pubs.acs.org/Langmuir © 2011 American Chemical Society
Effect of Chain Length on the Assembly of Mercaptoalkanoic Acid Multilayer Films Ligated through Divalent Cu Ions Steven Johnson,* Jocelyn Chan, David Evans, A. Giles Davies, and Christoph W€alti* School of Electronic and Electrical Engineering, University of Leeds, Leeds LS2 9JT, U.K. Received September 17, 2010. Revised Manuscript Received December 14, 2010 The structural stability of alkenthiolate monolayers assembled on gold surfaces is a result of the well-defined organization of the individual molecules within the film. The formation of three-dimensional films assembled by stacking multiple molecular monolayers is substantially more challenging because the correct organization of the molecular components is required not only within the individual monolayers but also between the monolayers of the film. In this paper we examine the structure of multilayer films based on mercaptoalkanoic acid monolayers in which ligation between adjacent monolayers is achieved using the interaction of carboxylic acid and thiol groups with a divalent Cu ion. Using contact angle analysis and atomic force microscopy, we show that the use of Cu2þ has profound implications on the properties and structure of the multilayer film. In particular, the divalent ions effectively prohibit the complete assembly of the next monolayer. For multilayer SAMs assembled from short alkane chains with six methylene groups, we find that molecules in the incomplete adlayer organize themselves randomly over the underlying monolayer. However, as the number of methylene groups increases (11 and 16 methylene groups), the upper layer tends to fracture into discrete islands which cover around 50% of the surface. The height of these islands is found to be equal to that expected for a complete, well-ordered monolayer assembled from the equivalent mercaptoalkanoic acid molecules. This relationship between chain length and island growth results from the migration of molecules into ordered aggregates driven by the reduction of free energy associated with maximizing intermolecular interactions.
Introduction Since the discovery that alkanethiols spontaneously form highly organized films on gold surfaces,1,2 these self-assembled materials have been employed in a wide range of applications3 from antifouling surface coatings4 and photoresists for lithography and patterning applications5-8 to molecular masking layers for nanoscale functionalization.9,10 The self-assembly process results in the formation of an essentially two-dimensional layer with a thickness determined by the length of the polymethylene backbone and the tilt angle between the substrate surface and the molecules within the layer. For many applications, however, it would be desirable to extend assembly into the third dimension, allowing the formation of high-quality ordered molecular films of increased thickness and the spontaneous assembly of complex three-dimensional objects structured on the molecular scale. One approach to achieve the assembly of 3D molecular layers is to create multilayer films by stacking multiple self-assembled monolayers. *Corresponding authors. E-mail:
[email protected] (S.J.); c.walti@ leeds.ac.uk (C.W.).
(1) Nuzzo, R. G.; Allara, D. L. J. Am. Chem. Soc. 1983, 105, 4481–4483. (2) Dubois, L. H.; Nuzzo, R. G. Annu. Rev. Phys. Chem. 1992, 43, 437–463. (3) Love, J. C.; Estroff, L. A.; Kreibel, J. K.; Nuzzo, R. G.; Whitesides, G. M. Chem. Rev. 2005, 105, 1103–1169. (4) Valiokas, R.; Svedhem, S.; Svensson, S. C. T.; Liedberg, B. Langmuir 1999, 15, 3390–3394. (5) Prompinit, P.; Achalkumar, A. S.; Han, X. J.; Bushby, R. J.; W€alti, C.; Evans, S. D. J. Phys. Chem. C 2009, 113, 21642–21647. (6) Johnson, S.; Evans, D.; Davies, A. G.; Linfield, E. H.; W€alti, C. Nanotechnology 2009, 20, 155304. (7) Piner, R. D.; Zhu, J.; Xu, F.; Hong, S. H.; Mirkin, C. A. Science 1999, 283, 661–663. (8) Saavedra, H. M.; Mullen, T. J.; Zhang, P.; Dewey, D. C.; Claridge, S. A.; Weiss, P. S. Rep. Prog. Phys. 2010, 73, 036501. (9) W€alti, C.; Wirtz, R.; Germishuizen, W. A.; Bailey, D. M. D.; Pepper, M.; Middelberg, A. P. J.; Davies, A. G. Langmuir 2003, 19, 981–984. (10) Evans, D.; Johnson, S.; Laurenson, S.; Davies, A. G.; Ko Ferrigno, P.; W€alti, C. J. Biol. 2008, 7(1), 3.
