- Email: [email protected]
Quantized double layer charging of dodecanethiol protected larger Au nanoclusters: combined investigations using differential pulse voltammetry, cyclic voltammetry and impedance technique
Electrochemistry Communications 6 (2004) 661–665 www.elsevier.com/locate/elecom
Quantized double layer charging of dodecanethiol protected larger Au nanoclusters: combined investigations using differential pulse voltammetry, cyclic voltammetry and impedance technique Nirmalya Kumar Chaki, Bhalchandra Kakade, Kunjukrishna P. Vijayamohanan
*
Physical and Materials Chemistry, National Chemical Laboratory, Homi Bhabha Road, Pashan, Pune 411008, India Received 7 April 2004; received in revised form 4 May 2004; accepted 4 May 2004 Available online 8 June 2004
Abstract The electrochemical quantized double layer charging of 4.63 0.04 nm sized Au nanoclusters protected with dodecanethiol [ca. Au2869 (DDT)541 ] has been investigated using differential pulse voltammetry, cyclic voltammetry and impedance measurements. Our results show for the first time that these particles are clearly accessible for single electron charging despite their higher capacitance (1.95 aF). Analysis of the impedance spectra using the Randles equivalent circuit reveals that the process is very sluggish, where the electron transfer rate constant is found to be of 8.27 106 cm s1 . Ó 2004 Elsevier B.V. All rights reserved. Keywords: QDL charging; Au2869 (DDT)541 ; Differential pulse voltammetry; Cyclic voltammetry; Impedance analysis
1. Introduction The monolayer protected nanometer sized metallic and semiconducting clusters have received extensive attention in recent years due to their promising applications as building blocks for designing nanoelectronic circuit components like molecular switches, single electron transistors, field effect transistors and resonant tunneling diodes [1–5]. All these applications are based on the fascinating single electron transfer property, which arises due to their smaller dimension and size dependent electronic behavior [1–6]. The voltammetric measurements of smaller sized MPCs in solution or self-assembled on electrode surface are known to show single electron transfer behavior at room temperature due to the sub atto-farad (aF) double layer capacitances (CCLU ) of these MPCs, which is prevalently known as quantized double layer charging (QDL) [7–10]. More importantly, QDL charging is comparable with the scanning tunneling microscope (STM) based
‘‘Coulomb Staircase’’ or ‘‘Coulomb Blockade’’ behavior [6,7]. The QDL phenomena have been explored for both metal (i.e., Au, Ag, Cu and Pd) and semiconductor (i.e., CdS, PbS, Si, etc.) MPCs, although the majority of the study is concerned with smaller sized AuMPCs due to their enhanced stability [7–16]. Although, there have been many reports available for the electrochemical behavior of these smaller (1.1–3 nm diameter) MPCs, no such report is known for larger (>3 nm) AuMPCs [7–10]. We report for the first time the electrochemical QDL charging behavior of the larger AuMPCs (4.63 0.04 nm) protected with dodecanethiol (DDT) in dichloromethane using differential pulse voltammetry (DPV), cyclic voltammetry (CV) and impedance measurements. Most importantly, our study summarizes that the QDL charging is evident for these MPCs, which show a large population of equally spaced charging events despite their higher capacitance (1.95 aF).
2. Experimental *
Corresponding author. Tel./fax: +91-020-25893044. E-mail address: [email protected] (K.P. Vijayamohanan).
1388-2481/$ - see front matter Ó 2004 Elsevier B.V. All rights reserved. doi:10.1016/j.elecom.2004.05.003
Dodecanethiol (99%), NaBH4 and tetrabutylammoniumhexaflurophosphate (TBAHFP) and
662
N.K. Chaki et al. / Electrochemistry Communications 6 (2004) 661–665
HAuCl4 3H2 O were obtained from Aldrich. Toluene, acetone and dichloromethane are of AR grade from Merk and were used after further purification. In all these experiments deionized water (16 MXÞ from MilliQ system was used. TEM micrographs of AuMPCs taken on a Philips CM 20 FEG (HRTEM) instrument operated at an accelerating voltage of 200 kV using dropcasted films from toluene on a carbon coated Cu grid. X-ray diffraction was carried out on a Philip1730 machine using Cu Ka radiation at a step of 0.02° (2h) at room temperature and NMR spectrum was recorded in a Bruker MSL300 MHz in CDCl3 solvent using TMS as an internal standard. UV–Vis absorption measurement was carried out at room temperature on a UV-2101PC double beam spectrometer (Shimadzu) using quartz cells (1 cm path length). All the electrochemical measurements were carried out in dichloromethane using 0.1 M tetrabutylammoniumhexaflurophosphate (TBAHFP) supporting electrolyte in an ice bath (0–3 °C) using a standard three electrode cell comprising Pt micro electrode (50 lm) as the working electrode, a Pt wire as counter electrode and another Pt wire as quasi-reference electrode (calibrated independently with an internal standard ferrocene/ferrocenium cation) unless mentioned separately. Differential pulse (modulation time 0.02 s, interval time 0.1 s, step potential 0.0025 V and modulation amplitude 0.02 V), and cyclic voltammetric experiments were performed on an Autolab PGSTAT30 (ECO CHEMIE) instrument, whereas electrochemical impedance measurements were carried out in an impedance analyzer (Autolab PGSTAT 30 with FRA software) employing a Pt disc (0.5 mm diameter) working electrode. The ac signal amplitude was 5 mV, and the frequency range employed was 1 Hz–50 kHz and the data recorded with averaging over three cycles for each frequency.
