Acid−Base Properties and Zeta Potentials of Self-Assembled

Langmuir 2004, 20, 8693-8698

8693

Acid-Base Properties and Zeta Potentials of Self-Assembled Monolayers Obtained via in Situ Transformations† Jing-Jong Shyue and Mark R. De Guire* Department of Materials Science & Engineering, Case Western Reserve University, 10900 Euclid Avenue, Cleveland, Ohio 44106-7204

Tsuyoshi Nakanishi, Yoshitake Masuda, and Kunihito Koumoto Department of Applied Chemistry, Graduate School of Engineering, Nagoya University, Furo-cho, Chikusa-ku, Nagoya 464-8603, Japan

Chaim N. Sukenik Chemistry Department, Bar-Ilan University, Ramat Gan 52900, Israel Received March 22, 2004. In Final Form: June 15, 2004 Siloxane-anchored, self-assembled monolayers (SAMs) on single crystal Si were prepared with a variety of surface functional groups using a single commercially available surfactant (1-bromo-11-(trichlorosilyl)undecane) followed by in situ transformations. Polar (thioacetate and thiol), nonpolar (methyl), acidic (sulfonic and carboxylic), basic (various amines), and ionic (alkylammonium) surface functionalities were prepared. For primary amine and sulfonate surfaces, the degree of surface charge as a function of pH was determined ex situ using X-ray photoelectron spectroscopy (XPS). Sulfonate SAMs exhibited much higher effective pKa (∼2) than dilute sulfonic acid (-5 to -6), and amine SAMs exhibited much lower pKa (∼3) than dilute organic amines (∼10). This is attributed to the stabilization of nonionized groups by adjacent ionized groups in the SAM. Zeta potentials of these SAMs as a function of pH were consistent with the XPS results and indicated that ionizable SAM surfaces can generate surface potentials much higher than those of nonionic SAMs (thioacetate, methyl) and typical oxide surfaces.

1. Introduction As close-packed, highly ordered arrays of long-chain hydrocarbon molecules anchored to a solid substrate,1 selfassembled organic monolayers (SAMs) represent one type of organically functionalized surfaces that have been used to promote deposition of inorganic oxide and nonoxide thin films from aqueous media at low temperatures. (For a recent review, see ref 2.) Several studies have reported a dependence of film formation on the type of SAM surface functionality3-18 as well as on the pH, temperature, and * To whom correspondence should be addressed. Telephone: +1(216)368-4221. E-mail: [email protected]. † Based in part on a thesis submitted for the M.S. degree in Materials Science and Engineering of J.-J. Shyue, Case Western Reserve University, 2002. (1) Ulman, A. Chem. Rev. 1996, 96, 1533-1554. (2) Niesen, T. P.; De Guire, M. R. J. Electroceram. 2001, 6, 169-207. (3) Rieke, P. C.; Tarasevich, B. J.; Wood, L. L.; Engelhard, M. H.; Baer, D. R.; Fryxell, G. E. Langmuir 1994, 10, 619-622. (4) Bunker, B. B.; Rieke, P. C.; Tarasevich, B. J.; Campbell, A. A.; Fryxell, G. E.; Graff, G. L.; Song, L.; Virden, J. W.; McVey, G. L. Science 1994, 264, 48-55. (5) Shin, H.; Collins, R. J.; De Guire, M. R.; Heuer, A. H.; Sukenik, C. N. J. Mater. Res. 1995, 10, 699-703. (6) Meldrum, F. C.; Flath, J.; Knoll, W. Langmuir 1997, 13, 20332049. (7) Nagtegaal, M.; Stroeve, P.; Tremel, W. Thin Solid Films 1998, 327-329, 571-575. (8) Nagtegaal, M.; Stroeve, P.; Ensling, J.; Gutlich, P.; Schurrer, M.; Voit, H.; Flath, J.; Kashammer, J.; Knoll, W.; Tremel, W. Chem. Eur. J. 1999, 5, 1331-1337. (9) Meldrum, F.; Flath, J.; Knoll, W. J. Mater. Chem. 1999, 9, 711724. (10) Meldrum, F. C.; Flath, J.; Knoll, W. Thin Solid Films 1999, 348, 188-195.