Langmuir 2011, 27(3), 1033–1037
Among the methods developed for the assembly of multilayer films, the most extensively studied are those based on LangmuirBlodgett (LB) monolayer transfer techniques.11,12 Although well established, it has been shown that multilayer films assembled via LB methods tend to suffer from problems of reproducibility, are susceptible to defects, and are limited in the range of potential substrates onto which a molecular layer can be assembled. Alternative approaches, for example based around ionic interactions between individual monolayers, have thus been developed. Mallouk et al.13 demonstrated the construction of multilayer films by exploiting the strong ionic bond between zirconium ions and organic phosphonates. Multilayer films assembled from up to eight individual monolayers were demonstrated, and the application of these multilayer films has been explored in areas from molecular and chemical recognition14 to novel materials for electronic devices.15 However, structural studies have since shown that the molecular ordering within these multilayer films is much reduced compared to single monolayer films.16 A similar approach for the assembly of multilayer films based on layer-by-layer assembly of alkanethiols has been developed where interlayer coupling is achieved through the interaction of Cu2þ with thiol and carboxylic acid functional groups.17,18 The approach is shown schematically in Figure 1. In detail, a (11) Blodgett, K. B. J. Am. Chem. Soc. 1935, 57, 1007–1022. (12) Blodgett, K. B.; Langmuir, I. Phys. Rev. 1937, 51, 964–982. (13) Lee, H.; Kepley, L. J.; Hong, H. G.; Mallouk, T. E. J. Am. Chem. Soc. 1988, 110, 618–620. (14) Cao, G.; Hong, H. G.; Mallouk, T. E. Acc. Chem. Res. 1992, 25, 420–427. (15) Kepley, L. J.; Sacket, D. D.; Bell, C. M.; Mallouk, T. E. Thin Solid Films 1992, 208, 132–136. (16) Bent, S. F.; Schilling, M. L.; Wilson, W. L.; Katz, H. E.; Harris, A. L. Chem. Mater. 1994, 6, 122–126. (17) Evans, S. D.; Ulman, A.; Goppertberarducci, K. E.; Gerenser, L. J. J. Am. Chem. Soc. 1991, 113, 5866–5868. (18) Freeman, T.; Evans, S.; Ulman, A. Thin Solid Films 1994, 244, 784–78.
Published on Web 01/07/2011
DOI: 10.1021/la103733j
1033
Article
Figure 1. Schematic diagram of a multilayer molecular film assembled from sequential adsorption of n-alkanethiolate monolayers and a divalent metal ion, here Cu2þ. The carboxylic acid surface exposed by a single monolayer of HS(CH2)nCOOH (a) is modified to form a Cu carboxylate by exposure to a solution containing Cu2þ ions (b). A second layer can be subsequently assembled on top of the first by formation of a covalent bond between the Cu ion and the thiol of the second HS(CH2)nCOOH SAM (c). In this figure, we show a 1:1 coordination between the number of molecules in the first SAM and the number of Cu ions, as suggested in early studies of multilayer assembly.
