Introduction

Nanotechnology is a burgeoning field which strives to push electronic and mechanical devices to their ultimate lower limit, the atomic scale. The explosive growth of research in this field over the past decade has proliferated the development of a variety of interdisciplinary approaches to molecular engineering and molecular electronics [1], Electrochemistry, as a bridge between chemistry and electronics, is uniquely positioned to advance the development of nanotechnology, particularly molecular electronics [2]. Electroactive molecules show considerable promise as nanotechnological building blocks due to the fact that the electrode potential can be employed as a means to induce transformations in their chemical, electronic, and/or structural properties. The ability to adsorb or chemically bind electroactive molecules to the surface of electrodes, the focus of considerable research effort over the past 25 years, represents an important first step toward molecular electronics [3 - 5].

Molecular redox films have already been incorporated into the design of thin-film batteries and electrochromic displays and are being developed for use in a variety of additional technologies including drug delivery, chemical sensing, and catalysis. Also, diode and field-effect transistor behavior has been demonstrated in systems integrating redox films with different electronic properties [6 - 7]. However, despite the many advances in the design and synthesis of molecular redox films, comparatively little is known about their electronic or molecular structure. The clear need for a deeper understanding of these systems necessitates the use of ultra-high resolution techniques such as scanning probe microscopy (SPM).

Since its inception, SPM has revolutionized the field of surface science and has provided an extraordinary impetus to nanotechnology by enabling the visualization of molecular structures with resolution at or approaching the atomic level [8 - 10]. The ability to operate this class of techniques in the presence of electrolyte solution has further benefited the interfacial sciences by facilitating in-situ studies of potential dependent variations of surface morphology [11]. In this brief account, we present a general overview of our efforts using scanning tunneling microscopy (STM), electrochemical scanning tunneling microscopy (electrochemical STM), and electrochemical force spectroscopy to study structural and chemical properties of molecular redox films.

2 Structural Studies of Molecular Redox Films

A vast literature exists describing numerous strategies for the adsorption and binding of electroactive molecules to electrode surfaces. Of those, the most commonly employed are covalent binding, polymerization, and self-assembly [2 - 5]. In this section, we briefly discuss the utility of these three immobilization techniques for studying redox-active species by STM and electrochemical STM, and present examples from our own studies.

For STM of molecular redox films, theoretical models predict enhanced tunneling through redox-active centers via resonant tunneling modes; specifically, reorganizations of the inner-sphere vibrational modes should significantly contribute to the tunneling current [12 - 14]. It is notable, however, that the observation of anomalous molecular corrugation for several electroactive self-assembling monolayers by Kim et al. [15] and our own observation of weak contrast for a self assembling redox-active osmium complex [16] suggest that the contrast mechanism is the result of several factors.

2.1 Covalent Binding

A convenient method of preparing molecular redox films is by covalently bonding them directly to an electrode surface. These techniques typically yield stable, disordered molecular films at coverages close to a monolayer. Numerous chemical strategies exist to covalently immobilize electroactive molecules to an electrode surface. A common one utilizes 1,3-dicyclohexylcarbodiimide (DCC), a dehydrating agent developed for peptide synthesis which facilitates amide bond formation between a primary amine and a carboxylic acid [17]. In a typical approach, the electrode is first chemically treated to generate a chemically reactive surface (with either the amine or carboxylic acid) and subsequently immersed into a solution containing DCC and a reactive redox molecule with a complementary pendant amine or carboxylic acid. Examples of covalent immobilization on Au(lll) and highly oriented pyrolytic graphite (HOPG) electrodes using DCC are detailed below

Figure 1(A) shows a typical electrochemical STM image for an Au(lll) electrode which was first modified with a monolayer of 3-mercaptopropionic acid (to form a carboxylate-terminated surface) followed by immersion overnight in a solution containing DCC and [Os(2,2'-bipyridine)2 [4-(aminomethyl)pyridine]Cl] (Fig. 1(B)). The high-contrast regions, commonly observed in these systems, most likely correspond to aggregates of the transition-metal complex, though it is not clear why aggregation occurs. A disadvantage of using this immobilization technique for STM studies is that it gives rise to disordered, chemically inhomogeneous surfaces which make image interpretation difficult.

