J. Phys. Chem. B 2008, 112, 11171–11176
11171
Study on Anion Electrochemical Recognition Based on a Novel Ferrocenyl Compound with Multiple Binding Sites Qiaohua Tan, Li Wang,* Liang Ma, Haojie Yu, Jianhua Ding, Qingquan Liu, Anguo Xiao, and Guoqing Ren State Key Laboratory of Chemical Engineering, College of Materials Science and Chemical Engineering, Zhejiang UniVersity, Hangzhou, P. R. China, 310027 ReceiVed: June 6, 2008; ReVised Manuscript ReceiVed: June 20, 2008
A novel ferrocenyl anion receptor N,N,N,N-(dimethyl, ethyl, ferrocenecarboxylic amidodimethylene) ammonium fluoborate 2 with multiple binding sites was synthesized. Its anion recognition behaviors were investigated by CV, 1H NMR and UV-vis spectrum. It was found that the combination of two interactions enforced the anion binding ability and the binding selectivity of 2 to phosphate anion. The effects of scan rate on the CV curves of 2 with phosphate were also investigated. In different scan rate, the CV curves kept stable which indicated the strong binding between 2 and phosphate. According to relationships of peak potential, peak currents and scan rate of 2 binding with phosphate, the kinetic parameters of electrode process such as diffusion coefficient Dapp, surface transfer coefficient RnR, and standard rate constant k0 were calculated. 1. Introduction Anion recognition has become a hot research area because anions play important roles in biology, medicine, and environment.1-6 People have tried to design and synthesize various artificial anion receptors to effectively and selectively recognize anions since the first anion receptor was synthesized in 1960s.7-17 To design a good anion receptor, the crucial point is to find proper signal unit and design effective binding site. In the electrochemical anion recognition area, ferrocene group has proved to be a good signal unit because of its stable electrochemical properties and easiness to be detected by electrochemical methods such as cyclic voltammogram (CV).8 Besides that, the ferrocene group can participate in and modulate the binding event through alternating between its two redox states.8 As for binding site, amide,7 urea,9 and the hydroxyl group10 that can form hydrogen bonds with anions are mostly used. Quarterized nitrogen11 and positively charged pyridine12 are also chosen as binding sites on the basis of the electrostatic interaction. Others that can provide shape complementation16 are also considered. However, for the anion receptors reported, sensitivity and selectivity, the two basic characters of anion receptors, are not so good and are still challenges for scientists. Now, people realize that one single interaction in the receptor molecule such as hydrogen bond or electrostatic interaction or shape complementation may not be enough to improve the selectivity and sensitivity, especially in aqueous environment.18 Thus, combination of several interactions is being considered. It is believed that multiple binding sites which involve several different binding groups in a certain spacial position, such as linking of hydrogen bond, forming group and electrostatic interaction, providing group according to a special position, will improve the sensitivity and selectivity of anion receptors. However, so far, anion receptors, especially ferrocenyl anion receptors, with multiple binding sites are still seldom reported.19 Ferrocenyl compounds in anion recognition are always investigated by CV method. Our previous work has shown that * Corresponding author. Tel: +86-571-8795-3200. Fax: +86-571-87951612. E-mail:
[email protected].