monolayer of alkanethiols functionalized with an acid headgroup [HS(CH2)nCOOH] is assembled onto a gold surface through the spontaneous formation of a gold-thiol bond.2,3 The exposed carboxylic acid surface is subsequently modified by exposure to a solution containing Cu2þ ions to form a Cu carboxylate, to which a second layer of HS(CH2)nCOOH molecules can be chemisorbed. By the sequential adsorption of mercaptoalkanoic acid molecules and Cu2þ ions, one can thus form multiple layers of the heterobifunctional alkyl monolayers ligated through carboxyl and thiol interactions with Cu ions. This approach has received further interest recently as a method for scaling of lithographically defined structures.19,20,22 Ellipsometry of mercaptoalkanoic acid/Cu2þ multilayer films demonstrated a linear relationship between the number of layers within a film and the overall average film thickness.17 However, in a study which focused on the interlayer coupling, Brust et al.23 showed that the Cu ions remain in their þ2 oxidation state after complexing with the carboxylic acid surface, reducing to Cu1þ upon chemisorption of the upper HS(CH2)nCOOH adlayer only. Significantly, the divalency of the Cu2þ ions associated with the acid surface effectively prohibits the complete assembly of the next monolayer. The reaction stoichiometry, RCOO-:Cu2þ = 2:1, suggests that the surface coverage of the adayer is at most 50% that of the underlying layer. Atomic force micrographs of bilayer SAMs have shown that rather than assembling into a uniform and homogeneous layer, as shown in Figure 1, the incomplete upper adlayer fractures into discrete islands.24 This results in significant surface roughening, which in turn severely limits the uniformity and integrity of the multilayer film. While it is appears that admolecules of the incomplete upper adlayer can migrate into ordered aggregates rather than form disordered and loosely packed monolayers, the mechanism (19) Hatzor, A.; Weiss, P. S. Science 2001, 291, 1019–1020. (20) Anderson, M. E.; Smith, R. K.; Donhauser, Z. J.; Hatzor, A.; Lewis, P. A.; Tan, L. P.; Tanaka, H.; Horn, M. W.; Weiss, P. S. J. Vac. Sci. Technol. B 2002, 20, 2739–2744. (21) Anderson, M. E.; Tan, L. P.; Tanaka, H.; Mihok, M.; Lee, H.; Horn, M. W.; Weiss, P. S. J. Vac. Sci. Technol. B 2003, 21, 3116–3119. (22) Negishi, R.; Hasegawa, T.; Terabe, K.; Aono, M.; Ebihara, T.; Tanaka, H.; Ogawa, T. Appl. Phys. Lett. 2006, 88, 223111–223113. (23) Brust, M.; Blass, P. M.; Bard, A. J. Langmuir 1997, 13, 5602–560. (24) Daniel, T. A.; Uppili, S.; Mccarty, G.; Allara, D. L. Langmuir 2007, 23, 638–648.
1034 DOI: 10.1021/la103733j
Johnson et al.
leading to migration and island assembly has not yet been established. With this in mind, we have examined the assembly of HS(CH2)nCOOH/Cu2þ multilayers for different numbers of methylene groups, n. In particular, we have examined the assembly of multilayer films assembled from mercaptohexanoic acid (MHA, n = 6), mercaptoundecanoic acid (MUA, n = 11), and mercaptohexadecanoic acid (MHDA, n = 16). We show that the degree of island formation is dependent on the number of methylene groups with 50% surface coverage occurring only for multilayer films formed from the longer molecules (n = 11 and n = 16). While some island growth is observed in multilayer films assembled from shorter molecules (n = 6), we show that the molecules forming the incomplete adlayer are on the whole distributed randomly over the underlying SAM. This relationship between chain length and island formation is a direct result of the reduction of free energy attained by maximizing intermolecular interactions within the molecular layer.
