«Kinetic Investigations of Thiolate Protected Gold Nanoparticles: Protein Interactions, Electron Transfer, and Precursor Formation By Brian N. Turner ...»
These peaks are associated, more specifically, with double layer charging. The capacitance on a single nanoparticle is, therefore, equal to the series capacitance of the stern layer and the diffuse layer. In the case of more polydisperse nanoparticles, the charging behavior still occurs, but the observation of discrete peaks is obscured, as a greater number of redox events are occurring and overlapping in the voltammogram.
Therefore, while the behavior is only observed for highly monodisperse nanoparticles, polydisperse nanoparticles should be useful as nanoscale capacitors as well. In order to use this property in a molecular electronics application, it is prudent to assess electron transfer dynamics associated with this charging.
In the solid phase, a number of studies have examined electron transfer with nanoparticles. The Murray group studied the electron self-exchange in deprotonated Au140(MUA)20(C6)33/Zn nanoparticle multilayer films on mercaptoundecanoic acid (MUA) self-assembled monolayers (SAMS) on Au disk electrode. Using Cottrell experiments, electron diffusion coefficients in the range of 10 -8 cm/s, electron hopping k values in the range of 106 s-1, and electron self-exchange k values of 108 M-1 s-1 were determined.139 Later, the same research group published a study proving that arenethiolate protecting ligands lead to films with faster electron transfer rates and higher conductivities than alkanethiolate ligands. 140 Furthermore, lengthening an alkyl spacer before the aryl group decreased the electron transfer rate. Other groups have investigated gold nanoparticle films with cytochrome C, 141 at a mercury junction,142 on 3mercatopropyltrimethoxysilane functionalized glass slides, 143 and on SAMs of 2Dimethylamino)ethanethiol,144 to highlight some of the many important contributions.
The difference between the molecules discussed emphasizes the importance of ligand structure to electron transfer. The importance of the current study, in the scope of the discussed investigations, is the characterization of nanoparticle electron transfer kinetics in the solution phase in order to eliminate effects associated with the solid phase such as packing geometry, interparticle interactions, film thickness, and solid-solid interfaces.
Molecular Wire Molecules Molecular wires are a class of small molecule that are characterized by delocalized HOMO orbitals, energetically contained within the Fermi level of an attached metal electrode. In the context of a metal-molecule-metal junction, this allows electrons to flow through the molecule in a manner known as “resonant tunneling.” Resonant tunneling is similar in speed to conductance in metals, and certainly faster than throughspace tunneling. Resonant tunneling exists in molecules that are rigid, conjugated, and fully coplanar with themselves. These structural properties lead to a high degree of πconjugation and, therefore, molecular orbitals that extend across the length of the molecule. The concept of a molecular wire is visually is illustrated in Figure 31.145 Figure 31: The molecular wire ligand (top, (S-[4-(2Phenylethynyl)phenyl]ethynylphenyl]thiol, PEPEPSH)) features closely spaced molecular orbitals (bottom) that are energetically within the Fermi level of a metal. 145 Copyright 2003, American Association for the Advancement of Science, reused with permission.
When thought of as a series of within molecule electron transfers, it is also important to note that these molecules tend to have low reorganization energies from a neutral to oxidized or reduced state, corresponding with small changes in bond angles and lengths.146 Theories on the electronic behavior of molecular wires have been evaluated in a number of studies, usually involving the formation of a metal-molecule-metal junction or junctions. Of note, Ratner and co-workers self-assembled mono- and dithioacetates on the tip of a 10 µm gold wire, and approached it within a small distance of a second gold wire, end to end.147 The approach of the two wires was controlled by a deflection wire within a magnetic field, bent by a Lorentz force provided by a small DC current. The approach was monitored by measuring current at 0.5 V. Having constructed such an apparatus allowed them to measure current-voltage profiles in a circuit that were dependent on how electrons negotiated the gap between the two wires, which was filled by the SAM molecules. Their experimental setup represents a multi-molecule version of the single molecule junction described previously. Current voltage profiles obtained using SAMs of monothioalkyl (2), dithioalkyl (1), conjugated monothiolalkynylaryl (4), and conjugated dithioalkynylaryl (3) molecules are displayed in Figure 32, along with structures of evaluated molecules (1-4).