3. Results and discussion 3.1. Synthesis of AuMPCs Monolayer protected Au nanoclusters (MPCs) were prepared by modified Brust synthesis route [17,18] using dodecanethiol as capping agent followed by a post heat treatment (digestive ripening [19]) at 120 °C in presence of excess unbounded thiol (Au:thiol; 1: 30). In brief, particles were synthesized in a biphasic mixture of water and toluene (v:v; 1:1; 50 ml each) using 1:3 mole ratio of gold salt to thiol in ice bath. The biphasic mixture under low temperature (ice bath) was stirred for 30 min and subsequently reduced using drop wise addition of 0.1 M 20 ml aqueous NaBH4 solution. Upon the slow addition of NaBH4 , the pale yellow colour of gold salt was
transformed to reddish-violet indicating Au0 cluster formation and after vigorous stirring for 3 h, the Au0 clusters were found to be shifted to the organic phase turning the upper layer reddish-violet. After separation of non-aqueous portion from the aqueous layer, the non-aqueous layer was collected and the process was repeated 20 times to finally obtain 1 l of toluene containing both AuMPC and unbound thiol. The nonaqueous layer was then concentrated under vacuum at 40 °C to a final volume of 100 ml, where the ratio of Au to unbound thiols is typically 1:30. This reaction mixture was further boiled at 120 °C for 1 h in an oil bath and was subsequently cooled slowly at room temperature to form a dark red solution. The reaction mixture was further concentrated (10 ml) and the MPCs were precipitated by the addition of excess acetone (250 ml). The particles were allowed to settle and excess acetone was decanted to reduce the volume to 20 ml, followed by repeated centrifugation (7 times) and decantation to remove excess unbound thiol, and other unwanted reaction byproducts. 3.2. Characterization Fig. 1 shows a comparison of the TEM image of these particles, where a uniform Gaussian distribution with an average particle size of 4.63 0.04 nm is evident. The composition of the average sized AuMPCs is also approximated from the TEM results by calculating the nearest stable magic number as the full shell clusters, since certain magic numbers are known to be stable [20]. The average composition of these MPCs is approximated as Au2869 (DDT)541 , by assuming the shape of these particles as spherical, radius of Au atom as 144 pm, perfect fcc packing and further considering 2 S atoms per 3 Au atoms on the surface as reported for twodimensional SAM [21]. Moreover, particles are found to have bulk fcc structure as evident by the XRD in Fig. 2(a), whereas the inset shows the surface plasmon band of these MPCs at 523 nm. The purity of these particles have been confirmed from the NMR result (1 H), which shows that the individual peak positions of various protons [d 1:56, 1.25 and 0.84 ppm corresponding to the proton of CH2 (C2 position) close to S, CH2 (C3 –C11 positions) at the middle of the chain and the terminal CH3 (C12 position) group respectively] have been shifted and broadened significantly, indicating effective capping. However, the minor peak at d 1:06 ppm might be observed due to the negligible presence of uncapped DDT. 3.3. Electrochemical measurements The consecutive one electron QDL peaks observed in solution is diffusion controlled, which also obeys the Nernst equation similar to the traditional redox systems
N.K. Chaki et al. / Electrochemistry Communications 6 (2004) 661–665
663
Fig. 1. (a)–(b) HRTEM images of the purified AuMPCs passivated with dodecanethiol along with their particle size distribution in the inset of (a), showing uniform Gaussian distribution with an average particle size of 4.63 0.04 nm.
Fig. 2. (a) XRD patterns of AuMPCs dropcasted on a glass slide revealing bulk fcc structure, where the inset shows the surface plasmon resonance of these particles. (b) 1 H NMR response of these purified AuMPC in CDCl3 , showing that the peaks are significantly broadened and shifted from their ideal position due to the surface passivation of the Au core.
[7–16]. Consequently, the individual peak potentials in the QDL process can be considered as formal redox 0 potential E0 for each z=z 1 ‘‘redox couple or charge state couple’’, which can be described by 00 Ez;zþ1 ¼ Epzc þ ðz 1=2Þe=CCLU ;
ð1Þ
where, Epzc is the potential of zero charge (i.e., z ¼ 0) of the MPC, and z is assigned such that z > 0 and z < 0 corresponding to core ‘‘oxidation’’ and ‘‘reduction’’, respectively, and CCLU is the capacitance of the MPCs [7–10].
