IR Spectroelectrochemical Cyclic Voltabsorptometry and Derivative

Apr 29, 2009 - Bao-Kang Jin,* Li Li, Jin-Ling Huang, Sheng-Yi Zhang, Yu-Peng Tian, and Jia-Xiang Yang. Department of Chemistry, Anhui University, Hefe...
0 downloads 0 Views 2MB Size
Download PDF
Anal. Chem. 2009, 81, 4476–4481

IR Spectroelectrochemical Cyclic Voltabsorptometry and Derivative Cyclic Voltabsorptometry Bao-Kang Jin,* Li Li, Jin-Ling Huang, Sheng-Yi Zhang, Yu-Peng Tian, and Jia-Xiang Yang Department of Chemistry, Anhui University, Hefei, 230039 China In this paper, we report the IR (infrared) CVA (cyclic voltabsorptometry) and DCVA (derivative cyclic voltabsorptometry) spectroelectrochemical techniques to elucidate an electrochemical mechanism. First we set potassium ferrocyanide as an example to explain the validity of this method. Then the electrochemical redox of two compounds, 1,4-benzoquinone and 1,4-bis(2-ferrocenylvinyl)benzene, was selected to be examined with this method. 1,4-Benzoquinone exhibits two single-electron waves in the cyclic voltammetric (CV) experiment, whereas two electroactive groups (Fc) are contained in p-(Fc-CHdCH)2BZ, but only one redox wave is observed. IR CVA results show that three IR absorption peaks in 1,4-benzoquinone, 1232 cm-1 (the absorption of final production), 1656 cm-1 (the absorption of original reactant), and 1510 cm-1 (the absorption of intermediate), and two IR absorption peaks in 1,4bis(2-ferrocenylvinyl)benzene, 1620 cm-1 (the absorption of final oxide production) and 1589 cm-1 (the absorption of intermediate), can be used to track the electron transfer. On the basis of the IR absorbance at the appropriate monitored wavelength (mentioned above), we can analyze simultaneously the concentration change of the corresponding redox transition during CV scans. Also the combination of the DCVA spectroelectrochemical technique with theory analysis allows reconstructing the current-potential (i-E) curve for each step of electron transfer. The reconstructed i-E curve can help us to understand the electrontransfer process. We believe IR CVA and DCVA spectroelectrochemical techniques can be applicable to the study of a wide range of complex electrochemistry processes. In situ spectroelectrochemistry, which is a combination of spectroscopy and electrochemistry, has been widely utilized to study various aspects of heterogeneous and/or homogeneous electron-transfer processes for 40 years.1 The method can provide more information about a redox reaction mechanism than can be acquired using traditional electrochemical methods alone. Many * To whom correspondence should be addressed. Phone and Fax: +86 551 5107342. E-mail: [email protected]. (1) Kuwana, T.; Dariington, R. K.; Leedy, D. W. Anal. Chem. 1964, 36, 2023– 2025.