Figure 32: Current-voltage profiles of molecular wires sandwiched between gold wires.
Two covalent sulfur-gold contacts shows higher current than single sulfur gold contacts for either case. The conjugated system (3 and 4, at right) shows greater current (µA vs.
nA) than the alkane system (1 and 2, at left) (molecular lengths are similar). Copyright 2003, John Wiley and Sons, reused with permission.
Summarily, this study found that having a thiol contact on the other end of the molecule (in addition to the one attached to the gold surface), as in molecules 1 and 3, improved conductivity by at least an order of magnitude over the monothio- molecules 2 and 4.
Additionally, the alkynylaryl “molecular wire” molecules, 3 and 4, had conductivities two orders of magnitude higher than the alkyl molecules. This important study highlights the concepts of a molecular wire attached to a classical conductor by use of sulfur “alligator clips” as a viable strategy for nanomolecular electronic device construction.
Tour and coworkers successfully integrated these types of molecules into a functional device.148 Using gold electrodes sandwiching a molecular wire, they were able to create a binary memory device with write, read, and erase capabilities, as shown in Figure 33.
Figure 33: Read-write computing system with gold electrodes sandwiching molecular wires.148 Writing is accomplished by reductive current, and erasing is accomplished by oxidative current. Copyright 2001, John Wiley and Sons, reused with permission.
Memory, in this case, is in the form of a stored conductance instead of a stored voltage.
The increased conductance state that is stored is the result of electron injection into the molecule. The device was designed as a way for their laboratory to test potential molecular wires in a pass/fail evaluation. Interestingly, the device was found to be robust, as no drop off in read signals occurred after an appreciable amount of time, and some devices can last continuously through a year of operation, which correlates with approximately one billion cycles. Aside from its use as a testing, the read/write system was a great proof of concept that small molecule molecular wires can be integrated into a useable device. With this motivation, the research proceeded to adapt molecules of this class to gold nanoparticles with the hope that both components could be integrated into a functional device for a materials or biological application.
Scanning Electrochemical Microscope Our group developed a method to evaluate electron transfer kinetics in the solution phase by using the scanning electrochemical microscope (SECM). The SECM was first described in 1989 by Bard and Kwak.149 The instrument consists of an ultramicroelectrode in a glass sheath that comes to a tip, controlled by a piezoelectric motor (or some other motor capable of micron spatial resolution). In feedback mode, it is operated in a solution that contains an electroactive substance, and it registers a normal chronoamperometric response when far away from a given substrate (it,∞), but it measures a differential feedback current as it approaches. The direction of the current difference depends on the substrate: positive for a conductive surface and negative for an insulating surface. Alternatively, one can measure redox active species generated at spatially discrete areas of a substrate by using the instrument in substrate generation/tip collection mode. The SECM has found use in a number of studies as an imaging technique in biological and materials applications. In an imaging study, the tip is rastered across a substrate, and differences in current are observed depending on whether it is over an insulator or conductor, or whether it is over an area of the substrate that is generating an electroactive species. Some brief examples should help clarify this concept. Mauzeroll, Bard, and coworkers were able to examine single cell metabolism of the cytotoxic compound menadione in Hep G2 cells (Figure 34a) using the substrate generation/tip collection mode.150 Wilburn and coworkers were able to image pores formed in lipid bilayer membranes by alamethicin (Figure 34b) using electroactive Ru-hex leakage from below the membrane as an imaging indicator. 151 Using feedback mode, Black and coworkers examined libraries of ruthenium-platinum thin films, potential fuel cell anode catalysts, in a 25 x 25 mm array (Figure 34c) with acid as the redox mediator. 152 Figure 34: SECM as a useful imaging technique in biological and materials applications.