Interestingly, the smaller sized particles (<1.35 nm diameters) behave like molecular redox system, where bonding interactions (electron orbital-shell effects or pairing effects, or both) are dominated over electrostatic charging of the metal cores covered with a dielectric film [8]. This has been reflected as a large separation of DV in the potential of zero charge (i.e., Epzc ) regions in (1:1 < d < 1:9 nm) size regime, which is comparable to the HOMO–LUMO gap as anticipated as ‘‘Electrochemical Band Gap’’ [8]. However, the electrostatic principles governs the electronic charging of the core for d > 1:9 (i.e., double layer dominated over bonding interactions), where the capacitance is associated with the ionic space charge formed around an MPC dissolved in the electrolyte solution [7–10]. Fig. 3(a) shows the superimposed DPV response of these larger sized AuMPCs protected with DDT in dichloromethane on a Pt microelectrode (50 lm) at 0.96 mg ml1 concentrations at (i) 30, (ii) 20 and (iii) 10 mV pulse amplitude. Accordingly, Fig. 3(b) shows the DPV response in 1.6 mg ml1 MPC concentration, where the dip indicates the Epzc region. These DPV plots reveal a highly populated evenly spaced (ca. 80–90 mV) current peaks, where the ()1/0) and (0/1) charging peaks are marked for comparison (corresponding potentials are referenced to the Pt quasi-reference electrode). In conjunction with the results of DPV, these high populations of charge steps are also accessible by cyclic voltammetry. For example, Fig. 3(c) shows the superimposed response of these MPCs at (i) 100 and (ii) 50 mV s1 scan rate using 0.96 mg ml1 MPC concentration, where the successive one electron process are visible as evenly spaced sigmoidal waves. The higher capacitance values of these larger sized particles might, in principle, prevent better resolution by CV as compared to DPV. However, the derivatives (iii) and (iv) corresponding to the CV shown at graph (i) and (ii) are comparable with the DPV results. Interestingly, these MPCs do not show any signature of the ‘‘Electrochemical Band Gap’’ as observed for smaller sized particles, indicating that the process is particularly
664
N.K. Chaki et al. / Electrochemistry Communications 6 (2004) 661–665
Fig. 3. (a) The superimposed DPV response of AuMPCs at (i) 30, (ii) 20 and (iii) 10 mV pulse amplitude recorded using 0.96 mg MPCs/ml. (b) At 20 mV pulse amplitude using 1.6 mg MPCs/ml concentrations, showing highly populated QDL charging, where ()1/0) and (0/1) charging peaks are marked for comparison. (c) Superimposed CV response at (i) 100 and (ii) 50 mV s1 scan rate at 0.96 mg MPCs/ml concentrations along with their derivative response (iii) and (iv), respectively. (d) Graphs (i)–(iv) shows the superimposed highly linear ‘‘Z-Plots’’ corresponding to the DPV response (a) and (b), suggesting ideal QDL behavior.
governs by the double layer charging principles [7,8,14,16]. The capacitance (CCLU ) of these MPCs can be de00 termined from the slope of the linear plot of Ez;z1 vs. charge state (‘‘Z-plot’’) using the Eq. (1) [10]. Fig. 3(d) shows the superimposed ‘‘Z-plots’’ (i)–(iv) for these AuMPCs corresponding to the DPV responses shown at Fig. 3(a) and (b), where linear ðR2 > 0:99Þ behavior can be seen as expected for the ideal QDL behavior [7–10]. The capacitance (CCLU ) values calculated from the slope of the ‘‘Z-plot’’ (Fig. 3d) are 2.13, 1.86, 1.75 and 1.87 aF corresponding to graphs (i)–(iv) respectively. The above experimentally obtained CCLU values are in good agreement with the estimated CCLU of 4.63 nm sized AuMPCs (1.95 aF) with the help of a concentric sphere capacitance model as CCLU ¼ ð4pe0 er Þ½rðr þ dÞ=d;
ð2Þ
where, e0 is the primitivity of free space, er is the dielectric constant of the medium (taken as 3) respectively, ‘r’ is the radius of the MPC core (2.315 nm) and ‘d’ is the length of the passivating agent (1.52 nm) [10]. Interestingly, the full width at half maximum (FWHM) of each peak is found to be of ca. 40–45 mV particularly signifying the QDL process for these larger MPCs, which differs from the earlier reports of smaller
sized AuMPCs (110–115 mV) or ideal (90.6 mV) one electron transfer process for conventional redox species [6–9,22]. In particular, these sharp DPV peaks might be arising due to several complex interactions along with the reorganization of charges in clusters by electron selfexchange or dispropotionation among these assemblies of mixvalent redox centers at the electrode solution interfaces during electrochemical process [23–25]. More specifically, these types of interactions are widely known for conducting polymers, electroactive films or adsorption electrochemistry, where depending on the degree of the adsorbate–adsorbate interaction, substantial broadening or sharpening of the voltammetric peaks are observed [23–25]. We believe, that the double layer charging of these larger metallic MPC core could in principle solely responsible for these types of interactions, which perhaps is the reason for the observed narrower DPV peaks. Fig. 4 shows complex plane impedance plots of AuMPCs in DCM on a Pt disc (0.5 mm diameter), where a suppressed semicircle at the high frequency domain and a straight line with more than 45° angle to the real axis in the low frequency region are observed. Subsequently, the impedance data have been divided into high and low frequency regions for the convenience of interpretation and analyzed using the Randles equivalent circuit,
N.K. Chaki et al. / Electrochemistry Communications 6 (2004) 661–665
665
larger MPCs, which can be useful for designing future nanoelectronic devices. Acknowledgements NKC and BK gratefully acknowledge the Council of Scientific and Industrial Research (CSIR) and University Grants Commission (UGC), New Delhi. We are thankful to Mrs. Renu Pasricha, T.G. Gopakumar and Dr. S. Shulze for TEM facilities. References
Fig. 4. Complex plane impedance plots of AuMPCs in DCM on a Pt disc (0.5 mm diameter), where the suppressed semicircle at the higher frequency domain is fitted to a semicircle and analyzed with Randles equivalent circuit. Inset shows the linearly fitted (R2 > 0:99) modulus of the real and imaginary part of impedance data (10.72–1 Hz) with x1=2 .