Sample homogeneity can be somewhat assured by using thermally oxidized HOPG as a substrate. Upon heating in the presence of oxygen, ketogenic oxidation causes the formation of pits and defects on the HOPG surface [18], enabling electroactive species to be directly bound to the step edges and defects using DCC. The disadvantage of this approach lies in image validation, since molecular species immobilized onto a step edge are often indistinguishable from the step itself. However, it is possible to find isolated molecular species which are bound directly to defects (or small, incipient pits). Figure 2(A) shows a single Fe(CN)5[4-(aminomethyl)pyridine] molecule (Fig. 2(B))

Fig. 1. (A) Unfiltered electrochemical STM image, taken in 0.1 M KC104 electrolyte solution, of an Au(l 11) electrode modified with [Os(bpy)2[4-(aminomethyl)pyridine]Cl](PF6) using 3-mercapto-propionic acid and DCC. Tip bias = -50 mV, tunneling current =1.0 NA, scan raté 5.1 Hz. (B) Structure and proposed binding mode of the covalently immobilized osmium complex.

Fig. 1. (A) Unfiltered electrochemical STM image, taken in 0.1 M KC104 electrolyte solution, of an Au(l 11) electrode modified with [Os(bpy)2[4-(aminomethyl)pyridine]Cl](PF6) using 3-mercapto-propionic acid and DCC. Tip bias = -50 mV, tunneling current =1.0 NA, scan raté 5.1 Hz. (B) Structure and proposed binding mode of the covalently immobilized osmium complex.

Fig. 2. (A) Fourier filtered ex-situ STM image, taken in constant-height mode, of a thermally oxidized HOPG electrode modified with K2[Fe(CN)5 [4-(aminomethyl)pyridine]] using DCC as a coupling reagent. Tip bias = -50 mV, tunneling current = 1.0 nA, scan rate = 30.5 Hz. (B) Structure and proposed binding mode of the covalently immobilized iron complex.

Fig. 2. (A) Fourier filtered ex-situ STM image, taken in constant-height mode, of a thermally oxidized HOPG electrode modified with K2[Fe(CN)5 [4-(aminomethyl)pyridine]] using DCC as a coupling reagent. Tip bias = -50 mV, tunneling current = 1.0 nA, scan rate = 30.5 Hz. (B) Structure and proposed binding mode of the covalently immobilized iron complex.

which was immobilized onto a thermally oxidized HOPG surface using DCC. The diameter of the imaged molecule (~10 Â) is somewhat larger than the calculated diameter of the transition metal heads group (~7 Â). We ascribe this difference to molecular motions by the tethered molecule. Similar effects have been reported by other investigators [19].

2.2 Polymerization

Electrode modification through electrochemical or chemical deposition of redox-active polymers is another area of active research since these methods typically produce highly stable films which can be prepared in thicknesses varying from nanometers to microns [20]. Polymer films are, however, not well suited for study by STM since the resulting structures are molecularly rough and often poorly conducting, especially at high coverages. Also, as is common for weakly bound molecular adsorbates, polymer films often have high surface mobility (compared with the time required to obtain an STM image) and are susceptible to tip induced interactions [21].

Fig. 3. (A) Unfiltered ex-situ STM image of an HOPG electrode modified with an electropolymerized film of [[(v-tpy)Ru]2(tppz)](PF6)4- Tip bias =100 mV, tunneling current = 500 pA, scan rate = 60.3 Hz. (B) Structure of the symmetric ruthenium polymer.

Fig. 3. (A) Unfiltered ex-situ STM image of an HOPG electrode modified with an electropolymerized film of [[(v-tpy)Ru]2(tppz)](PF6)4- Tip bias =100 mV, tunneling current = 500 pA, scan rate = 60.3 Hz. (B) Structure of the symmetric ruthenium polymer.

This is exemplified in Fig. 3(A), which shows a representative STM image of [((v-tpy)Ru)2(tppz)](PF6)4 (v-tpy is 4'-vinyl-2,2':6',2"-terpyridine, tppz is 2,3,5,6-tetrakis(2-pyridyl)pyrazine; Fig. 3(B)) electropolymerized onto an HOPG substrate according to a previously published procedure (coverage of approximately 75% of one equivalent monolayer). The poor resolution and streaking in the image, probably the result of tip-induced interactions, prevent the visualization of any discernible atomic structure. As a final note, image validation can be difficult for these systems since surface features such as defects and step edges can mimic molecular aggregates and polymer strands under certain conditions in the STM [23].

