«Kinetic Investigations of Thiolate Protected Gold Nanoparticles: Protein Interactions, Electron Transfer, and Precursor Formation By Brian N. Turner ...»
It was determined that the presence of both initial [H+] and [Cl-] retard the reaction rate, but not in a concentration dependent fashion (at least not within the range studied here).
Initial [Cl-] reduces the rate of reaction to about 1/3 and [H+] reduces the rate to about 2/3 of what it was without the presence of either species. Reduction in rate with the addition of substitution side products was consistent with a mechanism where both species were on the product side of an equilibrium process pre-reduction, which was suspected on the basis of previously mentioned mechanistic work on Au(III) reduction reactions. A variation of rate with concentration of either species might have been observed at lower concentrations of those species. Importantly, that the effects of [H+] on rate might have been attributable to other processes or species generated in the stock solution, such as methyl perchlorates, methyl esterified tiopronin, or hydrolyzed AuCl4 -, should be restated.
Additionally, the total ionic strength of the solution effects the kinetics of any reaction, especially those between charged species. In the current case, AuCl4 -, Cl-, and H+ were all charged and likely were involved in the reaction. Ionic strength influences the reaction by mediating collisions between charged and polar species. The effect on rate
constant was given by the following equation:
(5-8) where k is the rate constant as a function of ionic strength µ, k0 is the rate constant at zero ionic strength, z is the charge on reactant A or B, and α and β are free parameters. The βµ term is to correct for ion pairing effects. This equation could be useful to either extrapolate rate constants determined at a given ionic strength to the reaction at low ionic strength, or to back out the ionic strength given the experimentally determined rate constant.
Accomodations should be made if the total reaction chain is suspected to have prereduction equilibria, or a final equilibrium state, since the reaction could very well exhibit a kinetically limited behavior at elevated concentrations of either reactant. This phenomenon is known as saturation kinetics. The initial reaction velocity, in this case, is
given by a rate law of the form:
() () () (5-9) where a and b are experimentally determined parameters that relate to rate constants in different concentration regimes. If, then the second order reaction is
and b can be determined by plotting a linearized form of the rate law as follows:
(5-10) where k will be considered knet for n = m = 1and the slope and intercept are interpreted as 1/a and b/a, respectively.
The experiments outlined in Table 12 were performed to evaluate saturation kinetics at fixed ionic strength. Saturation behavior was apparent (via observation of a loss in linearity, R² = 0.9649) in a plot of kAu(III),1 versus [RSH] at higher [RSH] and higher [RSH]:[Au(III)], displayed in Figure 51.
Figure 51: Saturation behavior was suspected when kAu(III),1 lost a linear relationship with [RSH] at higher [RSH].
Table 12: Rate data for conditions designed to test saturation kinetics of the tiopronin/tetrachloroaurate system. All experiments in 101 mM NaClO4 in methanol.
kAu(III),1 is the first order rate constant in this table, not the fractional 2/3 order determined previously. Blue highlights data where [Au(III)] was varied and pink highlights data where [RSH] was varied. The calculated rate data is described in the following text.
large [RSH] span, so the rate law in equation 5-9 was applied by determining coefficients a and b by using the linearized version, equation 5-10. The plot is displayed in Figure 52.
Figure 52: Determination of parameters a and b for the saturation kinetics rate law. The slope is 1/a and the intercept is b/a.
The linear equation determined that the parameters are a = 0.162 ± 0.005 M-1 s-1 and b =
44.3 ± 5.7 M-1. This implies that the second order rate constant in the second order rate law, which is more closely followed at b[RSH] 1, is k2nd = a = 0.162 ± 0.005 M-1 s-1 and when b[B] 1 then the first order rate law is more closely followed with kAu(III) = a/b =
3.66 x 10-3 ± 0.46 s-1 at the given ionic strength in methanol.
