«Kinetic Investigations of Thiolate Protected Gold Nanoparticles: Protein Interactions, Electron Transfer, and Precursor Formation By Brian N. Turner ...»
In the current study, the theories discussed above in the context of standard kinetic and mechanistic study of transition metal complexes, as presented by Wilkins,175 were applied to the reaction between Au(III)Cl4- and tiopronin. With the results, the research aimed to provide insights on the pre-reduction steps of small gold nanoparticle syntheses of similar to the Brust method. Specifically, the Au(III)/tiopronin system was chosen as it leads to stable high-yield water-soluble clusters by straightforward methodology, as discussed in Chapter I. Knowledge of the pre-reduction conditions will aid researchers in developing standardized and better informed synthetic methods and, ultimately, routes to new nanomaterials. Indeed, Hainfeld and coworkers have already studied the variation in nanoparticle size with the variation in the size of the precursor polymer for the glutathione/tetrachloroaurate system.176 The hypotheses were motivated by Richard Feynman’s claim that “there is plenty of room at the bottom,” as studying the complexes prior to nanoparticle formation represents the basement of the nanoparticle synthesis picture.
Materials For this section only, HAuCl4- •3H2O was purchased from Sigma-Aldrich, along with optima methanol, 70% perchloric acid, and tiopronin. Sodium perchlorate (biology grade) and sodium chloride were purchased from Fisher scientific.
UV-Visible Spectroscopy Kinetics Experiments A Cary Bio-100 UV-Vis spectrophotometer was used to probe the initial rates, and full kinetic profiles of reactions between varying amounts of HAuCl4 - •3H2O and tiopronin.
Stock solutions of gold in methanol were prepared, and dilutions of this stock in methanol or salt/methanol were measured into 1 cm quartz cuvettes, and initial spectra were obtained. Then, an Eppendorf Research digital or Reference manual micropipette was used to inject an aliquot of a tiopronin stock solution in methanol or salt/methanol into the cuvette, followed by rapid mixing via blowing air through with a glass pipette and quickly shutting the instrument cover. Keeping the room dark allowed for reduction of the interference from other light sources just post-injection. Using the simple kinetics or scanning kinetics software packaged with the instrument, the change in the absorbance peak at λmax (local) = 320 nm is observed over a period of 15 minutes or up to 3 hours depending on the speed of the individual reaction and what information was desired.
Points were taken continuously for the first 1 to 3 minutes, and then every 30 seconds to 1 minute thereafter. In order to investigate medium effects, stock solutions of 100 mM NaCl in methanol, 100 mM sodium perchlorate in methanol, or 101 mM perchloric acid in methanol were occasionally used for dilution or added to achieve the desired concentrations/ionic strengths.
Preliminary Observations The reaction of Au(III)Cl4- with tiopronin in methanol has been followed at various stoichiometries, and the products have been characterized. At a [RSH]:[Au(III)] 3:1, fading of the solution, suggesting the reduction of Au(III) to Au(I) is observed. If allowed to stand longer, the solution returns to yellow and the deposition of gold metal on the reaction vessel confirms the disproportionation of Au(I) to Au(III) and Au metal, as described by Gammons and coworkers.177 At [RSH]:[Au(III)] ≥ 3:1, the solution fades over the course of about an hour to clear, followed by precipitation of a white solid over the course of subsequent hours. Increasing [RSH]:[Au(III)] from 3:1 to 5:1 changes the appearance of the white precipitate from more powdery to more ribbon like. It can be concluded from these observations that the stoichiometry of reaction is at least 3:1, which is consistent with the 4:1 ratio that is expected.
UV-Visible Spectroscopy Given the bright color of its complexes in solution, the spectroscopic determination of Au(III) is straightforward. In the case of tetrachloroaurate(III), two absorbance maxima in the UV-Visible spectrum are readily apparent at 320 nm and 271 nm. During reduction of Au(III) to Au(I) by tiopronin, the band at 320 nm decreased in absorbance intensity over time, and a saddle with the absorbance at 271 nm near 288 nm was observed. A second saddle between the 271 nm absorbance maxium and a shoulder to the right of that peak was observed at about 250 nm. Time resolved spectra are presented in Figure 44 below.
Figure 44: Time resolved UV-Vis spectra of AuCl4- undergoing reduction by tiopronin.