4476

Analytical Chemistry, Vol. 81, No. 11, June 1, 2009

spectroelectrochemical methods have been proposed.2-8 In 1981, Bancroft et al.9 developed a spectroelectrochemical technique which couples cyclic linear potential sweep perturbation of an optically transparent electrode (OTE) with optical monitoring of the light-absorbing product of the electrode reaction. For the O + ne- ) R couple, the absorbance of the product of the electrode reaction (AR) is differentiated with respect to the linearly varying potential (dAR/dE) and is displayed as a function of sweep potential. The techniques are termed derivative cyclic voltabsorptometry (DCVA).10-14 The main advantages of DCVA are the insensitivity to nonfaradaic charge-consuming processes and simultaneous multicomponent analysis with a single potential scan by appropriate selection of monitored wavelength. It has been demonstrated that dA/dt (or dA/dE) versus E is morphologically identical to a cyclic voltammogram12,13 and the dA/dt curve is “cleaner” than the corresponding cyclic voltammogram. DCVA has been extensively applied to mechanism diagnosis and kinetic characterization. Up to now, according to our knowledge, DCVA is limited to UV-vis spectroelectrochemistry.15-18 The purpose of the work reported here is to introduce of the IR (infrared) CVA (cyclic voltabsorptometry) and DCVA spectro(2) Zhu, Y.; Wolf, M. O. Chem. Mater. 1999, 11, 2995–3001. (3) Casado, J.; Miller, L. L.; Mann, K. R.; Pappenfus, T. M.; Hernandez, V.; Lopez Navarrete, J. T. J. Phys. Chem. B 2002, 106, 3597–3605. (4) Naudin, E.; Mehdi, N. E.; Soucy, C.; Breau, L.; Belanger, D. Chem. Mater. 2001, 13, 634–642. (5) Kim, S. H.; Lee, J. W.; Yeo, I. H. Electrochim. Acta 2000, 45, 2889–2895. (6) Kulesza, P. J.; Zamponi, S.; Malik, M. A.; Miecznikowski, K.; Berrettoni, M.; Marassi, R. J. Solid State Electrochem. 1997, 1, 88–93. (7) Cohen, D. J.; King, B. C.; Hawkridge, F. M. J. Electroanal. Chem. 1998, 447, 53–62. (8) Ding, Z. F.; Ferm, D. J.; Franc, P.; Brevet, O.; Girault, H. H. J. Electroanal. Chem. 1998, 458, 139–148. (9) Bancroft, E. E.; Sidwell, J. S.; Blount, H. N. Anal. Chem. 1981, 53, 1390– 1394. (10) Bowden, E. F.; Hawkridge, F. M.; Chlebowski, J. F.; Bancroft, E. E.; Colin, T.; Blount, H. N. J. Am. Chem. Soc. 1982, 104, 1641–7644. (11) He, J. B.; Wang, Y.; Deng, N.; Lin, X. Q. Bioelectrochemistry 2007, 71, 157–163. (12) Astuti, Y.; Topoglidis, E.; Briscoe, P. B.; Fantuzzi, A.; Gilardi, G.; Durrant, J. R. J. Am. Chem. Soc. 2004, 126, 8001–8009. (13) Astuti, Y.; Topoglidis, E.; Gilardi, G.; Durrant, J. R. Bioelectrochemistry 2004, 63, 55–59. (14) Nekrasov, A. A.; Ivanov, V. F.; Gribkova, O. L.; Vannikov, A. V. Electrochim. Acta 2005, 50, 1605–1613. (15) Ghica, M. E.; Brett, A. M. O. Electroanalysis 2005, 17, 313–318. (16) Brett, A. M. O.; Ghica, M. E. Electroanalysis 2003, 15, 1745–1750. (17) Janeiro, P.; Brett, A. M. O. Anal. Chim. Acta 2004, 518, 109–115. (18) Panicco, P.; Astuti, Y.; Fantuzzi, A.; Durrant, J. R.; Gilardi, G. J. Phys. Chem. B 2008, 112, 14063–14068. 10.1021/ac9003634 CCC: $40.75  2009 American Chemical Society Published on Web 04/29/2009

Figure 1. Thin-layer CV (A) of 5 mM potassium ferrocyanide in 0.2 M KCL and corresponding 3D spectra (B) of in situ FT-IR spectroelectrochemistry, potential scan rate 2 mV/s.

