Self-Assembled Monolayer and Multilayer Films of the Nanocluster

Jul 1, 2008 - Ramon Colorado, Jr., Christopher A. Crouse, Christopher N. Zeigler, and. Andrew R. Barron*. Richard E. Smalley Institute for Nanoscale ...
1 downloads 0 Views 1MB Size
Download PDF
8912

Langmuir 2008, 24, 8912-8917

Self-Assembled Monolayer and Multilayer Films of the Nanocluster [HxPMo12O40⊂H4Mo72Fe30(O2CMe)15O254(H2O)68] on Gold Ramon Colorado, Jr., Christopher A. Crouse, Christopher N. Zeigler, and Andrew R. Barron* Richard E. Smalley Institute for Nanoscale Science and Technology and Department of Chemistry, Rice UniVersity, Houston, Texas 77005 ReceiVed February 4, 2008. ReVised Manuscript ReceiVed May 16, 2008 Films of the molybdenum-iron nanocluster [HxPMo12O40⊂H4Mo72Fe30(O2CMe)15O254(H2O)68] (FeMoC) were generated on gold via the self-assembly technique using two divergent routes. The first route entails the self-assembly of unfunctionalized FeMoC onto a preprepared carboxyl-terminated SAM on gold. The second route involves the preparation of thiol-terminated functionalized FeMoC clusters, which are then allowed to self-assemble onto bare gold surfaces. Monolayer films of FeMoC clusters are attained via both routes, with the second route requiring shorter immersion times (2 days) than the first route (6 days). Multilayer films of FeMoC are formed via the second route for immersion times longer than 2 days. Characterization of these films using optical ellipsometry, X-ray photoelectron spectroscopy, and atomic force microscopy confirm the self-assembly of the clusters on the surfaces.

Introduction Early transition metal polyoxoanions, generally known as polyoxometalates (POMs), comprise a broad class of inorganic materials that exhibit properties useful for catalytic, electrochemical, and magnetic applications as a result of their welldefined uniform molecular structures.1 POMs are formed by the self-assembly of discrete metal oxide building blocks into nanoscale clusters that possess highly symmetrical frameworks and topologies.2 The composition and size of the POMs are easily tuned by varying the composition and concentrations of the dissolved metal oxide precursors (XaMbOcn-; M ) Mo, W, V; X ) P, As, Si, Ge, B, Co, Fe) employed in the solution-based synthesis.3 The resultant clusters consist of multiple transition metal centers separated from one another by distances determined by the size of the building blocks.4,5 By incorporating paramagnetic d transition metals into the structure, clusters possessing magnetic character can be attained.6 These features make POMs a particularly interesting set of molecular nanoparticles. Whereas progress in POM synthesis has flourished, strides in processing POMs with the ultimate aim of exploiting their unique properties in functional materials have been limited.7 The construction of POM-based devices will require precise isolation, deposition, and positioning of the clusters within the architecture. Although POMs exist as discrete clusters in solution, attempts to isolate them as solids for further manipulation are hampered by the tendency of the clusters to aggregate strongly and crystallize upon drying.8,9 Hence, these goals are advanced by the development of both physical and chemical methods for * Author to whom correspondence should be addressed. E-mail: [email protected]. (1) Gouzerh, P.; Che, M. L’Actualite´ Chim. 2006, 298, 9. (2) Pope, M. T.; Muller, A. Polyoxometalate Chemistry: From Topology Via Self-Assembly to Applications; Kluwer: Dordrecht, The Netherlands, 2002. (3) Pope, M. T. Heteropoly and Isopoly Oxometalates; Springer-Verlag: Heidelberg, 1983. (4) Borras-Almenar, J. J.; Coronado, E.; Muller, A.; Pope, M. T. Polyoxometalate Molecular Science; Kluwer: Dordrecht, The Netherlands, 2001. (5) Muller, A.; Roy, S. Russ. Chem. ReV. 2002, 71, 981. (6) Muller, A.; Kogerler, P.; Dress, A. Coord. Chem. ReV. 2001, 222, 193. (7) Liu, S.; Volkmer, D.; Kurth, D. G. Pure Appl. Chem. 2004, 76, 1847. (8) Muller, A.; Serain, C. Acc. Chem. Res. 2000, 33, 2. (9) Muller, A.; Krickemeyer, E.; Das, S. K.; Kogerler, P.; Sarkar, S.; Bogge, H.; Schmidtmann, M.; Sarkar, S. Angew. Chem., Int. Ed. 2000, 39, 1612.