For this specific experiment, the initial conditions were [tiopronin] = 16.9 mM, [HAuCl4] = 0.0955 mM, [HClO4] = 6.00 mM, [NaCl] = 1.00 mM, and [NaClO4] = 101 mM, all in methanol. Spectra were collected at approximately 30 s intervals for the first 1 min, every 1 min for the next 30 min, and every 5 min for the next 145 min. All spectra were collected at a scan rate of 20 nm/s with a spectral bandwidth of 2 nm. The local absorbance maxima at 320 nm (inset, zoomed in) started at A = 0.442, initially increased to A = 0.445 for the second scan, and decreases to a minimum of A = 0.0833 after 110 min of reaction time (±5 min). Isosbestic points at 250 nm and 288 nm are circled in red.
Upon further examination of the region around the saddle at 289 nm, the presence of an intermediate was apparent as the intensity just to right of the saddle initially increased, and just to the left initially decreased, as displayed in Figure 45.
Figure 45: UV-Vis region containing the saddle at 289 nm for the gold(III)/tiopronin system. Colored arrows and labels correspond with the start time of each scan from the reaction start time for the corresponding colored spectra. Note the initial increase in intensity to the right of the saddle followed by a rapid decrease, and the opposite phenomenon on the left.
Eventually, the direction of absorbance change became a decrease to the right of the saddle and an increase to the left of the saddle. The presence of the two saddles in combination with the deviations in intensity change direction during the first 1 minute of reaction is consistent with a transition that originated from an intermediate species. This intermediate species is probably a substitution product, but could possibly be a 5coordinate intermediate preceding the substitution product178 or an intermediate for the reduction process with a second or third thiol bound to a substituted thiol, or a combination of any or all of the mentioned species.
The nature of the absorbance changes and multiple isosbestic points presented problems with accurate spectroscopic determination of Au(III). For instance, the non-zero absorbance upon completion of the reduction reaction at 320 nm could have been interpreted as an incomplete reaction, dynamic equilibrium, or interference from the tail of a transition not associated with Au(III). Also, the changing dynamics of the 320 nm absorbance band suggests that a combination of interference from another transition and/or a change in extinction coefficient may have casued a significant deviation in accurate determination of Au(III) concentration from the spectrum. Given few options to circumvent these difficulties, a kinetic analysis of the data was undertaken with the assumption that Au(III)Cl4- was chiefly responsible for the absorbance band at 320 nm.
When tetrabromoauric acid (HAuBr4) was analyzed by identical methodology, a complete disappearance of the absorbance band at 390 nm was observed upon reduction with tiopronin, as illustrated in Figure 46. The kinetics of this system were difficult to evaluate quantitatively as the reaction was too fast to measure without a stop-flow system.
Figure 46: Time resolved UV-Vis spectra of tetrabromoauric acid undergoing reduction by tiopronin.
Given the fast kinetics and the lack of sufficient instrumentation to measure them accurately, the gold chloride system will be the focus of the discussion that follows.
The overall process that can be monitored with UV-visible spectroscopy by observing the decrease in absorbance at 320nm which corresponds to the reduction of Au(III) (yellow) to Au(I) colorless in the presence of excess tiopronin. Using Beer’s law, a calibration curve for Au(III)Cl4- is easily obtained yielding an extinction coefficient of ε 320nm = 4660 M-1 cm-1, which is reasonably consistent with reported literature values. The dynamic reaction profiles of three reactions at variable [RSH] and fixed [Au(III)Cl4 -] are shown below in Figure 47.
Figure 47: Dynamic reaction profiles of Au(III)Cl4- with tiopronin in a 3:1 (green), 6:1
(red), and 9:1 (blue) molar ratio.
Treating this reaction as a pseudo first order process, the rate law is:
() [ ] () [ ] (5-1) where t is time in seconds, k is the pseudo rate constant, y is an integer value from 0 to 2, and m is the integer order of reaction with respect to the corresponding species. For simplicity going forward, [Au(III)] will be considered as the concentration of all Au(III) species in solution. Pseudo first order conditions are followed when the dynamic reaction profile can be fit to the following linear equation if the process is first order with respect to AuCl4-, () () (5-2) or to the following linear equation in the event that it is second order with respect to
() () (5-3) Fitted plots are displayed below in Figure 48.