Referring back to Table 12, it is apparent that the saturation kinetics rate law, equation 5describes the data very accurately (on the basis of the % difference between the modeled and determined rates), with the exception of the reaction that was performed at the heaviest excess and highest concentration of RSH. This deviation is likely due to a second saturation point being reached, arising from the second equilibrium step, which is not accounted for in the rate law. A reaction scheme that can be associated with these observations, according to Wilkins,180 is one where there are two consecutive reactions with one element of reversibility. For this reaction, that corresponds with Scheme 5-4.
Scheme 5-4: The Au(III)Cl4- tiopronin reaction viewed as a one exchange equilibrium, one reduction process.
Scheme 5-4 is a simplification of Scheme 5-5, which follows more closely the form of that proposed by Elding and coworkers.
Scheme 5-5: The two consecutive exchange equilibrium, two simultaneous reduction mechanism for Au(III)Cl4- and tiopronin.
Scheme 5-5 is difficult to evaluate because the determination of ligand exchange products is not possible with the current methodology as the UV-Visible spectrum of the exchanged products is not significantly different from the starting material. This scheme is, however, supported by the rate retardation upon initial addition of Cl- or H+.
These Schemes are further supported by the fact that addition of Cl- or H+ reduces the rate of reaction. Examining the simplified Scheme 5-4, and using the following
( )( ) () (5-11) () () () () (5-12) where Ksub is the substitution reaction equilibrium constant, k = kAu(III) for the pseudo first order process (not the 2/3 order process), k1 is the substitution rate constant, k-1 is the dissociation rate constant, [Au(I)] represents all produced Au(I) species and [Au(III)] represents all Au(III) species, and kred is the rate constant for the reduction of all Au(III)
species to any Au(I) species, then equation 5-13 can be used to describe the system:
() (5-13) where a = kredKsub and b = Ksub. This equation is of the form of 5-9, therefore, kred = 3.66 ± 0.46 x 10-3 s-1 and Ksub = 44.3 M-1. The accuracy of this rate law (equation 5-9) in predicting the observed initial rates is displayed in Table 12. For all but the highest [RSH]:[Au(III)] (300:1), the model accurately predicted the initial rate within ±10%. The error in the 300:1 experiment might suggest that a second saturation kinetics limit was reached, dictated by the second equilibrium in Scheme 5-5. Experiments are ongoing to reproduce these results across a large range of [RSH] in identical media (excess [H +], [Cland µ). Studying the variation of initial rate with lower [Cl-] in order to determine equilibrium constants, such as Elding has done for the previously mentioned dimethyl sulfide and thiocyanate systems, is also desirable.169, 170 Examination of the kinetics of the analogous bromide complex was also attempted. The reaction happens too fast at room temperature in methanol to accurately measure the initial rates (t1/2 20s, significant conversion occurs during the mixing of solutions). In one set of experiments, a negative reaction order with respect to gold was observed, but on the basis of the chloride experiments and given the speed of reaction, these results were disregarded. There is much speculation in the nanoparticle synthesis community that AuBr4- would be a useful starting material for Brust type syntheses due to the decreased force constant of Br- with respect to Cl- making for a more labile Au(III) complex,181 and also because Br - complexes are less susceptible to hydrolysis reactions.182 Preliminary investigations have determined that Au(III)Br 4- is reduced much faster by tiopronin than Au(III)Cl4-, and, furthermore, more readily goes to completion (100% conversion for Br- as supposed to approximately 80% conversion for Cl-). Differences in percent conversion were noted for our homemade HAuCl4 trihydrate with the HAuCl4 trihydrate purchased from Sigma-Aldrich. These differences could arise from either the presence of nitrates (originating from nitric acid used in the original synthesis) or catalytic amounts of iron or palladium contaminants (arising from novice students using metal spatulas or contaminated spatulas for retrieval of tetrachloroauric acid) oxidatively driving the final conversion of Au(I)(RSH)2 to the polymeric or cyclic precursor material. Water content and excess acid should also be considered as possible sources of the differing behavior between the two samples.