modifying the surface of the clusters to improve the tractability of POMs and facilitate their controlled deposition onto surfaces. Our initial interest in POMs was triggered by a desire to utilize the clusters in the atypical application of serving as precatalysts with discrete size for the growth of single-walled carbon nanotubes (SWNTs) from surfaces.10 In these efforts, we concentrated on the molybdenum-iron POM, [HxPMo12O40⊂H4Mo72Fe30(O2CMe)15O254(H2O)68] (FeMoC), because it possessed a composition and size that was reported to be ideal for targeting the growth of small-diameter SWNTs.11 FeMoC is a spherical host-guest cluster (d ) 2.1 nm) consisting of the widely studied Keplerate POM Mo72Fe30 as the host and the Keggin ion [HxPMo12O40]3- as the guest to bear a unique magnetic nanoparticle possessing 30 distinct paramagnetic centers.12,13 Our results demonstrated that the nonspecific spin-coating method used to deposit the FeMoC clusters onto the surfaces promoted intercluster aggregation that needed to be controlled to optimize the growth of SWNTs.10 We were therefore prompted to explore methods for the controlled placement of FeMoC on surfaces. Several groups have contributed to advancing techniques for the soft encapsulation of Keplerate POMs within tailored surfactants or polyelectrolytes for tuning their solubilities and enabling their assembly onto surfaces.14–18 However, the assemblies constructed via these approaches have been restricted thus far to Langmuir-Blodgett (LB) monolayers and layer-by(10) Anderson, R. E.; Colorado, R.; Crouse, C.; Ogrin, D.; Maruyama, B.; Pender, M. J.; Edwards, C. L.; Whitsitt, E.; Moore, V. C.; Koveal, D.; Lupu, C.; Stewart, M. P.; Smalley, R. E.; Tour, J. M.; Barron, A. R. Dalton Trans. 2006, 3097. (11) An, L.; Owens, J. M.; McNeil, L. E.; Liu, J. J. Am. Chem. Soc. 2002, 124, 13688. (12) Muller, A.; Das, S. K.; Kogerler, P.; Bogge, H.; Schmidtmann, M.; Trautwein, A. X.; Schunemann, V.; Krickemeyer, E.; Preetz, W. Angew. Chem., Int. Ed. 2000, 39, 3413. (13) Muller, A.; Das, S. K.; Bogge, H.; Schmidtmann, M.; Botar, A.; Patrut, A. Chem. Commun. 2001, 657. (14) Volkmer, D.; Du Chesne, A.; Kurth, D. G.; Schnablegger, H.; Lehmann, P.; Koop, M. J.; Muller, A. J. Am. Chem. Soc. 2000, 122, 1995. (15) Kurth, D. G.; Lehmann, P.; Volkmer, D.; Muller, A.; Schwahn, D. Dalton Trans. 2000, 3989. (16) Kurth, D. G.; Lehmann, P.; Volkmer, D.; Colfen, H.; Koop, M. J.; Muller, A.; Du Chesne, A. Chem.sEur. J. 2000, 6, 385. (17) Liu, T. J. Clust. Sci. 2003, 14, 215. (18) Fan, D.; Jia, X.; Tang, P.; Hao, J.; Liu, T. Angew. Chem., Int. Ed. 2007, 46, 3342.