some factors such as potential scan rate and electrolyte concentration have influences on CV behaviors.10 However, in anion recognition study of ferrocenyl compounds by CV reported, no one has focused on these factors, and the kinetics of the electrode process in electrochemical anion recognition has not been reported yet. In this work, we designed and synthesized a ferrocenyl anion receptor 2 with multiple binding site composed of amide group and the quarterized nitrogen linked by dimethylene group. Its anion recognition behaviors, its sensitivity, and its selectivity to phosphate were studied. The influence of scan rate and electrolyte concentration on CV responses to anions are also discussed, and the kinetic parameters of the electrode process anion recognition were calculated. 2. Experimental Sections 2.1. Instruments and Reagents. The CV measurements of solution were carried out with a CHI-600A electrochemical analyzer (CH Instruments, Inc., Austin, TX) in an undivided three-electrode cell. The working electrode was a Teflonshrouded Pt disk electrode (2 mm, geometric area: 0.0314 cm2), which was polished to a mirror finish with 0.05 µm Al2Cl3 paste on felt, cleaned by ultrasonication successively in 0.1 M NaOH, 1:1 HNO3, anhydrous ethyl alcohol, and double-distilled water, then dried, and used for electrochemical measurements. Pt-wire counter electrode and Ag-wire reference electrode were used. The ohmic potential (iR) drop of the investigated solution was measured and autocompensated with CHI-600A electrochemical analyzer. All CV experiments have been done for three times to obtain the reproducible result. In every CV experiment, the electrolyte is Bu4NBF4, the scan rate is 0.1 V/s, and the ferrocenyl compound concentration is 0.5 mM, unless specifically indicated. 1H NMR spectra were recorded with an AVANCE DMX500 NMR spectrometer by using tetamethylsilane (TMS) as internal standard at room temperature. Mass spectrum was recorded with an esquireplus mass spectrometer. All the reagents in use are from commercial markets. Before use, ferrocene carbonate acid (AR) is dried under vacuum, and
10.1021/jp805002a CCC: $40.75 2008 American Chemical Society Published on Web 08/07/2008
11172 J. Phys. Chem. B, Vol. 112, No. 35, 2008
Tan et al.
SCHEME 1: Synthesis of Ferrocenyl Compounds 1 and 2
solvents such as dichloromethane CH2Cl2 (AR), tetrahydrofuran (THF, AR), pyridine (AR), petroleum ether (60-90 °C, AR), triethylamine (AR), and methanol CH3OH (AR) are distilled twice in presence of molecular sieve. Other solid reagents such as sodium bicarbonate (AR), anhydrous magnesium sulfate (AR) and tetrabutylammonium tetrafluoroborate Bu4NBF4 (AR), tetrabutylammonium phosphate, Bu4NH2PO4 (AR), and tetrabutylammonium hydrogen sulfate Bu4NHSO4 (AR) are directly used. 2.2. Synthesis and Characterization of Ferrocenyl Compounds 1 and 2. The synthesis routes of ferrocenyl compounds 1 and 2 are shown in Scheme 1. Synthesis of ferrocenyl carbonate chloride was performed as follows. A total of 2.00 g (8.7 mmol) of ferrocenyl carbonate acid was suspended in a 100 mL three-necked flask containing 30 mL of CH2Cl2, and 20 drops of pyridine were added. Then,
Figure 1. CV curves of ferrocenyl compounds 1 and 2 in CH2Cl2.
Figure 2. CV curves of 2 before and after addition of H2PO4- in CH2Cl2. (a) H2PO4-/[Fe] is 0; (b) H2PO4-/[Fe] is 1.8; (c) H2PO4-/[Fe] is 2.4.