Figure 48: Plots to determine the order of reaction with respect to Au(III)Cl4− in pseudo first order conditions, both first order (left) and second order (right). Blue represents points from earlier time points in the reaction and red represents later time points in the reaction.
As the entire profile would not fit to either equations 5-2 or 5-3 for the entire process, the common conclusion is that there are consecutive reactions occurring. Indeed, if the plot is broken into two parts, linear fits are found for either first or second order for either time regime during the reaction course. Discrimination of first and second order processes by this method can be difficult. In the analysis that follows, only observation of the reduction process by the method of initial rates under pseudo first order conditions was attempted. The first step in the process was to vary [Au(III)Cl4−] in deficiency with respect to tiopronin, shown in Figure 49 below.
Figure 49: Plot to determine the reaction order and pseudo first order rate constant with respect to [Au(III)].
The relationship was nearly linear (R² = 0.988), but when examined more closely, it was apparent that the reaction order with respect to [Au(III)] was closer to m = 2/3. In order to convert the pseudo rate constant, k, to the real rate constant(s), the concentration of tiopronin, [RSH], was systematically varied. The rate data is presented in Table 10, and the variation of kAu(III) with [RSH] is displayed in Figure 50.
Figure 50: Plot to determine the order of reaction with respect to [RSH], n, and the overall rate constant, knet. The linear relationship signifies n = 1 and the slope is thus interpreted as the overall 5/3 order rate constant.
The observed kinetic behavior corresponded with a reaction that was first order with respect to [RSH], and the overall second order rate constant was determined as the slope
of the plot in Figure 48, according to the following equation:
( ) (5-4) where knet is the second order rate constant and the pseudo first order rate constant discussed earlier, k, has been re-labeled as kAu(III) for convenience. Therefore, knet = 6.22 x 10-3 M-2/3s-1 for the reaction at standard nanoparticle synthesis conditions (RSH:Au(III) 5, room temperature, no adjustment to ion strength, no adjustment to pH).
In the event of a non-zero y-intercept, the following equation is necessary to describe the
full rate law:
() () () (5-5) where k1 and c describe a second process for the reduction that is independent of [RSH].
However, the y-intercept is however insignificant compared to the variability of rates determined by the full rate law from the observed rates (k1[Au(III)] % difference), and can be ignored. It should be noted that Au(III) solutions degrade over time, and, given more accurate analytical methods, a small but measurable k1 value would not be unexpected (e.g. solvolysis). That being said, a final rate law for the initial process in the
(5-6) While this rate law is practical for determining reaction velocities under conditions used during the synthesis of gold tiopronin nanoparticles, there are a number of shortcomings in using it to predict a mechanism. Furthermore, the the 2/3 order in [Au(III)] would imply an exotic mechanism, which raises skepticism.
Here, it is prudent to consider the effect of medium and other particpants in the reaction.
For instance, the variability of [H+] over the course of the reaction could have a varying medium or mechanistic effect on the rate, as both reagents are acidic, and, at some point, hydrogen atoms are lost from the tiopronin thiol groups, possibly as protons. It is also suspected that the [H+] concentration effects solvolysis and esterification reactions between primary reactants and solvent molecules (methanolysis, hydrolysis, 179 and methanol esterification). Another similar issue is with the [Cl-], as chloride is liberated from AuCl4- as tiopronin participates in equilibrium exchange reactions with gold prior to
the final reduction. This would lead to a rate constant of the type:
( ) (5-7) A handful of experiments were studied to assess the impact of [Cl-] and [H+] on the initial reaction under similar conditions to those above. The results are summarized in Table 11.
Table 11: Rate data at fixed [RSH] and [Au(III)] with variable [H+] (adjusted with perchloric acid/MeOH solution) and [Cl-] (adjusted with sodium chloride/MeOH solution). The concentrations are calculated assuming complete dissociation, although some ion pairing is expected to occur. Blue highlights the experiments where [Cl-] was varied and pink highlights experiments where [H+] was varied, while an experiment with no added [H+] and [Cl-] is in white. (a) indicates that while no [H+] was added, the exact [H+] concentration from dissociation of HAuCl4 and the carboxylic acid group of tiopronin was not determined.