10.1021/la800392a CCC: $40.75  2008 American Chemical Society Published on Web 07/01/2008

Monolayer and Multilayer Films of Nanocluster on Au

Langmuir, Vol. 24, No. 16, 2008 8913

Scheme 1. Schematic Depiction of the Formation of FeMoC-MHDA-SAMs on Gold

layer (LBL) multilayers, which offer limited control over the ordering of clusters on surfaces.19,20 The successful formation of self-assembled monolayers (SAMs) of [Mn12O12(OAc)16(H2O)4] · (H2O)4 · (AcOH)2 on gold has been formed by exchanging the acetate ligands with bifunctional 16-thioacetylhexadecanoic acid linkers possessing thioacetyl groups that bind to gold, thereby allowing these functionalized clusters to selfassemble onto the gold surfaces.21 Stoll and co-workers employed a similar technique to effect the controlled placement of [Mn12O12(OAc)16(H2O)4] · (H2O)4 · (AcOH)2 by first preparing SAMs on gold of the bifunctional adsorbate 11-mercaptoundecanoic acid and then dipping these monolayers into solutions of the dodecamanganese clusters.22 Inspired by these two approaches, we have investigated the SAM formation on gold of FeMoC in a desire to extend the methodologies available for the handling of these functional clusters.

Experimental Section Materials and Characterization. FeMoC was synthesized using a previously reported optimized procedure and purified via Soxhlet extraction with EtOH, giving the solvate H3PMo12O40⊂H4Mo72Fe30(O2CMe)15O254(H2O)68(EtOH)30, which is referred to simply as unfunctionalized FeMoC throughout this text.10 EtOH (Fisher Scientific) and 16-mercaptohexadecanoic acid (Sigma-Aldrich) were used as received. Gold shot (99.99%) was obtained from Americana Precious Metals, and chromium rods (99.9%) were acquired from R. D. Mathis Co. Polished silicon (100) wafers were purchased from North East Silicon Technologies, Inc. Gold substrates were prepared by thermally evaporating (at 8 × 10-6 Torr) an adhesion layer of 10 nm chromium followed by 200 nm gold at a rate of 1 Å · s-1 onto

Figure 1. Ellipsometric thickness of films generated when MHDASAMs are immersed in an MHDA solution (0) or a FeMoC solution (9) for 6 days. Linear regression through the MDHA-SAM data shows an average thickness of 2.6 nm.

the polished surfaces of the silicon wafers. For the AFM imaging substrates, 200 nm of gold was evaporated at a rate of 1 Å · s-1 directly onto the surface of freshly cleaved of mica. These goldon-mica surfaces were then flame annealed to produce flat gold. Freshly prepared gold-coated substrates were cut into slides (ca. 1 cm × 3 cm) using a diamond-tipped stylus (for silicon) or razor blade (for mica). The slides were washed with EtOH, blown dry with ultrapure N2, and immersed in the respective solutions for SAM formation over the designated times. Monolayer and multilayer thicknesses were measured using a Rudolf Research Auto EL III ellipsometer equipped with a He-Ne laser operating at 632.8 nm and an angle of incidence of 70°. The optical constants of the bare gold substrates were measured immediately after the evaporation of gold. To calculate the thicknesses of the monolayers and multilayers, a refractive index of 1.45 was assumed for all films. The data were collected and averaged from measurements on at least three separate slides using three spots per slide for each type of film. The measured ellipsometric thicknesses were always within (1 Å of the reported values. XPS spectra were collected using a PHI 5700 X-ray photoelectron spectrometer equipped with a monochromatic Al KR X-ray source (hν ) 1486.7 eV) incident at 90° relative to the axis of a hemispherical energy analyzer. The spectrometer was operated at high resolution using

Figure 2. High-resolution XPS spectra of the (a) Au 4f and (b) C 1s photoelectron regions for the MHDA-SAM (red line) and the FeMoCMHDA-SAM (blue line) after 6 days.

8914 Langmuir, Vol. 24, No. 16, 2008

Colorado et al.