In conclusion, a rate law that accurately describes the reaction at low [RSH]:[Au(III)] (equation 5-6), and a rate law that describes the reaction fairly accurately across a larger range of [RSH]:[Au(III)] (equation 5-9), were proposed. The latter rate law suggests that this reaction is mechanistically similar to the general Scheme proposed by Elding and coworkers for various substitution/reduction reactions of Au(III) complexes. In order to more completely support this mechanism, a thermodynamic study of this reaction would be necessary in order to observe whether the reduction proceeds via an inner-sphere or outer-sphere mechanism. If exchange must necessarily precede reduction, then one would expect changes in entropy consistent with an inner-sphere electron transfer. The
both to prove their existence and to accurately determine equilibrium constants for the exchange processes (Scheme 5-5), would yield important data in the future.
In the near future, study of this system will continue to be pursued. It would be useful to set up collaboration with a research group in possession of a stop-flow instrument with temperature control in order to direct a proper kinetic evaluation with thermodynamic information. Going forward, it would be of great interest to marry these studies with studies of the polymer or cyclic complex growth, much like Hainfeld and coworkers did with the glutathione/Au(III)Cl4 - system.176 It is reasonably expectable that by systematically varying pre-reduction reaction time, temperature, and starting materials to control the exact composition of the precursor species could have effects on the size and surface structure of the final nanoparticle. These are experiments that could prove useful in explaining the previous work of the Murray research group where they varied Brust nanoparticle synthesis conditions to obtain different sized nanoparticles.63
I would like to thank Dr. Kellen Harkness for his assistance with the background information, input on the kinetics experiments, and understanding of the reaction products. I would also like to acknowledge Andrzej Balinski for his assistance with the ongoing 1H NMR kinetics experiments (not discussed here), and Matt Casey for verification of some of the preliminary observations discussed.
As dicussed in Chapter I, thiolate protected gold nanoparticles have gathered much interest in the past decade due to their ease of synthesis, ability to be functionalized, their biological interactions, and unique electronic, optical, and chemical properties. In the described research, we have utilized these properties to create functional nanomaterials with applications in biomimetics and nanomolecular electronics. Additionally, fundamental aspects of nanoparticle synthesis and characterization were studied.
A peptide functionalized tiopronin protected gold nanoparticle, with approximate average composition of Au696Tiopronin261(CSGSGNSELLSLINDMPITNDQKKLMSNN)4 was synthesized and found to strongly and specifically bind to the pharmaceutical monoclonal antibody Palivizumab in a quartz crystal microbalance immunoassay, as described in Chapters II and III. The strength of the binding, as determined by a kinetic concentration-dependent study (Kd = 292 ± 177 nM) was competitive with a biomimetic peptide epitope scaffold which integrated the same linear peptide sequence (Kd = 87 nM)125 This biomimetic nanomaterial may prove useful as a novel subunit vaccine for the prevention of human respiratory syncytial virus, pending in vivo analysis.
Alkanethiolate protected gold nanoparticles functionalized with molecular wire thiolate ligands were synthesized and evaluated using scanning electrochemical microscopy (see Peterson et. al. 2006). Nanoparticles functionalized with wire molecules were found to exhibit faster electron transfer kinetics than unfunctionalized nanoparticles, described in Chapter IV. These “wired nanoparticles” might be useful in nanomolecular electronic devices.
For the case of water soluble tiopronin protected gold nanoparticles, which are useful in biological studies such as that presented in Chapter III, the complexation and reduction process preceding the final reduction step in their synthesis was studied, and a rate law was determined in Chapter V. In methanol/perchlorate solution, the reduction of tetrachloroauric acid trihydrate by tiopronin, as followed by UV-Visible spectroscopy, was found to follow saturation kinetics, consistent with a proposed mechanism where gold(III) is first complexed by one to two equivalents of tiopronin, followed by subsequent reduction by an additional equivalent of free tiopronin.
Additional accomplishments discussed include a comparative study of thermal gravimetric analysis and elemental analysis of thiolate protected gold nanoparticles (Appendix A), the synthesis and characterization of water soluble dendrimer-thiolate protected gold nanoparticles (Appendix B), and attempts at alternative synthetic routes to a molecular wire ligand (Appendix D).