concentration of the source solution.2,15,19 In several cases, electrostatic interactions between the SAM and the depositing inorganic materials appear to have a strong influence on the occurrence or rate of film growth. Therefore, the range of types and degrees of surface charge that are possible with various surface functionalities should enable considerable control to be demonstrated over inorganic film formation. Acidic or basic molecules confined to a surface may exhibit different degrees of reactivity than when in solution. A recent FTIR study of carboxylate SAMs20 reported that they exhibit two values of pKa (4.9 ( 0.4 and 9.3 ( 0.2). The lower value corresponds to the deprotonation of single acid groups (i.e., hydrogen bonded only (11) Koumoto, K.; Seo, S.; Sugiyama, T.; Seo, W. S.; Dressick, W. J. Chem. Mater. 1999, 11, 2305-2309. (12) Aizenberg, J.; Black, A. J.; Whitesides, G. M. J. Am. Chem. Soc. 1999, 121, 4500-4509. (13) Niesen, T. P.; Wolff, J.; Bill, J.; De Guire, M. R.; Aldinger, F. In Organic-Inorganic Hybrid Materials II; Klein, L. C., Francis, L. F., De Guire, M. R., Mark, J. E., Eds.; Materials Research Society: Warrendale, PA, 1999; Vol. 576, pp 197-202. (14) Kovtyukhova, N. I.; Buzaneva, E. V.; Waraksa, C. C.; Martin, B. R.; Mallouk, T. E. Chem. Mater. 2000, 12, 383-389. (15) Pizem, H.; Sukenik, C. N.; Sampathkumaran, U.; De Guire, M. R. Chem. Mater. 2002, 14, 2476-2485. (16) Masuda, Y.; Wakamatsu, S.; Koumoto, K. J. Eur. Ceram. Soc. 2004, 24, 301-307. (17) Masuda, Y.; Sugiyama, T.; Lin, H.; Seo, W. S.; Koumoto, K. Thin Solid Films 2001, 382, 153-157. (18) Masuda, Y.; Saito, N.; Hoffmann, R.; De Guire, M. R.; Koumoto, K. Sci. Technol. Adv. Mater. 2003, 4, 461-467. (19) Tarasevich, B. J.; Rieke, P. C.; Liu, J. Chem. Mater. 1996, 8, 292-300. (20) Gershevitz, O.; Sukenik, C. N. J. Am. Chem. Soc. 2004, 126, 482-483.

10.1021/la049247q CCC: $27.50 © 2004 American Chemical Society Published on Web 08/27/2004

8694

Langmuir, Vol. 20, No. 20, 2004

to surrounding water molecules), while the higher value corresponds to the deprotonation of dimers and oligomers formed among adjacent surface-bound molecules. The present study uses ex situ X-ray photoelectron spectroscopy (XPS) and zeta potential measurements on sulfonate and amine surfaces to measure their degree of deprotonation or protonation after exposure to solutions of various pH. Such studies not only can quantify the reactivity and surface potential of these SAMs for the design of film deposition processes but also may provide further insight into the interrelationship between reactivity and structural configuration of such relatively concentrated surfacebound molecular assemblies. Due to the high chemical reactivity of the trichlorosilane (-SiCl3) group to nucleophiles, siloxane-anchored sulfonate- or amine-terminated SAMs are usually obtained via in situ transformations of SAMs with less reactive surface functionalities, i.e., after the trichlorosilane has reacted fully with the substrate to form the siloxane anchoring network. (For a recent review of in situ transformations of SAMs, see ref 21.) In the present work, a wide variety of surfaces were obtained, starting with a single surfactant. The relatively rugged nature of the siloxane attachment to oxide surfaces (such as the native oxide on silicon used here) permits transformations to be carried out under more aggressive conditions of pH and/ or temperature than can be used with, e.g., thiol-anchored SAMs. 2. Experimental Procedures General Procedure. X-ray photoelectron spectra were recorded on a PHI model 5600 MultiTechnique System using monochromated 200 W Al X-rays for 5 min. (These analysis conditions are expected to have a negligible effect on the surfaces being analyzed here, based on studies of damage of siloxaneanchored SAMs during XPS measurements.22) Peak positions were referenced to silicon single crystal (100) 2p at 99.7 eV.23,24 The electron takeoff angle was fixed at 45°. A Rame-Hart goniometer was used for contact angle measurements. Advancing contact angles were determined by placing a drop of distilled water (pH ∼5, ∼3 µL) on the sample with a microsyringe and advancing the volume (∼2 µL more), keeping the area in contact with the substrate constant and leaving the syringe in the drop. Receding contact angles were determined by withdrawing the water until the lowest angle was achieved, without changing the area of the drop in contact with the substrate. The value reported was the average of at least 10 individual measurements. Zeta potentials of SAMs were measured on a Photal ELS7300K system at a field of 7.5 V/cm (40 V), following the procedure of Hozumi et al.25 The monitoring particle was polystyrene latex, with average diameter of 400.0 ( 72.4 nm as determined by dynamic light scattering. The zeta potential was calculated using the Smoluchowski method. The system error was (10 mV. Deposition of SAMs. Oxidized silicon surfaces (Si-OH) were cleaned as described previously.26 Alkyl bromide SAMs (-C11-Br) were deposited as previously reported,26 except that a commercially available surfactant (1-bromo-11-(trichlorosilyl)undecane, Gelest) was used. As a reference surface, methyl(21) Sullivan, T. P.; Huck, W. T. S. Eur. J. Org. Chem. 2003, 17-29. (22) Frydman, E.; Cohen, H.; Maoz, R.; Sagiv, J. Langmuir 1997, 13, 5089-5106. (23) Wagner, C. D., Bickham, D. M., Eds. NIST X-ray Photoelectron Spectroscopy Database; 1.0 ed.; U.S. Secretary of Commerce: Washington, DC, 1989. (24) Moulder, J. F.; Stickle, W. F.; Sobol, P. E.; Bomben, K. D. In Handbook of X-ray Photoelectron Spectroscopy: A Reference Book of Standard Spectra for Indentification and Interpretation of XPS Data; Chastain, J., Ed.; Physical Electronics: Chanhassen, MN, 1995; pp 213-242. (25) Hozumi, A.; Sugimura, H.; Yokogawa, Y.; Kameyama, T.; Takai, O. Colloids Surf., A 2001, 182, 257-261. (26) Balachander, N.; Sukenik, C. N. Langmuir 1990, 6, 1621-1627.