Scheme 2. Schematic Depiction of the Formation of FeMoC(MHDA)-SAMs on Gold

a pass energy of 23.5 eV, a photoelectron takeoff angle of 45° from the surface, and an analyzer spot diameter of 1.1 mm. The base pressure in the chamber during measurements was 2 × 10-9 Torr, and the spectra were collected at room temperature. The binding energies were referenced by setting the Au 4f7/2 peak to 84.0 eV. The peak intensities were quantified using standard curve fitting with Shirley background subtraction and Gaussian-Lorentzian profiles. AFM images were obtained using a Digital Instruments NanoScope IIIa scanning probe microscope in tapping mode. An RTESP-type Nanoprobe SPM tip with a drive frequency of 300 kHz was used. Images were taken at a scan frequency of 2 Hz and a scan resolution of 256 samples per line. Synthesis of Functionalized FeMoC(MHDA). A 0.75 mM solution of 16-mercaptohexadecanoic acid (10.8 mg, 37.5 mmol) in EtOH (50 mL) was prepared and transferred into a three-necked round-bottomed flask fitted with a condenser, argon inlet, and addition funnel. The addition funnel was charged with a 25 µM solution of FeMoC (22.5 mg, 1.25 mmol) in EtOH (50 mL). The flask was heated to reflux, and the FeMoC solution was added dropwise over a period of 15 min. Afterward, reflux was maintained for an additional 30 min prior to allowing the contents to cool to room temperature. This 25 µM solution of functionalized FeMoC(MHDA) was used as prepared for monolayer and multilayer preparation. Preparation of SAMs. A 1 mM solution of 16-mercaptohexadecanoic acid (14.4 mg, 50.0 mmol) in EtOH (50 mL) was prepared and used to generate MHDA-SAMs. To generate FeMoC-MHDASAMs, preformed MHDA-SAMs were immersed in a 25 µM solution of FeMoC (22.5 mg, 1.25 mmol) in EtOH (50 mL). To generate FeMoC(MHDA) monolayers and multilayers, bare gold-coated

Figure 3. Ellipsometric thickness of films generated on gold slides when immersed in an FeMoC(MHDA) solution (9) compared with that of films generated on MHDA-SAMs immersed in an FeMoC solution (0) for 6 days.

substrates were immersed in the 25 µM solution of FeMoC(MHDA) prepared as described above. All solutions were prepared in glass weighing bottles that were cleaned by soaking in piranha solution (3:1 H2SO4/H2O2) for 10 min. Caution! Piranha solution reacts Violently with organic materials and should be handled carefully! The bottles were then rinsed extensively with deionized water and absolute ethanol. After reaching the desired immersion time, all films were removed from their respective solutions and were rinsed thoroughly with EtOH to remove any physisorbed material. Finally, the films were blown dry with ultrapure nitrogen prior to immediate characterization. We have determined that the FeMoC solutions used for functionalization and SAM formation should be prepared from freshly purified FeMoC to achieve optimal activity. The observed lower reactivity of old FeMoC solutions likely arises from the known coupling of Mo72Fe30 clusters in solution with time.23,24

Results and Discussion Our first pathway (Scheme 1) requires the formation of a SAM of MHDA on gold, which is accomplished by immersing a gold-

Figure 4. High-resolution XPS spectra of the (a) Au 4f and (b) C 1s photoelectron regions for the 7.0 nm FeMoC(MHDA)-SAM (red line) and the 4.5 nm FeMoC-MHDA-SAM (blue line).

Monolayer and Multilayer Films of Nanocluster on Au

Figure 5. High-resolution XPS spectra of the (a) Au 4f, (b) C 1s, and (c) Mo 3d photoelectron regions for the 7.0 nm FeMoC(MHDA)-SAM (red line) and the 20 nm FeMoC(MHDA) multilayer (blue line).