2.50 mL (28.6 mmol) of oxalyl chloride was added dropwise to the suspension mixture. The solution was reacted 16 h in 50 °C; then, the solvent was removed, and the excess oxalyl chloride was obtained in vacuum. A total of 2 × 60 mL of petroleum ether (60-90 °C) was then added, and the ferroceneyl carbonate chloride was extracted after reflux for 20 min. We remove the petroleum ether in vacuum, obtaining the red solid product ferroceneyl carbonate chloride 1.68 g (Yield: 77.8%). Synthesis of ferrocenyl compound N,N-dimethylaminoethylferrocenyl-carboxylicamide 1 was performed as follows. A total of 1.68 g (6.7 mmol) of ferrocenyl carbonate chloride was resolved in 5 mL CH2Cl2 and then added dropwise to a mixture of 15 mL of CH2Cl2, 0.90 mL of triethylamine (6.7 mmol), and 0.75 mL (13.4 mmol) of N,N-dimethylaminoethylamine. After reaction for 16 h at room temperature, the solution was washed with 50 mL of NaHCO3 aqueous solution (0.30 wt%) three times and then dried by anhydrous MgSO4. A total of 1.56 g of brown solid product 1 was obtained by removing the solvent in vacuum (Yield: 76.9%). 1H NMR (DMSO-d6): δ 2.16 (s, -N-CH3, 6H), 2.34 (t, -N-CH2- 2H), 3.25 (q, -CONH-CH2-, 2H), 4.14 (s, C5H4, 5H), 4.30, 4.74 (t, C5H4, 4H), 7.67 (t, -CONH-, 1H). Synthesis of ferrocenyl compound N,N,N,N-(dimethyl,ethyl, ferrocenecarboxylic amidodimethylene) ammonium fluoborate 2 was performed as follows. A total of 0.60 g (2.0 mmol) of ferrocenyl compound 1 was resolved in 20 mL of CH3OH, and then, 0.25 mL (3.0 mmol) of ethyl iodide was added. After being refluxed for 8 h, the mixture was concentrated to 5 mL and then precipitated in 100 mL of THF for three times to obtain the iodide ammonium. The iodide ammonium was then resolved in 30 mL of CH3OH/water (v:v ) 50:50) mixture solvent, and the excess NH4BF4 was added to the solution. The solution was extracted by 5 × 30 mL of CH2Cl2. After drying by anhydrous MgSO4, 0.33 g of orange solid product ferrocenyl compound 2 was obtained by removing the CH2Cl2 in vacuum (Yield: 40%). 1H NMR (DMSO-d ): δ 1.27 (t, -CH CH -, 3H), 3.07 (s, 6 2 3 -N-CH3, 6H), 3.38-3.42 (-CH2-N-CH2-, 4H), 3.54 (q, CONH-CH2-, 2H), 4.17 (s, C5H4, 5H), 4.37, 4.75 (t, C5H4, 4H), 8.07 (t, -CONH-, 1H). Mass spectrum (ESI): 329 (positive fragment), 87 (negative fragment). 3. Result and Discussion 3.1. Anion Recognition Behaviors Investigation of 2 by CV. We used CV method to study the electrochemical behaviors and anion recognition properties of 2 in CH2Cl2. The CV curve of 2 in CH2Cl2 is shown in Figure 1, with a pair of reversible redox peak which is typical of ferrocene group. The average peak potential Ep1/2 (Ep1/2 ) (Epc + Epa)/2) is 0.891 V, which is larger than the 0.835 V of 1, which means that 2 is easier to
Anion Electrochemical Recognition
J. Phys. Chem. B, Vol. 112, No. 35, 2008 11173
TABLE 1: Electrochemical Data when 3 equiv anions existed in CH2Cl2 anion
Epc (V)
no anion H2PO4HSO4Br-
0.801 0.391 0.652 0.911
no anion H2PO4HSO4Br-
0.859 0.373 0.711 0.874
a
Ep1/2 a (V)
∆Ep1/2 b (V)
receptor 1 0.865 0.635 0.783 0.956
0.833 0.513 0.717 0.937
-0.320 -0.116 0.104
receptor 2 0.922 0.662 0.718 0.946
0.891 0.518 0.715 0.910
-0.373 -0.176 0.019
Epa (V)
Ep1/2 ) (Epc + Epa)/2. b ∆Ep1/2 ) Ep1/2(bound) - Ep1/2(free).
reduce and harder to oxidize. This may be the result of the repulsion between the positively charged ferrocenium and the quarterized nitrogen. The CV responses of 2 to H2PO4- are shown in Figure 2. With the addition of H2PO4-, the initial peaks of the CV curve weakened and gradually disappeared, whereas a new pair of peaks in the lower potential gradually appeared (the electrochemical data are listed in Table 1). The average potential shift value ∆Ep1/2 is 0.373 V, which indicates the strong binding of 2 to H2PO4-20. The response of 1 to H2PO4- appears to be the same two-pair-peak phenomena but with a smaller potential difference (∆E ) 0.320 V, Figure S1 in the Supporting Information), which indicates that the ability of 2 to bind H2PO4- is better than that of 1, and the involvement of positively charged quarterized nitrogen did enforce the interaction between H2PO4- and 2. The recognitions of HSO4- and Br- by 2 were also studied (Figure 3). From CV curves, we can find that addition of HSO4leads to the two-pair-peak phenomena as well but with a much smaller potential change compared to that for H2PO4- (∆E ) 0.176 V), indicating that the interaction between HSO4- and 2 is much weaker than H2PO4- because of the relatively weaker basicity.7,20 As for Br-, there’s no apparent change of CV curves, indicating no recognition of 2 to Br- because of the weaker basicity and the unmatching of the spatial complementation. By comparing CV responses of 2 to H2PO4-, HSO4-, and Br-, it can be concluded that 2 showed selectivity to H2PO4-. 