considering the electrode reaction to be a one-step, one electron process [25]. The standard rate constant of the one electron QDL charging process is found to be 8.27 106 cm s1 , calculated using the following relationship, k ¼ RT =n2 F 2 RCT C
ð3Þ
assuming n ¼ 1; C ¼ C ðOÞ ¼ C ðRÞ ¼ ð9:5 107 mol cm3 ), a ¼ 0:5, T ¼ 273 K, RCT ¼ 3:105 104 X cm2 from the arc of the fitted semicircle in the real axis. However, the reversible feature observed in voltammetry (Fig. 3) especially for scan rates below 100 mV s1 do not agree with this sluggish nature and this could be justified due to various deviations from the ideal semiinfinite diffusion boundary condition due to adsorption, assumption of symmetric energy barrier, involvement of complex interactions among these MPCs, additional convective contribution due to the charged nature of MPCs etc. Further, the diffusion coefficient is estimated to be 1.21 107 cm2 s1 from the slope of the linearly fitted modulus of the real and imaginary part of the impedance data (10.72–1 Hz) with x1=2 as shown in inset of the Fig. 4, which is lower than that of its smaller analogues [9].
4. Conclusion In conclusion, our electrochemical results show that QDL charging is indeed possible for larger sized AuMPCs despite their higher capacitance values. We believe that these results will help to fill up the lacuna of information about the single electron transfer features of
[1] M.-C. Daniel, D. Astruc, Chem. Rev. 104 (2004) 293. [2] G. Schmid, M. Baumle, M. Geerkens, I. Heim, C. Osemann, T. Sawitowski, Chem. Soc. Rev. 28 (1999) 179. [3] D.I. Gittins, D. Bethell, D.J. Schiffrin, R.J. Nichols, Nature 408 (2000) 67. [4] D.L. Feldheim, C.D. Keatig, Chem. Soc. Rev. 27 (1998) 1. [5] R.P. Andres, T. Bein, M. Dorogi, S. Feng, J.I. Jenderson, C.P. Kubiak, W. Mahoney, R.G. Osifchin, R. Reifenverger, Science 272 (1996) 1323. [6] C.N.R. Rao, G.U. Kulkarni, P.J. Thomas, P.P. Edwards, Chem. Soc. Rev. 29 (2000) 27. [7] R.S. Ingram, M.J. Hostetler, R.W. Murray, T.G. Schaaff, J.T. Khoury, R.L. Whetten, R.L. Bigioni, D.K. Guthrie, P.N. First, J. Am. Chem. Soc. 119 (1997) 9279. [8] S. Chen, R.S. Ingram, M.J. Hostetler, J.J. Pietron, R.W. Murray, T.G. Schaaff, J.T. Khoury, M.M. Alvarez, R.L. Whetten, Science 280 (1998) 2098. [9] B.M. Quinn, P. Lijeroth, V. Ruiz, T. Kaaksonen, K. Kontturi, J. Am. Chem. Soc. 125 (2003) 6644. [10] S. Chen, R.W. Murray, S.W. Feldberg, J. Phys. Chem. B 102 (1998) 9898. [11] W. Cheng, S. Dong, E. Wang, Electrochem. Commun. 412 (2002) 412. [12] S. Chen, J.M. Sommers, J. Phys. Chem. B 105 (2001) 8816. [13] F.P. Zamborini, S.M. Gross, R.W. Murray, Langmuir 17 (2001) 481. [14] S.K. Haram, B.M. Quinn, A.J. Bard, J. Am. Chem. Soc. 123 (2001) 8860. [15] S. Chen, L.A. Truax, J.M. Sommers, Chem. Mater. 12 (2000) 3864. [16] Z. Ding, B.M. Quinn, S.K. Haram, L.E. Pell, B.A. Korgel, A.J. Bard, Science 296 (2002) 1293. [17] M. Brust, M. Walker, D. Bethell, D.J. Schiffrin, R.J. Whyman, Chem. Soc. Chem. Commun (1994) 801. [18] N.K. Chaki, M. Aslam, T.G. Gopakumar, J. Sharma, R. Pasricha, I.S. Mulla, K. Vijayamohanan, J. Phys. Chem. B 107 (2003) 13567. [19] B.L.V. Prasad, S.I. Stoeva, C.M. Sorensen, K.J. Klabunde, Langmuir 18 (2002) 7515. [20] J.D. Aiken III, R.G. Finke, Mol. Cata. A: Chem. 145 (1999) 1. [21] R.H. Terrill, T.A. Postelethwaite, C. Chen, C.D. Poon, A. Terzis, A. Chen, J.E. Hutchison, M.R. Clark, G. Wignall, J.D. Londono, R. Superfine, M. Falvo, C.H. Johnson Jr., E.T. Samulski, R.W. Murray, J. Am. Chem. Soc. 117 (1995) 12537. [22] A.J. Bard, L.R. Faulkner, Electrochemical Methods, Fundamentals and Applications, second ed., Wiley, New York, 2001 (Chapter 7 and 10). [23] J.F. Hicks, F.P. Zamborini, A.J. Osisek, R.W. Murray, J. Am. Chem. Soc. 123 (2001) 7048. [24] C.E.D. Chidsey, R.W. Murray, J. Phys. Chem. 90 (1986) 1479. [25] E.J. Laviron, J. Electroanal. Chem. 122 (1981) 37.