Shyue et al. terminated SAMs were produced from octadecyltrichlorosilane (OTS) following Maoz and Sagiv.27 In Situ Transformations of Monolayer Functionality. Unless otherwise noted, all SAM-coated substrates (-C11-Br, or transformed as described below) were washed prior to further use with ethanol and dried with a stream of argon and then thoroughly cleaned with fresh chloroform-soaked tissues until the surfaces were perfectly mirrorlike. Undecyl Thioacetate Surface (-C11-SCOCH3). -C11-Br substrates were immersed in a slowly stirred solution of 0.7 g of potassium thioacetate in 10 mL of dimethyl sulfoxide (DMSO) in a sealed pressure tube at 80 °C overnight. Undecyl Sulfonate Surface (-C11-SO3H). -C11-SCOCH3 surfaces were immersed in a slowly stirred solution of saturated oxone (2KHSO5‚KHSO4‚K2SO4) at room temperature overnight.26 Undecyl Thiol Surface (-C11-SH). -C11-SCOCH3 substrates were immersed in a slowly stirred 1 M HCl aqueous solution at room temperature overnight. Undecyl Nitrile Surface (-C11-CN). -C11-Br substrates were immersed in a slowly stirred solution of 0.5 g of sodium cyanide in 10 mL of DMSO in a sealed pressure tube at 80 °C overnight. Dodecanoic Acid Surface (-C11-COOH). -C11-CN substrates were immersed in a slowly stirred solution of 0.5 g of sodium bicarbonate in 10 mL of distilled water in a sealed pressure tube at 40 °C for 2 days. Substrates were then washed with 1% hydrochloric acid prior to the normal cleaning procedure. Undecyl Azide Surface (-C11-N3). -C11-Br substrates were immersed in a solution of sodium azide in dimethylformamide.26 Undecylamine Surface (-C11-NH2). -C11-N3 substrates were immersed in a solution of lithium aluminum hydride in diethyl ether.26 Undecyldimethyl Amine Surface (-C11-N(CH3)2). Route I: -C11-NH2 substrates were immersed in a slowly stirred solution of 5 mL of aqueous 37% formaldehyde (HCOH) and 5 mL of 46% formic acid (HCOOH) solution at room temperature overnight. Route II: -C11-Br substrates were immersed in a slowly stirred solution of 10 mL of aqueous 45% dimethylamine solution at room temperature overnight. Route III: Undecyltrimethylammonium salt-functionalized substrates (-C11-N+(CH3)3; see below) were immersed in 5 mL of slowly stirred trimethylamine in a sealed pressure tube at 80 °C overnight. Undecyltrimethylammonium Salt Surface (-C11-N+(CH3)3/ SO4CH3-). -C11-NH2 or -C11-N(CH3)2 substrates were immersed in 5 mL of slowly stirred dimethyl sulfate in a sealed pressure tube at 80 °C overnight. Determining the Acid/Base Properties. To determine the degree of protonation or deprotonation of SAMs as a function of pH, -C11-NH2 and -C11-SO3H substrates were immersed in aqueous solutions of different pH (adjusted using HCl or NaOH). The solutions used for the -C11-SO3H substrates were 0.2 M NaCl. The samples were soaked and ultrasonically agitated in these various solutions for ∼30 min at room temperature. All specimens were then dried with a stream of argon (no ethanol rinse), cleaned with chloroform-soaked tissues, and then examined immediately (within 1 min) using XPS (N 1s peaks for amine surfaces; S 2s and Na 1s peaks for sulfonate surfaces).