(A) SECM image superimposed on a transparent optical image of Hep G2 cells challenged with menadione. Challenge causes Hep G2 cells to metabolize this compound into the redox active thiodione, which is ejected from the cell by ATP dependent pump (higher current, lighter green area). 140 Copyright 2004, National Academy of Sciences, reused with permission. (B) Alamethicin pore formation in a bilayer lipid membrane allows leakage of redox active Ru-hex, which is detected by the SECM tip. 141 Copyright 2006, Royal Society of Chemistry, reused with permission (DOI: 10.1039/B510649D).
(C) Combinatorial library of ruthenium-platinum thin films, imaged by SECM, in feedback mode, with H2SO4 containing buffer to look for optimal anode catalysts for fuel cells.143 Copyright 2005, IOP Publishing, Ltd., reused with permission.
SECM can also be used in the approach mode in order to determine information about the electroactive species in solution, as supposed to using the electroactive species to probe the substrate, as displayed in Figure 35.
Figure 35: SECM experimental design to measure electron transfer between platinum electrodes and solvated gold nanoparticles. Nanoparticles are initially oxidized at the tip by a cathodic voltage. Diffusion, mainly, carries the oxidized nanoparticles to the substrate electrode, where they are subsequently reduced. The tip current (it) increases as the tip approaches (i.e. distance, d, decreases) due to a higher local concentration of reduced species as the electrical double layers of both electrodes begin to merge.
Notably, in the scope of this study, Murray and coworkers showed how this technique was capable of measuring electron transfer kinetics between Au38 clusters and aqueous redox species at a liquid/liquid interface. 153 Using similar methodology in a single-phase organic solvent system, our lab was able to show that, for increasing alkanethiol chain length on a gold nanoparticle, the electron transfer rate slows.138 Approach curves for increasingly longer alkanethiolate protecting groups are displayed in Figure 36.
Figure 36: SECM approach curves of AuAlkanethiolate nanoparticles in anhydrous dichloromethane with 1V Pt UME, 0V substrate electrode (vs. Ag/Ag +). Solutions were 20 mg nanoparticle in 5 mL dichloromethane with 0.1 M TBAPF6.138 The dark blue line for mass transfer theory represents a case where feedback is not limited by diffusion and electron transfer rate, calculated from equation 4-6. Kinetic theory represents curves that are best fit from equation 4-5. Approach curves for nanoparticles fall further off of this line as electron transfer is slowed due to lower tunneling probabilities. Copyright 2006, American Chemical Society, reused with permission.
Peterson reasoned that the electron transfer rate became slower with protecting alkanethiol chain length because the tunneling probability is reduced with greater tunneling distance. This is represented in Figure 36, qualitatively, by the falling off of approach curves from that predicted by mass transfer theory, given by equation 4-5 in the discussion that follows.
The unique attributes of UMEs allow for examination of limiting effects on total current.
Quantitatively, the electron transfer rate can be backed out by modeling a best fit to the
approach curve. Steady-state current is defined by:
(4-1) where it,∞ is the steady-state tip current far from the substrate, n is the number of electrons transferred, F is Faraday’s constant, D is the diffusion coefficient of the electroactive species, C* is the bulk concentration of electroactive species, and a is the radius of the disk UME. Mass transfer limited current as the tip approaches the surface is
() (4-2) where it is the tip current while approaching the substrate, and m(L) is the mass transfer coefficient as a function of L, the normalized tip to substrate distance (L = d/a, where d is the raw tip to substrate distance). If current is kinetically limited, and the process is
unidirectional at the electrode, which is assumed to be true for a biased electrode, then:
(4-3) where kf is the forward electron transfer rate constant. In the case where the current is
both kinetically limited and mass transfer limited, it is useful to use:
(4-4) where it,exp is the experimental tip current at a given distance that is limited by both electron transfer kinetics and mass transfer.
The dimensionless normalized tip current is used to evaluate limiting behavior:
( )( ) () () (4-4)
where IT(L) is a computational fit to a non-limited experimental system,154 given by:
() ( )
cancel out. Equation 4-4 implies that the shape of an approach curve is not dependent upon concentration or the number of electrons transferred. Ultimately, the approach curve is only dependent upon the diffusion coefficient, the electron transfer rate constant, and the electrode geometry.