3.2. Anion Recognition Behaviors Investigation of 2 by 1H NMR and UV-vis. 1H NMR and UV-vis methods were used to also investigate the anion recognition of 2. The 1H NMR spectrum of 2 before and after the addition of H2PO4- is shown
in Figure 4. The addition of H2PO4- made the amide proton shift to low field from 8.07 to 8.33, and protons in ferrocenyl group and methylene group linked to the quarterized nitrogen were also influenced, indicating their involvement in the binding events. With the addition of H2PO4-, the UV-vis changes of 2 with the addition of H2PO4- showed interesting responses. The character peak of the amide group at about 237 nm changed apparently after H2PO4- was added; however, its intensity first decreased when less than 1 equiv H2PO4- was added (Figure 5a) and then gradually recovered when more H2PO4- was added (Figure 5b). The relationship of the intensity and the amount of H2PO4- is shown in Figure 5c. We can find that points for 1 equiv and 2 equiv are key points. Our previous work has shown that the involvement of carbonyl group in hydrogenbond forming leads to the decrease of the intensity of its character peak.17 Thus, we conclude that when H2PO4- was added, the carbonyl group of 2 was first involved in hydrogen bond, reflected by the decrease of the intensity of its character peak. When more H2PO4- was added, the change of binding model of 2 to H2PO4- from 1:1 mode to 1:2 mode somewhat weakened the hydrogen bond of the carbonyl group because of the steric effect, reflected by the recovery of intensity of its character peak. The possible binding mode of 2 to H2PO4- is illustrated by the molecular model and Kekule model in Figure 6. As for HSO4-, it was bound by 1:1 mode from the UV-vis spectrum (Figure S2 in the Supporting Information), indicating that the synergy effect of the positively charged quarterized nitrogen seems not apparent. 3.3. Kinetics of Electrode Process in Electrochemical Anion Recognition of 2. From our previous work, CV behaviors were influenced by several factors such as scan rate and electrolyte concentration.17 The influence of scan rate on anion electrochemical recognition of 2 was investigated. From Figure 2, we know that 3 equiv H2PO4- guaranteed the total binding of H2PO4- with compound 2. Thus, we studied the CV behaviors of 2 with 3 equiv H2PO4- in CH2Cl2 at different scan rates (Figure 7). We found that at different scan rates, the CV curves were stable, indicating that the binding of H2PO4- and 2 is strong and stable in the scan rate range, and we can treat them as a whole, named 2-H2PO4-. It was shown that with the increase of scan rate, the peak currents increased. Over a wide scan rate range, the peak current values ip of 2 are proportional to the square root of the scan rate, V1/2. (The results of the linear correlation between ip and V1/2 are listed in Table is similar to that of 2-H2PO4-. These relationships indicate
Figure 3. CV curves of 2 before and after addition of HSO4- (A) and Br- (B) in CH2Cl2. (a) [Anion]/[Fe] is 0; (b) [Anion]/[Fe] is 1.4; (c) [Anion]/[Fe] is 2.0.
11174 J. Phys. Chem. B, Vol. 112, No. 35, 2008
Figure 4.
1
Tan et al.
H NMR spectra of 2 before (a) and after (b) the addition of one equivalent of H2PO4-.
Figure 5. UV-vis spectrum of 2 with the addition of H2PO4-. (a) From top to bottom, H2PO4-/2 ranges from 0 to 1.2. (b) From top to bottom, H2PO4-/2 ranger from 1.2 to 4.0. (c) Changes of UV-vis intensity at 237nm with the addition of H2PO4-.