3. Results and Discussion In Situ Transformation of Surface Functionality. Table 1 summarizes the transformations used here, the contact angles of water with the surfaces, and the positions of characteristic XPS peaks for each surface. The results are in good agreement with literature values for SAMs formed by direct deposition and by in situ transformations.26-32,34 The XPS data identified the expected functional groups at each stage of the transformations. The wetting data show the expected relative hydrophobicity and/or hydrophilicity. In some cases the advancing (27) Maoz, R.; Sagiv, J. J. Colloid Interface Sci. 1984, 100, 465-496. (28) De Guire, M. R.; Niesen, T. P.; Supothina, S.; Wolff, J.; Bill, J.; Aldinger, F.; Ruhle, M. Metallkd. 1998, 89, 758-766. (29) Agarwal, M.; De Guire, M. R.; Heuer, A. H. J. Am. Ceram. Soc. 1997, 80, 2967-2981.

Properties of Siloxane-Anchored SAMs

Langmuir, Vol. 20, No. 20, 2004 8695

Table 1. Summary of Surfaces Studied in This Work, Their Contact Angles with Water, and XPS Results surface functionalitya Si-OH -C18-H -C11-Br -C11-SCOCH3 -C11-SO3H -C11-SH -C11-CN -C11-COOH -C11-N3 -C11-NH2 -C11-N(CH3)2 -C11-N+(CH3)3 a

reagent

transformation

Piranha Cl3SiC18H36H Cl3SiC11H22Br CH3COSK oxone 1 M HCl NaCN 5% NaHCO3 NaN3 LiAlH4 HCOH/HCO2H (CH3)2NH (C2H5)3N (CH3O)2SO2

Si f SiO2 alkyl SAM deposition alkyl bromide SAM deposition -Br f -SCOCH3 -SCOCH3 f -SO3H -SCOCH3 f -SH -Br f -CN -CNf -COOH -Br f -N3 -N3 f -NH2 -NH2 f -N(CH3)2 -Br f -N(CH3)2 -N+(CH3)3 f -N(CH3)2 -NH2-x(CH3)x f -N+(CH3)3

advancing angle, θa (deg)

receding angle, θr (deg)

35 ( 1 108 ( 1 87 ( 1 72 ( 4 32 ( 2 66 ( 2 76 ( 4 46 ( 2 76 ( 2 70 ( 1 74 ( 1

15 ( 1 100 ( 1 72 ( 1 66 ( 5 11 ( 3 37 ( 3 66 ( 5 14 ( 3 58 ( 1 40 ( 2 28 ( 2

70 ( 1

36 ( 1

XPS (eV) Si 2p 103.3 Br 3d 70.2 S 2p 164.8, 165.6 S 2p 168.8 S 2p 164.8, 165.6 N 1s 400.6 N 1s 405.7, 401.9 N 1s 401.1 N 1s 401.0 N 1s 403.4

On Si (100) substrates.