Quantized double layer charging of dodecanethiol protected larger Au nanoclusters: combined investigations using differential pulse voltammetry, cyclic voltammetry and impedance technique Nirmalya Kumar Chaki, Bhalchandra Kakade, Kunjukrishna P. Vijayamohanan
*
Physical and Materials Chemistry, National Chemical Laboratory, Homi Bhabha Road, Pashan, Pune 411008, India Received 7 April 2004; received in revised form 4 May 2004; accepted 4 May 2004 Available online 8 June 2004
Abstract The electrochemical quantized double layer charging of 4.63 0.04 nm sized Au nanoclusters protected with dodecanethiol [ca. Au2869 (DDT)541 ] has been investigated using differential pulse voltammetry, cyclic voltammetry and impedance measurements. Our results show for the first time that these particles are clearly accessible for single electron charging despite their higher capacitance (1.95 aF). Analysis of the impedance spectra using the Randles equivalent circuit reveals that the process is very sluggish, where the electron transfer rate constant is found to be of 8.27 106 cm s1 . Ó 2004 Elsevier B.V. All rights reserved. Keywords: QDL charging; Au2869 (DDT)541 ; Differential pulse voltammetry; Cyclic voltammetry; Impedance analysis
1. Introduction The monolayer protected nanometer sized metallic and semiconducting clusters have received extensive attention in recent years due to their promising applications as building blocks for designing nanoelectronic circuit components like molecular switches, single electron transistors, field effect transistors and resonant tunneling diodes [1–5]. All these applications are based on the fascinating single electron transfer property, which arises due to their smaller dimension and size dependent electronic behavior [1–6]. The voltammetric measurements of smaller sized MPCs in solution or self-assembled on electrode surface are known to show single electron transfer behavior at room temperature due to the sub atto-farad (aF) double layer capacitances (CCLU ) of these MPCs, which is prevalently known as quantized double layer charging (QDL) [7–10]. More importantly, QDL charging is comparable with the scanning tunneling microscope (STM) based
‘‘Coulomb Staircase’’ or ‘‘Coulomb Blockade’’ behavior [6,7]. The QDL phenomena have been explored for both metal (i.e., Au, Ag, Cu and Pd) and semiconductor (i.e., CdS, PbS, Si, etc.) MPCs, although the majority of the study is concerned with smaller sized AuMPCs due to their enhanced stability [7–16]. Although, there have been many reports available for the electrochemical behavior of these smaller (1.1–3 nm diameter) MPCs, no such report is known for larger (>3 nm) AuMPCs [7–10]. We report for the first time the electrochemical QDL charging behavior of the larger AuMPCs (4.63 0.04 nm) protected with dodecanethiol (DDT) in dichloromethane using differential pulse voltammetry (DPV), cyclic voltammetry (CV) and impedance measurements. Most importantly, our study summarizes that the QDL charging is evident for these MPCs, which show a large population of equally spaced charging events despite their higher capacitance (1.95 aF).
2. Experimental *
Corresponding author. Tel./fax: +91-020-25893044. E-mail address: [email protected] (K.P. Vijayamohanan).
1388-2481/$ - see front matter Ó 2004 Elsevier B.V. All rights reserved. doi:10.1016/j.elecom.2004.05.003
Dodecanethiol (99%), NaBH4 and tetrabutylammoniumhexaflurophosphate (TBAHFP) and
662
N.K. Chaki et al. / Electrochemistry Communications 6 (2004) 661–665
HAuCl4 3H2 O were obtained from Aldrich. Toluene, acetone and dichloromethane are of AR grade from Merk and were used after further purification. In all these experiments deionized water (16 MXÞ from MilliQ system was used. TEM micrographs of AuMPCs taken on a Philips CM 20 FEG (HRTEM) instrument operated at an accelerating voltage of 200 kV using dropcasted films from toluene on a carbon coated Cu grid. X-ray diffraction was carried out on a Philip1730 machine using Cu Ka radiation at a step of 0.02° (2h) at room temperature and NMR spectrum was recorded in a Bruker MSL300 MHz in CDCl3 solvent using TMS as an internal standard. UV–Vis absorption measurement was carried out at room temperature on a UV-2101PC double beam spectrometer (Shimadzu) using quartz cells (1 cm path length). All the electrochemical measurements were carried out in dichloromethane using 0.1 M tetrabutylammoniumhexaflurophosphate (TBAHFP) supporting electrolyte in an ice bath (0–3 °C) using a standard three electrode cell comprising Pt micro electrode (50 lm) as the working electrode, a Pt wire as counter electrode and another Pt wire as quasi-reference electrode (calibrated independently with an internal standard ferrocene/ferrocenium cation) unless mentioned separately. Differential pulse (modulation time 0.02 s, interval time 0.1 s, step potential 0.0025 V and modulation amplitude 0.02 V), and cyclic voltammetric experiments were performed on an Autolab PGSTAT30 (ECO CHEMIE) instrument, whereas electrochemical impedance measurements were carried out in an impedance analyzer (Autolab PGSTAT 30 with FRA software) employing a Pt disc (0.5 mm diameter) working electrode. The ac signal amplitude was 5 mV, and the frequency range employed was 1 Hz–50 kHz and the data recorded with averaging over three cycles for each frequency.