electrochemical techniques to elucidate electrochemical mechanism, and we demonstrate that the method can be used to investigate the complex electrochemical mechanism advantageously. In this paper, 1,4-benzoquinone and 1,4-bis(2-ferrocenylvinyl)benzene were selected to be studied by IR CVA and DCVA spectroelectrochemical techniques, and we can see the dAR/dt versus E waveforms are morphologically identical with cyclic voltammetric (CV) responses for the electrochemical process. Additionally, DCVA is a powerful method for understanding the multistep electron-transfer electrode reaction mechanism, but with CV, it is difficult to distinguish between the multiconsecutive one-electron process and the single-step multielectron process if a single wave is observed in the voltammogram. The results show that the DCVA method can provide much helpful information about electrode reactions. MATERIALS Potassium ferrocyanide (Mallinckrodt, analytical reagent) was used as received. 1,4-Benzoquinone (Aldrich, 98%) was used after it was recrystallized from ethanol and dried overnight. 1,4-Bis-(2ferrocenylvinyl)benzene was synthesized according to Zuo et al.19 The structure and purity of the complexes were confirmed by mass, IR, and 1H NMR spectroscopy. Tetrabutylammonium perchlorate (TBAP) was recrystallized from ethanol and dried overnight under reduced pressure at 100 °C before use. Acetonitrile and methylene dichloride were dried by vacuum distillations over P2O5. A solution of 0.2 mol/L KCl (in water) or TBAP (in CH3CN or CH2Cl2) was used as supporting electrolyte. ELECTROCHEMISTRY Electrochemical experiments were performed with an EG&G PAR model 283 potentiostat/galvanostat. A homemade thin-layer in situ FT-IR-RAS (reflection absorption spectroscopy) spectroelectrochemical cell was used with a 4 mm diameter platinum disk working electrodes, a platinum wire auxiliary, and a Ag wire or Ag/AgCl reference electrode placed symmetrically around the working electrode.20,21 FT-IR SPECTROELECTROCHEMISTRY In situ FT-IR spectroelectrochemistry experiments were carried out at the same time as the electrochemistry experiments. Rapid(19) Zuo, C.; Zhou, Y.; Wu, J.; Tian, Y. Chin. J. Inorg. Chem. 2004, 20, 1018– 1020. (20) Liu, P.; Jin, B.; Cheng, F. J. Electroanal. Chem. 2007, 603, 269–274. (21) Jin, B.; Liu, P.; Wang, Y.; Zhang, Z.; Tian, Y.; Yang, J.; Zhang, S.; Cheng, F. J. Phys. Chem. B 2007, 111, 1517–1522.

scan time-resolved spectroscopic measurements were performed on a Nicolet Nexus 870 spectrometer equipped with a variableangle specular reflectance accessory (VeeMax II) and a HgCdTe/A (MCT/A) detector cooled with liquid nitrogen, the incident angle was adjusted to 45°, 20-50 interferograms were collected for each spectrum, the sampling interval is 0.8-1.48 s, and the spectral resolution is 16 cm-1. Experimental results were dealt with Grams/3D software. RESULTS AND DISCUSSION Potassium Ferrocyanide. To demonstrate the validity of IR CVA and IR DCVA, the electrochemical oxidation of Fe(CN)64to Fe(CN)63- and subsequent reduction of Fe(CN)63- back to Fe(CN)64- was performed. Figure 1A shows the cyclic voltammogram of potassium ferrocyanide. The shape of the CV shown is in accord with theoretical expectations for a reversible electrontransfer reaction in a thin-layer cell. The rapid-scan IR spectra recorded in situ during the electrochemical process in the wavenumber range of 1800-2300 cm-1 are shown in Figure 1B. The three-dimensional (3D) spectra were gathered during the CV scan between 0 and 0.4 V using the spectrum recorded at 0 V as the reference spectrum. Two IR absorption peaks, 2116 and 2038 cm-1, can be employed to track the electrochemical reaction. The peak at 2038 cm-1 is employed to track directly the disappearance of Fe(CN)64-, and 2116 cm-1 is used to follow the formation of Fe(CN)63-. The corresponding cyclic voltabsorptomograms (CVAs) are shown in Figure 2A. It is apparent from Figure 2A that there is an increase and decrease of absorbance at 2116 and 2038 cm-1 subject to the sweep potential (time), corresponding to the redox process. The absorbance at 2116 cm-1 gradually increases in the oxidation process and reaches its maximum at nearly 0.35 V (175 s) after which it starts to decrease in the reduction process and vanish at 0.05 V (375 s). The absorbance at 2038 cm-1 gradually decreases during oxidation process and reaches its minimum at nearly 0.35 V (175 s) and begins to increase gradually during the reduction process. First derivatives of CVAs are morphologically identical to the “ideal” CV response for electrochemical processes.9,12 The corresponding derivatives of the CVA data (DCVAs) are shown in Figure 2B. It is apparent from Figure 2B that the shapes of DCVA at 2038 and 2116 cm-1 both are similar to that of the CV shown in Figure 1A. (Peak-peak separation, ∆Ep ) 35 mV (CV), whereas ∆Ep ) 11 mV (DCVA)). Analytical Chemistry, Vol. 81, No. 11, June 1, 2009