angles and/or the hysteresis (difference between advancing and receding angles) are higher than expected. To take advantage of commercially available material, the surfactant used here had a shorter hydrocarbon chain (C11) than is optimal (C16-C18) for obtaining highly ordered close-packed monolayer surfaces.33 Thus, the present surfaces are less well packed, leading to larger contact angle hysteresis. The contact angle of the tetraalkylammonium SAM is relatively high (70°) for a salt, as might be expected from the relative hydrophobicity of tetraalkylammonium salts. The wetting behavior of the amine SAMs reflects their partial protonation (see below) near neutral pH. The in situ transformation of bromide to thioacetate was based on the previously reported procedure,26 using DMSO instead of ethanol as the solvent. This change improves the nucleophilicity of the thioacetate anion. The reaction goes to completion: the Br 3d XPS peak at 72.1 eV disappeared and the S 2p signal appeared at 164.8 and 165.6 eV for 2p3 and 2p1, respectively. The contact angles are comparable to those reported for SAMs formed directly from a thioacetate surfactant.17,26,34 The thioacetate group can be hydrolyzed to thiol in acid solution. The C 1s peak in the XPS spectrum of the -C11-SCOCH3 surface showed a main signal at 285.5 eV from the alkyl C of the chain and a small but distinct shoulder at 288.5 eV, which can be assigned to the carbonyl C of the thioacetate group.23 After immersion in acid, this shoulder disappeared, consistent with hydrolysis of the thioacetate group. The newly formed thiol surface was more hydrophilic and showed a larger contact angle hysteresis, consistent with thiol surface functional groups. On the basis of this hydrolysis chemistry, it is clear that reports of deposition of tin oxide35 and zirconia36 films on thioacetate SAMs under strongly acidic conditions can be attributed more to the behavior of thiol, rather than (30) Sampathkumaran, U.; Supothina, S.; Wang, R.; De Guire, M. R. In Mineralization in Natural and Synthetic Biominerals (Proceedings of Symposium DD, MRS Fall Meeting, Boston, Massachusetts, 29 November - 3 December 1999; MRS Symposium Proceedings 599; Li, P., Calvert, P., Levy, R. J., Kokubo, T., Scheid, C. R., Eds.; Materials Research Society: Warrendale, PA, 2000; pp 177-188. (31) Baker, M. V.; Watling, J. D. Tetrahedron Lett. 1995, 36, 46234624. (32) Margel, S.; Sivan, O.; Dolitzky, Y. Langmuir 1991, 7, 23172322. (33) Schreiber, F. Prog. Surf. Sci. 2000, 65, 151-257. (34) Collins, R.; Sukenik, C. N. Langmuir 1995, 11, 2322-2324. (35) Supothina, S.; De Guire, M. R.; Heuer, A. H. J. Am. Ceram. Soc. 2003, 86, 2074-2081. (36) Niesen, T. P.; De Guire, M. R.; Bill, J.; Aldinger, F.; Ruhle, M.; Fischer, A.; Jentoft, F. C.; Schlogl, R. J. Mater. Res. 1999, 14, 24642475.

thioacetate, surfaces.34,37 Further confirmation of thiol formation comes from rinsing the substrate with 1 N NaOH. Under these conditions a Na 1s peak appeared in XPS, consistent with the expected behavior of surface thiols and their transformation into -SNa. The in situ transformation to cyanide also appears to have been complete using DMSO as solvent. (Margel et al. reported that this reaction cannot be done in ethanol solution.32) The Br 3d XPS signal at 72.1 eV disappeared, and the N 1s signal appeared at 400.6 eV with similar intensity to that of an amine. The hydrolysis of cyanide to carboxylate was hard to control. It is known that the cyanide group can be hydrolyzed to amide and then further to carboxylic acid in acidic or basic solutions.38 However, the siloxane network of the SAM is also prone to hydrolysis. For example, in 1% NaOH aqueous solution at room temperature overnight, the C 1s XPS peak was suppressed heavily, suggesting that the SAMs were at least partially stripped from the substrates. On the other hand, in relatively mild hydrolysis conditions such as 1% sodium bicarbonate at room temperature for 2 days, the contact angle changed to 65 ( 1° advancing, 46 ( 1° receding, but the N 1s XPS peak remained. This suggests incomplete hydrolysis and that cyanide and/or amide were still present along with some surface carboxylate. Using 5% sodium bicarbonate aqueous solution at 40 °C for 2 days eliminated the N 1s XPS peak, while the C 1s intensity remained. Furthermore, the protons from the carboxylic acid groups could be replaced by sodium ions by rinsing with sodium bicarbonate solution, consistent with the formation of carboxylate SAMs. Additional evidence for both in situ thiol and carboxylate formation was obtained through their zeta potentials as discussed below. Different degrees of methylation of amines (primary, secondary, and tertiary) can be achieved using different chemistries. The Eschweiler-Clarke reaction39 using formaldehyde and formic acid can produce either partially or fully methylated amine (-NH2-x(CH3)x). Dimethylamine was reacted with bromide SAMs to yield either -N(CH3)2 or nitrogen-bridged ammonium (-Cn-N+(CH3)2-Cn-) SAMs. An alkylammonium SAM was dem(37) Collins, R. J. Functionalized Self-Assembled Monolayers as Templates for Mineral Oxide Thin Film Deposition. Thesis, Department of Chemistry, Case Western Reserve University: Cleveland, OH, 1997. (38) Larock, R. C. Comprehensive Organic Transformation: A Guide to Functional Group Preparations, 2nd ed.; John Wiley and Sons: New York, 1999. (39) March, J. Advanced Organic ChemistrysReactions, Mechanisms, and Structure, 5th ed.; John Wiley and Sons: New York, 1992.