3. Results and discussion 3.1. Synthesis of AuMPCs Monolayer protected Au nanoclusters (MPCs) were prepared by modified Brust synthesis route [17,18] using dodecanethiol as capping agent followed by a post heat treatment (digestive ripening [19]) at 120 °C in presence of excess unbounded thiol (Au:thiol; 1: 30). In brief, particles were synthesized in a biphasic mixture of water and toluene (v:v; 1:1; 50 ml each) using 1:3 mole ratio of gold salt to thiol in ice bath. The biphasic mixture under low temperature (ice bath) was stirred for 30 min and subsequently reduced using drop wise addition of 0.1 M 20 ml aqueous NaBH4 solution. Upon the slow addition of NaBH4 , the pale yellow colour of gold salt was
transformed to reddish-violet indicating Au0 cluster formation and after vigorous stirring for 3 h, the Au0 clusters were found to be shifted to the organic phase turning the upper layer reddish-violet. After separation of non-aqueous portion from the aqueous layer, the non-aqueous layer was collected and the process was repeated 20 times to finally obtain 1 l of toluene containing both AuMPC and unbound thiol. The nonaqueous layer was then concentrated under vacuum at 40 °C to a final volume of 100 ml, where the ratio of Au to unbound thiols is typically 1:30. This reaction mixture was further boiled at 120 °C for 1 h in an oil bath and was subsequently cooled slowly at room temperature to form a dark red solution. The reaction mixture was further concentrated (10 ml) and the MPCs were precipitated by the addition of excess acetone (250 ml). The particles were allowed to settle and excess acetone was decanted to reduce the volume to 20 ml, followed by repeated centrifugation (7 times) and decantation to remove excess unbound thiol, and other unwanted reaction byproducts. 3.2. Characterization Fig. 1 shows a comparison of the TEM image of these particles, where a uniform Gaussian distribution with an average particle size of 4.63 0.04 nm is evident. The composition of the average sized AuMPCs is also approximated from the TEM results by calculating the nearest stable magic number as the full shell clusters, since certain magic numbers are known to be stable [20]. The average composition of these MPCs is approximated as Au2869 (DDT)541 , by assuming the shape of these particles as spherical, radius of Au atom as 144 pm, perfect fcc packing and further considering 2 S atoms per 3 Au atoms on the surface as reported for twodimensional SAM [21]. Moreover, particles are found to have bulk fcc structure as evident by the XRD in Fig. 2(a), whereas the inset shows the surface plasmon band of these MPCs at 523 nm. The purity of these particles have been confirmed from the NMR result (1 H), which shows that the individual peak positions of various protons [d 1:56, 1.25 and 0.84 ppm corresponding to the proton of CH2 (C2 position) close to S, CH2 (C3 –C11 positions) at the middle of the chain and the terminal CH3 (C12 position) group respectively] have been shifted and broadened significantly, indicating effective capping. However, the minor peak at d 1:06 ppm might be observed due to the negligible presence of uncapped DDT. 3.3. Electrochemical measurements The consecutive one electron QDL peaks observed in solution is diffusion controlled, which also obeys the Nernst equation similar to the traditional redox systems
N.K. Chaki et al. / Electrochemistry Communications 6 (2004) 661–665
663
Fig. 1. (a)–(b) HRTEM images of the purified AuMPCs passivated with dodecanethiol along with their particle size distribution in the inset of (a), showing uniform Gaussian distribution with an average particle size of 4.63 0.04 nm.
Fig. 2. (a) XRD patterns of AuMPCs dropcasted on a glass slide revealing bulk fcc structure, where the inset shows the surface plasmon resonance of these particles. (b) 1 H NMR response of these purified AuMPC in CDCl3 , showing that the peaks are significantly broadened and shifted from their ideal position due to the surface passivation of the Au core.
[7–16]. Consequently, the individual peak potentials in the QDL process can be considered as formal redox 0 potential E0 for each z=z 1 ‘‘redox couple or charge state couple’’, which can be described by 00 Ez;zþ1 ¼ Epzc þ ðz 1=2Þe=CCLU ;
ð1Þ
where, Epzc is the potential of zero charge (i.e., z ¼ 0) of the MPC, and z is assigned such that z > 0 and z < 0 corresponding to core ‘‘oxidation’’ and ‘‘reduction’’, respectively, and CCLU is the capacitance of the MPCs [7–10].