Fig. 4. Molecularly resolved ex-situ STM image of an Osdipy monolayer adsorbed onto a Pt(lll) electrode: (A) Fourier-filtered and (B) unfiltered. Tip bias = -150 mV, tunneling current = 500 pA, scan rate = 20 Hz. (C) Chemical structure and proposed binding mode of the Osdipy complex.

2.3 Self-assembly

The incorporation a redox-active transition metal head group into a self-assembling molecule provides a ready means for the immobilization of transition-metal complexes onto an electrode surface [13,24]. Unlike other approaches to immobilization which generally yield rough, unordered arrays of molecules, self-assembly techniques can produce well-ordered, atomically smooth molecular arrays.

Fig. 4. Molecularly resolved ex-situ STM image of an Osdipy monolayer adsorbed onto a Pt(lll) electrode: (A) Fourier-filtered and (B) unfiltered. Tip bias = -150 mV, tunneling current = 500 pA, scan rate = 20 Hz. (C) Chemical structure and proposed binding mode of the Osdipy complex.

The self-assembling complex [Os(bpy)2(dipy)Cl](PF6) (bpy is 2,2'-bipyridine, dipy is trimethylene-4,4'-bipyridine; abbreviated as Osdipy; Fig. 4(C)) chemisorbs onto platinum and gold surfaces through a dangling pyridine ring with a free energy of adsorption of the order of-50 kJ/mol to form well-ordered, redox-active monolayers

Fig. 5. electrochemical STM images of a Pt(l 11) surface during the deposition of Osdipy from a 0.1 M KCIO4 electrolyte solution with corresponding cyclic voltammograms taken immediately upon withdrawing the tip from the electrode surface. The integrated charge corresponds to (A) 0%, (B) 25%, (c) 49%, (D) 66%, and (E) 100% of a Ml monolayer. Tip bias = 50 mV, tunneling current = 2.5 nA, scan rate = 8.3 Hz.

Fig. 5. electrochemical STM images of a Pt(l 11) surface during the deposition of Osdipy from a 0.1 M KCIO4 electrolyte solution with corresponding cyclic voltammograms taken immediately upon withdrawing the tip from the electrode surface. The integrated charge corresponds to (A) 0%, (B) 25%, (c) 49%, (D) 66%, and (E) 100% of a Ml monolayer. Tip bias = 50 mV, tunneling current = 2.5 nA, scan rate = 8.3 Hz.

which are stable on the electrode surface in a wide range of solvents and electrolytes [25 - 27]. A representative unfiltered ex-situ STM image, taken in air, of an adsorbed

Osdipy monolayer deposited onto a Pt(l 11) substrate is shown in Fig. 4(B). The image clearly shows molecular-scale features in a rectangular close packed array with unit cell dimensions of 9.3Á by 12.4Á. These dimensions were found to be in excellent agreement with model calculations which assume that the head groups preferentially line up along their dipole [16].

The reversible redox activity enables the straightforward determination of adsórbate coverage at any point during the deposition process by integration of the charge under the voltammetric wave. In an effort to elucidate the mechanism of Osdipy monolayer formation, we utilized electrochemical STM to monitor simultaneously the coverage and structure of the depositing Osdipy layer. A typical series of electrochemical STM images with corresponding cyclic voltammograms (which exhibit the characteristic reversible metal-based oxidation of the transition-metal head group of Osdipy) are shown in Fig. 5.

Combining our electrochemical STM data with the results of previous voltammetric studies, we were able to determine that the monolayer forms through the random decoration of the electrode surface by the Osdipy adsórbate. The electrostatic repulsions between neighboring adsórbate molecules (each molecule carries a net positive charge), combined with the high mobility across the sample surface, causes the submonolayer to spread uniformly across the sample surface. At higher coverages, regions of the adsórbate layer lose part of their solvation to accommodate the increasing coverage (and perhaps counter ions) and become more tightly packed. This leads to the possible formation of domains within the submonolayer structure which appear as recessed defects. The high-density regions continue to grow, possibly by coalescing, until the full monolayer is formed. These observations are supported by electrochemical quartz crystal microbalance studies which show a large decrease in Osdipy solvation at high coverages [28]. Due to the rapid reordering of the adsórbate on the electrode surface, it was not possible to obtain molecularly resolved images of the full monolayer in electrolyte solution.

0 0

Post a comment