8696

Langmuir, Vol. 20, No. 20, 2004

Shyue et al.

Figure 1. XPS of amine (left) and sulfonate (right) SAMs after treatment with different pH solutions.

ethylated using triethylamine (see below). This process likely yields the tertiary amine (-N(CH3)2) but could also give some alkene (-CdC) via Hoffman elimination. Since the only common product of these three reactions is -N(CH3)2 and the product of all three reactions showed the same properties, it is inferred to be the same product, the tertiary amine. The methylation using dimethyl sulfate to make tetraalkylammonium salt SAMs on Si has not been reported, though Aizenberg et al. reported using thiol-anchored alkylammonium salt SAMs12 on gold or silver substrates. The contact angle of alkylammonium salt SAMs did not vary with pH (1 < pH < 9). In XPS, the position of the N 1s peak (403.4 eV) was also invariant. In a useful exchange reaction, -C11-N+(CH3)3 SAMs were demethylated with triethylamine. The alkylammonium surface effectively acts as a methylating agent (the triethylamine is methylated) and the SAMs are transformed to alkyldimethylamine SAMs (-N+(CH3)3 + N(C2H5)3 f -N(CH3)2 + (CH3)N+(C2H5)3). As stated above, the specimens made by this route were indistinguishable in their wetting behavior and XPS spectra from those made via replacement of bromide by dimethylamine. Intrinsic Surface Charge. Figure 1 shows the N 1s peak of -NH2 SAMs on Si after rinsing with aqueous solutions of different pH. Two peak positions (400 and 403 eV) are evident. These can be assigned to RNH2 and RN+H3,23 respectively. The area fraction of the 403-eV peak (A403/(A400 + A403)) represents the degree of protonation. First, we note that the degree of protonation of surface amine changed over a wide range of pH, unlike the simple acid/base behavior of amines in solution (completely protonated at pH < 9). Second, whereas the pKa for amine in dilute solution is typically ∼10,40 the surface amine was 50% protonated at pH ∼1, and pH below 0 was needed to achieve >90% protonation. At least part of this difference between surface amines and amines in dilute solution is attributable to the relatively dense packing of the surface amines. Once a surface amine group is protonated, its (40) Brown, H. C.; McDaniel, D. H.; Haflinger, O. Determination of Organic Structures by Physical Methods; Academic Press: New York, 1955; Vol. 1.

positive charge can suppress the protonation of neighboring groups. Also, poor solvation and bridging between neighboring charged groups would also suppress further protonation. (For further discussion, see below.) The degree of deprotonation of sulfonate SAMs after immersion in 0.2 M sodium chloride solutions of different pH was calculated as the atomic ratio of sulfur to sodium, i.e., NSO3Na/(NSO3H + NSO3Na) ) ANa/AS ≡ fraction of deprotonated sulfonate, where Ny denotes the number of species y and Ay denotes the atomic fraction of species y. The pKa of a sulfonate SAM was much higher than that of sulfonates in dilute solution (2 vs ∼-7) and full deprotonation was not observed until pH ∼ 5. Upward shifts of 4-6 decades in pKa have been reported for micellized fatty acids,41 in Langmuir films of fatty acids,42 and in geometrically constrained diacids.43 As with amine SAMs, it appears that the high concentration of surface groups in a sulfonate SAM tends to suppress the natural acid-base chemistry of the surface groups. This argument has been invoked in in situ FTIR studies of carboxylate surfaces.20,44 However, since sulfonate is a strong acid while amine is a weak base, the surface charge of sulfonate SAMs would still be much higher than that of amine SAMs for pH > 2 (Figure 2). While the direction of the shifts in pKa reported here is consistent with the constraints imposed by concentration of groups on a surface, the magnitude of the shifts here is unprecedentedly high. In this respect, we note that the present approach for measuring the surface charge of SAMs was ex situ. It presumes that the condition of the surface in the high vacuum of the XPS chamber was representative of that in solution. To check for possible artifacts, spectra were collected from sulfonate SAMs exposed to pH 2.5, 0.2 M NaCl solution under various experimental conditions. There were no differences be(41) Kanicky, J. R.; Shah, D. O. J. Colloid Interface Sci. 2002, 256, 201-207. (42) Miranda, P. B.; Du, Q.; Shen, Y. R. Chem. Phys. Lett. 1998, 286, 1-8. (43) Rebek, J. J.; Duff, R. J.; Gordon, W. E.; Parris, K. J. Am. Chem. Soc. 1986, 108, 6068-6069. (44) Cheng, S. S.; Scherson, D. A.; Sukenik, C. N. Langmuir 1995, 11, 1190-1195.