Experimental Section Materials. Ethanol (99%) and isopropanol (99%) from Fisher (Loughborough, UK); acetone (98%), mercaptohexadecanoic [MHDA] (99%), mercaptoundecanoic acid [MUA] (95%), mercaptohexanoic acid [MHA] (90%), copper(II) perchlorate hexahydrate (98%), sulfuric acid, and hydrogen peroxide from Sigma (Gillingham, UK); and n-doped silicon (100) capped with a 300 nm thick thermal oxide from Compart Technology (Peterborough, UK) were used as received. Contact Angle Analysis. The assembly of monolayer and multilayer films for contact angle analysis was performed using gold films prepared using an electron-beam evaporator. The gold films were fabricated by depositing a 20 nm thick titanium adhesion layer followed by a 100 nm thick gold layer under vacuum at a base pressure of 10-7 mbar onto silicon substrates capped with a 300 nm thick, thermally grown oxide. Substrates were cleaned prior to deposition by immersion in piranha solution (70% H2SO4, 30% H2O2) for 10 min before sonicating in deionized water, ethanol, and isopropyl alcohol for 10 min each. The freshly prepared substrates were placed into the monolayer solution immediately following removal from the deposition chamber. Contact angle analysis was performed using a custom-built apparatus. A 50 μL drop of Milli-Q water (18.2 MΩ 3 m) was dispensed onto a gold surface functionalized with a SAM using a flat-tipped micrometer syringe. Images of the advancing contact angle were recorded with the sample in close proximity to the needle apex. Retraction of the needle allowed for measurements of the receding angle while static contact angles were recorded using a free-standing water droplet formed following complete retraction of the needle. Images of the droplet were recorded digitally using a CCD camera, and analysis of the advancing, receding, and static contact angles was performed using the ImageJ 1.42q software (National Institutes of Health, Bethesda, MD) equipped with the Contact Angle plugin. At least three separate measurements were recorded for each sample and the experiments repeated on three different samples. Atomic Force Microscopy. Tapping mode AFM images were taken using a Dimension Series 3100 (Veeco Instruments, Cambridge, UK) scanning probe microscope using etched silicon tips (Veeco Instruments, Cambridge, UK) with a typical resonant frequency of 300 kHz. All imaging was performed using tapping mode AFM to avoid damage or restructuring of the functionalized surface. To minimize further tip-induced artifacts, care was taken to optimize the imaging conditions and, in particular, to optimize the tapping force. AFM measurements of gold films fabricated as described above showed typical rms surface roughness values of 1.2 nm with an average grain size of around 30 nm. This morphology is comparable to that observed following SAM island formation Langmuir 2011, 27(3), 1033–1037
Johnson et al.
Article
Figure 2. Plots showing (a) the measured static angle jstatic and (b) the difference between advancing and receding contact angles, jadv - jrec, for multilayer SAMs as a function of the number of monolayers in the multilayer stack. Data for multilayer SAMs formed from MHA, MUA, and MHDA are displayed. and thus limits accurate AFM characterization of the multilayer SAM surface. Therefore, alternative gold surfaces were prepared using a template stripping process based on that developed by Wagner et al.,25 and rms surface roughness values as low as 0.2 nm were achieved routinely. In brief, a 100 nm thick gold layer was deposited directly onto a SiO2-coated Si(100) substrate (the handle wafer) using an electron beam evaporator. Surface cleaning prior to evaporation was limited to degreasing in acetone and isopropyl alcohol. The gold-coated substrate was then glued onto a second SiO2-coated Si(100) wafer such that the gold layer was sandwiched between the two wafers. Following curing of the adhesive, the wafers were separated carefully such that the gold film was stripped from the handle wafer. Samples were used immediately after preparation to minimize further contamination.
Results and Discussion Contact Angle Analysis. Monolayers for contact angle analysis were formed by the spontaneous assembly of thiols onto the freshly prepared gold surface by immersion of the substrate into a 1 mM solution of MHA, MUA, or MHDA for 1 h. Multilayer films of MHA, MUA, and MHDA were prepared following a process of sequential functionalization: (1) immersion of the freshly prepared Au surface in a 1 mM ethanol solution of the MHA, MUA, or MHDA for 1 h; (2) rinsing of the sample in ethanol, followed by immersion in a 5 mM solution of Cu(ClO4)2 in ethanol for 5 min; (3) washing in ethanol, followed by immersion in a 1 mM MHA, MUA, or MHDA solution in ethanol for 1 h. Additional monolayers were assembled through repetition of steps 2 and 3. (25) P. Wagner, P.; Hegner, M.; Guntherodt, H.-J.; Semenza, G. Langmuir 1995, 11, 3867–3875.
Langmuir 2011, 27(3), 1033–1037
Contact angle analysis of single monolayers were found to be identical for the MHA, MUA, and MHDA solutions with a measured static angle of