Interestingly, the smaller sized particles (<1.35 nm diameters) behave like molecular redox system, where bonding interactions (electron orbital-shell effects or pairing effects, or both) are dominated over electrostatic charging of the metal cores covered with a dielectric film [8]. This has been reflected as a large separation of DV in the potential of zero charge (i.e., Epzc ) regions in (1:1 < d < 1:9 nm) size regime, which is comparable to the HOMO–LUMO gap as anticipated as ‘‘Electrochemical Band Gap’’ [8]. However, the electrostatic principles governs the electronic charging of the core for d > 1:9 (i.e., double layer dominated over bonding interactions), where the capacitance is associated with the ionic space charge formed around an MPC dissolved in the electrolyte solution [7–10]. Fig. 3(a) shows the superimposed DPV response of these larger sized AuMPCs protected with DDT in dichloromethane on a Pt microelectrode (50 lm) at 0.96 mg ml1 concentrations at (i) 30, (ii) 20 and (iii) 10 mV pulse amplitude. Accordingly, Fig. 3(b) shows the DPV response in 1.6 mg ml1 MPC concentration, where the dip indicates the Epzc region. These DPV plots reveal a highly populated evenly spaced (ca. 80–90 mV) current peaks, where the ()1/0) and (0/1) charging peaks are marked for comparison (corresponding potentials are referenced to the Pt quasi-reference electrode). In conjunction with the results of DPV, these high populations of charge steps are also accessible by cyclic voltammetry. For example, Fig. 3(c) shows the superimposed response of these MPCs at (i) 100 and (ii) 50 mV s1 scan rate using 0.96 mg ml1 MPC concentration, where the successive one electron process are visible as evenly spaced sigmoidal waves. The higher capacitance values of these larger sized particles might, in principle, prevent better resolution by CV as compared to DPV. However, the derivatives (iii) and (iv) corresponding to the CV shown at graph (i) and (ii) are comparable with the DPV results. Interestingly, these MPCs do not show any signature of the ‘‘Electrochemical Band Gap’’ as observed for smaller sized particles, indicating that the process is particularly
664
N.K. Chaki et al. / Electrochemistry Communications 6 (2004) 661–665
Fig. 3. (a) The superimposed DPV response of AuMPCs at (i) 30, (ii) 20 and (iii) 10 mV pulse amplitude recorded using 0.96 mg MPCs/ml. (b) At 20 mV pulse amplitude using 1.6 mg MPCs/ml concentrations, showing highly populated QDL charging, where ()1/0) and (0/1) charging peaks are marked for comparison. (c) Superimposed CV response at (i) 100 and (ii) 50 mV s1 scan rate at 0.96 mg MPCs/ml concentrations along with their derivative response (iii) and (iv), respectively. (d) Graphs (i)–(iv) shows the superimposed highly linear ‘‘Z-Plots’’ corresponding to the DPV response (a) and (b), suggesting ideal QDL behavior.
governs by the double layer charging principles [7,8,14,16]. The capacitance (CCLU ) of these MPCs can be de00 termined from the slope of the linear plot of Ez;z1 vs. charge state (‘‘Z-plot’’) using the Eq. (1) [10]. Fig. 3(d) shows the superimposed ‘‘Z-plots’’ (i)–(iv) for these AuMPCs corresponding to the DPV responses shown at Fig. 3(a) and (b), where linear ðR2 > 0:99Þ behavior can be seen as expected for the ideal QDL behavior [7–10]. The capacitance (CCLU ) values calculated from the slope of the ‘‘Z-plot’’ (Fig. 3d) are 2.13, 1.86, 1.75 and 1.87 aF corresponding to graphs (i)–(iv) respectively. The above experimentally obtained CCLU values are in good agreement with the estimated CCLU of 4.63 nm sized AuMPCs (1.95 aF) with the help of a concentric sphere capacitance model as CCLU ¼ ð4pe0 er Þ½rðr þ dÞ=d;
ð2Þ
where, e0 is the primitivity of free space, er is the dielectric constant of the medium (taken as 3) respectively, ‘r’ is the radius of the MPC core (2.315 nm) and ‘d’ is the length of the passivating agent (1.52 nm) [10]. Interestingly, the full width at half maximum (FWHM) of each peak is found to be of ca. 40–45 mV particularly signifying the QDL process for these larger MPCs, which differs from the earlier reports of smaller
sized AuMPCs (110–115 mV) or ideal (90.6 mV) one electron transfer process for conventional redox species [6–9,22]. In particular, these sharp DPV peaks might be arising due to several complex interactions along with the reorganization of charges in clusters by electron selfexchange or dispropotionation among these assemblies of mixvalent redox centers at the electrode solution interfaces during electrochemical process [23–25]. More specifically, these types of interactions are widely known for conducting polymers, electroactive films or adsorption electrochemistry, where depending on the degree of the adsorbate–adsorbate interaction, substantial broadening or sharpening of the voltammetric peaks are observed [23–25]. We believe, that the double layer charging of these larger metallic MPC core could in principle solely responsible for these types of interactions, which perhaps is the reason for the observed narrower DPV peaks. Fig. 4 shows complex plane impedance plots of AuMPCs in DCM on a Pt disc (0.5 mm diameter), where a suppressed semicircle at the high frequency domain and a straight line with more than 45° angle to the real axis in the low frequency region are observed. Subsequently, the impedance data have been divided into high and low frequency regions for the convenience of interpretation and analyzed using the Randles equivalent circuit,
N.K. Chaki et al. / Electrochemistry Communications 6 (2004) 661–665
665
larger MPCs, which can be useful for designing future nanoelectronic devices. Acknowledgements NKC and BK gratefully acknowledge the Council of Scientific and Industrial Research (CSIR) and University Grants Commission (UGC), New Delhi. We are thankful to Mrs. Renu Pasricha, T.G. Gopakumar and Dr. S. Shulze for TEM facilities. References
Fig. 4. Complex plane impedance plots of AuMPCs in DCM on a Pt disc (0.5 mm diameter), where the suppressed semicircle at the higher frequency domain is fitted to a semicircle and analyzed with Randles equivalent circuit. Inset shows the linearly fitted (R2 > 0:99) modulus of the real and imaginary part of impedance data (10.72–1 Hz) with x1=2 .
