760 spectrometer with a DTGS detector and a nitrogen-purged sample chamber, with 32 scans at 4 cm-1. GATR (germanium attenuated total reflection) IR spectra were collected with a GATR attachment from Harrick, taking 512 scans at 2 cm-1 resolution with a freshly etched [40% NH4F (aq)] Si(111) surface as background. XPS was taken on a Kratos Axis 165 X-ray photoelectron spectrometer in the Alberta Centre for Surface Engineering and Science (ACSES). SIMS (time-of-flight secondary ion mass spectroscopy) depth analysis was obtained on ToF-SIMS IV-100 (ION-TOF GmbH) at ACSES. The pSi sample was sputtered with a 1 kV Cs+ ion source, leading to a 300 × 300 µm2 crater with a central area of 120 × 120 µm2, analyzed by Ga gun. The depth of the hole after 2200 s of sputtering was determined to be 2.8 µm by a Zygo optical profilometer. Electron paramagnetic resonance (EPR) experiments were carried out on a Bruker Elexys E500 spectrometer operating in the X-band mode (9.5 GHz) at 10 K. A free-standing pSi sample (about 15-20 mg) was prepared for EPR to ensure good signal-to-noise ratio. Solution-phase NMR spectra were obtained on an INOVA 400 MHz instrument. All sample preparation for NMR analysis was carried out in a nitrogen-filled glovebox with CD3CN as a solvent. Porous Si Preparation. Prime-grade 150 mm Si(100) (n-type 0.5-1.2 Ω‚cm) wafers were cleaned before use by soaking and rinsing copiously with ethanol. Galvanostatic etching was carried out with a 24% HF/26% H2O/50% ethanol solution prepared from CMOS grade 49% aqueous HF (J. T. Baker). A Teflon sample holder exposing 1.1 or 0.28 cm2 of a silicon electrode, having a backside ohmic contact of heavy aluminum foil in contact with the unpolished side of the silicon wafer, was interchangeably used for the etching procedures, diazonium-mediated chemistry, and FTIR analysis. Unless stated otherwise, pSi samples were prepared by etching the (100) wafer anodically at 7.6 mA cm-2 for 1.5 min and then 76 mA cm-2 for 2 min with illumination. After the etching process, the samples were rinsed with ethanol and dried with N2. The free-standing pSi for EPR experiments was prepared by electrochemical etching of a p++ Si(100) wafer (10-20 mΩ‚cm) wafer in a 24% HF/26% H2O/50% ethanol solution at 37 mA cm-2 for 3 h, recycling the etching solution every 20 min without shorting the current cell, followed by a short high-current etch of 0.4 A/cm2 for 15 s to release the porous layer. After the etching process, the samples were rinsed with ethanol and pentane and dried with N2. Flat Si(111)-H/Si(100)-Hx Preparation. Prime-grade Si(111) and Si(100) wafers were cleaned via standard RCA procedures.32 The H-terminated Si surface was obtained by immersing the cleaned wafers in an aqueous fluoride source for 4-6 min: degassed 40% NH4F (aq, Transene) solution for Si(111)-H or 1% HF (aq) for Si(100)-Hx. The hydride-terminated surfaces were immediately dipped into water for several seconds and then transferred to the nitrogen-filled glovebox for derivation. Reactions with Diazonium Salts. The spontaneous grafting of aryl moieties on pSi was carried out under inert atmosphere by dropping 50 µL of a 50 mM diazonium solution (NBD, BBD, or FBD) in CH3CN on the pSi surface and leaving it for 30-60 min. A home-built glass autoclave was used to prevent solvent evaporation during reaction. Following completion of the reaction, the sample was brought out of the glovebox, rinsed with copious CH3CN, CH2Cl2, and pentane, and dried with N2. Silicon Surface Radical Trapping. All reactions were carried out in a nitrogen-filled glovebox. For the alkenes/alkynes reactions, neat or a 10% CH2Cl2 solution of alkene or alkyne was dropped (5-10 µL) onto the pSi or flat Si surface, followed by ∼40 µL of 100 mM 4-bromobenzenediazonium tetrafluoroborate (BBD) solution in CH3CN. The typical reaction time was 0.5-3 h. Surface radical trapping by alkyl/aryl chalcogenides was carried out in a similar manner except that 5 µL of neat ethyl 2-phenylselanylpropanoate or ethylphenylsulfide was used as the trapping agent. A home-built glass autoclave was used to prevent solvent evaporation during reaction. After completion of the reaction, the sample was brought (32) Kern, W.; Puotinen, D. RCA ReV. 1970, 31, 187.
Trapping Silicon Surface-Based Radicals out of the glovebox, rinsed with copious quantities of CH3CN, CH2Cl2, and pentane, and dried with N2. NMR Spectra. All sample preparations for NMR analysis were carried out in a nitrogen-filled glovebox with CD3CN as a solvent. A freshly prepared pSi sample was immersed in ∼1 mL of a 20 mM FBD solution in CD3CN. After the desired reaction time (24 h), the solution was passed over cotton wool to remove any suspended materials and sent for NMR analysis. The reaction between tris(trimethylsilyl)silane (TTMSS) and FBD in the absence or presence of dodecyne was carried out in a vial: an equimolar quantity of 1-dodecyne (0.025 mL) and TTMSS (0.032 mL) was added to a 20 mM FBD solution in CD3CN (5 mL). The final concentration for each molecule is 20 mM. After 24 h, a small amount of solution was sent for NMR analysis without any purification step.
Langmuir, Vol. 22, No. 14, 2006 6221
Acknowledgment. Generous financial support from the National Research Council (NRC), the National Institute for Nanotechnology, NSERC, CFI, and the University of Alberta is gratefully acknowledged. D.W. is an NSERC Visiting Fellow. The group of Professor D. Clive is thanked for their donation of the selenium reagent utilized in this study, and Greg Nilsson is thanked for assistance with the NMR experiments. Supporting Information Available: FTIR data for all surfaces not shown in the main text, XPS data, and NMR spectra. This material is available free of charge via the Internet at http://pubs.acs.org. LA060653E