coated substrate into a dilute solution of MHDA in EtOH and allowing it to incubate for 24 h at room temperature. The resultant SAM comprises MHDA adsorbates close packed into a hexagonal array with their thiol groups bound to the gold atoms, their hydrocarbon chains trans-extended upward, and their carboxylic acid groups composing the outermost surface of the film. An average film thickness of 2.6 nm was determined by optical ellipsometry, consistent with the formation of a monolayer of close-packed trans-extended MHDA adsorbates bound to gold. After being thoroughly rinsed to remove any unbound adsorbates, the MHDA-SAMs were dipped into a dilute solution of FeMoC in EtOH. The film thickness was monitored as a function of immersion time. Figure 1 shows that over an immersion time of 6 days the film thickness increases gradually to a maximum value of 4.5 nm. This progressive increase in thickness can be explained by the adsorption of FeMoC on the surface. The initial adsorption is presumably random; however, a tighter arrangement as more clusters adsorb leads to a final close-packed array consisting of a monolayer of FeMoC assembled on top of the MHDA film (denoted as FeMoC-MHDA-SAM). The final thickness of the FeMoC-MHDA-SAM corresponds well with

Langmuir, Vol. 24, No. 16, 2008 8915

that expected from adding the diameter of FeMoC to the length of MHDA. No evidence of multilayer formation (i.e., increased thickness) was observed for an immersion time longer than 6 days. The thickness of the MHDA-SAM remains constant when immersed in the MHDA solution for 6 days. The original MHDA-SAM and the FeMoC-MHDA-SAM were evaluated using X-ray photoelectron spectroscopy (XPS). The Au 4f peaks of the FeMoC-MHDA-SAM (Figure 2a) exhibit lower intensities than those corresponding to the MHDA-SAM, indicating that additional adsorption has occurred on top of the original film, causing increased attenuation of the gold signal. Additionally, the peaks in the C 1s spectra, which arise predominately from the carbon backbones of the MHDA adsorbate within the SAM, show the same attenuation (Figure 2b). As expected, the Mo 3d and Fe 2p spectra confirm the presence of Mo and Fe in the FeMoC-MHDA-SAM that are absent in the MHDA-SAM. Taken together, these data support the adsorption of FeMoC atop the MDHA-SAM. Our second approach (Scheme 2) begins with the functionalization of FeMoC with MHDA. We had previously developed methods for substituting the ethanolic solvent ligands on FeMoC with an assortment of monofunctional ligands25 and a bifunctional ligand, pyrazine.10 These approaches were adapted to the present ligand exchange, which is accomplished by refluxing FeMoC in a solution of MHDA in EtOH to yield FeMoC(MHDA) (d ) 7.0 nm). Next, a gold-coated substrate is immersed in an ethanolic solution of FeMoC(MHDA) and is left to sit at room temperature. Measurements of ellipsometric thickness were taken at intervals of increasing immersion time to determine if any film formation occurred on the gold surface. Figure 3 shows that, as with the FeMoC-MHDA-SAM, the measured thickness increases steadily with longer immersion time; however, there are a number of distinct differences indicated by this graph. First, a thickness of 7.0 nm, expected for a complete monolayer of FeMoC(MHDA), is reached after only 2 days whereas achieving a maximum thickness for the FeMoC-MHDA-SAM requires 6 days. Film formation occurs faster for the functionalized cluster on gold to generate an FeMoC(MHDA)-SAM than for FeMoC on the MHDA-SAM surface. This is expected because the former reaction (Scheme 2) involves the adsorption of thiol-functionalized FeMoC(MHDA) on a gold surface whereas the latter (Scheme 1) requires a ligand-exchange reaction (H2O and/or EtOH for carboxylate) to occur. The second difference between the two methods is that after 2.5 more days of immersion the film thickness of the FeMoC(MHDA) doubles wherease no growth beyond a monolayer is observed for the FeMoC(MHDA)-SAM. Furthermore, continued growth is seen such that after an additional 1.5 days (6 days total) a thickness of 20 nm is observed. Thus, the functionalized cluster continues to assemble on top of the (MHDA)-FeMoC-SAM to generate (MHDA)-FeMoC multilayers, which exhibit the larger thicknesses. A comparison of the 7.0-nm-thick FeMoC(MHDA)-SAM to the 4.5-nm-thick FeMoC-MHDA-SAM can be seen in the Au 4f, C 1s, Mo 3d, and Fe 2p regions of the XPS spectra of each film. The Au 4f signal is attenuated in the FeMoC(MHDA)SAM, reflecting the greater film thickness (Figure 4a). The C 1s signal is enhanced for the FeMoC(MHDA)-SAM compared to that of the FeMoC-MHDA-SAM because the former cluster presents MHDA ligands at the outermost surface (Scheme 2), in contrast with the FeMoC-MHDA-SAM, where the outermost unfunctionalized FeMoC clusters (Scheme 1) shield the C 1s photoelectrons emanating from the underlying MHDA-SAM (Figure 4b). The Mo 3d signals are consistent with the presence