4477

Figure 2. CVAs (A) and DCVAs (B) (smoothing with a fast Fourier transform smoothing algorithm) for the electrochemical reaction at 2038 cm-1 (black line) and at 2116 cm-1 (red line).

Figure 3. Relationship between (dA/dt)p and scan rate. Inset is the plot of E vs log[O]/[R].

The dependence of the (dA/dt)p (peak value of dA/dt) on scan rate is shown in Figure 3. The (dA/dt)p value is in direct proportion to the potential scan rate. The relationship is the same as peak current variation with scan rate under low potential scan rate. The logarithm of the resulting concentration ratio ([O]/[R] ) (A2038/A2038 max)/(A2116/A2116 max)) against the potential is also shown in the inset of Figure 3. A straight line with slope of 0.060 V, in accordance with Nernst behavior, can be observed. 1,4-Benzoquinone. 1,4-Benzoquinone is a well-studied example of an organic redox couple. A nine-membered square scheme is used to summarize the possible electrode reaction couple with proton-transfer reactions. In aprotic solvents, 1,4benzoquinone is known to undergo two successive one-electron reductions, corresponding to the formation of a radical anion intermediate and then the dianion. Figure 4A shows the thin-layer cyclic voltammogram of 1,4-benzoquinone. The rapid-scan IR spectra recorded in situ during the electrochemical process in the 1000-1800 cm-1 region are shown in Figure 4B. The 3D spectra were gathered during the scan between -1.2 and 0 V. The spectrum recorded at 0 V was used as the reference spectrum. Two distinct upward peaks at 1510, 1232 cm-1 were assigned to υC-O of the semiquinone anion radical and to υC-O of the final reduced production. Two downward peaks, at 1656 and 1109 cm-1, were assigned to υCdO of quinone and υCldO of ClO4(Figure 3B). The four IR absorption peaks can be employed to track the electrochemical reaction. The peak at 1656 cm-1 is employed to track directly the disappearance of original reactant 1,4-benzoquinone (BQ), 1232 cm-1 is used to follow the formation of dianion (Q2-), and 1510 cm-1 yields informa4478