considering the electrode reaction to be a one-step, one electron process [25]. The standard rate constant of the one electron QDL charging process is found to be 8.27 106 cm s1 , calculated using the following relationship, k ¼ RT =n2 F 2 RCT C
ð3Þ
assuming n ¼ 1; C ¼ C ðOÞ ¼ C ðRÞ ¼ ð9:5 107 mol cm3 ), a ¼ 0:5, T ¼ 273 K, RCT ¼ 3:105 104 X cm2 from the arc of the fitted semicircle in the real axis. However, the reversible feature observed in voltammetry (Fig. 3) especially for scan rates below 100 mV s1 do not agree with this sluggish nature and this could be justified due to various deviations from the ideal semiinfinite diffusion boundary condition due to adsorption, assumption of symmetric energy barrier, involvement of complex interactions among these MPCs, additional convective contribution due to the charged nature of MPCs etc. Further, the diffusion coefficient is estimated to be 1.21 107 cm2 s1 from the slope of the linearly fitted modulus of the real and imaginary part of the impedance data (10.72–1 Hz) with x1=2 as shown in inset of the Fig. 4, which is lower than that of its smaller analogues [9].
4. Conclusion In conclusion, our electrochemical results show that QDL charging is indeed possible for larger sized AuMPCs despite their higher capacitance values. We believe that these results will help to fill up the lacuna of information about the single electron transfer features of
[1] M.-C. Daniel, D. Astruc, Chem. Rev. 104 (2004) 293. [2] G. Schmid, M. Baumle, M. Geerkens, I. Heim, C. Osemann, T. Sawitowski, Chem. Soc. Rev. 28 (1999) 179. [3] D.I. Gittins, D. Bethell, D.J. Schiffrin, R.J. Nichols, Nature 408 (2000) 67. [4] D.L. Feldheim, C.D. Keatig, Chem. Soc. Rev. 27 (1998) 1. [5] R.P. Andres, T. Bein, M. Dorogi, S. Feng, J.I. Jenderson, C.P. Kubiak, W. Mahoney, R.G. Osifchin, R. Reifenverger, Science 272 (1996) 1323. [6] C.N.R. Rao, G.U. Kulkarni, P.J. Thomas, P.P. Edwards, Chem. Soc. Rev. 29 (2000) 27. [7] R.S. Ingram, M.J. Hostetler, R.W. Murray, T.G. Schaaff, J.T. Khoury, R.L. Whetten, R.L. Bigioni, D.K. Guthrie, P.N. First, J. Am. Chem. Soc. 119 (1997) 9279. [8] S. Chen, R.S. Ingram, M.J. Hostetler, J.J. Pietron, R.W. Murray, T.G. Schaaff, J.T. Khoury, M.M. Alvarez, R.L. Whetten, Science 280 (1998) 2098. [9] B.M. Quinn, P. Lijeroth, V. Ruiz, T. Kaaksonen, K. Kontturi, J. Am. Chem. Soc. 125 (2003) 6644. [10] S. Chen, R.W. Murray, S.W. Feldberg, J. Phys. Chem. B 102 (1998) 9898. [11] W. Cheng, S. Dong, E. Wang, Electrochem. Commun. 412 (2002) 412. [12] S. Chen, J.M. Sommers, J. Phys. Chem. B 105 (2001) 8816. [13] F.P. Zamborini, S.M. Gross, R.W. Murray, Langmuir 17 (2001) 481. [14] S.K. Haram, B.M. Quinn, A.J. Bard, J. Am. Chem. Soc. 123 (2001) 8860. [15] S. Chen, L.A. Truax, J.M. Sommers, Chem. Mater. 12 (2000) 3864. [16] Z. Ding, B.M. Quinn, S.K. Haram, L.E. Pell, B.A. Korgel, A.J. Bard, Science 296 (2002) 1293. [17] M. Brust, M. Walker, D. Bethell, D.J. Schiffrin, R.J. Whyman, Chem. Soc. Chem. Commun (1994) 801. [18] N.K. Chaki, M. Aslam, T.G. Gopakumar, J. Sharma, R. Pasricha, I.S. Mulla, K. Vijayamohanan, J. Phys. Chem. B 107 (2003) 13567. [19] B.L.V. Prasad, S.I. Stoeva, C.M. Sorensen, K.J. Klabunde, Langmuir 18 (2002) 7515. [20] J.D. Aiken III, R.G. Finke, Mol. Cata. A: Chem. 145 (1999) 1. [21] R.H. Terrill, T.A. Postelethwaite, C. Chen, C.D. Poon, A. Terzis, A. Chen, J.E. Hutchison, M.R. Clark, G. Wignall, J.D. Londono, R. Superfine, M. Falvo, C.H. Johnson Jr., E.T. Samulski, R.W. Murray, J. Am. Chem. Soc. 117 (1995) 12537. [22] A.J. Bard, L.R. Faulkner, Electrochemical Methods, Fundamentals and Applications, second ed., Wiley, New York, 2001 (Chapter 7 and 10). [23] J.F. Hicks, F.P. Zamborini, A.J. Osisek, R.W. Murray, J. Am. Chem. Soc. 123 (2001) 7048. [24] C.E.D. Chidsey, R.W. Murray, J. Phys. Chem. 90 (1986) 1479. [25] E.J. Laviron, J. Electroanal. Chem. 122 (1981) 37.