8916 Langmuir, Vol. 24, No. 16, 2008

Colorado et al.

Scheme 3. Schematic depiction of three possible routes for formation of FeMoC(MHDA) multilayers.

of FeMoC clusters whereas the Fe 2p signals in both samples are weak and suffer from poor signal-to-noise as a result of the

Figure 6. AFM images (1.0 µm × 1.0 µm) of (a) a bare gold on mica substrate, and (b) a 20 nm FeMoC(MHDA) multilayer formed on gold.

low concentration of Fe in the clusters. These results are consistent with the adsorption of FeMoC(MHDA) on gold as intact functionalized clusters. Although we have rationalized the observed differences in XPS signal for the two films as arising from differences in cluster size influencing the film thicknesses, the size differences could also conceivably influence the cluster coverage, which would also affect the XPS signals. However, the absolute Mo 3d signal intensities for both films are similar (Supporting Information), which suggests that the particle coverage is similarly maximized in both films. We then contrasted the XPS spectra of the 20 nm thick FeMoC(MHDA) multilayer with those of the 7.0 nm thick FeMoC(MHDA)-SAM. The Au 4f signal is barely detectable in the FeMoC(MHDA) multilayer indicating the significantly higher thickness of the multilayer over the monolayer (Figure 5a). The C 1s signals are equivalent since they both arise predominately from the MHDA ligands at the outermost surface (Figure 5b). The Mo 3d signal is higher in the FeMoC(MHDA) multilayer (Figure 5c) likely arising from the contribution of clusters from the layer lying under the outermost surface. The Fe 2p signals are weak and indistinguishable from one another due to poor signal-to-noise. Altogether, the data demonstrate the additional adsorption of FeMoC(MHDA) clusters onto the FeMoC(MHDA)SAM to form multilayers. Given the range of thicknesses observed for the FeMoC(MHDA) multilayers over the 6 day immersion period, we considered the pathways available for multilayer formation. Multilayers may form via three potential routes (Scheme 3): (a) intercalation of the MHDA chains; (b) formation of disulfide linkages between thiols; or (c) ligand exchange. We attempted to resolve the structure of the cluster assemblies on the gold surfaces using atomic force microscopy (AFM). Unfortunately, efforts to obtain clear images for either the FeMoCMHDA-SAM or the FeMoC(MHDA)-SAM were unsuccessful. In contrast, the FeMoC(MHDA) multilayer proved itself to be more easily resolved. Figure 6 shows AFM images for a 1 µm × 1 µm area obtained for a sample of bare flat gold on mica, and after being immersed in the FeMoC(MHDA) solution for 6 days. The surface of the bare gold is made up of a combination of small (20 nm wide) and large (100 nm wide) gold terraces, whose surfaces appear flat and pristine. The analogous area of the FeMoC(MHDA) multilayer, however, exhibits a surface

Monolayer and Multilayer Films of Nanocluster on Au

covered entirely by small spherical features of fairly uniform size. A particle analysis of this image (Supporting Information) reveals that these features exhibit an average height of 3.4 nm, which correlates well with the expected radius of the functionalized FeMoC(MHDA) cluster that would be exposed at the multilayer surface. Furthermore, the maximum observed height of 6.3 nm is near the value expected for a fully exposed functionalized FeMoC(MHDA) cluster sitting directly atop an underlying cluster. While the arrangement of the clusters appears random at first glance, closer inspection reveals that within the areas encompassed by the underlying flat gold terraces, the clusters appear close packed albeit not entirely symmetrical. These images demonstrate that the multilayers form by clusters first adsorbing to form full monolayers on the flat gold terraces, and then build upward by forming additional monolayers on top of these monolayers, while still following the original topology of the gold substrate.