Analytical Chemistry, Vol. 81, No. 11, June 1, 2009

tion about intermediate BQ-. The band at 1109 cm-1, assigned to υCldO, represents the change of ClO4- during the redox process. The corresponding cyclic voltabsorptomograms (CVA) of BQ are shown in Figure 5A. There is a periodic increase and decrease of absorbance at 1510 cm-1 with the sweep potential (time), corresponding to the reduction of BQ to intermediate BQ- and subsequent further reduction of BQ- to Q2-. Absorbance at 1232 cm-1 (assigned to υC-O with the formation of the benzene ring skeleton vibration) gradually increases in the reduction process and reaches its maximum at nearly -1.1 V (220 s). This absorbance diminishes during the oxidation process, and the absorption vanishes at 0.1 V (460 s). Also the absorbance at 1656 cm-1 gradually decreases in the reduction process and reaches a minimum at nearly -0.6 V (120 s). The intensity of the absorbance at 1656 cm-1 is steady from 120 to 360 s and begins to increase at nearly -0.6 V (360 s) in the oxidation process. This absorption vanishes at 0.1 V (460 s). The TBAP (ClO4-) absorbance changes during the redox process as deduced from the absorbance at 1109 cm-1. When BQ starts to be reduced to the intermediate BQ-, ClO4- anions decrease on the surface of the electrode to balance the charge equilibrium. As the reduction goes on, the intermediate BQ- is reduced to Q2- and more and more ClO4- anions leave the surface of the electrode. The absorbance at 1109 cm-1 gradually decreases in the reduction process and reaches a minimum at nearly -1.1 V (220 s). Also during the oxidation process, the oxidation of Q2- to intermediate BQ-, and the subsequent further oxidation of BQ- to BQ, ClO4- anions gradually return to the surface of electrode and reach a maximum at nearly 0.1 V (460 s) according to the absorbance at 1109 cm-1. It is also apparent from Figure 5B that the shape of DCVA at 1510 cm-1 corresponding to intermediate BQ- is similar to that of the CV shown in Figure 4A, where two redox couples are observed. Also, morphological DCVA couples the absorbance at 1656 cm1, which corresponds to the redox couple of BQ (the first electrochemical step), with the absorbance at 1232 cm-1, which is corresponds to the couple of Q2- (the second electrochemical step), and is the same with the corresponding CV shown in Figure 4A. We can devise a set of IR experiments to calculate the molar absorptivity coefficient ε at 1656 cm-1 using the Beer-Lambert law: A ) εcl

(1)

Figure 4. Thin-layer CV (A) of 5 mM 1,4-benzoquinone in 0.2 M TBAP CH3CN and corresponding 3D spectra (B) of in situ FT-IR spectroelectrochemistry, potential scan rate 5 mV/s.

Figure 5. CVAs (A) and calculated DCVAs (B) (smoothed with a fast Fourier transform smoothing algorithm) for the electrochemical reaction at 1232 cm-1 (red line), 1510 cm-1 (green line), and at 1656 cm-1 (black line). To make the DCVA data readily comparable to CV, the 1656 cm-1 (black line) and 1510 cm-1 (green line, the second reduction and the first oxidation) in the DCVA data were multiplied by -1.

where l, 25 × 10-4 cm, is the optical absorbance length, c is 20 mmol/L and is the concentration of BQ in anhydrous CH3CN solution, and ε is the molar absorptivity coefficient at 1656 cm-1. The experimental result shows that A ) 0.1002. The molar absorptivity coefficient in anhydrous CH3CN solution can be determined as follows: ε1656 cm-1 ) A/cl ) 2.0 × 103 L/mol·cm The absorption at 1656 cm-1 is equal to -0.015 in the potential range of -0.6 to -1.2 V (120-360 s). The potential range is sufficiently negative for complete (>99.9%) conversion from BQ to the Q2-. This data yields the optical absorbance length of the thin-layer electrolytic cell: l ) A1656 cm-1/(ε1656 cm-1) c ) 1.5 × 10-3 cm. Hence, the molar absorptivity coefficient at 1232 and 1510 cm-1 can be calculated to be 2.1 × 103 and 1.1 × 103 L mol-1 cm-1, respectively. These three peaks can be employed to track the concentration change of corresponding redox transition during CV. We can elucidate the relationship of the concentration and the sweep time (potential) as Figure 6. 1,4-Bis(2-ferrocenylvinyl)benzene. Figure 7A shows the cyclic voltammogram of 1,4-bis(2-ferrocenylvinyl)benzene. Only one redox couple is observed in CV. The shape of the CV shown in Figure 7A is a theoretical departure expected from theory for a reversible electron-transfer reaction in a thin-layer cell, due to uncompensated solution resistance. The phenomenon was first reported by Morrison et al.22 Flanagan et al.23 assumed, on the

Figure 6. Concentration distribution of BQ (red line), BQ- (green line), and Q2- (blue line) at different times (electrode potential) during a cyclic voltammetric experiment. Initial concentration (BQ) is 5 mM.