Figure 6. Possible step-by-step binding model of 2 to H2PO4-. (A) Molecular structure of 2 binding H2PO4-, and (B) Kekule model of 2 binding H2PO4-.
that the charge transport obeys Fick′s Law,21 and the electrode processes in solution were diffusion-controlled. According to the CV theories,21 for irreversible process, there are relationships between peak current, peak potential, and potential scan rate as follows,
( RTF )
1⁄2
ip ) 0.4985nFAC∗0
[
ip ) 0.227nFAC∗0 k0 exp -
1⁄2 (RnR)1⁄2D1⁄2 0 V
(1)
]
RnRF (E - E0′) RT p
(2)
where n, A, D0, RnR, and k0 are the electrochemical stoichiometry, surface area of the electrode (cm2), diffusion
coefficient of electroactive species (cm2/s), concentration of electroactive species in solution (mol/cm3), surface transfer coefficient, and standard rate constant (cm/s), respectively. RnR, D0, and k0 of 2 and 2-H2PO4- were calculated and are listed in Table 2. Generally, RnR represents rhw electrode surface electron exchange efficiency, reflects the reversibility of electrode process, and influences the peak potentials and the symmetry of the curves. From the CV curves of 2 and 2-H2PO4-, the reversibility of 2 is better than that of 2-H2PO4-, but the trend is the opposite for RnR, which can be explained by that in the sample of 2, where the potential scan rate did not reach below zero and adsorption happened
Anion Electrochemical Recognition
J. Phys. Chem. B, Vol. 112, No. 35, 2008 11175
Figure 7. CV curves of 2 only (A) and 2 with 3 equiv H2PO4- (B) in CH2Cl2 at different scan rates. From inner to outer, the scan rate are 0.14, 0.18, 0.22, 0.26, 0.30 V/s. Supporting electrolyte: 0.1 M Bu4NBF4, [Fe] ) 0.5 mM.
TABLE 2: Kinetic Parameters of Electrode Processes for 2 and 2-H2PO4- in CH2Cl2 peak type
RnR
k0
(cm/s)
5
2
Dapp (10 cm /s)
-
reduction peak oxidation peak
0.718 0.869
reduction peak oxidation peak
0.432 0.271
2–H2PO4 1.513 0.619
0.344 0.128
2 8.290 10.176
2.432 3.770
in the electrode surface which impeded the electron exchange. In the sample of 2, RnR of the reduction peak is larger than that of the oxidation peak, which is in accordance with the usual cases22-25 because of the relative instability of the oxidation state. However, from the sample of 2-H2PO4-, we discovered that the addition of H2PO4- can stabilize the oxidation state because of their combination. k0 represents the time that the redox system needs to reach the balance and can also reflect the reversibility of the electrode process. In our two samples, the much lower k0 of 2-H2PO4compared with that of 2 is reflected by the much larger ∆Ep, because the system with lower k0 needs a larger superpotential (represented by ∆Ep) to drive.21 The diffusion coefficient Dapp of the 2-H2PO4- is much lower than that of 2 because of the binding of anions making the fluidic volume much larger. 4. Conclusion In conclusion, a new ferrocenyl anion receptor 2 with specially designed multiple binding sites of amide and positively charged nitrogen was synthesized. Compared to its counterpart 1 with just a single binding site of amide, 2 showed higher sensitivity to H2PO4- and thus proved the enhancement effect of the multiple binding sites. The CV and UV-vis responses of 2 to H2PO4- indicated the unique step-by-step binding mode from 1:1 to 1:2 of 2 to H2PO4-, which differentiates the binding of H2PO4- by 2 compared to that of HSO4- and Br-. The good performance of 2 in anion recognition indicates a direction in the binding site design of anion receptor. Influences of scan rate on anion binding were also studied. It was found that at different scan rates, the CV curve of 2 with 3 H2PO4- remained stable. The linear relationship of the peak current with the square root of the scan rate in large a scan range indicates that the electrode process is diffusioncontrolled. Kinetic parameters of the electrode process such as the diffusion coefficient Dapp, surface transfer coefficient RnR, and standard rate constant k0 can be calculated and give