Conclusions Assemblies of discrete FeMoC clusters on gold were generated via two distinct methods using 16-mercaptohexadecanoic acid (MHDA) as a linker. Method one (Scheme 1) produces a surface composed of unfunctionalized FeMoC clusters, whereas method two (Scheme 2) yields a surface consisting of MHDAfunctionalized FeMoC clusters. Method one requires 6 days for the formation of a complete monolayer, but only 2 days are necessary with method 2. This disparity in rates reflects fundamental differences in the mechanism of cluster adsorption between the two methods. In method one, the bare FeMoC clusters must first impinge on the carboxyl-terminated surface of the MHDA-SAM followed by ligand exchange. In method two, the MDHA-functionalized FeMoC clusters present an outermost surface consisting of thiol groups that readily bind with the gold atoms to link the clusters to the surface. Method two can also be used to generate multilayers of functionalized FeMoC by extending the immersion times longer than 2 days. Under the reaction conditions used, this multilayer

Langmuir, Vol. 24, No. 16, 2008 8917

formation can plausibly occur by any combination of three mechanisms: (1) MDHA chain intercalation between functionalized clusters, (2) disulfide formation between thiol groups on different clusters, or (3) ligand exchange by a thiol group on a functionalized cluster in solution for an MHDA ligand on a functionalized cluster in the film. Our findings have significant applications for the preparation of POM-modified surfaces, which have application in the fields of catalysis, electrochemistry, and electromagnetism. The ability to build up multilayers in a controlled fashion is especially desirable where issues of cluster concentration are key to enhancing the performance of POMbased devices. Acknowledgment. Financial support for this work was provided by the National Science Foundation and the Robert A. Welch Foundation. The National Research Council and Ford Foundation are acknowledged for a postdoctoral fellowships (R.C.). We kindly thank Professor T. Randall Lee (University of Houston) for the use of his metal atom evaporator and ellipsometer and Dr. H. Justin Moore for assistance in preparing the gold-coated substrates. Supporting Information Available: Mo 3d and Fe 2p XPS spectra of the FeMoC-MHDA-SAM. Comparison of Fe 2p XPS spectra of the FeMoC(MHDA)-SAM with the FeMoC-MHDA-SAM. Comparison of Fe 2p XPS spectra of the FeMoC(MHDA)-SAM with the FeMoC(MHDA) multilayer. This material is available free of charge via the Internet at http://pubs.acs.org. LA800392A (19) Liu, S.; Kurth, D. G.; Bredenkotter, B.; Volkmer, D. J. Am. Chem. Soc. 2002, 124, 12279. (20) Liu, S.; Mohwald, H.; Volkmer, D.; Kurth, D. G. Langmuir 2006, 22, 1949. (21) Cornia, A.; Fabretti, A. C.; Pacchioni, M.; Zobbi, L.; Bonacchi, D.; Caneschi, A.; Gatteschi, D.; Biagi, R.; Del Pennino, U.; De Renzi, V.; Gurevich, L.; Van der Zant, H. S. J. Angew. Chem., Int. Ed. 2003, 42, 1645. (22) Steckel, J. S.; Persky, N. S.; Martinez, C. R.; Barnes, C. L.; Fry, E. A.; Kulkarni, J.; Burgess, J. D.; Pacheco, R. B.; Stoll, S. L. Nano Lett. 2004, 4, 399. (23) Liu, T. J. Am. Chem. Soc. 2003, 125, 312. (24) Liu, G.; Liu, T. J. Am. Chem. Soc. 2005, 127, 6942. (25) Ogrin, D.; Barron, A. R. J. Clust. Sci. 2007, 18, 113.