basis of theoretical analysis, that the electron transfer is two merged one-electron processes rather than a single two-electron transfer. The rapid-scan IR spectra recorded in situ during the CV scan in the 1750-1450 cm-1 region are shown in Figure 7B. The 3D spectra were gathered between 0.15 and 0.65 V using the spectrum recorded at 0.15 V as the reference spectrum. Two IR absorption peaks, 1620 and 1589 cm-1, can be employed to track the electrochemical reaction. The peak at 1620 cm-1 is employed to follow directly the formation of the final product, p-(Fc+-CHdCH)2BZ, and 1589 cm-1 is used to follow the formation of the intermediate, p-(Fc-CHd CH)2+BZ.21 Analytical Chemistry, Vol. 81, No. 11, June 1, 2009

4479

Figure 7. Thin-layer CV (A) of 5 mM 1,4-bis(2-ferrocenylvinyl)benzene in 0.2 M TBAP CH2Cl2 and corresponding 3D spectra (B) of in situ FT-IR spectroelectrochemistry, potential scan rate 2 mV/s.

Figure 8. CVAs (A), calculated DCVAs (B) (smoothed with a fast Fourier transform smoothing algorithm) for the electrochemical reaction at 1589 cm-1 (black line) and at 1620 cm-1 (red line), and reconstructed i-E curves (C) for i1, i2, and i () i1 + i2, total current). To make the DCVA data readily comparable to CV, the second reduction and the first oxidation in the DCVA data were multiplied by -1. All experiments were measured at scan rates of 2 mV/s.

The corresponding cyclic voltabsorptomograms (CVAs) are shown in Figure 8A. It is apparent from Figure 8A that there is a periodical increase and decrease of absorbance at 1589 cm-1 with the sweep potential (time), which corresponds to the oxidation of p-(Fc-CHdCH)2BZ to intermediate, p-(Fc-CHdCH)2+BZ, and subsequent further oxidation of p-(Fc-CHdCH)2+BZ to the final product, p-(Fc+-CHdCH)2BZ. These changes provide direct evidence for the assignment of each redox transition. The absorbance at 1620 cm-1 gradually increases in the oxidation process and reaches a maximum at nearly 0.55 V (200 s), and the absorbance begins to diminish at nearly 0.35 V (400 s) in the reduction process and vanishes at 0.18 V (480 s). It is apparent that about 60% of p-(Fc+-CHdCH)2BZ is produced before p-(Fc-CHdCH)2+BZ reaches its maximum concentration in the oxidation process. Hence, only a single wave can be observed in the CV (Figure 8A). It is apparent from Figure 8B that the shape of DCVA at 1620 cm-1 is similar to that of the CV shown in Figure 7A. However, the morphological DCVA at 1589 cm-1, in which two couples are observed, is quite different from the corresponding CV. The first oxidation (1) and reduction peaks (3) shown in DCVA indicate the formation of the intermediate, and the second oxidation (2) and reduction (4) imply the consumption of the intermediate. (22) Morrison, W. H., Jr.; Krogsrud, S.; Hendrickson, D. N. Inorg. Chem. 1973, 12, 1998–2004. (23) Flanagan, J. B.; Margel, S.; Bard, A. J.; Anson, F. C. J. Am. Chem. Soc. 1978, 100, 4248–4253.

4480

Analytical Chemistry, Vol. 81, No. 11, June 1, 2009

For an EE (consecutive electron-transfer) electrochemical process, the electrode reaction can be expressed as follows: A-e)B

(2)

B-e)C

(3)

Total current can be obtained by i ) i1 + i2

(4)

i1 and i2 are the faradic current produced in electrode reactions 2 and 3, respectively. From Faraday’s law, we can conclude that

Q1 ) nFs(

∫ C (x) dx + ∫ C (x) dx) ) ∫ i (t) dt l

0

l

B

0

t 0 1

C

(5)

Q2 ) nFs(

∫ C (x) dx) ) ∫ i (t) dt l

0

C

t 0 2

(6)

where s is the electrode area, l is the thickness of the thin-layer cell, and CB and CC are the concentrations of B and C, respectively. The relationship between the absorptivity and the concentration of B and C can be shown:

AB ) AC )

( )∫ ( )∫ 2εB 1000

0

2εC 1000

0

l

l

CB(x) dx ) KB′ cos θ

∫C

CC(x) dx ) KC′ cos θ

∫C

l

0

B

l

0

C

dx

(7)

dt

(8)

where εB and εC are the molar absorptivity coefficients at characteristic wavelengths, respectively, and θ is the angle of IR incidence. The “2” in eqs 7 and 8 is due to absorption of the solution before and after reflection. Substitution of eqs 7 and 8 into eqs 5 and 6 yields

(

Q1 ) nFs

AB AC + KB′ KC′

)

(10)

(

dAB /dt dQ1 dAC /dt ) nFs + dt KB′ KC′

(

dAC /dt dQ2 ) nFs dt KC′

)

)

(11)

(12)

The result demonstrates that the derivative cyclic voltabsorptometric curves for final production C are morphologically identical to cyclic voltammogram of the second electron-transfer process. Reconsidering the relationship of absorptivity and time (potential) in Figure 8A, we can conclude that product C exists only at the time from 200 to 300 s because all A and B are oxidized. Therefore CC ) 5 mM because the initial concentration of A is 5 mM and AC ) KC′

∫C l

0

C

dx ) 0.005KC′ l

(13)

yielding: K C′ ) AC/0.005l ) 3.94/l (Ac max ) 0.01972). If there are only B and C in the solution around 170 s, while CB + CC ) 5 mM, we can conduct that K B′ ) AB/(0.005 - AC/ 3.94)/l ) 23.8/l. Substitution of the constant into eqs 11 and 12 yields

(

i1 ) nFsl

) (

)

dAC /dt dAC dAB /dt dAB + + 0.254 ) K 0.042 23.8 3.94 dt dt (14)

(

i2 ) nFsl

)

dAB /dt dAC /dt + KB KC

(

i2 ) nFsl

The corresponding derivatives of eqs 9 and 10 show that

i2 )

(

i1 ) nFsl

(9)

AC Q2 ) nFs KC′

i1 )

The equation above shows that the derivative cyclic voltabsorptometric curves for final production C are morphologically identical to the cyclic voltammogram of the second electron transfer. The cyclic voltammogram of the first electron transfer is the linear combination of the derivative cyclic voltabsorptometric curves for intermediate B and final production C. Therefore, the relationship between faradic current and dA/ dt can be written as

dAC /dt dAC ) 0.254K 3.94 dt

dAC /dt KC

)

)

(16)

(17)

where KB and KC are constants which relate to the molar absorptivity of B and C, respectively, and the angle of IR incidence. Combining eqs 14 and 15 and the results of DCVA, we can reconstruct the voltammograms for each step of the two-electrontransfer process, and the result is shown in Figure 8C. Therefore, two midpoint potentials, E1 ) 343 mV and E2 ) 351 mV, for the first and the second electron-transfer step can be obtained. In addition, the apparent midpoint potential (current-potential (i-E) curve) is 348 mV and is in agreement with the midpoint potentials obtained by CV. CONCLUSION The approach presented here provides direct evidence for the assignment of each redox transition and represents a powerful approach to the study of the electron-transfer mechanism. We can track the concentration change of the corresponding redox transition during CV scans and distinguish the first and the second electron-transfer step based on the reconstructed i-E curves. We believe the IR CVA and DCVA spectroelectrochemical techniques should be applicable to the study of a wide range of complex electrochemistry process. ACKNOWLEDGMENT This work was supported by the National Nature Foundation of China (Grants 20775001, 20475001, 20875001, 50532030) and the Program for New Century Excellent Talents in University (China NECT-07-0002). We thank Professor Lin Xiangqin (USTC) for helpful discussion.

Received for review February 17, 2009. Accepted April 6, 2009. (15)

AC9003634

Analytical Chemistry, Vol. 81, No. 11, June 1, 2009

4481