Properties of Siloxane-Anchored SAMs

Langmuir, Vol. 20, No. 20, 2004 8697

Figure 2. Degree of surface charge of amine and sulfonate SAMs, as determined from XPS data of Figure 1.

Figure 3. Logarithm of the concentration ratio of amine to ammonium as a function of pH.

tween samples agitated ultrasonically in the solution for 1, 10, and 30 min, nor if the substrate was cleaned 10 times with a chloroform soaked tissue prior to XPS analysis, nor after exposing the specimen to air for 30 min prior to XPS examination. In contrast, substrates immersed for less than 1 min, without agitation, did show a significantly smaller Na/S ratio than the others. Therefore, it was concluded that a quasi-equilibrium was achieved during the 30-min immersion used for the specimens shown in Figures 1 and 2 and that the handling procedures reported herein were reliable and reproducible. The results of Figures 1 and 2 can be further analyzed using the Henderson-Hasselbalch equation. For species in dilute solution Figure 4. Effective pKa of ionized surface group as a function of pH.

-NH3+ f -NH2 + H+ Ka )

groups are in fact subject to an effective dissociation constant, Kaeff, defined45 so that

[H+][-NH2] [-NH3+]

-log Ka ) -log[H+] - log

pKaeff ) pH - log[θB/θA] [-NH2] +

[-NH3 ]

and

-SO3H f -SO3- + H+ Ka )

[H+][-SO3-] [-SO3H]

-log Ka ) -log[H+] - log

w pH ) pKa + log

[-SO3-] [-SO3H] θB θA

where θB is the fraction of the base form of the species (-NH2 or -SO3-) and θA is the fraction of the acid form of the species (-NH3+ or -SO3H). For species in dilute solution, a plot of pH vs log[θB/θA] (Figure 3) would have a slope of 1 and a y-intercept of pKa. However, data recorded in this study yielded a slope of 10 and a y-intercept of 1.4 for amines and a slope of 5.6 and a y-intercept of 2.42 for sulfonate. This implies that the surface-confined

Figure 4 shows a plot of pKaeff (calculated from experimental values of pH and the values of θB and θA measured from XPS) vs pH. Whereas pKa would be independent of pH for dilute species, Figure 4 shows that pKaeff is a function of pH for groups concentrated on a surface. This behavior can be qualitatively understood in terms of the suppression of ionization of groups adjacent to groups that are already ionized. The effective pKa defined here provides a way to characterize this behavior quantitatively for various surfaces. The results for amine and sulfonate surfaces from the present work fall on the same line for pKaeff vs pH, while the results reported by Cheng et al.44 for carboxylate SAMs (dashed line, Figure 4) show a similar trend but to a lesser degree. Zeta Potentials of SAMs. The hydrolyzed oxide surface of the Si substrates exhibited an isoelectric point of 3, consistent with published values for amorphous silica.46 At higher pH, this surface generated a negative potential that increased steadily to -118 mV at pH 11 (Figure 5). The methyl SAMs showed weak negative zeta potential (-30 to -50 mV) at pH g ∼4 (Figure 5). This suggests that the methyl-terminated surface preferentially adsorbed negative ions, since carbon anions cannot be (45) Holmes-Farley, S. R.; Reamey, R. H.; McCarthy, T. J.; Deutch, H.; Whitesides, G. M. Langmuir 1985, 1, 725-740. (46) Reed, J. S. Principles of Ceramics Processing; John Wiley & Sons: New York, 1995.

8698

Langmuir, Vol. 20, No. 20, 2004

Figure 5. Zeta potentials of nonionic SAMs.