us much information of the electrode process in the electrochemical anion recognition. Acknowledgment. This work is supported by the National Nature Science Foundation of China (no. 20572097 and o. 20672097) and Science and Technology Program of Ningbo. We also thank Prof. R. James Cross at Yale University for the kind discussions. Supporting Information Available: CV curves of 1 after addition of different anions, UV-vis spectra of 2 to sulfate, 1H NMR spectra, mass spectra, and linear relationship of peak potential, peak current,and scan rate atr shown. This information is available free of charge via the Internet at http:// pubs.acs.org. References and Notes (1) Quinton, P. M. FASEB J. 1990, 4, 2709–2711. (2) Renkawek, K.; Bosman, M. NeuroReport 1995, 6, 929–932. (3) Holm, E. Radiochim. Acta 1993, 63, 57–59. (4) Deng, L. B.; Wang, L.; Yu, H. J.; Wang, J. J.; Dong, X. C. J. Appl. Polym. Sci. 2008, 107, 1539–1546. (5) Stry, L. Biochemistryl, 4th ed.; W. H. Freeman & Co.: New York, 1988. (6) Deng, L. B.; Wang, L.; Yu, H. J.; Dong, X. C.; Huo, J. Des. Monomers Polym. 2007, 10, 131–143. (7) Beer, P. D.; Graydon, A. R.; Johnson, A. O. M.; Smith, D. K. Inorg. Chem. 1997, 36, 2112–2117. (8) Reynes, O.; Bucher, C.; Moutet, J. C.; Royal, G.; Aman, E. S.; Ungureau, E. M. J. Electroanal. Chem. 2005, 580, 291–296. (9) Oton, F.; Tarraga, A.; Espinosa, A.; Velasco, M. D.; Bautista, D.; Molina, P. J. Organomet. Chem. 2005, 70, 6603–6605. (10) Tan, Q. H.; Wang, L.; Yu, H. J.; Deng, L. B. J. Phy. Chem. B 2007, 111, 3904–3909. (11) Reynes, O.; Royal, G.; Chainet, E.; Moutet, J. C.; Amana, E. S. Electrochim. Acta 2003, 15, 65–68. (12) Reynes, O.; Buche, C.; Moutet, J. C.; Royal, G.; Aman, E. S. Chem. Commun. 2004, 428–429. (13) Fenniri, H.; Lehn, J.; Rigault, A. M. Angew. Chem., Int. Ed. Engl. 1996, 35, 337–339. (14) Quesada, S.; Seel, C.; Prados, P.; Mendoza, J.; Dalcol, I.; Giralt, E. J. Am. Chem. Soc. 1996, 118, 277–280. (15) Reetz, T.; Niemeyer, M.; Harms, K. Angew. Chem., Int. Ed. Engl. 1991, 30, 1472–1478. (16) Boiocchi, M.; Fabbrizzik, L.; Piovani, G.; Talietti, A. Angew. Chem., Int. Ed. 2004, 43, 3847–3851. (17) Valerio, C.; Fillaut, J. L.; Ruiz, J.; Guittard, J.; Blais, J. C.; Astruc, D. J. Am. Chem. Soc. 1997, 119, 2588–2595. (18) Zhang, S.; Echegoyen, L. J. Am. Chem. Soc. 2005, 127, 2006– 2011. (19) Steed, J. W. Chem. Commun. 2006, 2637–2649. (20) Labande, A.; Ruiz, J.; Astruc, D. J. Am. Chem. Soc. 2002, 124, 1728–1732.
11176 J. Phys. Chem. B, Vol. 112, No. 35, 2008 (21) Bard, A. J.; Faulkner, L. R. Electrochemical Methods; John Wiley & Sons Inc.: New York, 1980. (22) Wang, X. J.; Wang, L.; Wang., J. J.; Chen, T. J. Phys. Chem. B 2004, 108, 5627–5633. (23) Chen., T.; Wang, L.; Jiang, G. H.; Wang, J. J.; Dong, X. C.; Wang, X. J.; Zhou, J. F.; Wang, C. L.; Wang, W. J. Phys. Chem. B 2005, 109, 4624–4630.
Tan et al. (24) Chen, T.; Wang, L.; Jiang, G. H.; Wang, J. J.; Wang, X. J.; Zhou, J. F.; Wang, J. F.; Chen, C.; Wang, W. J. Electroanal. Chem. 2006, 586, 122–127. (25) Huo, J.; Wang, L.; Yu, H. J.; Deng, L. B.; Zhou, J. F.; Yang, Q. J. Polym. Sci., Part B: Polym. Phys. 2007, 45, 2880–2889.
JP805002A