Shyue et al.

pH, consistent with the weak acidity of -COOH groups. This result was similar to carboxylate SAMs on gold reported by Zhu et al.50 Cheng et al.44 reported that carboxylate SAMs at pH 3, 5, 7, 9, and 11 exhibited about 0%, 10%, 30%, 85%, and 100% deprotonation, respectively. The present results corroborate their findings: i.e., at pH 3, the zeta potential was zero; and with rising pH, the zeta potential became increasingly negative (-85 mV at pH 11, nearly that of sulfonate SAMs, -100 mV). As observed by Zhu et al.,50 amine SAMs exhibited positive zeta potential at low pH. However, here it decreased with increasing pH and became negative above neutral pH. This may be due to the presence of a Stern layer formed by negatively charged counterions, overcompensating the positive surface charge. Cuvillier et al. have reported evidence of Stern layers on amine-functionalized Langmuir-Blodgett films.51 Additional evidence52 supporting this interpretation is presented elsewhere. 4. Conclusion

Figure 6. Zeta potentials of ionic SAMs.

produced in aqueous solution. In general, preferentially absorbed ions are mostly negatively charged, because cations are more easily hydrated and retained in solution.47-49 Similar observations were reported by Hozumi et al.25 Thioacetate SAMs showed similar behavior to OTS SAMs as expected, since neither is ionizable and only preferential adsorption is possible. The thiol group is a very weak acid (and a polar group) and would therefore be expected to show a weak negative zeta potential, but stronger than that of thioacetate, at high pH, as observed (Figure 5). Thus, though XPS cannot distinguish the difference between thioacetate and thiol groups, the difference in zeta potential at pH g 7 gives further evidence that the transformation of thioacetate to thiol (Table 1) occurred. The zeta potential of alkylammonium SAMs (Figure 6) was large and positive (80-110 mV) at all values of pH studied. This is consistent with the behavior of alkylammonium as a cationic group. In XPS, the position of the N 1s peak was consistent with fully charged nitrogen, independent of pH. The zeta potential of sulfonate SAMs was negative even at low pH (-50 mV at pH 3) and became more negative with increasing pH (-100 mV at pH 11), as expected from its strong acidity and from Figure 2 (about 50% charged at pH 2 and fully ionized at pH above 5). The zeta potentials of carboxylate SAMs were near neutral at pH ) 3 and became negative with increasing (47) Hunter, R. J. Foundations of Colloid Science; Clarendon Press: Oxford, 1987; Vol. 1. (48) Hiemenz, P. C.; Rajagopalan, R. Principles of Colloid and Surface Chemistry, 3rd ed.; Marcel Dekker: New York, 1997. (49) Shaw, D. J. In Introduction to Colloid and Surface Chemistry, 3rd ed.; Butterworth-Heinemann: Oxford, 1992; pp 148-162.

The acid-base character of amine- and sulfonate-SAMs, as revealed by the present XPS studies, differs significantly from that of these groups in dilute solution, with the observed pKa differing by about 7 units from that of a dilute system. Moreover, the degree of protonation or deprotonation changed more gradually with pH than in solution. In most acidic environments, both surfaces are strongly (but oppositely) charged. The zeta potentials of ionic SAMs ((50-100 mV range) were much higher than nonionic SAMs ((0-40 mV). These zeta potentials can be expected to cause strong interactions between SAMs and ceramic particles in aqueous solution, which may account for some of the reported effects of functionalized surfaces on the formation of ceramic thin films from aqueous media.3-18 Nonionic OTS and thioacetate SAMs showed a moderate negative zeta potential, suggesting preferential adsorption of negatively charged species. The weakly acidic thiol SAMs showed a moderate negative zeta potential in strongly basic environments. On the other hand, alkylammonium SAMs showed a strong positive zeta potential at all pH, consistent with their cationic nature. The strongly acidic sulfonate SAMs showed highly negative surface charge at high pH and moderate negative surface charge even at pH 3. This result is supported by our XPS studies. In contrast, the weakly acidic carboxylate SAMs showed negative surface charge only in mildly acidic and basic environments. Although the intrinsic surface charge of amine is positive, the zeta potential of amine switched from positive to negative values as the pH shifted from acidic to basic (point of zero charge around pH 7.5). Acknowledgment. Sponsorships of the U.S. National Science Foundation through Grant Nos. DMR 9803851 and DMR 0203655, and Nature COE Open Cluster Project of the 21st Century of Nagoya University are gratefully acknowledged. C.N.S. acknowledges support of the Minerva Center for Microscale and Nanoscale Particles and Films as Tailored Biomaterial Interfaces. LA049247Q (50) Zhu, P.; Masuda, Y.; Yonezawa, T.; Koumoto, K. J. Am. Ceram. Soc. 2003, 86, 782-790. (51) Cuvillier, N.; Rondelez, F. Thin Solid Films 1998, 327-329, 19-23. (52) Shyue, J.-J. Deposition of Vanadium(V) Oxide Thin Films on Nitrogen-Containing Self-Assembled Monolayers. Thesis. Department of Materials Science and Engineering, Case Western Reserve University, Cleveland, OH, 2002.